Docstoc

UNITED STATES OF AMERICA 1 UNITED STATES

Document Sample
UNITED STATES OF AMERICA 1 UNITED STATES Powered By Docstoc
					                                                                       1


                    UNITED STATES OF AMERICA

                  FOOD AND DRUG ADMINISTRATION

                             + + + + +

                            WORKSHOP ON

       INNOVATIVE SYSTEMS FOR DELIVERY OF DRUGS AND

                            BIOLOGICS:

    SCIENTIFIC, CLINICAL, AND REGULATORY CHALLENGES

                             + + + + +

                     TUESDAY, JULY 8, 2003

                             + + + + +



                  The workshop was held in the Grand

Ballroom of the Bethesda Marriott, 5151 Pooks Hill

Road, Bethesda, Maryland, at 8:00 a.m.



PRESENT:

DAVID FEIGAL, JR., M.D., M.P.H., Center Director,

           CDRH

MARK McCLELLAN, M.D., Ph.D., Commissioner, FDA

ROBERT LANGER, Ph.D., Massachusetts Institute of

           Technology

DAVID C. KLONOFF, M.D., U.C. San Francisco

                            NEAL R. GROSS
                    COURT REPORTERS AND TRANSCRIBERS
                        1323 RHODE ISLAND AVE., N.W.
(202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       2


PRESENT (Continued):

JONATHAN B. KRUSKAL, M.D., Ph.D., Harvard Medical

           School

RICHARD E. KUNTZ, M.D., M.Sc., Harvard Medical

           School

AJAZ HUSSAIN, Ph.D., Nektar Therapeutics

CHET LEACH, Ph.D., Nektar Therapeutics

BILL VAN ANTWERP, Ph.D., Medtronic MiniMed

KEVIN C. SKINNER, V.M.D., Genzyme Corporation

JONATHAN S. KAHAN, ESQ., Hogan and Hartson, L.L.P.

KEITH SMITH, Becton, Dickinson, and Company

CHRISTINE ALLISON, M.S., RAC, Eli Lilly and Company

JOHN JENKINS, CDER

MARK KRAMER, FDA Office of Combination Products

ASHLEY B. BOAM, MSBE, CDRH

RICHARD P. FELTEN, CDRH

DAN SHAMES, M.D., CDER

JESSE GOODMAN, M.D., FDA

NANCY ISAAC, Aerogen




                            NEAL R. GROSS
                    COURT REPORTERS AND TRANSCRIBERS
                        1323 RHODE ISLAND AVE., N.W.
(202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                    3




                      C-O-N-T-E-N-T-S

                                                              PAGE

Welcome, David Feigal, Jr., M.D., M.P.H. ..........6

Presentation of Mark McClellan, M.D., Ph.D. .......9

Presentation of Robert Langer, Ph.D. .............26

Presentation of David C. Klonoff, M.D. ...........71

Presentation of Jonathan B. Kruskal, M.D., Ph.D. 119

Presentation of Richard E. Kuntz, M.D., M.Sc. ...155

Presentation of Chet Leach, Ph.D. ...............188

Presentation of Bill Van Antwerp, Ph.D.             ........210

Presentation of Kevin C. Skinner, V.M.D. ........224

Presentation of Ajaz Hussain, Ph.D. .............235

Public Comment, Dr. Paul Goldfarb ...............254

Presentation of Jonathan S. Kahan, Esq. .........271

Presentation of Keith Smith .....................297

Presentation of Christine Allison, M.S., RAC ....322

Presentation of Mark Kramer .....................339

Presentation of Ashley B. Boam ..................359




                         NEAL R. GROSS
                 COURT REPORTERS AND TRANSCRIBERS
                     1323 RHODE ISLAND AVE., N.W.
(202) 234-4433       WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                            4




1                        P-R-O-C-E-E-D-I-N-G-S

2                                                          (8:14 a.m..)

3                     DR. PROVOST:     Good morning.        We're going

4    to go ahead and get started.

5                     My name is Mariam Provost.           I work for

6    the FDA and the Center for Devices and Radiological

7    Health.

8                     And I just want to make a few

9    announcements before we introduce the first speaker.

10    I also want to say welcome to the people across the

11   hall on watching us on the TV and also people who

12   are phoning in.       We also are providing this

13   conference through an audio hookup.               So welcome to

14   everybody.

15                    And I also want to say thank you to all

16   of the speakers who have agreed to come today.                   I

17   think we have a very interesting program.               It's a

18   very full program.        So there's just a couple of

19   things that I want to mention.

20                    In order so that we can stay on time as

21   best we can, we have structured the morning session

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                               5


1    so that there's a question and answer period.                    So we

2    would ask that if you do have questions, if you

3    could wait until the question and answer period to

4    ask them, I think that will help us to keep on time.

5                     I also do want to mention there is going

6    to be a panel discussion at the end of the day.                      So

7    if you don't get a chance to ask some burning

8    questions because of the limited time, you can save

9    your questions for the end of the day, and we do

10   have 45 minutes set aside for panel discussion.

11                    We are, as I mentioned, audio

12   broadcasting this conference.             It's also being

13   transcribed.      So if you do have a question, we ask

14   that you identify yourself and also please speak

15   into the microphone so that everybody can hear.

16                    And, finally, very important, lunch.

17   We're a pretty big group here today.                  So that

18   everybody can get lunch and get lunch on time, we've

19   arranged with the hotel to provide a box lunch, and

20   there is an attendant from the hotel who is here out

21   in the hallway, and they ask that you order the

22   lunch by 9:30 this morning.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                            6


1                     So if you want to get a lunch, please, I

2    urge you to order your lunch now so that it will be

3    here when you need it.

4                     And that's all for the announcements of

5    that type.       I would just like to introduce Dr. David

6    Feigal, who is the Director of the Center for

7    Devices and Radiological Health, who is going to

8    give us some welcoming remarks.

9                     (Applause.)

10                    DR. FEIGAL:     Well, thanks.

11                    One of the most important things that I

12   could do this morning is to thank Mariam Provost and

13   Vickie Babb, who arranged this meeting on relatively

14   short notice as far as meetings go, and the

15   challenge of finding a room like this that's ideally

16   configured for talks and speeches.               The only one

17   that actually was more interesting was one once

18   where we had a lot of press with lights, and it had

19   mirrored columns all around the room.

20                    (Laughter.)

21                    DR. FEIGAL:     So every time they would

22   turn it was like begin inside a prism.                It was very

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                           7


1    interesting.

2                     One of the real challenges in medical

3    product development is to bring the first of a kind

4    to the market.      As we've looked back at what kinds

5    of applications are approved rapidly and which

6    applications take longer, it's quite clear that the

7    first of a kind products are often difficult.

8                     It's also very clear that if we can sit

9    down and have a discussion of what is needed to

10   establish the kinds of information that you need to

11   bring a product to market, that you can facilitate

12   this process.      And there are times when this is done

13   quite formally in the shape of developing guidances.

14                    There are other times when it seems to

15   evolve product by product.

16                    Now, it's challenging to bring one new

17   product to market.        It's even more challenging when

18   you have a combination of products.               When you have

19   two products, one of which may be novel, both of

20   which may be novel, you have particular challenges

21   to know exactly what is the regulatory path going to

22   be.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                               8


1                         It's also a challenge when you yoke

2    together a pharmaceutical, which often comes from a

3    very large and well resourced company, with the

4    device industry where many of the innovators are

5    small companies work on closer margins, on more

6    rapid cycles.

7                         There's a difference in the way that the

8    intellectual property of devices is protected than

9    drugs.           So there are many, many challenges, and as

10   we looked for topics that we could address and begin

11   to develop, the whole concept of combination

12   products and particularly products where there was a

13   novel mechanism of delivering a drug or a biological

14   therapeutic seem to be particularly timely.

15                        So my task this morning is to moderate

16   the session, introduce our speakers, and this

17   afternoon we'll come back and have time to actually

18   more explicitly talk about some of the regulatory

19   challenges.

20                        As you are aware, if you followed some

21   of the developments in the center in the last four

22   or five years, we view the regulatory process as an

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           9


1    intensely scientific one.           This isn't a type of a

2    decision-making that can be done simply by

3    developing checklists and looking for completeness

4    or other types of processes.

5                     So it's appropriate that we begin this

6    morning with an intense look at some of the science

7    and some of the exciting science in some of the

8    important disease areas, and that we begin the

9    morning with some remarks from Dr. Mark McClellan,

10   who is Commissioner of the FDA and who is

11   responsible for this meeting, which will probably

12   just be the start of a series of workshop-styled

13   meetings on product development.

14                    So with that, let me turn the mic over

15   to Mark and let's get started.

16                    DR. McCLELLAN:      Thank you, David.

17                    It's a pleasure to be here with you all

18   this morning at this new workshop on innovative

19   systems for the delivery of drugs and biologics.

20   This is a particularly important pleasure for me

21   because of all of the people here with our devices

22   center and with the biologics center and our drugs

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         10


1    center who have contributed to this effort.

2                     As David mentioned, this is an early

3    effort in what I think will be a series of programs

4    designed to focus on the important questions of

5    emerging technologies and our effective approaches

6    to regulating them, to demonstrating that they're

7    safe and effective in getting better treatments to

8    patients as quickly as possible.

9                     I also want to spend a minute thanking

10   Dr. Robert Langer for his work and his contribution

11   to FDA, in general, and to this meeting, in

12   particular.       Dr. Langer just finished as Chair of

13   our Science Board for several years, and that was

14   only one of many efforts in improving biomedical

15   technology.

16                    Dr. Langer is a professor at MIT who has

17   made many contributions in chemical and biological

18   engineering, ranging from insights in basic science

19   and improvements in biomedical technology to

20   actually bringing those products to market through

21   patents and through the development process.

22                    And I think Dr. Langer's efforts

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         11


1    exemplify the kind of work that we want to highlight

2    here and the kind of perspectives that we want to

3    bring to the FDA's efforts in these development

4    workshops.

5                     One of his recent articles, which is in

6    the packet included for this meeting, is on how to

7    get drugs where they need to go, and I think this

8    conference and the efforts that will follow from it

9    are an effort to build on that by figuring out how

10   to get drugs where they need to go quicker and more

11   effectively and more safely.            That's the goal that

12   we are attempting to fill with this workshop effort.

13                    As Dr. Feigal mentioned, this is the

14   first in a number of workshop that have developed

15   from a strategic planning process that we've

16   undertaken at the FDA over the last six months or

17   so.      This is an effort to develop clear guidance,

18   clear regulatory pathways for product developers in

19   a range of innovative areas.            It includes not only

20   novel systems for delivering drugs and biologics

21   where they need to go, but also such areas as

22   pharmacogenomics and cell and gene therapy, as well

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         12


1    as many priority areas for product development, such

2    as cancer treatments, obesity treatments, and

3    treatments for diabetes.

4                     In our strategic planning process, these

5    were areas where our staff felt that there were

6    opportunities if not to actually identify more clear

7    and effective regulatory pathways, at least a need

8    to take stock of recent developments in the sciences

9    as applied to product development.

10                    And so that's why we're having these

11   activities where we can get people together and

12   figure out if there are clear ways in which we could

13   improve regulatory pathways.            This is more important

14   today than ever because of the tremendous potential

15   out there for improvements in medical technology.

16                    You all are quite familiar with what

17   innovations in health care have brought to patients

18   in recent decades.        For example, treatment of heart

19   attack, which used to involve largely supportive

20   care as recently as a few decades ago, has

21   transformed as a result of innovations in drugs,

22   biologics, devices, and combination products, has

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         13


1    transformed heart attack care into a condition that

2    most people now should expect to survive.

3                     This is a big change in recent decades.

4    Diabetes is also an area where tremendous changes

5    have occurred, and some of the greatest improvements

6    in the treatment of these conditions have come from

7    combination products, devices and drugs, devices and

8    biologics working together.            These involve

9    treatments that may permit the delivery of

10   medications more accurately and effectively, as is

11   the case in a product that we approved yesterday

12   that combined a blood sugar monitoring system with a

13   continuous insulin delivery pump to permit more

14   accurate and at least the promise of more accurate

15   and timely delivery of insulin on an ongoing basis.

16                    It includes treatments that permit drugs

17   to get to the right place in the body more

18   accurately, as in some of the liposomal delivery

19   systems that have been developed recently.               It

20   includes ways of targeting particular cells more

21   effectively, for example, through new

22   nanotechnologies.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         14


1                     So there are many applications of new

2    technology in the area of combination products, and

3    the potential for these technologies to have an

4    impact on improving patient care in the years ahead,

5    I think, is even greater.           But it's not something

6    that's going to happen automatically.

7                     And one of the things that has concerned

8    me since coming to FDA is all that I've been able to

9    learn about some of the challenges facing product

10   development today.        If you look at just plain, old

11   drugs, small molecule drugs which in many ways are

12   not the only kind of innovative treatment coming

13   along now, the development process has gotten

14   considerably longer and more expensive, and this is

15   not something I think is the fault of regulation

16   primarily or maybe even at all, but it is a fact.

17                    It now costs, according to some

18   estimates, over $800 million to develop a new drug,

19   and while that number is somewhat controversial,

20   there's no arguing with the fact that it has gotten

21   a lot more expensive than it used to be because of a

22   more extensive preclinical development and testing

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         15


1    process.

2                     It has also gotten more uncertain than

3    it used to be with only a small fraction of the

4    drugs that enter clinical development actually

5    resulting in applications to the FDA and only less

6    than one in two that make it even to the advanced

7    phases of clinical testing, the so-called Phase III

8    trials, resulting in applications to FDA.

9                     And in the past few years, we've seen a

10   downturn in the number of new product applications

11   coming into the agency, and so that's an area of

12   concern where, on the one hand, the amount of

13   investment in new research and development both in

14   the private sector and in the government through

15   increases in the NIH budget have reached an all time

16   high, but on the other hand, we're not yet seeing

17   that translate into a significant upturn in the

18   number of valuable new products reaching patients.

19                    And this may be something that is just

20   going to take a matter of time to resolve.               I've got

21   a lot of long term confidence in the biomedical

22   industry to improve care, but this delay is

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          16


1    something that adds to health care costs because of

2    the cost that goes along with developing new

3    products and has an impact on quality of care

4    because it results in longer times before patients

5    can get access to safe and effective new treatments.

6                      So this is a real challenge as products

7    become more complex, and in meeting this challenge

8    FDA can and will maintain its gold standard for the

9    world for product approvals.             That means we will

10   continue to make sure that products are safe and

11   effective before we approve them.

12                     At the same time, with all of these

13   insights coming in the form of new products, I think

14   there are opportunities to find way to make that

15   development process work more efficiently.                Again,

16   this is not something that the agency can solve by

17   itself through just reducing review times and the

18   like.        It's something that will require some

19   creative thinking and efforts to make sure we are

20   applying the best and latest translational science

21   to our regulatory processes, to make sure that we

22   are using the most effective mechanisms for

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         17


1    designing studies, for developing endpoints, for

2    doing follow-up studies after approval and the like,

3    to get to our determinations of safety and

4    effectiveness as efficiently as possible.

5                     So as part of this effort, which is a

6    key element in FDA's strategic plan, we announced a

7    new FDA initiative on improving medical innovation

8    earlier this year.        The goal of this set of efforts

9    is to bring more clarity and consistency to the

10   review process for new and emerging medical

11   technologies, and we're aiming to do that in several

12   ways.

13                    First, as Dr. Feigal mentioned, we're

14   conducting an internal review, a root cause analysis

15   of cases where new products took more than one cycle

16   to reach a determination of safety and

17   effectiveness.

18                    In our preliminary results, it looks

19   like in a lot of cases the multiple cycles were

20   unavoidable.      New things were discovered in

21   development process late, clinical results that were

22   unanticipated, that required some further evaluation

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           18


1    and the like.

2                       But in some cases, it appeared that

3    earlier and clearer communication with product

4    developers about the standards for approval and

5    about what exactly was required for approval would

6    have helped, would have helped them get it right the

7    first time on their product applications.

8                       So a couple of the other major

9    components of this initiative are designed to try to

10   address that issue.

11                      We are also doing a number of guidance

12   development programs like this meeting here today.

13   This discussion is intended to lead to written

14   guidance that can help in the development of

15   products in the area of novel delivery systems for

16   drugs and biologics, and I mentioned some other

17   areas of emerging technology where we are conducting

18   similar kinds of activities.

19                      In addition, we are in the process of

20   implementing some quality systems for product

21   reviews.         We have a lot of expertise in the agency

22   on the best ways to approve and review new products,

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         19


1    and we want to make sure that the best practices in

2    various parts of our agency are shared throughout

3    the agency and are used to implement more efficient

4    regulatory processes.

5                     And this is something that we are

6    undertaking, in part, in conjunction with outside

7    consultants in developing better performance

8    measures, in part through internal work to identify

9    best practices, develop performance measures related

10   to them, and implement them more widely throughout

11   the agency.

12                    So these are all major elements of our

13   effort to improve the innovation process, but I

14   wanted to ask you to take a step back and think more

15   broadly about this.        Most of the time and product

16   development obviously doesn't occur in the review

17   times at the FDA.       Most of the time in product

18   development occurs between the time someone has a

19   good idea in the basic biomedical sciences of a

20   proof of concept and then starts moving that idea

21   into preclinical and then clinical testing.

22                    That's a process that can take many

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         20


1    years and, as I mentioned earlier, can be very

2    costly and have many uncertainties along the way.

3    Anything that we can do through clarifying what our

4    regulatory standards are to make that part of the

5    process work more efficiently as well will only add

6    to these potential savings and reductions in

7    uncertainty in the process of product development.

8                     So it's not just about our review time.

9    It's about clarity in what is needed for determining

10   that a product is safe and effective, and so that's

11   why it's very important to have many of you here

12   today who are involved in product development, who

13   have terrific experience in the regulatory process

14   and who can give us hopefully some insights that can

15   serve as a basis for our written guidance to make

16   this whole process work more efficiently in such

17   emerging areas of technology as novel systems for

18   delivering drugs and biologics.

19                    This is an area where these combination

20   product areas have not gone as smoothly as they

21   might in the past, and we are already taking some

22   steps to try to address that.             One of the first

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         21


1    things that I did as Commissioner was set up a new

2    Office of Combination Products, headed by the very

3    capable Mark Kramer,         in the Office of the

4    Commissioner to provide better oversight and to help

5    develop clear guidance about jurisdictional issues

6    and other issues that are unique to combination

7    products.

8                     One of the other things that I'm working

9    hard on now with Dr. Feigal and the rest of CDRH and

10   our Office of Combination Products is the effective

11   implementation of the new Medical Device User Fee

12   and Modernization Act.          This is a very important

13   piece of legislation that will give us additional

14   resources not only to turn around reviews more

15   quickly, but hopefully to spend some more time and

16   effort on identifying more efficient regulatory

17   practices, to have those kinds of early

18   conversations with product developers that help us

19   by making sure we understand some of the latest

20   technologies that are coming along and how to best

21   evaluate them, and to help product developers by

22   giving them some of our insights in terms of what it

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         22


1    actually takes to demonstrate that a product is safe

2    and effective and meets FDA's regulatory standards.

3                     We are fully committed to the goals of

4    the Medical Device Users Fee Modernization Act, and

5    we will implement this program successfully.                 We're

6    working closely with OMB and others on Capitol Hill

7    to make sure that the adequate funding will be there

8    to meet those program goals, and we're going to

9    succeed.

10                    So this is a very important time in

11   product development for combination products and

12   novel drug and biologic delivery systems for a

13   number of reasons.        We've got new resources.           We

14   have new programs in place already, and we have a

15   strong commitment from the people at CDRH, CBER, and

16   CDER to find more effective ways to implement, to

17   determine that these new technologies coming along

18   are safe and effective, and it couldn't happen at a

19   more critical time with the investment in biomedical

20   R&D in these combination product areas at the

21   highest levels ever and the potential for important

22   new technologies reaching patients.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          23


1                     If we can demonstrate they're safe and

2    effective, the potential is greater than ever.                  So

3    this is a critical time for health policy.               We think

4    that at FDA we're in a great position to help with

5    this innovation process, and we've got more

6    experience and data on the factors that influence

7    success or failure of new treatments than anyone

8    else, and we want to find ways to bring that

9    knowledge to bear and bring some of the new insights

10   in biomedical research to bear in these areas as

11   effectively as possible.

12                    So I am looking forward to hearing from

13   all of you here today and hearing about the results

14   of this conference.        It sounds like a great set of

15   sessions this morning on reviewing some promising

16   clinical applications in the areas of novel delivery

17   systems and some of the preclinical challenges, this

18   afternoon moving on to perspectives from product

19   developers and the FDA on challenges for product

20   development, all with the goal of finding a clear

21   basis for the regulatory processes that we require

22   for demonstrating the products are safe and

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                            24


1    effective, and getting safe and effective products

2    to market as inexpensively and with as little

3    uncertainty as possible.

4                        Thank you all for participating in this

5    effort.          As I said, this is, I hope, going to be an

6    early step in an ongoing effort to make sure that

7    our regulatory processes are up to date and are

8    helping patients get access to safe and effective

9    treatments as quickly as possible and at the lowest

10   possible cost.

11                       And we definitely need this to be a

12   collaborative effort.            We've got a lot of good ideas

13   internally.         We need to bounce them off people

14   outside the agency, and there are also a lot of good

15   ideas outside, given all of the progress that has

16   occurred recently in such areas as novel delivery

17   systems.

18                       So this seems like the right time and

19   the right topic for a kickoff conference on

20   improving innovation process, and I want to thank

21   you all again for coming here today and also for

22   listening to me this morning.

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         25


1                     Thanks very much.

2                     (Applause.)

3                     DR. FEIGAL:     Well, it's my pleasure this

4    morning to introduce our keynote speaker, Professor

5    Robert Langer.      Again, thanks are in order for the

6    service that he provided by chairing the Science

7    Board, and part of that time period was when the

8    center itself actually went through an external

9    review of our science program.             So we appreciated

10   his efforts in that very much.

11                    Dr. Langer is the Kenneth J.

12   Germeshausen, Professor of Chemical and Biomedical

13   Engineering at MIT and is a member of the National

14   Academy of Engineering, National Academy of

15   Sciences, and the Institute of Medicine, one of the

16   few people to hold memberships in all three of those

17   academies.

18                    The kinds and nature of the

19   contributions that Professor Langer have made are

20   particularly relevant to our program here today, and

21   without any further ado, let me ask Dr. Langer to

22   come and begin.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         26


1                     (Applause.)

2                     DR. LANGER:     David and Mark, thank you

3    very much.       It's an honor for me to be able to speak

4    to you all today, and it was certainly an honor for

5    me to work with the FDA as well.

6                     I've had the fortune over the years of

7    giving a lot of talks.          Usually I end up giving them

8    at universities, though I've given them at

9    companies, too.       And usually what I talk about are

10   drug delivery systems.

11                    A few years ago, though, I was giving a

12   talk at the University of California at Berkeley.

13   So it was the other side of the country, and I got

14   in very late at night, and I was trying to think how

15   to introduce my talk.         And I thought for a second

16   and I said, "Well, probably everyone here has taken

17   drugs."

18                    (Laughter.)

19                    DR. LANGER:     That's what they did, too.

20   They laughed, but of course, what I meant were drugs

21   like all the ones that are regulated by FDA, and

22   that is what I want to talk about today.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          27


1                     In particular, what I will try to do

2    this morning is to give you an overview, and

3    obviously it can't be complete, but what I'll try to

4    do is go over a little bit about why drug delivery

5    is important, where it is in terms of some of the

6    products, and where it's going.

7                     Let me start with a slide.           I just want

8    to make sure.      Do I do something to get this on?

9    I'll try to tell some more jokes in the meantime.

10                    (Pause in proceedings.)

11                    DR. LANGER:     So anyhow, I think I can do

12   this almost without slides, but at least the first

13   couple of ones.

14                    Thank you very much, Mary.

15                    So what I was going to say, and people

16   obviously have probably seen things like this

17   before, but if you take a drug, really any drug and

18   really by almost any means, mouth, skin, whatever,

19   the drug level starts out very low, reaches a peak,

20   and then goes down, and that peak and valley level,

21   the problem is those peaks can cause huge safety

22   problems, sometimes death, and the valleys, the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                           28


1    drugs, are not effective.

2                      One example I sometimes use in class

3    are, you know, sleeping pills.              If you take too much

4    you could die.        If you take too little, you don't go

5    to sleep.        I mean, there's various ones you could

6    think about.

7                      So that provides the motivation for

8    could you come up with a way -- and this is not

9    always what you want to do, but for a lot of cases

10   what you'd like to be able to do is take a drug and

11   have it go to the desired range and stay there for

12   as long as possible.

13                     Let me just give you a striking example

14   of that that ALZA did working with Pfizer.                So they

15   had a drug that was called Nifedipine, which also is

16   known as Procardia, a calcium channel blocker, and

17   all throughout the 1980s it was taken by a soft

18   gelatin capsule, so sort of immediate release.                   It

19   was quite a successful product.              It sold about $300

20   million a year.

21                     It was always, though, if you took it

22   and got the soft gelatin capsule, peaks and valleys

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          29


1    just like you saw.

2                     ALZA, using an osmotic pump system, and

3    I'll mention that a little bit more later, figured

4    out a way to get it at pretty constant release.

5    That product actually became very successful.                  Not

6    only was it used for angina.            It also got a new

7    approval for congestive heart failure.

8                     What happened if you looked at the side

9    effects are huge.       Here you can just look at the

10   comparison of the two, and if you compare things

11   like headaches or flushing or dizziness or

12   palpitations, there's a huge difference.

13                    Taking the controlled release form you

14   get many, many fewer side effects than the soft

15   gelatin capsule form, and from a cost standpoint,

16   from the company's standpoint, it became a $1.5

17   billion a year product rather quickly.

18                    Let me mention a few products to just

19   give you an idea, though I imagine people are aware

20   of this, of the range of things that are already

21   being used in this actually very young field.                 I

22   think if you look at this, almost all controlled

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                             30


1    release systems at least that I'll be talking about,

2    will be approved in the last 20 years or the last 21

3    years.           This was about one of the earliest ones.

4                         It's the nitroglycerine patch, one of a

5    number of transdermal systems that can deliver drugs

6    just passively in this case through the skin.                     Here

7    it does it for a 24-hour period.                  Over 500 million

8    of these were used last year.

9                         This is the longest.         Sometimes people

10   ask me how long can a controlled drug delivery

11   system go.           Well, this is the longest one that I

12   know of.           This is the Norplant.        These are little

13   silicone capsules that you can place underneath the

14   skin for contraception.              They're approved in over 50

15   countries.

16                        And what you can see here is these

17   capsules, which are simply the size of match sticks,

18   are able to release the drug for over 2,000 days or

19   five years from these tiny little implants.

20                        This is the very first controlled

21   release system for a protein.                 For many years people

22   didn't think you could ever deliver proteins.

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         31


1    Alcames, which is a company I've been associated

2    with for a number of years, developed along with

3    Genentec tiny little microcapsules that you could

4    put human growth hormone in.

5                     Normally a patient with pituitary

6    dwarfism would take the shots once a day.              Now with

7    this you could take them once a month.

8                     And another one, the last that I'll

9    mention at least right now is really a very

10   innovative thing that ALZA did for Ritalin.                If any

11   of you have children that have attention hyper

12   deficit disorder, they may take this.

13                    Normally what people had to do was take

14   Ritalin, you know, several times a day, and if

15   you're a small child that means you might have to go

16   to the nurse.      It might be embarrassing, and maybe

17   it doesn't even work as well in the regular forms.

18                    ALZA discovered that actually you don't

19   only want to get steady release.              You actually want

20   to have a time where the release is increasing, and

21   because they were able to design a special version

22   of an osmotic pump shown here where they've got an

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         32


1    overcoat, they're actually able to do this.

2                     So they developed a system called

3    Concerta based on an osmotic pill that you could

4    take, a child could take generally once a day to

5    treat ADHD.

6                     Let me just give you some statistics,

7    again, for part of this overview that adverse drug

8    effects where people take drugs kind of the way

9    they're supposed to, they can cause up to 15 percent

10   of hospital admissions.          This was in JAMA a few

11   years ago.       One hundred thousand deaths; that's more

12   than four times the number of deaths caused by AIDS

13   in this country, $136 billion in health care costs.

14                    Patient compliance, that can cause up to

15   ten percent of hospital admissions, particularly in

16   the elderly that forget to take drugs.

17                    And of course, one of the things that

18   motivates a lot of companies, which it should, is

19   can you make a profit in this area.               And if you look

20   at this, as I mentioned, controlled drug delivery

21   systems in the 1980s, the sales were about zero.                    In

22   2001, they're about $20 billion, and my expectation

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         33


1    is that number will go up rapidly for a single

2    reason that I'll mention later and will be talked

3    about later today.

4                     Just look at drug eluting stents,

5    totally based on controlled release technology.

6    Sales are projected to be five to seven billion

7    dollars rather quickly.          So there's enormous

8    opportunity in this area, as well.

9                     The advantages of controlled drug

10   delivery are reduction of adverse side effects,

11   which I've mentioned.         You can keep drug levels in

12   the desirable range, and much less drug is desired.

13                    You get improved patient compliance, and

14   as we will go over all day, new therapies are

15   possible.

16                    People -- I just want to grab myself

17   some water -- people, you know, you can approach

18   drug delivery from a number of standpoints.                One

19   standpoint is pretty much every part of the body --

20   this may be a little hard to see in the back -- but

21   I think that sometimes people ask me, "Will there

22   ever be an ideal delivery system that you could just

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         34


1    take by one way?"       And I think the answer is no.

2                     People have been successful at

3    delivering drugs orally, nasally, transdermally,

4    through the lung, transmucosally, like vaginal,

5    buccal, in the eye, by liposomes, by injection.                   All

6    of these, almost all of them are multi-billion

7    dollar markets in and of themselves.

8                     So there have been successful products

9    in almost all of these areas, and I expect that that

10   will continue because there's enormous opportunity

11   in each of these areas.          There's specific diseases

12   in some of those areas, and many of these areas can

13   be a portal to the rest of the body for delivering

14   drugs.

15                    I thought I would try to focus in the

16   interest of some of the goals of this meeting on

17   four areas:       the need for new materials;

18   nanotechnologies; noninvasive delivery; and high

19   throughput approaches.

20                    So I'll focus on each of these just in

21   context to illustrate what I see as some of the

22   ongoing work and some of the challenges ahead.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          35


1                      First, let me go over new materials.

2    You know, this is something that I got involved in

3    personally in the 1970s.           I was actually very

4    surprised to see this.           My own background is a

5    chemical engineer, but when I got done with my

6    degree I worked at Boston Children's Hospital, and

7    being a chemical engineer, I guess I just thought

8    naively that the people who were driving the work

9    for bringing new materials into medicine would have

10   been older chemical engineers or chemists or

11   material scientists.

12                     But when I looked into this, I found

13   that it was rarely the case.             Almost always the

14   driving force for bringing materials into medicine

15   were clinicians, and they wanted to solve a problem

16   and solve it as quickly as they could, which is

17   good.

18                     But what they would do is generally they

19   would take a material that was usually in their

20   house and that kind of resembled the organ or tissue

21   they wanted to fix, and they'd use it in the human

22   body.        And that led to some progress, but also to

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                             36


1    some problems.

2                         And just to give you some examples, this

3    may be cut off a little bit, but it's an artificial

4    heart.           But you probably figured that out.          Anyhow,

5    let me just tell you a story or two, and these are

6    all true.

7                         In 1967, clinicians at the NIH wanted to

8    come with an artificial heart, and they wanted

9    something with a good flex life, you know, for a

10   heart.           And they said, "What object has a good flex

11   life?"

12                        And they said a lady's girdle material.

13   What's that made out of?              It's made out of

14   polyetherurethane.            So that's what they began to

15   make the artificial heart out of.                  That was 1967.

16                        Now we're in 2003.        It's still made of

17   that, and you can imagine from a regulatory

18   standpoint once you start going down that path it's

19   not so easy to stop.             And that happens in many

20   different areas.           Dialysis tubing was originally

21   sausage casing.           Vascular graft, that's an

22   artificial blood vessel.              It was a surgeon in Texas

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           37


1    going to clothes store, and breast implants.                   One of

2    those was a lubricant -- oh, thank you -- one was a

3    lubricant.        The other was a mattress stuffing.

4    Probably you figured out the logic.

5                       But these are all true, and that's often

6    how materials have come into medicine, and I started

7    thinking in the '70s, well, you know, maybe you

8    could take a different approach, and I believe we're

9    going to start to see more and more of that today.

10                      And that is rather than take these

11   materials that might exist in your house, could you

12   actually ask the question what do you really want in

13   a biomedical material or drug delivery system from a

14   chemistry standpoint, biology standpoint and

15   engineering standpoint, and could you synthesize it

16   from first principles.

17                      I thought I'd give you an example.

18                      At any rate, let me give you that

19   example.         When we started in the 1970s, there was

20   only one material approved by the FDA that was

21   synthetic degradable material, suture materials like

22   polyesters, and they displayed bulk erosion kind of

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                            38


1    like this.       So it would start out -- now this isn't

2    working either.       That's okay.        I'll go over here.

3    I'll get some exercise this morning.                  Oh, but I

4    think -- is that going to be a problem?

5                     (Laughter.)

6                     DR. LANGER:     Maybe we had better get one

7    that works.      Anyhow, so it might be good if we got

8    one that works.       At any rate, I'll try to do this

9    with my pointing.        It might be harder on some of the

10   slides.

11                    Thank you very, very much.             Will you

12   remind me to give it back up?             Okay.       I got it.

13   Thank you very much.

14                    Okay.   So bulk erosion, usually you put

15   the drug uniformly throughout.             It starts out

16   getting spongy, and then it could fall apart.                    That,

17   by the way, is fine for a lot of drugs, but if you

18   had a really toxic drug like, say, insulin or a

19   cancer drug, it might not be so good because you

20   could get bursts of the drug coming out.

21                    So we said from an engineering

22   standpoint what you'd really like is this:                  surface

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                          39


1    erosion, kind of like the way a bar of soap

2    dissolves.       So the challenge is how could you do it.

3                     So I won't go through all of the

4    chemistry, but basically what we did is we took this

5    from an engineering design standpoint.                We said,

6    well, what are the right bonds.             We thought

7    anhydride bonds.       We thought what are the right

8    monomers, and we came up with a couple of monomers,

9    very hydrophobic ones that could keep water out.

10   This is extremely hydrophobic CPPP, and sebacic acid

11   is a little less so.

12                    What was interesting is that by simply

13   adjusting the ratio of those two not only could you

14   get surface erosion, but you could get these to

15   dissolve at almost any rate you want, from zero

16   percent sebacic acid.         So about eight percent is

17   gone in 14 weeks.       It will take three or four years

18   for one of these to dissolve fully.

19                    But if you add a little bit more sebacic

20   acid, it dissolves faster- that's 15 percent,                   55

21   percent, it dissolves faster, 79 percent it's all

22   gone in two weeks.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                         40


1                     So you could simply dial in your monomer

2    ratio and make these last for whatever length of

3    time you want.

4                     So with that, you could think about

5    using it for all kinds of applications.               One of the

6    early applications that came up was Henry Brown, a

7    young neurosurgeon -- he's now head of neurosurgery,

8    chief of neurosurgery at Johns Hopkins -- came to

9    see me in the 1980s, mid-1980s, and he said, "Could

10   we change the way people do chemotherapy with this

11   kind of approach?       Could we do local chemotherapy?"

12                    So here was the idea.         He would normally

13   go in, operate on patients, take as much of the

14   tumor out as he could in the brain.               He would always

15   do this, as would everybody else.              But he said is

16   after that, you       know, they have to give this drug,

17   BCNU, intravenously.         Could we do local

18   chemotherapy?      This drug is enormously toxic.

19                    So the idea was could you take polymers

20   like this, allow him as a neurosurgeon to put the

21   drug in in little wafers that would locally deliver

22   it to any remaining tumor he couldn't get.               The idea

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                             41


1    is that could tremendously spare the body the side

2    effects of this terribly toxic drug, but give high

3    concentrations right to the brain tumor where you

4    want it to be.

5                         So let me just show you that.          If

6    anybody is squeamish and doesn't like to look at

7    blood -- and I'm serious about this -- don't look,

8    but here is what it looks like, a little wafer the

9    size of a dime going in.              Usually you put seven or

10   eight and then close it up.

11                        I always show those slides rather

12   quickly.          You know, it's very hard to get good

13   advice when you give a talk, but a few years ago my

14   wife Laura came to one of my talks.                   She's a

15   neuroscientist, and I asked her at the end.                     I

16   actually was showing those slides.

17                        I said, "What did you think of the

18   talk?"

19                        And she said, "Well, Bob, the talk was

20   okay."           That's actually very high praise.

21                        (Laughter.)

22                        DR. LANGER:     But she said, "You know,

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                               42


1    there was this 12-minute period of that talk where

2    you had those two bloody slides on and you explained

3    every detail of it to the audience."                     This was all

4    chemical engineers I was speaking to, and she said,

5    "I don't know if you were looking, but they were all

6    turning green and looking at the floor."

7                        So ever after that I've done just what I

8    did today, showed them real quickly and I warn

9    people.          But I do want to tell you a sequel to that.

10    I give talks to lots of different groups, and I

11   happened to be giving a dinner speech to a group of

12   neurosurgeons and neurologists, all M.D.s, and I did

13   the same thing, and at the end of the talk a number

14   of the neurosurgeons came up to me and they said,

15   "You know those two bloody slides you showed?"

16                       I said, "Yes."

17                       They said, "Those were fine.             No

18   problem," but they said, "Those chemical formulas."

19                       (Laughter.)

20                       DR. LANGER:     You know, right after

21   dinner.          So you have to be very careful who you

22   speak to.

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                          43


1                      (Laughter.)

2                      DR. LANGER:     At any rate, there are a

3    lot of challenges -- I'm just going to go back a

4    slide -- that we had to overcome, which are typical

5    I think in any of these areas, and I'll just go over

6    those briefly.       These were all actually things at

7    the National Institutes of Health Study Sections,

8    other professors told us why we couldn't get it to

9    work, but basically it's just a synthesis,

10   reactivity, strength of the material, toxicity,

11   diffusion of the drug, manufacturing, and so forth.

12                     All of these were challenges that had to

13   be overcome.       Actually they made for a lot of good

14   theses in our lab.         Later on we licensed to do a

15   company, Guilford Pharmaceutical, and actually the

16   FDA did approve this originally in 1996.               It was the

17   first time that a local chemotherapy system got

18   approved.        That was approved in 1996 for recurrent

19   glioblastoma.       It was extended this year for primary

20   glioblastoma, but it's an example which I wanted to

21   pick of how you could use new materials to create

22   new therapies in drug delivery.

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         44


1                     Also, it illustrates a very early

2    example of local chemotherapy, which I think is very

3    powerful, and of course, I think the most powerful

4    example of that which you'll hear more about later

5    today is applying this idea to coated stents where

6    basically you put stents in to keep blood vessels

7    open, but the problem is, as people will hear about,

8    that for a fairly high percentage of the time those

9    vessels will close due to restenosis, smooth muscle

10   proliferation, and so forth.

11                    But you can take some pretty toxic drugs

12   like Taxol and repromicin or others, put them on a

13   tiny polymer film, locally deliver them, and the

14   results have been very, very dramatic by many

15   companies in terms of keeping these blood vessels

16   open.

17                    That's the first topic that I wanted to

18   mention, is this idea then of new materials and

19   local delivery.

20                    The second is nanotechnology.

21   Nanotechnology is something I'm sure everybody reads

22   about in the newspapers.          Probably everybody wonders

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         45


1    what it really is.        Even I wonder what it is

2    sometimes because it has so many definitions.

3                     But I think there's several ways

4    nanotechnology can make a huge impact, and I thought

5    I'd give you a couple of examples.

6                     The first actually is work that was done

7    originally by John Santini when he was my graduate

8    student at MIT and now president of MicroCHIPS,

9    which is a company I'm also affiliated with.                 I had

10   this idea about ten years ago.             I was watching this

11   TV show on how microchips are made in the computer

12   industry, like Intel, and I thought, gee, this would

13   be a very interesting way of doing drug delivery.

14                    So along with Michael Sema and John, we

15   came up with originally a very early design, which

16   I'm showing here, and the idea is rather than take a

17   chip for your television set or your computer, what

18   you could do is build little nanowells -- I'll show

19   you these in a minute -- into little chips.

20   Originally they were made of silicon and covered

21   with gold.

22                    They can, by the way -- some of our more

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          46


1    recent students have made them out of polymers.

2    They can be made out of almost anything, but

3    basically you can build little wells into them.

4    This is a cut-away.        So on one side there's an

5    impermeable epoxy, and here we're using gold as sort

6    of the cover.      They're hermetically sealed.             You can

7    actually keep them on the shelf or in the body for a

8    couple of years.       Nothing will happen.

9                     But if you apply selectively one volt to

10   any of these welds, they're all individually

11   addressable.      What will happen is the gold will come

12   off, and the drug will come right out.                And you can

13   program these to get almost any delivery pattern you

14   want because if you want to get instantaneous

15   release, well, then you just have the gold come off.

16                    But let's say you wanted the release to

17   be more slow.      You could put a polymer or a gel

18   right underneath it.         Also, you could deliver one

19   drug multiple times.         I'll show you an example of

20   that in a Pulsatile fashion, but if you wanted to

21   actually delivery many drugs, like say we've often

22   considered this like a pharmacy on a chip, you could

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                             47


1    do that.           If the drugs are potent enough, you could

2    put all of the drugs you want on such a chip.

3                         And you can make them very tiny like

4    nano, which is what I'm talking about, but you could

5    make them bigger, too, if you had drugs that were

6    less potent.

7                         Let me actually show you what they look

8    like in a picture.            This is from MicroCHIPS, and

9    this is a real good example of nanotechnology.

10   These are pencils, and here's the chip.                   Here's one

11   side and here's the other.               There's hundreds of

12   wells.           Each of these contains a different drug or

13   the same drug at a different dose.

14                        And then what's done to use these,

15   they're battery powered, and you could control them

16   by telemetry.           In other words, the same way you

17   might open up a garage door, you could open any of

18   these wells.           You could envision a day when you

19   might have a wristwatch or some unit like that that

20   you could just do remote control, and you could open

21   up any individual well whenever you want.                  And then

22   it's like encased in something like a little

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                            48


1    pacemaker.

2                     Also what I believe we'll see in the

3    further future is even very smart systems where you

4    could put biocensors -- I think Dr. Klonoff may talk

5    about this more -- where you could put biocensors on

6    these chips along with a microprocessor and a power

7    source, and you could get direct control.

8                     Let me actually show you how this works.

9     What I'm going to show you is a quick video.                    You

10   have to look quickly, but I'm going to just show you

11   a single well where you're going to be looking at

12   the top of the well, and then we're just going to

13   apply this one volt selectively, and what you're

14   going to see is the top dissolve, and then you'll

15   see a little conical bottom, and so let's take a

16   look at that.

17                    Here's the video.        This is the top.

18   Immediately the gold came right off.                  As soon as it

19   does, the drug can come out.            So basically it's just

20   that quick.      It can be made instantaneous.

21                    Here's an in vivo.        This is done in

22   animals, and this just shows you you can get very

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                             49


1    reproducible Pulsatile release of the single drug,

2    and you could do this with multiple drugs if you

3    wanted to.

4                         So this is, I think, one example of

5    nanotechnology where you have little nanowells, and

6    also a good example of another thing that people

7    sometimes hear, MEMS devices, micro electrical

8    mechanical system.

9                         The second example of nanotechnology

10   that I wanted to do is at a                more molecular level.

11   There are a whole range of different polymer

12   therapeutics that you might think about as nanosized

13   medicines.           Some of these are things like polymer

14   drugs.           You could actually take a drug, and I'll

15   give you examples in a minute, but I just wanted to

16   show you the range and also the sizes so that you

17   could have polymers combined to drugs which could

18   change the drug's properties.

19                        You can have polymer protein conjugates.

20    A very good example of this which we hear a lot

21   about are PEGylated, and I'll mention this more.

22   You could take a protein, which might normally have

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                            50


1    a short half life or might be immunogenic, and yet

2    you can conjugate polyethylene glycol to it, and

3    change both of those.

4                       Another example are polymers for DNA

5    delivery, and I'll talk about this more, but it's a

6    huge area because if you wanted to deliver DNA,

7    right now the only way to do it is with viral

8    vectors, and I think everybody here at the FDA and

9    elsewhere have made it pretty clear there are some

10   huge safety issues with it.

11                      Could you make nanosized polymers -- and

12   I'll mention a method later -- where you might be

13   able to make a polymer behave like a virus, but

14   without the safety problems associated with it?

15                      Polymer drug conjugates where you could

16   actually target the drug to a particular place in

17   the body, like, say, a tumor.

18                      And finally, micelles.

19                      So all of these things are being

20   studied.         I'll just go over one or two a little bit,

21   but some of them are already approved.                  The first

22   one is the drug attached to a polymer that's been

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                         51


1    approved for liver cancer in Japan, and then there

2    are a variety of pegylated molecules like

3    asparaginase, interferon on CGSF, which have been

4    approved by the FDA for different diseases.

5                     Now, these have made a very substantial

6    impact on medicine already, but there's more coming,

7    and some of the things that are coming are to really

8    engineer drugs with polymers to get very desired

9    properties.      So maybe you could put in a targeting

10   residue so that it would target to a receptor.

11   Maybe you would have a biodegradable linker, and

12   people are already studying these for cancer.

13                    Apart of the basis for treating cancer

14   is that a tumor has very leaky blood vessels.                 So

15   you might imagine a polymer, since it's big.                 Well,

16   if the blood vessel is not leaky, it's not going to

17   get out.

18                    But when it gets in the leaky region of

19   the polymer, it will get out, and that's called the

20   EPR effect.      And there are a variety of these

21   systems in various stages of clinical trials for

22   delivering drugs like Taxol or camptothecin, and

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                            52


1    some of them even have targeting moieties on them,

2    like lactosamine.

3                     So many, many that probably will keep

4    the FDA busy over the next number of years.

5                     The third thing that I wanted to go over

6    is the idea of noninvasive delivery of complex

7    molecules.       Can you do it orally?         This would be

8    like a protein or DNA.

9                     Can you do it transdermally?             And I'll

10   talk a little bit about both of these.

11                    Pulmonary, actually somebody from Nektar

12   is going to be speaking.          So I thought I'd let them

13   do that, and there's also various other routes that

14   we might consider, but I'm going to just focus a

15   little bit on these two today where there has been

16   probably a lot of work done, as has there been

17   there.

18                    So first, oral.       There has been at least

19   four strategies that have been used:                  carriers, and

20   I'll give an example of those; nanoparticles that

21   might be able to       be taken up by M cells, for

22   example in the Pyrus patch (phonetic); targeting to

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                          53


1    various receptors in the gut or Pyrus patches.

