even if sensitivity to drug effects does exist for a therapeutic class assay sensitivity in a particular study

Document Sample
even if sensitivity to drug effects does exist for a therapeutic class assay sensitivity in a particular study Powered By Docstoc


                        + + + + +




                        + + + + +


                        + + + + +


                        + + + + +


                    February 19, 2002

                        + + + + +

               The Advisory Committee was called to order

at 8:00 a.m., in the Conference Room of the Holiday Inn,

Two Montgomery Village Avenue, Gaithersburg, Maryland,

by Dr. L. Barth Reller, Chairman, presiding.


DR. L. BARTH RELLER                  Chairman

DR.   VINCENT T. ANDRIOLE           IDSA Representative
DR.   GORDON L. ARCHER              Member
DR.   DAVID M. BELL                 Consultant
DR.   JOHN E. BENNETT               Consultant
DR.   P. JOAN CHESNEY               Consultant
DR.   CHRISTY CHUANG-STEIN          PhRMA Representative
DR.   ALAN S. CROSS                Member
DR.   STEVEN EBERT                  Consumer Representative
DR.   JOHN E. EDWARDS, JR.         IDSA Representative
DR.   ROBERT J. FINK                Consultant
DR.   THOMAS R. FLEMING            Consultant
DR.   MARY GLODE                    Consultant
                       S A G CORP.
202/797-2525           Washington, D.C.         Fax: 202/797-2525


DR.   RICHARD GORMAN               Consultant
DR.   CATHERINE HARDALO            PhRMA Representative
DR.   JAMES E. LEGGETT, JR.        Member
DR.   CELIA MAXWELL                 Consultant
DR.   JOSHUA P. METLAY             Guest
DR.   ROBERT M. NELSON             Consultant
DR.   JUDITH R. O'FALLON           Member
DR.   JAN E. PATTERSON             Consultant
DR.   JULIO A. RAMIREZ             Member
DR.   COLEMAN ROTSTEIN              Guest
DR.   DAVID SHLAES                  PhRMA Representative
DR.   CIRO SUMAYA                  Consultant
DR.   GEORGE H. TALBOT             IDSA Representative
DR.   FRANCIS TALLY                Industry Representative
DR.   DENNIS D. WALLACE            IDSA Representative
DR.   JANET WITTES                 Consultant
DR.   LIANNG YUH                   PhRMA Representative
DR.   TARA P. TURNER               Executive Secretary

                      S A G CORP.
202/797-2525           Washington, D.C.       Fax: 202/797-2525


Call to Order and Introductions ................... 4

Conflict of Interest Statement .................... 7

Opening Comments by Dr. Goldberger ................ 12

Presentation by Dr. Albrecht ...................... 15

Presentation by Dr. Temple ........................ 24

Presentation by Dr. Lin ........................... 43

Presentation by Dr. Brittain ...................... 50

Presentation by Dr. McCracken ..................... 70

Industry Presentation by Dr. Shlaes ............... 85

Industry Presentation by Dr. Tally ................ 99

IDSA Presentation, Dr. Andriole & Dr. Edwards .... 137

Presentation by Dr. Fleming ...................... 150

Open Public Hearing ............................. 175

Presentation by Dr. John Powers .................. 180

Presentation by Dr. Susan Thompson ............... 209

Summary and Charge to Committee by .............. 236
      Dr. Goldberger

Committee Discussion ............................. 244

Adjournment ...................................... 373

                     S A G CORP.
202/797-2525         Washington, D.C.     Fax: 202/797-2525

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

2                                                       (8:11 a.m.)

3                   CHAIRMAN RELLER:        Good morning.   I'm Barth

4    Reller, in the Division of Infectious Disease, Professor

5    of Medicine and Pathology at Duke University Medical

6    Center, and Director of Clinical Microbiology.             I would

7    like to welcome you to this morning's and this afternoon's

8    Anti-Infective Advisory Committee of the U.S. FDA.

9                   We will begin this morning's meeting with

10   a conflict of interest statement read by our Executive

11   Secretary, Tara P. Turner.        Before that, however, I would

12   like to introduce or have the other panel members

13   introduce themselves.

14                  We will start at the right and continue

15   around, but in addition to that, there are three members

16   of the Pediatric Subcommittee for Anti-Infective Agents,

17   and after Dr. Glode, if those three members who are not

18   sitted at the table would please come up to a microphone

19   and introduce themselves.

20   We will start with Dr. Goldberger.

21                  DR. GOLDBERGER:       I am Mark Goldberger, from

22   the Office of Drug Evaluation IV, FDA.

23                  DR.    ALBRECHT:       Renata   Albrecht,    Acting

24   Director, Division of Special Pathogen and Immunologic

25   Drug Products, FDA.

                              S A G CORP.
     202/797-2525              Washington, D.C.        Fax: 202/797-2525

1                   DR. SORETH:        Good morning.          I am Janice

2    Soreth, the Division Director for Anti-Infectives at FDA.


4                   DR. LEGGETT:        Good morning.         Jim Leggett,

5    Infectious Diseases, in Portland, Oregon.

6                   DR. SUMAYA:       Ciro Sumaya, Dean, School of

7    World Public Health, Texas A&M University System Health

8    Science Center.

9                   DR. GLODE:     Mimi Glode, Pediatric Infectious

10   Disease, University of Colorado Medical Center.

11                  DR.   O'FALLON:         Judith    O'Fallon,     Cancer

12   Center Statistics, Mayo Clinic, Rochester, Minnesota.

13                  DR.     ARCHER:     Gordon      Archer,     Infectious

14   Diseases,      Adult      Infectious          Diseases,      Virginia

15   Commonwealth University, in Richmond, Virginia.

16                  DR. RAMIREZ:         Julio Ramirez, Division of

17   Infectious Diseases, University of Louisville, Kentucky.

18                  DR.     TURNER:         Tara     Turner,     Executive

19   Secretary for the Committee.

20                  CHAIRMAN RELLER:        And could we have the other

21   three members of the Pediatric Subcommittee come up to

22   a microphone and introduce themselves, please.

23                  DR. FINK:     Bob Fink, Pediatric Pulmonology,

24   Children's Hospital, in Washington, D.C.

25                  DR.   NELSON:         Robert     Nelson,     Pediatric

                               S A G CORP.
     202/797-2525               Washington, D.C.          Fax: 202/797-2525

1    Critical Care, Children's Hospital, Philadelphia.

2                     CHAIRMAN RELLER:           Thank you very much, and

3    we   look      forward    to     your    participation        in   today's

4    discussions.

5                     DR.     EBERT:          Steven     Ebert,     Infectious

6    Diseases Pharmacist, Meriter Hospital, and Clinical

7    Professor, University of Wisconsin, Madison.

8                     DR. BELL:          David Bell, Assistant to the

9    Director for Antimicrobial Resistance, National Center

10   for Infectious Diseases, at CDC in Atlanta.

11                    DR.     CROSS:         Alan      Cross,     Division     of

12   Infectious Diseases, University of Maryland at Baltimore.

13                    DR. PATTERSON:           Jan Patterson, Infectious

14   Diseases University of Texas Health Science Center, San

15   Antonio.

16                    DR.     CHESNEY:          Joan    Chesney,     Pediatric

17   Infectious Disease, at the University of Tennessee,

18   Health Science Center, in Memphis.

19                    DR. BENNETT:         Jack Bennett, NIH, Bethesda,

20   Maryland.

21                    DR. FLEMING:         Thomas Fleming, Department of

22   Biostatistics, University of Washington.

23                    DR. WITTES:          Janet Wittes, Statistician,

24   Statistics Collaborative, D.C.

25                    CHAIRMAN RELLER:           Thank you.       Dr. Turner.

                                    S A G CORP.
     202/797-2525                   Washington, D.C.            Fax: 202/797-2525

1                     DR. TURNER:         Thank you.          The Food and Drug

2    Administration has prepared general matters waivers for

3    the    following       special       Government          employees:     Julio

4    Ramirez, Steven Ebert, John Bennett, Jan Patterson, Celia

5    Maxwell, Ciro Sumaya, L. Barth Reller, Alan Cross, Gordon

6    Archer,        James    Leggett,       Jr.,       Joan    Chesney,      Celia

7    Christie-Samuels, Janet Wittes, Robert Fink, Richard

8    Gorman, Thomas Fleming, Robert Nelson, and Kathryn

9    Edwards, who are attending today's Anti-Infective Drugs

10   Advisory Committee Meeting on the proposed approach for

11   selection       of     delta    in    non-inferiority          equivalence

12   clinical trials.

13                    And the impact of this approach on studies

14   of anti-infective drug products, with a focus on acute

15   exacerbation            of         chronic           bronchitis            and

16   hospital-acquired-pneumonia being held by the Center for

17   Drug Evaluation and Research.

18                    A copy of the waiver statements may be

19   obtained by submitting a written request to the Agency's

20   Freedom of Information Office, Room 12A-30 of the Parklawn

21   Building.

22                    Unlike issues before a committee in which

23   a particular product is discussed, issues of broader

24   applicability, such as the topic of today's meeting,

25   involve        many     industrial         sponsors         and     academic

                                    S A G CORP.
     202/797-2525                   Washington, D.C.              Fax: 202/797-2525

1    institutions.

2                   The committee members have been screened for

3    their financial interests as they may apply to the general

4    topic at hand.       However, because general topics impact

5    on so many institutions, it is not prudent to recite all

6    potential conflicts as they apply to each member.

7                   The FDA acknowledges that there may be

8    potential conflicts of interest, but because of the

9    general nature of the discussion before the committee,

10   these potential conflicts re mitigated.

11                  With respect to FDA's invited guests, there

12   are reported interests which we believe should be made

13   public to allow the participants to objectively evaluate

14   their comments.

15                  Dr.    George       McCracken,        Junior.,      is        a

16   researcher     with    Bristol       Myers      Squibb     and     Abbott

17   Laboratories.          In      addition,        he    lectures           for

18   GlasxoSmithKline and serves as a scientific advisor for

19   GlasxoSmithKline, Abbott, Bristo Myers Squibb, Aventis

20   Pharmaceuticals, Bayer, and Johnson & Johnson.

21                  Dr.    Joshua      Metlay       lectures     and     is       a

22   scientific advisor for Aventis.

23                  Dr. Coleman Rotstein serves as a researcher

24   and has contracts and grant from Pfizer, Merck, ICOS,

25   Schering, Wyeth, and Fujisawa.            In addition, Dr. Rotstein

                              S A G CORP.
     202/797-2525                Washington, D.C.             Fax: 202/797-2525

1    consults for Merck, Schering Pfizer, and Pharmacia.         He

2    also lectures for Pharmacia, Pfizer, Bayer, Merck, and

3    Fujisawa.

4                   In addition, we would like to note for the

5    record that Drs. Catherine Hardalo, David Shlaes, Lianng

6    Yuh, and Christy Chuang-Stein from PhRMA, Dr. Francis

7    Tally from Cubist Pharmaceuticals, and Drs. Vincent

8    Andriole, George Talbot, Dennis Wallace, Louis Rice, and

9    John Edwards, Jr., from IDSA, are participating in this

10   meeting as industry representatives, acting on behalf

11   of regulated industry.

12                  As such, these participants have not been

13   screened for any conflicts of interest.     And I have two

14   announcements.    I just want to remind the participants

15   that when you want to speak into the microphone, please

16   pull the microphone towards you, and press the button

17   until the light turns on red.

18   And to be sure to turn it off when you finish speaking.

19                  Also, if you wish to enter a statement for

20   the record, comments on this meeting topic may be

21   submitted to Docket Number 98D-0548, Development of

22   Antimicrobial Drug Products, and there is a handout that

23   has been distributed at the front table.      Thank you.

24                  DR. ANDRIOLE:    Barth, I have a comment to

25   make about Ms. Turner's introduction of the four of us.

                            S A G CORP.
     202/797-2525           Washington, D.C.      Fax: 202/797-2525

1     We are here to represent the Infectious Diseases Society

2    of America and not any industry.

3                   CHAIRMAN RELLER:       Thank you, Dr. Andriole,

4    and actually this is a great segue to asking you and others

5    from IDSA at the invited guest table to introduce

6    themselves.    Could you start?

7                   DR. EDWARDS:     I am Jack Edwards, from Harbor

8    UCLA Infectious Diseases.

9                   DR. WALLACE:     I am Dennis Wallace, and I am

10   from Rho, Incorporated, in Chapel Hill.

11                  DR. TALBOT:    George Talbot, Talbot Advisors.


13                  DR.    ANDRIOLE:         Vince   Andriole,         Yale

14   University, and a previous member of this august and

15   dye infective advisory committee, and previous Secretary

16   of the Society, and President of the Society.

17                  And a person who was involved in the guideline

18   preparations in 1988 to 1990, and the four of us are here

19   to represent the Infectious Disease Society of America.

20                  CHAIRMAN RELLER:       And also at the table on

21   the far left, Dr. McCracken.

22                  DR.     MCCRACKEN:           George       McCracken,

23   University     of    Texas,   Southwestern      Medical      Center,

24   Pediatric and Infectious Disease, but also a member of

25   IDSA.

                              S A G CORP.
     202/797-2525             Washington, D.C.            Fax: 202/797-2525

1                   CHAIRMAN RELLER:        Thank you.        Tara.    We

2    have the facing table on the far right, and would Dr.

3    Tally begin, and then we will move to his right.                 Dr.

4    Tally.

5                   DR. TALLY:    Thank you, Barth.       I am Frank

6    Tally, from Cubist Pharmaseuticals, where I am the Chief

7    Scientific Officer.

8                   DR. SHLAES:     I am David Shlaes, and I am here

9    to today representing PhRMA, part of the PhRMA group.

10   I run the infectious disease discovery research group,

11   in the therapeutic area, for Wyeth-Ayerst.

12                  DR.   CHUANG-STEIN:           I      am     Christy

13   Chuang-Stein, Statistician, from Pharmacia Corporation,

14   here representing Pharmacia as well.

15                  DR.   METLAY:        Josh   Metlay,       from    the

16   University of Pennsylvania, from the Departments of

17   Medicine and Epidemiology.

18                  CHAIRMAN RELLER:      And lastly, Dr. Fleming.

19    He was here earlier, and Dr. Temple just joined us at

20   the table.

21                  DR. TEMPLE:     Bob Temple, Associate Director

22   for medical policy at FDA.

23                  CHAIRMAN RELLER:      Thank you.   We will begin

24   the presentation with opening comments from Dr. Mark

25   Goldberger, who is the acting director of the Office of

                             S A G CORP.
     202/797-2525            Washington, D.C.          Fax: 202/797-2525

1    Drug Evaluation for the FDA.                 Mark.

2                        DR. GOLDBERGER:        We would like to extend our

3    welcome to the advisory committee members, guests,

4    consultants, and everyone else in the audience who is

5    here    attending        what     has    been      a    reasonably       highly

6    anticipated event, I think.

7                        Our goal in having this meeting, which we

8    regard as the start of a process, is ultimately to ensure

9    that we have antimicrobial therapy that is in fact

10   adequate to meet the broad range of therapeutic challenges

11   that    we     face,     challenges        that    range       from    routine

12   infections,         to   very     difficult        to    treat     infections

13   illnesses, and of course some of the challenges having

14   only been heightened by some of the recent events in our

15   country.

16                       To accomplish this, we obviously need to

17   consider approaches to facilitate the development of new

18   antimicrobials, as well as to consider ways to preserve

19   the    usefulness        of   those     products        that     are   already

20   available.

21                       We regard this as the beginning of a process.

22    We are having today's meeting, tomorrow's meeting

23   dealing        on   issues      related      to    the    development         of

24   antimicrobials for resistant indications.

25                       As Dr. Turner noted, we have established a

                                   S A G CORP.
     202/797-2525                    Washington, D.C.               Fax: 202/797-2525

1    docket, which I think will be open for the next four months

2    or so to ensure that we get comments and participation

3    from the broadest range of individuals and organizations

4    who   are      involved   or   interested      in   the   process    of

5    antimicrobial drug development and infectious disease.


7                     We will be presenting some questions for

8    discussion today, but again these questions are really

9    for discussion, and we will not be asking for any formal

10   vote on them, nor do we anticipate reaching any decisions

11   as the results solely of the discussions today.

12                    I think that we certainly recognize the

13   issues that and that it is important to consider the

14   resources required to perform clinical trials, as well

15   as the types of information that we would like to be able

16   to get from such studies, and at times it appears as though

17   these two things, these two issues, have a certain tension

18   between them.

19                    And that is sort of the subject of much of

20   our discussion and I think some of the questions that

21   we will be asking this afternoon.             We certainly believe,

22   and the FDA has long used this approach, that the quantity

23   and strength of evidence should take into account the

24   seriousness of the disease, and the availability of

25   alternative therapy, and again we think that the questions

                                S A G CORP.
     202/797-2525               Washington, D.C.           Fax: 202/797-2525

1    we are posing, as well as the substance of much of the

2    discussion today, will focus on issues like that as well.


4                   And I would like to thank everybody.           We are

5    looking forward to a very interesting discussion today.


7                   CHAIRMAN RELLER:        Thank you, Mark.    Our next

8    speaker will be Dr. Renata Albrecht, who is the Acting

9    Director, Division of Special Pathogen and Immunologic

10   Drug Products as FDA.

11                  And   she     will     speak   to   the    "Historic

12   Perspective, Selection, and Implications of Delta."               Dr.

13   Albrecht.

14                  DR. ALBRECHT:        Thank you, Dr. Reller, and

15   good morning everyone.         I would like to add my words of

16   welcome to Dr. Reller and Members of the Committee,

17   guests, and colleagues.

18                  My task this morning will be to give you a

19   brief historical perspective on the selection of Delta,

20   and to talk about the implications of Delta on clinical

21   trials and patient care.          Next slide, please.

22                  Many of you may recall that originally this

23   meeting was scheduled for September 13th of last year,

24   and in fact the meeting had been planned for the better

25   part of the year, but needed to be postponed because of

                                S A G CORP.
     202/797-2525               Washington, D.C.         Fax: 202/797-2525

1    national events on September 11th of 2001.

2                   Discussion of Delta and related issues,

3    however, continued in the intervening 6 months, and

4    resulted in two letters being sent to clinical infectious

5    disease, which have been added to the background material

6    for this talk or for this meeting.

7                   Members of the Office of Drug Evaluation IV

8    had the occasion to have discussions with individuals

9    from academia and representatives from industry, and as

10   a result of these discussions, we have expanded the agenda

11   to include presentations by these groups.

12                  I would like to speak on a few broad areas.

13    One is the historical perspective on the role that Delta

14   has   played   in   regulatory    decision     making,    and    the

15   procedures used to select Delta as outlined in the 1992

16   points to consider document.

17                  Then I would like to speak about the impact

18   of delta on clinical trials, and finally the consequence

19   that Delta has on patient care.            Next slide.

20                  During most of today's presentations there

21   will be detailed discussions on the definition of Delta,

22   as well as the scientific and clinical issues important

23   in the process of selecting Deltas.          So I will not cover

24   these in my presentation.

25                  Instead, I wish to address the question of

                             S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    why is Delta important, and what role has Delta played

2    in the regulatory decision process.          Next.

3                     In general, the regulatory decision about

4    a particular product for a particular indication has been

5    that; if the Delta of the trial is met, the indication

6    is approved, and if the Delta is not met, the indication

7    is not approved.

8                     There have been rare exceptions to this

9    pattern.       For some drugs and indications, the Delta was

10   met, but the indication was not approved due to concerns

11   about the drug safety.       And in some examples, the Delta

12   was not met.

13                    Yet, the indication was approved due to an

14   overall risk benefit evaluation of the product, and in

15   those cases the results of the trials were reflected in

16   the product labeling.       Next.

17                    Thus, one may conclude that Delta has been

18   one of multiple important factors considered in making

19   a regulatory decision.        Next slide.

20                    So how do we select Delta?      The selection

21   of Delta has been guided by the 1992 points to consider

22   document, entitled, "Clinical Development and Labeling

23   of Anti-Infective Drug Products.             This document is

24   available on the FDA Guidance Website, and is also

25   included in the background material.         Next slide.

                              S A G CORP.
     202/797-2525              Washington, D.C.       Fax: 202/797-2525

1                      The       1992     points       to    consider        document

2    suggested        that       the    95    percent       confidence       interval

3    approach may be used, and recommended that Delta be based

4    on the observed success rate.                   So as shown in the green

5    rectangles to the left, for a 90 percent success rate,

6    the recommended Delta is 10 percent.

7                      For an 80 percent success rate, it is 15

8    percent.        And for 70 percent the Delta is 20 percent.

9    And   as       seen    in    the    rectangles         on   the    right,      the

10   corresponding sample size is 142 patients, 112 patients,

11   and 83 patients per arm, respectively.                       Next slide.

12                     The points to consider document also stated

13   that the design and conduct of clinical trials was

14   influenced by factors such as incidents of infection,

15   natural history of infection, realistic numbers of

16   patients available for study, cure rates of other; that

17   is, control drugs.

18                     In        addition,       one      has    to      take     into

19   consideration properties of the test drug, such as

20   pharmacokinetic and pharmacodynamic properties; in vitro

21   microbiology data, information from already approved

22   indications, and safety and efficacy data on other drugs

23   within the drug class.                  Next.

24                     The document also advised that demonstrating

25   effectiveness is one part of the burden of proof, and

                                       S A G CORP.
     202/797-2525                      Washington, D.C.               Fax: 202/797-2525

1    that a risk benefit profile for the drug must be

2    established.

3                    The document also stated that there are

4    situations where the morbidity and mortality of the

5    illness under evaluation will dictate that an absolute

6    difference      in     success      rates       will   be     clinically

7    unacceptable.        Next.

8                    However, over the years the step functions

9    specified in the points to consider document persisted,

10   while the other elements were lost, much like the body

11   of the Nike of Samothrace remains, while her head does

12   not.

13                   Therefore,       the    agency     held     an   advisory

14   committee meeting in 1998, during which a draft general

15   statistical guidance was presented, and then in February

16   of 2001, the agency published a disclaimer to the points

17   to consider document, stating that the sliding scale

18   method for determination of Delta was no longer used.

19                   Both    of    these       events   will     be     further

20   discussed      by    Drs.    Lin    and     Brittain      during     their

21   presentation.         So in 2001 then, the agency started

22   putting together motions and plans for this advisory

23   committee meeting to allow for a public discussion of

24   the selection and determination of Delta as Dr. Goldberger

25   stated in his introductory remarks.                Next.

                                  S A G CORP.
     202/797-2525                 Washington, D.C.             Fax: 202/797-2525

1                   As   we   hear    the        presentations    on    the

2    statistical and clinical issues for selecting Deltas,

3    it is important to keep in mind the impact these decisions

4    will have on clinical trials.

5                   This is the same slide that I showed earlier

6    about    the   95   percent   confidence        interval    approach

7    suggested in the 1992 points to consider document.

8                   This approach is familiar to industry, and

9    suggests the sample size of around a hundred to 150

10   patients per Arm for most clinical trials.           However, what

11   if an alternative Delta is selected.              Next.

12                  For the sake of illustration, and also in

13   the interest of time, I am going to focus on the impact

14   of selecting Deltas that are the same as or smaller than

15   suggested in the 1992 points to consider document.

16                  So if one were to say that a Delta of 10

17   percent should be used for all studies, meaning that the

18   test drug could be no more than 10 percent worse compared

19   to the control drug, the same size for the study with

20   a 90 percent success rate remains at a 142 patients per

21   arm.

22                  However, for a drug with an 80 percent success

23   rate, the sample size would double from 112 to 252 patients

24   per Arm; and for a 70 percent success rate, it would

25   increase four-found, from about 83 to approximately 330

                              S A G CORP.
     202/797-2525             Washington, D.C.            Fax: 202/797-2525

1    patients per Arm.          Next slide.

2                      And     if   one    were    to    take   an    even   more

3    conservative approach and select a Delta of 5 percent

4    for a trial, the sample size would increase four-fold

5    from 142 to 565 patients per Arm, with a success rate

6    of 90 percent.

7                      For an 80 percent success rate, the sample

8    size would go from 112 to 1,005, which is a nine-fold

9    increase; and finally if the study has a success rate

10   of    70       percent,    the       sample       size   would     increase

11   approximately 16-fold from 83 patients to 1,319 patients

12   per Arm.        Next, please.

13                     So the clinical trial implications of Delta

14   are the following.             For a given Delta, the lower the

15   success rate, the larger the sample size.                  And for a given

16   success rate, the smaller the Delta, the larger the sample

17   size.      Next slide.

18                     This relationship is nicely illustrated and

19   summarized in this graph, and I would like to thank Drs.

20   Lin and Brittain for making this slide for us.                      In this

21   graph the X-axis represents the success rate, and the

22   Y-axis the sample size, and the different colored bars

23   represent Deltas.

24                     And as one can see, going from a Delta of

25   20 percent, the light blue, to 15 percent, green, and

                                    S A G CORP.
     202/797-2525                   Washington, D.C.             Fax: 202/797-2525

1    10 percent, a darker blue; and 5 percent, the yellow,

2    and the sample size goes up.

3                     And the same pattern is seen as one goes from

4    a success rate of 80 percent, 70, 60, 50 percent, and

5    the sample size for all of the Deltas do go up.                   So as

6    we can see from these numbers, the demands on clinical

7    trials and the impact of these has impact on a variety

8    of groups and stakeholders.

9                     For industry and investigators, there is a

10   time commitment and a cost commitment of doing clinical

11   trials.        And the larger the trials, the more time and

12   resources       they   will    take.        Clinical   trials    impact

13   physicians, health care providers, and pharmacists who

14   rely on the availability of information from such studies

15   to guide their knowledge of drugs and use of drugs in

16   patient care.

17                    And clinical trials impact patients.               They

18   impact patients as participants in clinical trials.                  The

19   larger the study, the more patients need to participate.

20    And they impact patients as recipients of drug therapy.


22                    Clinical       trials      and   predefined     Deltas

23   determine the extent of information that is available

24   when making these treatment decisions for patients.                   So

25   in conducting a clinical trial, if one accepts a Delta

                                   S A G CORP.
     202/797-2525                  Washington, D.C.         Fax: 202/797-2525

1    of 15 percent instead of a Delta of 10 percent as evidence

2    of non-inferiority, the consequence may be that the drug

3    may be potentially 5 percent less effective than a drug

4    that would have been approved with a 10 percent Delta.