2                      And even bioadhesive approaches are

3    being worked on by a variety of groups.                I was just

4    going to mention one briefly, which is the carrier

5    approach, which is probably furthest along, and

6    David Klonoff may talk about this more, too, but

7    there's a company, Emispheres, which has synthesized

8    some molecules which they call eligens, and their

9    idea is to complex them to drugs, to deliver them

10   transcellularly without compromising cell integrity.

11                     And you see an example of this top panel

12   here where insulin is being delivered to a cell

13   monolayer at different concentrations, and you see

14   it getting through, and yet the lower panel seems to

15   show that it doesn't affect the tight junctions and

16   so forth.

17                     They've used this to deliver insulin in

18   patients.        This is oral insulin, and they have also

19   used it to deliver human growth hormone.

20                     Probably the big issue that will come up

21   here for oral delivery with any of these things are

22   two or three things.          One is bioavailability.          Do

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         54


1    you get enough in?        And two is safety, or do other

2    things also get transported?            And these will be some

3    of the key issues to look at.

4                     Nonetheless, I think it's very          exciting

5    that we're already seeing a delivery of large

6    molecules in people.

7                     The second area that I want to talk

8    about is the skin, and the skin has been a normally

9    incredibly impenetrable barrier though right now

10   there are ten transdermal products on the market,

11   all by passive delivery, a $3 billion market, and

12   some of them are huge, like Fentanyl patches, which

13   are for pain management.          People, I'm sure, are

14   familiar with smoking cessation, and so forth.

15                    Well, why is it so hard to get drugs

16   through the skin?       Well, first, I think we should

17   all be very glad that it's hard because otherwise

18   we'd get infected.

19                    But what makes it hard is all of the

20   resistance to the skin is the outermost part of it,

21   the stratum corneum.         It's only 15 cells layers

22   thick, and it looks like a brick wall if you looked

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          55


1    at it under the microscope.             You have deal cells,

2    keratinocytes, and lipid bilayers here.                So it's

3    like bricks and mortar, and it provides a

4    terrifically tight barrier to get through.

5                      That being said, a variety of groups are

6    trying to do this.         Vyteris is a company, a spinoff

7    of Becton-Dickinson, and they've developed an

8    approach using iontophoresis.              This is using an

9    electric field basically to transport drugs across

10   the skin.        This is their system, and they've

11   actually developed pretty advanced approaches.                   They

12   are actually delivering in this case calcitonin in

13   people this way.

14                     ALZA, which has certainly been a leader

15   in transdermal drug delivery, has developed what we

16   call the E-TRANS system, and here they're putting

17   Fentanyl, the drug I mentioned, which they've worked

18   on with Johnson & Johnson, which has been very, very

19   successful, and here they're putting it on the skin,

20   and it slowly delivers it, but as I'll show you, by

21   applying the electric field, and this is sort of

22   what it looks like; they've got an on demand button,

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           56


1    a red light diode.          This is kind of the structure of

2    it.

3                       But what you can do is if you want to

4    get a pulse of Fentanyl, feel more pain, you can

5    just do this rather quickly.              So this can actually

6    provide Pulsatile Fentanyl delivery using, again,

7    iontophoresis.

8                       And here we're looking at them

9    delivering another peptide.              This is luteinizing

10   hormone releasing hormone in a Pulsatile fashion.

11   This is a drug that people might use for fertility

12   control.         It has also been used for other treatments

13   as well.

14                      ALZA has also developed a kind of micro

15   needle approach.         As I mentioned, really if you want

16   to deliver a drug through the skin, if you could get

17   through relatively painlessly that thin stratum

18   corneum, you could deliver a drug.                 So they've

19   developed what's called the macroflux transdermal

20   patch, which has these little protrusions.                 You can

21   put it in the skin, and here they've done like an

22   example of delivering what are called anti-sensal

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           57


1    nucleotides (phonetic).            These are about 7,000

2    molecular weight through the skin this way.                  So

3    that's another approach.

4                       And the final transdermal approach that

5    I wanted to mention is one that I've been involved

6    in.      You'll see Coats in our laboratory and then

7    Santra Company that we've been involved in that has

8    developed what's called the SonoPrep system, and the

9    idea is you can take ultrasound, and this probably

10   gets an order of magnitude more increase in flux

11   than you get with iontophoresis, but it's at an

12   earlier stage.

13                      But the idea is that you can apply the

14   ultrasound maybe for about 15 seconds, and that will

15   permeabilize the skin.            You could then put a patch

16   on, and that patch, for example, could deliver

17   insulin.         Here's an example.       You could actually

18   lower blood sugar.          It's being tested on man.           You

19   could deliver pain medications, and also I think

20   what's particularly exciting is you could do

21   noninvasive extraction.

22                      Both Signas, a company in California,

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         58


1    and Santra have been developing noninvasive ways.

2    Say if you open up the barrier for delivery, you

3    could also open it up for getting interstitial fluid

4    out, and so you might be able to, for example, which

5    is in clinical trials now, detect glucose or many,

6    many other different substances.

7                     And again, I think we will see all

8    examples of this someday

9                     So the final thing that I wanted to go

10   over today before summarizing is could we look at

11   high throughput approaches, and I thought I'd give

12   you two examples of that.

13                    First, what I mean by high throughput

14   approaches.      If I looked at the pharmaceutical

15   industry and I don't mean to be insulting by this,

16   but from the formulation standpoint, it's sometimes

17   a little bit slow.        I mean, basically people make

18   the drug, and then you have a formulation

19   pharmaceutical R&D department, and they have to

20   formulate it.

21                    It's a huge, huge challenge as everybody

22   knows, and I thought I'd just pick a couple of

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           59


1    examples where you could maybe use high throughput

2    approaches to make these faster.              One of them is

3    gene therapy.      As I've already mentioned, gene

4    therapy is an area which I think is badly in need of

5    better delivery systems.

6                     In fact, it's interesting.           If you talk

7    to the gene therapy experts, like Endar Verma

8    (phonetic), for example, I remember he was quoted a

9    few years ago.      Somebody asked him what are the

10   three biggest problems in gene therapy.                In other

11   words, why is gene therapy not being used in

12   patients today.

13                    And he said, "Well, there are three big

14   problems," he said, "delivery, delivery, and

15   delivery."

16                    And that I think is true.            It has been a

17   huge, huge problem, and it is unsolved.                Richard

18   Mulligan, whom I work closely with, another gene

19   therapy expert, the same thing.             How could se solve

20   that?

21                    Well, one strategy might be to come up

22   with better viral vectors that would be safer.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                           60


1    That's a strategy.          Another might be nonviral

2    vectors.         Could you make polymers that could behave

3    like a virus, but without the safety problems of a

4    virus?

5                       And I won't go through that much of the

6    science.         I just want to illustrate the high

7    throughput idea, but David Lynn, one of the graduate

8    students in my lab -- he's now a professor at

9    Wisconsin -- came up with the idea of synthesizing

10   certain polymers that would, in fact, have many

11   properties of like a virus.              They would be taken up

12   by cells and so forth.

13                      But what was particularly interesting is

14   he came up with a synthesis approach where he could

15   make what are called polybeta amino esters, where he

16   would simply take amines like this and conjugate

17   them to diacrolates (phonetic) like this, and the

18   idea is that with this particular set of chemical

19   structures, he could take a whole range of

20   commercially available starting materials.                 He could

21   polymerize them in a single step.                Many

22   polymerizations, by the way, are done in many, many

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         61


1    steps, but the particular beauty of this is there's

2    no byproduct.      So no purification and no protection

3    and de-production steps.

4                     So you could actually set this up for

5    high throughput.       So Dan Anderson, another postdoc

6    of mine -- this slide is a little busy, but it will

7    get across the point -- basically what he did is he

8    took 909 amino monomers.          Those are shown here.

9    Twenty-five diacrolate (phonetic) monomers, and he

10   used a robot and developed high throughput synthesis

11   methods and then also high throughput screening

12   methods using cells, and what he was able to do in

13   really a couple of weeks is make over 2,000

14   structurally diverse polymers, like a whole polymer

15   library.

16                    And here's the robot, and he could take

17   these polymers, semi-automate it, cell based, and

18   screen up all of these polymers, 1,000 in a day, and

19   he found 46 new polymers.           This is just coming out

20   in publication very soon.           Forty-six new polymers

21   that could at least themselves deliver DNA as good

22   or better as polyethylene amine, which is one of the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         62


1    standard polymer vectors.

2                     Again, it's just an illustration, I

3    mean, of where high throughput could make an impact,

4    but I think there will be many other examples as

5    people think about this.

6                     And the final example that I wanted to

7    pick is a company that I've been associated with

8    called Transform Pharmaceuticals, and here the idea

9    is could you apply high throughput approaches and

10   robotics and bioinformatics to pharmaceutical

11   formulations.

12                    And Colin Gardner, who used to be VP of

13   R&D at Merck, Sharpe & Dome, and now is chief

14   scientific officer at Transform, has done this, but

15   basically if you look at what we'll call form, which

16   might be crystal structure, which is a real big deal

17   in drug delivery and formulation.              Traditionally,

18   you might take a month or two and there's almost no

19   informatics or data mining done.              This is classical.

20                    What Transform has done, and they're

21   working with Johnson & Johnson, Eli Lilly, many

22   other pharmaceutical companies, but basically

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         63


1    they're able to because of the use of robotics

2    approaches, and I'll show you this in a minute or

3    two, are able to do 200 to 20,000 experiments in two

4    to four weeks, and every time they do an experiment,

5    they use all of this informatics to tell them how to

6    do the next set of experiments.             So you can be

7    faster and faster and smarter and smarter, and I

8    think this is going to be the kind of way that will

9    make sense more and more.

10                    And let me illustrate that a little bit,

11   and what I'll also do is show you a video to just

12   give you a little feel for the high throughput idea

13   that they're doing.

14                    And in particular, what I want to show

15   you is how they're analyzed.            In other words,

16   they're using robotics to make these things, but you

17   also have to do analysis.           Let's say you want to

18   make a new crystal structure.             How do you know that

19   it's a new crystal structure or an old crystal

20   structure?

21                    So what they've done is they've used

22   Raman Microscopy to analyze these, and I just

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          64


1    thought I'd show you this video rather quickly.

2                     So the idea -- if I can do this -- so

3    here you're looking at the robot.              It just takes the

4    vial after all of these thousands and thousands of

5    things, puts it into -- it's all done robotically,

6    no people -- puts it in here, puts it under the

7    microscope.      Here's the crystal.

8                     Now this is real time, looking at Raman

9    spectrographs.      Then it takes all of the data like

10   this, mines them, and you see three crystal

11   structures.      So all done real time, and this, I

12   expect, is going to be more and more what you could

13   see in the future.

14                    Well, does it make a difference?             Let me

15   just show you a couple of key publications that

16   Transform has just done.          They took acetaminophen.

17   I think people are familiar with that.                That's

18   Tylenol, and they, by using this approach because

19   you could do it so much faster, so much better, came

20   out with the new crystal structure for Tylenol.

21                    They did 10,000 experiments in three

22   iterations in six weeks, published in JACS last

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                            65


1    year.        So that's one example.

2                      An even more striking example is this

3    one, which certainly the FDA probably would be

4    familiar with, Ritonavir.            This was an Abbott drug.

5    Here's what happened in initially.                Abbott launched

6    this drug for AIDS.         It was originally in crystal

7    form I, but 1.5 years after the launch, it converted

8    it into an unanticipated crystal form.                  Sometimes

9    this happens.       Form II, but that form was 50 percent

10   less soluble.

11                     So Abbott obviously was compelled to

12   recall and reformulate it.            It cost them hundreds of

13   millions of dollars and so forth.

14                     Using the high throughput approach, what

15   did Transform do in just a few weeks?                  Well, even

16   though Abbott could never get Form I back,

17   Transform, since they were able to do tens of

18   thousands of these, got Form I back, got Form II

19   obviously, also found three new forms that were

20   never found before.         So characterized them, you

21   know, and made them.

22                     This was just published in PNAS just a

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                         66


1    couple of months ago.         Again, an illustration of

2    what high throughput might be able to do for you.

3                     So let me end the talk by telling you

4    some even further future challenges about what I

5    expect, and I've been trying to keep on time.                  So I

6    think some of the challenges which I've begun to

7    touch on, they'll increase.            One of the tremendously

8    exciting things in the times we live in are all of

9    the new pharmaceuticals that are being made.

10                    Certainly we've had protein therapeutics

11   for years, and yet there are still very serious

12   delivery challenges.         Today there's still only, you

13   know, a few examples of the pegylated one and the

14   neutropin depo where you can deliver proteins, and I

15   think that the opportunity for delivering proteins

16   is just huge, whether you could do it by controlled

17   release to make it last longer or noninvasively.

18   That will be very big.

19                    DNA, obviously we've talked about that.

20   Delivery is probably the central problem, and so

21   there needs to be a lot of -- that's a big

22   challenge.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         67


1                     And there's even newer things that are

2    coming out.      People may be familiar with RNA

3    interference.      These are really incredibly exciting

4    molecules.       Our Science magazine called them the

5    molecule of the year last year.             These are very

6    powerful, very specific units, about 20 base pairs

7    of RNA that can interfere with cell function.

8                     Another big area, and this was mentioned

9    actually in Mark McClellan's introduction, is the

10   delivery of cells.        Could you deliver cells to the

11   human body to create new tissues, a field we call

12   tissue engineering?

13                    And finally, I think there is going to

14   be deliveries to new locales.             In particular, one of

15   the big areas I expect will be delivery to the back

16   of the eye, just picking this as an example, which

17   is clearly going to be critical because right now

18   there are many new drugs that people are developing

19   for treating diseases like macular degeneration or

20   diabetic retinopathy, but the target is the retina,

21   and these drugs, it's going to be very hard to get

22   the drugs back there.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         68


1                     So there's a number of companies

2    developing approaches to deliver the drug to the

3    back of the eye.

4                     I think there's also enormous

5    opportunities in other areas of the body, like could

6    you deliver a drug to the nerves, for example to

7    great herpes or other diseases, and I'm just really

8    giving you -- and probably many, many others,

9    delivery to the ears, i.e., you             know, for children.

10    You could go on and on, but I think there's many

11   great opportunities and yet many challenges where

12   drug delivery can make a huge difference.

13                    That's largely what I'd like to say to

14   you today, but what I want to end with is one of the

15   things that has been a personal pleasure for me is

16   to be having been involved in the drug delivery

17   field for now about 30 years.             It has been wonderful

18   to see the enormous progress that has been made by

19   so many scientists and so many people throughout the

20   world to the point where we see all of these kinds

21   of therapies and many more that you'll hear about

22   today that are really relieving suffering and

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           69


1    prolonging life and obviously will do so more and

2    more in the years to come.

3                       Thank you so very much.

4                       (Applause.)

5                       DR. FEIGAL:     You did get us almost back

6    on schedule, and we have time for some questions.

7                       DR. LANGER:     Maybe it was very clear and

8    people want to leave very much.

9                       DR. FEIGAL:     Yes.

10                      PARTICIPANT:     (Inaudible.)

11                      DR. LANGER:     So the question is what are

12   my thoughts on personalized medicine and individual

13   therapy.         Well, I think that that will come to pass.

14    I think that in the genomic era we will see

15   examples, you know, where we're learning more and

16   more about individual people.

17                      I think something like that microchip

18   that I mentioned is a very interesting example of

19   where you could some day, you know, create like this

20   pharmacy in a chip for individuals.                 You know, maybe

21   that would be a pill or a patch in some form, but I

22   think that that drug delivery can make a huge impact

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           70


1    in that area at some point.

2                       Yes.

3                       PARTICIPANT:     (Inaudible.)

4                       DR. LANGER:     Would I like to speculate

5    in the regulatory process?             Actually, you know, let

6    me make two points on that.              The answer is no, but

7    one thing I should --

8                       (Laughter.)

9                       DR. LANGER:     -- but one thing I probably

10   should have mentioned that I think also could be

11   interesting by these techniques is actually record

12   keeping.         You know, I also think that when you do

13   things by these kinds of smart medical things that,

14   let's say, it was a chip or something else, that

15   whenever somebody takes a medicine that you could

16   actually get a permanent record of whatever drug

17   somebody gets and, you know, have that transported

18   to your computer so that that would probably be

19   useful information for the FDA as well as the

20   patients and doctors and medical companies to have.

21                      It's hard to say.        You know, when we

22   deal with things like personalized medicine, it's,

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                            71


1    at this point, easy to say things will happen.                     I

2    think you have to break it down into individual

3    cases to say exactly what the regulatory process

4    would look like or should look like.                  I think it's

5    too diffuse at this point really for me or probably

6    anybody to comment on, other than to know that many

7    years from now we'll probably see something like

8    that.

9                     DR. FEIGAL:     Well, let me thank you very

10   much for coming and joining us this morning.

11                    DR. LANGER:     It's a pleasure to see you.

12                    DR. FEIGAL:     Thank you.

13                    DR. LANGER:     Thank you.

14                    (Applause.)

15                    DR. FEIGAL:     Well, continuing with our

16   program and actually even with the novelty of being

17   a little bit ahead of schedule, our next speaker

18   this morning is David Klonoff, who is here from the

19   Mills Peninsula Diabetes Research Institute in San

20   Francisco, talking about how these technologies will

21   have an impact on diabetes.

22                    DR. KLONOFF:     Good morning.          I'd like to

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                         72


1    thank Dr. Provost and the people at the FDA for

2    inviting me to come out from California to talk

3    about diabetes.

4                     I'm an endocrinologist.          My topic today

5    is novel technologies for treatment of diabetes.

6    I'll be discussing metabolic monitoring, the

7    artificial pancreas, and alternate routes for

8    administering insulin.

9                     This is where I'm from, and this is the

10   hospital that I work at in San Mateo, California.

11   I'm editor of Diabetes Technology and Therapeutics,

12   which is a journal that covers an area I'm very

13   passionate about, which is new technology to help

14   people with diabetes, and through the journal we

15   also organize an annual diabetes technology meeting,

16   and some of you in the room have attended that

17   meeting.

18                    When I was asked to discuss new

19   technologies for delivery of insulin, I thought

20   first that I would approach it from the standpoint

21   of asking three questions.           First, with respect to

22   insulin, why develop this type of new technology?

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                             73


1                         What is the new technology?

2                         And how good is the new technology?

3                         So the first question I'm going to

4    address is why develop the new technology.                   I think

5    everybody knows that giving insulin through shots

6    hurts.           It's even in the latest issue of Popular

7    Mechanics.           Certainly all of my patients know

8    insulin hurts, and basically that means that there

9    are barriers to the use of insulin.

10                        When I tell patients that they need to

11   use insulin, they usually don't want to go onto

12   insulin, and I think that if we can discover some

13   new routes of insulin delivery, we can help overcome

14   barriers to the use of insulin for patients who

15   really need this drug.

16                        And some of these barriers now include

17   the pain and trauma from being pricked by an insulin

18   injection needle, the inconvenience of carrying

19   needles, and also the risk of hypoglycemia from an

20   inadvertent excessive dose of insulin or

21   hypoglycemia from rapid absorption of insulin.

22                        So there's clearly room for new routes

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           74


1    of insulin delivery.           So the second question is:

2    what is the new technology?              How can insulin be

3    delivered?

4                       Well, I'm going to be covering metabolic

5    monitoring, as well as the other topics.                So first,

6    metabolic monitoring.

7                       This is the current status of metabolic

8    monitoring in diabetes.            First, in terms of

9    performance, the blood glucose monitors that are now

10   available require less blood than ever, less times

11   than ever to get the reading, and they provide more

12   data management, better performance.

13                      Second, they're greater options in terms

14   of body fluids that are tested.               It's not just blood

15   anymore.         In terms of sites, it's not just the

16   fingertip anymore.          And in terms of how automatic

17   the readings are, it's not just whenever the patient

18   remembers to check themself anymore.

19                      Finally, we're starting to see a trend

20   in integration.         There are two themes in diabetes

21   technology now.         One is better blood glucose

22   monitoring.         One is better insulin delivery.

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         75


1    They're starting to come together within products,

2    within the same product.

3                     And so what we see is that these blood

4    glucose monitors are now being linked to insulin

5    delivery systems.       So, again, the current status of

6    metabolic monitoring is greater performance, greater

7    options, greater integration.

8                     I'm going to discuss the performance.

9    The current level of performance of blood glucose

10   monitors is that the blood volume that's required is

11   now as little as 0.3 microliters, which is about one

12   tenth of the volume that was required ten years ago.

13    So we've come an order of magnitude in ten years.

14                    Second, the measurement time.         We can

15   now get a reading in as little as five seconds,

16   which is also one tenth of the time that was

17   required ten years ago.          This is quite a dramatic

18   improvement in just ten years.

19                    Finally, a typical blood glucose monitor

20   can store up to 3,000 results.             This goes along with

21   the issue Dr. Langer raised about storing data, and

22   this means that you can provide with this

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                            76


1    information mean blood glucose levels over the

2    previous one week, two weeks or four weeks, and this

3    type of data can now be downloaded into a computer

4    or to personal digital assistant.

5                     That's performance.        Next, as far as

6    metabolic monitoring is the options.                  What we're

7    seeing now is that we have options that are

8    available on current blood glucose monitors, but

9    we're seeing some new options that are emerging that

10   are starting to become available and will become

11   available in future blood glucose monitoring.                    These

12   involve the body fluid sampled, the sites that are

13   sampled, and what I call the automaticity, how often

14   and how it's tested.

15                    So where currently you've all seen a

16   blood glucose monitor, it's invasive.                  It's jabbing

17   the fingertip to get a sample of blood, and that

18   hurts also.      People don't like that.

19                    What we'd like to see is more

20   noninvasive or minimally invasive technology, and

21   it's coming.      It doesn't have to be the fingertip

22   where all of the nerve endings are.               People can now

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                          77


1    check blood glucose at alternate sites, such as the

2    forearm or the thigh.          Fewer nerve endings; patients

3    are more likely to want to test themselves.

4                      Finally, the glucose testing is not just

5    intermittent, but it can be done continuously, which

6    means it does not require patient effort to remember

7    to check themselves.          So I'm going to cover each of

8    these options.

9                      First, the non -- minimally invasive,

10   then the alternate site, and then the continuous.

11   So regarding noninvasive and minimally invasive

12   monitoring, first, I'm going to define noninvasive

13   blood glucose monitoring.            This is an area that all

14   of my patients ask about:            when is it going to be

15   here?

16                     A noninvasive blood glucose monitor

17   generates and processes optical signals.               It does

18   not harvest body fluids, and it measures three

19   compartments when optical energy is put into the

20   body.        It measures blood, interstitial fluid, and

21   skin cells as a blended reading, and most of the

22   noninvasive monitors that are being developed are

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         78


1    using optical energy applied to the skin.              There are

2    a few that are applied to the eye and instead could

3    measure aqueous humor.

4                     So what you see is you're applying an

5    energy source, and it goes to these three body

6    compartments, the blood, the interstitial fluid, and

7    the intercellular fluid.          They all reflect in some

8    way, depending on the property of optical energy

9    that's being used, the radiation into a detector,

10   and you get a blended reading. And it's up to the

11   engineers to sort this out and figure out how much

12   glucose is in each compartment and whether there's

13   any type of a lag between the reading that comes

14   back out of interstitial fluid or intercellular

15   fluid relative to blood.

16                    Now, when you apply optical energy,

17   there are various measurable effects of this light

18   that can be measured, and it turns out that there's

19   a scientist or company somewhere that's using every

20   one of these principles of light to measure glucose,

21   and part of the issue is finding the right wave

22   length and the right type of energy.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          79


1                      But I've made a list of the measurable

2    effects on light that glucose can have.                You can

3    measure absorption of light or scattering,

4    refraction, Raman scattering, rotation,

5    fluorescence, impedance, photoacoustic heating, and

6    there is at least one company working in each of

7    these areas.

8                      This is a picture from my hospital of a

9    patient who has given permission to show his

10   picture, having a noninvasive blood glucose test

11   done.        This is a glucose clamp study where we keep

12   the blood sugar constant by infusing high doses of

13   glucose and high doses of insulin, and then we can

14   keep the glucose level constant for a long period

15   and make multiple measurements.              His right arm has

16   three IVs in it.        His left arm is on the noninvasive

17   monitor which is purposely not being shown in this

18   picture, but he's going to be here 12 hours for this

19   study.

20                     Now I'm going to discuss minimally

21   invasive blood glucose monitoring.                 A minimal

22   invasive blood glucose monitor means that the skin

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                             80


1    barrier is disrupted, but not a blood vessel.                     So

2    you're digging, but not all the way into a blood

3    vessel.          So what you're measuring is either

4    interstitial fluid or some other extravascular fluid

5    that's harvested from skin.

6                         It turns out you can measure some type

7    of a diluted interstitial fluid or real interstitial

8    fluid.           You can make a reading either

9    intermittently, if you sort of drill intermittently,

10   or you can leave the sensor under the skin, and then

11   you can read a continuous reading.                   You can either

12   pull the interstitial fluid out so that it will

13   harvest it, so to speak, and measure it after it is

14   out of the body, measure continuously, or you can

15   leave a sensor under the skin and measure it

16   continuously in the body.

17                        There are a number of ways of pulling

18   interstitial fluid out of the body, different types

19   of ways of disrupting the skin so that you can

20   measure the glucose concentration of the

21   interstitial fluid, and it turns out in most cases

22   that the interstitial fluid glucose concentration

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                             81


1    and the blood glucose concentration are very

2    similar.          Now, they're not always the same, and

3    that's a subject of research in itself, but they're

4    often about the same, and that's just the first

5    approximation.

6                         So these are some methods that are being

7    used now to pull out interstitial fluid.                   You can

8    apply current, and that method is known as reverse

9    iontophoresis.           You can apply laser to drill a hole,

10   in effect.          That's called microporation.           You can

11   use ultrasound, such as what Santra Medical is

12   using.           We saw a picture of that.         That's referred

13   to as cavitation, creating a space between cells and

14   the interstitial fluid sort of comes up to the

15   surface.

16                        You can puncture the skin with a very

17   fine needle.          You can abrade the skin surface with

18   powder.          You can dissolve the lipid barrier of the

19   skin on the surface with chemicals.                   You can apply

20   very strong suction or you can penetrate the skin

21   with a fiber optic filament.                And these methods are

22   all ways of disrupting the skin, but not going so

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         82


1    deep as to draw blood.

2                     This is the last method that I mention.

3    This is a technology that we're about to be testing

4    at Mills Peninsula Health Services in which you have

5    a fiber optic filament that's inserted to measure

6    epidermal interstitial fluid glucose, and this

7    method just stays in the epidermis and through an

8    optical fluorescent reaction on the surface.                 You

9    actually send light down the needle it fluoresces

10   according to how much glucose there is, and you're

11   measuring the fluorescence within the needle.

12                    Okay.   I've talked about one type of

13   emerging option, which is noninvasive and minimally

14   invasive monitoring.         Another option that's

15   available for patients now is alternate site blood

16   glucose testing.

17                    This is one of my patients about to do

18   violence to herself, and imagine if you had diabetes

19   and you had to check yourself four times a day, or

20   maybe even seven times a day.             This would really be

21   a drag.

22                    Imagine if you could check a blood

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           83


1    sugar, but you didn't have to use your fingertip, if

2    you could go to a place where it just doesn't hurt

3    as much.         That's known as alternate site blood

4    glucose testing.

5                       The status is now that five of the six

6    major manufacturers have products approved for

7    alternate site by the FDA, and I know that the sixth

8    company is working on it, and we'll probably hear

9    something from them any time.               So basically I expect

10   that at least five, probably six of the six will

11   have products approved by the FDA.

12                      Clearly there's less pain than a

13   fingertip site.         As far as when is it okay to check

14   an alternate site, there's a question of the lag for

15   two hours after meals and during hypoglycemia, which

16   means that as the blood sugar is rising and you know

17   it's rising if you do fingertip blood glucose

18   levels, it doesn't rise as quickly from the

19   alternate site.

20                      So if you check the blood sugar at the

21   route of the alternate sites, using the forearm or

22   thigh, within the first two hours you might think

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                               84


1    it's not as high as it really is because of this

2    lag.       So most of the products that are approved for

3    alternate site recommend that you just not test the

4    alternate site within two hours of a meal.

5                         Most people with diabetes aren't

6    checking themselves within two hours after a meal.

7    So it doesn't affect their lifestyle, but they

8    might.           A pregnant woman would frequently check

9    during this time frame, and she should not use the

10   alternate site if she's testing within two hours and

11   if the product says not to test during that time

12   period.

13                        There's some thought that agitation -- I

14   actually went to college at Berkeley.                     We know a lot

15   about agitation there.              Does agitation correct the

16   problem?

17                        Well, they don't mean here protest

18   marches, but actually rubbing really hard so that

19   the skin actually becomes warm and red.                     There's

20   some question that this may improve the blood flow

21   in the forearm so much that that could eliminate the

22   lag.       So this is being studied, and it could be that

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                         85


1    the lag will be less of a problem.

2                     This is one of the first people in the

3    world to have an alternate site.              He's one of the

4    first 50 people.       He's one of my patients who also

5    gave me permission to show his picture here today,

6    and as you can see, he's not really sure if this is

7    going to work out, but he's being tested on the

8    forearm.

9                     This product is so historic in the field

10   of alternate site testing it has actually already

11   been on the market and gone off the market, but this

12   is the first alternate site blood glucose monitor.

13   It was called the Atlast by Amira Medical, and Amira

14   Medical was later purchased by Roche Diagnostics,

15   and one of the things Roche did was take this

16   product off the market.

17                    But this is the beginning of an era of

18   alternate site testing.

19                    Now, the next option that has become

20   available for patients is continuous blood glucose

21   monitoring.      There are four products that one can

22   use in the world for continuous blood glucose

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         86


1    monitoring or continuous glucose monitoring.                 Two of

2    them are available in the United States and Europe,

3    and the other two are available only in Europe and

4    are not FDA approved for use in the U.S.

5                     These are the products, and I'm going to

6    say something about each of them.              Starting with the

7    continuous glucose monitoring system and the

8    Glucowatch, the Glucoday and the Pendra.

9                     So first, the continuous glucose

10   monitoring system which has recently been upgraded

11   to a second generation product.             Now it's known as

12   continuous glucose monitoring system gold              And this

13   is manufactured by Medtronic MiniMed in Northridge,

14   California, and it consists of three parts.

15                    First is a sensor that goes under the

16   skin, and it can sit there for 72 hours in the

17   subcutaneous tissue.         There's a little wire from it,

18   and this wire is connected to a monitor that stays

19   out of the body.       It looks like a pager, about the

20   size of a pager, and that stores the data.               It

21   stores the data but does not project the data in

22   real time, just like a 24-hour Holter monitor stores

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           87


1    the heartbeat data but does not tell the patient

2    what the heartbeats are at that time.                 And it could

3    be, in future generations we could see real time

4    data, but what's on the market at this time is more

5    like a Holter monitor.

6                     And then you can take this monitor and

7    put it into a docking station, and then link it to

8    your PC to download the data, and what you get is a

9    picture that looks somewhat like this, which each

10   color throughout the day is a blood glucose level,

11   and what you look for is certain times of the day

12   when there's a pattern when things look really high,

13   like maybe at this time of day here, or maybe when

14   they look really low, like around this time of day

15   here.

16                    And thanks to some technology that has

17   to do with smoothing, Medtronic MiniMed has recently

18   found a way to connect the lines at midnight so that

19   there's less of a disparity.

20                    Next I'm going to talk about the

21   Glucowatch G-2 biographer.           Glucowatch is well known

22   among diabetes circles.          We're on our second

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                             88


1    generation device now.              This is what it looks like.

2                         It could be useful for a person at night

3    because it has an alarm to wake them during

4    hypoglycemia.           This is manufactured in the county I

5    live in, San Mateo County, by Cygnus in Redwood

6    City.

7                         So what you see here are three types of

8    input.           This device tells you the time, which is in

9    green.           It tells you the blood sugar level, and

10   there's a trend arrow, and this trend arrow to me is

11   a nice idea.          It's really easy to make, but what you

12   see is if your blood sugar is sort of borderline,

13   and it's on the way down, it's really important.

14   Now you've got to eat or take some action, or

15   conceivably if your blood sugar is high and it's on

16   the way up, it's time for some extra insulin.

17                        I should point out that under the terms

18   of this product being cleared, it's not cleared for

19   use such that you can take the blood sugar reading

20   and act upon it.           For example, if you see a 110 and

21   your doctor has said when the blood sugar is under

22   120 you've got to eat and you're seeing 110, does

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                            89


1    that mean now you should eat because the doctor said

2    under 120 you should eat?

3                        No, because the terms of the clearance,

4    what's said, is that this information, if you're

5    going take some action based on the reading, you

6    have to go check yourself with the traditional blood

7    glucose monitor and use that information to decide

8    what to do.

9                        So it tells you, in effect, if the

10   number looks alarming that it's time to check

11   yourself with the traditional blood glucose monitor.

12    I think this was possibly a step to give a patient

13   an extra measure of safety because this is an early

14   device that measures glucose in a nontraditional

15   way, and by putting in this extra step, if for some

16   reason the device was not giving you an accurate

17   reading, you're still protected because you're going

18   to go out and do a traditional blood glucose

19   reading.

20                       Here's how it works.         This is a cut-

21   away.        On the face you have a display unit in

22   purple.          Now we go to the bottom, and what's

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           90


1    happening is that the electrodes in yellow create an

2    electrical current.        The current pulls salt from the

3    skin toward the surface.          The salt carries water

4    that's dissolved, and water carries glucose that's

5    dissolved within the water.

6                     So what you're pulling up, in effect, is

7    a diluted form of interstitial fluid.                 The glucose

8    is trapped in these glucopads, also known as

9    autosensors, and a chemical reaction occurs with the

10   biosensor, and you get a glucose reading.

11                    The way this device operates is there's

12   a two-hour warm-up.        You put it on and it gives you

13   no readings.      Then after the two hours you still

14   need to do a single calibration every 13 hours,

15   which is how long it lasts.            That is, you get

16   readings for 13 hours.          It will deliver six glucose

17   measurements per hour that you can read.                 So every

18   ten minutes it gives you a reading.               It has a

19   programmable alarm for panic values, such as that

20   woman who was sleeping.          If the blood sugar gets

21   below a certain level, it will make a noise and wake

22   her up.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                             91


1                         Excessive sweat cuts off the

2    measurements, and it uses a triple A battery every

3    13 hours.           This is a product that's approved in

4    Italy.           It's a continuous glucose monitor known as

5    Glucoday.           It uses a principle called microdialysis

6    in which interstitial fluid is sort of rinsed, and

7    the fluid that comes out contains glucose and using

8    a proprietary formula, the concentration of glucose

9    in this rinsed fluid is supposed to be proportionate

10   to glucose in the interstitial fluid, which is

11   supposed to be proportionate to blood glucose.

12                        Another product which recently received

13   clearance in Europe is the Pendra.                   This is an

14   interesting device by Pendragon Medical in

15   Switzerland.           This is actually a noninvasive

16   monitor.           So there's no fluid.        The skin doesn't

17   become wet underneath, and it uses a method known as

18   radio frequency impedance, and it sends a radio wave

19   to the blood.

20                        And it turns out that, according to the

21   Pendra, that if the blood glucose level changes,

22   that changes the shape of red cells, and that

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         92


1    changes the dialectric properties of red cells and

2    changes the impedance.          So they're measuring glucose

3    indirectly by this method.

4                     I made a list of some other promising

5    technologies for implanted, either minimally

6    invasive or invasive, monitors that will be used as

7    continuous sensors in the future.              I think any one

8    of these on the list could be the next continuous

9    sensor that we're going to see.

10                    And the type of signal that they read

11   generally is either an electrochemical signal, such

12   as an enzyme sensor under the skin, or it measures

13   optical energy.

14                    So when we go through these, I see first

15   you can have a microdialysis catheter

16   subcutaneously.       That's what the Glucoday uses, and

17   recently Roche Diagnostics has announced that

18   they're very interested in this type of technology

19   as well, microdialysis.

20                    Viscometry is a method in which you're

21   rinsing interstitial fluid out as you would with

22   microdialysis, but there's sort of a catch to it, is

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          93


1    that you're using a solution that contains dextran

2    and Concanavalin A, and in effect, the                higher the

3    glucose concentration, the more glucose will be

4    pulled off the dextran onto the Con A, and the

5    viscosity is less, vice versa.

6                     So, in effect, high glucose is low

7    viscosity.       Low glucose is high viscosity, and

8    that's a viscometry method.            This method has been

9    developed by Disetronic in Switzerland, and Roche

10   Diagnostics recently acquired Disetronic, and we'll

11   see what happens to viscometry.

12                    Another method could be to implant a

13   Fluorophore, an agent that fluoresces under the skin

14   or put a tatoo under the skin and then you

15   interrogate this agent with light.               That could be a

16   continuous monitor.

17                    And there's a company near here in

18   Germantown, Maryland, called Centers for Medicine

19   and Science, that's working on this technology.

20                    You could use an enzyme tipped catheter

21   implant, which is essentially what the continuous

22   glucose monitoring system gold uses, and there can

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                         94


1    always be a new generation by Medtronic MiniMed or

2    some other company, and this could either e a

3    subcutaneous device or an intravenous sensor, and

4    here again Medtronic MiniMed is working on one and

5    so is Animas in Pennsylvania.

6                     And Animas' method is using a near

7    infrared spectrophotometer intravenously, which is

8    pretty interesting.        The idea is that you've got a

9    noninvasive type method, which is shining light, but

10   they actually have the light and the receptor so

11   small and so close to each other that they could

12   implant that as a unit into a blood vessel.

13                    Okay.   The last area where there are

14   more options in terms of metabolic monitoring are

15   integration.      I can tell you at the American

16   Diabetes Association meeting, which was in New

17   Orleans last month, for the first time I was struck

18   by a trend which is integration between blood

19   glucose monitors and insulin delivery systems.

20                    People are talking about it, but it was

21   everywhere you looked at the ADA meeting.

22                    Now, there's two kinds of integration.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         95


1    There's the mechanical integration where two

2    different systems are located next to each other,

3    but they don't necessarily work together, but even

4    more important is functional integration where they

5    actually function together.

6                     Now, the first mechanically integrated

7    system was known as the InDuo, and when it came out,

8    the manufacturers, Novo Nordisk, which is an insulin

9    company and Live Scan, which is primarily a blood

10   glucose monitoring company, asked the question, "How

11   did you manage before InDuo?"

12                    And they showed this mess of stuff that

13   you have to deal with because now they came out with

14   the first ever -- I insert the word "mechanically" -

15   - integrated blood glucose monitor plus insulin

16   delivery system, but there's no sharing of data.

17   This looks like a blood glucose monitor on the

18   outside, but it's a shell.           When you lift it up, it

19   contains an insulin delivery system, which is

20   basically a fancy pen system for delivering insulin.

21                    So it contains both, but there's still

22   just two things that are linked together.              They

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         96


1    don't necessarily work together.

2                     The second mechanically integrated blood

3    glucose monitoring plus insulin delivery system was

4    created by BD.      They call this the Latitude diabetes

5    management system, and this system contains various

6    features that a person with diabetes would need, a

7    glucose monitor; this is in millimoles per liter,

8    not milligrams per deciliter, and it contains a

9    lancet, and it contains an area where you can put a

10   pen.

11                    So here, again, these are mechanically

12   integrated, but not functionally integrated.                 Now

13   we're getting into functional integration, and these

14   are four recent alliances that I've become aware of

15   -- just some of them recently, some of them still

16   not very long ago.        I've listed them from top to

17   bottom according to the size of the insulin pump

18   company.

19                    Medtronic MiniMed is the largest insulin

20   pump company, and they've recently formed an

21   alliance with BD, which is a blood glucose

22   monitoring company.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           97


1                       The second largest is Disetronic out of

2    Switzerland.         They've recently not just formed an

3    alliance, but been acquired by Roche diagnostics.

4    That's definitely an alliance.

5                       The third largest, Animas from

6    Pennsylvania.         They formed an alliance with

7    Lifescan.

8                       And then the fourth largest, which is

9    pretty new on the market in insulin pumps is Deltec

10   in Minnesota, and they formed an alliance with

11   TheraSense.

12                      I'll say a little bit more about what

13   each of these alliances consist of.                 This slide was

14   stamped on the upper right corner "FDA Clearance

15   July 7th, 2003."         So as Dr. McClellan stated, we now

16   see that here we have FDA clearance for a

17   combination product which is a blood glucose monitor

18   developed by BD and MiniMed called the Paradigm Link

19   Monitor.         Here's the person sort of on the run

20   holding their blood glucose monitor, and by radio it

21   sends a message to the insulin pump that the person

22   is wearing, the Medtronic MiniMed Paradigm 512 pump.

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                             98


1                         So now for the first time you have an

2    FDA approved combination product which is

3    functionally integrated.              The glucose level is

4    projected into the pump.

5                         Medtronic MiniMed then took it even one

6    step further in terms of integration, and they

7    created a type of software which they say does --

8    they call it diabetes math.                It tells you based on -

9    - you know what's your blood sugar now, and it tells

10   you there.          You know what you want your blood sugar

11   to be.           You program that in.       You know what you're

12   about to eat.           The device is pre-programmed with how

13   sensitive you are to insulin and how sensitive you

14   are to calories.

15                        And with that information, as well as

16   one other factor which is how recently did you take

17   some insulin, it will tell you what type of a bolus

18   of insulin you need to get your blood sugar down to

19   this target level.

20                        And they actually did a study which they

21   published in Diabetes Technology and Therapeutics in

22   June where they took some experienced people with

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                            99


1    diabetes and said, "Okay.           You must know how much

2    regular insulin to give yourself at a meal, and so

3    half of them, use how much you think you should use.

4     The other half, don't use what you think you should

5    use.       Use what our diabetes math formula tells you."

6                     And it turns out that the control was

7    exactly the same, but from the standpoint of the

8    manufacturer, that was a good thing because there

9    are a lot of people with diabetes who aren't so

10   experienced and have a lot of trouble figuring it

11   out, and the machine --the software-- did as well as

12   what an experienced person could do.                  So that's a

13   promising sign as far as integration goes.

14                    Now, Disetronic and Roche, what do they

15   have planned?      This is a part of a press release on

16   May 2nd, after we see the Roche acquire Disetronic.

17   By combining the two businesses, Roche will be able

18   to offer comprehensive diabetes management solutions

19   from blood glucose meters for self-monitoring to

20   sophisticated, programmable insulin pumps that allow

21   patients to continually administer insulin doses

22   according to their individual needs.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                         100


1                     So to me this sounds like they are

2    getting ready to integrate their pumps and their

3    glucose measurement systems.

4                     Now, Animas and Lifescan.            I'm not sure

5    exactly what they're doing together, but I can tell

6    you that each company has a link to the other on

7    their Web site and to no other diabetes company.                      So

8    that tells me something is going on.

9                     Finally, Deltec and TheraSense have

10   created a product which is not FDA approved, but

11   people form both companies are optimistic, and what

12   you have here is an insulin pump made by Deltec and

13   then clipped over it, this basically just looks like

14   a little holster or clip.           It's actually a

15   TheraSense Freestyle blood glucose monitor.

16                    So we look at this from the front and

17   from the back.      So one thing that TheraSense is good

18   at doing is creating blood glucose monitors in

19   different shapes, and they have created a monitor

20   that basically looks like a clip, and you stick the

21   strip in the bottom of it, and you get a reading

22   here.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                          101


1                        So I'd say from a design standpoint this

2    is a nice integration of a glucose monitor and a

3    pump.        It looks and feels as if it's one unit.

4                        Okay.   The next topic I'm going to talk

5    about is the artificial pancreas.                 Now, we don't

6    have an artificial pancreas on the market yet, but

7    I'm going to tell you what the artificial pancreas

8    will look like in a broad sense when it is

9    available.

10                       First, it will contain a continuous

11   sensor.          It will contain an insulin delivery system,

12   which you can think of as a pump.                 There will be a

13   controlled processor which receives a glucose signal

14   and then uses an algorithm to drive the pump.                    That

15   links the glucose measurement with the insulin

16   delivery, and then there will be a radio that will

17   first link the sensor with the insulin delivery

18   system so that it knows how much insulin to give and

19   with an external monitor so that the patient will

20   know what their blood glucose level is at all times.

21                       This is a potential candidate to become

22   an artificial pancreas.             They still have a lot of

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                            102


1    work to do, but this is the Medtronic MiniMed long-

2    term implanted sensor pump or sensor and pump

3    system.          This round system is an insulin pump.               It's

4    implanted in the abdomen, and you see the different

5    parts of it.

6                        At the tip of it is an insulin delivery

7    catheter, which would be way out here.                    It's a

8    little bit cut off, and then it's also connected to

9    an intervascular glucose sensor here.                    So this

10   device is put in the abdomen.                The tip of the sensor

11   goes into the peritoneum, and the peritoneal

12   delivery of insulin has some advantages because it

13   goes right to the liver, and the other end of it is

14   an intravascular glucose sensor that's intended to

15   stay in the superior vena cava for a year.                    So

16   that's one way, but there's other ways.

17                       An artificial pancreas could contain an

18   external insulin pump.             The insulin could be

19   delivered subcutaneously, and so there's different

20   combinations, but there are some problems that have

21   to be solved in order to have a successful

22   artificial pancreas, and each component has

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                        103


1    problems.        The continuous sensor, for example, will

2    have calibration drift.           There has to be some way of

3    recalibrating regularly.           When you put a sensor in,

4    you can't just leave it.

5                      You can have a lag between dynamic

6    changes in blood glucose and interstitial fluid

7    glucose if the sensor tip is not in a blood vessel,

8    but in the skin, and the majority of artificial

9    pancreas systems that are being developed have the

10   sensor in the skin.

11                     There can be lag.        There can be fouling

12   of the sensor.        There can be immune rejection or

13   fibrosis of the sensor so that the body forms a

14   capsule around it, and then it's not reading true

15   interstitial fluid but just some kind of altered

16   fluid that's within the cap.             And there's local

17   complications.

18                     Insulin delivery in an artificial

19   pancreas could have some problems, namely,

20   nonphysiologic response to elevated blood sugar.

21   There are some other stimuli that affect insulin

22   beside glucose, and the current artificial

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       104


1    pancreases are not really taking that into account.

2                     Insulin can be denatured if it stays in

3    the body, which is nice and warm, for three months

4    at a time.       There's systemic complications, and

5    there's anesthesia and surgical risks of putting it

6    in and taking it out.

7                     And then additional problems with the

8    artificial pancreas is that you just can't have

9    hypoglycemia.       You're the manufacturer.          Your

10   algorithm must protect against severe hypoglycemia

11   or the patient is going to get sick and sue.                 There

12   could be product recalls.           A lot of bad things could

13   happen.

14                    So you have to run the sugar a little

15   higher than you need it, and yet the whole idea of

16   an artificial pancreas is to keep it normal.