5                     And which also means that an extra 5,000

6    patients       may   potentially      fail    therapy    for   each

7    hundred-thousand patients treated.            Next slide.

8                     But things are never one-sided.        What if the

9    Delta selected for a trial is small, so small as to be

10   unrealistic, and then no clinical trial is conducted.

11                    Then in fact no clinical data are available

12   to guide patient treatment.             And even under the 1992

13   points to consider approach, some diseases were rarely

14   studied,       including   endocarditis,      osteomyelitis,     and

15   meningitis.

16                    So, in summary, the selection of Delta

17   impacts not just clinical trials and all parties involved

18   in clinical trials, but impacts patients who then use

19   the agents approved on the basis of these studies.

20                    Selection of Delta raises a number of issues

21   and questions, and we would like the committee and our

22   guests to provide us with comments on these issues.               As

23   Dr. Goldberger said, we are not asking for any votes on

24   any of these topics today.

25                    And in addition as Tara Turner said, we are

                                S A G CORP.
     202/797-2525               Washington, D.C.        Fax: 202/797-2525

1    making available Docket 98D-0548 for those groups and

2    persons who wish to provide us with written comments.

3                     After this meeting, we do plan on reviewing

4    these comments, and plan at least one follow-up advisory

5    committee meeting, and plan to summarize the advice in

6    updated guidance documents.          Thank you.

7                     Thank you, Dr. Albrecht.         We will next hear

8    from Dr. Robert Temple, who is the Associate Director

9    for Medical Policy, Center for Drug Evaluation and

10   Research, at FDA.

11                    Dr. Temple will speak to us about Active

12   Control        Non-Inferiority      Studies:        Theory,      Assay

13   Sensitivity, Choice of Margin.               Dr. Temple.

14                    DR. TEMPLE:     Well, good morning.          It is a

15   pleasure to be here to talk about one of my favorite

16   subjects, which is active control trials and how to

17   interpret them.

18                    We have as an agency been interested in this

19   in a very long time.      I have been writing about it since

20   the early '80s, and we have hinted in regulation since

21   1985 that equivalence trials presents special problems,

22   and have written various guidances for years about how

23   to analyze such trials.

24                    Susan Ellenberg and I wrote an article in

25   the Annals of Internal Medicine in September of 2000 that

                              S A G CORP.
     202/797-2525              Washington, D.C.            Fax: 202/797-2525

1    discusses the theory of all of this.           But probably the

2    most prominent document that we have participated in is

3    an International Conference on Harmonization document

4    called, "E-10, Choice of Control Group and Related Issues

5    in Clinical Trials," that was issued in 1997, I guess.

6                      Just in case anybody doesn't know, a little

7    bit about what the ICH is, because this represented a

8    remarkable degree of international harmony.              It is the

9    International Conference on Harmonization.

10                     And three regions -- the U.S., Europe, and

11   Japan,     made    an   effort   to    harmonize   the   technical

12   requirements for the marketing of drugs.                   Not the

13   approval decisions, but the technical requirements, where

14   disharmonies appeared to be unnecessary.

15                     They focused on what they called quality,

16   which means manufacturing control, and safety, which

17   means pharm/tox, and efficacy, which means human efficacy

18   and safety.

19                     And produced a series of mutually agreed upon

20   guidelines.       The participants in this organization are

21   the three regulatory authorities and their respective

22   manufacture organizations, such as PhRMA for the U.S.

23                     The organization develops guidance documents

24   in the three scientific areas, and these are then adopted

25   more or less uniformly in the three regions, and sometimes

                                S A G CORP.
     202/797-2525               Washington, D.C.         Fax: 202/797-2525

1    as guidance in the U.S.

2                   Sometimes        you     need      to     change       your

3    regulations, and in the final stage, the guidances are

4    controlled by the three regulatory bodies, and not the

5    pharmaceutical organizations.

6                   And as I said, there is no attempt to make

7    the decisions to the same.            The ICH E-10 document, which

8    can be found on our website and the other parties, is

9    called "Choice of Control Group and Related Issues in

10   Clinical Trials."

11                  And it is actually a general discussion of

12   all   kinds    of   control      groups,       including     historical

13   controls, which it doesn't like very much.                 It discusses

14   the ethics of placebos,and a wide range of other matters.

15                  But it devotes particular attention to the

16   use of active control equivalents, sometimes called

17   non-inferiority designs.           Not to dehumanize them as has

18   been alleged, and to say that they can't be used, but

19   to    describe      their     logic      and     their     inferential

20   difficulties, and to emphasize the need for evidence of

21   assay sensitivity, which I will describe in a moment.

22                  Much of what follows is considered an ICH

23   E-10, but that document discusses the issue of margin

24   and the distinction between M1 and M2, which is here called

25   Delta-1 and Delta-2. amd we actually tried to call it

                                 S A G CORP.
     202/797-2525                Washington, D.C.             Fax: 202/797-2525

1    that    in     the   document,   but    it         was    considered      too

2    statistical.

3                     Anyway, it discusses that rather minimally,

4    and so this meeting and others like it are an important

5    next step in all of this.        When it comes to demonstrating

6    efficacy, there are two quite distinct approaches.

7                     One is to show a difference between two

8    treatments in a randomized trial, or for that matter in

9    an     historical     controlled     trial.              That    shows    the

10   superiority of the test drug to whatever the control is

11   -- placebo, active drug, or a lower dose of the same drug

12   -- and that demonstrates a drug effect if you show such

13   a difference.

14                    The second approach is to show that the new

15   therapy isn't worse or isn't much worse than some of

16   therapy.        Showing   similarity          to    a    known    effective

17   therapy, and that is an inactive control, and attributing

18   the efficacy of the active control to the new drug, and

19   that in-turn demonstrates drug effect.

20                    There is nothing wrong with that logic, but

21   it poses certain problems, at least in some cases.                              A

22   non-inferiority trial, which is really what equivalence

23   trials are, shows that the new drug is not worse than

24   the control by some defined amount.

25                    That amount being the margin, M or Delta,

                                S A G CORP.
     202/797-2525               Washington, D.C.                 Fax: 202/797-2525

1    and that amount can be no larger than the effect that

2    the active control would have had in the study.                   If you

3    can't rule out a difference that large, then you have

4    not shown that the new drug has any effect at all.

5                    And I just want to emphasize that I didn't

6    change all of my slides.      M and Delta are interchangeable

7    terms.     We are not so far from a time when the naive

8    approach in active control trials was in fact used, and

9    in fact one can discover such a naive use in the recent

10   New England Journal of Medicine article, comparing

11   coumadin and aspirin in prevention of stroke.

12                   The idea is that you compare the new and

13   control drug, and if there is no significant difference,

14   then you declare the new and old drugs equivalent, and

15   the new drug is effective.

16                   The problem with that is that increase in

17   variance all by itself -- that is, making the study too

18   small -- will lead to success.              And that is now widely

19   understood.     So what is done now is that a non-inferiority

20   study specifies as a null-hypothesis that the new drug

21   is   inferior     by   some   margin,        M,   and     tests     this

22   statistically.

23                   So if the 95 percent confidence interval

24   upper bound for the degree of inferiority, that is, the

25   control drug minus the test drug, is less than M, then

                              S A G CORP.
     202/797-2525             Washington, D.C.               Fax: 202/797-2525

1    the null hypothesis of inferiority is rejected, and if

2    it were greater than M, then of course or then the

3    hypothesis is not rejected.

4                   If the confidence interval is very wide,

5    because the sample size is too small, the study will not

6    declare non-inferiority.      So it solves the size problem.

7     But it doesn't solve what I will describe as the assay

8    sensitivity problem.

9                   Any   time   you     do     an   equivalence      or

10   non-inferiority trial there is a question.               Did the

11   active control drug have an effect of the size expected

12   in the trial that you actually carried out.

13                  That may not seem like a pertinent question

14   in many antibiotic settings, but it is in lots of others,

15   most symptomatic treatments.        If the active drug didn't

16   have that expected effect, then showing equivalence or

17   non-inferiority by the expected margin -- and that is

18   a typo there, sorry -- by the expected effect, that's

19   meaningless, because the equivalent or non-inferior drug

20   could have no effect at all, and this study just is one

21   that could not tell anything from anything.

22                  So if no difference greater than the margin

23   is seen, does that mean that both drugs work or that

24   neither drug worked, and you have to know something from

25   outside the study to answer that question.

                             S A G CORP.
     202/797-2525            Washington, D.C.          Fax: 202/797-2525

1                   Assay sensitivity is a property of a clinical

2    trial, and it is the ability of the trial to distinguish

3    effective from ineffective drugs.            Assay sensitivity

4    depends on the effect size that you need to detect.                      A

5    trial may have assay sensitivity for an effect of 10,

6    but not an effect of five.

7                   So you really need to know what the effect

8    of the control drug was in that study, and of course,

9    you are not measuring it in an equivalence trial, and

10   so you have to learn it from historical information.

11                  So there is an unstated assumption in any

12   non-inferiority trial, which is actually nowadays it is

13   stated, but it used to not be stated, that the active

14   control was effective in the particular study.

15                  That    is,   that      the   trial      had     assay

16   sensitivity, and that is not necessarily true for all

17   effective drugs.      It is not testable in the data collected

18   because there isn't any placebo group.

19                  And it gives an active control study some

20   elements of a historically controlled study.                Again, I

21   know that I am repeating myself, but superiority equals

22   efficacy as long as the control is better than placebo,

23   which is usually safe to assume.

24                  And non-inferiority doesn't equal efficacy

25   unless assay sensitivity is present.          Assay sensitivity

                              S A G CORP.
     202/797-2525             Washington, D.C.            Fax: 202/797-2525

1    has   to       be   deduced   or   assumed       based   on   historical

2    experience showing sensitivity to drug effects, and that

3    means that it is usually possible to distinguish the

4    control drug from placebo.

5                        And then you have to do the study in a way

6    that doesn't mess it up.             If, for example, nobody took

7    the drug, then even an effective drug would not be

8    effective in a given trial.

9                        And it is important to make the new trial

10   as similar as possible with respect to patient population

11   and end-points as the trials in which the active control

12   was effective.

13                       This is just an advert from three-arm trials,

14   where there is both an active control and a placebo, which

15   is nice if you can do it.                So what one component of

16   deciding that a drug -- that a control drug -- that a

17   study has assay sensitivity, is to look for historical

18   evidence of sensitivity to drug effects.

19                       That means that well-designed trials pretty

20   regularly can distinguish the active drug from placebo.

21    Sensitivity of drug effects is an abstract conclusion,

22   and assay sensitivity is a conclusion about a particular

23   trial that takes historical evidence of sensitivity to

24   drug effects, and adds to it a proper study quality.

25                       Now, many people don't appreciate this.

                                   S A G CORP.
     202/797-2525                  Washington, D.C.           Fax: 202/797-2525

1    When you raise the issue of assay sensitivity, and say,

2    well, not every drug is effective against placebo every

3    time, and the next question is, well, why did you approve

4    a drug that bad.

5                      And the answer is that is the best that we

6    can do.        There is some settings in which it is not easy

7    to   distinguish        drug    from     placebo.     Some     of    these

8    situations are very well understood, and antihistamines

9    are very hard to show a difference between drug and

10   placebo, because the pollen blows away, and then you can't

11   see anything.

12                     And we know that studies of antidepressants,

13   even the effective antidepressants that we all know and

14   love, fail a significant fraction of the time.

15                     We have looked over several years, and in

16   about almost 50 percent of well done trials, or apparently

17   well-done trials, and they are as done as near we can

18   tell, of effective antidepressants, can't tell drug from

19   placebo.

20                     And   no     one   yet    knows   how   to    choose        a

21   population sample size or design that would alter that

22   state and everybody would like to, because failed trials

23   are a burden for everyone.

24                     And just a list of situations in which studies

25   of current drugs cannot be assumed to have sensitivity

                                    S A G CORP.
     202/797-2525                   Washington, D.C.           Fax: 202/797-2525

1    to drug effects include depression, anxiety, dementia,

2    symptomatic      congestive     heart      failure,     seasonal

3    allergies, GERD, which is the devil to study.         Systematic

4    GERD, I mean.

5                   It is post-infarction beta blockade, and only

6    about post-infraction aspirin, and only about 5 out of

7    35 studies have actually shown a benefit.

8    Post-infarction aspirin, only occasional studies show

9    an effect on survival, and the largest study ever leaned

10   the wrong way.

11                  That doesn't mean that the drugs that are

12   approved aren't effective for these conditions.         It means

13   that you have a problem if you are going to do an active

14   control trial, because you can't be sure that the drug

15   would have an effect in your particular study.

16                  It is always worth remembering that even if

17   sensitivity to drug effects does exist for a therapeutic

18   class assay sensitivity in a particular study, can be

19   undermined by a variety of study, conduct factors, that

20   give you a bias towards the null.

21                  That is, obscure true differences between

22   treatments just to illustrate these.          Poor compliance,

23   and nobody takes the drug, and the drug can't tell drug

24   from placebo.

25                  Too many cross-overs, and a population for

                             S A G CORP.
     202/797-2525            Washington, D.C.         Fax: 202/797-2525

1    one   reason   or   another    improves     very   rapidly,      and

2    spontaneously.      On the other hand, a population that is

3    very resistant, and too much use of concomitant medication

4    that treats everybody independent of the drugs so that

5    you can't see a drug effect anymore.

6                   Poor diagnostic criteria.      You put the wrong

7    people into the trial.        Insensitive measures of a drug

8    effect, and poor quality of measurements, mixing up the

9    treatment.     All of these things don't necessarily affect

10   variance very much, but they might affect the treatment

11   size.

12                  It is also worth remembering that what you

13   think you know about historical evidence really applies

14   only to trials of a particular design, and different

15   trials may or may not have that property.             Changes in

16   these can effect the size of the active control effect,

17   and therefore one's choice of margin, or in fact even

18   completely undermine assay sensitivity.

19                  So the non-inferiority margin, or delta, and

20   completely equivalent terms, is the degree of inferiority

21   of test drug to control drug that the trial is going to

22   exclude statistically.

23                  In other words, if you take the 95 percent

24   confidence interval for the difference between control

25   and test drug, it has to be less than that margin, whatever

                             S A G CORP.
     202/797-2525             Washington, D.C.          Fax: 202/797-2525

1    that margin or delta is.

2                      Obviously the margin can't be any larger than

3    the effect the control drug would be reliably expected

4    to have.        And we will call that M-1 or Delta-1, and if

5    M-1 is the entire effect, the control drug can be presumed

6    to have in the study.

7                      And if C minus T is greater than M-1, then

8    the new drug has no effect at all.             And it is always worth

9    remembering that the choice of margin is very critical

10   for everybody, including regulatory agencies.

11                     And if you allege that the control drug has

12   an effect on M-1, and find that the control minus test

13   drug is less than M-1, then the test drug is effective,

14   which is what we want.

15                     But if in this trial you are wrong, and it

16   really only had an effect size of half of M-1, then the

17   test drug will not really have been shown to be effective,

18   and it will only look at that way, and that is why we

19   worry.

20                     The margin used in a trial could be the entire

21   effect     of     the   control   drug        for   many     symptomatic

22   conditions.        We are content to know that the drug has

23   any effect.       But the margin chose could be smaller, and

24   we have been calling that M-2, or delta-2, if there were

25   a clinical need to assure preservation of more than just

                                S A G CORP.
     202/797-2525               Washington, D.C.                Fax: 202/797-2525

1    some of the control drug effect.

2                    That is, preservation of some fraction of

3    the effect of the control drug, or some absolute benefit.

4     Choosing an M-2 smaller than the whole effect of the

5    control may be important when the effect is clinically

6    critical.      For example, mortality.

7                    It might then be 50 percent, which would be

8    25 percent, and you wouldn't want to lose more than 25

9    percent of the effect of the control agent, or as you

10   will see, sometimes even less.

11                   Just to illustrate these, these are five

12   examples.      What you see on the left axis is the difference

13   in effect of what I have been calling C minus T, and there

14   are five examples, 1 through 5 across the top.

15                   And the top dotted line is M-1, and that is

16   the whole effect of the control drug, and M-2 is some

17   smaller effect, because you want to preserve more than

18   just any effect.      And M zero is the line of equivalence.

19                   In example number one -- and this is the point

20   estimate, plus a confidence interval for the difference

21   between C minus T.      The drugs look about the same, and

22   the confidence interval is narrow, and so you have shown

23   that the effect is at least M-2.

24                   And if that is what you were trying to do,

25   you are happy.        In Number 2, the point estimate is

                              S A G CORP.
     202/797-2525             Washington, D.C.       Fax: 202/797-2525

1    somewhat adverse to the new drug, and the confidence

2    interval includes a value larger than M-2.     So you have

3    not ruled out loss as drawn here, say 50 percent of the

4    effect, although it does look as if it has some effect.

5                   In M-3, the point estimate is adverse to the

6    new drug, and now the confidence interval includes a value

7    that is even worse than the whole effect of the drug.

8    So in this case, you haven't really shown that the drug

9    does anything.

10                  Example Number 4 shows superiority to the

11   control drug and that is always good.      And in the fifth

12   example, it shows a point estimate that is favorable,

13   but the study is too small or something else is wrong

14   with it, so that you haven't excluded a loss of all of

15   the effect of the drug.

16                  Just briefly, this will be a point discussed

17   later.     In the past, and actually some of the original

18   descriptions of non-inferiority studies, the margin was

19   chosen clinically.

20                  That is, you decide how much difference you

21   were willing to accept, and you rule that out.          Where

22   the effect of the drug is very large, which is certainly

23   the case in many antibiotic settings, and certain highly

24   responsive tumors, that is okay.

25                  You don't have to worry about losing all of

                            S A G CORP.
     202/797-2525            Washington, D.C.     Fax: 202/797-2525

1    the effect of the drug, because it is very easy to tell

2    the    difference   between      an    effective     drug     and   an

3    ineffective drug.

4                   If   you   are     looking      at   urinary     tract

5    infections, you don't have to worry about whether your

6    effect side is 10 percent or 20 percent.              You would be

7    able to tell an ineffective drug from an effective drug.

8                   So the only thing you are really interested

9    in is how much of that effect you are willing to lose.

10    That is, M-2 becomes the matter of interest.                       In

11   oncology, for many years, we considered assurance that

12   you hadn't lost more than 20 percent of the survival of

13   the population, an acceptable evidence of effectiveness.

14                  The trouble with that was that the drugs that

15   were being used as the control drugs didn't have an

16   effective survival that large.               So that what we were

17   ruling out in many cases didn't rule out the possibility

18   that the drug had no effect at all.

19                  Anyway, that is going to be an important

20   discussion later.     So I won't dwell on it now, except

21   with this one slide.      In many situations, the effect is

22   very large, and there isn't really a problem in knowing

23   what the historical -- in knowing that a trial has assay

24   sensitivity.

25                  If acute lymphocytic leukemia has a complete

                               S A G CORP.
     202/797-2525              Washington, D.C.           Fax: 202/797-2525

1    response rate of 80 or 90 percent, you don't have to worry

2    about ruling out a difference of 50 percent of that, or

3    60 percent.      You are going to worry about how much

4    clinical difference you are willing to accept, and so

5    you are going to worry mostly about M-2 or Delta-2.

6                   Similarly,   for    testicular   cancer,    acute

7    response to bronchodilators, anesthetic effects, and even

8    in the case of thrombolytics, a look at the available

9    data shows that it is fairly easy to tell whether an active

10   drug is -- to be sure that a drug is active in a particular

11   study.

12                  But you know more about this than I do, but

13   that that would be equally true for urinary tract

14   infections, meningitis, and lots of other situations.

15   One of the things you will talk about is how much effect

16   needs to be retained in situations where the effect size

17   is large.

18                  And of course it is worth remembering that

19   the very reason that you can't do a placebo control trial

20   is the reason for assuring that you are preserving a good

21   part of the effect of the control agent.

22                  So, for thrombolytics, we have said that you

23   need to show that you are not -- that you have not lost

24   50 percent of the effect, and in certain cancer drugs,

25   we have asked for retention of 50 percent of the survival

                            S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    effect, where that is a matter of a few months.

2                     In adjuvant breast cancer, however, we have

3    asked that you preserve at least 75 percent of the effect

4    because one does not want to lose more than 25 percent.

5                     This is in some sense a practical question,

6    and one doesn't actually want to lose any of the effect

7    of the control when it has an important effect, but sample

8    size has become rapidly out of sight when you try to do

9    better.

10                    Thrombolytic trials show that you preserve

11   50 percent of the effect in 14,000 people, if you wanted

12   to preserve 75 percent of the effect, you would get into

13   the 70,000 range.

14                    Again, as you will hear, there are at least

15   a few situations where the effects of active agents is

16   not so large, hard to discern, and hard to demonstrate.

17    And when that's true, then a non-inferiority design

18   becomes a problem, and one does have to think both about

19   M-1 and M-2, and it may be very difficult to use a

20   non-inferiority design, and therefore placebo controls

21   need to be considered.

22                    One   question    is    whether   those     will    be

23   ethical.       So a brief word about ethics, which ICH E-10

24   considers      at   some   length.        That   document     clearly

25   distinguishes       between   available       drugs   that    prevent

                                S A G CORP.
     202/797-2525               Washington, D.C.           Fax: 202/797-2525

1    serious harm and those that treat symptoms.

2                   As a general matter, where an available

3    treatment is known to prevent serious harm, death, or

4    irreversible morbidity in the study population, you

5    really can't use a placebo control.

6                   The   only   generally      is   a   hedge    because

7    sometimes the drug is so toxic that people will reject

8    it anyway.     Where there is no serious harm, however, it

9    is generally considered ethical to ask patients to

10   participate in a placebo control trial even if they may

11   be uncomfortable, provided the setting is non-cohesive,

12   and that patients are fully informed about available

13   therapies.

14                  Of course, it is also true that whether a

15   particular placebo control trial will be acceptable to

16   patients and investigators is a matter of investigator,

17   patient, and IRB judgments.       So it might be ethical, but

18   it might be that no one would be in it.

19                  One question again, and this is just the

20   briefest introduction, but it may be possible to design

21   trials where it is impossible or difficult to specify

22   M-1 that randomize patients to drug and placebo, and

23   preferably with an active control as well, and that allow

24   early escape for any one not doing well.

25                  For example, failing to respond by time-X

                             S A G CORP.
     202/797-2525            Washington, D.C.             Fax: 202/797-2525

1    or something like that.            Again, you will hear a great

2    deal more about all of this.              Thank you.

3                     CHAIRMAN RELLER:           Thank you, Dr. Temple.

4    We will next hear from the Statistical Team Leader, Dr.

5    Daphne Lin, and Dr. Erica Brittain, the Statistical

6    Reviewer for FDA on Statistica Issues in Specification

7    of Delta.       Dr. Lin.

8                     DR. LIN:       Thank you, Dr. Reller.             Good

9    morning.       This is a joint work with Dr. Erica Brittain.

10    We     are    going   to    present       statistical    issues      in

11   specification of delta.

12                    I am going to give the first part of the talk,

13   and later Dr. Brittain will cover the second part.                   The

14   outline of our talk.           First, briefly, an introduction

15   to non-inferiority trials, and non-authority margin; that

16   is, delta.

17                    Later I will give a brief introduction, a

18   brief history, of the reaction in FDA's anti-infective

19   drug product area.          Later, Dr. Brittain will talk about

20   the principles for determining Delta, and difficulties

21   in practice, and alternative design, and finally a summary

22   will be made.

23                    If there is a new drug, how can we show the

24   new drug has identical efficacy to the standard of the

25   drug?      A short answer is that we can't.                   And the

                                  S A G CORP.
     202/797-2525                 Washington, D.C.          Fax: 202/797-2525

1    alternative     availability      in       the   clinical     trial,

2    statistically, we cannot prove the effect of treatments.

3                   So what can we do?      The short answer is that

4    we must allow for some potential difference in efficacy,

5    and that is Delta, the topic of today's talk.

6                   So what is delta?       ICH E-9 has a definition

7    of delta, which is that it is the largest clinically

8    acceptable difference, and it should be smaller than

9    differences observed in superiority trials of active

10   comparator.

11                  Or Delta can be described as the largest

12   acceptable line in efficacy between tests and the active

13   counter drug.     For example, if we tried to design a

14   meningitis trial, then what is the largest clinically

15   acceptable difference between tests and the active

16   counter drug?

17                  We can design a non-inferiority trial to

18   answer the previous question.          A non-inferiority trial

19   is designed to ensure that the new drug is not worse than

20   the standard drug by some margin delta.

21                  In the anti-infective drug product area, in

22   general, what defines treatment effect as the absolute

23   difference is the absolute difference of percent cure

24   rates.

25                  For example, if an observed success rate in

                             S A G CORP.
     202/797-2525            Washington, D.C.            Fax: 202/797-2525

1    control is 85 percent, and the observed success rate of

2    test drug is 75 percent, then the point estimate of

3    difference is 10 percentage points.

4                   An in general a confidence interval around

5    this estimate of treatment effect is used as the primary

6    analysis for non-inferiority trials.               So what is a

7    confidence interval?

8                   The 95 percent confidence interval for the

9    difference in success rate between two drugs means we

10   are 95 percent confident of that.          Now the true difference

11   in efficacy between these two drugs is contained in the

12   confidence interval.

13                  Next,   let me give you two examples to

14   illustrate and how to use the 95 percent confidence

15   interval to interpret the result from a non-inferiority

16   trials.

17                  The first example is if a trial of two hundred

18   patients per Arm, designed with a delta of 10 percent,

19   and if the trial results shows the success rate of the

20   test drug is 88 percent.      Control drug, 90 percent, and

21   if the point estimate of the difference is minus 2 percent,

22   and the 95 percent confidence interval along this point

23   estimate is between minus 8.6 and the 4.6 percent, and

24   in this example, since the 95 percent lower limit is no

25   less than 10 percent -- I'm sorry.

                             S A G CORP.
     202/797-2525            Washington, D.C.            Fax: 202/797-2525

1                     So in this example, which can concur the test

2    drug is non-inferior to the contour.                 The second example

3    is similar design with a trial of 200 patients per Arm,

4    with 12 percent.

5                     However, in this example, the trial results

6    show the success rate of test drug is 84 percent and

7    control drug 90 percent, and the point estimate of the

8    difference is minus 6 percent.

9                     And the 95 percent confidence interval falls

10   between minus 13 and the 1.1 percent.                       And in this

11   example, since -- I'm sorry, I just don't know how to

12   operate this.

13                    So in this example, 95 percent lower limit

14   is   less      than   10    percent,     and    so    we    concur     that

15   non-inferiority        is    not    demonstrated.            From     these

16   examples, we can see that the decision of non-inferiority

17   depends not only on the success rate of test and control

18   drugs also depends on how Delta is chosen.

19                    There are two objectives in non-inferiority

20   trials, and the first objective is that non-inferiority

21   indirectly determine if the test drug is better than

22   placebo.

23                    And it directly determines if the test drug

24   is similar to the active control drug.                     So we need to

25   choose delta appropriately to achieve both objectives.

                                  S A G CORP.
     202/797-2525                 Washington, D.C.              Fax: 202/797-2525

1                   Next,   the history of the history of a

2    selection in FDA's Division of Anti-Infective Drug

3    Products area.     As Dr. Albrecht mentioned in her talk,

4    in 1992, points to consider in her document used the staff

5    step function approach.

6                   This slide shows the relationship between

7    Delta and the success rate described in the points to

8    consider document.     Choice of delta only depends on the

9    success rate.     If the success rate is greater or equal

10   to 90 percent, delta is 10 percent.

11                  If the success rate is in the 80 percent

12   range, delta is 15 percent.        if the success rate is in

13   the 70 percent range, delta is 20 percent.          Since this

14   is a step function which can lead to problems of

15   interpretation, and if a few outcomes are changed, then

16   a different standard will be used for evaluation.

17                  For example, if the success rates is changed

18   from 80 percent to 79 percent, and delta will be changed

19   to 15 percent to 20 percent.         Since delta in points to

20   consider has been chosen primarily based on success rate,

21   it did not take into account the seriousness of disease

22   and the consequence of treatment value.

23                  And whether delta was small enough that a

24   drug with no efficacy could meet the standard was not

25   considered.

                             S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1                   In addition, as I described previously, this

2    step    function   approach   has    undesirable   statistical

3    properties.    Another concern.       If the active control arm

4    and the delta are not appropriately chosen, then the

5    so-called "Bio-Creep phenomena may happen.

6                   And that is that if trials over time used

7    progressively less effective control arms, and the delta

8    is not appropriately chosen, then they are already in

9    attenuation of efficacy.

10                  For example, if Drug 1, with a success rate

11   of 70 percent, is used as an active comparator to compare

12   with the new test drug Number 2, with a success rate of

13   60 percent, and if a delta of 20 percent is used, then

14   in this case, Drug 2 is not inferior to Drug 1.

15                  And if later on there is another test drug,

16   Test Drug Number 3, and if Drug Number 2 is used as an

17   active comparator, and if a delta of 20 percent is still

18   being used, then we might approve a drug with a success

19   rate of 48 percent, which is much lower than the success

20   rate of Drug Number 1.

21                  Another case, and the worst case scenario,

22   how about if the placebo rate is here, and that is that

23   we might have another drug which is not much different

24   from the placebo.

25                  In July of 1998, on the advice of the

                             S A G CORP.
     202/797-2525            Washington, D.C.         Fax: 202/797-2525

1    committee, we have discussed that, and the choice of delta

2    should reflect many important clinical factors, such as

3    historical cure rate with and without therapy, risk

4    associated with treatment failure, and advantages and

5    disadvantages of study drug.

6                      In addition, in '98, on the advice of the

7    committee, we also proposed that when delta is chosen

8    for simple size computation, it should be clinically

9    relevant, and since delta will be picked based on clinical

10   issues, it may need to be indication specific, and they

11   are some special situations for individual indications

12   when delta may need to be chosen on a case by case basis.

13                     In addition, we also encourage sponsor to

14   discuss the choice of delta with the Medical Division

15   during protocol development.                   And a sponsor should

16   provide the rationale for selection of control arm.

17                     The CPMP, counterpart of the FDA, published

18   a guidance on the evaluation of anti-bacterial medicinal

19   products in 1997.       And this guidance recommended a delta

20   of    10       percentage   points      for      common     non-serious

21   infections.

22                     But it needed to be smaller for very high

23   cure rates.        Also, this guidance recommended the choice

24   of delta should be based on the clinical judgment, and

25   it is based on a minimum clinically relevant difference,

                                 S A G CORP.
     202/797-2525                Washington, D.C.              Fax: 202/797-2525

1    and should be justified in the protocol.

2                         For the past two years, we have worked with

3    sponsors on a case-by-case basis to specify delta.                      In

4    February of last year, a disclaimer was added to the points

5    to consider document, stated that the step function

6    approach has been phased out, and the choice of delta

7    should follow the ICH E-10 principles, and there is a

8    need to establish standards.             This is the end of my talk.

9                         Next, Dr. Brittain will talk about a general

10   principle for selection of delta.                  Thank you for your

11   attention.

12                        DR BRITTAIN:    Okay.       So, now what?   Here is

13   a road map for the rest of the talk.                  I am going to be

14   talking about principles for determining delta, and these

15   are going to be based on the ICH E-10 principles, and

16   then the very real difficulties in practice.

17                        This is the hard part; how you apply this

18   in practice, and this is where we need your advice.                   Then

19   I will mention alternate designs, a summary, and I also

20   want to say that one of my main goals here is to get across

21   the idea that the choice of delta is not a technical

22   matter,        but    actually     one   that     potentially    impacts

23   patients.

24                        Again,   to     demonstrate       efficacy,       the

25   experimental drug needs to be better than placebo, and

                                   S A G CORP.
     202/797-2525                  Washington, D.C.           Fax: 202/797-2525

1    in some settings, it should have similar efficacy to the

2    existing therapy, and so we want to choose a delta to

3    assure that both of these goals are met.

4                    Here is an important quote from the E-10.

5    This design, "is appropriate and reliable only when the

6    historical estimate of the drug effect size can be well

7    supported by reference to results of previous studies

8    of the control drug."

9                    So what does this mean?       We must know with

10   good precision the magnitude of the advantage of the

11   active control drug over placebo in the setting of the

12   clinical trial.

13                   Now, in practice, as Dr. Temple was talking

14   about, if the advantage is very large, the precision of

15   this estimate probably won't matter.          On the other hand,

16   if it is potentially modest, the precision is critical.


18                   And the sort of unfortunate corollary of this

19   is the active control that is based on a single trial

20   with borderline efficacy, we are going to have poor

21   information       about   the     magnitude    to      support          a

22   non-inferiority trial.

23                   So here is some important principles from

24   the E-10.      First, a delta could based on both statistical

25   reasoning and clinical judgment; and, second, it cannot

                               S A G CORP.
     202/797-2525              Washington, D.C.          Fax: 202/797-2525

1    be larger than the advantage of the "active drug would

2    be reliably expected to have compared with placebo in

3    the setting of the planned trial."

4                   And it goes on to say that we usually choose

5    delta to be even smaller to ensure that some clinically

6    acceptable treatment benefit is maintained.

7    This is a very artificial example, but I hope that it

8    will convey some important concepts.

9                   Say we actually knew the true success rate

10   of the placebo was 70 percent, and the true success rate

11   of the active control was 85 percent.      So the difference

12   between 85 and 70 is 15.        So that is the advantage of

13   the active control over placebo.

14                  One could choose a delta of 15 percent, but

15   you could not use a delta larger than 15, because a drug

16   that has no efficacy has too high a chance of being

17   successful.

18                  And then you might say, well, I don't want

19   to have a drug that is down near the placebo rate.                I

20   would like to keep it up closer to that 85 percent rate.

21    So maybe you would want to preserve half the benefit

22   and have a delta of 7 percent.

23                  And then somebody else might think, well,

24   in a particular situation we don't want to lose much of

25   the benefit of the active control, and then you would

                            S A G CORP.
     202/797-2525            Washington, D.C.      Fax: 202/797-2525

1    want a delta of 3 percent.

2                   The main point here is that you can't go

3    bigger than 15, and there might be -- there are all sorts

4    of infinite choices of delta smaller than 15, depending

5    on the objective.

6                   And we have been using this approach to delta

7    as a two-step process and have found this way of looking

8    at it very useful.

9                   We first determine a conservative estimate

10   of the advantage of active control over placebo, the delta

11   one; and this is data based.           And then we select the

12   largest clinically acceptable difference between the

13   active control and the experimental drug, and we call

14   that delta two, and that is judgment based.

15                  And then the smaller of these two values would

16   be the delta that we would use in the non-inferiority

17   trial.     So what is this benefit of active control over

18   placebo.

19                  You could define that as a true success rate

20   of the active control, minus the true success rate of

21   the placebo in the setting of the clinical trial.

22                  In other words, by how much is the active

23   control better than placebo in the non-inferiority trial

24   setting if the placebo were actually present.

25   And again I want to emphasize that this is based on

                             S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    historical data.

2                     And it is not a judgment.               It is not a choice.

3     At some level there is a right answer.                    We may just have

4    trouble finding out what it is.                 And again it is not that

5    critical to get it just right if the benefit is very large.

6                     So why did I say conservative estimate?

7    Well,     E-10     says        delta,       quote,        should     reflect

8    uncertainties in evidence on which the choice is based,

9    and should be suitably conservative.

10                    The     problem      is    that     if     the    delta    is

11   overestimated, the chance of concluding efficacy when

12   the new drug is no better than placebo is too high.

13                    So if we are going to err at all, we want

14   to err on the side of underestimating the benefit.                          So

15   what    this     means    is     that      we     have    poor    historical

16   information.

17                    We are not going to use our best guess of

18   the estimate.          We want to use some smallest of the

19   reasonable values.          I know that I am being very vague

20   here, partly because even in the statistical community

21   there isn't agreement about exactly how to do that.

22                    Okay.    So what is the best information for

23   estimating the benefit of the active control, the delta

24   one?    The best case would be if you had a whole bunch

25   of placebo control trials, with exactly the same design

                                    S A G CORP.
     202/797-2525                   Washington, D.C.              Fax: 202/797-2525

1    that you want to use in the non-inferiority design.

2                    We just -- I don't think there is any

3    situation      that    in     anti-infectives         that     meets    that

4    situation.      Sort of halfway down this list would be if

5    you have multiple placebo control trials, but not with

6    the same design that you would want to use in the

7    non-inferiority trial, and maybe not with the same design

8    that the others have used.

9                    And     then     at    the       bottom   would    be    the

10   observational         data,    and    antidotal       data,     and     this

11   obviously is not the best situation, but again if we are

12   talking about large treatment effects, it is probably

13   fine.

14                   But the case that in a way is the most

15   interesting case for anti-infectives, what if we have

16   some placebo control data in the literature, but there

17   is some problems with it.

18                   The     trials     are    old,      and   so   antibiotic

19   resistance that is taking place in the meantime changes

20   in clinical care management may mean that the values in

21   the old trials aren't that valid or relevant.

22                   The proposed active controls may not be

23   studied because these trials were old, and there may

24   not be very many of them, and so we would not know if

25   the treatment effect is consistent.

                                   S A G CORP.
     202/797-2525                  Washington, D.C.             Fax: 202/797-2525

1                   And very importantly, there are probably

2    differences in entry criteria, assessment criteria, the

3    timing of the assessments, and the populations.             So as

4    wonderful as these data are compared to having no

5    information, we have to take the data with a big grain

6    of salt.

7                   So how do we then come up with an estimate

8    of this delta one with this situation, and we don't know.

9     We are hoping that you can give us some advice.           So the

10   bottom line for estimating delta one is we want to use

11   historic data, preferably from placebo control trials

12   with    similar   designs   as   possible   as   the    upcoming

13   non-inferiority trial.

14                  The bad news is that in anti-infectives, your

15   historic data is often poor, and maybe poor for good reason

16   because of ethical constraints in doing placebo control

17   trials.

18                  But the fact that the data is not there makes

19   it hard for us to come up with this conservative estimate.

20    And again the good news is the precision of this is

21   probably irrelevant for those indications where the

22   benefit is known to be very large.

23                  So again let me take you back.        This was a

24   two step process, and we are just talking about step one,

25   the determination of the estimate of the advantage of

                             S A G CORP.
     202/797-2525            Washington, D.C.         Fax: 202/797-2525

1    active control over placebo, delta one.

2                   And the second step is the acceptable loss

3    from active control delta two, and delta is the smaller

4    of these two components.           Now, the selection of delta

5    two is going to be the primary concern for the majority

6    of anti-infective indications probably.

7                   I    want   to     emphasize      that     unlike      the

8    delta-one, which really is pretty much a statistical

9    decision, this delta one, because of the clinically

10   acceptable loss, is not.         It is really a clinical judgment

11   of the largest acceptable difference between active

12   control and the new drug.

13                  It is a difference that is such that it is

14   so important clinically that it must be ruled out, or

15   you could think of it as a borderline value between just

16   barely acceptable and not acceptable.

17                  So   what    is    important      to     think     about?

18   Certainly the consequential treatment failure.                  If most

19   err study failures are deaths or very serious morbidity,

20   you would probably want to use a smaller delta two.

21                  If treatment failure can be easily reversed

22   or addressed, we could be more lenient.                 And then this

23   is an important way to look at it.            It is kind of obvious,

24   but if in fact the true loss in efficacy of the new drug

25   from the active control drug were say five percent, if

                                S A G CORP.
     202/797-2525               Washington, D.C.             Fax: 202/797-2525

1    a hundred-thousand patients used the new drug instead

2    of the active control, 5,000 extra patients would have

3    failures than if they had used the active control drug

4    and so on.

5                     And if the true loss is 10 percent, then there

6    would be 10,000 extra patient failures.        You could kind

7    of go down the right side and say what is the worst case

8    scenario that we could accept, and then see what delta

9    would correspond to that.

10                    Then there is another issue to think bout

11   with the clinically acceptable loss, and it is a little

12   more subtle, and I kind of call it clinical trial reality.


14                    It is clinical trials that measure the

15   abstract concept that we might be thinking about in our

16   minds.     For example -- and this would be one example of

17   a clinical trial reality.           And for those indications

18   where there are going to be patients in the studies who

19   do not have disease, and where the indications are hard

20   to diagnose the disease exactly.

21                    Say in a case where the treatment difference

22   among patients with a bacterial infection were 12 percent,

23   and a case with patients without a bacterial infection

24   is zero.       So if you had a 50-50 mix in your trial, the

25   treatment difference that you should be measuring would

                               S A G CORP.
     202/797-2525              Washington, D.C.      Fax: 202/797-2525

1    be six percent.

2                   So if you had selected a delta of 10 percent,

3    you may end up concluding the new drug is sufficiently

4    efficacious.    But notice that in the key population the

5    patients with the bacterial infections, the treatment

6    difference was actually greater than 10 percent.

7                   So we have to think about -- we can't just

8    think about the clinically acceptable loss in an abstract

9    way.    We need to know about how or what you are actually

10   measuring in the clinical trial.           And there are other

11   factors that can dilute treatment effects as well.

12                  So, in summary, for the selection of the

13   clinically acceptable loss, certainly the consequence

14   of treatment failures is primary in this consideration.

15    And then the potentially large impact on patient care.


17                  And then we have to be careful about these

18   clinical trial realities, and again I want to emphasize

19   that unlike the delta one, this component, the clinical

20   judgment is really the primary judgment.

21                  Now, for a long time we have been thinking

22   about selecting for each indication its own delta, and

23   this would provide regulatory consistency, but we want

24   to acknowledge that even once we have finally decided

25   what the delta should be for each indication, we are not

                             S A G CORP.
     202/797-2525            Washington, D.C.         Fax: 202/797-2525

1    going to be done, because we are going to have to stay

2    vigilant because we could have the bio-creep problem that

3    Dr. Lin mentioned.

4                   And that if we could keep changing the active

5    control, and that the delta may not be small enough.

6    And then emerging resistance on other temporal changes

7    can diminish the efficacy of any active control.

8                   So we are going to have to stay on top of

9    this unfortunately.     You are going to hear a lot today

10   about consequence to sample size.           When you assume that

11   the cure rates are the same in the active control and

12   the new drug, when you cut the delta in half, your sample

13   size quadruples.

14                  One other important thing to mention though

15   is that is the new drug, if it is reasonable to assume

16   that the new drug is slightly better than the active

17   control, the sample size can be sharply reduced.

18                  For example, in this particular case, say

19   you are using 80 percent power and you were using a delta

20   of 10 percent.    If you assumed that both cure rates were

21   80 percent, you would have about 250 in a group.

22                  But if you assumed that the new drug cure

23   was just a little bit better at 82 percent, your sample

24   size would be cut by one-third.            So what is the biggest

25   challenges?    And we have plenty of challenges for you.

                             S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1                   The biggest challenges are indications where

2    the treatment effect is potentially modest, but not

3    precisely known, and on sort of the flip side, serious

4    indications where we may be pretty comfortable that there

5    is a large treatment effect, but there is low incidence,

6    and so it is hard to do the kind of size studies that

7    we might want to do.

8                   Now,   superiority      designs   may   offer     an

9    important alternative to the non-inferiority design,

10   particularly in the first case.             They can provide

11   stronger evidence, and it in some situations with smaller

12   sample size.

13                  So the question is can they be done ethically.

14    The early escape approach that Dr. Temple mentioned is

15   something that we have been thinking about for quite some

16   time, and I know that it was discussed in the previous

17   advisory committee on a titus media, and a few people

18   brought this up as a possible situation for a titus media.

19                  But the question is whether it is ethical,

20   and this is applicable probably only to a handful of our

21   indications, the less serious ones, or potentially

22   applicable.

23                  But these are big indications in terms of

24   numbers of millions of prescriptions a year.             So these

25   are important indications.        The two arms, experimental

                             S A G CORP.
     202/797-2525            Washington, D.C.          Fax: 202/797-2525

1    versus placebo, the key element is that patients are seen

2    several days after baseline, and at that time if a blind

3    assessment      shows   no   improvement,   the     patient      is

4    considered a failure in the analysis, and then the therapy

5    is switched.

6                    Now, this is ethically consistent with the

7    way and see practice of medicine.            So if you are

8    comfortable with wait and see, you can be comfortable

9    with this.

10                   A variant of this would be an early escape

11   with three arms, where you would add the active control

12   arm, and obviously that would be the most informative

13   design.

14                   I just wanted to mention other superiority

15   designs.       I just want to encourage people to consider

16   superiority designs, even though the non-inferiority

17   design has been the mainstay in this area for so long,

18   we think it would be important to you to open to

19   considering other designs.

20                   One design could be like the placebo add-on

21   design, where the existing drug -- one arm is the existing

22   drug, plus the new drug, versus the existing drug, plus

23   placebo, which answers the question does the new drug

24   have benefit in the presence of the existing therapy.

25                   And a question would be labeling implications

                              S A G CORP.
     202/797-2525             Washington, D.C.         Fax: 202/797-2525

1    with that design.      But the dose response design versus

2    low dose situation, superiority to some comparator, or

3    perhaps some combination of these.

4                   Okay.   I want to move back to summarize the

5    selection of delta, the big picture.          Again, choice of

6    delta impacts patients.        If delta is incorrectly chosen

7    so that it is greater than the advantage of active control

8    over placebo, patients may end up getting drugs with no

9    benefit, while being exposed to toxicity, and there is

10   potential for development of resistance.

11                  And even in those situations where we are

12   not so concerned about the placebo rate, there is still

13   potential      benefits      of      using   smaller     deltas.

14   Potentially, more patients are cured overall and there

15   are higher survival rates, and subtle, but important,

16   differences are detected that might not be detected with

17   bigger deltas.

18                  Of course, other consequences of this, of

19   the smaller delta, would be larger and longer studies

20   which may impact drug development, as of course we will

21   be hearing more about today.

22                  And as a final slide here, as an absolute,

23   delta must be smaller than the conservative estimate of

24   the advantage of the active comparator over placebo,a

25   nd the challenge here is that we really do not have very

                               S A G CORP.
     202/797-2525              Washington, D.C.       Fax: 202/797-2525

1    good historic data to know what that advantage is.

2                   And so we really need your advice about how

3    to handle that, and then using clinical judgment, we may

4    want to increase delta further to rule out important loss

5    in efficacy.

6                   And again we need your advice in determining

7    what is an important loss in efficacy.

8    And finally that superiority designs can play an important

9    role in some settings.        Thank you.

10                  CHAIRMAN      RELLER:          Thank       you,      Dr.

11   Brittain,and Dr. Lin, and to the other speakers this

12   morning for their insightful presentations.                Are there

13   any questions from the committee specifically on the

14   material presented thus far?           Yes?

15                  DR. FINK:     I guess my question is that in

16   terms of the issue of bio-creep, which I think is an

17   important one, could a propagation of errors analysis

18   be applied to this data if one could define an initial

19   gold standard?

20                  Propagation of error analysis is commonly

21   used in more defined settings, such as manufacturing or

22   in physical chemistry, but it doesn't seem like it would

23   be impossible to apply it potentially to biologic systems.

24                  CHAIRMAN RELLER:       Thank you, Dr. Fink.         Drs.

25   Lin or Brittain?     Dr. Albrecht, any comment?

                               S A G CORP.
     202/797-2525              Washington, D.C.            Fax: 202/797-2525

1                   DR. ALBRECHT:     In reviewing and approving

2    of new drug products, we don't actually have gold

3    standards that would apply in this case.

4                   CHAIRMAN RELLER:      Dr. Temple.

5                   DR. TEMPLE:     In a lot of situations, what

6    you are looking at is hazards ratios where you are very

7    worried that you don't know what the actual rate of the

8    untreated condition would be.

9                   It seems to me, but I don't really know the

10   field very well, that in antibiotic treatment that you

11   might set a minimum response rate that would apply to

12   whether you count the study at all.

13                  If you were dealing with urinary tract

14   infections, for example, and you had a 60 percent response

15   rate, you might say, oh, well, that is not typical, and

16   you would throw it out, and it just would be a null study,

17   and you would insist that it be 80 or 85, or whatever

18   you are familiar with.

19                  That might prevent bio-creep to a degree.

20   I don't know how that relates to propagation of errors.


22                  CHAIRMAN RELLER:      Dr. Bennett.

23                  DR. BENNETT:     I wonder if I could ask Dr.

24   Brittain to clarify something, and that is the early

25   escape with three arms that she alluded to.        One arm would

                            S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    be the control drug, the active control, and the other

2    new drug, and do I assume the third arm would be a placebo,

3    because if you have got an early escape clause, you

4    wouldn't want to then go to placebo would you?

5                   DR. BRITTAIN:       This is the early escape

6    placebo design, and what I was saying in the two arm study

7    is that it is the new drug versus placebo.           The three

8    arm version of that would be new placebo and an active

9    control.

10                  And the idea being again that after maybe

11   two days after base line, patients are determined to see

12   whether they have improved or not.         And if they are not

13   improved, they would be put on other therapy.

14                  In other words, no one could stay on a drug

15   that wasn't working for them for more than two days.

16                  DR. TEMPLE:    You have to introduce a time

17   element into those kinds of studies.           It isn't total

18   response rate, because everybody is going to respond

19   before you are done.    It is how many responded three days

20   or five days, or whatever, or time to response, or

21   something like that.

22                  CHAIRMAN RELLER:      Dr. Leggett.

23                  DR. LEGGETT:    Just a historical question and

24   a couple of things.     Have we actually seen evidence of

25   bio-creep, and have we -- and by we I mean you or the

                            S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    society, or the Europeans, or the Japanese, have we

2    actually seen cases where the step function has resulted

3    in retrospective analysis of saying, oh, I wish we hadn't

4    done that, or is this all still hypothetical/theoretical?

5                    CHAIRMAN RELLER:            Dr. Brittain and Dr.

6    Goldberger.

7                    DR. BRITTAIN:        I just want to add one

8    comment.       I think the worst case of bio-creep is when

9    you can't see it, when you don't know that it is there,

10   and that is the most insidious form of it.

11                   CHAIRMAN RELLER:            In listening to this

12   morning's      presentation,   the     language    is   remarkably

13   similar to some of the dilemmas faced in the practice

14   of evidence-based medicine, evidence based on regulatory

15   process.

16                   And the best available evidence, which may

17   not be ideal, and then plus experience, and then after

18   the break we will hear the experience component from

19   industry, and infectious disease practitioners, to blend

20   these together to try to come to a full and complete

21   discussion with all perspectives presented.

22                   So that the agency and other interested

23   groups over time can come to a reasonable approach, though

24   not necessarily a perfect one, with a continuing evolution

25   of the evidence on which these decisions can be based.

                              S A G CORP.
     202/797-2525             Washington, D.C.           Fax: 202/797-2525

1                   Dr. Soreth, you had a comment before we take

2    our 15 minute break?

3                   DR. SORETH:   To answer further Dr. Leggett's

4    question about whether or not we have evidence, hard

5    evidence of bio-creek.       I think there is one approval

6    that we took a number of years ago that illustrates this.


8                   It was a drug, Monul, used as a single dose

9    for the treatment of cystitis in women, and there were

10   three trials submitted in that package.            Two, which

11   compared the use of that drug, with 7 days of ciprofloxacin

12   and 10 days of bactrim, in which the drug proved itself

13   to be inferior to those treatments.

14                  And a third trial in which Macrodatin or

15   Nitrofurantoin was chosen as the comparator, and which

16   equivalence was shown.        The product label gives the

17   results of those clinical trials, and so hopefully one,

18   a prescriber would understand where it fits in the

19   spectrum of treatments for urinary tract infections.

20                  But I think that could be -- that is an

21   illustration of having a drug on the market that is

22   inferior to other treatments, and equivalent to another.


24                  CHAIRMAN RELLER:     We will reconvene at 9:50.

25    Thank you.

                            S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1                     (Whereupon, at 9:40 a.m., a recess was taken

2    and the meeting was resumed at 10:02 a.m.)

3                     CHAIRMAN RELLER:       We will begin the second

4    half of this morning's presentations with a presentation

5    on the Medical Perspective: Bacterial Meningitis, by Dr.

6    George McCracken.

7                     DR.    MCCRACKEN:        Dr.   Reller,    Committee

8    Members,       Ladies    and   Gentlemen,       the   title    of    my

9    presentation is evaluation of antibiotic treatment of

10   bacterial meningitis, an increasing challenge.

11                    At the outset, I want to repeat that what

12   was made, the comment that was made originally at the

13   outset of the meeting that the reason for presentation

14   -- and you can see that I am going to touch briefly on

15   fluoroquinolone, and there is a protocol in front of the

16   FDA for gatifloxacin therapy in meningitis.

17                    I hope to be the principal investigator if

18   it is approved, and thus have potential or conflict of

19   interest with regard to that, and I am an advisor to

20   Bristol-Meyers Squibb, and several other companies that

21   were mentioned to help develop drugs.