17                    Currently the artificial pancreas is

18   being developed to treat low blood sugar because

19   it's so important to avoid low blood sugar means

20   that in effect you're going to have more high blood

21   sugar than you want, and then finally there's the

22   economic impact of improving control from current

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       105


1    levels to better levels with the artificial pancreas

2    is unknown.      This can be very expensive.          It's not

3    clear who's going to pay for this technology.

4                     Another device that's being developed is

5    a bioartificial pancreas, and this is a device that

6    would substitute for an endocrine pancreas, but

7    instead of being purely bioengineered, it contains

8    synthetic materials and functional islet cells that

9    are encapsulated within a semi-permeable membrane to

10   protect them from immune rejection.

11                    So within the membrane, glucose comes

12   in.      The eyelet cells see it.         They figure out how

13   much insulin to make.         The insulin goes out, and

14   this membrane protects the eyelet cells from being

15   destroyed by antibodies or lymphocytes.               The results

16   look good in rodents, but we don't have good results

17   in larger animals or in humans.

18                    We need better immunoisolation to

19   protect these cells.         Every year I go one year

20   further out.      So you come back next year and it will

21   say 2009 maybe, and it's certainly going to be

22   expensive, about $20,000 a year.              I'll show you a

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           106


1    picture of an artificial pancreas.

2                       This is produced by a company in San

3    Francisco called Islet Sheets Medical.                  We'll look

4    at a liver, a dog liver, and on it is this sheet,

5    and within the sheet there's a little cuff that's

6    dark, and then this sort of milky white square.

7    This milky white square are islet cells, and this

8    sheet was sutured to the liver in a

9    pancreatectomized dog in the hope that these eyelet

10   cells would protect it from hyperglycemia.

11                      Unfortunately in this particular

12   experiment the sheet fell off.               The sutures broke,

13   and they don't know why this tends to happen.                    So

14   that's a problem they're working on.

15                      The last area I want to discuss is

16   alternate routes for administering insulin.                   Dr.

17   Langer covered some alternate routes for drugs in

18   general.         Insulin has some areas that I think are, I

19   think, interesting.

20                      Some promising technologies include

21   inhaled, oral, buccal, nasal, transdermal, all of

22   these ways of getting insulin into a person other

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                           107


1    than with a needle.

2                         Now, here's why inhaled insulin looks

3    promising.           If you give a person, say, in the

4    hospital intravenous insulin, which is red here,

5    what happens is it gets in very quickly.                  You want

6    rapid action.

7                         If you give the person subcutaneous

8    insulin, which is yellow, it lasts for a long time.

9    So that can be good in some situations.

10                        If you give inhaled insulin, what tends

11   to happen is you get rapid absorption of insulin so

12   that what you're seeing is similar to IVs.                   So it

13   gets in quickly the way IV insulin gets in, and it

14   lasts for a long time the way subcutaneous insulin

15   lasts.           So in theory inhaled insulin would be very

16   useful for people, especially at mealtime.

17                        Now, I'm going to show you what the

18   system looks like from what used to be called

19   Inhaled Therapeutics, now known as Nektar.                   I was an

20   investigator with three of their trials that they

21   did with Pfizer.

22                        This is the lady taking out the device.

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           108


1    It looks like an asthma spray device, but it's a

2    little bigger.           She's putting in an insulin-like

3    little sheet.           This is powdered insulin, and there's

4    a bubble that's going to go inside the device.                      So

5    she's putting that in.

6                         Now she's sort of getting the trigger

7    pulled back, and when she pressed the button it's

8    going to fire.           She's turning the mouthpiece.            It's

9    going to be facing her, and now she's firing the

10   trigger, and what's happening now is that the

11   blister of insulin is ripped.                 Air comes in, and

12   suddenly disburses the insulin into a cloud, and now

13   you see a cloud of insulin.                This is correct.       It's

14   white.           They call this a standing cloud.         It's

15   inhaled insulin, and she's inhaling, and in just a

16   moment it has gone clear.               I'll show you that again.

17                        Here it is, a cloud of insulin.           It's

18   clear.           Where did that go?      It went into her lungs.

19   So that's inhaling dry powdered insulin.

20                        Now she's finished.        She puts the two

21   cylinders one on top of the other and puts it away.

22   So that's one way of delivering inhaled insulin.

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                             109


1    She's all finished.

2                         Another way is being developed with

3    liquid insulin.           This is by a company -- I should

4    say Inhaled Therapeutics, Inc. is in San Carlos,

5    California.          This is being developed by Aradigm,

6    which is in Hayward, California.                  This is a first

7    generation device.            This is a second generation

8    device with liquid insulin.

9                         They're putting a blister in here.               The

10   insulin blister strip is inserted.                   Now you rotate

11   this mouthpiece, and a pin punches the blister

12   strip, and when the person inhales, they're getting

13   an aerosol of liquid insulin.

14                        This is a third generation device by

15   Aradigm.          They call it the AERx pulmonary drug

16   delivery system.           In that you're going to have

17   buttons and a mouthpiece and a screen.

18                        But an interesting feature here is this

19   green light.          This is the breath control guidance

20   light.           Here's why this is important.            In order to

21   make inhaled insulin work, to get it into the

22   alveoli          where you want it and not have it land in

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                         110


1    your mouth or in the trachea, you have to breathe at

2    the right speed and without turbulence.                 It has to

3    be even and at the right speed.             If you breathe fast

4    and jerk, it's going to go too fast and it won't get

5    into the alveoli.

6                     So people are trained to breathe

7    properly, and the idea of this device is as the

8    manufacturer claims, that only if you're breathing

9    the right way will it fire and deliver the insulin,

10   and if you're the patient, you don't know whether it

11   fired or not.      You can't even taste it.             So if you

12   see a green light, you know you got your insulin.

13   If you see a red light, you have to take another

14   dose until it gives you a green light.

15                    This is a method known as PDC

16   Technospheres.      This company has been known as PDC,

17   Pharmaceutical Discovery Corporation.                 Recently it

18   has been acquired by Mannkind.             Now these are

19   Mannkind technospheres.          We're about to do a Phase

20   II trial at Mills Peninsula on these spheres.

21                    This is an interesting technology.               You

22   take fumaric acid.        You polymerize it, and you form

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                         111


1    a shell around powdered insulin.                You get an insulin

2    loaded Technosphere, and the fumaric acid was

3    selected because at the pH of alveolar air it melts,

4    turns into liquid, and now the insulin is in the

5    alveoli.         It gets absorbed.       Fumaric acid is

6    absorbed.

7                       And according to what the company has

8    told me, that the fumaric acid is not toxic, and so

9    they found another way of delivering powdered

10   insulin to the alveoli.            This is what their inhaled

11   device looks like.

12                      Another method that actually Dr. Lander

13   is associated with, I'll just say a word about it,

14   is Alkermes' air particle.             This is an interesting

15   particle.        You want an aerodynamic diameter of one

16   to five microns if you want this powder to be

17   absorbed.        This particle has a larger geometric

18   diameter, five to 30 microns, but it's very fluffy.

19   It's looks like a flower, and it functions as if it

20   has the small aerodynamic diameter, and this device

21   uses an inhaler air dispersion chamber which

22   delivers porous powders.

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          112


1                        And they're working with Eli Lilly, and

2    one of the scientists form Lilly showed me this

3    device at the American Diabetes Association meeting

4    a couple of weeks ago, and he put in like an empty

5    capsule into the cap and he started breathing, and

6    it sounded as if there was something wrong with his

7    hygiene.

8                        But as it turned out it wasn't his

9    hygiene.         It's this capsule is designed to rotate

10   around.          The cup that it's in is slightly eccentric

11   and as it rotates, it spins off the insulin.                    So

12   it's designed that way, and they seem to be making

13   good progress with this technology.

14                       This is the last company I'm going to

15   mention, Aerogen in Sunnydale, California.                  The Air

16   Alkermes is in Massachusetts.                 They were in the air

17   inhaled insulin business.              We did a user study for

18   them, but they recently announced in December that

19   they're going out of the inhaled insulin business.

20   They're just going to work on inhaled drugs other

21   than insulin but use a Piazo electric effect that,

22   in effect, shakes insulin, and it sprays out.

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          113


1                      Okay.   Now, oral insulin.           Oral insulin

2    would be very attractive.            No needles.       People are

3    used to pills.        Why can't insulin be needles or why

4    can't insulin be pills?

5                      Well, if you can have an oral insulin,

6    you would need to avoid the acidic degradation of

7    the stomach, the enzymatic degradation of                 the

8    intestines, but preserve the potency of the insulin

9    molecule.        That's the challenge.

10                     So three different solutions have been

11   proposed.        One is to conjugate a low molecular

12   weight polymer to the insulin to preserve adequate

13   activity and resist digestion.              That's what Nobex

14   Corporation is doing.

15                     Or you can have a delivery agent that

16   carries intact insulin into intestinal cells as Dr.

17   Langer showed.        That's what Amesphere is doing, or

18   you can PEGylate -- that means conjugate with

19   polyethylene gylcol -- the molecule and then create

20   a micelle with Casein, and this will increase

21   transport to the gut epithelium.

22                     This is an example of the polymer where

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                          114


1    you've put a polymer onto insulin.                 This is an

2    example of how you have a delivery agent mixed with

3    insulin.         You've just got a plain, old pill, and

4    this is an example of a calcium phosphate insulin

5    that has been pegylated and you've formed a micelle,

6    and basically because you have a casing coating

7    around these little blue insulin balls, this means

8    that you can pass through the stomach of the

9    intestine, and it sort of falls apart.                  It stays

10   intact in the stomach, but it falls apart in the

11   intestine, and then because it has been pegylated,

12   it can get into the small intestine.

13                      Here's buccal insulin delivery.            It

14   looks like you're spraying it into the -- as if

15   you're inhaling it, but actually you're not.                    You're

16   aiming at the buccal mucosa here.                It contains

17   permeability-enhancing agent.               It gets absorbed very

18   rapidly just like we know nitroglycerine from buccal

19   mucosa gets absorbed rapidly.

20                      Nasal insulin requires dissolving

21   insulin with some type of calcium carbonate, and

22   there's different forms of calcium carbonate.

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                         115


1                       Finally, there's transdermal routes of

2    injection, that is, getting insulin to the skin

3    without a needle.         You could use a jet injector or a

4    patch or an implanted chip, which you've seen, or

5    micro needle.

6                       This is the Med Ejector Vision.           We've

7    done a study on this one at Mills Peninsula Health

8    Service.         The ideas are injecting the insulin not as

9    a puddle, but as a spray, and that perhaps the

10   insulin can get absorbed more quickly than if it was

11   injected by a needle.           That's being studied.

12                      This is using encapsulation systems with

13   an ultrasound to         break the skin cell barrier.             This

14   is similar to what Santra Medical is doing.                  This is

15   a company called Encapsulation Systems, Inc., in

16   Pennsylvania.

17                      This is using the MicroCHIPS technology,

18   which Dr. Langer discussed and showing how this

19   could be applied to insulin.              Each of these pyramids

20   here, which are sort of small, here you see blown up

21   in this case contains insulin, and when you put the

22   right charge on it, the gold cap in the presence of

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        116


1    a high concentration of electricity just blows off,

2    and now the contents, which are here, this spray,

3    the insulin, are strayed into the body.

4                      So a person could program how much

5    insulin they need with a wristwatch or you could use

6    different kinds of microneedles.               This is a human

7    hair to show that micro needles are not much

8    different in size than a hair.              This is a 25 gauge

9    needle, which you think of as small, but it's

10   massive compared to these microneedles.

11                     And this is one other type of device

12   which uses a microneedle, and it's so small you

13   can't even touch the needle.             So you program it with

14   a wrist watch.

15                     Okay.   the last question I want to ask

16   now that I've shown you all of the different toys

17   that we endocrinologists have to work with is, how

18   good is the new technology, and there are three

19   types of questions that I think should be answered

20   with new technology.

21                     Is the patient receiving the desired

22   dose?        Is the innovatively delivered insulin safe?

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          117


1    And is the innovatively delivered insulin effective?

2                      So regarding the dose, if you have a

3    blood glucose meter determining the insulin dose, is

4    that really the amount that's needed by the patient?

5     We need to be sure.

6                      Also, is this innovatively delivered

7    dose predictable and consistent?               People want the

8    same amount every time.           Is this innovatively

9    delivered insulin lost to the environment?                  And if

10   so, how much is lost?

11                     And is absorption of the alternately

12   administered insulin predictable and sufficient?

13                     These alternate routes tend to not have

14   as good bioavailability as injection.                  It all gets

15   in.      If you give it by mouth or by nose or by

16   inhaled, only a small percentage gets into the body.

17                     Safety.   Is there local toxicity of the

18   innovative insulin delivery system?                Is that system

19   itself irritating to the body?              Are there immune

20   problems?        Is the insulin itself causing local

21   toxicity?        Could it even be causing cancer because

22   it's a growth factor?

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                            118


1                         And finally, effectiveness.          Is the

2    bioavailability of this alternatively administered

3    insulin, is it adequate and consistent?                   Is the

4    availability affected by common environmental

5    factors, such as perhaps inhaled insulin?                   Could it

6    be affected by a person with asthma or smoking?

7                         Do the pharmacodynamics and

8    pharmacokinetics resemble subcutaneous insulin, and

9    are both types of doses, bolus, which is short

10   acting, and basal, which is continuous dosing

11   options, available for the patient?

12                        So I raise some questions.           I'm going to

13   show you how one man's approach to this, and this is

14   Dilbert.           This next to the last slide shows

15   innovative technology according to Dilbert, and here

16   Dilbert is getting a report.

17                        The new product brochures have already

18   won design awards.            Dilbert is going, "That's great,

19   but our product won't do any of the things you claim

20   here."           I wonder who says that all the time.

21                        "Well, who should we believe, the award-

22   winning designer or the guy who can't stop

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                        119


1    complaining?"

2                     (Laughter.)

3                     DR. KLONOFF:     So in conclusion,

4    regarding new technologies for innovative insulin

5    delivery, improved metabolic monitoring now allows

6    improved bolus dosing.          Continuous monitoring will

7    allow improved basal dose adjustments.                Closed loop

8    artificial and bi-artificial pancreas systems are

9    coming, and new routes of administration will remove

10   barriers to use of insulin.

11                    And if we do these things and have

12   better methods for delivering insulin, then all of

13   our patients will have better glucose.

14                    Thank you very much.

15                    (Applause.)

16                    DR. FEIGAL:     Well, thank you.

17                    Our next speaker, changing topics, is

18   going to take a look at the emerging techniques and

19   technologies for treatment of solid tumors.                 Dr.

20   Jonathan Kruskal from Harvard University.

21                    DR. KRUSKAL:     Dr. Feigal, colleagues, I,

22   too, would like to thank the organizers for inviting

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                        120


1    me to participate in today's seminar.

2                     One hat I wear is that of an

3    interventional radiologist performing minimally

4    invasive tumor oblations in solid human organs, and

5    I'd like to share with you this morning in the time

6    remaining some of the exciting emerging new

7    techniques and new technologies that we are using

8    both in the laboratory and already in the clinical

9    setting.

10                    Some of the challenges that we face in a

11   daily basis for treating solid tumors include, first

12   of all, vector engineering.            How do we optimally

13   take drugs or genes to get these to a site in the

14   body for optimal efficacy?

15                    Secondly, how do we deliver these?              What

16   are the options available to us as interventional

17   radiologists that allow us to deliver drugs or genes

18   into solid tumors in pretty deep cavities of the

19   body?

20                    What you've heard so far this morning

21   are the transdermal, the inhalational.                They're

22   pretty superficial ways of delivering drugs in

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                           121


1    genes, but in the real world setting with solid

2    tumors, you really need to get deeper, and image

3    guidance provides us with opportunities to get

4    needles pretty deep into the body and to deliver

5    locally.

6                         And finally, how can we inhibit efflux?

7    It's all very well dropping the payload into a

8    tumor.           It's all very well trying to enhance uptake

9    of that payload into a tumor, but if we just leave

10   it, it's simply going to be washed out or

11   metabolized, and we need to see what options are

12   available to us now in terms of inhibiting efflux of

13   drugs out of solid tumors.

14                        What I teach our fellows in residence in

15   terms of drug delivery into tumors is ways of an

16   approach to enhancing the payload efficacy, and the

17   way we would like to look at it is simply how do we

18   deliver drugs.           How do we deposit these into tumors?

19    How do we get these to be detained within the

20   tumors?          And how can we ultimately destroy these

21   tumor?

22                        Some of the innovative techniques that

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           122


1    we're now using for treating solid tumors can be

2    categorized either into the intervascular area,

3    interstitial treatments and efflux inhibition, and

4    I'll go through all of these in the remaining time

5    and show you what we are already doing and how some

6    of these can be approached.

7                      Well, let's start off with payload with

8    efficacy.        How can we look at the new strategies

9    available to us in terms of delivering drugs with

10   genes into tumors?

11                     These tumors on the left, you can see

12   this is a typical conventional delivery of drugs

13   into liver tumors.         This is a catheter inserted by

14   the groin all the way up the aorta into the hepatic

15   artery supplying the liver, and you then deliver --

16   you can see these lines over here of the pacified

17   arteries going into the tumor.              You can deliver drug

18   into these large round liver tumors.                   This is drug

19   that we on a daily basis deliver in a poppy seed oil

20   extract called ethiodol, which is a depo delivery

21   system for enhancing retention of drug in these

22   tumors.

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                           123


1                         We can image this.        We can see exactly

2    where the drug is going.              We can look at the

3    efficacy of the drug in terms of serial CTOMR to

4    know if a tumor is being made any smaller.

5                         But what we don't know at this point is,

6    in fact, is the drug getting to where we want it,

7    and on this complementary electromicrograph, you can

8    see this small lipid particle, this liposomal

9    aggregate which has got into the tumor cell and is

10   actually adjacent to the cell nucleus.

11                        So what are the ways that we can do

12   right now to enhance delivery both from delivering

13   it in an endovascular route all the way into the

14   nucleus of the cell to effectively get the treatment

15   we want?

16                        Well, let's look at some of these ways.

17   Catheter design.           There are some remarkable new

18   advances in terms of catheter design for delivering

19   drugs.           We will be hearing a little bit later on

20   today about some of the drug-eluting stents.                     These

21   right now are primarily for cardiovascular or

22   angiogenic type treatments, drug eluting stents or

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         124


1    other deliver chemotherapeutic agents, those that

2    will prevent stenosis.          We are putting stents into

3    livers to, in fact, prevent portal hypertension in

4    patients with cirrhosis.

5                     But what's equally important is to

6    deliver drugs into the wall of these stents that

7    will prevent these from occluding and allow these

8    patients to continue living good quality existence.

9                     We are currently seeking further

10   oncologic applications.          These are minimal right

11   now, and I'm sure there's a huge amount of

12   opportunity for oncologic applications of these

13   drug-eluting stents.

14                    Intervascular circled in vivo

15   bioengineering, which is where genes are delivered

16   into endothelial cells via catheters.                 The catheters

17   are inserted into specific vessels in the body.                     You

18   can then implode.       You can drive these genes into

19   the cells lining the vessels, endothelial cells,

20   effectively to create, for instance, a situation

21   where these blood vessels will not be blocked off.

22                    And, once again, we have not taken

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                       125


1    adequate advantage of the entire field of

2    angiogenesis.      Right now in tumors a lot of the

3    theory behind tumor treatment right now is

4    unblocking the blood vessels, destroying the blood

5    vessels to the tumor.

6                     But a lot of the patients we see, again,

7    on a daily basis, the minute the blood vessels have

8    been knocked out supplying the tumor, it effectively

9    takes away a lot of the options we have for treating

10   these tumors.      Since we are delivering a lot of

11   drugs via the vessels by blocking these major

12   vessels going to the tumors, we've effectively taken

13   away several major options for our patients, which

14   is not an optimal situation.

15                    So there are ways of taking advantage of

16   angiogenesis to find a nice match between the two.

17                    This is two examples I've taken from an

18   article of John Thomas in radiographics in 1998, and

19   these are types of catheters which are being

20   developed now for drug or gene delivery.              You can

21   see over here this is simulated vessels.              Two

22   balloons are blown up in this catheter, and you can

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       126


1    then perfuse a drug or gene mixture in the vessel to

2    allow it to deliver into the endothelial cells.

3                     More exciting is this type of catheter,

4    this patch type catheter where the wire is inserted

5    into a vessel, it's blown up, and you can see this

6    loop which develops, it does not block the vessel.

7    It allows the blood to          continuously pass through

8    the vessel without causing any ischemia or

9    occlusion, and you can then profuse your drug or

10   gene in this helical tube, and it then leaks out.

11   It's a very permeable membrane, and it leaks out

12   into this little cavity over here, and it will then

13   allow it to basically be taken up by the endothelial

14   cells.

15                    These are the types of systems that are

16   now being delivered and explored for local delivery

17   of drugs or gene product and peptides into the

18   endothelial cells lining vessels.

19                    What about some of the therapeutic

20   vectors, the therapeutic ways in which we delivery

21   payload into tumors?

22                    And the four categories I will be

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          127


1    talking about will be radio immunotherapy, vector

2    engineering and design, some of the new cell

3    delivery techniques, and some of the new gene

4    delivery enhancement techniques.

5                     Selective internal radiation therapy,

6    I'm sure many of you have heard about this.                  As an

7    example I've just selected the Yttrium microspheres.

8     These are very small, 32 approximately micron resin

9    microspheres onto which is bound some radiation,

10   Yttrium 90.

11                    This is then delivered.          We put a

12   catheter all the way up, again, up the aorta.                   We

13   target this catheter with guide wires into the

14   tumor, and then you can deliver these small, little

15   microspheres directly into the tumor.                 There's

16   preferential deposition in very vascular angiogenic

17   tissue, and we can deliver, therefore, therapeutic

18   dose of radiation to the tumor and not to the entire

19   organ.

20                    The liver, as an example, is a very

21   sensitive organ.       If you expose the liver to

22   conventional doses or radiation treatment, you're

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                         128


1    going to wipe out the liver function, and the

2    patient might succumb.          However, if you can deliver

3    this local radiation treatment to solid vascular

4    tumors, it allows you to then subject this to a much

5    higher radiation exposure than conventional

6    radiation treatment.

7                     However, this technology certainly needs

8    to be optimized.       There are lots of companies out

9    there which are exploring it.             We need to see some

10   good comparative prospective studies.                 We need to

11   see the technology optimized before I would

12   certainly be happy about administering this to any

13   of our patients.

14                    Immunocongugates monoclonal antibody

15   therapy also is being used right now, not with too

16   much success in our experience, and as an example,

17   if you take colon cancer, which expresses what's

18   called a carcinoembryonic antigen on its cell

19   surface, you combine radiation Iodine 131 to these

20   monoclonal antibodies.          You can deliver these

21   intravenously, and these will then bind onto the

22   cell surface of any tumor cell which is expressing

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                        129


1    this antigen.

2                     The problem, of course, is that many

3    other normal cells in the body might express it,

4    such as the colon, and so we need to basically

5    improve ways of targeting the immunoconjugates.

6    It's not sensitive enough at this time.               The

7    monoclonal antibodies need to be worked on.                 It's

8    not enough to simply use a rather specific

9    monoclonal-type antibody.           You need to use antibody

10   fragments and small, little peptide fragments,

11   cyclic peptides as well, and this might improve the

12   localization.

13                    The other area which is explored in many

14   laboratories is once you've actually delivered these

15   onto the surface of the tumor cell, how do you get

16   these inside.      How can you internalize either this

17   radiation or, in fact, whatever you might put on it.

18    This might be drugs.         This might be other types of

19   therapeutic agents.        How do you get these in?

20                    And the areas which are being looked at

21   now with some, in fact, quite optimistic early

22   results include radio frequency or heat,

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         130


1    sonoporation using focused ultrasound, and UV light.

2     All of these techniques are being explored in the

3    laboratory setting for enhancing uptake and

4    internalization of delivered immunoconjugates.

5                       Vector engineering is another area which

6    is receiving a lot of interest in the laboratory

7    setting.         I'll give an example of what we refer to

8    as immunoliposomes.          Some of the very good work has

9    come out of David Cheresh's group in La Jola, and

10   what they've done is they've taken advantage of

11   tumor angiogenesis.          The integren off of E-beta-3 is

12   expressed on very early angiogenic vessels.

13                      What they've done is they've bound a

14   monoclonal antibody to this integren, to a small,

15   little liposome which contains gadolinium.                 We can

16   see gadolinium with MRI, and therefore, if you give

17   the small immunoliposome into an animal at this

18   stage, it will actually localize in areas where

19   there are integrens being expressed in very early

20   angiogenic territories, and you can see it because

21   of the gadolinium.

22                      In further studies, what they've done is

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       131


1    they've also then bound doxorubicin, the

2    chemotherapeutic agent doxorubicin, to this same

3    agent, and this, again, will then target the

4    doxorubicin to the integren which is being

5    expressed.

6                     Phage display technology is a very

7    exciting, I'd like to say, new technique.              In fact,

8    it has been around for a while, which really allows

9    us to target far more specifically than monoclonal

10   antibodies would, and in using phased( )display

11   technology, that group and others have certainly

12   been able to identify small what they call cyclic

13   peptides, and these will target not only small

14   integrens, but as more work is done, in fact,

15   they're finding that these probes target multiple

16   different receptors.

17                    They're able to target angiogenesis.

18   They're able to target receptors on tumor cells.

19   They're able to target other enzymes which might be

20   expressed prior to angiogenesis, such as the so-

21   called metalloproteinases.

22                    So, in fact, a more and more basic

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       132


1    science is being performed, they're identifying more

2    and more applications for each of these probes.

3                     Similarly, tumor receptor is another

4    big, exciting area.        A lot of work has been done on

5    tumor proteases.       Ralph Weissleder and his lab in

6    Boston has developed a lot of imaging probes to the

7    cathepsins and other proteases.             Metrics

8    metalloproteinase is one of our own optical imaging

9    probes actually showing a circular room of matrix

10   metalloproteinases being expressed around the

11   periphery of a colon cancer metastasis in this video

12   micrograph of a colon metastasis in a mouse liver.

13                    And there are also a variety of growth

14   factor receptors which are now being targeted, and

15   remember we can use these not only for diagnostic

16   purposes, but also for therapeutic purposes.                 So we

17   can try and look at developing probes which show us

18   on an imaging basis where these receptors are,

19   confirm that they're being expressed, and then block

20   them with a lot of these very exciting, new factors

21   which are being engineered.

22                    VEGF, the vascular endothelial growth

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         133


1    factor, also very exciting.              VEGF is being used.

2    You'll hear in subsequent talks this morning about

3    the way in which it's being used in Hans

4    angiogenesis.

5                       VEGF can also be targeted for gene

6    therapy.         We use VEGF; in fact, we drop it onto

7    tumors with needles, and it enhances the

8    permeability of the leakiness of tumors, and we can

9    then pulse this with drugs off to its enhanced

10   delivery of drugs into tumors.

11                      So whereas VEGF might not be the ideal

12   agent being expressed by tumor cells because it

13   enhances angiogenesis in growth, we're also

14   administering it to enhance delivery of drugs into

15   these tumors.

16                      Targeting tumor-associated cells, this

17   is something that we hit on inadvertently a couple

18   of year ago through our radio frequency ablation

19   program.         It's well know that many solid tumors,

20   breast, for instance, will recruit systemic

21   macrophages.         Systemic macrophages are recruited

22   into the center of solid tumors, and these then

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         134


1    might play either a pro or an anti-tumor effect

2    depending on which specific population of

3    macrophages these are.

4                     However, we have now found, in fact,

5    that when you ablate a tumor with radio frequency

6    ablation, you can actually recruit specific types of

7    macrophages that would have an anti-tumoral effect

8    on the tumor.

9                     And we have taken advantage of this.

10   This is a small colon cancer metastasis.                This is a

11   video micrograph of an exteriorized mouse liver with

12   colon cancer, and by sticking a needle in and

13   ablating this for about 30 seconds and waiting for a

14   few days, we've recruited these very Agard

15   phagocytic macrophages into the cell.                 These black

16   cells infect all systemic macrophages which have

17   taken up these small carbon micro particles, and

18   this is a different population of macrophages to

19   which reside in the typical growing antiogenic tumor

20   cell.

21                    So therapeutic macrophage recruitment is

22   interesting not only because of its anti-tumoral

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                         135


1    effects, but because these avidly phagocytic cells,

2    to me, seem to represent a wonderful delivery site

3    for drugs or for genes.

4                      Taking advantage of tumor permeability,

5    you have already heard in the previous two talks

6    about pegylated liposomes.            We have certainly played

7    around with these a lot.           This is just an image.

8    You can see this is a diagrammatic illustration of a

9    liposome.        These yellow bands along the periphery

10   are the polyethylene glycol.

11                     And what this does is they provide

12   stearic hindrance.         What this means is that if you

13   just inject these into the blood stream, they will

14   circulate.        They will have a prolonged intravascular

15   residence, and these thin strands of polyethylene

16   glycol will prevent these from being taken up by

17   macrophages throughout the body.               They, therefore,

18   would stay in the blood stream for up to two days.

19                     The illustration on the right, again, is

20   one of our small -- this is about a two millimeter

21   colon cancer tumor growing in a mouse liver.                  You

22   can see PV is the portal vein, is the blood vessel

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       136


1    supplying the tumor, labeled as T, and what we have

2    done is we have simply given these animals an

3    injection of a small amount of these pegylated

4    liposomes containing doxorubicin, and these will

5    simply leak out because of the leaky vessels within

6    the tumor.

7                     And more interesting, in fact, is that

8    the doxorubicin will only fluoresce once liberated

9    from the actual liposome, and all of this bright

10   white area is the liberated doxorubicin which we can

11   see in real time.

12                    So taking advantage of tumor

13   permeability is another broad area that to me seems

14   quite optimistic and hopeful.

15                    So we've looked at the vector

16   engineering.      We looked at the catheters.          Now let's

17   look at cell transplantation.             Cell transplantation

18   certainly we've heard in this previous talk.

19   There's a lot of opportunities for diabetes.

20                    We are injecting islet cells into

21   patients in our institution, but what's sort of

22   strange and bizarre to me as a radiologist is that

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         137


1    clinicians come to us; they give us a little vial;

2    they provide the patient' and they say, "Please

3    inject this into the spleen."

4                       And we inject these eyelet cells into

5    the spleen, and we have no idea where these cells

6    are going, and this, of course, I think is one of

7    the big challenges we're dealing with in liver cells

8    as well.         We're injecting hepatocytes into the

9    spleen, and there's a lot of work that needs to be

10   done in the laboratory to know exactly where these

11   cells are going.         They seem to be working in some

12   patients, not working in others.

13                      And interestingly, we're finding with

14   our liver cells, which we're giving to patients to

15   tide them over prior to transplantation, that they

16   seem to reside within the spleen and do quite well

17   and actually work.

18                      So that opens up another whole

19   possibility.         You can have ectopic location of

20   normal functioning cells.             They don't need to be in

21   the organ where they normally function.

22                      In our oncology patients, we're

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       138


1    injecting the fibroblasts and the dendritic cells

2    into the peritoneal cavity.            We do this under image

3    or ultrasound or CT guidance, and again, these are

4    cells which have been transduced to produce things

5    like human growth factor, some of the clotting

6    factors in our hemophiliac patients, and this again

7    provides a wonderful opportunity.

8                     However, as has been said before, we

9    certainly await new techniques for improved

10   targeting of these cells, and I think this is

11   another big area that a lot of work needs to be

12   done.

13                    So recruitment I've mentioned here.

14   Some cells can be recruited.            Certainly image-guided

15   MCF delivery; what I mean by MCF is the macrophage

16   chemotactic factors.         You can literally pick up the

17   sigma biochemicals catalogue and purchase overnight

18   a whole variety of different chemotactic peptides,

19   and a lot of these now that we inject in an image

20   guidance into a solid organ in the body will then

21   recruit macrophages, which might have an anti- or

22   pro- tumoral effect.         And we need to explore this

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                              139


1    area further.           There's a lot of opportunity here.

2                         Radio frequency tumor ablation we've

3    shown.           Our own institution recruits macrophages,

4    and this, again, was data that was sitting in front

5    of our eyes for years and years, since every time we

6    did this to an animal or patient we would get

7    histology that would show a lot of macrophages, and

8    the assumption that we made, that this was simply

9    the RF-induced inflammatory response.

10                        So certainly there's a lot of data out

11   there that we just need to look at again and take

12   advantage of.

13                        And these cells, again, are a wonderful

14   depo for drug and gene delivery.                  These are two

15   micrographs, again, in our little mice in the lab.

16   This is an exteriorized mouse liver.                      You can see

17   the vessels draining out.               This is the portal vein

18   coming into the liver.              These are the individual

19   liver cells, and these small white dots, in fact,

20   are the liver macrophages, also known as the Kupfer

21   cells, and we've delivered a fluorescent peptide to

22   these, and you can see the broad delivery of these.

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                       140


1                     Whereas once we let a tumor grow inside,

2    we ablate this tumor with RF.             You can see a

3    different population of macrophages which takes up a

4    different dye, which has been localized around these

5    tumor cells.

6                     So depos for drug and gene delivery, I

7    think, are another bit area that deserves some

8    further work, and this is, again, one of our images.

9     This is radio frequency recruited into two

10   macrophages, and what these have now done is they've

11   taken up liposomal doxorubicin, and it is being

12   released in these macrophages.

13                    So this is a one millimeter tumor.

14   These are macrophages which are being recruited

15   often within the center of the tumor for about two

16   to three days after RF ablation, and these are not

17   there before, and you can then deliver drugs to

18   these.

19                    And these are also a rich population for

20   delivery of gene products.

21                    Adoptive immunotherapy, I don't want to

22   get into this in too much detail, but it is

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        141


1    certainly being performed in patients in our

2    institution.      What we mean by this is one of several

3    things.

4                     First of all, you can take natural

5    killer cells from the patient or others.                You could

6    activate these with lymphokines, reinject these into

7    the patient, and then hope that these will somehow

8    attack the tumor for some therapeutic purpose.

9                     The trouble is the nonspecificity of

10   these cells, and again, to improve targeting of

11   these natural killer cells.

12                    And then lastly, in this category, the

13   so-called TIL, the tumor infiltrating lymphocytes.

14   What we have in our institution is one of the basic

15   science researchers takes lymphocytes.                He

16   transfixed them with a cDNA of carcinary rheonic

17   antigen, and then what they do is they actually

18   ultimately start making an antibody for the

19   carcinary embryonic antigen, and we then reinject

20   these back into the patients, and they will then

21   home in on our patients with colorectal cancer

22   metastases in the liver.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                           142


1                         And we are just sharing these, and this,

2    again, is one of our micrographs of a small mouse

3    liver.           This is looking directly into a live tumor

4    in the liver through a microscope, and these small,

5    little cells here are the lymphocytes which, in

6    fact, fluoresce under the appropriate conditions,

7    and we can target these to the tumor.

8                         However, clinically is it successful?

9    I'm not convinced.            It seems to target other parts

10   of the body, such as the colon, and it's an area

11   richly in need of good research and optimizing this

12   technology.

13                        Gene-based therapies.         We hear earlier

14   that gene therapy has not been performed that much

15   in humans.           Certainly in our institution it appears

16   to be.           We've seen some major hurdles over the last

17   couple of years, but with a lot of trepidation and

18   being extremely gentle with the patients, we

19   certainly are delivering genes to patients.

20                        Two of the major innovations that I

21   think we're going to hear about for treating solid

22   tumors are the use of tissue specific promoters and

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       143


1    the use of inducible enhancers.             And what I mean by

2    this is the ways in which genes are being

3    synthesized now are to allow specific factors on

4    them to promote gene expression, and one which is

5    being used is VEGF, the vascular endothelial growth

6    factor.

7                     And what this means is that in an animal

8    model you could introduce genes into solid tumors,

9    wait for these to become angiogenic, become

10   invasive, and the minute VEGF starts being

11   expressed, it turns on therapeutic anti-tumoral

12   genes.

13                    And then what we'll also look at is how

14   we can actually enhance delivery of genes, and the

15   areas which are being looked at with most interest

16   are heat, hypoxia, and ultrasound.

17                    The inducible enhancers of gene

18   expression, a little gene fragment, a little cDNA

19   fragment consists of an enhancer subunit, promoter

20   subunit, and the actual gene.

21                    And what you can do is, if you can

22   basically subject this enhancer subunit to one of

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          144


1    many ways of activation, it will, in turn, activate

2    the promoter subunit, will activate expression of

3    the gene product, which will then be released and go

4    off and have the therapeutic effect.

5                        How can we take advantage of this?

6    Well, certainly with hypoxia.                Hypoxia inducible

7    factors can be inserted on the enhancer unit, and

8    then in the presence of hypoxia, these will then be

9    activated to express genes, such as the gene for

10   VEGF of a variety of other genes.

11                       Believe it or not, in the year 2003, we

12   are delivering chemotherapy to patients with solid

13   tumors.          We're then blocking the vessels in the hope

14   that this will occlude the blood supply and kill the

15   tumor.

16                       But as I've just shown you, in fact, to

17   make a tumor hypoxic, it, in fact, stimulates VEGF

18   expression and should, in reality, induce further

19   growth of the tumor.            And this really is sort of the

20   take-home point I'd like to leave us all with, is

21   that a lot of things that we are doing to patients

22   right now, they seem to have a wonderful, positive

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           145


1    effect on a lot of patients, and in theory some of

2    these might not work that well.

3                         Ultrasound is something that Bob Langer

4    mentioned, and certainly he deserves even more

5    credit than we can give him for what he has done in

6    this field, but heat shock protein is another

7    protein which has recently been identified as a

8    protein which can be up-regulated by the presence of

9    the heat delivered by ultrasound.                  If you can make a

10   gene that has heat shock protein inserted into it,

11   you can then target ultrasound directly to this gene

12   and it will inactivate this and induce gene

13   expression.

14                        The trouble is that this has not been

15   done with too much efficacy at this point, and we

16   need to look at all of the entire spectrum of other

17   available heat opportunities for this.

18                        So we've not delivered vectors.           We've

19   delivered genes.           We've delivered drugs into the

20   tumor.           How come we enhance the delivery here?

21                        First of all, drugs, which can enhance

22   permeability and, secondly, mechanical; there's a

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           146


1    variety of different pre-targeting drugs that we can

2    look at.         VEGF again, as I said, we drop it onto

3    tumors to increase endothelial pores.                   We can

4    actually deliver via catheters transient

5    permeability enhancers.            You can see all of these

6    that I've mentioned over here on this slide:

7    platelet activating factors, bradykinin, all of

8    these will, in fact, enhance permeability.

9                       Mannitol is used by neurosurgeons to a

10   large extent to disrupt the endothelium, and then

11   mechanical enhancement.            It's well known that RF

12   ablation as well as electrophoresis or antiphoresis,

13   all of these will enhance permeability to allow

14   drugs to be delivered.

15                      This is one of our tumors we have

16   subjected to 30 seconds of RF ablation and changed

17   this with small fluorescent microbeads, and all that

18   you can see the track of the needle inside the solid

19   tumor, and you can see how the microbeads, they leak

20   out around the tumor.           So certainly RF can enhance

21   permeability.

22                      Something I suspect we might be hearing

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                          147


1    a little bit more about later on, these so-called

2    magnetic targeted carrier particles.                  These are

3    small, little magnetized particles onto which

4    different chemotherapeutic drugs can be bound.                     This

5    is then delivered via catheter into a patient's

6    blood system, and then these magnetic particles can

7    effectively be sucked out by a magnetic field placed

8    onto the patient's surface.

9                     Here's an example of this, a catheter

10   that has been delivered into an artery supply in

11   these liver tumors.        These magnetic targeted

12   carriers are delivered into the liver tumors.

13   Magnetic field is placed over there that would suck

14   these out, and then these are delivered into the

15   tumor.

16                    And you can use MRI to actually see this

17   small, little magnetic particles in the tumor.                     What

18   needs to be looked at, in fact, not only is the

19   system being fully optimized, but once you've got

20   small magnetic ion particles in the liver, what

21   effect would this have on other therapies?

22                    For instance, if you use ion and RF

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                       148


1    ablation, what effect would ion and RF ablation?

2    Would this be synergistic?           Would this be

3    antagonistic?

4                     There's a lot of additional exciting

5    work that can be done here to further optimize this,

6    and this sort of falls into the category of what I

7    call cooperative therapies, something that hasn't

8    received much attention, but for an example, RF can

9    be used to recruit targetable macrophages.

10                    We already are injecting the genes for

11   P53 into solid tumors, and what these do is they

12   then allow the tumors to, in theory, re-get into the

13   normal way of dying, but P53 also allows us to

14   subject these tumors to a lower level of radiation.

15                    Radiation-inducible promoters are

16   another entire area.         Thermally-activated vectors,

17   vectors which can be delivered in the blood system,

18   into solid tumors and then shattered by subjecting

19   these to different heat techniques.

20                    In vivo electroporation, sticking a

21   needle into a solid tumor, delivering drugs

22   systemically, and then by subjecting this to a local

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           149


1    electric field, allow these drugs, just as we do in

2    the laboratory, to be taken up into the tumor cells.

3                         And then, of course, a nice combination

4    that we have done and published last week, in fact,

5    is a combination of radio frequency and liposomal

6    doxorubicin, and our theory here was that once you

7    have a tumor in the liver, you can give the patient

8    liposomal doxorubicin or, in fact, any liposomal

9    agent.           It will then surround the periphery of the

10   tumor.

11                        We then, using image guidance, stick a

12   needle into this tumor.              We turn on the RF ablation.

13    You can see the red heat, and then what this does

14   is it actually extends all the way out to ablate the

15   entire tumor.

16                        And I was also actually very excited.

17   We've done this in quite a few patients.                   The

18   regulatory issues in and of themselves are very

19   interesting because RF ablation is approved.

20   Liposomal doxorubicin is approved.                   So we've taken

21   two approved technologies, and what we're getting

22   over here, this is one of our patients, and it's

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       150


1    showing us some very surprising results.

2                     This is a tumor which has been ablated.

3    This is the liver.        This is a CAT scan through the

4    patient's upper abdomen.          This big, black area is

5    the dead tumor, but you can still see a few blood

6    vessels within it.

7                     And about two weeks later these blood

8    vessels have disappeared completely, and the types

9    of results we're seeing, in fact, is that              whereas a

10   couple of months ago we could only ablate tumors up

11   to four centimeters in size, we're now getting up to

12   eight centimeters in size.           So a 100 percent

13   increase in tumor size.

14                    We've even showing in our animal studies

15   that the survival of the animals has increased.

16   We're also getting slowed growth not only when the

17   entire tumor is ablated, but when parts of the tumor

18   are ablated, and we're also knocking out blood

19   vessels which may be residual.

20                    So the combination of interstitial

21   treatment, such a microwave or radio frequency

22   ablation and drug therapy, certainly is being used

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         151


1    at this point in patients and deserves further

2    investigation.

3                       In such activation of expression of

4    drugs or genes, you can certainly induce local

5    liberation of contents of drugs with

6    photoactivation, radiation of sound radio frequency,

7    heat sensitive liposomes, a lot of great work being

8    done by Needham's group down in the Duke hypothermia

9    project, and here they are using special liposomes

10   which are activated or shattered apart by heat.

11                      And of course, sonoporation of using

12   ultrasound to shatter liposomes, and this is an

13   example.         Some of the ultrasound contrast agents are

14   being designed to have a biomaterial on the outside,

15   which are antibodies which can target these to

16   specific surfaces of tumor cells.

17                      They have a polymer inside which is

18   specifically designed to be shattered by using

19   conventional ultrasound waves, and then inside they

20   could have a drug or a gene.

21                      And then what you do is you subject this

22   to ultrasound waves.           This will then break it apart,

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        152


1    release the small, little peptides, and allow local

2    release of gene or drug inside a tumor.

3                      And we, in fact, are doing this in the

4    laboratory.       This is the liver ultrasound delivered

5    doxorubicin.       This is a small liver in a rat, and

6    there's no ultrasounds being given when you subject

7    this to conventional ultrasound, and by

8    conventional, exactly the same ultrasound that many

9    in this room may have gone to have your fetus, your

10   embryo imaged.        It's not using any higher frequency

11   ultrasound whatsoever, and you can show the marked

12   increase in the fluorescence of this doxorubicin

13   when this is subjected to approximately 30 seconds

14   of conventional ultrasound.

15                     What we have shown that's even more

16   interesting, in fact, is that in the presence of a

17   tumor, you can get even further delivery.               So this

18   really opens up a whole new ball game where we can

19   use conventional ultrasound, and already we're

20   exploring this.

21                     The patient comes in.         We can image the

22   tumor        in the liver.    We can then give a drug and

                                  NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          153


1    actually use that exact same ultrasound while we're

2    imaging it, target the beam, and try to deliver

3    this, get local delivery and implosion of the

4    ultrasound contrast agent.

5                      Detention of the payload.            We're almost

6    done.        There's certainly a lot of pharmacologic

7    inhibitors.        These are efflux inhibitors.            Once

8    you've got the drugs into the set tumor cells, we

9    could take advantage of the ATP dependent pumps, P-

10   glycoprotein multi-drug resistance pump is something

11   that a lot of drugs being used for other purposes

12   will block, and there are a variety of these multi-

13   drug resistance-associated proteins.

14                     Any of these infective, once the drug is

15   inside the tumor by giving these to the patient or

16   to the animal, it will inhibit efflux of these drugs

17   out, and of course, the mechanical inhibitors.

18                     And there's some very good work that has

19   come out of the laboratories of Genzyme in Boston

20   showing that gene delivery intravenously in animals

21   by inhibiting flow out of the liver, by occluding

22   the hepatic veins, will cause significant increase

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                        154


1    in the uptake of genes into these cells.

2                      So, of course, using catheters and other

3    engineering techniques to cause local increase in

4    interstitial pressures certainly may have a positive

5    effect on gene and drug delivery, and this is,

6    again, one of our small colon cancer cells, and what

7    we've done is we've given verapamil and Cyclosporin

8    A, and this has inhibited efflux of doxorubicin out

9    of this tumor cell.

10                     So these are types f therapies, types of

11   approaches that need to be looked at once you have

12   delivered the payload, once you've deposited in the

13   cell.        You need to prevent it from being released.

14                     So in summary, this was a very brief

15   overview.        For the treatment of solid tumors there

16   really are a variety of emerging techniques and new

17   technologies.        There are a huge amount of

18   opportunities for optimization of these techniques,

19   especially these combination therapies.                However,

20   someone who is doing these on a daily basis -- and I

21   think this is where the challenge really is -- we

22   still do await some good quality, peer reviewed,

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          155


1    published science showing which techniques are the

2    best.        We need to compare the techniques, and we

3    would really as clinicians love to get involved in

4    some good, prospective, randomized studies to see

5    which are really going to be best for our patients.

6                      Thank you very much.

7                      (Applause.)

8                      DR. FEIGAL:     Thank you.

9                      Our final speaker before the break is

10   Richard Kuntz, who will be talking about the novel

11   technologies for the treatment of cardiovascular

12   disease.

13                     DR. KUNTZ:     Good morning.         I'd like to

14   thank Dr. Feigal and Dr. Provost for inviting me to

15   this wonderful session.

16                     And I'd like to talk in the next few

17   minutes about the clinical impact of some of the

18   technologies that you heard about this morning,

19   mainly focusing on the drug eluting stent

20   experience.

21                     We all know that coronary stents use

22   funny, little metal cages that have been around for

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                       156


1    about 15 years, made of about three different types

2    of materials, mainly stainless steel 316L or Nitinol

3    or recently cobalt chromium.            These materials are

4    now referred to as bare metal stents because of the

5    drug-eluting stent environment, have basically

6    revolutionized the treatment of coronary disease

7    throughout the world.

8                     That is, these cages basically open

9    lumens that are blocked in the coronary arteries and

10   maintain, because of their physical properties and

11   mechanical properties of plastic deformation, can

12   maintain an opening in the artery despite injury

13   sustained by the stent, and overcoming the reaction

14   of vascular injury.

15                    Now, one of the problems is that when

16   you start to expand any new therapy, you start to

17   see a problem associated with expansion of the

18   clinical outcomes.        We initially evaluated stents in

19   basically simple patients, and they could be defined

20   by patients with large vessels and generally non-

21   diabetics.       They had rates of failure that were

22   very, very good and basically were associated with

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         157


1    pretty much a breakthrough therapy in coronary

2    disease.         That is, only about ten to 20 percent of

3    the patients who were treated with coronary stents

4    in the simplest lesions would ever fail over the

5    course of the restenosis period, which is about six

6    months.

7                       But as expansion included diabetics and

8    longer lesions and vessels that are smaller, we

9    started seeing that these parameters are actually

10   quite influential on the geometry of renarrowing.

11   So that when you have patients who are diabetics

12   with long vessels and small lesions, failure rates

13   approach 50 percent.

14                      So this is, I think, a pretty typical

15   cycle of any new technology, that when it is

16   initially introduced it is with really fantastic

17   results.         Clinicians figure out a way to expand it

18   to patient populations where it fails again.

19                      (Laughter.)

20                      DR. KUNTZ:     And then it's time for us to

21   now engender a new need for a new breakthrough

22   therapy.

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       158


1                     So the drug-eluting stent process

2    started out, and it wasn't necessarily that it was a

3    drug-eluting approach.          Early on we know the biology

4    of thrombus and neoplasia, which is the renarrowing

5    process of restenosis, is guided by four different

6    types of pathological processes.

7                     One is that when you put a stent or

8    injure any artery, you get initially thrombus that

9    forms on the artery.         This engenders an inflammatory

10   process at the site with recriminative white cells

11   and macrophages.       This leads to stimulation of the

12   deeper tissue in the vasculature of proliferation,

13   both of in situ perivascular cells and also media

14   which transform to macrophages in the fibroblast and

15   recruit more cells and they basically heap up the

16   scar that if you're in a vascular bed, generally it

17   causes a reduction in the lumen size.

18                    And then finally, arteries that don't

19   get stented actually can contract around the

20   inflammation itself so that there are these four

21   process that we have known for years cause a

22   problem.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       159


1                     The problem has been that almost every

2    drug available in the last 15 to 20 years has been

3    tested in over 40 or 50 multi-center randomized

4    trials, and all have failed.            So the notion in the

5    mid-'90s was that maybe we should reevaluate some of

6    these drugs with the emerging technology of local

7    drug delivery.

8                     That was always in the back of the mind

9    of many of the scientists that not enough drug was

10   getting to the tissue site because it had to be

11   given systemically.        So the notion of local

12   delivery really has been manifested as a success and

13   the poster child for drug delivery at this point is

14   the drug-eluting stent.

15                    Now, in conjunction with this concept

16   that local delivery was important was even more

17   science that was added by Nurse, Hartwell, and Hunt,

18   who ultimately ended up winning the Nobel Prize in

19   1991 for their similar work on understanding the

20   importance of specific key proteins orchestrating

21   cell division.      These include Cyclin CDK, CDK1, and

22   a variety of P proteins.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       160


1                     Simplistically one can look at a variety

2    of compounds that have been around for a while and

3    look at their impact using this model on the cell

4    cycle and, in general, knowing that the

5    implementation of a stent would cause activation of

6    inflammation followed by cell division, and trying

7    to process some of the data from those Nobel Prize

8    winning science, we could see that potentially these

9    drugs that have been used in other areas, including

10   immunosuppression and chemotherapy, might be

11   valuable loading a stent to stop a cell from getting

12   into mitosis.

13                    Now, early on we know the radiation

14   therapy is extremely effective in that, and there

15   was a heads-up with respect to that working because

16   radiation therapy is extremely effective in the

17   prevention of in stent restenosis, that is

18   restenosis happening a second time.

19                    So we do know that we can inhibit

20   mitosis, and radiation therapy is kind of a no

21   brainer approach, but we can reduce this problem of

22   repeat failure after stenting.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       161


1                     A variety of different drugs that are

2    mentioned here include Sirolimus, which is the brand

3    name for rapamycin; paclitaxel and actinomycin D.

4                     Now, if we look specifically at the

5    first compound extensively studied, which is

6    rapamycin, Sirolimus, we know that processing some

7    of this data that a variety of cell receptors, both

8    stimulated by white cells and by platelets lead to

9    activation of some of these key proteins that are

10   synthesized at some unknown protein enzyme, and this

11   has been referred to as the target of rapamycin

12   because it is felt that rapamycin works after

13   combining with a KPB12 to inhibit the function of

14   TOR in leading to the synthesis of these key

15   proteins, which lead to cell division.

16                    So one had to utilize this science with

17   the emerging technology, as was pointed out by

18   previous speakers, of polymers that can hold and

19   deliver the drug.

20                    So the concept of drug-eluting stent was

21   started, pioneered throughout several centers

22   throughout the world, including MIT, with some of

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       162


1    Dr. Langer's students, including Elazar Edelman at

2    the Biotechnology Center.

3                     And these agents were felt to be part of

4    a three-part process of combination, including the

5    initial stent itself, which was generally just a

6    stainless steel stent on the market; a pharmacologic

7    agent which was going to work and have some

8    theoretical advantage to prevent mitosis, and, of

9    course, the most critical thing was the drug

10   vehicle.

11                    And if you follow the coronary field in

12   polymer science in the last 15 years, we actually

13   didn't get off to a good start initially.              Polymers

14   were probably the harder nut to crack rather than

15   the drug itself because the initial polymers were so

16   toxic that they in themselves would cause dramatic

17   vascular responses.

18                    Well, after a lot of work, and this is

19   almost ten years of work at Cordis in conjunction

20   with Wyeth-AIRS, there had been multiple efforts to

21   try to develop the ultimate polymer-holding drug

22   with a top coat that would allow for delivery to

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       163


1    stent without rubbing off the drug, and ultimately

2    release of drug over the course of 30 days that

3    would, in fact, interfere with the process of

4    thrombus and inflammation, which was the kind of

5    ring leader of the restenosis process that occurred

6    subsequently for six months.

7                     The notion was, in fact, if you could

8    stop the upstream processes of cell division, you

9    wouldn't get the manifestation of heaped up

10   neomyplasia after six months.             So the notion was to

11   develop a rapidly releasing polymer that would get

12   drug into the vasculature within the first seven

13   days and possibly as late as 30 days.

14                    Now, I'll jump right to the clinical

15   trials because we could spend a lot of time on the

16   polymer science here, and there are better speakers

17   than me to talk about that, but with respect to how

18   this has manifested itself out, early on there were

19   some studies done in South America, as are a lot of

20   kind of under the radar screen studies that are done

21   outside the United States, and one of the initial

22   studies with this drug showed up as a winner.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        164


1                      The first in-man analysis demonstrated

2    that after treatment of 40 patients there was

3    absolutely no latent loss that would be expected to

4    be seen at six months, and this triggered initially

5    Cordis to start two prospective studies.

6                      Now, the prospective studies were first

7    a study called RAVEL done in Europe, and then the

8    FDA regulated study in America called SIRIUS, which

9    was more of a pivotal trial study.

10                     The RAVEL study was actually designed to

11   demonstrate reduction in a surrogate of restenosis,

12   which is angiographic narrowing.               A 200-patient

13   study generally wouldn't show reductions in clinical

14   outcomes, and it was substantially and markedly

15   positive.        That is, if we look at the classical

16   measures of narrowing, which is the crossing of the

17   50 percent narrowing diameter stenosis at angiograph

18   at follow-up, it rate was 26 and 27 percent, as we

19   would expect, in the control arm, and in the active

20   arm it was zero.

21                     Now, there are a variety of ways of

22   measuring narrowing within the stent and outside the

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         165


1    stent, but regardless of how we measured it, it was

2    quite fantastic, and this study was performed by Dr.

3    Serois in Rotterdam using his European colleagues,

4    and it was probably the most substantial

5    breakthrough in the field of interventional

6    cardiology in the last 30 years.

7                     Now, this was in tandem and slightly

8    frame shifted behind, performed with a study called

9    SIRIUS, which was the American study.                 Again, this

10   study is a lot larger because it's powered to

11   demonstrate reductions in the clinical restenosis

12   rates, which are lower and less powerful endpoints

13   than that established from angiographic measures,

14   and we see that the restenosis rates

15   angiographically were also substantially reduced.

16   You can see the reductions here, almost 90 percent,

17   depending on how we measure restenosis.

18                    This, again, is unprecedented not only

19   in coronary cardiology, but in medicine in general.

20                    If we look at other measures of what the

21   target was, which is this amount of neomyplasia best

22   measured by three dimensional intervascular

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                       166


1    ultrasound reconstruction, you can see that when the

2    patients were exposed to normal stenting, they had

3    34 cubic millimeters on average of neomyplasia

4    compared to 2.6 from the other group, again showing

5    substantial reductions.

6                     And then if we go to the robust clinical

7    measures, that is, does the patient have to be

8    revascularized, what about if they had a heart

9    attack and other kinds of very robust measures?

10                    This is the major clinical outcome

11   called target lesion revectorization, and that was

12   reduced almost fourfold, from 16 to four.                And if

13   we look at that event plus anything else that can

14   happen to the patient, including small heart

15   attacks, it was still substantially reduced.

16                    Now, it was interesting because we have

17   a paper pending in the New England Journal of

18   Medicine that should be out next month, and in the

19   initial review the editors asked us to remove the

20   words "marked" and "substantial" that we were using

21   in the manuscript because they said it sounded like

22   a marketing brochure rather than a scientific paper.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       167


1                     And we tried to figure out a way to

2    describe the 91 percent treatment effect without

3    using the word "substantial" or "marked."              It was

4    pretty hard.

5                     (Laughter.)

6                     DR. KUNTZ:     So you'll see sentences

7    like, "A treatment one effect was found, 91

8    percent."

9                     What's interesting is that this is

10   almost a dream come true from an initial

11   perspective, and that is the field of DES, I think,

12   is more so than just SIRIUS itself, Sirolimus.

13   These drugs in their initial incarnation so far

14   appear to work without any increase in adverse

15   events, and stent thrombosis was something of great

16   concern because we were putting a polymer on top of

17   the surface of the stent, and that might be a

18   problem.

19                    And in a variety of different studies

20   from Europe and Canada, America, and others, the

21   pooled analysis shows the same thrombosis rate or

22   even lower from what we would expect at least on the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       168


1    patients we've studied so far.

2                     So in general, the inclusion criteria

3    for this trial, which included relatively sick

4    patients, had fantastic results from a stent

5    thrombosis perspective.

6                     What also is interesting was that if we

7    looked back at those predictors clinically of

8    increased restenosis, which is the length of the

9    lesion, the size of the vessel of the person with

10   diabetes, there was a really uniform treatment

11   effect -- this is looking at clinical restenosis --

12   across the board.

13                    That is, if we looked at linear,

14   nonlinear modeling, if we looked at actual results

15   and we tried to smooth them in a variety of

16   statistical ways, we would find this consistent

17   effect.

18                    So this, again, is a little bit unusual

19   to see in medicine where almost all subgroups

20   benefit to some degree.

21                    Another way to look at that is just to

22   break them down by the observed outcomes, and this

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       169


1    is the classical odds ratios analysis, and, again,

2    this is a familiar graph that one takes a positive

3    study like this with its odds ratio reduction from

4    the active arm and its confidence intervals, and

5    then measures it against the unity line, and then

6    looks at a variety of subsets.

7                     And it's very hard to come up with any

8    other study in medicine I know of that has all of

9    these subsets located so far to the left.              So it was

10   very hard for us to find any subsets that didn't

11   have substantial advantage in this group overall.

12                    What's more interesting mechanically is

13   that we've always known that with the advent of

14   stenting and its ability to prevent abrupt closure

15   and other acute complications, many interventional

16   cardiologists use a lot of stents because they could

17   really get themselves out of problems.

18                    But there's a price that you pay, that

19   is, the increase in stent length was associated with

20   substantial increase in restenosis, and this is

21   mainly a probabilistic reason statistically.

22                    Well, this was almost negated by our

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       170


1    experience so far with the Sirolimus stent,

2    suggesting that now the interventional cardiologists

3    can have their cake and eat it, too, that they can

4    put the long stents in, the so-called full metal

5    jackets, and not pay the price they have before with

6    substantial increases in restenosis per se.

7                     Now, we don't want these interventional

8    cardiologists to go hog wild and start putting a lot

9    of stents in.      Surgeons certainly don't want that,

10   but at least when one is concerning themselves about

11   an acute complication, like an edge dissection, and

12   you're always debating as to whether you should put

13   that extra stent in, we feel that the patient can

14   actually benefit from having a safe approach by

15   putting the extra stent length in because the price

16   we see so far of restenosis is very minimal for

17   extra stent length.

18                    We followed this for now a year, and

19   what we see is that even from the initial nine month

20   outcomes which were reported to the Food and Drug

21   Administration and led to approval of the one-year

22   data, still is maintained, and if anything, we still

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       171


1    see a slight reduction in freedom from restenosis in

2    the control arm by the main endpoints, and it is

3    still maintained, I assume, more robustly in the

4    active arm.

5                     So our treatment effects actually have

6    lightened, interestingly enough, even from nine

7    months to 12 months, to suggest that there is no

8    evident catch-up phenomenon.

9                     If we look at the RAVEL study, the one

10   that was started slightly before, the two-year data

11   suggests that we have still maintenance of good

12   clinical outcomes, and there's clearly in all of the

13   angiographic analyses no evidence that this process

14   of delay or narrowing that occurs in six months is

15   delayed any more than what we normally see in six

16   months.

17                    Now, European studies have just been

18   reported a few months ago.           Again, a new data set;

19   again, phenomenal results overall, and I think

20   overall the results of rapamycin with three

21   randomized trials now suggest that this is a good

22   drug.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        172


1                      Well, what about other drugs?         Does it

2    work?        Is the answer local drug delivery or is the

3    answer Sirolimus?

4                      Well, paclitaxel is another important

5    therapy, and its first study was a 500-patient study

6    done in Europe, and it also showed marked reductions

7    in restenosis.        The FDA study called TAXUS-4 in

8    America, which has, again, over 1,000 patients will

9    be presented relatively soon, whose results, I

10   think, are being filed if not now, to the Food and

11   Drug Administration, and I think they'll be

12   presented some time in           August or September.

13                     But if it does follow this initial

14   European experience overall, we're looking at

15   probably another 50 to 60 percent reduction in

16   restenosis.       We're the second drug now attached by

17   polymer to a stent.

18                     Does that mean that every drug-stent

19   combination now works?           The answer is no.        Actually

20   it doesn't.       The same drug, paclitaxel, was shown

21   not to have substantial reduction in restenosis, 13

22   versus ten, when directly applied to the stent

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                              173


1    surface.          Okay?    Paclitaxel is a sticky molecule,

2    and if you spray it on and then put it in the body,

3    it actually doesn't seem to prevent restenosis to

4    the same degree that we certain saw with rapamycin

5    or the other formulation of Boston Scientific TAXUS

6    stent.

7                         So I think the polymer technology is

8    critical, at least from my limited perspective, so

9    far.       It looks like that is an important component

10   rather than just drug and stent alone.

11                        There are lots of other polymers out

12   there.           I just want to give you a little sampling

13   now of what they look like.                Abbott, in

14   collaboration with Biocompatibles in the U.K., has

15   access to phosphatidylcholine, which this agent is

16   like a sponge.            It essentially is easy to apply.                 It

17   holds molecules up to 2,000 Daltons.                      It is a

18   natural reservoir and can be easily manipulated to

19   change its kinetics of release.

20                        Abbott, in conjunction with Medtronic,

21   are looking at a variety of different compounds,

22   including a rapamycin analogue called Rapalog, or

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                       174


1    ABT 578, and both of them have licensed this

2    compound, and there are two studies that are ongoing

3    right now in Europe.

4                     Interestingly enough, there is some

5    interesting data from basic old drugs that are off

6    patent and have been studied before and were

7    negative, and when combined with a polymer looks

8    initially like it might have good results as well,

9    and they include dexamethasone estradiol.

10                    And of course, Guidat has another

11   rapamycin analogue in a polymer called everolamus,

12   and this in a study called FUTURE in Germany has

13   demonstrated fantastic results so far.

14                    If we look at the overall experience so

15   far, we can start to classify them, and this is from

16   Peter Fitzgerald, who is virtually the intervascular

17   ultrasound core laboratory in Stanford for almost

18   all of these studies, and what he's seeing is that

19   he's got a marked reduction in neomyplasia using

20   either paclitaxel or the limus family.

21                    Now, I don't know that there's a

22   difference between these two.             These are very small

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         175


1    sample sizes overall.           I'm a little skeptical about

2    that.        I think when we find the actual results from

3    the TAXUS-4 study we'll be able to tell whether, in

4    fact, they're all in the same class or not.                  My

5    guess is they probably are.

6                       In any event, they're substantially

7    lower than that seen in the bare metal stent.

8    Again, polymer is the key for a variety of these

9    drugs that work.

10                      Now, I just want to point out one other

11   stent just to show how the technology can go

12   further.         This is just an interesting company that

13   has a stent in which the struts now have little

14   holes in them, and what these holes are are little

15   wells that can contain drug.

16                      And there is a manufacturing process

17   that can precisely place in these tiny holes levels

18   of drugs with different levels of polymer and

19   different elution characteristics so that one could

20   stack a variety of different drugs with different

21   release kinetics so that if you want to have a drug

22   for the first three days, it would be released, a

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         176


1    drug for the next week would be released below that,

2    and so on and differential release both to abluminal

3    and vessel size.

4                       This is a very interesting type of new

5    technology, and I think we'll see more and more of

6    this.        Trying to design a trial, I think, to deal

7    with all of these permutations may be difficult, but

8    in general if one comes up with a theoretical nice

9    combination of drugs, such approach might be

10   something interesting and may stimulate other people

11   to think about likewise approaches.

12                      Now, one of the important things is how

13   does drug-eluting stents, even as in its infancy

14   right now, how does that impact on how we take care

15   of patients with coronary disease per se.                Well, as

16   an interventional cardiologist, we're constantly

17   measuring ourselves against the surgeons, and early

18   on we felt that we owned a single vessel disease

19   problem.         That is, the heart usually has three

20   vessels, and if one is blocked, you generally don't

21   want to send someone to surgery for that.

22                      Well, there have been a variety of

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       177


1    studies done on patients with multi-vessel disease

2    and comparisons with surgery, and in general,

3    there's not much of a difference except for maybe a

4    subset of diabetics with severe vessel disease.

5    There's not much difference between mortality or

6    other major adverse events between the two

7    therapies.

8                     That is, angioplasty or bypass surgery

9    tend to be extremely effective with respect to the

10   ability to revascularize and also has about the same

11   major adverse event outcomes.

12                    But the main problem with angioplasty

13   has been that the restenosis process requires that

14   it be reintervened on, and that gap was 32 percent

15   when balloon angioplasty was initially out there.

16                    This slide, by the way, I borrowed from

17   Dr. Serois in Rotterdam who made this up.              Now, Dr.

18   Serois is also the PI of the ARTS study, which is

19   the first stent study versus bypass surgery, and

20   that gap for revascularization repeat in

21   intervention has narrowed to 14 percent.

22                    Even with conservative predictions of

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       178


1    what the drug with the stent world can look like, it

2    now appears that even under multi-vessel angioplasty

3    and stenting we not only will be as safe as surgery

4    for many multi-vessel diseases, but possibly even

5    have fewer revascularization failures than surgery

6    alone, and this is going to have a tremendous

7    impact, I think, in how patients with multi-vessel

8    disease are going to be treated, and slowly we'll

9    have to do clinical trials to prove that one can

10   shift into the coronary surgical arena.

11                    And, in general, I think that this is

12   very good for patients because the noninvasive

13   approaches or less invasive approaches, I think, are

14   going to take over in a big way from the more

15   invasive surgical procedures.

16                    Now, if you're a stent company with a

17   new drug-eluting stent, the question is how are you

18   going to do your study, and            if you are around a

19   year or two ago, you could do this study, which is

20   like TAXUS or SIRIUS, and do a 1,200 patient study

21   compared to bare metal stent.

22                    But now that the first drug-eluting

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       179


1    stent is out of the bag and CMS is paying for it,

2    it's hard to do a study against bare metal stent

3    because everybody is going to get a drug-eluting

4    stent in America.       It seems that way, at least.

5                     So we have to consider looking at

6    equivalency studies overall, but if you look at

7    trying to be equivalent to something that only has a

8    five or six percent rate of failure clinically, you

9    need to do a big study, four or 5,000 patients, or

10   if you try to beat the five percent, you              know,

11   failure rate, which would be very hard to do, that

12   still requires four to 5,000 patients overall.

13                    Well, I think what you're also going to

14   see if you're interested in the clinical field here

15   is that I think in collaboration with the FDA there

16   are going to be several clinical investigators and

17   others working with a large group at the FDA

18   interested in surrogate outcomes, and we'll try to

19   make a case for angiography and also intervascular

20   ultrasound as very powerful measures of looking at

21   how these stents work and prevent people from having

22   failures, and they include measures of narrowing of

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       180


1    the artery.

2                     And we do have a long history of well

3    designed studies with good follow-up that

4    demonstrates angiographic outcomes actually very

5    good, and when we employ these kinds of outcomes, we

6    can reduce the sample size substantially and I think

7    still do something there, but we have to go through

8    the classical analysis that will support surrogacy

9    for these endpoints overall.

10                    Right now, what some companies are doing

11   is, they are trying to either go through a U.S.

12   dominant approach, which would be to try to do a

13   large scale equivalency trial at the FDA or go to

14   Europe where the bare metal stent is not being paid

15   for by any third party payers, and you can still do

16   a bare metal stent study.

17                    So the drug eluting stent still can be

18   randomized against a bare metal stent, and there's a

19   lot of kinks in these approaches, and they're all

20   trying to work out both in collaboration with

21   notified bodies in Europe as well as the FDA, but I

22   think that this is kind of the current status right

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        181


1    now, and I think we'll work ourselves out a little

2    bit better.

3                      I just want to spend the last few

4    minutes on potentially other applications overall,

5    and this is very speculative.              So I don't want to

6    say that this is proven at all, but I think that

7    with the advent of drug-eluting stents we can

8    actually get into completely new uses of these

9    little vehicles.

10                     To me, and I think to others, now that

11   we've essentially solved restenosis to some degree,

12   and I think we have largely, maybe we can start to

13   do things that make sense.            As interventional

14   cardiologists, we have never really helped extend

15   anybody's lives.        We basically make them feel better

16   when they play the 18th hole, or maybe they can

17   walk, you        know, 18 without using a cart.         We make

18   their quality of life better, and that's really what

19   angioplasty does.

20                     But still, almost a million people a

21   year die of heart attacks, and heart attacks occur

22   because of plaque ruptures, not at the sites where

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       182


1    blockages occur.       Usually they don't rupture, but at

2    sites that we don't treat, the ones that don't cause

3    obstructions.

4                     Well, we analyzed a variety of different

5    locations for these MIs, and this is my fellow John

6    Wang who had done this, and we found that the

7    distribution of MIs is mainly in the LAD and RCA if

8    we look at a consecutive series of a couple of

9    hundred patients at the Brigham, for example.

10                    And interestingly enough, there seems to

11   be some clustering.        That is, we can see if you look

12   at the LAD most of the MIs occur in the first couple

13   ten, 20, 30 millimeters of the artery itself, and

14   that's been kind of observed by a lot of people for

15   a while.

16                    If we apply a continuous frequency

17   distribution curve to the location in the LAD, for

18   example, of where these occur, we can see that about

19   80 percent of the MIs occur in the first 30

20   millimeters of the vessel itself.

21                    So the notion might be that we actually

22   have vulnerable hot spots in the artery.              Not

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       183


1    actually vulnerable hot lesions, and that we don't

2    have to really try to search out to find the plaque

3    that's going to rupture tonight.              Just use some

4    basic shoe leather epidemiology and say that this is

5    where the heart attacks occur.

6                     And if you are to look at the other

7    notion that once you put a stent in the artery, the

8    neomyplasia that occurs there or the scar that

9    happens makes it impossible for atherosclerosis to

10   grow anymore.      I mean, you have basically ruined the

11   fertile ground of atherosclerosis, and we have good

12   evidence for this.

13                    We can actually take arteries and remove

14   their ability to have plaque rupture by just putting

15   a stent there, and hopefully if we have a stent that

16   reduces restenosis, we can have a nice, thin layer

17   of neomyplasia and basically prevent that segment

18   from ever having an MI.

19                    So you know where I'm going on this one.

20    If we are to actually look at the instantaneous

21   probabilities of restenosis overall and apply a

22   variety of different simulated models, this is a

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       184


1    model for an eight millimeter single stent.                We have

2    them up to three or four stents now.

3                     We can see that the placement of a

4    stent, and it's eight millimeters subsequent, can

5    actually reduce -- we can actually optimize and find

6    where to place the stent.

7                     Well, to make a long story short, our

8    initial analysis has suggested that with the use of

9    two stents, a 28 and 23 millimeter stent, we can

10   reduce someone's MI risk by almost 50 percent, just

11   placing them in the proximal LAD and in the proximal

12   right coronary artery.

13                    Now, if you're a diabetic with three

14   vessels, it's easier.         MI risk is something like --

15   it was in the Berry study -- which was 70 percent of

16   five years or your fatality risk is close to 30

17   percent of five years if you're diabetic.              A 50

18   percent reduction in MI could be a substantial

19   thing.

20                    So I think that what you're going to see

21   is a wide expansion of these new stents with anti-

22   restenosis therapies to potentially prevent heart

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       185


1    attacks in the future, and how we get to those

2    patients I think will be the $64,000 question, and

3    how we utilize other diagnostic approaches such as

4    imaging techniques I think will be quite

5    interesting.

6                     So let me just conclude with our

7    experience so far with drug-eluting stents.                Drug-

8    eluting stents can definitely reduce restenosis, and

9    right now the Level I evidence is for the CYPHER

10   stent or rapamycin, and there's Level II evidence

11   and hopefully Level I pretty soon for paclitaxel.

12                    The long-term effects at this point

13   appear not to be problematic, that is, we do have

14   data out to three years          for the first in man, two

15   years for this RAVEL study done in Europe, and one

16   year for the SIRIUS study, and we see no catch-up

17   phenomenon.      We see no later aneurism formation, and

18   we see no late thrombosis problems.               So far it is

19   almost a dream come true.

20                    Other drugs are certainly going to work.

21    There's no question that with the wide formulation

22   of the polymer, which I think is the key component

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           186


1    here, drugs that we always thought should have

2    worked that didn't in the past are now going to be

3    given a second chance, and they include paclitaxel,

4    rapamycin, and possibly even other basic and

5    inexpensive therapies, such as steroids.

6                         Finally, cost effectiveness, which I

7    didn't review here, actually looks quite good, and

8    that's because restenosis is a costly event, and

9    even at the prices that are being charged now for

10   the Cypher stent, they're still cost effective, and

11   hopefully with more approvals of proven therapies

12   the prices will come down, which is what's important

13   for most patients overall.

14                        And I think ultimately drug-eluting

15   stents will be used for other functions and

16   indications in the future, including potentially to

17   take a bite out of            MIs in the future.

18                        And I'll stop there.         Thank you.

19                        (Applause.)

20                        DR. FEIGAL:     Well, I think you'll agree

21   with me this morning has really been a tour de

22   force.           I think almost every type of therapeutic

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           187


1    product has been mentioned in one respect or

2    another.

3                         We've run a little bit over time.             So if

4    you have questions, seek out the speakers during the

5    break.           We will reconvene at 11:30.

6                         (Whereupon, the foregoing matter went

7                         off the record at 11:15 a.m. and went

8                         back on the record at 11:33 a.m.)

9                         DR. HUSSAIN:     Good morning.       We are

10   ready to start the second session on preclinical

11   challenges.          Please take your seats.

12                        We had planned for four presentations on

13   different issues with respect to preclinical

14   challenges, and these presentations are roughly

15   about 20 minutes.           So if we get started on time,

16   we'll have lunch on time.               And I was told that if we

17   don't start on time, lunch is on yourself.

18                        (Laughter.)

19                        DR. HUSSAIN:     My name is Ajaz Hussain.

20   I'm with the Office of Pharmaceutical Science at

21   Center for Drugs, and I'd like to welcome our first

22   speaker, Dr. Leach.            He will be speaking on

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         188


1    preclinical development and considerations for

2    preliminary delivery of drugs approved for other

3    routes of administration.

4                     Dr. Leach.

5                     DR. LEACH:     Thank you very much.         And

6    thanks to Dr. Provost and the other organizers for

7    inviting me to speak.

8                     It's been an interesting morning.

9                     I'll go pretty quickly here because I

10   doubt that a      lot of people are interested in the

11   nitty-gritty details of preclinical sciences.                  So

12   I'll try and give you an overview of some programs

13   that have been successfully done, as well as some

14   ones that are in the development process, as well as

15   some that are in the early research stage, and you

16   get to choose which is which.

17                    Okay.   So to begin with the obvious,

18   maybe it's a good time to always state the obvious.

19   A lot of thought really needs to go into any of

20   these program a priori.

21                    The first thing you need to know is has

22   the drug been to the site before.              Particularly with

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       189


1    the lung, a lot of people have nebulized things

2    before and have gotten some amount of drug to some

3    areas of the lung, and that information may be very

4    valuable.

5                     Is the local concentration at the new

6    site higher than before?          Well, almost always yes.

7    We're trying to get more drug into the lung for

8    targeted lung disease, as well as new systemic

9    applications of drugs, existing drugs delivered by

10   the lung.

11                    The next thing is are the metabolic

12   pathways present in the new site.              There are usually

13   less metabolic pathways present, for example, in the

14   lung than there are in other tissues, like the liver

15   or the kidney or serum enzymes, that sort of thing.

16   But you have to make sure.           Maybe your drug is a PRO

17   drug by the IV route.         You have to make sure you

18   have the enzymes to metabolize it to the active

19   form.

20                    Are there new susceptible cell types?

21   We heard before that insulin is a growth factor, and

22   is a growth factor given in concentration of the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         190


1    lung which has never been there before an issue?

2                       Will new or existing excipients cause

3    problems?         This is a huge area.        For example, some

4    excipients         which are normally benign cause

5    bronchospasm in asthmatics or even normal

6    individuals.

7                       And, of course, our favorite, membrane

8    disruptors.         Those are usually a no-no in lungs.

9    You can get away with them in other areas, but

10   membrane disruptors in a lung, which may be part of

11   a normal formulation is a major issue.

12                      And of course, my personal favorite,

13   which is antibodies to proteins and peptides.                   Will

14   antibodies form?           Will they be neutralizing or

15   anaphylactic?         If they're anaphylactic, of course,

16   you're out of business, and if they're neutralizing,

17   to a large extent, then with repeated exposure your

18   dose must go up and, therefore, it might be

19   impractical.

20                      Okay.    So let's start out with a couple

21   of simple examples and work our way towards the more

22   complex.         First would be approve drug, Proventil

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       191


1    HFA.       It's called Air Amair (phonetic) in Europe,

2    versus the existing albuterol CFC products.                It was

3    the same drug.       It was in a different propellant.

4    It was the same amount of drug delivered, same

5    particle size distribution, but it did have some

6    improved dosing characteristics.              Okay?

7                     So if you look at Ventolin on the bottom

8    versus Proventil HFA, you can see there's a clear

9    difference there in what we call the plume, and in

10   fact, there's only about half the propellant in the

11   Proventil HFA as there is in Ventolin, and this

12   resulted in a warmer spray and with less force

13   behind it.

14                    The thought here was that there's a cold

15   freon effect that causes some asthmatics to have a

16   cough or mild bronchospasm, and then if you reduce

17   that, then you could get more drug in more

18   consistently.       Pretty simple.

19                    So to support that, we embarked -- this

20   is a 3M pharmaceuticals product, and we embarked on

21   a program and again went to the regulatory

22   authorities, and this is the first time there had

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           192


1    been a switch from CFCs to HFAs, and essentially

2    they said, "Gee, we have no idea what to do.                     Go and

3    do something and come back to us and we'll tell you

4    if it's okay or not."

5                         Hopefully from the talk we heard this

6    morning we won't be doing that anymore and we'll

7    have a lot better communication on new things in the

8    future.

9                         So we designed our own program, and it

10   basically was this.            It entailed, fundamentally,

11   what you would do with an NCE at the very beginning

12   stages, say, through Phase I, maybe early Phase II.

13                        And the studies we designed were

14   actually fairly complicated in the sense that we

15   included safety pharmacology in them, as well as

16   recovery periods, and tried to design very well

17   targeted studies to answer specific questions that

18   we thought of beforehand.

19                        There was an inhalation teratology study

20   in rats done, which of course was negative for

21   albuterol, and by and large unnecessary in our

22   minds.           But at that time reproductive studies were

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       193


1    in vogue in the '90s, and everybody wanted a

2    reproductive study on everything regardless of

3    whether there was an indication or not.

4                     Okay.   Just to pick out one clinical

5    study to prove the point that this was, indeed, the

6    same product as the old product, this clinical study

7    was a 12-week clinical study where half of the

8    patients were exposed to the HFA product and the

9    other half the old CFC product, and this is a

10   durational effect in terms of FEV, and I've actually

11   shown you the back half of this.

12                    The first half was when the yellow ones

13   are the HFA.      They had no difference in duration of

14   effect through the 12 weeks, but then we did a

15   split-off study where we took those patients who

16   were the CFC patients at the end of this 12 weeks

17   and then split them in half, continued one half on

18   the CFC and put the other half on HFA, and again, we

19   see no difference here in duration of effect.

20                    And of course, there were many

21   parameters involved in the study.              This is just one

22   of them.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         194


1                     So for this particular study compound,

2    then we had no preclinical surprises.                 We knew

3    exactly what the old CFC version produced in

4    animals, and we had no surprises in the animal

5    studies that we did conduct.

6                     We had no PK/ADME clinical surprises,

7    and we had no efficacy surprises.              So no further

8    preclinical studies were necessary, as deemed by the

9    developers, us and the regulatory authorities around

10   the world.

11                    Pretty simple, right?         Well, three and a

12   half years after we started this, we made a

13   submission, and about one and a half years later it

14   was approved.      So this was a five-year program, and

15   I think one of the simplest that's ever been done.

16                    If we go on to the next most complicated

17   one, this is QVAR.        It's also approved in about 40

18   countries now, versus the old CFC product.                 Here we

19   have the same drug, different propellant, a

20   different amount of drug, different particle size

21   distribution.      Therefore, it went to different

22   places in the lung, as I'll show you in a minute, as

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                        195


1    well as some improved dosing characteristics, which

2    I won't go into.

3                      Okay.    So here we're going to see a very

4    large difference then.           If you look at the old CFC

5    products, they were about three and a half microns,

6    which is actually fairly large for pulmonary

7    delivery.        Greater than 90 percent of it actually

8    went into the mouth, and less than ten percent went

9    into the lungs.

10                     Not only that, but you can see a big

11   difference here.          That doesn't even cover the large

12   airways which actually extend to the periphery in

13   two dimensions of the lungs as opposed to the QVAR

14   product, which is 1.1 microns, a very small amount

15   relatively speaking, only 30 percent in the mouth

16   and 60 percent in the lungs.

17                     And you can see that the lungs were

18   covered very well.         Well, this was terrific, except

19   it did raise some preclinical safety issues.                  This

20   drug is going to all the airways, as well as the

21   alveoli, and what are the safety consequences of

22   that?

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       196


1                     It should be great efficacy-wise, but

2    this did raise a lot of questions.               So we performed

3    the following preclinical program, which was, again,

4    sort of a modification of what you would do for any

5    NCE, range finding studies, 14-day studies, and then

6    a 12-month inhalation study.

7                     And the rationale behind the 12-month

8    study was that this could cause some endocrine

9    disruption in young animals, and there needed to be

10   some long-term exposure.          There was no scientific

11   rationale to speak of behind this, but nonetheless,

12   there were people who thought this was important.

13                    The other maybe more applicable

14   explanation for requiring such a long, hard study

15   was that it might have an effect on the developing

16   one on branching.       Again, there wasn't any real

17   precedence for this, but some people felt like it

18   was important, and of course, again, in the middle

19   '90s, being we conducted an inhalation teratology

20   study in rats, again, reproductive studies being in

21   vogue then.

22                    In fact, it was negative in that

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           197


1    teratology study, but because the class of steroids

2    is labeled as having reproductive effects, this

3    ended up with a label anyway.              So I'm not sure why

4    we did the study.

5                      Okay.   Well, let's take that product

6    then.        We did a preclinical program.             We showed that

7    it really wasn't any different once you understood

8    the dosing between the CFC and the HFA product.

9    What happens when you go to Phase I?

10                     And I don't really separate preclinical

11   from clinical very well.           They should fuse right

12   into each other and sometimes feed back.                  So, in

13   other words, if you set up your preclinical program

14   and you find clinical results in your early phases,

15   they should go back to the preclinical, explore

16   those differences and then come back to clinical,

17   and so forth, and have an exchange that way.

18                     So this is a prediction then of what

19   would happen.        If you give the beclomethasone to the

20   lungs, it's 100 percent bioavailable.                  It is about

21   20 percent bioavailable by the oral route.                   So if

22   you come up with these, you can do a projection here

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                       198


1    and say if you believe the dosing, if you believe

2    the deposition studies, then when you do your Phase

3    I TK study, if you give the same amount of Beclovent

4    100, which is the old CFC product, versus the HFA

5    product, you should get about 2.6 times as much in

6    the serum with the QVAR product.

7                     So we tested this hypothesis, and we

8    actually gave 400 microgram of the BDP, old BDP

9    against 200 and looked at the pharmacokinetics, and

10   you can see that, indeed, when you adjust for double

11   the dose here, it was about two and a half to one

12   ratio with the BDP-HFA being the yellow line here.

13                    Now, there's a couple of things you

14   might notice.      First of all, the Tmax happens

15   quicker with this than it does with the CFC, and

16   that's because of the oral contribution.              So this

17   actually did confirm not only by the AUC two and a

18   half difference, but also by looking at the

19   Cmax/Tmax values and showing that our hypothesis did

20   appear to be correct.

21                    Okay.   So then we're ready to go into

22   the clinic, and so we did a dose response

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          199


1    relationship between the QVAR and the old product,

2    got these lines, drew the equivalence there and saw

3    that it was as efficacious at about 2.6 times less

4    dose.

5                     So, again, this is fitting with our

6    preclinical, with our Phase I, and so forth.                    And in

7    fact, when you go on to long term clinical studies,

8    you can see breakthrough of asthma here, and you can

9    see the yellow line being the QVAR.               You can see

10   that at a two to one switch here there was actually

11   less breakthrough of asthma than there was with the

12   old product.

13                    So the safety parameter.             We looked at

14   many, but of course, urinary free cortisol is one of

15   the major ones, and so you worry about that kind of

16   dose being given, and is it different?

17                    When we looked at the urinary free

18   cortisone, this is the placebo, and these are the

19   different doses, and in fact, we found that the --

20   boy, I switched colors here, yellow and red, just to

21   see if you're awake.

22                    In this case the yellow -- oh, the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                             200


1    yellow is the HFA.           Sorry.     The CFC is the red, and

2    you can see that there was no additional safety

3    concern matching doses of 800 versus 800, even

4    though clinically 400 was equivalent to 800.

5                        Okay.   So once again, we had no

6    preclinical surprises in the two species.                     We were

7    able to predict the PK and the ADME clinical

8    results, and there were no efficacy surprises.                        So

9    there was no further preclinical studies required.

10                       Now, this program, again, took about

11   five and a half years to complete and another almost

12   two years to get registered once it was submitted.

13                       So even these simple cases have not

14   turned out to be so simple or cheap.

15                       So now let's move into some of the

16   things that are being worked on.                 You've heard a lot

17   about proteins and peptides and insulin.                     Everybody

18   is very, very excited, as are we because there are

19   just so many proteins and peptides that are being

20   explored now with so many exciting results, but they

21   have very serious delivery problems.                     They need to

22   inject.          No one wants that.

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                       201


1                     The time of action is too short.            Native

2    structures are just too susceptible to the

3    peptidases, and all of those severely limit the

4    application.

5                     Normally we're not allowed to show

6    animals in distress, but you know, this is not too

7    distressful.      This would be the same thing that you

8    would see in a human being.            You've probably heard

9    that the average diabetic may inject themselves 40

10   or 50,000 times during a lifetime.               No one likes

11   that.

12                    One of the things you probably don't

13   hear enough about is the fact that there are so many

14   borderline diabetics or diabetics who just plain

15   refuse treatment because they don't want needles,

16   period.

17                    And that actually is a very important

18   segment of population, in my opinion.

19                    Well, why don't you just swallow it?

20   Well, with the bioavailability orally of growth

21   hormone, you'd need about $120,000 a day and quite a

22   bit of eating.      That's not very practical.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           202


1                      So why not inhale?        People say it's too

2    hard.        It takes too much education, whatever.               The

3    fact of it is an 18 month old can use an inhaler,

4    sometimes even by themselves.

5                      Just to back up a little bit, too, on

6    what is the lung, the lung has 23 generations of

7    airways, and those airways' surface area would be

8    the equivalent of that towel thrown on the tennis

9    court, and the tennis court would be the equivalent

10   of your lung surface area, and if you talk about the

11   volume of lung surfactant, it's about 30 mils.

12                     So when people say they're worried about

13   high doses to the lung, if you have a very

14   dispersable, well aerosolized product, that whole

15   product can actually use that whole lung, and you

16   can imagine that a milligram or three milligrams --

17   whoa, five minutes?         He's vicious.         Okay.    I'm going

18   to have to skip some.          I think the introduction went

19   into my time.

20                     (Laughter.)

21                     DR. LEACH:     So you can see that even

22   though you're talking about a lot of surface area,

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                        203


1    and sometimes you're talking about three milligrams,

2    four milligrams of drug, when you imagine those

3    little particles spread over that tennis court, it's

4    really not a high concentration.

5                      In fact, the concentration you're given

6    by IV or Sub-Q is much greater than this.

7                      Okay.   I'm going to have to skip some

8    here.

9                      The rule of thumb though is that for

10   about two to five percent of the IV dose actually

11   reaches a lung.        That's not very efficient, not very

12   attractive.

13                     Five minutes.      So I'm going to have to

14   skip some of this.

15                     Let me give you a couple of examples

16   here.        Leuprolide, which you've already heard about,

17   is very, very limited by its side effects.                We

18   wanted to see if we could get inhalation

19   bioavailability to match an IV dose in this, and in

20   fact, this is a human clinical study, and I'm sorry

21   it didn't show up that well.             But we showed that we

22   could do a dose by IV injection and match that dose

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        204


1    by inhalation very well.

2                      As you know, the side effects can be

3    severe, headaches, and especially with these

4    implantable devices.          Once you get them implanted,

5    you're going to live with the side effects for a

6    very long time, as opposed to inhalation product

7    where you can titrate yourself down or even stop

8    temporarily.

9                      PulmoSpheres are our version of what

10   you've heard about this morning.               They're wonderful

11   materials.       They're hollow.        They're porous.

12   They're ultra low density.            They're able to get to

13   the deep lungs so that you can take advantage of

14   that huge surface area.

15                     They are actually made of lung

16   surfactant themselves.           DSPC and DPPC are natural

17   components of the lung excipient.

18                     And so what are the preclinical issues

19   here?        Well, again, there's larger lung

20   concentrations that are going to be seen from

21   leuprolide.       Lung doesn't normally see leuprolide,

22   but again, you get big doses spread over that tennis

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                            205


1    court, gives you reassurance that it's not going to

2    be too bad.        One must certainly do the work anyway.

3                       There's the antibody question, and then

4    there's the excipient question here.                    That's why

5    people choose excipients that are very compatible,

6    biocompatible.

7                       Okay.   One of my personal favorites I

8    want to spend a minute on is antibiotics for lung

9    disease.         It's so unattractive to give an antibiotic

10   either orally or IV for a lung disease that I'm

11   surprised that we haven't gone a lot further with

12   inhalation antibiotics than we actually have.

13                      Here's an example of one that's actually

14   on the market.        If you look at the blue lines here,

15   they're what normally happens when you inject it by

16   IV.      You can see that you get a nice, good curve

17   here.

18                      But when you go up and look at the lung,

19   the lung values here -- I'm not sure it's showing

20   up.      My angle is not good here -- it's about 20

21   micrograms per gram of tissue.               Okay?       This actually

22   is not too much approaching the MIC.

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                         206


1                      Now, if you look at the lung lavage, you

2    can see they're almost not detectable.                 Well, you

3    can take that same dose and give it by the lung and

4    you get none in the plasma.             You get a fairly

5    significant amount in the lung lavage, which means

6    that it's on the mucosal side where the actual bug

7    is, and a huge number in the lung.

8                      In fact, there is a line broken here,

9    and this number is 1,500 versus about 20 on this

10   line.        So you can see that if you just have a good

11   powder, and I emphasize you just can't put these

12   things in nebulizers and expect to get these kinds

13   of results because they're notoriously inefficient.

14   They don't get to the deep lung, et cetera.

15                     If we go on to a more sophisticated

16   study in dogs, this is an actual tox. study, PK

17   study and whatever kind of name we could put on it

18   to get to our endpoints.           We see the same thing.             We

19   could get a nice, good dose response relationship.

20   We get plasma half-lives at 28 hours, and we get

21   lung tissue half-lives of 19 days.

22                     And I think the important point here is

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                          207


1    we can get four orders of magnitude difference

2    between lung and plasma.          So the hypothesis here is

3    that if we want two to three times the MIC, the

4    plasma levels are likely to be undetectable, and the

5    plasma levels, again, are the limiting side effect

6    of this particular drug, and so those have the

7    potential of going completely away.