22                    I would take some issue with the comment that

23   I speak for companies.          I speak for no company.             The

24   companies provide money to institutions where I speak,

25   but there is a difference in how that is said.

                                S A G CORP.
     202/797-2525               Washington, D.C.           Fax: 202/797-2525

1                   So   fluoroquinolones         are     coming       to

2    pediatrics, whether we like it or not, and I have some

3    reservations, but for some conditions it is critical,

4    and bacterial meningitis is one of those.

5                   So why fluoroquinolone therapy for bacterial

6    meningitis?    Well, increasing resistance of pneumococci

7    is a problem worldwide and these drugs are active, at

8    least the newer generation compounds are.

9                   They have expanded coverage against many of

10   the meningeal pathogens, including coliforms, and it can

11   be used in a simplified regimen of a step-down from IV

12   to oral in some settings, in which this would be feasible.

13                  And it certainly penetrates well and has

14   superior or at least comparable bactericidal activity

15   in spinal fluid.     Next slide.

16                  Now, how do we study a drug for bacterial

17   meningitis?     The first step is in a rabbit model of

18   meningitis, which has been used for many, many years,

19   for more than 25 years, and we are able to apply the

20   pharmacogenetic     and      pharmacodynamic       principle      of

21   relevance, which for the fluoroquinolones is area under

22   the curve, and over the MBC, and not MIC, but MBC.

23                  We want cidal activity, and we apply this

24   to spinal fluid, and we adjust the regimen in order to

25   achieve a dosage that has concentrations in plasma or

                               S A G CORP.
     202/797-2525              Washington, D.C.         Fax: 202/797-2525

1    serum that are comparable to those in adults, and the

2    actual amount given to the animal is irrelevant to what

3    we use in humans.

4                   It is only to achieve that concentration,

5    and then we think the regimen in order to achieve the

6    AUC over MBC, and that would be optimal.       Now, we can

7    pretty much predict what that would be when you look at

8    dosing intervals, and half-life those, and then we can

9    predict from that what the dosage will be in humans, in

10   infants and children.

11                  So I am going to show you now the next step

12   in which we looked at one drug, which was trovafloxacin

13   just recently published in the January of the Pediatric

14   and Infectious Disease Journal, in which we evaluated

15   trovafloxacin, and compared to the comparator, which was

16   ceftriaxone, with or without vancomycin.

17                  The dosages was exactly what was predicted

18   from the animal models.     Now, we had chosen a 20 percent

19   difference in proportions as the end-point which we were

20   achieving in clinical results.

21                  It was a multi-center trial of 30 centers,

22   in 11 different nations, and it could not be performed

23   in the United States because we don't see enough cases

24   of the disease.

25                  And we had desired to have 284 evaluable

                            S A G CORP.
     202/797-2525            Washington, D.C.     Fax: 202/797-2525

1    patients.       We enrolled 311 patients, and the study was

2    stopped because of the concern for liver toxicity in

3    adults, but it was not observed in infants and children.


5                     But because of that concern, we stopped the

6    study at 311, and 65 percent of the patients were

7    evaluable, which gave a total of 203 at the time of the

8    end of therapy, 203 patients, which was underpowered then

9    for even a 20 percent difference in proportions.

10                    However, there is important lessons to learn

11   from this study that apply directly to any consideration

12   of   a   drug    in   the   future.        Here   are   some    of    the

13   demographics.

14                    The age is comparable by 2-1/2 years, and

15   that is about reasonable for infants and children.

16   Symptoms.       The number of days to enrollment, 3.1 and 3.2,

17   is long, because the standard deviation, you can see,

18   is broad.

19                    And there were at least three institutions

20   in the study from other countries in which the delay in

21   diagnosis was 4 to 6 days, and the outcome in that group

22   was clearly inferior, and that is a problem when you go

23   outside the country, that the duration of illness is often

24   longer.

25                    Approximately 40 to 50 percent of patients

                                 S A G CORP.
     202/797-2525                Washington, D.C.            Fax: 202/797-2525

1    received prior antibiotic therapy, and by definition they

2    could receive no more than one dose.          But let me remind

3    you that one dose intermuscularly of ceftriaxone will

4    sterilize the spinal fluid of meningococcus disease in

5    many of the patients.

6                   And in those that it does not sterilize it,

7    or any drug, we know that it drops the log concentration

8    of bacterium CSF, a study that we did in the '70s, Bill

9    Feldman and others, that showed clearly a two log drop,

10   even with oral ampisone, with a number of the different

11   agents.

12                  So if you drop the log concentration, a drug

13   is going to look easier because you are dealing with many

14   10 to the 4, or 10 to the 5, on admission with the study

15   drug; compared to 10 to the 7, which is the average

16   concentration in spinal fluid of bacterium.

17                  Looking at etiologic agents, it is reasonably

18   distributed, but let me remind you that we really want

19   to see Strep pneumoniae.         That is the most difficult to

20   treat, and it is the one that is resistant, and we see

21   that it is not always easy to get, and it is not going

22   to get easier.

23                  Meningococcus is nice to have, but anything

24   works for that disease, and so it doesn't tell you much.

25    If   a    single   dose   of    a   sulfonamide   works     for       a

                                S A G CORP.
     202/797-2525               Washington, D.C.        Fax: 202/797-2525

1    bacteriologic cure, I am not going to be too interested

2    in whether a comparator works to an experimental drug,

3    because they all work for that.

4    So it is a very important consideration.

5                     Now, here is the clinical and microbiologic

6    end-points.        Now,   remember        we    chose    a   20   percent

7    difference in proportion, and by the FDA standard of 10

8    percent, the trovafloxacin would have looked inferior.


10                    Now, there are two mistakes here.                     This

11   should be minus 2.9 percent, and this is minus 4.8 percent.

12    So    they      are   all     minus      here,    tilting        against

13   trovafloxacin.

14                    You can see the 95 percent confidence limits

15   do not exceed the 20 percent, but clearly the 10 percent

16   it does.       So, does this mean that trovafloxacin in this

17   particular study was inferior to the comparator, which

18   was ceftriaxone, with or without vancomycin?

19                    I don't think so, and let me explain why.

20   First of all, look at bacteriologic success, and I ask

21   you a simple question.         What is the purpose of antibiotic

22   therapy for bacterial meningitis?                  To eradicate the

23   bacteria.       It does nothing else.

24                    So, bacteriologic eradication, 98 percent,

25   minus than 1 percent, very tight bounds.                There were eight

                                  S A G CORP.
     202/797-2525                 Washington, D.C.             Fax: 202/797-2525

1    patients who had a delay in bacteriologic eradication.

2     And 6 of those 8 had poor outcome, totally expected.

3                     Now, let's look at the ITT analysis.                    The

4    last was for protocol.                 And here we encounter some

5    problems.       You can see here at the end of the therapy

6    there was clearly a big difference.                  Now, why is that?

7                     Well,     if    you    look   at    the    designation,

8    clinical success, and then come down and say 13 patients

9    were considered clinical failures.                   Those 13 patients

10   were in two centers in one country outside the United

11   States.

12                    And 11 were in the trovafloxacin arm, and

13   two were in the ceftriaxone arm.           And nine had haemophilus

14   meningitis.       All 13 had immediate sterilization of their

15   spinal fluid.

16                    And 11 of the 13 had follow-up at 5 to 7 weeks,

17   and at 6 months, were considered normal.                    And yet they

18   were called clinical failures, which we had to designate.

19    And that is because the investigator had a concept of

20   what was expected.

21                    It wasn't correct.        Subdural effusions were

22   called         failures,        and     subdural       effusions         are

23   part-and-parcel       of    meningitis         and   portend      no   poor

24   prognosis, and have no bearing on prognosis.

25                    So it must be very -- when you go outside

                                 S A G CORP.
     202/797-2525                Washington, D.C.               Fax: 202/797-2525

1    the   country    to    do    these     studies,   it   becomes     very

2    difficult.        We        had   an     oversight     committee      of

3    non-investigators in the study.

4                    We chose not to act on this because the drug

5    was not going to be used again anyways, and so we decided

6    to show all the data, and not eliminate those patients,

7    but it represents an important point to consider.

8                    This one shows the adverse event profile,

9    and the only significant difference was in abdominal pain,

10   and more common in trova.              I would point to the joint

11   abnormalities which we followed.

12                   This is at 5 to 7 weeks, but even following

13   out to six months, there is no difference.               In fact, it

14   was a little higher in the ceftriaxone group.             Next slide.

15                   There are many restrictions on performing

16   studies of antibiotic therapy for meningitis, and the

17   first and most important in the United States, and in

18   any developed country, is the development of the conjugate

19   vaccines.

20                   They have been a blessing.             We don't see

21   haemophilus disease in the United States.                I have seen

22   on meningitis as of Memorial Day, 1999.                 That was the

23   last case.

24                   Now we have pneumococcal vaccine, a conjugate

25   vaccine, and it has been in the United States for two

                                  S A G CORP.
     202/797-2525                 Washington, D.C.          Fax: 202/797-2525

1    years, almost two years now.              With the implementation

2    of these vaccines throughout the world with time, we will

3    virtually eliminate haemophilus, which we have where it

4    is used.

5                   And   certainly       it       will    reduce,     if    not

6    eliminate, pneumococcal.           Probably not eliminate.               At

7    least 50 percent of the patients are pre-treated, and

8    I told you what the issue is there.                       It drops the

9    concentration or will sterilize if ceftriaxone is the

10   drug administered.

11                  The necessity to have large numbers as

12   required by the FDA for a 10 percent difference in

13   proportion is simply not possible.                   A requirement for

14   a   clinical   end-point,       rather        than    a   bacteriologic

15   end-point,     I   think   is    not      reasonable       any    longer,

16   particularly when you understand what the effect of

17   antibiotics are in bacterial meningitis.

18                  And of course we know the logistical problems

19   performing studies anywhere, but most especially outside

20   the United States.     However, it is necessary to have study

21   centers outside of the United States, outside of North

22   America.

23                  But to have those, we must do them, we must

24   enroll them, we must conduct the study in the following

25   ways.    This is my opinion, and I feel very strongly about

                                S A G CORP.
     202/797-2525               Washington, D.C.               Fax: 202/797-2525

1    it.    It must be FDA approved obviously.

2                   We must have participation of U.S. centers,

3    and most especially the principal investigator.            They

4    must have his center or her center involved.    IRB approval

5    in all centers.

6                   Informed consent for every patient.           And

7    there must be a preliminary investigators meetings.

8    Everyone there to go over word by word the protocol for

9    approval.

10                  Now, the next two slides we can skip because

11   they were covered beautifully before me, and probably

12   much more authoritatively.       Let's just go to the sample

13   size estimates.

14                  Now, we are talking about an 80 percent

15   response rate, but let me remind you as we move outside

16   the country, and we go to developing nations for these

17   studies, 80 percent is not going to be the end point.

18                  I just reviewed a study from Malawi, 582

19   patients with meningitis, and 40 percent response rate.

20    Now, that is because of underlying conditions obviously,

21   and this becomes a very important point, malnutrition,

22   HIV, other conditions, have impact on the outcome.

23                  So 80 percent is really a little high now,

24   and I am going to use multi-center trials.       And we knew

25   that from the Trova study.     Nevertheless, let's just take

                            S A G CORP.
     202/797-2525            Washington, D.C.       Fax: 202/797-2525

1    80 percent.

2                   And we know that the evaluation rate is

3    actually 65 percent, and may even go lower than that

4    because of prior treatment.        It is become very common.

5     So if you use 80 percent, 10 percent difference in

6    proportions is over a thousand patients.

7                   If it is 15 percent across the board, then

8    65 percent evaluation, and it would be 462.           If it is

9    20 percent, 262.    So it shows you the range.      I can tell

10   you in a simple word that there will never be a meningitis

11   study where 500 or more patients need to be enrolled.

12   It is simply not possible.        Next slide.

13                  There is one paper looking at equivalence

14   and randomized control trials of therapy for bacterial

15   meningitis.    It has not been published, but will be in

16   our journal, the Pediatric Infectious Disease Journal

17   sometime this hear by Kryson and Kemper, from the

18   University of Michigan.

19                  They looked at 25 trials since 1980, and all

20   of these trials claimed equivalence among control and

21   investigational drugs.      Only two studies were designed

22   to test true equivalence.

23                  And 24 had sufficient sample size to exclude

24   a 20 percent difference in case fatality rate, and three

25   trials could exclude a 10 percent difference.           Proving

                            S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    therapeutic equivalence will be a challenge.                    Next slide.

2                      So the potential problems with enrolling

3    centers        from   outside    the     United      States,     mainly     in

4    developing nations where these conjugate vaccines will

5    not have been instituted yet.

6                      And     even     in        some   that    were     in    the

7    Trovafloxacin study that were large contributors to the

8    trial are now using the conjugate vaccines.                   The problems

9    include non-adherence to the protocol, and monitoring

10   issues, and severity of illness.

11                     And let me remind you that at least a third,

12   if not more, will have underlying conditions in children,

13   which will impact outcome.

14                     Performing       appropriate           audiometric       and

15   psychometric          evaluations, complete follow-up is often

16   difficult.        There is no system, and no infrastructure

17   to be able to do that.

18                     There     will        be     larger      percentages      of

19   meningococcus haemophilus cases, and lower pneumococcal,

20   and of course storage of specimens.                     So let me again go

21   back to what I think is the essential point here.                           An

22   antibiotic has only one effect; to eradicate bacteria

23   from the CSF, and we can very objectively measure that.

24                     And we have found in the multiple studies

25   that we have done that they follow the prediction from

                                  S A G CORP.
     202/797-2525                 Washington, D.C.                Fax: 202/797-2525

1    the animal models beautifully.             Next slide.

2                   This just shows a further breakdown from the

3    trovafloxacin study that I showed you.             So that in 18

4    to 36 hours, this was the difference trova versus

5    ceftriaxone.     Very close.     This should be a minus 1.5

6    percent.

7                   The bounds are very tight, and at 72 hours,

8    even closer, very tight bounds.              So this was a very

9    objective end-point, and I think should be considered

10   the primary endpoint in bacterial meningitis.

11                  It is not to say that there shouldn't be a

12   clinical harm to that as well.      Now, I made a point earlier

13   that the eight children in the trova study who had delayed

14   sterilization, 6 of those 8 had poor outcome, death or

15   severe sequelae.

16                  We knew that and it is based on many studies,

17   and this summarizes many of those studies, and shows that

18   the positive or rather negative bacteriologic cure or

19   positive culture at 18 to 48 hours and is on average is

20   8 percent, with a range of 2 to 23 percent, depending

21   on the antibiotic.

22                  And in a study that we looked at here, we

23   looked at four control trials in Dallas.             We had a 6.7

24   percent positive culture at 18 to 48 hours.              These are

25   all significantly different.

                             S A G CORP.
     202/797-2525            Washington, D.C.            Fax: 202/797-2525

1                    A higher rate of neurologic abnormalities

2    at discharge, 45 versus 19 percent, and 45 percent in

3    those with delayed sterilization; and at follow-up, 41

4    versus 13 percent.

5                    So a very big difference, and so one of the

6    determinants     of     clinical    outcome    is   bacteriologic

7    response.      So, in summary, the critical end-points for

8    assessing bacterial meningitis, and the antibiotics for

9    bacterial meningitis, are the following.

10                   One, bacteriologic eradication at 18 to 30

11   hours.     It validates the data in animal studies.           Again,

12   in my estimation, this should be the primary end point.

13    We obviously must study tolerance and safety, and

14   clinical outcomes should be evaluated at 6 weeks and 6

15   months.

16                   The end of therapy is not very important,

17   and 6 weeks and 6 months is by far the better end point.

18    However, let me again point out that clinical outcome

19   is very subjective.         There are many variables, many

20   variables that determine clinical outcome that have no

21   bearing on which antibiotic was used.

22                   These   include     duration   of   illness,      and

23   etiology, severity of illness at the time of admission,

24   fluid and electrolyte balance, availability of intensive

25   care management, underlying conditions, just to mention

                               S A G CORP.
     202/797-2525              Washington, D.C.          Fax: 202/797-2525

1    a few.

2                        They are all independent of the antibiotic

3    given.     However, the one determinant that is objective

4    and does influence outcome is eradication of the pathogen.


6                        My suggestion is to enroll approximately 300

7    patients to distinguish a 20 percent difference in

8    proportion, and this is currently achievable using many

9    centers outside the United States.

10                       It will also provide enough patients to

11   determine           tolerance      and     safety,      and      of     course

12   bacteriologic success.                 A 10 percent difference in

13   proportions currently, and in the future, is not feasible.


15                       It cannot be accomplished in the type of

16   setting        in    which   we    now     have    to   study      bacterial

17   meningitis, because of the availability of conjugate

18   vaccines and other factors that I have mentioned.                       Thanks

19   very much for your attention.

20                       CHAIRMAN RELLER:         Thank you, Dr. McCracken.

21    At the end of the presentations, and this afternoon,

22   we will have ample time for questions and a thorough

23   discussion of all of the issues presented.

24                       Our next speaker is Dr. David Shlaes, who

25   will give the industry presentation for PhRMA.                              Dr.

                                     S A G CORP.
     202/797-2525                    Washington, D.C.              Fax: 202/797-2525

1    Shlaes.

2                      DR. SHLAES:     Hi, and thank you very much.

3    My name is David Shlaes, and I am presenting PhRMA today.

4     Just a little bit about me.          I spent 16 years in academic

5    medicine, working mainly on antimicrobial resistance,

6    but also treating a fair number of patients in a Veterans

7    Administration Medical Center with infectious diseases.

8                      So today I am representing the Antimicrobial

9    Working        Group   of   the   Pharmaceutical    Research      and

10   Manufacturers of America.           Next slide, please.

11                     This group offers a forum for exchange of

12   scientific information among PhRMA companies, and our

13   deep commitment to anti-infective drug products.                   It

14   provides industry's scientific perspective in response

15   to proposed rules, draft guidances, and relevant issues

16   affecting anti-infective drug products.            Next slide.

17                     In our working group, there have been a large

18   number of companies involved.           We have had prior meetings

19   with the FDA and a number of teleconferences and other

20   meetings within our Antimicrobial Working Group.                Next

21   slide.

22                     Today I want to cover three topics, and just

23   a little background on the antibacterial clinical trials

24   and the selection of delta.            Implications of the delta

25   in antimicrobial development, including a number of

                                 S A G CORP.
     202/797-2525                Washington, D.C.        Fax: 202/797-2525

1    unintended consequences I think, some of which have

2    already been discussed.

3                     And then I would like to present a number

4    of alternative proposals that one could consider going

5    forward.       Next slide.

6                     So the key or bottom line messages that I

7    will try and support during the talk are what in our view

8    is the current system for designing clinical studies and

9    registering antibacterial drugs has worked well.

10                    In fact, we recognize that there is always

11   room for improvement here, but in our view this system

12   has worked well, and a lot of the considerations that

13   you are hearing about today are mainly theoretical ones.


15                    What you are also hearing is that a single

16   approach       for   all     antibacterial      drugs,    for     all

17   indications, is unlikely to be an optimal one because

18   of the differences in patient populations, variability

19   from one patient population to another, and even within

20   the population that you are studying.

21                    Clinical studies must be feasible as you just

22   heard from Dr. McCracken.               The sample sizes must be

23   practical.      We have to be able to get these studies done

24   in some reasonable period of time for a variety of reasons.

25                    And also we need to be able to do studies

                                S A G CORP.
     202/797-2525                 Washington, D.C.       Fax: 202/797-2525

1    that direct our attention to areas of public health need,

2    something that we will talk about more tomorrow.             Now,

3    one of the major ways that we can address the worry about

4    bio-creep is in choice of comparator.

5                   And I would say in the example that Dr. Soreth

6    cited that this may have been just a problem of choice

7    of comparator and poor study design, rather than actual

8    bio-creep related to statistical concerns around the

9    delta.     So PhRMA's proposals are offered in this context.

10    Next slide.

11                  Now, there are a few differences comparing

12   anti-infective drugs with drugs in a lot of other

13   therapeutic areas.       First of all, in the case of

14   anti-infectives, we can get considerable information

15   about activity against targeted pathogens from our in

16   vitro      testing,   from    animal       models,   and     from

17   pharmacokinetics and pharmacodynamics.

18                  And this is something that is not shared by

19   many other therapeutic areas.          We do carry out trials

20   with rigorous design, usually using an active control.


22                  And it is important to keep in mind that the

23   magnitude of efficacy observed in a given study as you

24   have already heard varies with the severity of the

25   pathogen, or of the infection rather, the specific

                             S A G CORP.
     202/797-2525            Washington, D.C.         Fax: 202/797-2525

1    pathogens that are involved, and a variety of other

2    conditions.

3                   And therefore within any given population

4    there is going to be a certain variability.        Next slide.


6                   Now, the approach of the FDA throughout the

7    '90s as you have heard is the following.            Regulatory

8    approval has been based on evidence from multiple clinical

9    studies, typically from multiple indications.          So in most

10   cases, there are two well controlled clinical trials for

11   each indication.

12                  The evidence must show that the success rate

13   of the new drug is reasonably close to the success rate

14   of an active control statistically; that is, that the

15   new drug is not inferior to the control drug by more than

16   a predetermined amount.

17                  And that is the delta essentially, and the

18   main assessment is to compare the lower bound of a

19   two-sided      95   percent   confidence    interval     on    the

20   difference in success rates for the new drug, versus the

21   active control, to a pre-specified limit, or the delta.

22    And this was explained actually by Dr. Temple.               Next

23   slide.

24                  This just shows again the step functions to

25   remind you as explained in the FDA's 1992 points to

                              S A G CORP.
     202/797-2525             Washington, D.C.       Fax: 202/797-2525

1    consider, which we think is still a very reasonable way

2    to approach clinical trial design actually, where we have

3    a sliding scale of delta, with a cure rate.

4                   This does allow for reasonable trial sizes,

5    varying with severity of infection and cure rate.                    Next

6    slide.

7                   One of the major merits of the step function

8    is that it recognizes that one size does not fit all.

9    So that there is a smaller margin when comparative success

10   rates are higher, and therefore a higher hurdle for new

11   treatments, compared with very effective controls.

12                  The step function recognizes the magnitude

13   and    variability    of   the     success     rate     to    establish

14   non-inferiority criteria.           It recognizes the need for

15   both    statistical   and     clinical        aspects    of    efficacy

16   evaluation.

17                  It supports study design using realistically

18   achievable sample sizes, which I think as you have heard

19   is a clearly important consideration.

20                  And the approach in fact has been used

21   effective for a decade of drug development, and we as

22   you heard earlier, I don't think anybody is aware of any

23   evidence that newer agents approved to treat serious

24   infections,     especially        those       involving       resistant

25   pathogens, are less effective than previously approved

                                S A G CORP.
     202/797-2525               Washington, D.C.             Fax: 202/797-2525

1    products.

2                   This is just a list of some effective products

3    that have been developed and approved sine the early 1990s

4    using this approach, and again I don't think there is

5    evidence that this approach results in the approval of

6    inferior products.     Next slide.

7                   Now, there are some implications of a smaller

8    delta, and I would like to go through a few of those.

9    Clearly, there is an increased time to drug availability.


11                  So that if you carry out a trial, for example,

12   in the example that Dr. McCracken mentioned, where if

13   you carried out a meningitis trial for a 10 percent delta,

14   even at an 80 percent power, that trial might last for

15   5 or 6 years, if you could do it at all.

16                  And the question is would the comparator that

17   you chose at the start of that trial be relevant at the

18   end of 5 or 6 years.      Is that relevant?    So there was

19   a question about the validity of a trial being carried

20   over a number of years, and this adds further to the

21   inherent variability in a given infectious disease

22   indication.

23                  And the other problem is the increased number

24   of investigators that are required, which gives another

25   source of variability.      So basically what you get is a

                             S A G CORP.
     202/797-2525            Washington, D.C.      Fax: 202/797-2525

1    smaller delta, larger sample size, increased development,

2    time, costs, and variability.

3                   And   as   Dr.     McCracken    also   mentioned,

4    frequently increased numbers of investigators outside

5    the United States, because you simply cannot gather or

6    enroll the number of patients that you need to enroll

7    for many of these trials within the United States alone.

8     Next slide.

9                   And I won't go over this because Dr. McCracken

10   covered this in great detail.           Next slide, please.

11                  So what do you gain by reducing the delta?

12    If you have a control cure rate of 85 percent, and a

13   new cure rate of 75 percent, you run a 90 percent powered

14   study with 120 available patients per group; and two

15   trials, powered at 50 percent delta; and the risk of

16   incorrectly concluding non-inferiority is 2.7 percent.

17                  Therefore, I think in this design there is

18   very little risk of approving non-inferior products.

19   So I am not sure how much advantage you get by reducing

20   that delta to 10 percent.

21                  The other thing that I will point out is that

22   a lot of the examples that have been shown today assumed

23   an 80 percent beta power trial.

24                  If you run an 80 percent beta power trial,

25   at a 10 percent delta, your chance of falsely concluding

                               S A G CORP.
     202/797-2525              Washington, D.C.        Fax: 202/797-2525

1    inferiority is about 30 percent, and most of us in the

2    PhRMA group wouldn't run such a trial.                 Next slide.

3                      So disadvantages will require considerably

4    larger     sample       sizes.     It   is      unrealistic    for     some

5    indications in patient populations, and there is a

6    disincentive        therefore     to    develop     new    antibiotics,

7    particularly for indications with inherently low success

8    rates.

9                      You just heard about meningitis, but that

10   is not the only one.         There are a variety of others, where

11   you have seen very few clinical trials in the last decade.


13                     Endocarditis, osteomyelitis, and those are

14   neglected areas because of already statistical design

15   requirements.           The other problem is that by increasing

16   the trial size, you could potentially unnecessarily

17   expose patients to investigational treatments for longer

18   than what might be otherwise required.                 Next.

19                     An increased cost and time will further

20   disadvantage investment in new antibiotics and company's

21   portfolios relative to other therapeutic areas.                     We are

22   already        seeing    this,   and    fewer     companies      will    be

23   developing new antibiotics.