8                     Skipping some of the good stuff, I added

9    this in because of the mention of PEG.                 PEG insulin

10   is a very important thing right now.                  If you have a

11   long acting PEG, then you can provide basal levels

12   to diabetics, and I think most of Type Is and about

13   20 percent of Type IIs actually require some basal

14   injection.

15                    And even when the inhaled product comes

16   out, they're still going to require that unless we

17   come up with a longer acting.             We think PEG is one

18   of the ways we can do this.            PEG is a really

19   interesting.      They're very, very safe.             They've been

20   in many approved products, and the PEG is actually a

21   long chain here, and this would be the drug.

22                    And not shown here is the hydrodynamic

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                           208


1    diameter.         There's actually about five to ten times

2    the molecular weight of water that actually

3    surrounds this.          So it protects it from the immune

4    system.          It also protects it from degradation in the

5    lungs, et cetera.

6                        And I'll just show you one piece of data

7    on that, and that is glucose suppression.                 Again, I

8    don't have time to go into a lot of details.                    For

9    those of you who know about insulin and know about

10   glucose suppression, if we give normal insulin, we

11   normally can suppress glucose that would go down to

12   these sorts of levels, and it lasts about two hours

13   and then comes back up.             You saw a similar graph

14   earlier today.

15                       If we use PEG insulin here, then we can

16   go down, suppress it, and stay out here.                 And we've

17   gone out to eight to ten hours, and we presented

18   this at the ADA meeting about a month ago.                  So we're

19   very hopeful that we can come up with a pegylated

20   insulin that might last as much as ten hours and get

21   people through the night.

22                       Okay.   In summary, a route changed

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          209


1    inhalation can offer fast onset.                 I didn't get a

2    chance to really talk about that, but, for example,

3    nicotine by inhalation only takes six heartbeats to

4    reach the brain.          So things like fentanyl might be

5    very interest for instant relief.

6                        Higher bioavailability than some other

7    routes.          Freedom from ejection, less side effects.

8    The preclinical requirements should be unique to

9    each new change in route.              I don't believe there's

10   ever going to be a cookie cutter approach to these

11   issues.          It needs very close work with the

12   regulatory authorities.

13                       Preclinical programs should stress the

14   exploration of known differences, not

15   unsubstantiated speculation or not what's

16   particularly in vogue.             These things add up to a

17   fear of the unknown and unreasonable preclinical and

18   clinical requirements that keeps many new drugs from

19   really happening, especially for the non-blockbuster

20   category drugs.

21                       I can't tell you how many conversations

22   we've had with drugs that we know we can make

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       210


1    significantly better or we can give by the pulmonary

2    route and really improve and meet an unmet clinical

3    need, but if they're in the 50 to $100 million

4    range, nobody wants to touch them.               There are many,

5    many, many like that, and it's really heartbreaking

6    to know that we could do such a better job than

7    what's out there, but the economics are driving it.

8                     And the fear of the unknown, which is my

9    last slide.

10                    (Laughter.)

11                    DR. LEACH:     Thank you.

12                    (Applause.)

13                    DR. HUSSAIN:     Thank you for the

14   excellent presentation.          I'm sorry I had to show you

15   the five minute page.

16                    The next presentation is entitled

17   "Protein Delivery from Implantable Devices:

18   Challenges and Opportunities," to be presented by

19   Bill Van Antwerp, Vice President and Chief

20   Scientific Officer of Medtronic MiniMed.

21                    DR. VAN ANTWERP:       Well, thank you.        I'd

22   like to thank Miriam and the FDA for inviting us

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           211


1    here to tell you a little bit about our view on

2    protein drug delivery.

3                         You've heard a lot this morning about

4    things that might happen in the future.                   Bob Langer,

5    in particular, gave us a vision that's incredibly

6    long seeing.

7                         I'm going to tell you a little bit more

8    about the grunt work that you have to do in the lab

9    to make some of these products possible.

10                        Okay.   So why protein drugs and why

11   protein drugs and devices?               Proteins are becoming

12   increasingly important for a variety of disease

13   states:          diabetes, which is near and dear to our

14   heart and everyone else's; cancer; cardiovascular

15   treatments; inflammation; HIV/AIDS; Hepatitis C, for

16   example.

17                        Those are drugs that are now coming or

18   now approved.           There's a variety of drugs from a

19   variety of companies also coming on line that are

20   proteins.          Proteins need delivery, as we have all

21   heard.           They need delivery.       They're not very

22   bioavailable.           They get denatured.          They get

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         212


1    hydrolyzed.      They get degraded by enzymes, and if

2    those escape some of those routes, they're not

3    absorbed very well either due to their size or due

4    to their polar or charged distribution.

5                     There's a variety of companies

6    developing novel technologies.             We just heard about

7    pulmonary delivery.        There's a variety of depo

8    injection and other technologies.              We're going to

9    talk about the old fashioned way, which is basically

10   delivering through the skin through a subcutaneous

11   or interperitoneal infusion using mechanical

12   devices, pumps.

13                    Bob Langer showed you something like

14   this slide earlier.        This is a classic case where we

15   have a drug that has about a six-hour half-life, and

16   if I deliver it via injection and then I have in

17   blue here a therapeutic range, I need to give

18   another injection 12 hours later when I'm just at

19   the nadir of activity.          Well, I have to deliver 14

20   times as much drug.

21                    More importantly, the side effects are

22   often due to the peak concentrations.                 Enzyme

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                       213


1    activation is incredibly important.               In fact, in

2    Colin Denney's group at M.D. Anderson he's shown

3    that high concentrations of drug actually deactivate

4    enzyme or activate enzymes that deactivate the drug,

5    and I can deliver via mechanical devices drugs with

6    a perfect matching of the drug to the therapeutic

7    range.

8                     Today parenteral delivery of drugs are

9    done via two old routes, IV administration,

10   subcutaneous injection.          Two routes that have had

11   some success, continuous subcutaneous infusion via

12   mechanical pumps and continuous interperitoneal

13   infusion, both of these mostly for insulin, but

14   they've been used for a number of other compounds as

15   well.

16                    And we've heard a lot about subcutaneous

17   depos, PLGA microspheres, PEG attached peptides,

18   micro emulsions, pulmonary delivery, and also

19   there's some new routes, intrathecal and

20   intraparenchymal delivery.

21                    Medtronic has a significant business in

22   intrathecal delivery of small molecules, morphine

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        214


1    and baclofen, although just recently we're starting

2    to look at those routes to get drugs into the brain,

3    cross the blood-brain barrier using proteins.

4                     Well, what are the challenges?           Well,

5    Bill Clinton said it about the economy.               Here I

6    would tell all of you involved in drug device

7    combinations that it's the formulation.

8    Formulation, formulation, formulation.

9                     Old challenges, formulation stability,

10   chemical stability, clearance issues in the body

11   once you inject it, but when you start to give drugs

12   by mechanical devices, you run into two new

13   problems.

14                    One is physical stability.           If you pull

15   a syringe full of insulin out of a bottle and I

16   inject it Sub-Q, I don't have to worry too much

17   about the physical stability of that insulin.

18                    If I put it in a mechanical device, I

19   have to worry a lot.         I also have to worry about

20   some PKPD issues because now I'm giving it

21   continuously in a trial that we just finished, the

22   Phase I trial.      We had a dose escalation study

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                          215


1    planned.         It turned out to be a dose de-escalation

2    study because when I gave the drug continuously, it

3    was much more effective than we had thought by

4    continuous or by multiple injections.

5                       We found that patients had to down-dose

6    rather than up-dose, and we have to think a little

7    bit about toxicity in a different way.                  This is not

8    systemic toxicity, but if your formulation isn't

9    right, we need to worry about localized site

10   reactions.        If I have got an injection catheter

11   that's in the subcutaneous tissue and it's supposed

12   to be there for three days, I have to make sure that

13   the formulation of the drug is suitable for those

14   three days of delivery.

15                      Regulatory hurdles, let's not reinvent

16   the wheel.        We build devices.         The device physics

17   are what they are.          If we build a pump, it turns out

18   every time we want to put a new drug in the pump we

19   have to prove that the pump pumps again, even though

20   we've shown in the laboratory that it pumps with a

21   wide range of viscosities.             It's always an

22   indication that we have to prove that the pump pumps

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                         216


1    again.

2                     The same thing is happening with drug

3    chemistry.       We're developing a prefilled insulin

4    cartridge.       The insulin degradation chemistry has

5    been well known since the late 1920s.                 Yet we have

6    to show that the impurities in our insulin are

7    exactly the same as the impurities in all the other

8    insulin formulations that have ever been developed.

9                     The same with drug packaging.           We try to

10   use for pumps the kind of packaging materials that

11   people have been using for the drugs for a long

12   time, but again, we need to show stability.

13                    There are, however, two areas where we

14   need to pay much more attention.              One is pump-drug

15   interactions and drug physical stability.                What we

16   like to say in our laboratory is that when God

17   invented insulin, She didn't design it to be stable

18   for 90 days at body temperature sloshing around in a

19   metal can.

20                    These are the kind of things that

21   traditional drug systems don't need to think about.

22   We need to think about physical stability.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                        217


1                      Stability in pumps has two components:

2    chemical stability, which in our hands looks very

3    much like stability in primary packaging.               We don't

4    see chemical changes in formulations that we don't

5    see at the same temperature in primary packaging.

6                      Physical stability is important.            Why is

7    it important?        I'll show you in a minute some

8    results from some studies, but physical stability

9    generally leads to things like soluble aggregates,

10   Soluble aggregates are well known to lead to

11   antibody issues that you don't see, for example,

12   with noncontinuous infusion.

13                     There have been a wide variety of

14   measurements of physical stability in the protein

15   business.        Every protein company has five or six in

16   their labs.       None of them seem to give you exactly

17   the same results, at least in our hands, and I want

18   to propose some testing that I think makes sense in

19   a lot of situations.

20                     People have looked at turbidity,

21   concentration changes, fluorescent spectroscopy,

22   microcalorimetry, and a whole variety of other

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       218


1    things.

2                     As I said before, chemical stability

3    determined by the molecule, by the formulation.                   One

4    important point to note, that relatively

5    straightforward formulation changes can affect

6    stability, and what we have seen in devices ranging

7    over a wide range of molecular types, interferons,

8    insulin, interleukins, and a variety of peptides,

9    large and small, is that the stability in the device

10   is pretty much the stability in the primary

11   packaging.

12                    Physical stability, however, isn't.              It

13   depends on a number of things.             Probably most

14   important:       the physics of the device, whether

15   they're sheer or compliance in the system; what the

16   materials of contact are, Teflon, titanium,

17   polyolefins, silicone oil.           All of these are common

18   in medical devices.

19                    Agitation is incredibly important, as is

20   body temperature storage.

21                    We believe that in physical interactions

22   there are a couple of steps that are important.                   The

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         219


1    first is absorption.         The next is denaturation,

2    typically on the surface, and we believe that a lot

3    of the story in terms of physical stability of

4    proteins and devices can be told by looking at

5    partially unfolded intermediates.

6                     Tony Fink's group up at U.C.-Santa Cruz

7    has been a leader in this idea, and we concur with

8    some of what he has done.           Once we get these

9    partially unfolded intermediates they lead to

10   aggregation on the surface, which then leads to

11   aggregation in solution.

12                    We have a model here.         Part of this

13   model was originally proposed by Bob Langer many

14   years ago now, but we have a protein.                 It sticks to

15   the surface, then unfolds, falls back into solution,

16   forms aggregates, and the model is autocatalytic.

17                    And we test this in the laboratory now.

18   Five years ago it took six months to a year to

19   understand all of the physical stability of a

20   protein in a pump.        We can now test it in a few

21   hours or a few days.

22                    And basically this is the autocatalytic

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                           220


1    curve.           We put the protein in a 96 well plate with

2    some Thioflavin T.            Thioflavin T is a fluorescent

3    molecule that only fluoresces when it's bound to

4    aggregated proteins.

5                         We look at the fluorescence as a

6    function of time, and we curve fit this to the

7    autocatalytic curve model.               You see that the

8    correlation coefficient is .99, which is quite nice.

9                         What's the point of all this testing?

10   Well, the point of all this testing, one point here

11   is to look at the physical stability in contact with

12   a number of materials so that when you're designing

13   devices you always have to design with materials

14   that are available.            You know, FDA doesn't like to

15   see new chemical entities particularly that might

16   end up in your drugs.

17                        So here we've taken a formulation of

18   insulin and compared it in the same experiment with

19   Teflon, polyethylene, glass and titanium, and what

20   you see is that the Teflon is by far the most

21   susceptible to aggregation, whereas glass and

22   titanium are quite nice.

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         221


1                      And this difference here, this

2    difference between 50 hours and 150 hours, we have

3    good correlation to stability in clinical testing in

4    pumps.

5                      Similarly, and this is the formulation

6    issue with a different compound, different

7    formulation, we have two drug substances, a new one

8    and an old one, and we formulated them two different

9    ways.        In one case we simply dissolved the protein

10   at high pH and then pH'ed it down to pH 7.4, and

11   then the one we call low pH, we took the drug

12   substance, dissolved it in acid and then took it up

13   through the PI to the appropriate pH.

14                     And you see that even though the end

15   formulation is exactly the same by any chemical

16   tests that we can do, the physical stability when I

17   start out at low pH versus when I start out at high

18   ph, this is a factor of four or five more stable,

19   which has significant implications in the clinic.

20                     Okay.   So where does that leave us?

21   Formulation, formulation, formulation.                 We're a

22   device company.        We're not a drug company.           All of

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                       222


1    the products that we put in our devices come from

2    biotech companies, typically not big PhRMA, although

3    insulin is obviously not the case.

4                     If you want to talk to someone about

5    devices, if you're a PhRMA company or if you're a

6    device company wanting to talk to PhRMA, start with

7    the formulation.       There are multiple interactions

8    that you need to study.          Control of the material

9    interface is the most important thing, and what's

10   very important from the regulatory standpoint,

11   device design and formulation need to work together

12   and be regulated together.

13                    We always talk about our devices

14   breaking proteins.        This is a picture of a protein

15   that actually broke the device.             This is a seal.

16   This is a titanium seal, and you can see the

17   titanium here.      This is a deposit of insulin

18   crystals that formed on this seal, and this was ten

19   years ago now, on an implantable pump, and you see

20   that this seal worked perfectly fine, except where

21   this crack was.

22                    This crack allowed actually insulin to

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          223


1    flow out.        The seal no longer worked.            This caused

2    us a lot of headaches.           It turns out that it was a

3    materials and formulation issue which has now been

4    solved thankfully.

5                      So in conclusion, interactions need to

6    be managed.       They   need to be understood.            Pump

7    design and formulation need to work together.

8    Combination product components can be evaluated

9    separately using historical data.               We have pumps.

10   We know how they work.

11                     However, we need to pay appropriate

12   attention to the drug-device interactions.                  Those

13   are the things that are critical.

14                     And when I talk about "we," I really

15   mean "they."       This is the Protein Formulation and

16   Stability Group at Medtronic MiniMed.                  They only let

17   me in the lab now to get coffee for my coffee

18   machine.

19                     (Laughter and applause.)

20                     DR. HUSSAIN:     Our next topic is

21   developing a local drug delivery combination product

22   for postoperative atrial fibrillation, preclinical

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                           224


1    challenges, by Dr. Kevin Skinner.

2                         DR. SKINNER:     Thank you very much, and,

3    Miriam, thank you for organizing this conference.

4                         We've been working on this project for

5    two years.          So I'm going to discuss the development

6    process, and at this point in time we've gone from

7    concept to a bench research level, and we're at a

8    preclinical research level.

9                         And the concept actually came from

10   clinicians and marketers, and they brought that idea

11   to us.           We started evaluating this concept of

12   marrying biomaterials with an old drug entity called

13   amiodarone, and then we took it into preclinical

14   research to get a proof of concept.

15                        In the near future we'll take it into a

16   preclinical development phase, which will enter into

17   a history design, and then move it into the clinic.

18                        So postoperative atrial fibrillation,

19   it's a kind of tachycardia that you see in patients

20   following CABG and bowel surgery.                  It happens around

21   20 to 30 percent of the time, and usually happens

22   within three to five days, but it can happen up to

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        225


1    two weeks following surgery.            It increases the

2    length of stay for the hospital or for the patient

3    up to 1.5 days, and people have assessed or have

4    followed this from an economic standpoint, and it

5    can cost the patient about 8,000 more dollars.

6                     You get a decrease in cardiac output.

7    You get an increase in stroke due to stasis in the

8    atria, and there have been prophylactic treatments,

9    but they're not widely accepted.              Amiodarone is a

10   drug of choice for treating postoperative atrial

11   fibrillation, but it requires at least for oral

12   dosing seven days prior to surgery, and if you use

13   the IV formulation, you have some severe side

14   effects.

15                    So amiodarone is probably the most

16   widely used antiarrhythmic for clinical use.                 Its

17   label indication is for ventricular tachycardia and

18   for ventricular fibrillation and super ventricular

19   tachycardia.      However, it is used off label for

20   atrial fibrillation.         In fact, it's the most common

21   use of or amiodarone is the most common use for AF.

22                    It's a Class III drug, which means that

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       226


1    it increases the action potential, and increases the

2    effective refractory period.

3                     But, as I said, one of the drawbacks

4    with the drug is it has high toxicity for pulmonary,

5    and it also causes bradyarrhythmias following

6    loading doses.       The systemic doses you actually have

7    to load up to gram quantities within the first week,

8    and then it tapers down to between 800 and 400

9    milligrams.

10                    So the thought was:        could we deliver

11   that drug locally?        And there was some basic

12   research done by Ayers and Zipes, where they locally

13   delivered the drug into the pericardial sac.

14   However, they had to load it for three hours, and

15   they looked at several doses.             They looked at the EP

16   parameters following the administration of

17   amiodarone locally, and they measured myocardial

18   drug levels.

19                    So what we have here is we have both the

20   atrial refractory period and also the dose level,

21   and so what you see is increasing from the control

22   up to the five milligram dose you get an increase in

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       227


1    atrial refractory period and also an increase in

2    tissue levels.

3                     And the effective doses or therapeutic

4    levels that they saw in the tissue was between 20

5    and 120 micrograms per gram of tissue, and this is

6    the data that you see in humans, patients that have

7    died, and they have posted their tissues and have

8    posted the level of amiodarone.             So, you know,

9    that's the therapeutic level of drug.

10                    And in these animals they only had a

11   small amount of trace drug that was found following

12   the dosing for three hours.            So the thought was

13   could we find a biomaterial that we could put

14   amiodarone into it.

15                    Genzyme had a collaborative research

16   with a company called Focal and subsequently

17   acquired the company, and the technology is a

18   bioreabsorbable PEG based hydrogel.               It's actually

19   approved in the United States and in Europe for lung

20   sealants for pulmonary leaks, and in Europe it's

21   approved for dural sealants.            It's tissue adherent.

22   It's compatible with drugs and biologics.               You spray

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       228


1    it on or you can drop it on as a liquid and you

2    photopolymerize it with light.             The product can be

3    tailored to whatever application you'd like to use

4    it for.

5                     So the questions we wanted to answer

6    were could amiodarone be delivered via this tissue

7    adherent hydrogel and could we get effective doses,

8    and can we reduce the amount of drug levels, and

9    would it not be systemically found?

10                    And can these drug levels cause an EP

11   effect that would prevent AF?

12                    So the product characteristics were

13   could it adhere to cardiac tissue.               We have a

14   pumping structure.        So that was a challenge for us.

15   Could we deliver the drug locally?               Were we able to

16   reduce the level of drug?           And could we deliver it

17   up to 14 days?      And was it compatible with cardiac

18   tissue?

19                    So before we even went into doing animal

20   studies, we wanted to make sure that there wasn't a

21   drug-device interference or a device-drug

22   interference.      So we wanted to make sure that the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           229


1    hydrogel did not affect the amiodarone.                   So we did

2    HPLC mass spec analysis and demonstrated that the

3    drug was not affected by the hydrogel or its

4    individual components.

5                         We also made sure that the amiodarone

6    didn't affect the in situ polymerization of the

7    hydrogel or other properties of that hydrogel, and

8    we could load up to five percent of amiodarone into

9    the gel without affecting those properties, and then

10   we demonstrated in vitro release that we could get

11   up to two to three weeks of drug being delivered out

12   of that hydrogel, up to one percent of the

13   amiodarone being loaded into the hydrogel.

14                        So in the first study that we did

15   preclinically, we implanted the hydrogel amiodarone

16   at a half a percent and one percent onto the canine

17   heart.           We came back seven days later, looked for

18   levels of amiodarone in the cardiac tissue and also

19   its active metabolite desethyl-amiodarone, and we

20   also looked at other tissues in the body, lung,

21   kidney, and liver, and also urine and blood, and

22   then we observed for any adverse events in animals.

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       230


1                     So after seven days at the half percent

2    we got around 64 micrograms per gram of tissue in

3    the half percent loaded gel, and the one percent gel

4    gave us around 230 micrograms per gram of tissue.

5                     The gel itself had only eluted only 30

6    percent of the drug at day seven, and we found four

7    to six percent of the metabolite desethyl-amiodarone

8    in the treated tissue, and there were no measurable

9    drug levels in the lung, liver or kidneys.               We did

10   see some in the cardiac pad around the pericardium,

11   but it's known that amiodarone, because it's fat

12   soluble, will reside there.            And there were no

13   adverse events seen in any of the dogs.

14                    So our next study was, you know, we can

15   deliver the drug, and the amount of drug that we can

16   deliver, would it have an EP effect on that?

17                    So we looked at four groups:         just the

18   hydrogel itself, the hydrogel loaded at half a

19   percent and one percent, and then we did a surgical

20   control group.

21                    And we measured EP parameters

22   preoperatively, postoperatively, three to five days,

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       231


1    ten to 14 days, three to six weeks, and collected

2    the tissues for drug level.

3                     So what I have here is a chart that

4    shows pre-op.       Basically there's no difference

5    between any of those groups.

6                     Immediately after implanting the

7    hydrogel or the hydrogel and drug, you see an

8    elevation in EP, but the sham group also shows an

9    elevation.       So just the act of surgery increased the

10   atrial refractory period, but by day three and five

11   we see a significant increase in the treated group

12   of almost a 50 percent increase in the EP in the

13   treated group, and by day 14 you see the elevation

14   of the EP relative to the control group, and we also

15   see the effect out to three weeks.

16                    And then when we harvested tissue at

17   three weeks, we had therapeutic levels of the drug

18   within that tissue.

19                    So what we had shown in the preclinical

20   research aspect of this project is that we're able

21   to deliver amiodarone to the cardiac tissue at

22   therapeutic levels, which is significantly lower

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       232


1    than IV and PO routes.          As I said previously, you

2    have to deliver gram quantities of amiodarone orally

3    or milligram quantities IV.            But for us, we were

4    delivering milligrams, 16 milligrams, 32 milligrams,

5    and we were delivering it once over a three-week

6    period.

7                     And all of those studies that we have

8    done so far have shown no systemic levels of the

9    drug other than where we placed the material.

10                    And the product has been well tolerated

11   in all of the animal studies, and we demonstrated

12   that we were able to elevate the effective

13   refractory period, which is indicative of the proof

14   of concept to reduce the incidence of AFIB.

15                    So where are we today?          We're getting

16   ready to plan the preclinical development strategy,

17   and one is to leverage the existing data from

18   FocalSeal.       It has been approved in the United

19   States and also in Europe, and the amiodarone has

20   been approved for IV and oral formulation, and

21   there's a generic form out there already.

22                    So what we'd like to do is bridge those

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       233


1    existing data that's out there and just do studies

2    that are necessary to really evaluate it for the

3    specific use we're using.

4                     In this chart, you know, when we were

5    coming up with a strategy, we were trying to figure

6    out, you know, was this going to be ruled a device

7    or is it going to be ruled a drug, and somebody in

8    our regulatory department came up with this cartoon.

9                     And so if you look, what you have is the

10   drug, the potential drug, and the way of delivery,

11   and what you're looking at is a generic drug which

12   is amiodarone or a new indication or a new drug

13   entity, and if you, you know, look at how it would

14   be delivered, whether it be chemically modified or

15   would it be a depo effect or does it also have a

16   device action.

17                    So when we talked about drug eluting

18   stents today, you either can look at it as a dip

19   coated stent where it was a generic drug and it had

20   a device action and it was ruled by CDER or whether

21   it's a drug coated stent with a polymer that's being

22   ruled by CDER or, in our particular situation, we

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          234


1    have a generic drug, and it has a depo effect.                     So

2    it's being ruled by CDER.

3                     So, you know, what are the bridging

4    studies that we think we need to do?                  One is to look

5    at the long-term degradation of this product on the

6    cardiac tissue, and what are the acute toxicity

7    issues of placing a biomaterial on the heart with

8    the drug being delivered to the specific part of the

9    heart?

10                    And then what are the temporal drug

11   deliveries?      We've only looked at very short-term

12   delivery of that drug, and we need to look at long

13   term.

14                    And then we would do confirmatory EP

15   studies once we finalize the formulation.

16                    So in summary, post AFIB is a serious

17   unmet medical need which may benefit from the

18   advances in therapeutics that are delivered at the

19   time of surgery.       The combination product of

20   amiodarone and a synthetic adherent PEG based

21   hydrogel shows promise for safety and efficacy in

22   preclinical models.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                       235


1                     We'd like to leverage prior studies and

2    perform appropriate bridging studies that should

3    provide facilitated regulatory approval of this

4    drug-hydrogel combination.

5                     And this combination product is a good

6    example of a device/drug combination with a primary

7    pharmacologic mode of action.

8                     Thank you very much.

9                     (Applause.)

10                    DR. HUSSAIN:     Well, I was conflicted.             I

11   wanted to keep everybody else's time on the thing so

12   I could use all of the time.

13                    (Laughter.)

14                    DR. HUSSAIN:     No, what I'd like to do is

15   sort of in some ways connect the various

16   presentations that occurred this morning and also

17   hopefully help set the tone for the afternoon

18   discussion on regulatory.

19                    Although I'm from FDA, I'm not actually

20   taking a regulatory perspective, but more of a

21   scientific, broad, almost an academic perspective.

22   So the title of my talk that I've selected is

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                            236


1    different from what's in the brochure.                 I

2    essentially would like to sort of take a step back

3    and summarize for you some of the discussions this

4    morning and present to you the concept of quality by

5    design, which I think is a preclinical opportunity

6    to address many of the challenges.

7                      In this session we had three wonderful

8    presentations before mine, and we looked a

9    preclinical challenges with respect to pharm tox and

10   the need for doing additional pharm tox studies when

11   there is a route of administration change or when

12   there is a potential for change in exposure, and the

13   exposures may lead to or trigger some safety

14   concerns.        And I think if there is a better way of

15   addressing that, that would be a step forward.

16                     And the second presentation was, I

17   think, very important from my perspective to sort of

18   highlight the importance of physical stability and

19   in some ways I have sort of built some information

20   on that from my perspective also.               And I think

21   physical stability is a gap in terms of our ability

22   to analyze, do testing, which is proper and

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701        www.nealrgross.com
                                                                       237


1    relevant, and I think there's a significant

2    opportunity for collaboration there.

3                     And then finally we had a presentation

4    on local drug delivery and how does one sort of not

5    only start with in vitro methods that start

6    screening interactions as well as moving towards a

7    methodology that sort of demonstrates the local

8    effect, and local effectiveness is also a

9    significant challenge for us as we move forward.

10                    So to sort of summarize some of the

11   discussion and looking at quality by design

12   concepts, what I thought I'd do is share with you

13   the current FDA initiatives.            This workshop is

14   focused on the initiative as Dr. McClellan talked

15   about, improving innovation in medical technology.

16   This is the workshop sort of starting this

17   initiative, but there are two other initiatives, and

18   there are synergistic interactions between these

19   initiatives, and I think hopefully you'll see a

20   linkage between these two.

21                    I would like to sort of put on my

22   academic hat and use a very old slide that I used to

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       238


1    use when I used to teach, and this was sort of an

2    evolutionary step in pharmaceutical products and

3    process development, and I want to sort of use that

4    as a framework for defining quality by design.

5                     Pharmaceutical manufacturing, as opposed

6    to, say, device manufacturing and so forth,

7    essentially originated in the other pharmacy

8    compounding, and it has moved over the last 30 years

9    to more science and engineering based.

10                    So now you can start talking about

11   pharmaceutical engineering, and I think there is a

12   big advantage of thinking of developing products

13   from an engineering perspective.

14                    In that vein, I think we have moved from

15   dosage forms to now what we call drug delivery

16   systems in the late '80s.           And now we're moving

17   towards innovative or more intelligent drug delivery

18   systems, and I think that's the drug delivery

19   systems and intelligent drug delivery systems which

20   is essentially the focus of this initiative and

21   workshop.

22                    But in terms of, I think, quality by

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           239


1    design, we have to take a step back and see how we

2    are developing these products and what impact does

3    that have on efficiency of development and time to

4    market.

5                         Traditionally pharmaceutical development

6    started with trial and error type of

7    experimentations where it's often difficult to

8    manage the multi-variables and the interactions

9    between those variables.              We moved to a more of

10   design of experiments, more empirical statistical

11   designs in the mid-'70s, but yet we have not moved

12   to computer aided design, and I think we have an

13   opportunity to start thinking in those terms, and it

14   can have a very significant impact on not only the

15   development time, but I believe on the regulatory

16   assessment itself.

17                        If we're able to move in this direction,

18   I think we will have our resources focused on

19   testing more creative options, and that's what I

20   want to sort of convey with this slide at this

21   point.           The other aspects,      I think, end product

22   testing, a focus on testing to document quality as

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       240


1    opposed to real time quality assurance also has some

2    bearing and is part of the other initiatives I

3    talked to you about, but I will not get into that in

4    detail here.

5                     If you look at the traditional approach

6    to formulation development, Professor Langer had a

7    slide in his second to last slide, I believe, where

8    he had a black box, and through strategic

9    experimentation and so forth, if you notice the

10   black box became transparent, then you could see

11   inside the black box, and I think if we are focused

12   on trial and error and being part of the art of

13   product development, then I think we have a black

14   box to deal with, and that poses significant

15   regulatory assessment challenges and leads to

16   questions which may not really be in the best

17   interest of the development program.

18                    So if you look at traditional dosage

19   forms, a typical pharmaceutical focus would be

20   making sure it's stable and then it's bioavailable,

21   and we approach that formation development looking

22   at drug and excipients, the physical and chemical

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       241


1    properties to develop a formulation and then try to

2    understand the in vivo and in vitro attributes of

3    those products that we develop and screen, and how

4    are they absorbed and are they bioavailable or not?

5                     So there are many aspects that sort of

6    bring in the physics and the chemistry as well as

7    the test methodologies or also the physiology that

8    comes into consideration to develop a formulation

9    which is bioavailable and stable.

10                    But this is relatively simple when it

11   comes to drug delivery systems.             I think the

12   challenges get confounded and have significantly

13   much more than that, and typically I think the CMC

14   and GMP considerations that I think we struggle with

15   is to insure consistent quality and performance is

16   the objective.      How we design and how we develop

17   specification for a given product, how do we

18   manufacture and how do we establish manufacturing

19   processes and their controls, test methods and shelf

20   life are key challenges.

21                    And then once we have an approval, you

22   know, process validation, manufacturing under GMPs

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       242


1    and making sure that the manufacturing remains

2    consistent are significant challenges, too.

3                     Studies during development can have a

4    significant impact on development time.               Many

5    speakers before me have touched upon that, but I

6    think I'm talking about bridging studies with

7    respect to bioavailability characterization from a

8    chemistry perspective.          So you have to address some

9    of those, and in absence of good analytical methods

10   that relate to in vivo of performance or to shelf

11   life, it becomes very difficult to manage changes

12   that are necessary during the development program,

13   and the bridging studies can become very elaborate

14   and can often be clinical studies themselves.

15                    So unless I think we think of new

16   methods, I think these are potential bottlenecks in

17   the development program that I think we will face.

18                    Post approval changes often is not on

19   the minds of people who are focused on developing

20   formulations and doing the clinical studies.                 But I

21   think thinking about post approval changes is

22   important.       Change is part of life, and changes lead

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       243


1    to improvement at the same time, but if you're not

2    able to change and justify those changes, that can

3    lead to significant problems, too.

4                     And I think that's the point I was

5    trying to make with continuous improvement, is if

6    the regulatory process is tedious, the methodologies

7    that we use to define comparability or establish

8    comparabilities are difficult.             Then the technology,

9    the innovation is hindered, and I think we have to

10   start thinking more proactively in terms of how we

11   move forward here.

12                    And this is the point I want to make, is

13   when you start bringing drug and drug delivery

14   systems, you have not only a large number of factors

15   to understand and optimize, but you have an even

16   larger number of interaction terms, and these

17   include, I think, considerations from anatomy,

18   physiology and pathology, pharmacology of the drug,

19   pharmamechanics of the drug, biopharaceutics,

20   physical and chemical attributes of the drug, the

21   polymer, and your device.

22                    And in fact, the drug delivery system

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         244


1    itself is quite complicated, and if we remain in

2    sort of a black box mode, bridging studies, post

3    approval changes, even establishing specifications

4    can be very challenging.

5                      So I think we have to start thinking of

6    more of an engineering approach to designing these

7    systems, and that's the phrase I have used, is

8    quality by design.         We all know Quality 101.           You

9    cannot test quality in the product.                I mean, that's

10   well established.        You have a design for quality.

11                     And I think I have defined quality by

12   design as achievement of product and process

13   performance characteristics that are adequate for

14   the intended use through scientific understanding

15   and management of sources of variation and other

16   risk factors due to manufacture.

17                     Most of this process gets started in the

18   development itself, and based on my experience at

19   FDA, much of this information is not either shared

20   with FDA or there's a strong hesitation to share

21   this.        So the regulatory assessment without some of

22   this information can sometimes become quite

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       245


1    challenging.

2                     So I was recently looking at Los Alamos

3    Laboratory presentations on designing missile

4    systems and so forth, and I stole the plane from

5    that slide, and I think the key here is what are the

6    design objectives, the target to reach our target

7    goal, and in the case of drug delivery systems, they

8    have very exciting design objectives, and I think

9    the hypotheses out there are mind boggling, and I

10   think the innovation that will occur in the next ten

11   years is going to be amazingly productive and useful

12   for public health.

13                    But I think we have to be very diligent

14   in moving towards this in a structured, scientific

15   way to make sure the innovation is not hindered

16   because of regulatory concerns or, as one of the

17   speakers used, fear of the unknown.

18                    If I take a look at drug delivery

19   systems now, past and present, our focus has been on

20   changes in route of administration and bringing drug

21   delivery systems through different routes.               Clearly

22   the deployment, how we administer this has been

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           246


1    relatively simple.

2                         For example, if you have a transdermal

3    drug delivery system, some of the key features that

4    are important for deployment is the adhesive

5    performance, but the deployment can get more

6    complicated, say, if you think about a drug eluting

7    stent.           It's a procedure that require additional

8    deployment attributes to be considered.

9                         Drug delivery in the current situation

10   is primarily based on PK and PD, and the intention

11   or the design objectivity of the drug delivery

12   system is between proof compliance, patient

13   compliant, and also to improve safety and efficacy.

14   Many have used examples of the peak concentration

15   and potentially that relating to safety, and I think

16   more controlled release allows you in many ways to

17   improve safety.

18                        But the future, I think I see the

19   deployment attributes could get more complicated

20   because now you're looking at more sophisticated

21   devices either implanted or otherwise, and you may

22   have to have considerations for what are the right

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          247


1    deployment attributes and how does a drug coating or

2    a drug combination alter that or how do we manage

3    that from a chemical stability, shelf life

4    perspective also?

5                        And clearly I think the desire is to

6    move towards more target oriented drug delivery

7    system.          The challenges and the opportunities

8    associated with those challenges are currently in

9    the pharmaceutical development quality and

10   performance consistency has been based on

11   traditional chemistry testing.                I feel, and I think

12   the previous presentation made a good point for

13   that, that there are gaps in physical test.                   We have

14   a difficult time addressing physical changes which

15   are important and establishing shelf life based on

16   physical changes are still more complicated.

17                       The way forward in my opinion is we have

18   to be proactive and look for these challenges and

19   start working on those now.               If we don't then we

20   create bottlenecks and its difficult to get over

21   those bottlenecks.

22                       Clearly, if I just use a quick example

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         248


1    of drug coated stent, but starting with stents

2    themselves, I think if you look at what are the

3    design considerations here, intracoronary stents are

4    deployed to form a scaffolding for the coronary

5    artery vessel wall during coronary angioplasty.                     So

6    I think the applied and the procedure leads to a

7    number of issues that I think we have to think

8    about.

9                       How does the drug coating affect this

10   process?         Or is this process affecting the drug

11   coating itself, and so forth?               That sort of comes

12   through that in the design consideration.

13                      I'm going to skip this slide.

14                      So as we start thinking about how do we

15   identify and optimize critical factors, trial and

16   error experimentation under all selected conditions

17   is one way, but I don't think it's practical.                    There

18   are significant opportunities where I think quality

19   by design brings in an engineering approach where

20   computer analysis employs numerical techniques of

21   finite elements coupled with completion of fluid

22   dynamics can help us understand our systems better

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                             249


1    and actually help us control or design systems that

2    will address the sources of variation quickly and up

3    front and also have this information available to at

4    least discuss with the agency.

5                         One of the issues, i think, which is

6    quite important, is the release rate.                     Many speakers

7    have essentially argued the importance of the

8    release rate from either drug eluting stent or any

9    other delivery system.              But what are the design

10   objectives? What is optimal in vivo release profile?

11    What is the mechanism and rate and duration of this

12   release?          How do we establish specifications?

13                        These are important questions, and

14   unless we think of different ways, the way today is

15   to establish these specifications based on a limited

16   amount of information.              The opportunity is there to

17   actually get to the mechanisms of these release

18   profiles and actually start building back into the

19   decision making criteria both in the companies and

20   FDA.

21                        So how do we establish controls and

22   tests?           Factors that influence release profile in

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                         250


1    vivo as well as design feature itself and

2    manufacturing factors, in vitro test methods,

3    quality assurance, and in vivo relevance I think are

4    important questions.

5                     And with a focus on testing to document

6    quality, these would be quite challenging, but with

7    a move towards quality by design through scientific

8    understanding, I think we can find a better way of

9    moving forward.

10                    For example, I think with drug eluting

11   stents what is an appropriate in vitro method?                  I

12   think that is a significant discussion point and a

13   debating point of how does one start addressing that

14   question.

15                    Is that the right way of dealing with

16   the quality issues or even establishing in vivo, in

17   vitro, and real correlation?            I think these are

18   topics that I think we need further discussion.

19                    The only point I want to make here is if

20   we assume traditional drug release profiles and use,

21   for example, bulk elution models, this is a

22   publication from MIT-Harvard.             Professor Hwang is

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          251


1    one of the authors of this, and this was published

2    in Circulation, essentially identifying some of the

3    challenges in terms of drug release.

4                     If you use traditional approaches, we

5    get a flat concentration profile, but if we examine

6    the coronary artery after application of a stent,

7    there's a potential for localize effect, which may

8    be very different from and is not picked up by the

9    traditional pharmacokinetics modeling.

10                    So if we establish an in vitro release

11   profile or an in vivo release relevant to a

12   traditional in vitro/in vivo correlation, is that

13   the right question?        Is that the right thing or are

14   we even asking the right question?

15                    So these issues come up.             So I think

16   there is a wonderful connection between the new

17   initiative and the initiative on drug quality

18   system, and I want to sort of end my presentation --

19   I'm on time -- end my presentation with a couple of

20   slides sort of explaining the other initiative and

21   so that you can see the connection between the two.

22                    In the direct quality system for the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                       252


1    21st Century, I think what we have articulated here

2    is a vision.      Pharmaceutical manufacturing is

3    evolving from an art form to one that is now science

4    and engineering based.          Effectively using this

5    knowledge in regulatory decisions as we establish

6    specifications and evaluating manufacturing

7    processes which can substantially improve the

8    efficiency of both manufacturing and I would argue

9    development manufacturing and the regulatory

10   process.

11                    This initiative is designed to do just

12   that through an integrated systems of product

13   quality regulation founded on sound science and

14   engineering principles for assessing and mitigating

15   risk of poor product and process quality in the

16   context of intended user of pharmaceutical products

17                    So the desired state essentially as we

18   have defined here is product quality and performance

19   achieved and assured by design of effective and

20   efficient manufacturing processes, and this is the

21   point I was making.        Product specification based on

22   mechanistic understanding of how formulation process

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       253


1    factors impact product performance, guarantees real

2    time assurance of quality, but in order to get

3    there, I think the regulatory system that has to

4    evolve, our regulatory policy should be tailored to

5    recognize the level of scientific knowledge,

6    supporting product applications, process validation,

7    and process capability.

8                     Risk based regulatory scrutiny then

9    relates to level of scientific understanding of how

10   formulation and manufacturing process factors affect

11   product quality and performance, and the capability

12   of process control strategy is to prevent or

13   mitigate risk of producing a poor quality product.

14                    With that I'll stop.         I know we're

15   running late.      If you have any questions, I think

16   why don't we have you sort of contact the speakers

17   directly?

18                    So we will hold the questions to

19   individual questions if you can catch us.              If not,

20   then have a great lunch.          Thank you.

21                    (Whereupon, at 12:49 p.m., the meeting

22   was recessed for lunch, to reconvene at 2:00 p.m.,

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                    254


1   the same day.)

2




                             NEAL R. GROSS
                     COURT REPORTERS AND TRANSCRIBERS
                         1323 RHODE ISLAND AVE., N.W.
    (202) 234-4433       WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                             255


1                     A-F-T-E-R-N-O-O-N       S-E-S-S-I-O-N

2                                                              (1:47 p.m.)

3                      DR. PROVOST:      We did offer the public

4    the opportunity to comment, and we did have one

5    request to speak, and that is Dr. Paul Goldfarb.

6    He's with Oncology Associates and is a clinical

7    professor of medicine at U.C.-San Diego, and Dr.

8    Goldfarb will make his presentation now.

9                      DR. GOLDFARB:       Thank you.

10                     My name is Paul Goldfarb.             I'm a surgeon

11   actually, and I do cancer surgery.                 I trained at

12   Memorial Sloan-Kettering, and so I guess in the

13   context of today's meeting I'm a maximally invasive

14   radiologist.

15                     (Laughter.)

16                     DR. GOLDFARB:       I have had the

17   opportunity to work with two different companies

18   that deal with ablation of tumors using drugs.                      I

19   find it intriguing because in doing surgical

20   oncology we're always looking, despite what most

21   other physicians think, we're actually looking for

22   new ways of achieving the same goals using less

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                         256


1    invasive technologies and trying to find new ways of

2    dealing with it.

3                       I've been to the agency several times

4    with Genetronics, and I've certainly been aware of

5    the work at FeRx and have helped them do one of

6    their trials that we'll discuss today as well, and

7    the reason that I've come again is because I think

8    these are critically important issues to us who do

9    clinical medicine and surgery that need to be

10   addressed.

11                      Today I'm using a computer generated

12   presentation.         The last time I came in November I

13   did it with overheads.            So even I have moved forward

14   with the technology.

15                      I think there's a pressing clinical need

16   to develop new technologies to control localized

17   disease.         I think more and more we're finding other

18   needs to control local manifestations of disease,

19   either primary or recurrent.

20                      We're looking for less invasive ways of

21   doing it, and we want to find ways that are more

22   protective of normal tissues.               The rapid adoption of

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       257


1    thermal ablation targeted radionucleotides,

2    hypothermia, embolic agents, and cryosurgery

3    reflects the fact that all of us are looking for

4    these noninterventional ways of approaching these

5    kinds of tumors.

6                     The drug-device combinations as novel

7    drug delivery systems provide the potential to

8    enhance the effectiveness and reduce the adverse

9    events of intertumoral delivery.              Right now we have

10   tumor ablation systems that combine drug delivery

11   systems in multiple parts of the body, and as you

12   can see from the slide, those are all of the organs,

13   all of the solid organs that we're now looking at

14   using drug delivery systems to try to treat with

15   local therapy.

16                    Now, we are able to achieve high local

17   drug concentrations.         There's low systemic exposure,

18   and we have equivalent response in the tumors to

19   other ablative forms of therapy, including surgery,

20   the advantage being that we're able to preserve

21   adjacent normal tissue.

22                    This is one of the ways I actually got

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         258


1    into doing radio frequency, was that it seemed like

2    such an obvious move that if I could use an ablative

3    technology that would preserve the half of the liver

4    that the tumor was sitting in, that I'd be able to

5    manage the patient much better than doing right

6    hepatic lobectomies, even though it pays less.

7                       The two companies that we want to talk

8    about are FeRx.         FeRx you've already heard described

9    briefly this morning in the discussion by the

10   radiologists.         It's an intertumoral drug delivery

11   system which takes doxorubicin and uses small

12   magnetic pellets to put the drug directly into the

13   tumor.

14                      And as you saw this morning, it's easy

15   to target the tumor using the technology, using a

16   simple external magnet.

17                      Genetronics is a company that uses

18   electroporation as its way of enhancing the delivery

19   of drug.         Electroporation is the technique where you

20   create an electric field within the tumor by using a

21   series of needles.          You inject the bleomycin into

22   the tumor initially, and then by creating the field

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       259


1    you allow the drug to enter the cell, and you

2    essentially get ablation of the malignant tissue

3    with protection of the normal healthy tissue around

4    it.

5                     In both of these systems -- and that's

6    why I came back, because now we really have two

7    different products that address things the same way.

8     Utilizing well characterized drugs with known

9    safety profiles, we deliver the drug to a localized

10   area with minimal systemic exposure.

11                    What we're really doing is utilizing

12   novel devices to deliver this well established drug.

13    In both systems we have an ablative effect that's

14   confined to the area of the drug delivery and

15   affects malignant tissues independent of histology

16   and demonstrates a clinical benefit analogous to

17   that of thermal ablation or surgical resection.

18                    The issues that I want to talk to you

19   about for a few minutes are the regulatory pathways

20   and the standards that we're using for these sorts

21   of products I believe are inappropriate for the

22   perceived clinical benefit; that in both cases CDER

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         260


1    is the lead review agency for both of these

2    combination products, and I understand why CDER is

3    the review agency, and my education in this that has

4    gone on for the last five years has taught me that

5    the issue is really not which agency reviews it or

6    which division reviews it, but how it's reviewed,

7    and so I don't think that's an issue.

8                     There are no other products that have

9    localized ablative effects at disease sites that

10   have been required to do such extensive testing and

11   have such extensive review.

12                    The drug components of these combination

13   products in both cases that have been approved and

14   used clinically for decades, they have safety

15   profiles that are well characterized.                 They have

16   extensive scientific and medical therapy supporting

17   multiple therapeutic applications, and the

18   technologies in these two cases are being developed

19   by using reduced therapeutic doses of drugs.