24                     Because of this, there is a risk that existing

25   drugs will continue to be used in lieu of a constant

                                  S A G CORP.
     202/797-2525                 Washington, D.C.             Fax: 202/797-2525

1    pipeline of new drugs, and even if there is an invest

2    so that we get new drugs that delay an availability, we

3    will continue to put pressure on the existing drugs just

4    because        of    the    increased     stringency       of   the    trial

5    requirements.

6                        And obviously the fewer new anti-infectives

7    will be exacerbated by the current trend in industry

8    towards         dis-investment            in       anti-infective         R&D

9    infrastructure.

10                       And    this     all   leads      to    public     health

11   considerations, which I think we have to keep in mind.

12   And we must have an ability to respond to these public

13   health conditions going forward.                   Next.

14                       Just to point out that anti-bacterial drugs

15   are already disadvantaged in the R&D portfolios of the

16   pharmaceutical industry.              The reason for that is that

17   the antibacterial drugs are usually intended for short

18   duration        of    use     for    acute        diseases,     unlike     an

19   anti-depressant, which you take for a very long time;

20   and an antihypertensive, which you take forever, et

21   cetera.

22                       The size of patient population is relatively

23   unpredictable and can vary dramatically from year to year,

24   depending on the indication.                And as I pointed out, an

25   economic justification within companies is stronger for

                                    S A G CORP.
     202/797-2525                   Washington, D.C.             Fax: 202/797-2525

1    the development of drugs in other therapeutic areas.

2                   So this therapeutic area is a therapeutic

3    area within the industry that always sits on the brink.

4     It is always on the brink, and it doesn't take much to

5    push it over the edge.       Next.

6                   So what PhRMA would like to suggest is a

7    number of alternatives.       One is to continue to use the

8    step function approach until an optimal alternative is

9    agreed upon, and we think this basically works.

10                  As I pointed out, the comparator agent should

11   be a consensus standard of care and this should thereby

12   address concerns about bio-creep in our view.                 And for

13   indication specific deltas, a consideration of the

14   seriousness of the disease, the variability of the

15   response rate, and the feasibility of conducting the

16   trials, must be undertaken for each indication.                   Next

17   slide.

18                  There   are   several       options.       One    could

19   conduct two independent Phase III trials with a delta

20   of 15 or 20 percent for each trial, which essentially

21   is included in the step function as it stands now.

22                  There is a low risk of incorrectly including

23   non-inferiority in this case.              One could conduct two

24   independent Phase III trials, one larger and one smaller,

25   with a combined analysis or Meta-analysis, providing a

                             S A G CORP.
     202/797-2525            Washington, D.C.              Fax: 202/797-2525

1    power of 95 percent, and a combined sample size using

2    a delta of 10 percent to assess non-inferiorities.          So

3    you could achieve an analysis in that way.     Next slide.

4                   One could analyze results of trials by

5    comparing the lower bound of a one-sided 95 percent

6    confidence interval on the difference in success rates

7    for new drugs, instead of using the two-sized confidence

8    interval, and this in fact was suggested in the ICH E9

9    document.

10                  Another approach would be to use the FDA's

11   general equivalence definition for selected indications,

12   and I will show the nosocomial pneumonia one on the next

13   slide.

14                  So this is just to summarize the general

15   equivalence for nosocomial pneumonia, where you would

16   use one well controlled trial and an absolute clinical

17   success rate of new drug no more than 5 percent in absolute

18   terms, less effective than an agreed active comparator

19   agent.

20                  And this requires at least 80 patients in

21   each arm, and clearly well-defined patients, and this

22   sample size in fact, in measure of equivalence, describes

23   an 80 percent power design and a 20 percent delta.

24                  This would be quite feasible, and we believe

25   we could do these trials in nosocomial pneumonia, and

                            S A G CORP.
     202/797-2525            Washington, D.C.     Fax: 202/797-2525

1    they would be valid.     Next slide.

2                   Now, we agree with a lot of the previous

3    speakers, in terms of alternate designs for diseases where

4    there may be placebo effects, such as acute bronchitis,

5    acute exacerbation of chronic bronchitis, acute otitis

6    media.

7                   People have talked about a so-called rapid

8    cure design, where again you could do a 50 patient per

9    arm study, and evaluation at some time point, and we chose

10   day four to five year, but it could be 2 to 3, or whatever

11   the time point is, to show that active treatment provides

12   a two-fold increase in success rate, compared to placebo.

13                  And then a no improvement would be failure,

14   and then failures are treated with open label antibiotics.

15    Also, a time to cure design, where a placebo controlled

16   study is done to demonstrate a 50 percent reduction in

17   time to symptom resolution.

18                  Obviously, this would have to take into

19   account the severity of infection within these specific

20   indications somehow.    But these are approaches to getting

21   placebo designed, placebo controlled, trials, and some

22   indications for not serious infections.

23                  So, in summary, PhRMA recognizes the medical

24   need for discovering development of new antibacterial

25   drugs.     I think nobody more than me.    PhRMA companies'

                            S A G CORP.
     202/797-2525            Washington, D.C.      Fax: 202/797-2525

1    welcome and rely on informative and realistic guidances

2    to provide the latest thinking of FDA and its advisors.

3                     This is terribly important to us because it

4    allows us to know the path forward in the development

5    of new drugs.        We are planning a workshop for industry,

6    FDA, IDSA, and other stakeholders, in order to define

7    clinical       and   statistical    standards      consistent      with

8    efficient development of safe and effective antibacterial

9    drugs.

10                    And we hope that this will be part of the

11   process of coming to consensus on how we can go forward

12   from here.       And I think that is all that I have to say.

13    Thank you very much.

14                    CHAIRMAN RELLER:         Thank you, Dr. Shlaes.

15   Our next speaker with an industry presentation will be

16   Dr. Francis Tally.         At the completion of Dr. Tally's

17   presentation, and before the IDSA presentation, I would

18   like to have questions directed at the first three

19   speakers, if there be any, including Dr. McCracken's

20   presentation for him.         Dr. Tally.

21                    DR. TALLY:    Thank you, Dr. Reller.          I would

22   like to thank the FDA for inviting me to participate in

23   this advisory committee meeting.              What I am going to talk

24   about to day is the biotech approach to this topic.

25                    The difference between big Pharma and biotech

                                S A G CORP.
     202/797-2525               Washington, D.C.            Fax: 202/797-2525

1    is that biotech companies usually focus in one area, and

2    doesn't have the luxury of having several of the areas

3    to support the research structure in the development group

4    involved.

5                   We also have a lower threshold for getting

6    drugs into development, but we need to have a threshold.

7     And we have strong influences to have frequent dialogue

8    with regulatory bodies so we can take the most focused

9    path in achieving a registration of our drugs, because

10   we don't have the luxury of studying eight different

11   indications.

12                  What I would like to do today is give a view

13   from our perspective.      The disclaimer about companies

14   is on every slide, and I am the chief scientific officer

15   of Cubist Pharmaceuticals.

16                  But like David, I had a 15 year history in

17   the academic world, studying a number of different drugs,

18   and like Vince Andriole, was on the committee, the

19   ISDA-FDA Committee, back in the mid-1980s to early-1990s.

20                  I then went into industry and first worked

21   in big pharma, and had the pleasure of registering a large

22   drug for resistant infections with piperacillin or

23   tazobacam, and also doing some discovery.

24                  And for the last seven years, I have been

25   at a small pharmaceutical company or biotech company,

                            S A G CORP.
     202/797-2525            Washington, D.C.     Fax: 202/797-2525

1    and we are currently developing a drug for the treatment

2    of serious Gram-positive infections.

3                   The majority of antibiotics developed over

4    the last several years, or last 40 years, have been broad

5    spectrum drugs, and we have had a number of "me-too" drugs

6    in the same area, which I know has brought up a problem

7    with development.

8                   But now we are looking at different drugs

9    that we have both broad spectrum and narrow spectrum,

10   and it is going more towards the narrow spectrum.                  We

11   also have oral and/or IV, and there are special problems

12   when you have an IV only drug with the practice of medicine

13   in the United States, and now we are seeing the same

14   problem in Western Europe.

15                  And   finally   you     will   see    existing      --

16   modification of existing drugs, but what the big effort

17   now in research is to develop novel classes of drugs with

18   novel targets.

19                  And I will touch on that a little more

20   tomorrow in the resistance discussion.         But I am listing

21   some of the drugs here, and a couple that have been

22   recently approved -- quinopristin, dalfopristin, and

23   linezolid, representing an old class streptogramins, and

24   a new class, the oxazolidinones.

25                  On the other drugs that have been from

                             S A G CORP.
     202/797-2525            Washington, D.C.            Fax: 202/797-2525

1    existing classes, Wyeth and David's shop has tigecucome.

2    amd we have dalbavancin and oritavancin, which are analogs

3    of glycopeptides.

4                   And ertapenem that Merck had approved was

5    the pharmacological advantage of an important class of

6    drugs.     The other new classes we see are daptomycin and

7    telithromcin.

8                   The   details    of    some     of   the    drugs     in

9    development to cover both VRE and MRSA are listed on this

10   slide.     I am not going to go into the details.                It is

11   in the handout.

12                  But what I would like to do is to look at

13   what justifies in 2000 the development of new drugs.

14   First, you have to have microbiological superiority.

15   I think the days of a lot of "me-too" drugs in the same

16   area are over.

17                  And    particularly          with    microbiological

18   superiority is going through resistance, and we will talk

19   a lot more about that tomorrow.               You could look for

20   pharmacological      advantages,     and      clearly     one   a   day

21   carbapenem that Merck just got approved is an improvement

22   in therapy patients.

23                  And so ease of administration, and finally

24   safety advantages are always looked for at different

25   classes of drugs.     There are a number of different drugs

                             S A G CORP.
     202/797-2525             Washington, D.C.             Fax: 202/797-2525

1    around, and the only reason that I put this slide up is

2    there are some cephalosporins coming along with MRSA

3    activity.

4                    And so I think you will be seeing a couple

5    of these drugs come down to see whether or not they can

6    hold out for MRSA, because as you will see tomorrow, one

7    of the main problems we have in the future is at MRSA.


9                    We have heard a lot about protocol design,

10   and I think the drug's characteristics actually dictate

11   in   protocol      design.      Specifically,   spectrum      and

12   distribution of drug is going to dictate what clinical

13   indications you use.

14                   You heard about the PK/PD guides to therapy,

15   and they are just guides, because we need also dosing

16   studies.       And a preclinical safety profile is whether

17   or not you are going to have this drug developed for broad

18   indications and outpatient, or a restricted drug for use

19   in serious infections.

20                   We have heard a lot about superiority and

21   non-inferiority today, and I think superiority trials

22   are very limited in anti-infectives, probably to the

23   out-patient oral drugs that David Shlaes just talked

24   about, and some areas.

25                   But in sick patients in hospitals where you

                              S A G CORP.
     202/797-2525             Washington, D.C.       Fax: 202/797-2525

1    have a known mortality rate, superior trials using placebo

2    are    not     possible.         And     that's   why    we     do    the

3    non-inferiority trials for almost all of the antibiotic

4    trials for serious infections.

5                    And I think there are a lot of data out there

6    in the serious infections where we can look at rates.

7    Finally, in considering these infections, you have to

8    consider whether the infection is a monomicrobial or

9    polymicrobial.

10                   My scientific area was in the study of mixed

11   anaerobic infections, and depending on the type of

12   infection, it presents a number of different challenges

13   on control agents, and covering all of the infecting

14   flora, because if you don't cover all of the infecting

15   flora, you will have a higher failure rate.

16                   And this is particularly true when you are

17   picking the comparative agents to prevent the bio-creep

18   that we have heard about.              And it really dictates the

19   comparative agents.

20                   If you look at the narrow selection rate,

21   such as complicated skin and soft tissue, with Staph

22   aureus, and Group A beta strep, are the main pathogens.

23                   We    have    very     selected   therapy      in    that

24   particular area, depending upon whether you have an MSSA,

25   or    MRSA.     And   so     it is either an amoxicillin or

                                  S A G CORP.
     202/797-2525                 Washington, D.C.           Fax: 202/797-2525

1    vancomycin, and that is what you are limited to.

2                   But   when    you    go       to   community-acquired

3    pneumonia, or nosocomial pneumonia, because of the

4    diversity of pathogens that you see in this disease, you

5    run into a much different problem.

6                   And   when you run into this problem in

7    different countries, you are also running into different

8    types of patients, which we have recently seen.

9                   Indeed, in community-acquired pneumonia, you

10   have    Gram-positives,     and    Gram-negatives,         atypicals,

11   intracellular, cell wall minus, and so there is a whole

12   host of therapies that could complicate your choice of

13   comparative agents.

14                  It is similar in nosocomial pneumonia, but

15   it is much more limited because of the predominance of

16   Staph aureus and Gram negatives, and with the high

17   mortality rate that you see in these groups of patients.


19                  When we are looking at trial design, to prove

20   non-inferiority, you are looking at blinding. Everybody

21   would like the Holy Grail of randomized perspective

22   double-blinded studies.

23                  However, with narrow spectrum drugs, you run

24   into problems in your comparative therapy, and in the

25   companion therapy for the potential pathogens that are

                             S A G CORP.
     202/797-2525              Washington, D.C.             Fax: 202/797-2525

1    not covered by a narrow spectrum drug.

2                    I covered that a couple of years ago in one

3    of the ICAHC meetings.          You can get around some of those

4    by investigative blinding, and it is not quite as good

5    as double-blinding, but still you can come up with

6    dialogue with regulatory authorities to establish a well

7    controlled study.

8                    Open    label       studies     are    reserved      for

9    end-points which are hard microbiological end-points.

10   You   keep     the   microbiologist         blinded,   but    not    the

11   physician.

12                   We have heard a tremendous amount about

13   sample size today of the patients enrolled in your study,

14   and it is driven by delta.            I don't have any numbers in

15   my slides.      I was trusting that everybody in front of

16   me would have beaten that to death, and I am pleased that

17   they have.

18                   We are looking at 95 percent confidence

19   levels, and then project efficacy rates, and we have heard

20   a lot about that.             And finally we are looking at

21   end-points, be it microbiological or clinical.

22                   And we heard from Dr. McCracken about the

23   importance      of     the     microbiological         end-point      in

24   meningitis.      We have also heard about the challenges with

25   when you have a small delta.

                                  S A G CORP.
     202/797-2525                 Washington, D.C.          Fax: 202/797-2525

1                   In challenges of selecting a delta, you can

2    look at is it better than placebo, and that is a

3    superiority trial.       It just requires a monitoring board

4    because if you reach the statistical significance that

5    the drug is working better than the placebo, you should

6    stop the trial.

7                   Like     the   Pharmaceutical     Manufacturer's

8    Association opinions that David just presented, I think

9    the seriousness of the infection affects the delta.

10                  You can look at mild infections, severe

11   infections, or moderate infections, or severe infections,

12   and you want to see that the drug is equal to the standard

13   of care, and this is the concept of bio-creed.

14                  Outside of the people in this audience, you

15   really have to define what bio-creep is, and I think with

16   serious infections that you want to select the best

17   therapy.

18                  I   am   going    to    skip   bio-creep    because

19   everybody knows what it is, and the fear is that we will

20   approve a drug that is no better than placebo, and I think

21   that was nicely presented by the statistical group from

22   the FDA.

23                  And I think that it is important -- and one

24   of the things that has to be developed -- and I would

25   agree with David's recommendation, is that we should try

                               S A G CORP.
     202/797-2525              Washington, D.C.         Fax: 202/797-2525

1    and wipe out the bio-creep that has occurred.

2                   And I know of a couple of other bio-creeps,

3    particularly in impetigo, and cutaneous ulcers, where

4    when you are measuring the effect of drugs, when you give

5    adequate care to these diseases, it is no better than

6    good soap and water, and good nursing care.

7                   And   so it is important to prevent the

8    bio-creep in this particular area.         Once again, I am not

9    going to go into the 1992 recommendations.        That has been

10   beat to death this morning.

11                  I would like though to look at the impact

12   of a small delta as David did, and the number of patients

13   is greatly enlarged, to the point where it drives expenses

14   way up, and for a small farmer, raising all their money

15   on the open market, it puts added pressure.

16                  But that's not a reason for not having a small

17   delta.     The time to complete studies may be in years,

18   and I think this is a major impediment that has been

19   pointed out previously.

20                  One, you are losing investigator interest

21   in the study, and if it stretches out over a couple of

22   years, and you start to get poor patient selection, you

23   may no longer have the appropriate comparator agent.

24                  And when you are finished, you may not have

25   the proper study after all that time.       We have heard about

                             S A G CORP.
     202/797-2525            Washington, D.C.         Fax: 202/797-2525

1    enrollment outside of the United States, and we have

2    recently        experienced    that      in     community-acquired

3    pneumonia by getting a very different patient population

4    in other parts of the world, and as shown from our

5    sub-analysis.

6                     And that is because of the size of the study,

7    we could not hope to enroll all of the patients in the

8    United     States.      And    finally        the   costs    of    drug

9    development.

10                    It is a burden on big farmer and on biotech

11   and specialty firms, but that is something that I think

12   -- my fear at electronic presentations.             And this is Frank

13   Tally's opinion now in collaboration with several of my

14   colleagues at Cubist Pharmaseuticals.

15                    And what would be my opinion on looking at

16   deltas?        I think for oral drugs for common community

17   diseases listed here, such as skin and soft tissue

18   infections, sinusitis and otitis media, bronchitis, UTI,

19   and gonorrhea, this is the area where 10 percent deltas

20   make a lot of sense.

21                    There is big patient populations, easily

22   enrolled, and you can clearly define the character at

23   stake, and it doesn't take years to do the studies, and

24   these studies can be done in the United States.

25                    Indeed, I would even say that in some urinary

                               S A G CORP.
     202/797-2525              Washington, D.C.             Fax: 202/797-2525

1    tract    infection    studies,       and       in   the   treatment      of

2    gonorrhea, where the cure rates are very high, even a

3    delta at 5 percent may be acceptable in these particular

4    areas.

5                   For IV drugs for more serious infections

6    though, I would agree with the recommendation that David

7    just put forth for PhRMA.                When we are looking at

8    different -- I am jumping all around.                  Let me go back.


10                  (Brief Pause.)

11                  One of the other ways to stop bio-creep is

12   when you select a comparative agent, and I think it is

13   important to select the standard of care, and I think

14   there is a lot of guidelines coming from a number of the

15   academic societies.

16                  And I think this is an area that should be

17   worked on to work out the standard of therapy to prevent

18   the bio-creep from going forward.                   With looking at IV

19   drugs for serious infections, what I did was look back

20   at 2 or 3 of the drugs that have just been approved, and

21   looked    at   the   cure    rates     in      nosocomial     pneumonia,

22   hospitalized           community-acquired                     pneumonia,

23   intra-abdominal infections, and complicated skin and soft

24   tissue infections.

25                  And most of them are not in the 90 percent

                                 S A G CORP.
     202/797-2525                Washington, D.C.              Fax: 202/797-2525

1    area.    Most are in the 75 to high 80s, and I think the

2    delta    for    these   should      be    carefully     selected      in

3    consultation with the regulatory bodies based on the

4    clinical knowledge of the disease in the hard end points.

5                    And I think the sliding scale that David

6    talked about that was exposed and published in 1992 still

7    fits, and that there has been very little bio-creep in

8    the IV drugs.

9                    And that's because IV drugs only in the United

10   States present major problems in doing the clinical

11   studies.       And if we put very small deltas on them, we

12   won't be able to achieve enrollment of enough patients

13   to come to the appropriate conclusions.

14                   And   the    patient      population    is    limited,

15   although there are large numbers of patients out there,

16   it is difficult to get them into these studies.                    Here

17   it is imperative that you select the best therapy, because

18   in these infections, there is an attendant mortality that

19   you can affect.

20                   And I think that this is an area where you

21   have to go with the current standard of care based upon

22   the bacteria involved, the resistance rates, and proven

23   efficacy.

24                   We have the further problem with IV drugs

25   of in-hospital use and home IV use, and finally a number

                                 S A G CORP.
     202/797-2525                Washington, D.C.           Fax: 202/797-2525

1    of these patients have switched to oral step-down, and

2    for drugs without an oral component, if you switch them

3    to another drug, it is currently considered a failure.

4                   Whereas, really this has been a switch to

5    oral therapy because of a clinical response.         And I think

6    this is an area which has to be worked on also in the

7    development of drugs going forward.

8                   Finally, we heard from Dr. McCracken about

9    the problems with doing studies for meningitis.              These

10   are hard end-points when we look at meningitis.             People

11   die from this, particularly with strep pneumo.

12                  We have been looking at endocarditis because

13   of the characteristics of our drug, and we have been

14   working closely with the FDA, and I think we have come

15   up with an approach to this, because there has not been

16   an endocarditis approval since the mid-1980s.

17                  And a couple of companies have tried to study

18   this area, but have been unsuccessful.           And this is an

19   area of unmet medical need.       Why?     Because when you look

20   at endocarditis, there has been a change.          Staph auerus

21   is now a major problem with endocarditis, and this is

22   because of our sicker patients in hospital, and the higher

23   incidents of endocarditis in hospitalized patients.

24                  And with mortalities of 24 to 40 percent in

25   Staph aureus and endocarditis, there is a major unmet

                             S A G CORP.
     202/797-2525            Washington, D.C.          Fax: 202/797-2525

1    medical need in this particular area.              And so I think

2    getting the widest delta in order to study this is

3    appropriate, because like meningitis, you have a hard

4    end-point due to bacteremia.

5                    And there are a bunch of other confounding

6    factors that go into this, but the hard end-point of

7    clearing the bacteremia, because if you don't, you have

8    the hard end-point that the patient has failed.

9                    And so in conclusion, I think community-based

10   common infections are where the most bio-creep has

11   occurred.      Therefore, small deltas are appropriate and

12   the best comparative agents should be selected.

13                   For   intravenous       therapy,    and     serious

14   infections, the main problem is the clinical development,

15   and where the physician should select the best therapy.

16                   And in human studies committees, and the FDA,

17   and the physicians themselves, will ensure that you select

18   the best comparative agent.          Thus, I don't think that

19   bio-creep comes in in 2000 and into this particular area.

20                   The delta should be based on the statistical

21   considerations that we heard, and clinical considerations

22   in a comparative therapy should represent that standard

23   of care.

24                   And finally severe infections require the

25   widest deltas, and it is fortunate in those that we have

                              S A G CORP.
     202/797-2525             Washington, D.C.           Fax: 202/797-2525

1    higher microbiological end-points, and the incidence of

2    infection; that is, the patient population to do these

3    studies is very low, and if you put a small delta in this

4    particular area, it will continue to be an unmet medical

5    need.

6                   Finally, I think one of the things that I

7    have been trying to bring about is it really takes a closer

8    interaction between industry and FDA to come up with the

9    appropriate design of the clinical studies for new agents,

10   and I think we will hear more about this tomorrow when

11   we are talking about the evaluation to drugs for resistant

12   organisms.     Thank you.

13                  CHAIRMAN RELLER:        Thank you, Dr. Tally.

14   Questions for the first three speakers in this session?

15    Yes, Dr. Goldberger.

16                  DR. GOLDBERGER:      Given that Dr. McCracken

17   was kind enough to come all the way here for just

18   essentially one day, we would be remiss if we didn't make

19   sure that we got the maximum use from his advice.

20                  I first wanted to ask just a couple of basic

21   questions.     You were talking about important issues on

22   severity of illness, patient's underlying status, et

23   cetera, as being important components of outcome in

24   meningitis,     and   not   impacted,      for   instance,      by

25   antimicrobial therapy.

                            S A G CORP.
     202/797-2525            Washington, D.C.         Fax: 202/797-2525

1                    Is it fair then to conclude from your comments

2    that you don't believe there are drug disease interactions

3    with regards to treatment of bacterial meningitis?               That

4    all of the information basically is simply captured by

5    what happens in the spinal fluid a X-hours?

6                    DR. MCCRACKEN:         Well, it is hard to be a

7    hundred percent about anything when you deal with a

8    complicated disease.         but certain features of patients

9    with meningitis that have clear impact are irrelevant

10   to the antibiotic, and duration of illness, before the

11   doctor ever sees them and they are enrolled.

12                   The severity of the disease at the time of

13   enrollment can be a one hour illness with meningococcemia

14   shock and meningitis, and the antibiotic is -- the only

15   effect it is going to have is on that bacterium.

16                   Underlying HIV and underlying malnutrition,

17   availability of intensive care management, all of these

18   things are really peripheral to the central issue of

19   whether an antibiotic is effective or not.

20                   Now it is not to say that an antibiotic

21   doesn't have interaction, and of course there are people

22   who    are     interested     in    the    possibilities    of    the

23   anti-inflammatory aspects of the drugs, et cetera.

24                   But at this point, I think the clearest and

25   most objective end-point is bacteriologic cure in the

                                 S A G CORP.
     202/797-2525                Washington, D.C.        Fax: 202/797-2525

1    spinal fluid.            And we know that is one of the variables,

2    and probably the only variable, that an antibiotic has

3    clear impact on.             It eradicates that bacterium.

4                        And in fact I feel so strongly about that,

5    that I think you could use a delta 5 percent for that,

6    and if a comparator is inferior, and is less than 5 percent

7    on the 95 percent confidence interval, I don't think that

8    drug should be considered.                    I think it should be very

9    narrow, but the clinical one is much more difficult.

10                       DR. GOLDBERGER:           Well, our concern might be

11   to use an example.                    If you had an infection with

12   haemophilus influenzae in a person with bronchitis, and

13   assuming you felt that the patient needed to be treated,

14   you    might    be        comfortable         with    using    a    macrolide

15   antimicrobial.

16                       If       you      had      established         haemophilus

17   influenzae pneumonia, you might very well want to be

18   looking        at        a    different          class    fluosoquinolone

19   third-generation cephalosporin.

20                       I just wanted to get your feel whether issues

21   like that exist within the area of bacterial meningitis

22   from your perception.

23                       DR. MCCRACKEN:           Yes.     I would not consider

24   the use of a bacteriostatic agent.                         You want cidal

25   activity, and so although the general concept, and

                                        S A G CORP.
     202/797-2525                       Washington, D.C.           Fax: 202/797-2525

1    beautifully illuminated by Bill Craig, is the AUC over

2    MIC for consideration of fluoroquinolones for systemic

3    infection.

4                   I won't accept MIC.        It has to be MBC.          I

5    want cidal activity.      So as that goes in classes of

6    antibiotics, there would be some that I would consider

7    clearly inferior and should not be studied.          Within the

8    classes, it would depend on which the agent is.

9                   But as long as it has two characteristics

10   -- well, three, but two characteristics from a meningitis

11   standpoint.     One, it penetrates well.         It maybe has

12   lipophilic activities, much like the lipophilicity, like

13   the fluroquinolones.

14                  And so it gets into the spinal fluid, and

15   two, it has demonstrated cidal activity; first in the

16   animals and then in the human.      Of course, there are other

17   features; safety and tolerance, and all of those.

18                  But other than those two, which you can

19   clearly demonstrate before you even get to a patient,

20   I don't think the class matters as long as it is cidal.


22                  DR. GOLDBERGER:     I just wanted to make an

23   observation.     You were kind enough to go through some

24   of the trovafloxacin data in some detail, and we are sort

25   of forced to be in the position regrettably of having

                            S A G CORP.
     202/797-2525           Washington, D.C.          Fax: 202/797-2525

1    to be at times skeptical when we look at information.

2                      But   looking at that data, the kind of

3    questions that probably would come up if someone here

4    were reviewing that, for instance, to get that indication

5    for trovafloxacin, were the proportion of the retreated

6    patients in the trovafloxacin arm was noticeably higher.

7                      The proportion of pneumococcal infections

8    in   the       trovafloxacin   arm   was      notably   lower.        You

9    correctly brought up this issue of the early failures,

10   and how it didn't seem as though that was related to

11   microbiology.

12                     Yet, the kind of thing that would always

13   bother us is that there were 13 early failures, and 11

14   were in the trovafloxacin arm, and only two in the

15   comparator.

16                     And as you can imagine, when we look at data,

17   we are forced to just look at that and wonder, well, why

18   did it turn out that way.            And I was wondering if you

19   had any observations about that, and also just to give

20   you our perspective.

21                     And although we agree with you, that big

22   trials are a big problem.        These are the kinds of problems

23   that come up when you have smaller amounts of data.

24                     DR. MCCRACKEN:        I think those are very

25   justified concerns.            Indeed, the smaller number of

                                S A G CORP.
     202/797-2525               Washington, D.C.             Fax: 202/797-2525

1    pneumococci is worrisome, because that is the one pathogen

2    that you would like to have for bacterial meningitis.

3                   I mean, meningococcus, when I reviewed the

4    data from Malawi for a paper in the Lancet, the case

5    fatality rate for meningococcus meningitis was 4 percent.

6     The case fatality for haemophilus was 30 percent, and

7    35 percent for strep pneumo.

8                   Well,   30 and 35 percent for those two

9    organisms, in the United States, it is 4 percent for

10   haemophilus, and 8 percent for pneumococci, and yet you

11   see the huge difference.

12                  And so one agent, given as a pre-dose, or

13   prior therapy, can have a huge impact on meningococcus.

14    So I tend to discount that and look more to the other

15   two agents, and most especially pneumococcus.        So there

16   was that issue.

17                  This early failure thing gets down to one

18   issue.     I mean, I hate to mention it, but it was a bias

19   of the investigator.      He did not like fluoroquinolone,

20   and he should never have been allowed in that study.

21                  He did not come to the investigators meeting,

22   and that is the issue.     And that's why I pointed out that

23   it is unacceptable, totally unacceptable to do a study

24   now where an investigator is not part of the original

25   description and review of the protocol.

                             S A G CORP.
     202/797-2525             Washington, D.C.     Fax: 202/797-2525

1                     And if that investigator feels that the

2    protocol is not suitable for his or her institution, fine,

3    it shouldn't be in it.             But that wasn't what happened

4    there, and so we had to go back and look at that, and

5    see why was there a failure.

6                     And he just had a bias towards the other drug

7    to   compare.          It    is    unfortunate,    but     fortunately

8    trovafloxacin is never going to be used for bacterial

9    meningitis.       So it wasn't an issue.

10                    It would have been an issue had it been --

11   I would have made a big issue of this, and probably

12   appealed to the FDA if it had ever come to them for this.

13    It was purely an error in that regard.

14                    CHAIRMAN RELLER:           I would like to ask the

15   same question and comments from Drs. Tally, McCracken,

16   and Dr. Shlaes.             In your presentations, there was a

17   recurring themes that for some of the most serious

18   infections, where the numbers of plausible patients

19   enrolled       would   be    the   smallest,     such    as   infective

20   endocarditis, meningitis, the deltas should be larger.

21                    But paradoxically those infections also, at

22   least some of them, have the most objective end-points.

23    Where on the other hand, Dr. McCracken has emphasized

24   that deltas could be very small, 5 percent or less.

25                    And then the analogy to a not so serious

                                   S A G CORP.
     202/797-2525                  Washington, D.C.           Fax: 202/797-2525

1    infection, where in fact there are specific threshold

2    criteria for even considering the efficacy of the drug,

3    and    specifically     gonococcal     infections,   where     the

4    eradication rate must be 95 percent, or any other

5    considerations in approval of the compound are not

6    considered.

7                   So my question is this.       Should we consider

8    different      deltas    for      clinical    end-points       and

9    bacteriologic end-points with specific infections?

10   And also pathogen specific.

11                  So, for example, with meningitis, that if

12   there were approval, there would have to be X-number of

13   patients with pneumococcal infection, and they would have

14   to have a 95 percent or delta 5 percent eradication of

15   the organism by specific methods at particular points

16   after initiation of therapy.

17                  And that other considerations of second

18   end-points for clinical outcomes at 6 weeks, 6 months,

19   follow-up blood cultures at X-number of months with

20   endocarditis, might have different criteria.

21                  Because it seemed to me that one of the

22   driving issues for considering wider deltas was not a

23   clinical reason, but rather a practical reason having

24   to do with economics and number of enrollable patients.


                              S A G CORP.
     202/797-2525             Washington, D.C.        Fax: 202/797-2525

1                   So    how    does     one       bring   those      clinical

2    necessities, objective possibilities of really tight

3    criteria       for     efficacy         microbiologically              into

4    consideration with the realities of the numbers and the

5    economics?     Drs. Tally, McCracken, and Shlaes, comments

6    on those possibilities?

7                   DR. TALLY:       That's why I put the paradox on

8    the severe infections, because if you look in the response

9    from the FDA in the beginning of this material that was

10   handed out, that is the paradox.

11                  You want more surety in the most severe

12   infections.     But the fact is that when you put that tight

13   clinical delta, you are increasing the size where you

14   never are going to have that study to be even -- to measure

15   anything.

16                  So what are some of the alternatives?                     And

17   one of the reasons that has been pointed out is that if

18   there is a hard microbiological end-point, and I think

19   we    should    talk       about    your        proposal      with     that

20   microbiological endpoint, because it is going to be clear

21   early on that if somebody doesn't clear their bacteremia

22   by the fifth or sixth day with endocarditis, I mean, that

23   is a clear failure.

24                  And as you move along -- and it may come down

25   to a smaller delta with that clear number of patients.

                                 S A G CORP.
     202/797-2525                Washington, D.C.               Fax: 202/797-2525

1     It is in designing a study to say that you have to enroll

2    600 patients in a study, I think you probably with these

3    various serious illnesses, with the hard base line, that

4    you can do lower numbers, and draw valid conclusions from

5    those lower numbers.

6                   And based upon everything that Mark was just

7    saying, taking everything into consideration, and the

8    different pathogens, and the predicted outcome in those.

9                   But a priority to say that you have to do

10   a 700 patient study in endocarditis, you are never going

11   to see that based on that small delta.          So I do think

12   you open it up for different approaches.         David, do you

13   want to comment?

14                  DR. SHLAES:   Yes.     I think the comments that

15   we made were based on the current clinical outcome at

16   trial design.      Clearly, if you have microbiological

17   end-points, and one of the points that we are going to

18   make tomorrow, and which we will start make tomorrow,

19   is that it is about time for us to be using surrogate

20   end-points in trials of anti-bacterials, one of which

21   could be bacterial eradication.

22                  And it is something that has been done in

23   the anti-viral group for a very long time already, and

24   so I don't see any reason why we can't do it.            I think

25   you could have smaller trials with hard end-points using

                            S A G CORP.
     202/797-2525            Washington, D.C.         Fax: 202/797-2525

1    microbiological end-points for certain infections.

2                   I think though that your suggestion of having

3    different deltas for each specific pathogen within an

4    indication is going to get down to being difficult to

5    get the appropriate number of patients for those cuts.

6                   So   you    probably          will   have    to    take

7    microbiological end-points in all-comers for a number

8    of those infections.        The other limiting factor would

9    be, for example, an osteomyelitis, to getting follow-up

10   cultures will be technically an issue.

11                  And having enough centers in the case of

12   otitis media that could do tabs to support all of the

13   development that might be going on might also be an

14   issue.

15                  But I agree with the idea.           I think we all

16   agree with the idea that microbiological end-points is

17   a very good way of going forward, and it is long overdue.

18                  CHAIRMAN RELLER:        Dr. McCracken.

19                  DR. MCCRACKEN:       Well, it is an interesting

20   question, Barth.     I hadn't really thought of you quite

21   the way that you put it.        But I can tell you that in 30

22   years, seeing I don't know how many hundreds of cases

23   of meningococcal meningitis, I have only seen delayed

24   sterilization once, and that was because the wrong drug

25   and the wrong dosage was used.

                               S A G CORP.
     202/797-2525              Washington, D.C.            Fax: 202/797-2525

1                     So there you could have a delta of one

2    percent.       I mean, that is a rule.       You get bacteriologic

3    cure.     With pneumococcitis, and haemophilus, it is not

4    a rule.

5                     The studies in the late '70s and early '80s

6    by Ken Altland showed about an 18 percent to 20 percent

7    delayed sterilization at 18 to 24 hours.           But by 36 hours,

8    it was a hundred percent.

9                     So it depends on when that end-point is taken,

10   and I would definitely never go out beyond 36 to 48 hours.

11    I think the end-point, if it is taken at 18 to 24 hours,

12   can be a little broader.         Maybe 5 to 7 percent.

13                    But if it is taken at 30 to 36 hours, then

14   it should be very tight, because by that time you have

15   cure.    I am talking about meningitis only, and I am not

16   addressing issues of endocarditis or other diseases.

17                    CHAIRMAN RELLER:      Dr. Wittes.

18                    DR. WITTES:    Yes, I have a question about

19   when you develop a new drug, do you in fact expect that

20   it is no better in terms of cure than what is on the table?


22                    And the reason that I am asking this is that

23   as Dr. Brittain pointed out in her presentation, the

24   sample size is really driven by the assumption that the

25   underlying rates are identical.              And that's what makes

                               S A G CORP.
     202/797-2525              Washington, D.C.           Fax: 202/797-2525

1    the sample sizes really high.

2                   But if in development you have seen better

3    bacteriologic end-points, and you really believe that

4    the efficacy, the clinical efficacy, is even slightly

5    better than the comparator, then the sample size goes

6    way down.

7                   So my question is that in development are

8    you aiming for improvement that you can't see, or are

9    you aiming at equality?

10                  DR. SHLAES:   Okay.     So I think, at least from

11   our point of view, and I have a few colleagues who will

12   chime in, I hope, when you look at the variability in

13   the population within and an indication is such that it

14   is very hard to prove a superiority in terms of clinical

15   end-points, such as a cure.

16                  And if you look at other end-points, such

17   as time to cure, you might be able to do superiority

18   trials, or there may be other end-points that may be more

19   applicable to a superiority study.

20                  But   if you look at the usual clinical

21   end-points, superiority is difficult to show.          The other

22   issue is that if you actually run the numbers on a

23   superiority trial, taking all-comers in a clinical study

24   for clinical end-points, they are actually not all that

25   much difference.

                             S A G CORP.
     202/797-2525            Washington, D.C.         Fax: 202/797-2525

1                   Again, at the 90 percent power.                 So if you

2    do a 2 percent superiority study and a 90 percent power,

3    and you account for non-evaluable patients, the numbers

4    actually get to be just as large as they would be in the

5    current step function study.

6                   So I am not sure that in terms of patient

7    numbers that there is an advantage there anyway.

8                   DR.    WITTES:      But       you    are    answering          a

9    different question.       Can I clarify the question?

10                  DR.    HARDALO:       Maybe     I     could     also     add

11   something in.    When we develop a drug, we really believe

12   based on our animal data, and our lab data, that it is

13   better than what exists.

14                  However, real life often times gets in the

15   way of proving that.      And as Dr. McCracken said, and as

16   I am sure as Dr. Talbot has experienced, that in diseases

17   where there is a significant mortality rate, like VRE

18   infections,          or     bacterial              meningitis           and

19   immunocompromised hosts, or I can name a whole list of

20   infections, including endocarditis with Staph aureus,

21   and hospital-acquired pneumonia.

22                  The inflammatory sequelae caused by the

23   bacteria is responsible for the vast majority of the

24   morbidity and mortality that ensues.                Therefore, I take

25   a clinical only based end-point is going to be very

                               S A G CORP.
     202/797-2525              Washington, D.C.                Fax: 202/797-2525

1    difficult for you to prove that significant differences

2    between the treatment groups exists.

3                   And there is no way that one could do a placebo

4    controlled     trial,    and    because       there   is     inherent

5    variability in the patient populations, you will enroll,

6    least of which is the standard of care in the center that

7    you are having in your study.

8                   And it can present significant issues for

9    trial design, and it is not always something that you

10   can take care of in a prospective stratified, randomized,

11   clinical trial.

12                  DR. FLEMING:      Can I make a suggestion in the

13   interest of time?       I think Dr. Wittes is raising a very

14   key point.     I am going to be discussing this in some detail

15   in my presentation, and maybe we can return to it after

16   that if there are still remaining issues?

17                  DR. WITTES:      Sure.        I just wanted to make

18   it clear that you both answered a question different from

19   the one that I have asked.

20                  DR. FLEMING:      Yes.

21                  CHAIRMAN RELLER:         Thank you, Dr. Fleming,

22   and that is the approach that we will take.                 We heard

23   from Dr. Hardalo, and we had earlier hands up.                      Dr.

24   Maxwell, do you have a question, and then Dr. Bell, and

25   then Dr. O'Fallon, and then we will get on to the next

                              S A G CORP.
     202/797-2525              Washington, D.C.            Fax: 202/797-2525

1    presentation.

2                     DR. MAXWELL:    Yes.        The question is for Dr.

3    McCracken, just to clarify for me.              Would bacteriologic

4    outcomes, and let's say in the case of haemophilus

5    meningitis, be the same in a child that had the vaccine,

6    and one that didn't?         Should it have the same exact

7    measure?

8                     DR. MCCRACKEN:      Well, one would hope that

9    the child who received the vaccine wouldn't develop the

10   disease.       With haemophilus, they wouldn't, most likely.

11    With pneumococcus, we are seeing a couple of failures,

12   and their disease looks identical to those who had gotten

13   no vaccine.

14                    And the reason is that the spinal fluid is

15   a sequestered or privileged site, where there is no native

16   immune function.       Antibody compliment white cells are

17   not present.

18                    So the organism, once it gains footing there,

19   can multiply without any control from immune function

20   until late in the course.         So if it develops, which is

21   less likely in the vaccinated child, it probably would

22   have a similar course.

23                    And this is not true necessarily in the

24   systemic disease, but for meningitis, I think it is, yes.

25                    DR. BELL:   I wonder if the speakers could

                               S A G CORP.
     202/797-2525              Washington, D.C.            Fax: 202/797-2525

1    comment on how these issues apply to the development of

2    drugs for resistant infections in particular.         Are those

3    study designs considered to be superiority trials, in

4    the sense that the new drug has to be better than the

5    drug for which the drugs are now becoming resistant?

6                   Do they also have to meet non-inferiority

7    criteria in the treatment of sensitive infections?           What

8    are the implications of some of what you have been

9    discussing specifically for resistance?           How do you

10   address that issue?

11                  DR. SHLAES:     Actually, I think we are going

12   to have a whole day on resistance tomorrow.         Can I hold

13   -- are you going to be here tomorrow?          Can I hold you

14   off until tomorrow?

15                  CHAIRMAN RELLER:      We will do that tomorrow.

16    Dr. O'Fallon.

17                  DR. O'FALLON:    I have a couple of questions.

18    I am trying to understand the thinking process that has

19   been processed in the documents that we have seen from

20   industry in our packet.

21                  The first one is I was a little surprised,

22   or there was some support that has been voiced for the

23   delta procedure.     Now, I am a little bit puzzled as to

24   why that is considered a good idea to be able to when

25   you have a very successful comparator, that you would

                            S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    want to spent a lot of patients to try to prove a very

2    small difference.

3                    Whereas, you want to spend far fewer patients

4    when the successful rate is down closer to 50 percent.

5     You know, 70 percent, 65 percent, and that sort of thing.

6     You are willing to spend half as many patients to try

7    to   prove     what   you    would     call    efficacy   as    being

8    non-inferior to the other thing.

9                    Why are you not asking instead to just hold

10   a sample size constant for your study, and then take the

11   delta that comes out of that?

12                   CHAIRMAN RELLER:         Dr. McCracken.

13                   DR. MCCRACKEN:       I don't know exactly how that

14   applies to what I am -- well, I don't know what you are

15   leading to with regard to --

16                   DR. O'FALLON:        I don't think you spoke in

17   favor of the delta method, and some of the others did.

18                   DR. MCCRACKEN:         Oh, I am not against the

19   delta method.     I just want a broader limit.            I think it

20   is wonderful, and I just propose that it be a 20 percent

21   difference in proportions for clinical outcome and a much

22   narrower one for bacteriologic outcome.

23                   DR. O'FALLON:        But the delta is defined to

24   be a step function where you spent fewer and fewer patients

25   in order to establish a bigger delta.            Why do you go with

                                 S A G CORP.
     202/797-2525                Washington, D.C.         Fax: 202/797-2525

1    fewer patients around when there is a lower success rate?

2     What is considered to be, or why is that a good idea?

3     It is not obvious to me.

4                    DR. MCCRACKEN:     Well, I don't know if it is

5    a good idea or not, but unfortunately what you are faced

6    with, with bacterial meningitis, when you leave the United

7    States and go to developing nations is a very good outcome.

8                    That is to say that the clinical outcome there

9    is probably in the rage of 60 to 70 percent success, and

10   maybe not even that high.         Therefore, it is easier to

11   do a study because you might be able to show a difference

12   with the smaller numbers.

13                   But my point was only that using a 10 percent

14   difference in proportions for a disease in the United

15   States, or even throughout, we can't get a thousand

16   patients.      We just cannot do that any longer.

17                   We need -- we -- I -- it is not me, but to

18   do a study, and for me to be a principal investigator

19   of that study, I can't -- 10 years is too long.

20                   DR. O'FALLON:     I understand that part.

21                   DR. MCCRACKEN:    I may not be here in 10 years.

22                   DR. O'FALLON:     But why not go for a, sat,

23   set number of patients; that you are going to serve a

24   minimum sample size, and then take whatever the delta

25   is that you can buy with that.              Spend fewer and fewer

                              S A G CORP.
     202/797-2525             Washington, D.C.           Fax: 202/797-2525

1    patients, the harder it is to distinguish the differences.


3                     DR. MCCRACKEN:        Well, I guess my -- and

4    probably       statisticians     can    answer   this   far    more

5    competently than I could, but I am afraid that if I used

6    -- whatever that defined number of patients would be,

7    I am afraid that you might be surprised by the outcome.

8                     It could by chance be that you have a much

9    better outcome in the countries that were selected, and

10   therefore, it is in the 80 to 85 percent range, and small

11   numbers would give you inferior data, and you couldn't

12   tell the difference.

13                    So, therefore, you would shoot yourself in

14   the foot by preselecting without knowing exactly where

15   you stand.      And that would worry me.

16                    CHAIRMAN RELLER:      We need to get on to the

17   same presentation, and Dr. Albrecht had a comment that

18   she wished to make.

19                    DR. ALBRECHT:     Actually, I wanted to just

20   follow up on the microbiological discussion that we had

21   earlier.        Dr. McCracken, you indicated during your

22   presentation of the trovafloxacin data that the patients

23   that were pretreated even with a single dose,           you could

24   often see up to a two-fold reduction in the colony count

25   when the patients were entered.

                              S A G CORP.
     202/797-2525             Washington, D.C.          Fax: 202/797-2525

1                   So I just wanted to use that as an opportunity

2    to ask whether we might consider if we are going to hear

3    suggestions about microbiology a quantitative approach

4    to microbiology.

5                   And I just wanted to mention that we use that

6    in the evaluation of urinary tract infection agents

7    currently, but not in other sites, and in meningitis,

8    a sterile site, I would appreciate comments on that.

9                   But also then in the afternoon as we hear

10   other presentations, I would like to raise that same issue

11   relative to sites that are not normally sterile.

12                  DR. MCCRACKEN:     I mentioned that there can

13   be up to a two or even larger log count drop in the

14   pre-treated, and that was based on data int he '70s by

15   Bill Feldman, in which ceftriaxone was not one of the

16   agents used.

17                  It was mainly ampicillin and other drugs,

18   and amoxicillin, which had an impact.       Ceftriaxone might

19   even have a greater impact.           The problem with doing

20   quantification of bacteria in CSF is that it is not as

21   simple a thing to do.

22                  The investigator who did that study was up

23   all night.     He came in whenever a patient came in, and

24   that is a tough chore.            You could put it in the

25   refrigerator.     It is doable, but it is very difficult,

                             S A G CORP.
     202/797-2525            Washington, D.C.       Fax: 202/797-2525

1    particularly when you get outside the country to actually

2    do quantification.

3                   You do get a rough estimate of bacteria by

4    just looking at the stains smear, knowing that the break

5    point, and seeing bacteria per field, is about 10 to the

6    5.

7                   So if you see multiple organisms, which we

8    have a child in the hospital now, probably has 10 to the

9    8, or 10 to the 9, and we know the outcome there is very

10   poor.

11                  CHAIRMAN RELLER:      It is time to hear from

12   the Infectious Diseases Society of America, a group that

13   is very much involved, both in the development and

14   carrying out of clinical trials, as well as importantly

15   in the use of these agents in clinical practice.                Dr.

16   Andriole, your team.

17                  DR. ANDRIOLE:      Thank you, Barth.           As I

18   pointed out earlier this morning, we are here to represent

19   the   Infectious   Disease   Society      of   America,   and    my

20   colleagues, Jack Edwards, and Dennis Wallace, and George

21   Talbot.

22                  As you know the society has now more than

23   7,000 members, and it was founded 40 years ago.              I was

24   one of the founding fathers.       No comments, please.         And

25   the member really cover all of the areas of infectious

                            S A G CORP.
     202/797-2525           Washington, D.C.           Fax: 202/797-2525

1    disease.

2                   And without being arrogant, they are people

3    who have contributed their life to studying particular

4    issues, and I know that you recognize this.                      Seven of

5    you on this committee are members of the society.

6                   And so the agency has to recognize this, and

7    one, as a past president.          In addition, we have some very

8    excellent members from the pharmaceutical industry who

9    are members of this society.

10                  And that we would like to help the agency

11   accomplish the goals that it has set out to do.                          My

12   involvement    with    the      agency,         as   secretary    of    the

13   Infectious Disease Society -- and Lillian will remember

14   this if she is -- yes, she is right here.

15                  Awe    were     very     concerned       about    clinical

16   investigation, and the guidelines that had been written

17   in 1977 were pretty much outdated.                      And so in the

18   mid-1980s, or actually the late-1980s, the Society and

19   the Agency put together a task force to redo the

20   guidelines.

21                  The late Tom Beam was our liaison, with Matt

22   Lufkin, and Lillian, and Dr. Peck.                   And we volunteered

23   -- all of the members of the society volunteered to come

24   down and to write guidelines.

25                  We were given two years to do it, and in two

                              S A G CORP.
     202/797-2525                 Washington, D.C.             Fax: 202/797-2525

1    years, and this is the flow sheet -- this is a classic

2    paper -- we wrote 13 guidelines.           And they were finished

3    in 1990, June 24th.

4                   Now, that is a decade ago, and I think they

5    have served us well for the majority of that decade.

6    I have also been co-author of one of those guidelines,

7    and I was a member of this committee for 3 years, and

8    paid my dues, and did all of that.

9                   And I have been doing clinical research in

10   Phase III and IV trials for 43 years.             But now that I

11   have joined the more mature population, I don't do that

12   any more.

13                  How I wound up being the spokesperson for

14   this meeting is not clear to me, and I just have drawn

15   the short straw, and once I was told by the council that

16   I was going to be the speaker, reminded me of a story.

17                  When I was teaching in Kenya, on the edge

18   of the Serengeti, I had wanted to go down and visit a

19   village of Masai warriors.        So my wife and I went down

20   there, and I was talking to the chief, and my wife was

21   playing with the children and talking to the women.

22                  And the chief looked very sad, and I said

23   to him what is the matter, and he said, well, I just lost

24   one of my best warriors.        I said, oh, that's too bad.

25   What happened?

                            S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1                   Well, he said that he was running across the

2    Serengeti to come back to the village, and he came around

3    a clump of trees and there was a lion.            And he looked

4    to his left and he looked to his right, and he looked

5    behind him, and it was clear.         There was no escape.

6                   So he dropped to his knees and clasped his

7    hands, and started to pray.           And after five minutes

8    passed, nothing happened.       And so he looked up and there

9    was the lion on his knees with his hands clasped.

10                  And the warrior said to the lion why are you

11   doing what I am doing, and the lion said to the warrior,

12   I don't know what you are doing, but I am saying grace.

13    Well, that's how I feel right now.

14                  (Laughter.)

15                  DR. ANDRIOLE:    First, I will be very brief,

16   Barth, because somebody asked me before the meeting

17   started what are you going to say.         I said, well, I don't

18   know what I am going to say until I hear what everybody

19   else has to say.

20                  And I don't have any slides, and so you are

21   just going to have to pay attention to me, or fantasize,

22   or whatever you want to do.        But the fact of the matter

23   is that everybody has touched on all of the issues that

24   I have been instructed to tell you from the Infectious

25   Disease Society of America.

                            S A G CORP.
     202/797-2525            Washington, D.C.          Fax: 202/797-2525

1                     I want to make a couple of points clear.

2    One, as an organization, we have no vested interest in

3    this agency, or in the pharmaceutical industry.                I am

4    here as a representative of the society for two major

5    reasons.       One, we want to be able to treat our patients

6    with the best medical care.