20                    So really the dose of Adriamycin or

21   bleomycin used in these technologies is essentially

22   homeopathic, and that they have minimal systemic

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                       261


1    exposure to the drug.         Both of these products which

2    have a local effect are currently held to the same

3    evidentiary standards and regulatory burdens of new

4    drugs having untested and potentially significant

5    systemic effects.

6                     FeRx is in the process of conducting a

7    Phase III study of over 200 patients with

8    hepatocellular carcinoma, comparing their local

9    therapy to a systemic chemotherapy in patients with

10   end stage disease, and the study is using the

11   survival endpoint.

12                    The Phase I and II studies have already

13   been done, demonstrated efficient tumor targeting

14   using their product with Adriamycin; showed durable

15   local disease control; and showed that the dosing

16   paradigm was really based on the size of the tumor

17   and not on the patient weight.

18                    The new paradigm that we're looking at

19   is to use ablation therapies regardless of what they

20   are in terms of hepatocellular cancer because we now

21   use it as a bridge to liver transplant.               Liver

22   transplant is perceived as the gold standard for the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       262


1    treatment for liver cancer.            The role of ablation

2    technologies has really become one to stabilize the

3    patient until the liver is available.

4                     And so in a sense, that's the clinical

5    arm.       That's where we would be using it clinically.

6    We'd be much less likely to use this local ablation

7    therapy in people with far advanced disease.

8                     And as I say, stabilization of the

9    disease then becomes a viable surrogate clinical

10   endpoint because that's what we would be doing in a

11   clinical environment.

12                    Here's an example of how the FeRx

13   product works.       Here's a tumor.        You see the blood

14   supply on the left.

15                    Since I took my pointer back, I'll have

16   to use -- there's the tumor, and you're able to

17   actually put the drug just where the tumor is and

18   have the clinical effect of ablating that tumor.

19                    What's been interesting and what I've

20   worked with FeRx on is using the same technology in

21   a group of patients who have metastatic cancer.                   So

22   these are people who have non-hepatomas, and the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       263


1    question would be:        can this drug Adriamycin, which

2    we normally would not use in these other settings,

3    be of value?

4                     And, two, the thought had always been

5    that the blood supply to metastatic tumors was such

6    that it didn't allow for easy interarterial therapy.

7     In fact, what we demonstrate, and these are studies

8    using PET scans, and so what you're really saying is

9    that this is the tumor before treatment, and the

10   patient after treatment.          At least physiologically

11   you can say that the tumor is not viable.

12                    These are early studies, but we

13   certainly plan on following up on this, and I think

14   this is the future for this kind of therapy.                 I use

15   it to highlight the issue that the therapy works

16   independent of histology just as radio frequency

17   works independent of histology.

18                    This was the second patient where,

19   again, there's the tumor and there was the effect on

20   PET scan.

21                    Genetronics is a company that has a

22   local therapy, and they embarked upon treating head

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         264


1    and neck cancer as the model that they wanted to

2    test in, and we were initially involved in a classic

3    Phase III randomized trial in which we were going to

4    take people with far advanced head and neck cancer.

5    Half would get this local therapy.                 All would get

6    systemic chemotherapy, and we would try to

7    demonstrate a survival advantage.

8                       I must say as a surgical oncologist I

9    thought that the study was inappropriate in the

10   sense that that's not where I would use a local

11   ablation therapy, and I thought the chance of

12   meeting that goal was unlikely to occur.

13                      And so I called Mark Kramer about a week

14   after he got his new job and said, "You're in

15   Combination Products.           I've got a combination

16   product.         We need to figure out where do we go from

17   here."

18                      And so with Mark's help and in

19   renegotiating with CDER, we have now evolved a study

20   which I think is more clinically relevant in which

21   we take people with early recurrence or second

22   primaries, and we're looking at comparing the role

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       265


1    of this ablation technology to surgery because

2    that's really the standard.

3                     I would not expect that the ablation to

4    do better.       So as we saw this morning when they were

5    talking about stents, we're sort of trapped because

6    we have this positive gold standard comparator

7    rather than comparing it to nothing.

8                     But I think what we could show is that

9    the control rate of these tumors will be no worse

10   than it is with surgery, and arguably since we're

11   able to do a much smaller, less invasive procedure

12   than what I would normally do as a surgical

13   oncologist, we should show functional improvement,

14   and we should show pharmacoeconomic advantages that

15   it should be cheaper and easier to achieve the same

16   goals.

17                    Now, the ongoing challenge is this is a

18   stretch for the people at CDER just as it's a

19   stretch for all of us, and so it has required

20   ongoing negotiations about what really is a

21   functional benefit and how do we measure this and

22   how will we really know what's going to happen.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        266


1                     What we had shown originally with

2    electroporation is -- I'll go over it.                At the

3    agency's behest, we took a bunch of people and

4    injected bleomycin into the tumor with no

5    electroporation, and we got essentially no result.

6                     We then took people with far advanced

7    cancer and injected bleomycin, electroporated them,

8    and essentially we got a 50 percent objective

9    response rate.

10                    But we also had a group of people in

11   Europe who had primary head and neck cancer, and

12   these people were treated in a way that's very

13   similar to what's been done with ablation.                They had

14   their tumors electroporated, which consists of

15   injecting drug, putting the needles in, treating the

16   whole tumor.      They were electroporated, and then

17   several weeks later the tumors were cut out.                  So we

18   basically had a treat and resect model, which is the

19   classic ablation technology.

20                    In those now 20 patients there's nobody

21   who has had a local recurrence out to two years.

22   There were three who had microscopic cells in the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                       267


1    resected specimen.

2                     I argue that if I was coming with a new

3    form of ablation technology that used luke warm

4    temperature instead of hot or cold and I said we

5    have 20 people where we treated and resected, we'd

6    have a discussion about whether this is approvable

7    instead of embarking upon a 400-patient study.

8                     I realize the challenge that I'm

9    presenting, but I think that these are issues that

10   need to be raised, and since the afternoon is set

11   aside for discussion of regulatory issues, I hope

12   this is a good lead-in to that discussion.

13                    Now, my suggestions are both of these

14   products are subject to review standards typically

15   applied to novel drugs with unknown risk and safety

16   profiles, with large numbers of patients, and a

17   survival endpoint.        The device products that have

18   been approved for local ablation type effects have

19   been subjected to much less extensive clinical data

20   requirements.

21                    Given that the safety profiles of both

22   drugs are well characterized, there's minimal

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       268


1    systemic exposure.        The requirements for approval

2    should be comparable to devices that have an

3    ablation effect.

4                     When I came in November, my approach was

5    basically at that meeting that we should look at all

6    of the -- everything we're dealing with either has a

7    local effect, a regional effect, or a systemic

8    effect, and so it doesn't matter, I would argue

9    whether it's a drug, a device or a biologic.                 We

10   look at the effect on the patient, and that sort of

11   defines how we should look at it in terms of

12   regulation, and that might make it easier.

13                    Both products are innovative device-drug

14   combinations that utilize a new route of

15   administration for old drugs, drugs that have been

16   formerly administered intravenously and should have

17   reduced time in clinical development and reduced

18   evidentiary requirements.

19                    Recommendations.       New therapies need to

20   be compared to other therapies that have a similar

21   effect on the patient.          Therapies which are local,

22   regional and systemic in their effect should be

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       269


1    compared to therapies with a like effect regardless

2    of which division is assigned as the lead, Device

3    Drugs or Biologics.

4                     To expedite the review and approval of

5    innovative devices for the delivery of known drugs,

6    the evidentiary standards must be appropriate to the

7    potential risk-benefit in cancer patients.               And I

8    speak basically as a surgical oncologist.

9                     We need to implement new regulatory

10   pathways and least burdensome principles for

11   innovative technologies that allow for rapid market

12   entry and for patient benefit.

13                    I've been working with these products

14   for over five years.         It seems to me that after five

15   years and several hundred patients were treated it

16   would be nice if we could find a way to move this

17   forward in a more expeditious manner.

18                    I understand what the barriers are, and

19   certainly we're living within those guidelines and

20   moving forward, but I think as a surgical

21   oncologist, first because of my surgical

22   personality, and then, two, because of my ongoing

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       270


1    clinical needs, I come to say to you we need to find

2    a better way to do it.

3                     I'd close by saying it reminds me of

4    what Yogi Berra said.         Yogi Berra said that in

5    baseball 50 percent of baseball is 90 percent

6    mental.

7                     (Laughter.)

8                     DR. GOLDFARB:      And so I think regulatory

9    approval is the same in a sense in that regulatory

10   approval, 50 percent of regulatory approval is 90

11   percent negotiation, and so I hope that this opens

12   the door so that we can continue that process.

13                    Thank you very much for allowing me the

14   time.

15                    (Applause.)

16                    DR. PROVOST:     Thank you.

17                    And now I'd like to introduce the

18   moderator for the first session of this afternoon on

19   regulatory issues, the industry perspective, and

20   we're very pleased to have Dr. Liz Jacobsen here.

21   Liz is a former FDAer, was at FDA for a long time,

22   and is now at AdvaMed as the Executive Vice

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       271


1    President for Technology and Regulatory Affairs.

2                     DR. JACOBSEN:      Well, thank you very

3    much, and it always bothers me a little when they

4    say "a long time."

5                     (Laughter.)

6                     DR. JACOBSEN:      Welcome to the industry

7    perspective session.         It's my pleasure to be the

8    moderator for this segment and also for the final

9    session, which is going to be the FDA-industry kind

10   of Q&A session.

11                    And we're hoping to get some good

12   discussion going at the end of the day, and first

13   we're going to have remarks from sort of a legal

14   perspective from the device and drug industries and

15   from FDA.

16                    So we are going to ask you if you would

17   hold your questions for those sessions, either hold

18   them in your head so you can go up to the microphone

19   at the last session of the day or write them down

20   and you can give them to Miriam at the break or

21   whenever you see her, and she'll make sure that they

22   get up here, and we're hoping that that will work

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       272


1    because obviously we would like to have some good

2    Q&A.

3                     Okay.   Well, first up in this session,

4    the industry perspective, is Jonathan Kahan, partner

5    at Hogan & Hartson, and he'll be talking about

6    regulatory and legal challenges for the developers

7    of drug delivery systems.

8                     MR. KAHAN:     Thank you very much.

9                     Good afternoon.       I want to thank Dr.

10   Feigal and Dr. Provost for inviting me to speak this

11   afternoon.       I promise to be on my best behavior.

12                    And there is good news and bad news, I

13   think, in my presentation.           I think the good news is

14   I will have no slides of blood fields or tumors, and

15   the bad news is I'm going to try to walk you through

16   some fairly dry legal and regulatory issues,

17   although I'm also going to try to give you, I think,

18   the perspective, at least my perspective, on some of

19   the significant issues that industry has faced over

20   the years in this area.

21                    I'm going to start out by talking very

22   briefly about the legal framework.               I'm then going

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       273


1    to talk about the historical approach that FDA has

2    taken to the regulation of combination products and

3    drug delivery devices over the years.

4                     And then I'm going to talk about the

5    obstacles and challenges that we're all facing in

6    this area and try to talk about some new policies

7    and procedures which may be appropriate in this area

8    to try to change around what I think a lot of us,

9    including many at FDA feel is not an optimal area

10   right now.       There are many, many delays and

11   inefficiencies in the process, and I think we're

12   going to have a good discussion about that this

13   afternoon.

14                    Just for those of you who are interested

15   in definitions, a lot of what we're going to be

16   talking about this afternoon has to do with the

17   definitions of drugs, devices, and biologics.                  And

18   without going into too much detail and putting

19   everybody to sleep, basically drugs are articles

20   intended to prevent, cure, and treat disease,

21   intended to affect the structure and function of the

22   human body.       It's basically the same definitions for

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        274


1    devices, except the devices do not typically achieve

2    their primary purposes through chemical or metabolic

3    action.

4                      I'd say the rule of thumb is if it's

5    more mechanical, it's a device.              If it's more

6    metabolic and chemical, it's a drug, although they,

7    as we'll talk about probably in depth this

8    afternoon, they very often tend to merge, and it

9    becomes a very metaphysical discussion as to whether

10   the action of the product is chemical, metabolic, or

11   physical.        In many cases, as we'll discuss, it's all

12   three.

13                     Biologics, I have no clue as to what a

14   biologic is.

15                     (Laughter.)

16                     MR. KAHAN:     This is the definition of

17   biologic.        It's sort of like pornography.         You know

18   it when you see it, but it's hard to define, and

19   biologics are basically derived -- there are

20   definitions under the Public Health Services Act and

21   we'll talk about in a minute that are actually

22   products that are combinations of drugs, devices,

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       275


1    and biologics, all in one specific product.

2                     Just historically, just to give you the

3    perspective of what we've all been facing for years

4    and years, back before 1990, we sort of addressed

5    all combination products, including drug delivery

6    devices on sort of a case-by-case basis, and you've

7    heard that very often it was a question of

8    negotiation, and that's absolutely true.

9                     In many of these cases, there was

10   negotiation not only between the companies and FDA

11   as to how the product was going to be regulated, but

12   there were also negotiations within FDA as to how

13   the product was going to be regulated.

14                    I'll give you just one example.           I'm

15   going to try to keep the war stories to a bare

16   minimum, but biliary lithotripters is just a good

17   example to start out on and combination products

18   generally because with respect to that product you

19   had a lithotripter that could fragment gallstones,

20   but it needed to be used with a litholitic agent,

21   which at that time was ursodiol or Actigol, the

22   product that was on the market at that time.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                            276


1                       And the companies came to FDA and said,

2    "We want PMAs for our lithotripters.                    Clear these

3    devices."

4                       And FDA said, "No, this is a combination

5    product, and it needs to be used with the drug," and

6    they said, "But the drug has already been approved

7    for the dissolution of the stones."

8                       And Steven Fred then in Gastro at CDR

9    came back and said, "Wait a minute.                 It was cleared

10   for nonfragmented gallstones.               We need an NDA

11   supplement for fragmented stones."

12                      That was basically the end of the

13   process.         The drug company was not willing to work

14   with the device companies, and 12 years later,

15   probably 13 years later, the drug company finally

16   decided to allow access to its NDA files, and that

17   product was approved.

18                      But that roadblock, which I'll talk

19   about again in a minute between access to drug files

20   and master files and IND files is one of the key

21   factors that has led to many, many problems over the

22   years in the drug delivery area and in combination

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                       277


1    products generally.

2                     How did we seek to resolve this issue?

3    I believe that the disaster we face on biliary

4    lithotripters was one of the reasons that Congress

5    decided to address the law in the Safe Medical

6    Devices Act of 1990, which added the combination

7    product regulations.

8                     As a matter of fact, the person who

9    actually drafted the first combination product

10   statutory division was Pat Schraeder, who's not here

11   today, when she was working with Senator Kennedy's

12   committee on that, and at that time, the first draft

13   of this regulation and statute was essentially

14   designed to allow one filing.             There was not going

15   to be an NDA and a PMA for one product.

16                    Congress backed down on that probably

17   through the second or third draft of the law, but

18   essentially what came out of the law was we need

19   some structure to combination products, and the way

20   we're going to add structure is we're going to work

21   with FDA and we're going to say that the primary

22   mode of action is going to be the key standard for

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       278


1    determining whether a product is going to be

2    regulated by the device center, by the drug center,

3    or the biologics center and under which statutory

4    authorities is that product going to be regulated.

5                     The problem is that Congress never

6    defined primary mode of action, and to this day it

7    has never been defined, and FDA, as I understand it

8    and will talk about this this afternoon probably

9    during Mark Kramer's presentation, FDA, I believe,

10   is now starting down the road of seeking to actually

11   define primary mode of action, and I think we all

12   welcome that.

13                    Mark is also going to talk about the

14   definition of combination products.               So I'm not

15   going to get into it very much, but simply to say

16   that the technologies that are coming along right

17   now are mind boggling.          You've only heard of some of

18   them, and I never cease to be amazed by the

19   combinations of drugs, devices and biologics that

20   are presently on the drawing board and which FDA is

21   going to be facing very shortly.

22                    And one of the things that we're going

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         279


1    to talk about today is whether the 20th Century

2    standards and regulations that FDA now has in place

3    are going to be adequate to handle the 21st Century

4    technologies that are presently coming down the

5    pike.        It's no longer going to be are we looking at

6    prefilled syringes, but are we going to be looking

7    at products, for example, a dopaminergic cell that's

8    encapsulated in a semi-permeable polymer that elutes

9    dopamine.        So you have a drug, dopamine.         You have a

10   dopaminergic cell, which is a biologic, and you have

11   a semi-permeable polymer, which is a device.                  And

12   there are many, many products that are presently

13   coming down the road that are going to be

14   combinations of many different kinds of tissues,

15   drugs, deices, and biologics.

16                     So I'm going to let Mark handle the rest

17   of that one.

18                     (Laughter.)

19                     MR. KAHAN:     With respect to exactly the

20   regulatory structure, again, others are going to be

21   better able to deal with this.              I'm just going to

22   very quickly talk about sort of what really happens

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       280


1    when a company has a combination product.

2                     Over the years, we basically have dealt

3    initially with the product jurisdiction officers.

4    That would be Warren Rumble right now in drugs, Gene

5    Burke over in Devices, and Cheryl Lord Weiford over

6    in Biologics, and often we simply seek to get an

7    indication or a feeling from them when we initially

8    have a product.

9                     What do you guys think, based upon your

10   institutional memories?          Do we need to file a

11   request for designation?

12                    And under the Safe Medical Devices Act

13   of 1990, Congress said, "Wait a minute," and FDA

14   implemented regulations under Part 3 that said,

15   okay, if you're not sure, you can file an RFD, a

16   request for designation, and we will tell you within

17   60 days how we're going to regulate your product,

18   what's the primary mode of action, and in some cases

19   in those letters, they actually tell you whether

20   it's going to be a PMA, a 510(k), and what the NDA

21   process may look like.

22                    And we often start out with discussing

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       281


1    these issues with the product jurisdiction officers.

2     We find that helpful.          We then move on to the

3    request for designation, and the next stage is, once

4    you've gotten the designation, we then go to a pre-

5    IDE or a pre-IND meeting to actually flesh out

6    exactly what the data requirements are going to be.

7                     Now, if you're a smart company, you

8    start thinking about the data requirements at the

9    early developmental stages of your product, not when

10   you're sitting down with FDA and talking to them in

11   a pre-IDE or a pre-IND meeting.

12                    And, therefore, I think the most

13   important thing I can probably say today is start

14   early and communicate very well with both your

15   clinicians, your engineers, your regulatory affairs

16   people, and try to prophesy early on what you might

17   need for that pre-IDE, pre-IND meeting later on.

18                    And then you're going to later have to

19   face, if it is a combination product, coordination

20   between CDRH, CBER and CDER.            Mark will talk

21   probably more about exactly what they call their

22   consultative meetings and their collaborative

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       282


1    reviews and consultative reviews.

2                     But the bottom line is if it's going to

3    be one center with primary jurisdiction, they will

4    consult with another center, and if it is a

5    collaborative review with two primary reviewers,

6    you're going to have input from two centers at once,

7    and that's often not the best way to do it.

8                     So I think most of us would try to seek

9    to have one center with primary jurisdiction and the

10   other center consulting so that you're not whipsawed

11   between two centers during the process.

12                    Just to again give you the historical

13   picture here, over the years FDA has sort of

14   developed their own gestalt internally, some of

15   which is reflected in the inter-center agreements,

16   which I would urge all of you to read, although I'm

17   about to tell you I think they are out of date,

18   outmoded and need to be revised.

19                    But they give you a picture of what

20   FDA's thinking actually is, and if you look at the

21   inter-center agreements, you'll see that things like

22   prefilled syringes and infusion pumps and

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       283


1    transdermal patches, those kinds of things have all

2    been pretty much defined as to how FDA is going to

3    look at them.      That's what I call the early

4    generation products, although ionaphoresis devices

5    have given FDA heartburn for quite some time, and I

6    won't go into how FDA has regulated ionaphoresis

7    devices over the years, but it's not a pretty

8    picture.

9                     And with the new products coming down

10   the road with regard to ionaphoresis devices,

11   hopefully meetings like this can help develop new

12   paradigms for how to regulate products like that.

13                    Simply another very quick example on

14   metered dose inhalers, if those of you who remember

15   those products were originally all regulated in the

16   device center through the 510(k) process, there were

17   then I wouldn't say it was a fight.               It was a very

18   cordial discussion between the device center and the

19   drug center about whether the droplets and the size

20   of the droplets and the efficacy of the drug that is

21   taken in through the metered dose inhaler requires

22   the inhaler to be regulated in the drug process with

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         284


1    the approval of the drug, like albuterol sulfate,

2    for example.

3                     And that's what happened.            All got

4    shifted over to the drug center based upon the

5    safety and efficacy issues that CDER thought were

6    raised as part of the drug delivery process of the

7    drug.

8                     The second generation products, most of

9    you didn't realize that your cigarette was a drug

10   delivery device.       Neither did the Supreme Court,

11   and --

12                    (Laughter.)

13                    MR. KAHAN:     -- therefore, that issue is

14   no longer on the table, but there are many other

15   products during the second round, which I believe

16   FDA has been thinking quite a bit about.                The drug-

17   coated catheters and stents have primarily been

18   regulated through the device center where the drug

19   coating on a catheter, for example, if you had an

20   antimicrobial catheter, if it's an approved

21   antimicrobial the device center has pretty much kept

22   jurisdiction.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                       285


1                     On the drug-coated stent, you've heard a

2    lot about that.       We're going to talk a lot more

3    about it later.       The bottom line is there that I

4    believe that what FDA did there they should be

5    congratulated on.       It was a very well thought out, a

6    very common-sensical approach.             The studies and the

7    way FDA is handling that I think is optimal, and I

8    hope that that kind of paradigm can be used further

9    in the future.

10                    I did sit in in one meeting where Dr.

11   Lipicky, the head of Cardiovascular Drugs, indicated

12   that he wanted a 10,000 patient study, and I think

13   the device company fell out of their chair at that

14   point, but what we ended up with was studies that

15   started out initially with 1,000 patients, with

16   post-market requirements up to a couple of thousand

17   more patients.

18                    And while it is true that a two to 3,000

19   patient study cannot meet the ICH guidelines for a

20   one in 10,000 adverse event rate, identification

21   rate, I believe that the approach that's been taken

22   here with the coordination between the drug center

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       286


1    and the device center and the working groups that

2    have been set up is absolutely a paradigm that can

3    be applied to other drug delivery devices in the

4    future.

5                     I'll just mention a couple of others.

6    There are drug delivery lollipops.               In working on

7    that one, FDA's primary concern was that Grandpa

8    would leave his fentanyl lollipop on the stand and

9    his grandson would get a nice dose of a controlled

10   substance, but chewing gums and lollipops and other

11   drug delivery devices through the oral mucosa is

12   another way that we're going to see in the second

13   generation those are already now on the market and

14   in use.

15                    Now, this third generation of products

16   is one that I think is going to cause FDA a lot of

17   trouble, and it's going to cause the companies a lot

18   of trouble, and it's going to require a lot of

19   creative thinking, and we're going to talk about

20   this in a second, but one of the major issues that

21   we're going to be facing now is that some of these

22   drug delivery devices are going to be delivering

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         287


1    multiple drugs at one time.

2                     And let's say that you had -- I think

3    Dr. Langer talked about the microchips device which

4    could have 100 wells, and let's say you're going to

5    have 20 different drugs in those 100 wells.                  Do we

6    need to get an NDA supplement or an NDA for each

7    drug that's going to be in that little pacemaker?

8                     I think those are the kinds of issues

9    that we're going to be facing in the very immediate

10   future.

11                    I'm now going to talk about the

12   challenges that we're facing with these products,

13   and I'll try to be quick because I don't want to

14   take too much more time.

15                    Drug delivery devices are often

16   developed initially by the drug companies for uses

17   with approved drugs or biologics, and that usually

18   is the easiest paradigm to deal with.                 If it's an

19   approved drug, usually CDER and CDRH are pretty

20   comfortable with it, and that's why you will see

21   some of the silver-coated wound dressings or

22   antimicrobial bone cements.            There's not too much

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                       288


1    heartburn at FDA about that, and the agency has been

2    able to regulate those products fairly well.

3                     However, when you switch to new or

4    different indications for the drug or you have a

5    different mode of delivery or a different drug or

6    dosage schedule, all hell breaks loose, and then you

7    have to really start in what is essentially a

8    scientific regulatory negotiating process with the

9    agency.

10                    And the question then is when you modify

11   the drug formulation to optimize delivery with the

12   device, are you now having, as a couple of people

13   have said, are you now about to reinvent the wheel

14   and have to start over with, let's say you can skip

15   Phase I of the drug process and go to a Phase

16   II/Phase III drug trial, at the same time that

17   you're demonstrating the safety and efficacy of your

18   device.

19                    That is not something that most device

20   companies want to do.         They do not want to reinvent

21   the drug wheel.       And so the question is:         is a new

22   NDA required for the drug if you have a different

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       289


1    delivery mechanism than the mechanism that was

2    described in the NDA-approved label?

3                     In other words, the NDA was approved for

4    IV use or subcutaneous injection and now you want to

5    deliver it in that little pacemaker that's

6    implanted.       Does that require you to have to go

7    through an entirely new NDA process?

8                     Let's say that you change nothing with

9    respect to the drug that's being delivered, although

10   there may be stability issues and a tiny bit of a

11   reformulation.       Are you going to have to start the

12   NDA process over again?

13                    I'm going to raise a lot of questions.

14   I'm not even going to pretend I have the answers to

15   all of these questions.          I tell my kids I have all

16   the answers.      They don't believe me either, but I'm

17   not going to try to answer all of these.              Maybe this

18   afternoon with people smarter than I we'll be able

19   to try to answer some of these questions.

20                    All I can say is that it is not optimal

21   to start over when you have a new drug delivery

22   device, to start over in the entire new NDA process,

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       290


1    and I'll talk about a couple of alternatives in a

2    second.

3                     So the question is, when you have this

4    sort of combination of a drug delivery device and

5    either a new drug or a drug which has been modified,

6    which predominates in the review process?              Is it the

7    PMA for that new novel MicroCHIPS pacemaker type

8    device or is the NDA process going to predominate?

9                     And does the device labeling have to

10   conform, mutually conform to the drug labeling?

11   This is an issue which Mark has on his plate right

12   now for several different companies and the inter-

13   center agreements say that the drug labeling and the

14   device labeling have to mutually conform.

15                    So you couldn't clear a device,

16   theoretically, unless the device's labeling was in

17   conformance with the drug labeling, and the inter-

18   center agreement primarily talks about conforming in

19   terms of formulations, dosage and schedule, but it

20   doesn't necessarily address all of the issues.

21                    For example, if you have a device that's

22   now being delivered subcutaneously and you now want

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       291


1    to have it implanted in a pacemaker to deliver the

2    drug over time, does that now mean that you have to

3    have an NDA supplement or change the drug labeling?

4                     And if you're not the drug company, what

5    do you do?       You can't change that drug company's

6    labeling if you're a device manufacturer.              So what

7    do we do?

8                     Here's the challenge.         The challenge is

9    that if the pharmaceutical manufacturer authorizes

10   access to their master files, their DMFs or they

11   authorize access to their NDAs and their INDs, all

12   of the world would be a lot easier, and it is not

13   very often that you have the drug and the device

14   company in the same shop.           I mean, there are

15   companies like Johnson & Johnson and others that are

16   lucky enough to have both drugs and devices in the

17   same company, but very often the device company

18   doesn't have access to the drug company's files.

19                    So especially if the device allows a

20   broadened use of the drug the pharmaceutical company

21   would probably likely agree.            That's more drug sales

22   and, therefore, they're more likely to grant access

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           292


1    or authorize access to their master files or their

2    NDAs.

3                      But in some cases drug delivery devices

4    allow a more optimal and efficient delivery of the

5    drug, and therefore, less drug is going to be sold

6    if that drug delivery device is approved by FDA, and

7    that's a disincentive for PhRMA to cooperate with

8    the device industry.

9                      So what are the regulatory implications

10   here?        Without    PhRMA cooperation the device

11   companies have a very difficult time obtaining NDA

12   approval, as we saw with the example I used earlier

13   with respect to Actigol and the biliary lipotripsy

14   paradigm.

15                     The applicability of Section 505(b)(2)

16   is an issue presently on the table.                   505(b)(2), it's

17   not really a paper NDA, but it's like a paper NDA

18   where you rely upon literature and existing data to

19   avoid having to file a 505(b)(1) brand new and

20   spanking new NDA.

21                     And query whether a device company using

22   the 505(b)(2) process can with a different, let's

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          293


1    say, route of administration and a clear drug

2    product, can they then rely on 505(b)(2) without a

3    drug manufacturer even on the horizon to get their

4    product through?

5                        A real tough issue.        I don't have the

6    answer.          It is something that a lot of companies are

7    looking at, and it is one way for the companies to

8    proceed.

9                        The regulatory pathway conundrum:

10   should a 510(k) or PMA be required with an NDA for

11   each new drug delivery device?                In the QLT example

12   with photodynamic therapy we had sort of a pullout

13   PMA for the lasers and the fiber optics that went in

14   at the same time as the drug NDA, and believe it or

15   not, it was a miracle.             The three PMAs and the NDA

16   were all approved on the same day.

17                       That example worked out well there.

18   There have been other examples.                There was actually

19   a 510(k) with a pullout NDA for these H. pylori

20   breath detection devices where you had C-13 labeled

21   "urea" having to be approved by the device center.

22   That pullout didn't work real well.

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         294


1                     So I think what I'm pushing for is more

2    along the lines of what the FDA has done with drug

3    eluting stents.       If you look at the Cypher labeling,

4    it looks like drug labeling through a lot of the

5    package insert, and I believe that with these

6    combination products, you can mix drug labeling with

7    device labeling to appropriately reflect the

8    intended use of the device with the appropriate

9    precautions and warnings such that the user of the

10   product will have information that's appropriate for

11   both the drug and the device side.

12                    The lead center conundrum, I'm going to

13   let Mark address this since I'm just about out of

14   time, but let me just say that we need a new

15   definition of primary mode of action.                 Primary mode

16   of action is one of those areas where we need

17   guidance from FDA.

18                    There is a very, very extensive database

19   of primary mode of action decisions under the RFD

20   process that have never been made public, and in

21   fact, I don't think FDA ever put them in one spot.

22   I think Mark is now gathering the historical

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                       295


1    precedents in this area, but I think we would all

2    feel very much more comfortable if we entered into

3    the process knowing more about what primary mode of

4    action means.

5                     Finally, just to sort of sum up here, I

6    believe that there should always be a preference for

7    one submission.       The idea that you have to go

8    through the NDA process and the PMA process or

9    510(k) process at the same time to me is not

10   optimal.

11                    And the idea that we had back in 1990

12   for a unitary approval mechanism, I don't know

13   whether you want to call it a CPA or a combination

14   product approval, but maybe we need new legislation

15   that would allow us to look at whether we still want

16   to keep primary mode of action as the standard.

17                    Do we want to have a new statutory

18   provision that would replace primary mode of action

19   and go to a uniform, a unitary combination product

20   approval to avoid what was just stated in the last

21   presentation, where you end up with a disconnect

22   between the way the drug center would treat the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         296


1    product and the way the device center would treat

2    the product?

3                       And I believe that many in industry

4    believe that there is a different approach to

5    product approvals within the device center and the

6    drug center.

7                       Conclusions.     Dual approvals, not

8    optimal.         I think most people would agree with that.

9     Primary mode of action, standard.                 We need a new

10   guidance.        We think it's outdated.

11                      I believe that guidance documents with

12   respect to specific classes of drug delivery devices

13   would be very, very helpful.

14                      How about guidances with respect to drug

15   eluting stents?         What do you expect on the drug

16   side?        What do you expect on the device side?

17                      Nasal inhalation devices, what do you

18   expect on the drug side?            What do you expect on the

19   device side?

20                      A lot of work.      It's going to require a

21   lot of coordination between the centers, but

22   specific product area guidances would be very, very

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       297


1    helpful to the companies going through the process.

2                     A uniform, unitary drug delivery device

3    mechanism, such as a combination product approval,

4    that would be great.         More involvement by the Office

5    of Combination Products.          I'm not trying to get more

6    staff for Mark, but I believe that it would be

7    extremely helpful for the really novel drug delivery

8    devices and combination products for somebody in the

9    Office of Combination Products to have liaison

10   responsibility with the centers, not that they need

11   adult supervision.        It's just it would be helpful to

12   have some liaison and someone that's involved in the

13   process from the very beginning to help negotiate

14   and have a liaison between the centers.

15                    Thank you very much.

16                    (Applause.)

17                    DR. JACOBSEN:      Thank you very much,

18   Jonathan.

19                    Jon mentioned in his talk about Pat

20   Schraeder being an early player in combination

21   products, and Pat sends her apologies to everyone.

22   She intended to be here to represent AdvaMed and to

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        298


1    give the device industry's perspective, but she had

2    to cancel at the last minute, and we're very

3    fortunate to have her colleague, Keith Smith, who is

4    Director of Regulatory Affairs from BD who has

5    graciously agreed to present this perspective in

6    Pat's place.

7                      And then Nancy Isaac, Vice President for

8    Regulatory Affairs and Quality at Aerogen, is going

9    to take her place later today on the FDA industry

10   panel.

11                     So with that, we'll turn it over to

12   Keith.

13                     MR. SMITH:     Thanks.

14                     Good afternoon.       I'm sure most of you

15   know Pat.        So I certainly don't look like Pat or

16   talk like Pat, but I'm going to do my best.

17                     Okay.   My name is Keith Smith.         I'm

18   Director, Regulatory Affairs at Beck and Dickinson,

19   but I am here today as a member spokesman on behalf

20   of AdvaMed or Advanced Medical Technological

21   Association.

22                     AdvaMed is the largest medical

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       299


1    technology association in the world, representing

2    more than 1,100 innovators and manufacturers of

3    medical devices.

4                     One of AdvaMed's principal roles is to

5    support laws and policies that foster innovation and

6    bring safe and effective technologies, including

7    novel delivery systems, expeditiously to the market.

8                     In January, the FDA announced a new

9    initiative to help make certain innovative medical

10   technologies available sooner and to reduce the cost

11   of developing safe and effective medical products.

12   While still maintaining FDA's traditional high

13   standards of consumer protection, we applaud the

14   agency for identifying as one of the core areas of

15   attention of this initiative novel drug delivery

16   systems.

17                    Novel delivery systems are an important

18   subset of combination technologies ranging from

19   implantable infusion pumps to magnetically based

20   delivery devices, to systems that automatically

21   deliver anesthesia drugs in response to a patient's

22   vital signs.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       300


1                     The new technology intended to improve

2    targeting of chemotherapeutics by blocking blood

3    flow, novel delivery systems were identified as a

4    priority area for FDA's initiative because they

5    represent an exciting area of technology development

6    with potential to significantly improve patient

7    therapy and public health yet are often slow to

8    reach market due to complexities and uncertainties

9    in the pre-market review process.

10                    Our discussion today focuses on these

11   pre-market complexities and uncertainties and how we

12   might improve our regulatory processes so as to

13   further the Commissioner's goals of encouraging

14   delivery system innovation.

15                    The comments we provide summarize the

16   principal concerns and recommendations received from

17   AdvaMed member companies on the three questions

18   identified in the June 5th Federal Register notice.

19                    The first and most general and

20   overarching of the agency's questions asked that we

21   identify current critical challenges in developing

22   and bringing to the market novel delivery devices.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       301


1                     As an initial comment, we are gratified

2    that some of the historical challenges relating to

3    regulatory processes are beginning to be addressed.

4    As you know, AdvaMed, working closely with FDA and

5    Congress, helped implement Section 204 of MDUFMA

6    which, among other things, created for the first

7    time in the Office of Combination Products having as

8    one of its key functions to serve as an advocate for

9    combination technology, including novel delivery

10   systems.

11                    MDUFMA also provided a statutory

12   directive for the office to help ensure timely and

13   efficient premarket process and to establish dispute

14   resolution mechanisms should impediments arise

15   during those processes.

16                    With this new law, we have an important

17   first step to refining and improving premarket

18   systems in this area.

19                    Challenges, however, remain; four in

20   particular, all relating to the fundamental

21   framework of premarket review, still requiring

22   further consideration, clarification, and consensus

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       302


1    if we are to preserve the progress we have made thus

2    far in combination technology and achieve further

3    improvements.

4                     First, we have conveyed previously we

5    must reaffirm the agency's past interpretation of

6    primary mode of action, which has allowed so many

7    innovative and important combinations to reach

8    market using device jurisdictional standards.

9                     Second, we must refine and preserve the

10   agency's historic inter-center practice of applying

11   flexible approaches to cross-labeling, an issue that

12   arises not just at the end of a premarket review,

13   but also early on in setting jurisdiction and

14   defining pathways for many novel delivery systems.

15                    Third, we need to create new guidances

16   that allow for more creative and flexible approaches

17   to data development for this class of products, with

18   the device authorities clearly and consistently

19   applied for the device and/or the device component

20   parts of these reviews.

21                    And finally, we need better

22   understanding and clarification of those

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       303


1    circumstances where parallel path review may or may

2    not be appropriate.

3                     These four framework issues are, of

4    course, not the only instances that require

5    continued collaboration attention.               For example,

6    enhanced communication and transparency, greater

7    predictability of data requirements, and further

8    efforts to reduce the number of review cycles are

9    all important areas for ongoing improvement.

10                    However, our focus today is on a broader

11   framework of challenges, using the premise that if

12   the framework itself is first optimized to foster

13   innovation and to reduce needless data burdens and

14   avoidable delays, secondary product improvements

15   more easily fall into place.

16                    First and foremost among industry

17   challenges in the jurisdictional standard of primary

18   mode of action and reaching consensus with the

19   agency on the appropriate interpretation of this

20   term.

21                    Before discussing this issue, however, a

22   brief comment on the agency's reference to novel

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        304


1    drug delivery systems in this context.                The term

2    suggests devices serving to deliver a drug which may

3    inadvertently misdirect primary mode of action

4    analysis and thus inadvertently misdirect

5    jurisdiction.

6                     For example, some of the devices listed

7    in the Federal Register and agency press releases

8    and Web announcements leading up to this meeting,

9    including orthopedic products containing

10   biomaterials, hyperthermia/drug combinations, and

11   drug eluting stents.

12                    In each of these cases, the device

13   component has been determined to provide the primary

14   mode of action with the drug facilitating the

15   device's performance.         These are not drug delivery

16   systems for purpose of jurisdictional

17   determinations.       For this reason, we suggest not

18   using the term "delivery systems" unless the primary

19   intended use of the device is, in fact, to deliver a

20   drug.

21                    Without this subtle but important

22   clarification, there may be undue and potentially

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                       305


1    misleading emphasis on the jurisdictional role of

2    the drug component.

3                     In interpreting primary mode of action,

4    and as we have conveyed on a number of prior

5    occasions, AdvaMed's member companies have come to

6    rely and build their combination business around two

7    fundamental interpretation standards that have now

8    been replaced for more than a decade.

9                     First, the combined product, that is,

10   the product as a whole is analyzed for purposes of

11   determining the primary mode of action.

12                    And second, mode of action is determined

13   based on the primary intended function of the

14   combined product.

15                    The principal theme of the CDRH-CDER

16   inter-center agreement, as you know, provides that

17   products which are primarily structure, physical

18   repair or reconstruction purpose should be regulated

19   as devices.

20                    For the inter-center agreements from our

21   RFD decisions and from informal center assignments

22   over the years, there has emerged a long and varied

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       306


1    list of combination products granted primary device

2    status based on the intended function of the

3    composite product.        Among them, human fibroblast

4    derived skin substitutes, bone cements containing

5    antimicrobial agents, spinal fusion products

6    containing biomaterials, dental devices with

7    fluoride, and condoms with contraceptive agents.

8    All of these examples may deliver a drug or a

9    biologic, but that function was not deemed the

10   primary intended function of the combined product

11   for jurisdictional purposes.

12                    FDA's historic interpretations of

13   primary mode of action have served both the agency

14   and industry well.        They have fostered innovation,

15   on one hand, and protected and preserved public

16   health on the other, the precise two goals of the

17   Commissioner's new initiative

18                    Innovation has been fostered because of

19   the legal and policy initiatives that are uniquely

20   available under our device premarket review

21   structure, including early collaboration meetings,

22   100-day meetings, modular reviews, least burdensome

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       307


1    review principles, and humanitarian device exemption

2    initiatives.

3                     From the public health perspective, we

4    have had over a decade of combination assignments to

5    CDRH, and to our knowledge, not a single post-market

6    safety issue has arisen as a result of these

7    assignments.

8                     For these reasons, maximum use of device

9    jurisdiction authority should be encouraged.                 If

10   Commissioner McClellan is to truly accomplish his

11   initiative in making innovative medical technology

12   sooner and reducing the cost of developing safe and

13   effective medical products while maintaining

14   standards of consumer production since CDRH

15   jurisdiction over a combination has a demonstrated

16   effective review history in these instances where

17   primary mode of action is otherwise unclear.

18                    And companies believe that a device

19   assignment would serve to foster and advance their

20   technologies.      Strong deference should be given to

21   this principle.

22                    For the subset of combination products

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       308


1    that, in fact, serve to deliver a drug, for example,

2    new aerosolized insulin systems and lasers to

3    deliver topical anesthetics, there are other

4    jurisdictional principles that have been placed over

5    the years which like the agency's primary mode of

6    action, interpretation will be important to

7    preserve.

8                     For example, the inter-center agreement

9    provides that for drug delivery devices intended for

10   use with marketed drugs and used together as a

11   system, CDRH will have jurisdiction if the device

12   technology predominates.          From this jurisdictional

13   interpretation, whole industries and, indeed, whole

14   new standards of care have been born.

15                    Elastomeric infusion pumps, for example,

16   are delivered systems that historically have been

17   granted device review.          CDRH jurisdiction and

18   related innovations under our device authorities

19   have allowed this delivery system technology to

20   progress and evolve quickly from hospital to home-

21   based patient use, bringing improved standards of

22   patient care and significant cost savings to our

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       309


1    health economy.

2                     The challenge of cross-labeling.             A

3    second and particularly significant challenge for

4    novel drug delivery systems is cross-labeling.

5    Since 1991, when the agency first articulated its

6    framework for combination products, including how

7    labeling must conform for these products, market

8    introduction of novel delivery systems have been

9    aided tremendously by FDA's flexible approach to

10   cross-labeling issues.

11                    For the last decade, cross-

12   labeling/mutual conformance have been defined

13   through the inter-center agreement.               From this

14   inter-center agreement important guidance has been

15   provided both to FDA and industry on the issue of

16   cross-labeling, used not simply for final labeling

17   discussions, but also early on and concerning

18   framework/jurisdictional issues for novel devices

19   intended primarily to deliver drugs.

20                    The inter-center cross-labeling

21   standards are fourfold.          First, the inter-center

22   standard states there are three essential aspects of

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       310


1    drug labeling requiring mutual conformance:

2    indications, general mode of delivery, drug

3    doses/schedule equivalence.

4                     If device labeling is generally

5    consistent with these key parameters of drug

6    labeling, the essential elements of mutual

7    conformance will be assumed.            When there is general

8    mutual conformance, the agreement states that the

9    FDA should do two things.           It should grant CDRH

10   jurisdiction for the product, and it generally

11   should waive additional clinical showing of drug

12   effectiveness.

13                    A second standard.        The agency has

14   recognized that as delivery system technology

15   evolves, models of delivery and dose schedules for

16   drugs may inevitably be refined.              To accommodate

17   these refinements, two of the three key drug

18   parameters in the standard are described with some

19   flexibility.

20                    Specifically, the mode of delivery need

21   only been the same general mode of delivery, and the

22   doses/schedule need only be equivalent.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         311


1                     Also, the term conformance.            Using the

2    standard does not convey verbatim replication of or

3    precise equivalence to drug labeling.                 Device

4    labeling need only be generally consistent with the

5    labeling of the drug intended to be delivered.

6                     Examples of the precedents that have

7    relied on the flexibility of this labeling standards

8    include continuous delivery devices for insulin and

9    fibron sealant mixing and delivery systems.

10                    As a third inter-center cross-labeling

11   principle, even if there are changes to these three

12   critical drug parameters described in the cross-

13   labeling standard, the standard nevertheless affords

14   CDRH further flexibility to consult with CDER and to

15   resolve those issues through device labeling.

16                    And finally, the inter-center agreement

17   on cross-labeling does not purport to address any

18   other secondary aspects of the drug labeling beyond

19   the three stated parameters of indications, mode of

20   administration and dosage.           Under this

21   interpretation second drug labeling issues have been

22   available to be addressed through device labeling

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                       312


1    and review.

2                     In keeping with the Commissioner's goal

3    of encouraging innovation in this area, we ask that

4    these four historic cross-labeling standards,

5    reaffirmed through agency device reviews over the

6    years, continue as policy practice in this area.

7    Without these flexible policy approaches,

8    significant new challenges will be added to pathway

9    development for this category of products.

10                    A third challenge for novel delivery

11   system relates to data burdens and the need for new

12   guidance that permit more flexible, more

13   predictable, and more consistent approaches to data

14   development.      In the novel delivery system context,

15   data challenges can sometimes be very different

16   depending on whether CDRH or another center has

17   received primary jurisdiction for the composite

18   product.

19                    Compounding these challenges is the

20   reality that, in contrast to certain other forms of

21   combination products, delivery system technology

22   often involves two severable components, and review

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       313


1    standards for those components are not always clear

2    or even applied.

3                     If CDRH jurisdiction has been granted

4    for delivery systems, historically it has been

5    because the drug is marketed, has been generally

6    approved, and the device issues thus predominate. In

7    this context two data challenges have emerged.

8                     First, our members believe that there

9    needs to be stronger emphasis on the principle first

10   articulated in our inter-center agreement that

11   whenever possible, delivery systems need not reprove

12   the fundamental efficacy of a drug already approved

13   for the same general mode of administration, dosage,

14   and indication.

15                    In reaffirming this historic standard,

16   our members ask that the agency provide concrete and

17   specific guidance through examples as to how this

18   principle can be more effectively and consistently

19   applied.

20                    As a second challenge our members also

21   feel strongly that any CDER consult process, while

22   important to resolving unsettled drug issues, not be

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          314


1    permitted, directly or indirectly, to set the review

2    standards for the composite product.                  CDRH product

3    jurisdiction, if it is to be meaningful, necessarily

4    must involve device authorities.

5                     Defining the combined product.             In this

6    context we need to make certain that the tried and

7    true drug standards are not applied to combination

8    technologies.      These technologies represent

9    breakthrough thinking and application of established

10   drug standards may not in most instances be an

11   appropriate standard for review.

12                    CDRH has a long history of establishing

13   flexible standards because of the nature of the

14   products they regulate.          This history gives CDRH a

15   unique experience based on the development of review

16   criteria for those novel products.