7                     And without the continued development of

8    anti-infective agents, forget it.             We will be out of

9    business.       We want to help people, and we know that the

10   agency doesn't want to embarrass itself by preventing

11   the development of new agents.           That would be a tragedy.

12                    Number 2, you can sit here and talk all you

13   want all day long; the industry and the agency, who does

14   the work?       We do.   We are the clinical investigators.

15                    So we beg you, we know that he current

16   guidelines should be updated in different ways, but we

17   are a little concerned about the criteria, because if

18   you set the bark too high, you can't do the work.

19                    And George McCracken said that very clearly,

20   as have others, by discussing the mathematical approach

21   to clinical investigation.          We would like to the agency

22   to adopt a scientifically and statistically appropriate,

23   but also a clinical practical approach, to determining

24   efficacy.

25                    I don't care whether you want to call it a

                               S A G CORP.
     202/797-2525               Washington, D.C.        Fax: 202/797-2525

1    delta, or a mega, or a zero, or whatever.                  But that is

2    what we would like to see.         That you have when you review

3    these     NDAs    that    come    into        you   in   trucks,       and

4    electronically now, that you have a reasonable chance

5    of evaluating this data to determine whether we are going

6    to get to use it in our patients.

7                     Now, is it -- do we really need to focus on

8    a delta?       Is that going to be the end point for clinical

9    investigation?       I mean, you just raised that question.

10    Is that he end all and the be all of what we should be

11   doing?     I don't think so, and neither does the Society.

12                    We think that you have to evaluate, one, the

13   frequency of the disease.               If the disease is very

14   frequent, make the delta whatever you want.                   Number 2,

15   if the patients are not available in order to study

16   thousands of them, then you have to come up with a

17   different plan.       You really do.

18                    Otherwise, there is not going to be any more

19   anti-infective research for the kinds of diseases that

20   we need to treat.        Well, how can we do that?           We are not

21   going to settle that today, but some of the suggestions

22   have been already nicely stated by our colleagues who

23   have already presented.

24                    And some of the suggestions, and the details

25   of all of this, the nuts and bolts in working it out can

                                S A G CORP.
     202/797-2525               Washington, D.C.              Fax: 202/797-2525

1    be done later.      But we need to know what surrogate

2    end-points we should be using based on the type of

3    infection that we are treating.            We have to really look

4    at that.

5                   And what are surrogate end-points?             Well,

6    George pointed out that clearance of the bacteria from

7    the cerebral spinal fluid in meningitis.             Others have

8    asked the question can we do quantitative microbiology

9    and other infectious diseases.

10                  That's a hard thing to do from a practical

11   point of view, and there are other ways that you can use

12   surrogate end-points; rapidity to cure is one that people

13   are now looking at.    Those are just some of the examples.

14                  The second thing is that animal models of

15   disease have been the bridge between Phase II studies

16   and Phase III studies for years.             And many of us have

17   spent our lives developing animal models, which the agency

18   has used in hits deliberation before a Phase III protocol

19   is designed.

20                  Pharmacokinetics and pharmacodynamics are

21   extremely important.    I am now speaking for the society,

22   and they really feel that that kind of data is very helpful

23   in determining whether a particular Phase III study is

24   likely to work.

25                  And finally the level of anti-microbial

                            S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    resistance     in   your   ability       to   determine   what     the

2    comparative agent is going to be.             In patients who have

3    very serious illness, we have to lower the bar.            We really

4    do.

5                   An example -- Frank gave examples of this,

6    and David gave examples of this, and these are very

7    important things in our view.            We wanted to compliment

8    the agency actually on the paper that you wrote on

9    resistant pathogens.

10                  We all went through that in great detail,

11   and we thought that was really good, and we hoped that

12   it could be refined just a little bit more.          But the final

13   message from the Society is the Infectious Disease Society

14   of America is here to help you.           That is the message that

15   we want to leave you with.

16                  We want to help in any possible way.            We are

17   prepared to volunteer any member of the Society.                   You

18   tell us what you want us to do, and we will make a list

19   of people that you can call on to help you solve some

20   of these problems.

21                  We have done this in the past, and Lillian

22   knows that, and we worked very hard for two years to get

23   done what had to be done, and we are prepared to do that

24   now.

25                  We will update your guidelines, and we will

                                S A G CORP.
     202/797-2525               Washington, D.C.          Fax: 202/797-2525

1    help you work out a delta.           I don't think that can be

2    accomplished in a big meeting like this.                 So we are

3    suggesting that maybe the agency might want to consider

4    a task force to meet with representatives from the

5    Infectious      Disease      Society          of   America,      with

6    representatives of PhRMA.

7                   After all, they are integral players in this,

8    and with representatives from the agency, to try to fix

9    the issues that have been raised so clearly today.                  We

10   have many qualified members who are really willing to

11   volunteer their time, just like they did 12 years ago.

12                  And that is probably the most important

13   message that I have, Barth, from the Society.                      Any

14   questions that you have, and I don't know, one, Barth

15   wants to have the questions.

16                  I have three distinguished colleagues who

17   will be very happy to answer them, and I am very happy

18   to have escaped the lion.         Thank you.

19                  CHAIRMAN    RELLER:           I think it would be

20   actually a good time for questions for Dr. Andriole and

21   other members of the IDSA.           Not that they aren't also

22   included on our advisory committee as Vince pointed out.

23    Questions?    Yes.


25                  DR. NELSON:      I would be interested in some

                               S A G CORP.
     202/797-2525              Washington, D.C.           Fax: 202/797-2525

1    comments       on    the   surrogate    end-point      issue,   and    in

2    particular whether one can extrapolate microbiological

3    end-points from meningitis, which I thought was well

4    argued based on clinical data, to other infectious

5    diseases.

6                        Working in an ICU and seeing the result of

7    host response, I would have to be convinced that there

8    is no drug disease interaction that would have to be

9    considered in some of these other conditions.

10                       There was clinical data to support that use

11   of surrogate end-point meningitis, but does that data

12   exist in a lot of these other conditions?

13                       DR. ANDRIOLE:      Well, that is one of the

14   issues that needs to be hammered out, and that is a very

15   important question.            Microbiologic endpoints in the

16   intensive care unit in patients with a hospital-acquired

17   pneumonia, forget it.

18                       You can't even get the pathogen to begin with.

19    You don't know what you are treating.                  But there are

20   other surrogate markers that can be looked at, such as

21   APACHE         scores,     temperature          response,   radiologic

22   clearance, improvement, oxygen saturation.

23                       Now, you can say, well, that might happen

24   anyway, but it doesn't.           That is a disease with a high

25   mortality and you know that.              But this is what we need

                                  S A G CORP.
     202/797-2525                 Washington, D.C.           Fax: 202/797-2525

1    to do to sit down and talk about what are the surrogate

2    endpoints for each type of disease that are acceptable,

3    and will provide information to help with the agency

4    decide on efficacy.        But I don't have any specific

5    criteria.

6                    DR. MCCRACKEN:       I can give one.               Acute

7    otitis media.       The data are quite clear now that a

8    double-tap      study    giving      bacteriologic           endpoints

9    correlates beautifully with clinical outcome.

10                   Now, studies are not easy, particularly in

11   the United States, but that is a very good example of

12   bacteriologic eradication in clinical cure.

13                   CHAIRMAN RELLER:        It was Dr. Nelson who

14   imposed that question to Drs. Andriole and McCracken.

15   Other comments from the IDSA in response to this query,

16   or other questions?      Yes.

17                   DR. EDWARDS:    Just to cite another example

18   of consideration is the resolution of candidemia, which

19   is   in    a   problematical    area        for   studying    of    the

20   antifungals.

21                   It is a complex issue again, but the surrogate

22   endpoint of just the resolution of the candidemia is a

23   factor to consider.

24                   CHAIRMAN RELLER:     That was Dr. Edwards.           One

25   of the constraints with the less commonly encountered,

                              S A G CORP.
     202/797-2525             Washington, D.C.             Fax: 202/797-2525

1    and often requiring many patients enrolled from outside

2    of the United States, specifically meningitis, is there

3    any room for looking at it from the direction of what

4    are practical numbers of patients, and then what criteria

5    experienced individuals would be comfortable with that

6    would demonstrate reasonable efficacy.

7                    For example, the concept of if you had

8    X-hundreds of patients, to demonstrate efficacy, you

9    would need these etiologies, these deltas as regards

10   eradication of organism at 24 and 48 hours, or 24 this

11   delta, and 48 this delta.

12           And this latitude of clinical assessments out at

13   six weeks or six months.        Basically, not starting with

14   a delta in one or the other areas, but starting with this

15   is the maximum number of patients that are possible, and

16   then how much information?

17                   I mean, basically, it is issues of numbers

18   versus quality of information in smaller numbers of

19   patients.      Dr. McCracken, any thoughts on that approach?

20                   DR. MCCRACKEN:       Well, I think it is an

21   interesting approach.       When I sort of threw out those

22   numbers of up to 24 hours, or 24 to 36 hours, I really

23   wasn't proposing those.

24                   And I would really have to think about that

25   in terms of numbers, because it gets a little tricky,

                              S A G CORP.
     202/797-2525             Washington, D.C.      Fax: 202/797-2525

1    particularly as you get the pneumococcal disease.

2                     I think that approach is a very reasonable

3    one, and I would echo Vince's comments that surrogate

4    markers become more and more critical as we try to evaluate

5    diseases that are becoming less and less common.

6                     I would think five years from now that there

7    will be no meningitis studies in any developed nation

8    with the prospect of a meningococcal vaccine, and already

9    there    are     conjugate      meningococcal         and     haemophilus

10   vaccines, and that disease will be in small numbers.

11                    And one could argue then immediately, well,

12   why even worry about it.          Well, it doesn't mean that it

13   disappears.          And   it   is    in      other    countries,         and

14   resistance, and we all know when you disappear, or when

15   one pathogen disappears, something pops up sometimes in

16   its place.

17                    So they are necessary.           But your approach,

18   Barth, I think, is an appropriate one, but I am not willing

19   to give numbers yet because I really have not given it

20   enough thought.

21                    CHAIRMAN RELLER:         This is only a concept to

22   increase       the   repertoire      of    things     that       could     be

23   considered.      It looks like it is time to hear from Dr.

24   Thomas Fleming from the University of Washington on issues

25   regarding choice of the margin in non-inferiority trials.

                                S A G CORP.
     202/797-2525               Washington, D.C.                 Fax: 202/797-2525

1     Dr. Fleming.

2                     DR. FLEMING:     Thank you, Barth.        Well, as

3    Vince has pointed out, there has already been -- much

4    has been said, and what I would like to try to do is

5    highlight and amplify several of the key issues that are

6    important in the choice of the margin.              Next slide.

7                     I think it is important when we are thinking

8    about choice of margins to keep in mind as has been stated

9    today there really is a dual goal here in non-inferiority

10   trials.

11                    First, to enable a direct evaluation as to

12   whether or not the benefit to risk profile of the

13   experimental therapy truly is adequate relative to the

14   benefit to risk profile of the active comparator.

15                    And also to contribute evidence to evaluating

16   whether or not the experimental truly is superior to the

17   placebo.       Well, what I would like to do, and it is going

18   to be kind of a quick overview, because a number of these

19   issues    have    been   covered,    looking   at    factors     that

20   influence the choice of margin.

21                    I will be talking about issues of clinical

22   relevance, as well as active control effects, and I will

23   be briefly talking about some issues that impact the

24   interpretation of non-inferiority trial results.                  Next

25   slide.

                               S A G CORP.
     202/797-2525              Washington, D.C.           Fax: 202/797-2525

1                   So if we look first at issues of clinical

2    relevance, and in choosing the margin, it is very

3    important to consider the clinical relevance of the

4    primary end point.

5                   If it is a morbidity, major morbidity or

6    mortality end point, even most changes in efficacy can

7    have considerable clinical importance.     At the same time,

8    it is important to consider when thinking about the

9    experimental against the active comparator, do we expect

10   an alteration and hopefully an improvement may be in the

11   safety or tolerance profile, and convenience of the

12   administration, or other issues such as resistance or

13   drug interactions.

14                  If in fact there are important improvements

15   in these areas to be expected by the experimental, that

16   should in fact be factored in, in the choice of the margin,

17   and it could influence choice of margin.      Next slide.

18                  The ICH guidelines also point out that

19   factors relevant or related to the active control effect

20   should influence the choice of margin.     And essentially

21   they are arguing that ideally we want well designed

22   superiority trials to clearly establish the efficacy of

23   the active comparator.

24                  And that ideally, and this assay sensitivity

25   issue that Dr. Temple referred to, we would like those

                            S A G CORP.
     202/797-2525            Washington, D.C.      Fax: 202/797-2525

1    estimates to be reliably predictive of what the estimates

2    or what the actual efficacy of the active comparator would

3    be in the non-inferiority trial.           So, the next slide.

4                   I would like to on this slide illustrate then

5    three factors related to the active control effect that

6    really should be influential in our choice of the margin.

7                   First of all, ideally we would like to be

8    doing active comparator trials in settings where the

9    active comparator is very effective with a precisely

10   estimated level of efficacy.

11                  So, for example, to illustrate.            Suppose

12   that a placebo has a 45 percent cure rate, and the active

13   comparator increases that to an 80 percent cure rate.

14   And this is estimated to within plus or minus 10 percent.


16                  So, for plotting here along this X-axis down

17   at the bottom, the cure rate on placebo relative to active

18   comparator, then the placebo is 35 percent less effective,

19   with estimates consistent to as much as 25 percent less

20   effective.

21                  Now, Dr. Temple has pointed out, as has Dr.

22   Brittain, that in some settings that you might set the

23   margin when you are choosing the margin to be specific

24   to preserving a fraction of the effect.           Let's say it

25   is half of the effect.

                             S A G CORP.
     202/797-2525            Washington, D.C.          Fax: 202/797-2525

1                   If we use this 25 percent estimate, and we

2    choose half of the effect, we might choose the margin

3    to   be    12-1/2   percent.        Using    this    then    in    the

4    non-inferiority     trial,     if   the     experimental     or    the

5    estimate of the experimental efficacy is favorable

6    relative to the active comparator, such that the lower

7    limit rules out this margin, this is a positive result.

8                   Now, this margin is greater than 10 percent,

9    and part of what justifies this is we are dealing with

10   an active comparator that is highly effective.

11                  And if in fact it could be clinically argued

12   that losing this much efficacy would be acceptable, then

13   one would have a margin of this size.               You might note

14   that when I derive this margin that I used the 25 percent

15   rather than the 35 percent estimate as a rationale for

16   that caution.

17                  And part of it is this assay sensitivity

18   issue.     Is the estimate of the active comparator obtained

19   from these historical or placebo controlled studies

20   relevant to the actual efficacy of the active comparator

21   in the non-inferiority trial.

22                  So specifically suppose in these historical

23   control trials we were looking at patients that were at

24   lower risk than the patients that would be looked at in

25   the non-inferior trial.

                             S A G CORP.
     202/797-2525            Washington, D.C.             Fax: 202/797-2525

1                        It might be that the active comparator is

2    more effective in lower risk patients than in the higher

3    risk patients in the non-inferiority trial.                   And there

4    may be other differences as well in the non-inferiority

5    trial from the active comparator trials.

6                        Why are these issues important?            Well, it

7    may be that the active comparator provided a very big

8    effect         in    the    historical       trials,    but     in     the

9    non-inferiority trial, its effect might be much more

10   modest.

11                       To position the placebo in green here might

12   be much closer to zero, compromising then the ability

13   or the integrity of using a margin of 12-1/2 percent.

14                       In this setting it may be that using the

15   margin of 12-1/2 percent not only assures us that we are

16   maintaining half of the effect, but we may not even be

17   able to conclude that we are maintaining any of the effect.

18                       Other issues also relate to being cautious

19   when doing non-inferiority trials, and that is the quality

20   of the design and conduct of a non-inferiority trial also

21   raises     factors         that   influence      the   interpretation,

22   particularly in non-inferiority trials.

23                       As the ICH Guideline E-9 indicates, many

24   flaws in design or conduct of the trial will tend to bias

25   results toward a conclusion of equivalence, such as

                                   S A G CORP.
     202/797-2525                  Washington, D.C.           Fax: 202/797-2525

1    eligibility criteria violations, non-compliance, loss

2    to follow-up.

3                   Why    is    that    especially      important      here?

4    Well, these types of biases in a superiority trial lead

5    to an increased risk of false negative conclusions.                  They

6    lead to an increased risk of false positive conclusions

7    though in a non-inferiority trial.

8                   I might focus for a moment on this issue of

9    loss to follow-up.         Next slide.         And it is not uncommon

10   in   antibiotic      non-inferiority           trials   for    valuable

11   datasets to involve maybe only 75 percent to 50 percent

12   of the overall randomized ITT dataset.

13                  If one is in fact excluding patients because

14   of the absence of the targeted pathogen, then that

15   probably just leads to an increase in variability.

16                  But if we are much more seriously, and if

17   we are excluding from the ITT, and if we are including

18   in the invaluable, but excluding patients who are not

19   assessed due to termination of treatment for reasons such

20   as     adverse       clinical        events,        perceived       drug

21   ineffectiveness, or because patients took prohibitive

22   concomitant meds, this is at risk of being what we would

23   call informative censoring.

24                  And it can substantially increase the bias,

25   and hence in non-inferiority trials, these issues arise

                                 S A G CORP.
     202/797-2525                Washington, D.C.            Fax: 202/797-2525

1    and should lead to greater caution in choices of margins,

2    and in particular in interpretation of results in such

3    studies.       Next slide.

4                     I would like to touch on an issue that was

5    motivated by a question from Dr. Wittes, and that is on

6    the issue of sample sizes, what we have heard a lot of

7    discussion about is that non-inferiority trials, if we

8    use scientifically rigorous margins, will always require

9    very large sample sizes.            Fact or myth?   Next slide.

10                    To address this, let's look at an active

11   control antibiotic that has an 80 percent cure rate, and

12   what I am plotting here along this X-axis is the

13   experimental, minus the active control cure rate.

14                    So, let's suppose that the experimental

15   improves this cure rate by 10 percent.                    Then the

16   experimental will have a 50 percent relative reduction

17   in non-cure rates, reducing the non-cure rate from 20

18   to 10.

19                    On the other hand, suppose the experimental

20   has a 10 percent or 15 percent lower cure rate than the

21   active comparator.       One would then have a 50 to a 75

22   percent relative increase in the non-cure rate, issues

23   that would generally would be viewed to be of concern.

24                    Well, let's look at in the setting of doing

25   superiority trials and non-inferiority trials when one

                              S A G CORP.
     202/797-2525                 Washington, D.C.       Fax: 202/797-2525

1    has an 80 percent cure rate.              Next slide.

2                      Well, in this setting, I am again along this

3    X-axis, and I am plotting the experimental, minus the

4    active, control cure rate.               And in a superiority trial

5    one is trying to rule out the no-hypothesis of equality.

6                      Let's suppose that the experimental arm truly

7    provides a 12 percent improvement over active control

8    in the cure rate.        One can then obtain 90 percent power

9    to rule out equality if one has about 340 evaluable

10   patients in the pool sample.

11                     A   reasonable     or    acceptable    sample       size

12   generally, and yet one is having to presume a very

13   substantial       effect     of    the    experimental.         So,    an

14   alternative to this approach would be scenario two.                   Next

15   slide.

16                     And that would be a non-inferiority design,

17   where one assumes a non-inferiority margin, and where

18   one    is      essentially    trying      to    rule   out   that     the

19   experimental arm has a 15 percent lower cure rate than

20   the active comparator.

21                     And in this setting, if the experimental

22   truly is the same as the active comparator in the cure

23   rate, then one would have 90 percent probability or power

24   to rule out this margin with the sample size of about

25   300 patients.

                                  S A G CORP.
     202/797-2525                 Washington, D.C.           Fax: 202/797-2525

1                   A concern that often arises in this setting

2    those is what if the experimental is 10 percent worse

3    in cure rate, which is a relative 50 percent increase

4    in non-cure.

5                   One   has almost a 20 percent chance of

6    achieving a false positive conclusion.                Next slide.     And

7    as a result, most rigorous non-inferiority margins of

8    10 percent have been advocated, and in that setting with

9    a 10 percent margin, if the experimental truly is the

10   same as the active comparator, one can have 90 percent

11   power to rule this margin out.

12                  But as has been noted, a substantially

13   increased sample size is the price.             Well, as Dr. Wittes

14   was really getting at in her question, the issue is that

15   in the superiority trial, we were having to presume a

16   12 percent improvement in cure rate in order to have good

17   power.

18                  Whereas,    if    that        might    not   be    highly

19   plausible, what if it is highly plausible that the

20   experimental    is   moderately       better         than   the   active

21   comparator.

22                  Wouldn't then we be able to rule out this

23   rigorous margin with reasonable sample sizes, and the

24   answer is yes, and that is scenario number four.                   Let's

25   suppose in fact that the experimental is only 3 percent

                               S A G CORP.
     202/797-2525              Washington, D.C.              Fax: 202/797-2525

1    better than the active comparator and cure rates.

2                   Then one would be able to achieve 90 percent

3    power then to rule out this more rigorous non-inferiority

4    margin with sample sizes that are in fact not a lot larger

5    than what would have been required in the scenarios one

6    and two.

7                   It is important to recognize when one is

8    looking at scenario number four these numbers in green.

9     Essentially    what   these   represent   are   what    is    the

10   estimated success rate on the experimental, in terms of

11   cure rate, relative to the active comparator.

12                  And in the superiority trial, one would have

13   to estimate that the experimental arm provides a 7.3

14   percent increase in cure rate relative to the active

15   comparator for this study to be positive.

16                  Whereas, in scenario number four, a result

17   would be positive if the experimental arm has a cure rate

18   that is even two percent less than the active comparator,

19   or a relative 10 percent increase in non-cure would still

20   give a positive result.

21                  It is interesting to compare that to the

22   lenient criterion that you would have in scenario number

23   two for non-inferiority, and in this setting one would

24   achieve positivity even if you had a 6 percent lower cure

25   rate, or a 30 percent relative increase in non-cure, would

                             S A G CORP.
     202/797-2525            Washington, D.C.         Fax: 202/797-2525

1    still yield a positive result.

2                   And it is in these settings where positive

3    results are a conclusion, even when you have a meaningful

4    reduction in the post estimate that lead to concerns about

5    bio-creep.     Next slide.

6                   We have heard about bio-creep and the fact

7    that it can arise in repeated non-inferiority trials.

8    Is this a hypothetical that we would have repeated

9    non-inferiority trials?

10                  Well, to give an illustration from last

11   October, the Anti-Viral Drugs Advisory Committee was

12   asked to consider voriconazole as an empiric anti-fungal

13   therapy, and the data that was provided, and the basis

14   for this, was in essence from three generations of

15   studies.

16                  The first generation were control trials of

17   Amphotericin B.     The second generation was looking at

18   the    liposomal   version    of    Amphotericin   B    against

19   Amphotericin B.

20                  And then the third generation was looking

21   at voriconazole against the liposomal version.         Now, what

22   were some of the complexities that this advisory committee

23   had to face?

24                  The first is that there were control trials

25   of the efficacy of amphotericin B, and the Pizzo study

                            S A G CORP.
     202/797-2525            Washington, D.C.       Fax: 202/797-2525

1    and EORTC studies, did yield evidence that amphotericin

2    B yielded a reduced breakthrough infection rate.

3                   However, the studies were very small, not

4    reliable, and so there is considerable variability or

5    uncertainty in what the level of efficacy would be.           Also,

6    this study was done in patients from 15 to 20 years ago.

7                   So there are lots of uncertainties about the

8    relevance of these data, interpretability of these data,

9    in the context of present day studies.

10                  The second generation study, and pardon the

11   typo here, was done by the Mycosis Study Group, an

12   important study looking at ambisome against amphotericin

13   B.

14                  One issue that was very relevant is that the

15   definition of the end-point in this second generation

16   study was somewhat different than the third generation

17   study, so that ambisome had a very different response

18   rate, a much lower success rate in the third generation

19   study, rather than the second generation study.

20                  The success rate was essentially a composite

21   end   point    looking   at   persistent     fever,   death,     and

22   breakthrough fungal infections.             Furthermore, it this

23   third generation study, voriconazole was estimated to

24   have a 6 percent lower success rate, with a lower level

25   of the confidence interval of minus 12 percent.

                              S A G CORP.
     202/797-2525             Washington, D.C.          Fax: 202/797-2525

1                   And guided by the proposed use of a margin

2    of minus 10 percent, and many other considerations, the

3    Anti-Viral Advisory Committee voted unanimously against

4    approval of voriconazole in the setting of empiric

5    anti-fungal therapy.

6                   It is interesting to speculate what decisions

7    would have been if more lenient margins of minus 15 percent

8    had been used, and it is also interesting to speculate

9    that if voriconazole became a standard therapy in use,

10   and there was now a fourth generation study looking at

11   a new empiric anti-fungal therapy, what would be the

12   choice of margin that you would use when comparing against

13   voriconazole that would provide a reliable estimate of

14   efficacy or sense of efficacy of that fourth generation

15   agent.     Next slide.

16                  In closing, just to highlight a couple of

17   the key conclusions.       Non-inferiority trials that use

18   scientifically    rigorous     margins      do   not   necessarily

19   require very large sample sizes, particularly as we were

20   hearing before if we are developing new agents that we

21   are hoping are better, but aren't so confident that they

22   are so much better that we could provide superiority with

23   high power, but are just modestly better.

24                  If they are just modestly better, we can rule

25   out that they are meaningfully worse without having an

                              S A G CORP.
     202/797-2525             Washington, D.C.          Fax: 202/797-2525

1    inordinately large sample size.              And finally as ICH E-10

2    indicated,       the   determination         of   the    margin     in       a

3    non-inferiority trial needs to be based on a wide array

4    of issues, issues that relate to clinical judgment.

5                     What is the clinical importance of losing

6    a given level of efficacy.          That is one key issue, and

7    another key issue is do we expect major important tangible

8    benefits to patients, in terms of safety, tolerability,

9    convenience       of    administration,           resistance,        drug

10   interactions, et cetera, that would allow us to give up

11   some margin or some level of efficacy on the primary

12   end-point.