17                    In instances where CBER and CDER have

18   granted jurisdiction of novel delivery systems, it

19   is generally the case that both aspects of the

20   product, both the device and the drug that are

21   biologically being delivered, have been deemed

22   investigational.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                       315


1                     In this context, industry data

2    challenges are somewhat different.               First, for the

3    device combination of delivery system combination,

4    the agency needs to be clear that device

5    authorities, including least burdensome principles,

6    frame the review for the aspect of the product.

7    This component part of the evaluation could occur

8    through separate review and/or consultation process

9    at the sponsor's discretion, but it is important

10   that it be undertaken effectively.

11                    Too many of our members have expressed

12   concern with the agency's internal assessment,

13   published last October, which acknowledged that some

14   reviewers in CDER and CBER lacked fundamental

15   understanding or appreciation of advice premarket

16   review authorities.

17                    The ability to ensure proper device

18   review becomes more important the more complicated

19   the device design, and complexities are increasing

20   reality for delivery system as many new technologies

21   involve software electronics, electromagnetic

22   principles, ultrasound energy, and other

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       316


1    sophisticated forms of device engineering.

2                     As a second more general challenge, our

3    members convey that when CDER and CBER jurisdiction

4    is granted, there is little, if any, incentive at

5    the moment for reviewers in those centers for

6    seeking mechanisms or employing standards and

7    encouraging the development of novel drug delivery

8    systems.

9                     If we are to achieve meaningful

10   premarket improvements in this area, it will be

11   important to develop new guidance specifically

12   addressing delivery system combinations and

13   acknowledging the sentiments expressed at the

14   agency's January 31st press release which launched

15   this initiative.

16                    At the risk of repeating ourselves in

17   that release, the Commissioner described the

18   agency's desire to help make innovative delivery

19   systems available sooner and to reduce needless

20   costs and burdens while ensuring safe and effective

21   medical products.

22                    We believe this initiative represents a

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       317


1    form of least burdensome philosophy now sanctioned

2    expressly under our device laws.              New guidance for

3    novel delivery systems should attempt to reflect

4    this standard as appropriate and consistent with the

5    current law.

6                     We believe that separate guidance

7    specifically encouraging and promoting novel

8    delivery system development will give CDER and CBER

9    reviewers one more reason to think creatively and

10   flexibly about data issues and to avoid any

11   temptation for more doctrinnaire data demands.

12                    A fourth and final issue that challenges

13   the framework for premarket review of novel delivery

14   systems is the subject of separate parallel past

15   submissions, and better understanding those

16   circumstances where parallel review may or may not

17   be appropriate.

18                    In the November hearing on combination

19   products, we let the agency know that our member

20   companies see the advantages and disadvantages of

21   separate applications in different ways at different

22   times, depending upon the specific regulatory,

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       318


1    factual, and business circumstances presented by the

2    particular combination.

3                     We believe, however, that these

4    differing views may be fully reconciled by

5    distinguishing required separate filings that may be

6    an option of the sponsor.           Several specific

7    recommendations highlight and explain how this

8    distinction would be implemented.

9                     First, in order to avoid redundant

10   reviews and excessive regulation, only one filing

11   should be required in the majority of the cases.

12   Indeed, we believe that as the consultative process

13   continues to be regulated and improved and held

14   accountable, there should be fewer and fewer

15   mandated separate applications.

16                    There are certain selected

17   circumstances, particularly for novel delivery

18   systems, where a company at its option might see a

19   separate filing as useful for regulatory

20   business/marketing reasons.            Factors include:

21                    One, where two different companies, for

22   example, a drug company and a device company, are

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       319


1    involved in the manufacturer of a combination drug

2    delivery system.

3                     Two, where delivery system components

4    are expected to have separate distribution and

5    use/reuse patterns.

6                     And, three, where primary jurisdiction

7    for the combination delivery system has been given

8    to the center other than CDRH and the delivery

9    device component is capable of being separately

10   defined and reviewed.

11                    Examples include novel ultrasound

12   infusing catheters, nebulizers, jet injectors,

13   insulin pens, and drug delivery systems that monitor

14   a patient's vital signs.          In these circumstances

15   AdvaMed believes that the separate filings are

16   appropriate.

17                    The key to this recommendation, however,

18   is that the option of the dual filings is left up to

19   the sponsor.      We believe this theme of flexibility

20   and sponsor discretion is important if we are to

21   encourage the development of novel drug delivery

22   systems in an industry with such a wide array of

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       320


1    corporate and technological interests.

2                     Your next two questions inquire about

3    areas where guidance would be helpful on how the

4    agency can best collaborate with industry and other

5    institutions in the development and encouragement of

6    novel delivery system technology.

7                     Given the commonality of themes

8    presented by these two questions, we have

9    consolidated a response and have several

10   recommendations to provide.            We believe the four

11   framework issues just discussed should be reaffirmed

12   in separate or consolidated guidance documents, and

13   those documents should be developed following notice

14   and comment processes required by good guidance

15   practices.

16                    Further agency collaboration with

17   industry on development of these documents also

18   would be beneficial.         We also agree with the agency

19   that a drafting process which is as interactive as

20   possible, for example, through stakeholder meetings,

21   would allow for further debate and reiterative

22   refinement of FDA and public views on those

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       321


1    important issues.

2                     As part of the guidance process we also

3    recommend that the agency's initiative and intent to

4    encourage novel delivery systems be fairly stated

5    and specifically supported.            In particular, industry

6    would appreciate receiving concrete examples of how

7    the agency process to reduce needless delays and

8    avoidable product developments cost in the premarket

9    process.

10                    We believe suggestions that the agency

11   already has made concerning improvement and review

12   in communications and proceduralizing combination

13   reviews will facilitate the agency's goals, but we

14   request that additional mechanisms for more

15   efficient review processes and further encouragement

16   of flexible review standards be considered as well.

17                    Implementation of this initiative will

18   work best if all aspects of the agency's review

19   chain are trained well on the principles adopted.

20   If the agency is to have all reviewers consistently

21   thinking creatively and flexibly in this area, there

22   must be regular, internal reminders of this goal to

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        322


1    all three centers involved.

2                      Industry would also appreciate ongoing

3    efforts to make combination product include a novel

4    delivery system database as transparent and as

5    informative as possible, consistent with FDA's

6    nondisclosure obligations and the proprietary

7    interests of sponsoring companies.

8                      Data on approved products should convey

9    primary jurisdiction, time frame for reviews,

10   available information on consultative or

11   collaborative processes invoked, the number of

12   review cycles involved, and public summaries for

13   review.

14                     This database should be separate and

15   apart from other databases for approved products to

16   facilitate industry's efficient review of

17   combination precedents.

18                     With those recommendations, AdvaMed

19   thanks the agency for its consideration of our

20   comments.        Our members strongly support the agency's

21   ongoing efforts in this area, and we look forward to

22   working closely with you to further reduce

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       323


1    regulatory challenges and to improve premarket

2    processes so as to foster and facilitate innovation

3    of delivery systems and other forms of combination

4    technology.

5                     Thank you.

6                     (Applause.)

7                     DR. JACOBSEN:      Thank you, Keith.

8                     And finally, Christine Allison from the

9    Global Regulatory Affairs Group at Eli Lilly is

10   going to discuss the drug industry's perspective on

11   combination products.

12                    MS. ALLISON:     Thank you.

13                    First, I'd like to thank FDA for

14   sponsoring this very important workshop.              For those

15   of us that have been working on this type of

16   combination products for years, this is exactly what

17   we have been looking for, an opportunity to have an

18   open dialogue with the agency and to discuss about

19   some of the issues and challenges we have been

20   dealing with and struggling with on a daily basis.

21                    I'm also very honored to be invited as a

22   speaker today.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       324


1                     My presentation goals today is first to

2    give you a brief introduction of the type of

3    products that my company has been working with, and

4    also hopefully walk you through some of the

5    regulatory challenges that we have experienced

6    during development and market applications and post-

7    approval, and then touch a little bit on the

8    challenges that we have experienced working with

9    partners, and some global challenges, and then I'll

10   summarize some key points           and conclude it with our

11   recommendations to the agency.

12                    Lilly's experience on the combination

13   product is mainly on the drug-device combination.

14   We are currently working on several innovative

15   products, for example, pulmonary inhalation system

16   for systemic delivery of drugs and also other, you

17   know, interesting, innovative products, and those

18   are all at the development stage.

19                    We also have many years of experience

20   working on the pen injectors, which is already in

21   the market, and for those products we have post

22   development and post approval experience.              Although

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       325


1    I don't consider pen injectors as an innovative

2    product, however, I believe that some of the post

3    approval experience that we have will be good

4    examples for us to look forward once the innovative

5    product has been approved in the market.

6                     My presentation will also be focused on

7    the CMC issues.

8                     For innovative products, a lot of time

9    the questions surface very early in the development

10   stage, even before we are ready to request for lead

11   center designation.        A lot of time we will have a

12   lot of questions, sometimes drug questions or device

13   questions, and we often struggle which center we

14   should go to to ask those type of questions.

15                    So it would be very nice if we have a

16   single focal point so we can just channel those

17   questions to.      So we recommend that the Office of

18   Combination Products be the coordinator and

19   facilitator for identifying the appropriate centers

20   for technical consultation prior to lead center

21   designation.

22                    Some of the challenges we have

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       326


1    experienced is also the consistency in lead center

2    designation.      We understand that the statutory

3    history does not mandate that the sponsor has to go

4    to the Office of Combination Products to request for

5    designation of lead center.

6                     So sometimes the sponsor can choose to

7    go to individual center instead of go to the Office

8    of Combination Products.          So this could result in

9    some similar products, combination products that

10   result in different lead center assignments.                 It

11   depends on which center the sponsor goes to first.

12                    And often the consultation centers are

13   not defined at the time of lead center assignments.

14   So we suggest that internal procedure be developed

15   to guide each center for routing those requests to

16   the Office of Combination Products for review to

17   ensure the consistency of lead center designation.

18                    And also we recommend that Office of

19   Combination Products also identify those consulting

20   centers at the time of lead center assignments.

21                    Some of the major CMC challenges that we

22   have experienced for innovative drug delivery

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       327


1    systems, because this type of product is very new

2    and is kind of unique, and because a lot of times

3    the agency has no experience with dealing with this

4    type of product, a lot of time the agency will

5    request a commercial system to be used for pivotal

6    studies.

7                     And this creates a lot of technical

8    challenges for us.        It means that we have to lock in

9    the CMC development process in a very early stage of

10   development, and this also prevents us to continue

11   to improve the process during the clinical phase and

12   feed it back to our design.

13                    In addition, early resource commitment

14   is required.      Sometimes we have to purchase

15   commercial equipment or even build manufacturing

16   sites in the very early stage of development,

17   sometimes even as early as Phase II or III.                It

18   depends on what kind of clinical plan that we have.

19                    And so this is a very typical approach.

20   It compares to the normal product development

21   process, and if we have a change that is not

22   avoidable, then we have to make those changes.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       328


1                     The difficulty is to establish

2    successful, satisfactory comparability protocol

3    between pre-commercial and commercial systems to

4    satisfy agency's expectation.             So we encourage the

5    agency to consider the role of reaching strategies

6    to allow product process improvements during the

7    development through commercialization, and also

8    clear and documented expectations from the agencies

9    are needed.

10                    And this is not an easy task.         We

11   realize this is not an easy task.              Therefore, I

12   think frequent dialogue with the agency regarding

13   specific issues is very critical throughout the

14   entire development process.

15                    The traditional pre-IND meetings, or end

16   of Phase II meetings, it is just not sufficient for

17   us.      Therefore, we encourage that the agency to be

18   flexible in granting the request for meetings and

19   consultations when dealing with this type of

20   product.

21                    Another major CMC challenge for us in

22   dealing with this type of innovative drug delivery

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           329


1    system is that the drug and device is an integrated

2    system.          This is an integrated system.           It's got

3    drug components in there and the device components

4    in there, and both components have to work together

5    as a system.

6                        And, therefore, it requires a lot of

7    time, with frequent consultation with multiple FDA

8    centers and sometimes multiple divisions within the

9    same center.

10                       Currently, based on our experience,

11   alignments and communication with multiple centers

12   and divisions has been a challenge, and therefore,

13   we believe that it would be very beneficial if

14   agency's review team can include members from all

15   relevant centers and divisions from the very first

16   sponsor meeting.

17                       Another major challenge that we face is

18   quality systems.          The question is which regulations

19   you apply for the drug-device combination.                   Is it

20   drug cGMP or should we apply the device QSR or both?

21    And which compliance guidance will be used during

22   the preapproval inspection?

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                          330


1                     It is our opinion that the drug

2    regulations should apply to the drug portions of the

3    products, and the device regulations should apply to

4    the device portions.

5                     And we also believe that clear policy is

6    needed with regard to the FDA inspections for

7    preapproval inspection of the combination products.

8                     And also, we encourage that the

9    investigators to be trained and to perform

10   combination product inspections using the

11   appropriate regulation for each component of the

12   combination.

13                    Another challenge is during the

14   development, is the regulatory reporting.                  It's

15   unclear what are the requirements for AE and device

16   reporting during the clinical study.                  Should we

17   follow the 21 CFR 312 or 21 CFR 812?

18                    Especially when it comes to the device

19   expedited reporting requirements for the device

20   portion submitted under the IND in terms of the

21   device malfunction and the inclusion of the device

22   investigation results.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                       331


1                     It is our opinion that in principle,

2    reporting requirements for both drugs and devices

3    should be applied as appropriate.              We believe that

4    if device malfunction is reportable under the device

5    regulation, it should be also reportable even if

6    submitted under the IND.

7                     We also   believe that device

8    investigation results should be included in the

9    report, and this report should be directed to the

10   lead center doing the review.

11                    As far as the reporting time, we have no

12   preference one way or the other as long as it is

13   clear to us what kind of reporting time we need to

14   follow.

15                    Moving on to the challenges during the

16   market applications, the question is always is it a

17   single or dual submission, and if it's dual

18   submission, would dual user fees apply?

19                    And also, what is the format we should

20   use to include those device information in a CTD

21   submission?      And what kind of device information

22   needs to be included in drug submissions?

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        332


1                      We support the concept of a single

2    premarket review mechanism leading to a single

3    approval of combination products.               We understand

4    that there will be exceptions when the DOS

5    submissions may be more appropriate.

6                      In terms of the formats, submission

7    formats, we recommend that standardized formats and

8    also provide guidance for us to include the data,

9    the data requirements for the device information to

10   be included in the CTD submission.

11                     In addition, the phase-appropriate data

12   requirements for the device to be included in the

13   INDs.        From our experience, a lot of times we find

14   out it's very        beneficial if we prepare the device

15   information in a format that is familiar with the

16   CDRH reviewer.        So if we prepare that information,

17   for example, in a 510(k) format, it will be much

18   easier for a CDER reviewer to hand it over to the

19   CDRH reviewer for consultation.

20                     Moving on to some of the post approval

21   challenges, the difficulties we have encountered the

22   most is when we have to deal with device changes,

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           333


1    when this device information is part of the NDA

2    submission, and currently because there is no clear

3    guidance on how to handle this, it's been a

4    challenge for us.

5                         We recommend the use of the CDRH 510(k)

6    decision tree as a guidance.                If we go through the

7    510(k) decision tree and the conclusion is we don't

8    need to have a 510(k), we suggest this type of

9    change will be communicated to the CDER reviewer

10   through the annual report.

11                        And if we go through the 510(k) decision

12   trees and the 510(k) is required,, then we suggest

13   that this type of change will be communicated to the

14    CDER reviewer through the NDA Supplement B or C.

15                        Again, talk about the post-approval

16   regulatory reporting requirements.                   There's

17   currently no clear guidance on how to conduct those

18   AE reporting and device reporting for drug-device

19   combination products.               Which regulation should we

20   apply?           Is it 21 CFR 314 or 21 CFR 803?

21                        The same challenges when we talk about

22   doing the IND stage is for those expedited

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        334


1    reporting, device reporting, do we include the

2    device investigation result in the report?

3                     It is our opinion that, in principle,

4    reporting requirements for both drugs and devices

5    should be applied as appropriate.              We believe that

6    if a device malfunction is reportable under the

7    device regulation, it should be reportable as well

8    when it is submitted through the NDA, and we also

9    believe that the device investigation results should

10   be included in the report.

11                    And those reports should be directed to

12   the lead center that has reviewed the submission and

13   approved the products.          Again, for the reporting

14   time, we have no preference one way or the other as

15   long as it's one clear reporting time that we have

16   to follow.

17                    Another challenge we have experienced

18   for the post-approval is the cross-labeling of

19   products intended to be used together.                An example

20   of some cases is that some of the 510(k) devices

21   approved in the market, cleared in the market can be

22   used for multiple products.            When the drug company

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                         335


1    wants to include those devices into the drug label,

2    there's no clear guidance how to do this, to reach

3    conforming labeling.

4                     And since there is not a user fee

5    associated with this type of labeling change,

6    therefore, there's no set reviewing times, and in

7    some cases it takes a long time to have this

8    labeling change accomplished.             Sometimes when we are

9    waiting for the approval the device that we try to

10   include in there, the model already is obsolete.                      We

11   know it is very dynamic in the device                 work.    So

12   that means that we have to then restart it again for

13   this whole entire reviewing process to include the

14   new versions of the model.

15                    And sometimes the reviewer may

16   repeatedly review the data set that has already been

17   reviewed by the other center, and in some occasions

18   the reviewer may request additional data beyond what

19   was required by the other centers.

20                    Therefore, we believe that clear

21   guidance is needed on how to obtain mutually

22   conforming labeling.         And we suggest that allowing

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                       336


1    the 510(k)-cleared device to be included in the drug

2    label is appropriate, and communication of this type

3    of labeling change be made in the annual report.

4                     To touch a little bit on the challenges

5    we have experienced working with partners, very

6    often a device company may work with multiple drug

7    companies with the same device platform.              So those

8    device information will be considered proprietary,

9    and when we have to submit an NDA to include this

10   device information, we will not be able to describe

11   those information in detail without reference to a

12   DMF.

13                    And the regulatory challenges come into

14   play when the reviewer wanted to discussion or has

15   questions regarding to those informations.               It would

16   be very difficult for the FDA reviewer to discuss

17   those issues with the sponsor due to the

18   confidentialities.        And this is the same challenge

19   once the product is in the market and we have

20   changes made in the device portion.

21                    And sometimes the device company may

22   have a different approach, regulatory approach or

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       337


1    interpretation than the drug company, especially in

2    a controversial area, such as we mentioned earlier,

3    some of the GNP requirements or regulatory reporting

4    requirements.

5                     And I believe that those differences can

6    be minimized once the agency has a clear guidance on

7    how to deal with those issues.

8                     Some of the global challenges that we

9    have experienced.       A drug device combination product

10   approved under the CDER NDA may require a market

11   authorization for the drugs and a CD marking for the

12   device in EU, and this reports a lot of challenges

13   in terms of submission document preparations,

14   quality system requirements, post-approval changes,

15   regulatory reporting, labeling, and compliance

16   inspections.

17                    And, therefore, it would be very nice if

18   the agency, when dealing with certain policies and

19   guidance, if you can work with your counterparts in

20   the other parts of the world and work toward a

21   direction of having a global harmonization.                That

22   would be very nice.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       338


1                     Therefore, in summary, we have

2    identified many challenges throughout the entire

3    product life cycle for innovative device products.

4    We also identify many areas that we need guidance,

5    such as quality system requirements, post approval

6    changes, regulatory reportings, and cross-labeling.

7                     And we believe continued dialogue

8    between sponsors and the agency is critical to

9    ensure successful development and timely review of

10   market applications.

11                    And in conclusion, we would like to

12   recommend that the agency when setting policy and

13   guidance, please consider using Office of

14   Combination Products as a single focal point to

15   handle the issues regarding the combination

16   products, and keep it simple.             If we can do it with

17   one process let's not use two processes.

18                    Reduce redundancy, especially in the

19   reviewing process.        If one center already reviewed

20   the data, the other center does not need to review

21   it again.

22                    And also when setting guidance and

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          339


1    policy, not only market applications, please

2    consider also post approval requirements and think

3    through the entire product life cycle.

4                       And the last if not the least, please

5    consider global harmonization needs.

6                       Thank you.

7                       (Applause.)

8                       DR. JACOBSEN:      Okay.     Thank you very

9    much.

10                      That's the end of this session, and I

11   think we're scheduled to have a break now.                 So if

12   you could all be back at 3:30, we'll get started

13   again with the FDA session.

14                      Thanks.

15                      (Whereupon, the foregoing matter went

16                      off the record at 3:17 p.m. and went

17                      back on the record at 3:33 p.m.)

18                      DR. JENKINS:     We will begin the FDA

19   session.         I'm John Jenkins.       I'm the Director of the

20   Office of New Drugs in the Center for Drugs.                   I'm

21   here substituting for Dr. Woodcock, the Center

22   Director, who was not able to be here because of a

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       340


1    conflicting schedule on her calendar.

2                     It's a pleasure to be here, and I think

3    from what I've been hearing in the hallway, this is

4    the session you've been waiting for, which is FDA

5    perspective on all of these issues.

6                     So I'm going to serve as a moderator,

7    and let me introduce our first speaker.               Mark Kramer

8    I think you all know.         He's the Director of FDA's

9    new Office of Combination Products.

10                    Mark.

11                    (Applause.)

12                    DR. KRAMER:     Thank you, John, and I'd

13   like to thank Dr. Feigal and Dr. Provost for

14   inviting me to be with you here today and talk about

15   what we're doing.

16                    The first thing I have to do is start

17   out by saying that I've been asked to focus on the

18   role of the Office of Combination Products and the

19   kinds of things we're doing.            What I wished we had,

20   and this is what I'm going to cover today, is just

21   to give an overview of what is and what is not a

22   combination product; give an overview of how we

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       341


1    regulate combination products at FDA; talk about the

2    role of our office and also some of the current

3    initiatives that we have gotten underway.

4                     I really wish I had a fifth bullet here,

5    which are the answers, the answers to all of the

6    questions that people have been raising today, but

7    as I think you'll hear, we are beginning to work on

8    these issues, and clearly these kinds of sessions

9    really help give us the kind of input that we need

10   in order to anticipate the products that are coming

11   down the line, and there's clearly a lot of new

12   things that I've heard today in terms of products

13   that we have to anticipate.

14                    So the first thing I wanted to do is

15   start out with a question.           Are novel drug and

16   biologic delivery systems combination products?

17                    And I think the answer is it depends,

18   and as I'm going to lay out the definition of a

19   combination product, I think we use the term

20   loosely, but really many of the products that we

21   talk about as being combinations might not meet the

22   regulatory definition of a combination product.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       342


1    However, they may still raise complicated regulatory

2    issues and, therefore, we sort of, you know, are

3    trying to address these types of products at the

4    same time.

5                     This is a regulatory definition of a

6    combination product, and the regulation provides --

7    this is in 21 CFR Part 3, 3.2 -- there are really

8    four different types of combination products, and

9    the third and the fourth bullets are tied together.

10                    But sort of the quintessential

11   combination product is a product where the product

12   itself comprises two or more regulated components

13   that are physically, chemically, or otherwise

14   combined or mixed as a single entity, and a good

15   example there would be the drug eluting stent that

16   has been discussed much today.

17                    But it has its forbears, things like

18   antimicrobial coated catheters, heparin coated

19   catheters, condoms with spermicide that have been

20   around really for a long time, and those are

21   combination products, too, and we have been

22   effectively regulating those for quite a long time.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       343


1                     The second type of combination is where

2    we have a kit or a co-package that in itself is

3    comprised of separate products, drugs and devices,

4    devices and biologics, or drugs and biologics.

5    Together they create one product which is a

6    combination product because it contains different

7    types of regulated articles.

8                     And the third category is really one of

9    the most complicated for us, and this is where you

10   have separate products, often provided by or

11   manufactured by separate companies that their use

12   together constitutes a combination product.                Both

13   products are required to achieve the intended use,

14   indication, or effect, and where upon approval of

15   the proposed product, the labeling of the approved

16   product would need to be changed.              And this is the

17   so-called cross-labeling issue that's been raised by

18   a couple of the speakers earlier this afternoon.

19                    In the interest of time I'm going to run

20   through these examples pretty quickly, but they are

21   in your notebook, devices coated, impregnated or

22   otherwise physically combined with a drug or

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         344


1    biologic.        I gave some examples of the drug-device

2    combinations.       There are also drug-device biologic

3    examples, such as skin substitutes with cellular

4    components, orthopedic implant with growth factors,

5    and there was one of those that was recently

6    approved in the least year.

7                      Prefilled drug or biologic delivery

8    devices are also combination products.                 Some of the

9    simplest ones that we have are just prefilled

10   syringes, a syringe that is filled with a drug or

11   biologic is a combination product because the

12   syringe is a device and the drug or biologic

13   obviously is regulated separately.

14                     But we also have insulin, epinephrine,

15   interferon injector pens, metered dose inhaler,

16   transdermal patches, again, all examples of

17   combinations.

18                     Drug or biologics that are provided with

19   an applicator or delivery device; drug-biologic

20   combinations.       We haven't spoken about that too much

21   today, but radiopharmaceuticals combined with a

22   biologic or monoclonal antibodies combined with a

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                       345


1    chemotherapeutic drug, interferon-ribavirin

2    combination for Hepatitis C.            These are drug-

3    biologic combinations.

4                     And then again in that last category of

5    combinations, separate products that may constitute

6    a combination:       a hyperthermia device used with a

7    chemotherapeutic drug; photodynamic therapy drug and

8    laser light source, and you'll be hearing from

9    Richard Felten about that soon after my talk.

10                    Diagnostic devices that require the

11   administration of a particular drug or biologic, or

12   a drug requiring a specific diagnostic device.

13   These are examples of separate products that used

14   together might constitute a combination.

15                    What are not combination products?

16   Well, combinations of two drugs, two devices, or two

17   biologics.       They may raise some of these types of

18   regulatory issues as well, but in order to be a

19   combination product by the regulation, you have to

20   comprise different types of regulated articles:                   a

21   drug and a device; a drug and a biologic; or a

22   device and a biologic.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       346


1                     Most concomitant use of drugs, devices

2    and biologics is not a combination product, and also

3    general drug or biologic delivery devices, such as

4    an infusion pump that's not intended for use with an

5    individually specified drug or biologic product,

6    they don't meet the definition of a combination

7    product.

8                     And I think those are some of the types

9    of things we're discussing here today that may

10   actually fall in that last bullet and not

11   technically meet the definition of a combination,

12   but may pose some of the very same issues that

13   combination products raise.

14                    These are the various regulatory

15   approaches we have in our armamentarium.               Devices

16   generally get approved under the PMA or 510(k)

17   process and are investigated under IDE.               Drugs

18   approved via NDA, studied under IND, and biologics

19   under BLA and studied under an IND.

20                    And what somebody once told me at the

21   last talk I gave where I had this same slide was

22   it's like worlds colliding.            If anybody remembers

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         347


1    Seinfeld, there was an episode where George felt his

2    worlds were colliding.          His girlfriend was getting

3    to know his friends and his work people, and he

4    didn't like that.

5                     And somebody told me once that's what

6    happens with combination products.

7                     (Laughter.)

8                     MR. KRAMER:     But when we look at the

9    intersection here, I think it's what makes

10   combinations unique because we have the regulatory

11   flexibility to apply the most appropriate regulation

12   to a combination product by tailoring the approach

13   to taking pieces of drug regulation, pieces of

14   device regulation or biologics as appropriate.

15                    But what we haven't had in the past was

16   a very consistent way of doing that, and that's what

17   we're in the process of doing.

18                    Some of the things that are unique about

19   the way we regulate combination products, first, as

20   you heard earlier, they're assigned to a lead center

21   based on the s-called primary mode of action.                  As

22   Jonathan pointed out, not defined in the law, not

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       348


1    defined in our regulations, although we are in the

2    process of formulating a regulatory definition for

3    primary mode of action that will be made available

4    for public review and comment, and we feel that's a

5    very important first step in the process of ensuring

6    that these products are appropriately regulated and

7    that we have a good way for determining which center

8    will have lead review responsibility.

9                     Another one of the hallmarks of

10   combination product regulation is that we often, but

11   not always, have consultation or collaboration

12   between the centers.         That is a way of applying the

13   best mixture of expertise to insure that one center

14   can supplement its expertise in order to best

15   understand and tackle the review issues associated

16   with that product.

17                    Jonathan asked me to touch on the

18   difference between consultation and collaboration,

19   and I'll try to do it in ten seconds, but

20   consultation is what we generally have.               In most of

21   the cases this is where the lead center is

22   ultimately responsible for all decision-making on an

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       349


1    application.

2                     But we also have an approach of

3    collaboration that allows basically the two centers

4    to have equal votes at the table, and both centers

5    would need to reach agreement on the outcome of a

6    submission in order to approve or disapprove the

7    product.

8                     I mentioned earlier that we do have the

9    flexibility to tailor the premarket regulatory

10   authorities, and we have the same flexibility to

11   tailor the post-market regulatory authorities.                  So

12   we may have a product that might be subject, for

13   example, to elements of the quality system

14   regulation and to elements of CGMPs.

15                    And we have done that, in fact, with

16   drug eluting stents where the drug substance needs

17   to conform to drug GMPs up until the time that it's

18   coated on the stent, and once it's a combination

19   product, then the combination is subject to the

20   quality system regulation.

21                    The other thing is one application

22   versus two, and this is an important issue not only

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         350


1    because it affects, you know, really the whole

2    regulatory landscape of how these products are

3    regulated, but some companies, as AdvaMed pointed

4    out, prefer one application and some prefer two.

5    And sometimes there are business reasons that affect

6    what a company's preference is.

7                     And we're trying to look at tailoring

8    our approach in terms of making it as -- our

9    ultimate goal is to try to have one application

10   whenever we can, but we recognize that there will be

11   instances where two will be most appropriate in

12   order to regulate a product, and there are cases

13   where a company may actually prefer to have two even

14   if we feel they only need one.

15                    There are user fee issues associated

16   with that.       I think Christine mentioned that in her

17   presentation, and therefore, there are important

18   ramifications to a decision as to whether to require

19   one or two applications.

20                    The Office of Combination Products was

21   established in December of this year.                 We have six

22   main roles that are outlined in the statute, and

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                       351


1    this was all part of the Medical Device User Fee and

2    Modernization Act.

3                     We have responsibility for assigning

4    combination products based on the primary mode of

5    action to ensure the timely and effective premarket

6    review of combinations, to ensure the consistent and

7    appropriate post market regulation of combination

8    products.

9                     We also have a role with respect to

10   dispute resolution, review and update of guidance

11   agreements and practices relative to the assignment

12   of combination products, and we have to report to

13   Congress on an annual basis on the activities and

14   impact of the Office and provide some prescribed

15   data.

16                    In terms of the assignment of

17   combination products, the statute tells us that we

18   have to promptly assign an agency center with

19   responsibility for jurisdiction of a combination

20   products, and our goals there is to have as

21   efficient an RFD process -- that's the request for

22   designation process -- as possible and to make it as

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         352


1    consistent, transparent and predictable as possible.

2                      Some of the things we're doing I already

3    mentioned.       We are developing a definition of

4    primary mode of action, which we think is probably

5    one of the most important steps of this process, but

6    we're also working on guidance on the selection of

7    premarket authorities so that our reviewers

8    understand what tools are available to them in order

9    to regulate a combination, but that we have a

10   framework in order to do it in a consistent and

11   transparent and more predictable way so sponsors

12   will have a better understanding of how their

13   product will be regulated.

14                     Similarly working on guidance for the

15   one application versus two, we're in the process of

16   continuing to make the RFD process as efficient as

17   possible.        We just modified 21 CFR Part 3 for some

18   administrative changes to implement MDUFMA and

19   recognize the Office's role here, and we are

20   documenting the various precedents -- I think

21   Jonathan mentioned this earlier -- to make them much

22   more searchable and readily available.                 So we were

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                         353


1    able to have a much better assessment of what we've

2    done in the past with similar products and can help

3    ensure better consistency.

4                       In terms of review of combination

5    products, again, here what does the statute tell us?

6     Ensure the timely and effective premarket review by

7    overseeing the timeliness of and coordinating

8    reviews involving more than one center, and these

9    are the kinds of things that we're doing in that

10   regard.

11                      We do have an SOP on the inter-center

12   consultative collaborative review process, and on my

13   next slide I'll just give a few quick bullets about

14   that.

15                      We're also in the process now since

16   February 14th actually when we modified our SOP in

17   order to allow us to do this, is to monitor the

18   consultative process between the centers, and what

19   we're doing is when the centers initiate a consult

20   request to another center, they copy us on that

21   request.         We sort of have a low-tech way that we're

22   doing this right now, but we're in the process of

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       354


1    developing a data base, basically an automated way

2    of doing this so that we'll know in the background

3    every time one of these consults is going on.

4                     And what we do is we take a pretty

5    active approach when they come in to make sure that

6    the request is clear, that the second center, that

7    is, the center that is being asked to help

8    understands what's expected of them, what the time

9    frames are, and then we monitor that process to make

10   sure that the originating center actually got the

11   input that it was expecting and on time.

12                    We also have an effort underway where

13   we're reporting and tracking other combination

14   products, that is, combinations that don't require

15   consultation, but are combinations nevertheless, and

16   these include things like prefilled syringes and

17   transdermal patches which are typically not

18   consulted out to another center, but there's also a

19   lot more sophisticated combinations as well, where

20   the lead center has developed an expertise over the

21   years and doesn't require consultation.

22                    Well, what we have underway now is a

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       355


1    process where every major type of premarket

2    submission in all three centers as of May 1st is

3    categorized as to whether it concerns a combination

4    product or not and if so, what type, and we actually

5    have eight different types of combination products

6    that we're categorizing, and our first annual report

7    to Congress is due on October 26th, and we'll be

8    providing that data.

9                     This first year we won't have very much

10   to report just because of the time we've started.

11   There won't be a lot of combinations that are

12   actually approved by then, just given the statutory

13   time frames, but the data are being collected and we

14   will really have for the first time knowledge of how

15   many combination products we get each year, how many

16   we approve, what types they are, which centers are

17   doing them, and all of that.            And up until now we

18   really haven't had that kind of data.

19                    We're also available as a resource to

20   sponsors and review staff for combination products,

21   issues and questions.         We're in the process of

22   developing reviewer tools and training.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       356


1                     For example, on the consultation

2    process, by actively monitoring the consults, we're

3    seeing first-hand a lot of the issues that are

4    presented by consulting reviews, and we have some

5    lessons learned beyond what we thought we had

6    already addressed in the SOP, but real-life

7    practical issues that when somebody is actually

8    looking and seeing every one of these, some of the

9    common denominators of problems, and we're going to

10   be disseminating those to review folks.

11                    The SOP, in two words about the

12   consultation process, says that consults count.

13   Consulting reviews need to be given due priority,

14   and it's all part of the agency's work.               So if one

15   center asks another for help, the second center

16   needs to do its part in order to make sure that the

17   lead center is able to meet its review goals.

18                    And the consulted centers are expected

19   to be consulted with respect to the time lines for

20   the consulting review, and in turn, held accountable

21   for the timeliness and the quality of their

22   consulting review.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       357


1                     Very quickly on post-market regulation,

2    the law tells us to ensure the consistency and

3    appropriateness here.         What we've begun doing where

4    possible is in our RFD letters providing preliminary

5    determinations of what GNPs and adverse event

6    reporting requirements a combination product will be

7    subject to.

8                     We have two active working groups in

9    this area, one on GNPs, one on adverse event

10   reporting, to be able to provide some of the

11   guidance that Christine Allison from Lilly said was

12   badly needed in this area.

13                    Some general considerations.          Really my

14   point here is that one size doesn't fit all.                 I

15   think that point was made this morning.               There is no

16   cookie cutter approach.          These products, you know, I

17   think we heard this morning that they really run the

18   scope of a wide variety of different products in the

19   kinds of issues that they raise, and therefore,

20   there is not a one size fits all approach.

21                    We've had combinations approved under

22   HDE with as few as 30 or 40 patients, and then we've

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       358


1    had combinations reviewed with much larger

2    randomized controlled trials.

3                     So I think consultation with FDA is very

4    important.       I think some of the technologies that

5    were described this morning I personally hadn't

6    heard about before, and I think it's very important,

7    not that that means much, but I think it's important

8    that these dialogues, you know, the dialogue with

9    FDA begin as early as possible so that we can help

10   work with you on what regulatory pathway will be

11   followed.

12                    Just very briefly, collaboration between

13   the device and drug or biologic sponsors I think is

14   very important.       That was mentioned earlier, and I

15   have seen first hand that when the collaboration or

16   cooperation exists, the process does work much

17   better.

18                    It is very difficult, as Jonathan

19   pointed out, for a device company without any drug

20   company or biologic company partnership to be able

21   to independently work on changing the label of an

22   approved drug.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       359


1                     And this is going back to the question

2    in the beginning.       If my product is not a

3    combination product, will these initiatives still

4    help?

5                     And I hope the answer is yes.         It's our

6    intent that they will.          I think they present some of

7    the same issues even though a product may not

8    technically be a combination, and I'd encourage you

9    to contact our office.          We do have the liaison role

10   that I think that Jonathan mentioned earlier in

11   working with the centers, and I'm hopeful that we'll

12   be there to help make a difference and make the

13   development program easier for your product.

14                    So I think there will probably be a lot

15   of issues that come up in the Q&A section, and you

16   know, if you have issues you'd like me to address

17   afterwards, I'll be happy to stay after the

18   conference and address them at that time, too.

19                    Thanks.

20                    (Applause.)

21                    MR. JENKINS:     Thanks, Mark, for that

22   overview of the Combination Products Office.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       360


1                     I think you're doing a great job there,

2    and we're all very fortunate to have you there.

3                     We're going to enter into a series now

4    of some vignettes about the current approach to

5    review of some of the combination products, and

6    we'll start off with one that has been very hot in

7    the news recently, and it sounds like it has been

8    the topic of much discussion already here today, and

9    that's the drug eluting stents.

10                    We're fortunate to have Ashley Boam, who

11   is the Chief of the Interventional Cardiology Branch

12   in CDRH to give us that overview.

13                    Thank you.

14                    DR. BOAM:    Thanks, Dr. Jenkins.

15                    I thank Dr. Provost and Dr. Feigal for

16   setting up this workshop today, and I appreciate

17   being asked to speak on this very hot topic, drug

18   eluting stents.

19                    We've heard a lot about it today, and we

20   have a few more little items on this today.

21                    This is kind of a redundant slide at

22   this point.      I should have realized talking at four

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       361


1    o'clock you would have known the answer to this

2    question by now, but for those of you who maybe

3    stepped out this morning, this is an example of a

4    drug eluting stent.

5                     This is a diagram of the Cordis' CYPHER

6    Sirolimus-eluting stent which was approved just this

7    last April.      As you can see, the stent consists of a

8    bare metal stent platform with a polymeric carrier

9    in which the drug is loaded, and the drug elutes

10   from the polymeric carrier on the surface of the

11   stent.

12                    One of the things we found to be very

13   important when looking at applications for drug

14   eluting stents is that this really is a three

15   component system.       There is the stent platform and

16   delivery system which has traditionally fallen to

17   CDRH for review.       There is the polymeric carrier in

18   which the drug is loaded.           That has also kind of

19   fallen to CDRH review.          And then there is the drug

20   substance which has fallen under CDER review.

21                    So today I wanted to talk about some of

22   the review challenges for drug eluting stents kind

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       362


1    of as a real life example that we've been through in

2    the last couple of years.           Some of those topics

3    include regulatory jurisdiction, inspectional

4    authority and site readiness, disparity in statutory

5    and regulatory requirements between the two centers

6    involved, then appropriate leveraging of information

7    from other sources, appropriate preclinical and

8    clinical trial design issues, and then a little bit

9    on post market studies and surveillance.

10                    First, as Mark just ably described,

11   combination products fall under Part 3 of 21 CFR.

12   Request for jurisdiction was made for these products

13   pretty early on, and jurisdiction was granted to

14   CDRH as lead center with Center for Drugs'

15   consultation.

16                    But as you can see, there are quite a

17   number of divisions in both centers that are

18   involved in the review of these devices ranging from

19   the Division of Cardiovascular Devices and the

20   Division of Mechanics and Materials Science within

21   CDRH to the Cardiorenal Drug Products Group, the New

22   Drug Chemistry Group, and Pharmaceutical Evaluation

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       363


1    folks from CDER.

2                     Since CDRH does have the lead

3    jurisdiction for these products, the appropriate

4    marketing submission is a PMA, and the appropriate

5    application to investigate these devices is under

6    IDE.

7                     Just to give you a hint as to some of

8    the complexities of these devices, there are quite a

9    number of areas that require expertise from

10   mechanical performance and testing to drug substance

11   and polymeric carrier chemistry, to animal studies,

12   to PK/PD clinical trial design, and not the least of

13   which, manufacturing.

14                    I guess all of this is to really say

15   that it has really been a successful collaboration

16   between the two centers that has really led to the

17   success that we've had in the review of these

18   applications thus far.

19                    One question that has come up earlier

20   today, Christine mentioned this in her talk, was the

21   question of inspectional authority, and for drug

22   eluting stents, as I believe Mark mentioned, the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       364


1    inspections are conducted by CDRH's lead center, but

2    have involved participation from reviewers from

3    CDER's Office of New Drug Chemistry.

4                     It's important to note that for these

5    devices, as Mark mentioned, the drug regulations

6    have been applied to the drug substance, and then

7    the device QSR regulations to the finished product.

8                     It's also important for companies who

9    are making these devices to have their validations

10   complete prior to inspection.             We have worked very

11   interactively with the two centers and with

12   companies to try to get inspections done as quickly

13   as possible, but it's very important to have all of

14   those validations done.

15                    We understand these are very complex

16   products, but if there are subsequent changes and

17   subsequent validations, we may have to go out for a

18   second time, and it's the best use of all of our

19   resources if we go out once and get you taken care

20   of.

21                    There are a number of differences

22   between the different centers, CDRH, CDER and CBER,

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       365


1    when it comes to marketing applications and the

2    statutory authorities.          As you can see, for CDRH an

3    IDE is what's filed to start an investigation,

4    whereas in CDER and CBER you have an IND.

5                     And then for a marketing application for

6    these devices it would be a PMA as opposed to an NDA

7    in CDER or a BLA in CBER.

8                     There are also a number of differences

9    in development.       I think this has been pointed out

10   today as well.      The rate of technology change for

11   devices is much faster than that for drugs.                I

12   believe there was an example earlier that devices

13   can become obsolete within six to 12 months, whereas

14   a drug might be on the market for ten, 15, or even

15   20 years.

16                    There are a number of other differences

17   that are very important in our consideration of

18   these devices, and that relates to the influence of

19   physician technique on the results, on the number of

20   full scale studies that are usually required, and

21   how we regulate products in CDRH according to risk,

22   in which there are three classes versus one class

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          366


1    for new molecular entities in Center for Drugs.

2                     Companies that are investigating these

3    new products often want to know, well, what kind of

4    information do I need and where can I get it.                    How

5    do I not have to reinvent the wheel?

6                     And whether you have to reinvent the

7    wheel or not really depends on a couple of items

8    here, and it's really in this table.                  It depends on

9    whether your stent platform is approved or not

10   approved, and it depends on whether your drug

11   substance is approved or, as we say, unstudied.

12                    The easiest scenario in terms of being

13   able to leverage information from other applications

14   is the box marked one where both your stent platform

15   and your drug substance are approved.

16                    The most difficult situation is where

17   both the stent platform and the drug substance are

18   unapproved or unstudied, and that's the box marked

19   four.

20                    As you can see, there are guidance

21   documents for both centers that can help you to put

22   together the right information to get started with

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                       367


1    the study for these devices.

2                     So if I have an unstudied or unapproved

3    drug, what type of information do I need that I may

4    not be able to get from somewhere else?               For all of

5    these drug eluting stents, we're requiring that

6    sponsors provide the equivalent information that

7    would be required in a Phase I IND for CDER.

8                     It's also important to understand that

9    an analogue of an approved drug is considered to be

10   a new molecular entity by both CDER and CDRH for

11   these products.       So Phase I IND information for an

12   analogue to an approved drug would also be required.

13                    There are several categories of safety

14   information that fall under the Phase I IND

15   requirements.      That includes chemistry manufacturing

16   controls, otherwise known as CMC.              Both preclinical

17   pharm-tox and systemic clinical exposure in normals

18   are required prior to doing human investigations of

19   the finished product as a device.

20                    It's also important to note that if your

21   drug substances has not been studied before, this

22   Phase I IND safety information could very well

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       368


1    inform on the clinical trial design that would be

2    necessary for the finished product.               If there are

3    toxicity issues or potential drug-drug interactions,

4    that are identified during the Phase I safety

5    information gathering.          It may be necessary to alter

6    your clinical study to look for those when

7    evaluating the drug eluting stent.

8                     In terms of preclinical testing, I think

9    one of the most important messages we try to get out

10   is that characterization of the finished sterilized

11   product as it is to be studied is really essential.

12   We realize that a lot of changes and design

13   improvements go on during the design and

14   development, product development process, but by the

15   time you're ready to do a clinical trial we really

16   need you to characterize the actual device you want

17   to study so that we know when you get clinical trial

18   results what they really represent.

19                    And a couple of those areas include

20   characterization of the coating and the drug

21   substance, in vitro and in vivo elution

22   characteristics with release rate of the drug, and

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       369


1    also methods and some initial specifications for

2    stability of both the drug substance and the

3    polymeric carrier, if applicable.

4                     Also, adequate animal studies are really

5    needed to assess safety prior to going into human

6    clinical trials for these devices.

7                     A few more specifics here in the

8    preclinical area.       Some of the common deficiencies

9    that we see are inadequate stent platform testing in

10   terms of looking at fatigue and corrosion testing.

11   This is not testing that can be leveraged simply

12   from the bare stent platform.             With the addition of

13   a coating, it becomes important to look for fatigue

14   and the effects of corrosion through potential

15   cracks in the coating.

16                    We also see inadequate analysis of any

17   surface modifications made to the device, either

18   through application of the coating with the drug

19   substance in it, and so this relates to coating

20   integrity, durability testing, and also

21   characterization of both drug content and its

22   uniformity along the length of the stent.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       370


1                     We also see incomplete in vitro

2    pharmacokinetics both in terms of methodology, and

3    we strongly recommend that sponsors attempt to

4    develop an in vitro/in vivo correlation if at all

5    possible.

6                     This becomes very important in terms of

7    scale up from a clinical trial batch or

8    precommercialization to commercialization

9    manufacturing.       It also becomes very important if

10   there are changes or improvements that you want to

11   make in the device because the better you can

12   characterize the device you study, the better you

13   can evaluate what those changes might look like in

14   terms of clinical sequelae.

15                    And also CMC issues not being adequately

16   addressed, stability and shelf life are very

17   important.       I think device companies are very much

18   used to a device paradigm where accelerated aging

19   with real time aging to confirm those results have

20   been accepted.

21                    We typically review protocols, and when

22   we're very comfortable with device materials and

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       371


1    device packaging materials, they're pretty

2    straightforward protocols, but when you introduce

3    the drug substance and a polymeric substance, there

4    are a lot of new issues that need to be looked at in

5    terms of stability, and we recommend that sponsors

6    follow the ICH guidelines for evaluation of

7    stability especially of the drug substance.

8                     In terms of animal studies that we

9    receive, we often receive inadequate reports to

10   allow us to make an assessment of safety, to know

11   whether it's appropriate to start a clinical trial,

12   and that involves lack of an evaluation of the doses

13   that are intended for use in the clinical study, and

14   we also require doses higher than this.               We require

15   overdosage to make sure that we understand what the

16   toxicity limits are in an animal model.

17                    We look for serial sections of

18   myocardium, arterial histopath, and necropsy reports

19   for any deaths that might have occurred during the

20   study.

21                    As we move to the clinical evaluation of

22   drug eluting stents, first and foremost, we're

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       372


1    looking for a reasonable assurance of safety and

2    effectiveness, and it's important to remember that

3    your clinical trial design should look to meet both

4    of these objectives.

5                     The usual standard of evidence for these

6    products at this point is the randomized controlled

7    clinical trial, and in terms of study endpoints for

8    coronary drug eluting stents, we're looking for

9    primary endpoint or endpoints that include at least

10   a clinically meaningful endpoint.

11                    We're also evaluating the use of

12   surrogate or co-primary endpoints at this point.                    As

13   I believe Dr. Kuntz mentioned this morning, now that

14   we have an approved drug eluting stent on the

15   market, there are a lot of questions about how do

16   you design a study to be performed in the U.S. if I

17   don't feel I can do a placebo trial because the

18   penetrance has been so remarkable.

19                    And these are some of the areas that

20   we're looking at in working with sponsors as well as

21   our statistics group to come up with reasonable

22   clinical trial designs that will still give us

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           373


1    evidence of both effectiveness, but also, and most

2    importantly, of safety.

3                         We also recommend the use of independent

4    core labs, clinical event committees, and an on-

5    line, very active data safety monitoring board.

6                         I wanted to mention that TPLC is really

7    critical for drug eluting stents.                  The first drug

8    eluting stent is estimated to have been implanted in

9    over 50,000 patients since it was approved in late

10   April, which is a pretty remarkable roll-out for a

11   new product.

12                        And compare that 50,000 number to the

13   1,100 patients that we saw in the U.S. clinical

14   trial.           There's a lot of information you can get

15   from 1,100, but it's never going to tell you

16   everything about what happens when it gets to 50,000

17   or 100,000 patients.

18                        And so we feel that information gathered

19   in the post market is very important for these

20   products.           We are requiring five-year follow-up for

21   all of the cohorts that have been enrolled in

22   support of an application.

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       374


1                     We're also requiring additional data

2    collection in the post market period to get a

3    further understanding of what happens when real docs

4    put these in real patients because we know that

5    sometimes there are differences between what happens

6    in a clinical trial and what happens in the real

7    world.

8                     It's important to note though that in

9    the post market as folks are looking at new

10   indications and new patient populations for these

11   products, those indications and those patient

12   populations should be studied under the IDE process.

13                    There was a question about adverse

14   events earlier, and for these particular products,

15   in collaboration with our folks in the Office of

16   Surveillance and Biometrics and the people in the

17   Office of Drug Safety over in CDER, we have made a

18   determination that for coronary drug eluting stents

19   reports will come to CDRH through the MDR process,

20   but we have made arrangements for data to be shared

21   with CDER, both preapproval and post approval, as

22   information is gained on both sides about drug

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          375


1    substances.

2                     And so in closing, if you have

3    questions, certainly we encourage very early

4    meetings with us.       We're happy to meet with you.

5    We're happy to talk with you very early in your

6    process and then as needed again as you get further

7    through your product development.

8                     I'm the Branch Chief that handles the

9    coronary drug eluting stent program.                  Lisa Harvey is

10   handling the peripheral drug eluting stent program,

11   which I didn't really speak about today, but it has

12   its own set of challenges as you get great big

13   stents with lots more drug and an area where bare

14   stenting doesn't have approved products.                 So a lot

15   of their own challenges in that group.

16                    But I encourage you to contact the folks

17   on this slide if you have questions, and we look

18   forward to continuing our collaboration with CDER

19   and with the Office of Combination Products to make

20   efficient and effective review of these

21   applications.

22                    Thank you.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                          376


1                     (Applause.)

2                     MR. JENKINS:     Thanks, Ashley.

3                     We're going to move along quickly.

4    We're running a little long on time.                  I recognize

5    that, but I see a lot of attention in the audience.

6    So I think that's okay.

7                     Next we're going to talk about

8    photodynamic therapy systems.             We have Richard

9    Felten from the General Surgery Devices Branch of

10   CDRH.

11                    Richard.

12                    MR. FELTEN:     I'll just actually go to

13   the next slide very quickly.

14                    This I think, hopefully, is a success

15   story, but it also gives a good idea of how we got

16   where we are in this particular area.                  If you'll

17   notice from the slide, there is a pre-1984 date.

18   The drug that initiated photodynamic therapy and the

19   combination product review that we used to get this

20   finally to market was originally submitted as an IND

21   in 1978.

22                    We became involved from the standpoint

                                 NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                       377


1    of devices with this product in early 1980, '81, by

2    being asked by the Center for Drugs to look at the

3    light source that was being used to activate that

4    drug.

5                     We formalized that arrangement in 1985,

6    where I was actually designated as the lead reviewer

7    from the Center for Devices to look at these

8    products at the time.

9                     In 1989, we developed a collaborative

10   process for review with the Center for Drugs

11   following lots of conversations between Center for

12   Drugs, Center for Devices, and the company on how

13   best to proceed with these products, and as

14   formalized through the interagency agreement in

15   1981.

16                    The reason we did this is -- and you've

17   already seen this slide sort of -- Center for

18   Devices has a very involved process in getting

19   things to market.       You have premarket notification,

20   PMAs, the premarket notification that's in our

21   510(k), the premarket approval which is PMAs, and

22   some other sources, where essentially drugs has NDA.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          378


1     The generic drug approval process is not even close

2    to something we do.

3                      The problem, therefore, was how to take

4    these products and have them reviewed efficiently

5    and at the same time make sure that both the device

6    and the drug were being addressed in the appropriate

7    ways.

8                      The important thing to remember in this

9    area is all of these drugs are brand new drugs.

10   These are not already marketed products.                  These do

11   not have a history.         These are brand new drugs.

12                     In some cases they are derivatives of

13   biological products like human blood.                  Other times

14   they are simply chemistries that somebody has

15   developed on a lab bench that has a photosensitive

16   property.        So that alone is what led us to decide

17   that Center for Drugs would have lead in all of

18   these products with Center for Devices acting as the

19   consultant.

20                     The way this worked was we encouraged

21   the companies through conversations to have these

22   meetings with us.        Center for Devices took lead.                 We

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                          379


1    consulted the Center for Devices on the device

2    section.         All official correspondence during the

3    initial IND stage was sent through Center for Drugs

4    to the drug company.           The device companies were

5    involved peripherally through the drug company to

6    submit their device section as part of the IND.                      We

7    reviewed the device section, sent our comments to

8    Center for Drugs, who reported back to the drug

9    company, who talked to the device company.

10                      As Jonathan has mentioned, although

11   nobody believed it was going to work, it apparently

12   worked very well and very efficiently.                  We did have

13   an oral arrangement with Center for Drugs that

14   allowed me to talk directly to device companies if I

15   needed to, but official correspondence was always

16   through Center for Drugs.

17                      Once the clinical trials were completed,

18   the issue came up then how do we submit this

19   application and what do we do with it.                  One of the

20   interesting parts of this particular initial

21   application which was for a drug eventually called

22   Photofrin was that the drug company very clearly

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701     www.nealrgross.com
                                                                         380


1    told the agency, "We do not want to be a device

2    company.         We want to have nothing to do with the

3    devices once you approve this."

4                       And the reason for that was that in the

5    initial first one of these drugs, which was

6    Photofrin, the light source were commercially

7    available surgical lasers.             Surgical lasers, as

8    laser products, have regulatory responsibilities

9    under the Center for Devices' performance standard

10   for light emitting products.              The drug company

11   didn't want to be a device manufacturer because they

12   didn't want to be responsible for all of the

13   reporting requirements lasers have once you market

14   them and post market.

15                      In those conversations with the drug

16   company, who very clearly said, "We don't want to be

17   a device manufacturer," between Center for Drugs and

18   Center for Devices there was an agreement reached

19   that what would happen would be that a single

20   application would be submitted as an NDA.

21                      We, the Center for Devices would take

22   the device section out of the NDA and convert those

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        381


1    into PMAs.       We made three PMAs out of the device

2    section, two laser PMAs and a fiber optic PMA.

3                      Now, the reason we did that was, of

4    course, to make this work more efficiently for the

5    drug company after the approval process because once

6    they sold off the device sections, we needed a way

7    to be able to track those post marketly in case

8    there was design changes to the lasers, if there was

9    design changes to the fiber optics, if there are new

10   indications for use to come along with a different

11   drug.        We needed a way to be able to at least track

12   the devices.

13                     For the drug company, of course, they

14   wanted to not have anything to do with the devices

15   after the fact.        Historically actually what has

16   happened since then is we have had two new

17   indications for use added to the devices since the

18   original approval, and the fiber optic systems have

19   had three PMA supplements for design changes in the

20   fiber optics.

21                     This has made it much easier for them to

22   make these changes to the devices because they could

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       382


1    come directly to a device document that existed, and

2    it makes it easier for the Center for Devices to

3    track these changes, and this is why we've done it.

4                     That process has continued for

5    subsequent drugs.       Presently we have three approved

6    photodynamic therapy device-drug combinations.                  The

7    original one, which was by QLT, was for palliative

8    treatment of esophageal carcinoma.               The second one

9    is for a topical drug for treatment of actinic

10   keratosis, where we've done the same process where

11   the NDA was submitted.          We pulled out the device

12   section, created a PMA for the light system, and we

13   have continued to follow that device separately, and

14   we have some suggestions again that the company has

15   come in subsequently to the original approval and

16   made a device modification to that device under the

17   PMA as a PMA supplement, which again allowed that to

18   work very smoothly for us.

19                    And then the most recent approval, which

20   I cannot remember now, is like three or four years

21   ago.       It was again from QLT for the treatment of age

22   related macular degeneration, again using a laser

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       383


1    light source with a drug, which again we created the

2    PMA process.

3                     We've found that to work very well

4    because in most cases so far, lasers have been the

5    light source of preference because most of these

6    systems so far require the ability to transmit light

7    down fiber optics, and lasers give you that very

8    nice ability to do that.

9                     It has also allowed these companies to

10   sell off the laser part of their approvals so that

11   they don't have to be laser manufacturers.

12                    Whether or not that's the future to

13   where we will continue I have no way of knowing

14   because the laser is a very unique part of this

15   system and has unique responsibilities with the CDRH

16   requirements under the light performance standard,

17   but this history for the photodynamic therapy is how

18   we got to where we are today with what we're doing.

19   It is mainly because this was the first one of these

20   to come into the system in 1978 actually, and when

21   we first started this process, we had no previous

22   histories.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       384


1                     And it did require a lot of

2    collaboration between the Center for Drugs and the

3    Center for Devices, and I will repeat what everybody

4    else is telling you.         With these kinds of products

5    you have to get involved with the centers early on

6    in discussions to find out not only where you're

7    going to be placed as far as jurisdiction, but to

8    find out from the reviewing centers what are going

9    to be your responsibilities.

10                    And I thank you for your attention.

11                    (Applause.)

12                    MR. JENKINS:     Thank you, Richard.

13                    Our last speaker is Dan Shames, who is

14   the Director of the Division of Reproductive and

15   Urologic Drug Products in CDER, who is going to talk

16   to us about his experience with contraceptive

17   delivery systems.

18                    Dan.

19                    DR. SHAMES:     Thank you.

20                    I just noticed that the title of this

21   conference is "Innovative Systems for Drug

22   Delivery."       I'm the only person from CDER talking

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       385


1    about the drug portion of the review, and I'm the

2    last speaker.      So I guess this is what everybody is

3    waiting for right here.

4                     I want to thank the device folks for

5    asking us to give our perspective on the review of

6    drug device combinations.           I'm going to discuss the

7    particular experiences of our division in this area.

8     I think our group has a particular interest and

9    perhaps expertise since the gynecologists and

10   urologists in our division have significant clinical

11   experience with devices and drug-device

12   combinations.

13                    I was asked to evaluate our review

14   process regarding contraceptive implants, and

15   actually let me go back.          I was asked to review

16   contraceptive implants, but what I did was I

17   expanded this to contraceptive drug-device

18   combinations and other devices that we've had

19   experience with in our division.

20                    I'm going to take the experience that

21   we've had over the last five or six years and give

22   you what lessons we've learned regarding these

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         386


1    combination products.

2                     I'm first going to describe five

3    contraceptive systems that involve cooperative

4    review of device and drug components.                 The approval

5    year is in parentheses.          The first are a group of

6    device drug combinations that are all variations of

7    subdermal progestin releasing implants for

8    contraception.      They deliver the drugs systemically.

9     These are rods that are made up of a co-polymer

10   core surrounded by thin walled elastic tubing.

11                    I think that the lesson I got from this

12   group was my personal education regarding the

13   chemistry and the sort of engineering and

14   pharmacokinetic experience related to quantifying

15   manufacturing processes.          So I personally learned a

16   lot, and our division learned a lot by reviewing

17   these materials.

18                    Now, all of these and all of the

19   products I'm talking about went through the same

20   type of clinical trial experiences that oral

21   contraceptives would go through.              So the review

22   standard was the same for these products as they

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                       387


1    would be for any other kind of contraceptive device.

2                     Next is Lunelle, which is a monthly

3    injectable that delivers a combination of estrogen

4    and progestin for contraception.              The take-away

5    message here is that although this was a prefilled

6    syringe and should have been fairly straightforward,

7    there were issues related to manufacturing the

8    syringe and the vial, et cetera, which did make the

9    review challenging.

10                    The other thing that should be a lesson

11   here is that there were two drugs involved, and we

12   had to deal with the combination rule regarding

13   drugs in themselves.         So we not only dealt with the

14   device issue.      We dealt with the fact that we have

15   to show that each individual drug adds to the safety

16   and efficacy of the product.

17                    Mirena, which is the next one, is an

18   intrauterine system which delivers levonorgestoral

19   both locally and systemically.             I think the

20   important take-home message for this one is that

21   although there have been IUDs around for some time,

22   there was a challenge here to show that the addition

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         388


1    of the drug added to the effectiveness and safety of

2    the product itself, which is something that products

3    have to do, combination products have to do in

4    general.

5                     This next one is an intravaginal ring,

6    which delivers estrogen and progestin systemically.

7    This was relatively recently approved.                 It can be

8    inserted by the individual themselves.                 Vaginal

9    contraceptive rings have been studied for decades,

10   but it took some innovation on the part of the

11   developer to get it quite right regarding both

12   placing the estrogen and the progestin, and the

13   right combination of materials to make this work

14   properly.

15                    I think this is our last example.               This

16   is a transdermal system which delivers estrogen and

17   progestin, and although you might think at first

18   glance, well, this, you know, should be fairly easy,

19   we have a lot of transdermal systems.                 We've had

20   transdermal systems for menopausal symptoms for many

21   years, but as you may or may not realize, it

22   requires more drug delivery for contraception than

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701      www.nealrgross.com
                                                                       389


1    for menopausal symptoms, and it took some

2    significant innovation for the developer to create a

3    system which is only about the size of a match box

4    and deliver the appropriate amount of estrogen and

5    progestin.

6                     So I was asked to look at all of this

7    and look at our experience and ask myself what did

8    we learn from this.        Well, the good news is that as

9    far as CDER, we can work with device technical

10   experts in a productive and efficient manner to

11   review innovative drug delivery systems, and we have

12   done it, and we continue to do it, and most of the

13   time it goes fairly well.

14                    In this case, with contraception, it was

15   relatively easy because we used the same standards

16   for review, the same clinical trial standards, and

17   that was not terribly burdensome on the device

18   systems.

19                    And also, as many people have said, it

20   goes much better when we discuss these issues a

21   priori, before we start.

22                    The other news, the good news and then

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        390


1    the other news, when it doesn't go well.               It doesn't

2    go well, I find, when things are a little out of the

3    box, which, of course, is happening more and more.

4    I find that at least with our division the problem

5    is not necessarily the scientific challenges.                  It's

6    how to fit the scientific issues into the regulatory

7    constraints that we all seem to have, and I think

8    that's improving.

9                      Then, of course, we have the issue about

10   who's in charge, which we've talked about, CDER or

11   CDRH.        I've never seen a turf battle, but I guess

12   that could possibly happen.

13                     We also have what I call culture

14   clashes, and I didn't see Mark's slides.               He calls

15   it war of the worlds or colliding of the worlds.

16   Sometimes critics of CDER might characterize our

17   reviewers as what I have here, the pointy headed

18   bureaucrats whose main goal is to keep things off

19   the market because they have no regard for the

20   entrepreneurial spirit and good old American

21   ingenuity.        We're just obstructionists, et cetera,

22   et cetera, you know, versus the critics of CDRH who

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           391


1    think the reviewers are those people who will

2    approve anything that doesn't blow up when they plug

3    it in regardless of consequences to patient safety.

4                         Of course, none of that is true, but

5    there is a bit of a culture clash, and you know, I

6    try to think what is the origin of this culture

7    clash.           Well, maybe it's engineers and doctors, you

8    know.        There's a lot of doctors in CDER.            There's a

9    lot of engineers in CDRH, and I really don't think

10   that's the basic -- I think it's a matter of working

11   together more because the doctors in CDER seem to

12   get along well with the chemists, which we deal

13   with.        We've had a long history with, and I just

14   think we should be able to get along with the

15   engineers also.

16                        Perhaps another issue that may be more

17   important has to do with this reasonable versus

18   substantial evidence, which is in our regulations.

19   I'm not sure that's supposed to be an issue.                     I've

20   had sponsors and lawyers for sponsors tell me, well,

21   that's only in the eye of the reviewer, you know.

22   You determine what's reasonable and substantial.

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       392


1                     And then there's the issue of the big

2    PhRMA versus small firms, and it is true that we

3    deal with larger companies in general, but we do

4    deal with smaller companies, and I think that's

5    something maybe we all have to learn how to deal

6    with better.

7                     However, the future looks good in my

8    estimation.      I think Mark's group has actually

9    improved things.       We've had a very difficult issue

10   that had been essentially in regulatory and

11   scientific limbo for years, and Mark was able to get

12   us to move forward on this issue, find a way to move

13   forward, and I think the regulatory hang-ups can

14   often be the most difficult hang-ups.

15                    I think we are improving in terms of the

16   culture gaps or culture differences between the two

17   areas, and I think getting the message out that we

18   have to be talking about development very early on

19   with companies is important.

20                    However, I think most staff at CDER and

21   CDRH enjoy working on innovative products and are

22   well motivated to assist the sponsors and improve

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          393


1    our internal processes.

2                     Thank you.

3                     (Applause.)

4                     MR. JENKINS:     All right.          Thanks, Dan.

5                     I guess we're done with this section,

6    and we'll move on to the panel discussion.

7                     DR. JACOBSEN:      While she's collecting

8    the questions, let me just welcome everybody back to

9    this final session where we have both FDA and

10   industry on the panel up here.             We have tried to put

11   FDA and industry folks on both sides of the table so

12   that we're coming across with a message that this is

13   not an us against them.          This is an exciting new

14   area or an exciting old area really if you listened

15   to Richard Felten, and just listening to the talks

16   this morning, it was, I think extremely exciting to

17   see all of the things coming down the road.

18                    Clearly, we're going to have lots and

19   lots of issues to work through together, and I think

20   together ought to be the underlying take-home

21   message from today.

22                    I have one.     Do you have more written

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                       394


1    questions?

2                     I think the appropriate thing to do is

3    this is your part of the meeting, to let people ask

4    questions, and as I said before, you can either walk

5    up to the mic and ask it out loud or you can send

6    your question up on paper and we'll try to, you

7    know, get them answered.

8                     I mean, I can do a kick-off question

9    from the paper if that makes everybody more

10   comfortable.      I have one.

11                    Miriam, do you have something?

12                    Stuart, go ahead.

13                    DR. PORTNOY:     Hi.     My name is Stuart

14   Portnoy, and until a year ago I worked at CDRH for

15   eight years, most recently as a Branch Chief of

16   International Cardiology Devices.

17                    While I was at the FDA and in the past

18   year since I joined PharmaNet as a medical device

19   consultant, I have closely monitored the regulatory

20   and scientific requirements for drug-device

21   combination products, especially drug eluting

22   stents.

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       395


1                     And it has been my observation that

2    while CDRH has been designated as both the lead

3    review center and the regulatory authority for drug

4    eluting stents, the agency has clearly and

5    consistently raised the bar so that these products

6    are actually regulated more like drugs than devices.

7                     I'm concerned that if this trend were to

8    continue, that some new and potentially breakthrough

9    combination technologies may face significant delays

10   in making it to the marketplace and putting it to

11   clinical use because of unrealistic agency

12   expectations and requirements.

13                    And specifically, I've noticed a trend

14   of what I consider to be overly burdensome

15   requirements for things like kinetic drug release

16   testing, stability and lot release testing, and

17   other traditional drug testing requirements.

18                    Now, while I agree with the FDA that

19   such testing is absolutely necessary and fundamental

20   to demonstrating acceptable product performance, I

21   still believe that the agency is already going too

22   far in their requirements that the drug-device

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           396


1    combination products are held to the same standards

2    and level of quality control as pharmaceuticals.

3                         And let me emphasize that this is for

4    drug agents that have previously been demonstrated

5    to be safe in an NDA.

6                         So to address this concern, I hope and

7    recommend that the agency and specifically the

8    Office of Combination Products considers a

9    reasonable and feasible approach to regulating

10   combination products that lies somewhere between the

11   current requirements for traditional pharmaceuticals

12   and perhaps the less burdensome standards for

13   medical devices, and I invite panel discussion of

14   this important issue.

15                        DR. JACOBSEN:      Just let me add that we

16   are joined at the panel -- I should have done this

17   before           -- we have two center Directors joining the

18   panelists who have already spoken here today:                     David

19   Feigal, the CDRH Director, and Jesse Goodman, the

20   CBER Director.          Obviously they really need no

21   introduction, but I figured I'd better do it anyway.

22                        Anybody want to start off with that?

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         397


1                       DR. FEIGAL:     Well, let me make a

2    comment.         I don't think the standards are the same.

3    If the standards were, then some of the things you

4    have asked for would not get alternatives.                 If you

5    didn't understand the release kinetics and couldn't

6    do bridging between release kinetics of two

7    formulations, the only thing you could do if you

8    changed manufacturing would be to repeat the

9    clinical trials.

10                      And so I think there is a real vested

11   interest for us to make this as scientifically based

12   a process as possible because we have not taken that

13   stance, nor have we taken some of the other types of

14   traditional pharmaceutical approaches of requiring

15   the manufacture of three complete commercial batches

16   prior to NDA approval.            We haven't required only

17   testing, only doing clinicals on products

18   manufactured in the final formulation.

19                      So I think there still is a fair amount

20   of flexibility there.           The one sort of wish that we

21   often hear is to say we'd like to have a

22   breakthrough product and we'd like to not have to

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       398


1    submit any evidence for it.

2                     (Laughter.)

3                     DR. FEIGAL:     And to me it has always

4    struck me as sort of strange when someone says we

5    have enough evidence to show that it's safe, but

6    this is a breakthrough product and we haven't seen

7    the evidence of effectiveness yet.

8                     It would strike me that if you've got

9    enough evidence for safety so that you've begun to

10   see what the side effects are and so that you're

11   seeing side effects but you haven't seen any

12   benefits yet, you must not have a positive risk-

13   benefit ratio.

14                    So I think that all of the centers, when

15   you have a product that's dramatically different

16   than the existing therapies, treat those products

17   differently.      The evidence requirements are

18   generally less, and I think it would have been a

19   mistake to have simply stopped with the European

20   experience.      The limitations there weren't so much

21   the small numbers, but the very careful limitation

22   of the types of patients studied and have led the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       399


1    public to believe that that's the expectation of

2    what to expect from the product.

3                     There's a need to characterize how these

4    products basically work.          In other words, I think,

5    you      know, we can join the advertising promotion

6    staff of the companies, as many of our former

7    employees do.

8                     (Laughter.)

9                     MR. HUNTER:     Richard Hunter, and I'm not

10   a former employee.

11                    (Laughter.)

12                    MR. HUNTER:     I have been in the business

13   for 35 years in industry to get products on the

14   market, but just to ask my question, can you use

15   your imaginations here to determine how a company

16   like mine, Altea Therapeutics, that has a technology

17   for delivering drugs and other products through the

18   skin, can avoid double jeopardy, triple jeopardy,

19   whatever, every time we go to a different division,

20   a different center, which we will and have to some

21   degree already?

22                    In terms of some of the major questions,

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           400


1    I know that there will be tailored questions per

2    patient group and per drug, but the major questions,

3    the blockbusters that would put us back to square

4    one in that particular area.                Can you imagine a

5    better world, is what I'm asking.

6                         MR. KRAMER:     I think we can imagine that

7    world.           If I understand the question correctly, I

8    guess what you were indicating was that when you

9    have a new indication for an existing product, that

10   that indication is going to different divisions

11   within a center, and therefore, new questions are

12   generated each time.

13                        I think the intent would be not to have

14   to, you know, reinvent the wheel.                  We've heard that

15   before.          If that's a problem that you're

16   encountering, then you should definitely bring that

17   to the agency's attention so that we can look into

18   that.

19                        I mean, clearly indication-specific

20   questions will arise, but if fundamental questions,

21   as you say, have been answered, then I don't think

22   the intent would be to have to review those all over

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       401


1    again.

2                     DR. JACOBSEN:      Okay.

3                     DR. GOODMAN:     I'd just make one

4    addition, which is certainly in our world you can

5    refer to master files and data generated in other

6    settings to support an application involving

7    portions of the same product.             So to the extent that

8    the developers of these products are the same or are

9    cooperating, you know, we're very open to looking at

10   data broadly, but as Mark said, there are going to

11   be some kinds of data that are distinct for a new

12   combination.

13                    DR. VAN ANTWERP:       We've talked a lot

14   about today --

15                    DR. JACOBSEN:      Could you identify

16   yourself?

17                    DR. VAN ANTWERP:       Oh, I'm Bill Van

18   Antwerp from MiniMed, Medtronic MiniMed.

19                    We've talked a lot today about drug

20   delivery systems, but we all, or at least those of

21   us in the diabetes world, believe what David showed

22   us this morning, that devices delivering drugs or

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          402


1    biologics controlled by diagnostics are going to be

2    the future of diabetes therapy, at least, and

3    perhaps in a whole bunch             of hopefully other

4    therapeutic regimes.

5                        So how does the addition of a diagnostic

6    tool change what we've talked a little bit about

7    here today?         So I have a drug that's made by one

8    company, Christine's perhaps, and then we have a

9    device, a pump, and then we have now a diagnostic

10   device.          Does that change anything in your thinking,

11   your fundamental thinking about how these systems

12   get approved?

13                       And where I'm going is is the ultimate

14   endgame therapy, delivery?              Is there going to be the

15   possibility of approval for a therapy that includes

16   some of these kinds of systems?

17                       DR. FEIGAL:     I can give you an example

18   of one that's already on the market that has got a

19   diagnostic, and that's the variations of different

20   pacemakers that sense rhythms, sometimes deliver

21   intermittent shocking therapies, decide whether or

22   not to pace the heart, and it does introduce a whole

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       403


1    additional number of issues in terms of the way the

2    software is written.         There's an amazing number of

3    lines of code imbedded in the people's chest as they

4    walk around, and you need to make sure that the

5    software behaves properly in addition to the sensors

6    behaving properly, in addition to the whole logic.

7                     And then you have to prove that the

8    whole strategy has a net benefit, and that I think

9    has been one of the successful areas where the

10   devices are actually starting to look better than

11   the drugs that used to be used for arrhythmia.

12                    So I think that's possible.          I think

13   that some of the challenges laid out this morning

14   specifically for diabetes identifies, you know, that

15   not all side effects are created equally; that

16   hypoglycemia is potentially fatal and much more

17   devastating than loose control.             And so how do you

18   back into this and how do you do this in ways?                  And

19   it probably isn't even so much a matter of whether

20   it's an implantable, tiny device, which we'd

21   eventually like to see, or initially if it's

22   something that's done in a more controlled

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       404


1    environment.

2                     But I think that's where some of the

3    paradigms are.      I think what's interesting is you

4    watch and you see these things being developed

5    incrementally and you see changes, and this is

6    different than drug development.              You'll see a

7    change in pacemaker features from the same

8    manufacturer every six to nine months, and you'll

9    see new strategies that are unproven being planned

10   to be imbedded in the future models to treat

11   different types of things.           I imagine there will be

12   that sort of incremental benefits in developing

13   software for diabetes management.              You may not try

14   to do anything very complicated at first and deal

15   with the safer sort of things that you can treat and

16   then gradually work into the other things as you

17   develop the safety track record for that.

18                    What often you don't have is a sense yet

19   of sort of what will the clinical and the patient

20   population and the public bear in terms of

21   complications.      There are some products -- Jesse has

22   unfortunately a couple of them -- where it's

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       405


1    national news if there's a single product failure.

2    You know, one patient gets an HIV transfusion, and

3    it's news.

4                     And there's other products that are over

5    the counter drugs that we tolerate a certain serious

6    complication rate and even death rate, you know, a

7    death rate from, and so how the technology is

8    developed and the comfort that people have with the

9    technology so that we don't make our patients into

10   Luddites who think, "Oh, it must be the technology

11   that is going to be bad."

12                    How we build that trust as we build that

13   to say that the products are safe and effective is

14   very important, and it's a complex process.                It even

15   involves things like handling recalls responsibly

16   and safety alerts responsibly.

17                    There have been a lot of pacemaker

18   safety alerts, recalls over the years that haven't

19   undermined the confidence in the products because

20   they've been viewed largely as proactive measures to

21   deal with problems as they're discovered, as opposed

22   to manufacturing problems that weren't anticipated

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       406


1    and other kinds of problems.

2                     MR. KAHAN:     Can I add just one

3    regulatory point to that?           What you're talking about

4    in a regulatory sense is a closed-loop system where

5    the actual control of the release of the drug is by

6    a diagnostic feedback, and our discussions with FDA

7    over the years on closed loop systems is that they

8    certainly can be cleared through the agency.

9    However, the approval process will be one that will

10   be extremely rigorous because the potential for

11   underdosing or overdosing if somehow there's a gap

12   or a data glitch in the loop through a software or

13   other problem has raised the agency's hurdles here.

14                    And I think we've been talking about

15   these products for at least ten to 15 years, and now

16   they're about to come to be very, very quickly, and

17   so can we think out of the box?             I think the good

18   news is that you're going to be, especially with

19   insulin, you're going to be delivering a drug that

20   has a well-known character and a well known profile.

21                    On the other hand, the closed-loop side

22   of this is going to lead to possibly FDA

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       407


1    scrutinizing the product more than they would

2    scrutinize a pacemaker or an automatic implantable

3    defibrillator because you're relying totally on the

4    software and the feedback.

5                     DR. JACOBSEN:      Let's take this question,

6    and then I have a written one that I want to ask.

7                     MS. ITANI:     Temima Itani with Ethicon

8    Endo-Surgery.

9                     I was struck this morning by the

10   complexity of the programs that were presented, and

11   I believe that they will undoubtedly present a big

12   challenge to the regulatory system.               I'm interested

13   in hearing from the various center Directors here

14   what are their thoughts on where FDA needs to go to

15   meet these challenges.

16                    What are the changes that need to be

17   made, the competencies, et cetera?

18                    DR. FEIGAL:     Jon's taking the easy way

19   out.       We'll make you Deputy Center Director for the

20   hour.

21                    (Laughter.)

22                    DR. FEIGAL:     So that you can answer the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       408


1    question, Jon.

2                     But I think the hardest thing for CDRH,

3    one of the things, we were talking about the culture

4    differences.      It isn't just the fact there's

5    different application processes and things.                The

6    thing that is different and was alluded to a little

7    bit in Ashley's slides is our responsibility to make

8    risk-based determinations in an application.

9                     And so even within an application not

10   every question has to be settled with clinical data.

11    So one of the hardest things is to decide which

12   kind of things are actually better determined with

13   performance specifications, engineering

14   specifications.

15                    And sometimes it's thought of as a

16   lesser standard, but you know, I would argue there

17   are some things like radiation therapy equipment

18   where you'd rather have a physicist measure the beam

19   than try and figure out how sharp the beam is by

20   testing it on patients.          You're better off with

21   performance standards in that kind of setting once

22   you've established that a beam has some therapeutic

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       409


1    uses.

2                     So I think a lot of the strategies that

3    were presented this morning, which included drugs

4    which were activated by the use of energy, by the

5    use of light, that included many new sort of novel

6    fabrication technologies to make needles that were

7    smaller than were possible before.

8                     A lot of that, I think, comes down to

9    really identifying what are the different

10   characteristics of those products that are really

11   going to be essential to their performance and that

12   will make them safe and effective, and to figure out

13   which of the things, even though they're new, are

14   probably better determined by looking closely at the

15   engineering than at the clinical data.

16                    So I think that's probably going to be

17   one of the challenges, is making that sort of risk

18   based assessment.       I think the fortunate thing for

19   devices is that they are built incrementally and

20   iteratively, change by change, and that gives us the

21   ability to creep up on some of those technologies,

22   but some of them seem awfully slow in the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       410


1    development.

2                     Was that in the 1840s that those laser -

3    - that those light activated drugs started?                That

4    was a long time ago.

5                     John, do you     have any comments on new

6    technology and how CDER can learn about it?

7                     MR. JENKINS:     Now that I've been

8    promoted to head pointy head bureaucrat, I guess.

9                     (Laughter.)

10                    MR. JENKINS:     I took offense to that

11   remark.

12                    I think the biggest challenge that we

13   face in CDER is becoming more familiar and aware of

14   the CDRH regulations and statutory provisions.                  Most

15   of our reviewers really have very little knowledge

16   about the CDRH process.          So when they get asked to

17   do a consult or a collaborative review for a drug

18   device combination may be where CDRH is the lead

19   center, it's really a whole new world for them.

20                    I've watched the collaboration that's

21   been going on for the last six or 12 months between

22   Ashley Boam's group in CDRH and the Cardiorenal

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         411


1    Drugs Division in CDER, and I think they've

2    developed a really good working relationship, a good

3    understanding of the procedures, the regulatory

4    hurdles, and the pathways, and I think that's worked

5    very well.

6                       So at some point you develop a critical

7    mass of relationships and understanding that make it

8    go well.         All too often most of our divisions see

9    one of these, you know, every year or once every two

10   or three years.         So you don't really develop that

11   critical mass of knowledge.

12                      One of the other things that struck me

13   as I was thinking of answering this question is it

14   may not be apparent to most of the people in the

15   audience, but most of the people at CDER don't even

16   know people at CDRH.           We're not in the same physical

17   location.        We rarely run into each other in the

18   cafeteria or whatever.            In fact most of us don't

19   even know where CDRH is located.

20                      (Laughter.)

21                      MR. JENKINS:     So it would be, I think,

22   really nice if, down the road, the White Oak campus

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          412


1    does actually bring us all together on the same

2    campus where there can be shared training

3    opportunities, where you can kind of walk across the

4    courtyard and go to a device meeting rather than now

5    trying to figure out how to get your way up 270 to

6    go to a device meeting.

7                        So I think training, opportunity to

8    interact and experience go a long way to making

9    these collaborations work well.

10                       DR. GOODMAN:     Well, you know, I think

11   CBER has some unique perspectives on this that I

12   think are relevant to this in terms of constantly

13   dealing with a lot of new technologies and cutting

14   edge technologies where risk is often uncertain, and

15   where as David said, a risk based approach and an

16   iterative approach is important.

17                       I think these are big challenges for the

18   agency.          I think everything the agency does is a big

19   challenge for it, but I think new technologies are

20   particularly big challenges, and then new

21   technologies that cross regulatory lines are even

22   more difficult ones.

                                   NEAL R. GROSS
                           COURT REPORTERS AND TRANSCRIBERS
                               1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433            WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       413


1                     To me some of the things we need to

2    strive for in FDA and you outside need to help us

3    with are our expertise, you know, and when you're

4    dealing with new technology, with new material

5    science, with new biologics and cells or drugs, you

6    really need people who are cutting edge and have

7    stayed current.

8                     So we need to invest in our own people

9    in terms of being scientifically up to date, and I

10   include there not just the technology, but in being

11   in touch as much as possible with clinical reality,

12   clinical trials, et cetera.

13                    And I think most people at FDA would

14   like to see that, but when people are working very

15   hard and don't have a lot of time, that's one of the

16   things that tends to suffer.            It also suffers from

17   the resource point of view, but I know all of the

18   people sitting up here from the agency are very

19   conscious of trying to support our people to be as

20   expert as possible.

21                    Anther part of that, I think, is

22   collaboration and consultation both within the

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        414


1    agency and then outside, and how can we find

2    nonconflicted ways, for instance, to collaborate

3    more and get more outside the agency, and to me for

4    CBER that's a real priority.

5                      And finally, as I think both previous

6    people said, I think, you know, this is sort of

7    "Brave New World" technology that many of you have

8    talked about earlier today, and it really has to be,

9    as David said, in devices you see this all the time,

10   but in the other areas we don't see it as much; that

11   there needs to be this iterative approach to how we

12   evaluate products and react to new information and a

13   degree of flexibility that one needs to strive for.

14                     But I think all of those things to do

15   them, you know, have required expertise and good

16   communication, all very resource-intensive stuff,

17   but I think it's stuff ideally we want to work with

18   you to do.

19                     DR. JACOBSEN:      I'm not a center

20   Director.        In fact, I don't even work for FDA

21   anymore, but can I make a comment on this question

22   anyway?

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                        415


1                      DR. GOODMAN:     Sure.

2                      DR. JACOBSEN:      Even though it was asked

3    for the center Directors.

4                      Mark said earlier that he had really

5    liked the talks this morning and this early

6    afternoon, that he hadn't heard of a lot of the

7    technologies, and that he thought that his take-away

8    message as a result of seeing all of those exciting

9    technologies was that companies need to dialogue

10   with FDA.        I think he said begin early, and I agree

11   with that.

12                     But I also would add that it seems to me

13   that this kind of open meeting really helps that

14   dialogue start to happen and maybe we should do more

15   meetings with industry and FDA staff like this one

16   where you really get a chance to hear the talks on

17   new technologies like we heard this morning, maybe

18   even have the products, you know, area specific.

19                     I don't know, but sort of talking

20   together about the technologies that are leading to

21   these new and interesting combination products.                    I

22   mean, the platform presentations were really

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         416


1    terrific, but the hallway conversations were just as

2    terrific.

3                       So that would be my suggestion, but I

4    don't know how you all feel about that.

5                       DR. GOODMAN:     Yeah, we think it's great,

6    and you know, the other thing some people have done

7    is just come in and talk to us about their future

8    plans and portfolios, and it's a little bit of, you

9    know, meet and greet kind of thing.

10                      On the other hand, we find it very

11   informative to be aware of not just what's there,

12   but what's coming to be sure we have the right kind

13   of expertise.

14                      DR. JACOBSEN:      I have a couple of other

15   written.         I don't see anybody else at the mic.

16                      The question is insulin is currently not

17   FDA approved for IV route of administration.                   IV is

18   an off label use.         The insulin manufacturers don't

19   seem interested in filing with FDA to do the studies

20   for IV insulin to be approved, yet it's widely used.

21                      If IV insulin was approved, then that

22   would open the door for novel IV insulin devices to

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                           417


1    be developed for hospital patients.                   Can IV insulin

2    be cleared without much initiative from insulin

3    manufacturers?

4                         IV insulin devices are not approvable

5    now with IV insulin being used off label.

6                         MR. JENKINS:     Sounds like a drug

7    question.

8                         (Laughter.)

9                         DR. GOODMAN:     We would be very open to

10   having sponsors of the insulins come forward to

11   develop, you know, approved indications for use of

12   insulin IV.          I think we already have the dosage

13   forms.           I think the forms that are available may be

14   appropriate, although I'm not sure of that.                    There

15   may be some modifications that need to be made in

16   the preservatives or whatever.

17                        Sometimes the agency finds itself in the

18   situation where sponsors don't come forward, and

19   sometimes we find that we have to develop the data

20   ourselves.          It may be possible that there's adequate

21   data in published literature that someone could put

22   together and come forward and submit a supplemental

                                    NEAL R. GROSS
                            COURT REPORTERS AND TRANSCRIBERS
                                1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433             WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       418


1    application to get that approved.

2                     Sometimes it comes down to we have to do

3    it ourselves, which is obviously a very resource-

4    intensive process to go through reviewing the

5    literature and developing an understanding of

6    whether the product is felt to be safe and

7    effective, and then we can put out calls for

8    applications.

9                     So that's a question we could take back

10   to our Metabolic and Endocrine Division, but I think

11   we also have a representative from one of the major

12   insulin manufacturers on the panel.               So she might

13   want to address coming to us for an indication.

14                    MS. ALLISON:     I think I'm probably not

15   the proper person to answer that question, but it

16   would still be welcome if anybody wants to discuss

17   about this approach to our company, and we can talk

18   about that.

19                    DR. KLONOFF:     David Klonoff from Mills

20   Peninsula.

21                    That was actually my question, and I

22   just wanted to have a follow-up to that, which is:

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       419


1    do you think that if a device company came forward

2    because they have a method of delivering insulin by

3    an alternate route, namely, intravenously for

4    hospital patients, that this would be sufficient for

5    you to look into the IV insulin indication or would

6    you still say that this device company must bring on

7    board an insulin manufacturer?

8                     MR. JENKINS:     Well, I think there are

9    different ways that you can approach it.              Clearly

10   the most straightforward way is as the question was

11   written, is if the insulin manufacturers would get

12   approval for an IV indication that would help,

13   obviously, the device manufacturers.

14                    The other approach would be for you to

15   come in in partnership with an insulin manufacturer

16   or maybe not even in partnership; just, you know,

17   some of the pumps are not in partnership with the

18   insulin manufacturers.

19                    You yourself could be the one who could

20   summarize the literature and try to present the

21   evidence to support approval that, you know, IV

22   insulin for whatever indication you're seeking is

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                         420


1    safe and effective, and there maybe adequate data in

2    the literature to help support much, if not all of

3    that indication.

4                       So I would encourage you to, you know,

5    consider talking to the Metabolic and Endocrine

6    Division about what they might need to feel

7    comfortable for that indication.

8                       DR. KLONOFF:     Okay.     Thank you.

9                       DR. FEIGAL:     There is one historical

10   example.         The very first H. pylori approvals were

11   done based on literature reviewed by an FDA

12   reviewer.        It didn't occur to me at the time, but

13   that might have had some user fee implications --

14                      (Laughter.)

15                      DR. FEIGAL:     -- because if you were to

16   come in with an efficacy supplement for insulin,

17   wouldn't he need a drug user fee for that?

18                      MR. JENKINS:     Probably if you're

19   submitting the simple clinical literature to try to

20   support an indication.            That would probably meet the

21   definition of clinical data for review, but there

22   obviously are also provisions for waivers of fees in

                                  NEAL R. GROSS
                          COURT REPORTERS AND TRANSCRIBERS
                              1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433           WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       421


1    some cases.

2                     We have taken the approach occasionally

3    of, you know, developing the data ourselves.                 We did

4    that with levothyroxine.          We published a Federal

5    Register notice saying that we, you know, based on

6    the accumulated scientific evidence found

7    levothyroxine to be safe and effective, and what we

8    needed were manufacturers to submit NDAs to show

9    that they could manufacture a quality product that

10   was stable over time.

11                    We did that recently with Prussian Blue

12   for the indication for elimination of radiation from

13   the body after accidental exposure.               So we published

14   a Federal Register notice saying that we had

15   reviewed the scientific literature and concluded

16   that Prussian Blue was safe and effective for that

17   use, and now we're looking for manufacturers to come

18   in and basically do the manufacturing package, the

19   CMC package.

20                    DR. JACOBSEN:      Okay.     I think we have

21   time for one more question.

22                    (Participant speaking from an unmiked

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                          422


1    location.)

2                     DR. GOODMAN:     Well, you know, I think

3    the question, because that mic doesn't seem to be

4    working, was about vaccine delivery devices, and

5    we've actually talked recently about potentially

6    having a public workshop about this.                  I think it's a

7    very rich area.

8                     I think there are several different

9    technologies out there that are quite exciting that

10   offer promise of more rapid or less complicated

11   vaccine delivery.

12                    I think with vaccines the general point

13   of view has been that each vaccine is a new product,

14   but I think just like a syringe is a vaccine

15   delivery device, some of these formats readily lend

16   themselves to multiple vaccines.

17                    So we do want to both hear more broadly

18   about some of the technologies that are out there

19   being developed as was suggested and then discussion

20   some of the regulatory implications.

21                    But as I said, it is very exciting.

22   When you think of, for instance, we've had

                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701       www.nealrgross.com
                                                                        423


1    discussion about this, you know, there are issues in

2    the Third World about reduction of needle

3    transmission of infections, and potentially some of

4    these devices if they were not too costly could have

5    tremendous promise in alleviating global health

6    problems.

7                      There are some suggestions that some of

8    these devices may be able to deliver equivalent

9    immunogenicity at lower antigen levels.                That's a

10   hope.        So I think it's a very exciting area, and as

11   I said, we may be able within the next year or so to

12   be thinking about a workshop just on that subject.

13                     DR. JACOBSEN:      Well, it's five o'clock,

14   and the agenda promised that you would be out by

15   five.

16                     I'd like to thank all of the panelists,

17   and also I'm sure that if you have individual

18   questions, they probably would be willing to hang

19   around for a few minutes if you want to grab them

20   before they can get out the door.

21                     I don't know if there are any other

22   wrap-up comments.        Are there any other wrap-up

                                 NEAL R. GROSS
                         COURT REPORTERS AND TRANSCRIBERS
                             1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433          WASHINGTON, D.C. 20005-3701    www.nealrgross.com
                                                                       424


1    comments?

2                     (No response.)

3                     DR. JACOBSEN:      Okay, and I'd like to say

4    again thanks to Mariam and to Vickie for putting on

5    such a good workshop in such a short time.

6                     (Applause.)

7                     (Whereupon, at 5:05 p.m., the         meeting

8    in the above-entitled matter was concluded.)

9

10

11

12

13

14

15

16

17

18

19

20

21




                                NEAL R. GROSS
                        COURT REPORTERS AND TRANSCRIBERS
                            1323 RHODE ISLAND AVE., N.W.
     (202) 234-4433         WASHINGTON, D.C. 20005-3701    www.nealrgross.com