13                    In addition, there are important statistical

14   issues.        What is in fact a reliable estimate of the

15   efficacy of the active comparator.                      If the active

16   comparator is highly effective, with precisely estimated

17   efficacy, where we have assay sensitivity, where we can

18   believe that that estimate of efficacy in the historical

19   trials reliably predict what the efficacy would be in

20   the non-inferiority trials, then we would be able to with

21   confidence have larger margins.

22                    However, as the ICH guideline indicates, to

23   the extent there are uncertainties in these issues, that

24   should influence the size of margin that we are willing

25   to use.

                               S A G CORP.
     202/797-2525              Washington, D.C.               Fax: 202/797-2525

1                     Finally, the question or finally the comment

2    here    is     the   choice    of    margins     should    be    suitably

3    conservative.          It is certainly the case that we would

4    want to have efficient and timely development of new

5    agents.

6                     But     to   follow      this    concept       of   being

7    conservative, the question arises isn't public health

8    best served by using approval standards that do reliably

9    rule out experimental therapies that do have an inferior

10   benefit to risk profile relative to standard of care.

11   Thanks.

12                    CHAIRMAN RELLER:         Questions for Dr. Fleming?

13    Jim.

14                    DR. LEGGETT:        In terms of the practicality,

15   from the PhRMA and the other speakers, they talked about

16   the impracticality of having a smaller delta.                   What about

17   the factors of having a practicality for an agency such

18   as the FDA when you want to factor in the other things

19   that you talked about?

20                    How do you make the hurdle the same for Drug

21   A, Drug B, Drug C, that come into these same designated

22   indications?         If Drug A is a much better tolerant, and

23   Drug B you can give once a year, and Drug C -- well, how

24   can you bring those in so that there is one hurdle?

25                    DR. FLEMING:       You mean so there is one hurdle

                                   S A G CORP.
     202/797-2525                  Washington, D.C.            Fax: 202/797-2525

1    for all agents in a class, or for agents across classes?

2                   DR. LEGGETT:     How do you determine when a

3    particular drug company wants to present something to

4    the FDA about what kind of numbers they should go for?

5                   DR. FLEMING:       Right.   Well, what I am

6    arguing here is that there are a myriad of issues that

7    need to be considered, and the actual choice of a margin

8    really should be specific to a given agent and a given

9    indication.

10                  And the ideal time for this is in the planning

11   process for the trial, as opposed to after data are

12   available in the trial.       Clearly there is a requirement

13   here for both clinical and statistical judgment, and that

14   clinical judgment I believe needs to take into account

15   the trade-off's between what are the negatives for

16   allowing a loss of a certain level in the primary end

17   point, the primary efficacy end point.

18                  And weighed against what are the perceived

19   or expected benefits that the experimental therapy is

20   going to provide.     And if that experimental therapy is

21   providing      significant      improvements   in      safety,

22   tolerability, resistance to drug interactions, et cetera,

23   one, I believe should have a willingness to allow a

24   somewhat larger margin.

25                  If on the other hand we are looking at a new

                             S A G CORP.
     202/797-2525            Washington, D.C.       Fax: 202/797-2525

1    agent that is not anticipated to be any different, then

2    I am arguing that if in fact the efficacy of that is thought

3    to be modestly better, then you can have a rigorous lower

4    limit, or a lower margin, and have very reasonable sample

5    sizes.

6                        On the other hand, if it isn't any better,

7    then admittedly there would be either the need for a larger

8    sample size, or a risk of a false negative conclusion

9    if the new agent truly isn't any better and doesn't provide

10   any tangible benefits relative to standard of care.

11                       CHAIRMAN RELLER:         Dr. Bell.

12                       DR. BELL:     I am wondering if somebody from

13   the FDA could answer how much leeway does the agency have,

14   either legally or practically, to set different deltas

15   for    different          --    for    the        myriad    of    different

16   considerations, including different drugs for different

17   -- I mean, how uniform do they have to be?

18                       DR. GOLDBERGER:       Actually, our last question

19   this afternoon deals with some of these issues about the

20   factors that ought to be taken into account beyond simply

21   delta in making regulatory decisions.

22                       But to answer your question, products are

23   supposed       to    be   substantial        evidence      of    safety   and

24   efficacy.       There is in fact a lot of flexibility that

25   can be applied.

                                    S A G CORP.
     202/797-2525                   Washington, D.C.             Fax: 202/797-2525

1                   I think one of the things that you have heard

2    this morning, and that you will hear again this afternoon,

3    is we have to be satisfied that the drug is more effective

4    than placebo or no treatment would be in that situation.

5                   I mean, that is sort of the minimum standard.

6     Beyond that, there is just a lot of flexibility.              It

7    would depend if this is the tenth drug for an indication,

8    and it doesn't appear to be any different, in terms of

9    tolerability, activity, pharmokinetics, et cetera.

10                  And there is not a whole lot of reason to

11   necessarily be that flexible.         If on the other hand

12   -- and we have done this in the past, the drug may in

13   fact be less effective than comparator.

14                  And the example that comes to mind is in

15   trials for pneumocystis, where we have in the past

16   approved drugs that were less effective on a mortality

17   end-point than the comparator, because the drugs offered

18   the opportunity to treat patients who could not otherwise

19   be treated by the comparator, which was trimethethum

20   sulfur.

21                  So that represents a lot of the flexibility,

22   and that we can actually approve a drug that may be worse

23   than comparator, with of course including information

24   in labeling to the point where we would expect a reasonably

25   tight delta in a situation where there might be 10 other

                             S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    drugs, and in fact this drug offers no advantage.

2                   CHAIRMAN RELLER:        Dr. Temple.

3                   DR. TEMPLE:     The people who wrote the Food,

4    Drug, and Cosmetic Act, made it very clear that they were

5    not trying to impose a relative effectiveness standard.

6                   So   for   symptomatic         treatments,     we    are

7    interested in whether the drug works at all.                It can be

8    less effective than available therapy as long as it is

9    effective.

10                  But when lack of efficacy has important

11   consequences, safety consequences, then the implications

12   are somewhat different.         And the very reason that you

13   can't do placebo controlled trials in some pneumonia is

14   the reason why you are not willing to accept too much

15   less effectiveness.

16                  And so there is a complex of judgments made

17   about how much evidence you need.            It is worth remembering

18   that when you have a delta, what you are excluding out

19   is the lower bound of a 95 percent confidence interval.

20                  The exclusion of 10 percent, it doesn't mean

21   that you are likely or it is likely that the drug is 10

22   percent worse.      It is more -- I mean, in fact, the point

23   estimates in general would be right on top of each other.

24                  Which means that it is most likely they are

25   fairly close, and the question then becomes how much risk

                               S A G CORP.
     202/797-2525              Washington, D.C.            Fax: 202/797-2525

1    are we willing to accept that the drug is a little bit

2    worse, and as Tom was saying, and that Mark said, you

3    accept more risk if there is some comparative benefit;

4    greater ease of use, less of an important side effect,

5    and those things.

6                        But in general -- and actually this was all

7    described in a Presidential Proclamation about 3 years

8    ago that I have been trying to find.                    But what it said

9    was that relative efficacy is not what we do unless lack

10   of efficacy represents a safety consequence.

11                       And   then      we    consider      it,    and    we    ask

12   sophisticated advisory committees for help in thinking

13   those questions through.

14                       DR. FLEMING:         But just to follow up on what

15   Dr. Temple just said, we talk a lot about margins.                         They

16   are very important issues.                      But it is important to

17   understand for any given margin what does this really

18   mean, the point estimate has to be in order for you to

19   satisfy the criterion of non-inferiority.

20                       And where I worry is when we are choosing

21   margins        so   large    that        the    point   estimate      can    be

22   substantially             less      or         substantially       negative,

23   substantially less favorable for the experimental, versus

24   the active comparator, and still be viewed to be a positive

25   result.

                                      S A G CORP.
     202/797-2525                     Washington, D.C.             Fax: 202/797-2525

1                     That's the setting that leads to this risk

2    of bio-creep.

3                     CHAIRMAN RELLER:      Dr. Bennett.

4                     DR. BENNETT:    Could I ask Dr. Temple about

5    the power function in selecting or estimating sample size?

6     I think I heard Dr. Shlaes said that the examples that

7    the FDA was giving, you are using a power of .8, but that

8    PhRMA    would    find   that   unacceptable   because    of    the

9    possibility of accepting too many ineffective drugs.

10                    Is it true in your experience that PhRMA

11   generally insists on a power of .9 in estimating sample

12   size?

13                    DR. TEMPLE:    Well, Tom has probably helped

14   a lot more companies figuring out what power they should

15   use than we have.

16                    My experience is that in many settings --

17   for example, in different show and trials, that companies

18   often do use a power of something like 80 percent.

19                    And perhaps because they are going to do

20   multiple trials and figure that it will work out all right.

21    But nobody wants to have a substantial chance of losing.


23                    So I think a tendency towards getting the

24   best power you can manage is certainly there.              What I

25   would say we find more -- and this again applies mostly

                               S A G CORP.
     202/797-2525              Washington, D.C.        Fax: 202/797-2525

1    to different show and trials, is an estimate of the effect

2    size that is optimistic.

3                   So if you estimate that you are going to have

4    50 percent effect on something, well, then your power

5    looks terrific, even in a modest sized study.           And where

6    failures occur is where people have been over-optimistic,

7    and not realistic, and haven't done a large enough trial.


9                   In the setting or in these settings, the fear

10   would be that you are going to come out a little bit worse

11   for your point estimate, and therefore, will not be able

12   to exclude the margin that you are talking about.                And

13   I would think companies would worry about that.

14                  CHAIRMAN RELLER:      Dr. Shlaes.

15                  DR. SHLAES:   Just to clarify.       I think what

16   I said was that if you do an 80 percent power at a 10

17   percent delta, and that sort of study, then you have a

18   32 percent chance of falsely concluding inferiority based

19   on these set point considerations.

20                  I think that is what I was trying to say,

21   and so that most companies wouldn't do a 10 percent delta

22   trial powered at 80 percent.

23                  In the old step function, obviously many

24   trials were done at 20 percent, or 15 percent deltas,

25   and then you can tolerate a risk of an 80 percent power

                             S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    because your chance of falsely concluding inferiority

2    is lower.

3                     CHAIRMAN RELLER:      Dr. Glode.

4                     DR. GLODE:    I was just going to mention that

5    I brought with me to this meeting, because I thought it

6    was very informative and Dr. Fleming just mentioned it,

7    which is the article published in the January 24th New

8    England Journal of Medicine, on voriconazole compared

9    to ambisome.

10                    And where in the discussion it mentions

11   exactly the conclusion that you mentioned, that it fails

12   the test of non-inferiority.           However, in the abstract

13   of the article and in the conclusion that is never

14   mentioned, but rather that it is a suitable alternative

15   to amphotericin B preparation.

16                    Now, there is a lot in this article to explain

17   that conclusion, but it still brings up the complexity

18   of selecting the appropriate end point.             Anyway, that

19   is a good example.

20                    CHAIRMAN RELLER:     Thank you.    It is time for

21   lunch.     Let's reconvene promptly at 1:15, and not one

22   o'clock.       We will pick up the time probably during the

23   public hearing.

24                    A reminder.    There are 30 seats set aside

25   in the restaurant reserved for committee members to enable

                               S A G CORP.
     202/797-2525              Washington, D.C.          Fax: 202/797-2525

1   people to get back at 1:15.         And also the discussions

2   about the issues that we have addressed should be kept

3   in the public arena here and not outside of this public

4   arena.     Thank you.

5                  (Whereupon, at 12:22 p.m., a luncheon recess

6   was taken.)




                            S A G CORP.
    202/797-2525            Washington, D.C.       Fax: 202/797-2525

1                   A-F-T-E-R-N-O-O-N       S-E-S-S-I-O-N

2                                                       (1:24 p.m.)

3                    CHAIRMAN RELLER:       I would like to open this

4    afternoon's component of our Advisory Committee Meeting

5    and ask for the Open Public hearing.                We have one

6    scheduled speaker, Dr. Kem Phillips, from Advanced

7    Biologics.      Dr. Phillips.

8                    DR.   PHILLIPS:        I am Kem Phillips from

9    Advanced Biologics.      We, meaning myself and Dr. Michael

10   Corrado, submitted a paper to the committee, and we

11   thought this was going to be a kind of stealth paper that

12   would go under everybody else's radar right into their

13   laps.

14                   But apparently if you do this, it has to get

15   presented, and so to save time from actually having to

16   read this thing to you, I will give a brief presentation.

17    I am just hoping that the lion isn't looking for desert

18   here.

19                   Our    paper     was     titled,   "Should       the

20   Non-Inferiority Margin Vary With the Comparator Rate."

21    There were a lot of good presentations this morning on

22   the clinical issues involved in this issue.

23                   And some of the things that came up were that

24   you would have a difficult time establishing a comparator

25   rate, because for one thing, you might have an increase

                              S A G CORP.
     202/797-2525              Washington, D.C.         Fax: 202/797-2525

1    in resistance.

2                   You might have difficult indications, or you

3    might have new designs.       For example, one design for a

4    drug that only targets GRAM positive organisms.         So all

5    of these lead to an inability to predict response rates.


7                   In some cases, you might have a good rate,

8    a well-established rate , and you wouldn't have a problem.

9     But if you can't, you have a difficulty, and for us

10   statisticians, the question is how to set the sample size.


12                  Drs. Lin, Brittain, and Fleming discussed

13   statistics earlier today, and did an excellent job, and

14   I don't have anything to add to what they have said about

15   a fixed delta method.

16                  But how are you going to set that delta when

17   you can't predict the success rates?       And as they have

18   said several times, if you have a 10 percent delta and

19   a 70 percent underlying rate, you need 330 patients.

20                  And if it is a 90 percent underlying rate,

21   then you need 142.      So that is a big disparity.         The

22   points to consider had one main feature that has been

23   discussed a little bit, and that is that based on observed

24   rates.

25                  You would set the delta to be 10, 15, or 20

                            S A G CORP.
     202/797-2525            Washington, D.C.      Fax: 202/797-2525

1    percent.       Now, one of the things that I don't think did

2    get discussed is this issue of the observed rates.             Any

3    many of us would interpret that as meaning if you observe

4    in your trial, say, an 85 percent rate, then you would

5    in the better of the two arms, then you would use a 15

6    percent delta and so forth.

7                     That leads to sort of an odd test, and among

8    other people, Rohmel, in a '98 Statistics in Medicine

9    paper, outlined some of the problems with that procedure.

10                    The main thing that comes up is this.          We

11   have seen before where we have this discontinuities at

12   80 percent and 90 percent.             So, for example, if you

13   observe a 91 percent success rate in your trial, and maybe

14   you wished it was an 89 percent so you could use the 15

15   percent delta, and various other things happened.

16                    So Rohmel says -- and it discusses a little

17   bit about the possibility of adapting delta to the

18   observed rates, and he says that there were two criteria.


20                    One, there should be good reasons, clinically

21   and statistically, for the non-inferiority margin should

22   vary with the response rate of the standard drug, or the

23   better of the two.

24                    And, number two, the boundary curve of the

25   equivalence margin should be smooth.              The standard

                               S A G CORP.
     202/797-2525              Washington, D.C.       Fax: 202/797-2525

1    approach takes a null hypothesis that the test rate be

2    at least the comparator rate, minus delta, and T is greater

3    than C minus delta.

4                    And in that case, we get these various

5    characteristics that we have seen.                 And you will notice

6    that C minus delta is a linear function of the comparator

7    rates, C.      So why not think of it as being a more general

8    linear function, A times C, plus B.

9                    And if you do that, you can actually establish

10   a valid test, and it doesn't have these problems that

11   you have with the points to consider procedure.

12                   You could even fit that linear function to

13   the points to consider deltas, and get some approximates

14   very    clearly,    but   it    still        has    good     statistical

15   properties.

16                   Another thing you can get out of this test

17   is by setting these parameters A and B appropriately,

18   and you can get something that satisfies something you

19   might call the Lewis criteria.

20                   Rohmel quotes J.A. Lewis as saying that you

21   might adopt the equivalence margin in such a way that

22   the response rate of the better of the two agents that

23   the power of the study remains constant over a wide range

24   of potential response rates, and is thus independent of

25   the later observed response rates.

                               S A G CORP.
     202/797-2525              Washington, D.C.                 Fax: 202/797-2525

1                   And you can set these parameters of this more

2    general test to be able to do that.          So this again is

3    a valid statistical test, and it approximates the points

4    to consider or some other set of criterion that you might

5    like.

6                   But one main problem with it that came up,

7    and I believe that Dr. Fleming mentioned briefly this

8    morning, is that at least if you look at the ITT

9    population, if you get worse success rates, and perhaps

10   intentionally, because you are getting bigger deltas with

11   lower success rates, you might actually increase your

12   probability of showing equivalence bogusly.

13                  But in the evaluable population, you are

14   probably throwing those cases out anyway.              So that

15   probably isn't so much of a problem.         So, anyway, that

16   is all that we wanted to say, that we believe that it

17   might be a good idea to be able to adapt delta to the

18   comparative rates, and that we do have a valid statistical

19   test for doing that.

20                  CHAIRMAN RELLER:       Are there any questions

21   for Dr. Phillips or comments on this approach?

22                  (No audible response.)

23                  CHAIRMAN RELLER:      Were there other persons

24   who wish to present at the open public hearing?         If not,

25   we will move to the FDA's presentations.      First, Dr. John

                             S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    Powers, who is a Medical Officer with the Division of

2    Special Pathogen and Immunologic Drug Products at FDA,

3    who     will      present      a      medical       perspective       on

4    hospital-acquired pneumonia and meningitis.                 John.

5                     DR. POWERS:       Okay.       We're on.   Thank you,

6    Dr. Reller.       This afternoon, we would like to give two

7    presentations, the first of which will be mine, looking

8    at two serious diseases with high mortality rates, and

9    that is acute bacterial meningitis and hospital-acquired

10   pneumonia.

11                    And then after my talk, Dr. Susan Thompson

12   will present some similar information on a less severe

13   disease,       acute   bacterial      exacerbations        of   chronic

14   bronchitis.

15                    And our goal with these two talks is actually

16   to try to give you a framework to hang some of these

17   principles on that we have talked about earlier this

18   morning.

19                    So what I would like to do first off is to

20   reiterate what the definition of delta is, and its various

21   components, and then talk about the impact of deltas in

22   the clinical setting, and what it means to patients.

23                    And then we will go through the selection

24   of delta, or some of the issues in the selection of delta,

25   looking at the two components that were explained this

                                 S A G CORP.
     202/797-2525                Washington, D.C.           Fax: 202/797-2525

1    morning, the delta one, or the historical sensitivity

2    to   drug      effects   in    acute     bacterial    meningitis        and

3    hospital-acquired pneumonia.

4                     And we will look at that by examining some

5    information from the pre-antibiotic era, and from the

6    antibiotic era, to try to get a feel for what is the

7    magnitude of the benefit for antibiotic therapy in these

8    two indications.

9                     And also talk about what are some of the

10   confounders in determining the efficacy of control

11   regimens in these particular diseases.               Then we will talk

12   about the issues of delta two, or that judgment related

13   issue of acceptable loss in these two diseases, by

14   focusing on what are the consequences of less effective

15   therapy in these two diseases.

16                    And then finally finish up with some of the

17   practical issues in selecting deltas.                 It is important

18   I think to start with an idea of what is the purpose of

19   a clinical trial in the first place.

20                    And     a    clinical      trial    is   supposed       to

21   distinguish the effects of a drug from other influences,

22   such as spontaneous change in the course of the disease,

23   placebo effect, or biased observations.

24                    One could ask the question, well, why can't

25   clinicians just do this on their own once the drug gets

                                   S A G CORP.
     202/797-2525                  Washington, D.C.            Fax: 202/797-2525

1    into common usage.          And it actually can be quite difficult

2    for clinicians to make judgments on the efficacy and

3    safety of a drug outside of the setting of a clinical

4    trial, and there are several reasons for this.

5                         In a disease that has a high spontaneous cure

6    rate, if a patient receives antibiotic X or Y, they may

7    get better anyway, regardless of which drug they get,

8    and   it       may    actually   be    impossible    to   discern     an

9    ineffective therapy given that most patients will resolve

10   spontaneously.

11                        Also in diseases that are more serious, and

12   that have high mortality rates, at least in today's realm,

13   most of those people have serious underlying diseases

14   which can be a confounding factor.

15                        So if a patient dies on therapy, is that

16   because of their underlying disease, or was it because

17   of progression of that infectious disease, and that can

18   be quite difficult to tell, even with autopsy data that

19   can sometimes be hard to tell what the patient died from.

20                        And finally it can also be very difficult

21   to tell what the safety of a drug is compared to another

22   drug just in the clinical realm.             If you give your patient

23   a particular drug, and they get a rash, that is pretty

24   clear.

25                        But the real question is how does that compare

                                   S A G CORP.
     202/797-2525                  Washington, D.C.         Fax: 202/797-2525

1    to another therapy, and what is the rate of rash in a

2    controlled regime, and it is really hard to do that outside

3    of the setting of a clinical trial.

4                     And just to add a point.              This morning we

5    were talking about antibiotics and their ability to

6    eradicate      bacteria.        Some     would     also     argue     that

7    antibiotics also have other effects.

8                     And   as   Dr.     McCracken      mentioned,          some

9    antibiotics have anti-inflammatory effects, or sometimes

10   they go in the opposite direction.             And there is actually

11   some in vitro data with amphotericin B that says that

12   if you incubate amphotericin B with white cells, that

13   it releases massive amounts of tumor necrosis factor.

14                    Whether this has an impact on clinical

15   outcomes or not really isn't clear, and hasn't been

16   studied.       The other reason for clinical trials is that

17   sometimes we see a result that just wouldn't be intuitive

18   based on what we would think going into the trial.

19                    And probably one of the best examples of this

20   is    clarithromycin        studied       in     the    treatment        of

21   disseminated      microbacterium         avian    disease        in    AIDS

22   patients.       And in that trial, there were three doses

23   tested; a low, an intermediate, and a high dose.

24                    And in that trial the low dose had no effect

25   on eradication of MAC.           The moderate dose did have an

                                 S A G CORP.
     202/797-2525                Washington, D.C.              Fax: 202/797-2525

1    effect, and actually the mortality was higher in the high

2    dose than it was in the moderate dose.

3                   And one would never have guessed that going

4    into the trial based on the pre-clinical data.                         So

5    sometimes we see results from clinical trials that we

6    just wouldn't predict from some of the preclinical

7    information.

8                   And in non-inferiority trials -- and again,

9    Dr. Fleming said this as well -- we are attempting to

10   prove that the test drug is not inferior to the control

11   drug by some margin, and we can't prove that two drugs

12   are absolutely statistically identical in efficacy.

13                  So    we   need   some        way   to   estimate      the

14   variability around the difference between those two

15   treatments.    And the way we do this is again looking at

16   the non-inferiority margin or delta, which we are defining

17   as the maximum degree of inferiority of the test drug,

18   compared to the control drug the trial attempts to exclude

19   statistically.

20                  And   again    this     is     specified      prior     to

21   initiation of the trial.          Once the trial is over, we

22   calculate the difference in the point estimates of the

23   efficacy of the test agent, minus the control agent, and

24   again I am using the convention that Drs. Brittain, Lin,

25   and Fleming used.

                               S A G CORP.
     202/797-2525              Washington, D.C.              Fax: 202/797-2525

1                   Dr. Temple used the opposite of this, but

2    I am using the test agent, minus the control agent.             And

3    here on this slide, we can see just as an example, I am

4    showing that the point estimate of the difference of the

5    test minus the control agent is minus 8 percent.

6                   We then calculate 95 percent confidence

7    intervals around the difference in the point estimate,

8    which gives us some idea of the variability around this

9    estimate.

10                  And then we compare the lower bound of the

11   95 percent confidence interval to this pre-specified

12   non-inferiority margin, which in this example is minus

13   15 percent.

14                  So again just to reiterate what you heard

15   this morning, since we are all sleepy after lunch, delta-1

16   is a conservative estimate of the advantage of active

17   control over placebo that is based on data.

18                  Delta-2 is the largest clinically acceptable

19   difference     between    the      active   control     and     the

20   experimental drug, which is based on judgment.          And again

21   that    judgment   is    in-turn    based   on   what    are    the

22   consequences to patrons of treatment failure.

23                  So overall selecting a delta for the clinical

24   trial, if the delta-1 is very large, or in other words,

25   is there is a huge benefit of drug treatment over placebo,

                              S A G CORP.
     202/797-2525             Washington, D.C.         Fax: 202/797-2525

1    then what really matters is selecting the delta based

2    on the delta-2.

3                   So if we then go on to talk about delta-1,

4    which is historically-based data, we can ask the question

5    do we really know what we think we know about the

6    historical information.

7                   And again the important point to remember

8    here is that it is not whether an antibiotic actually

9    helps patients or not.     It is what is the magnitude of

10   that benefit, and when one actually goes through the

11   literature, trying to tack a number on to this, it can

12   be actually quite a daunting task, I can tell you, having

13   spent hours in the library looking this stuff up.

14                  So one of the problems is that for some

15   diseases that we deal with, there is no data from the

16   pre-antibiotic era.    These are really diseases of modern

17   medical care in some cases.

18                  The second thing is that there has been

19   changes in the resistance patterns of the common organisms

20   causing these diseases, and also the epidemiology of the

21   disease itself.

22                  Thirdly, there can be differing response

23   rates in various sub-populations with the disease.

24   Fourthly, there can be changes in the practice of

25   medicine, or supportive care, of patients with that

                            S A G CORP.
     202/797-2525           Washington, D.C.     Fax: 202/797-2525

1    disease.

2                     And then also there can be problems in

3    defining patients who actually have bacterial infections,

4    versus either non-bacterial causes of the same kind of

5    infection, or non-infectious causes that may mimic that

6    disease.

7                     And finally a point that was brought up

8    several times this morning, is that sometimes we use

9    different definitions of success and failure in our

10   current        trials,   compared      to     the   end   point      in

11   pre-antibiotic trials were, which is mostly mortality

12   for the main part.

13                    The delta-2 is the judgment based acceptable

14   loss relative to current therapy.              In an ideal world,

15   one could make the assumption that for more severe

16   diseases one would like to see a smaller delta, because

17   the consequence of treatment failure in those severe

18   diseases could be increased morbidity and mortality to

19   patients.

20                    On the other hand, in less severe diseases,

21   one would be tempted to accept a larger delta because

22   even though there may be greater loss relative to current

23   therapy, that may not translate into mortality for

24   patients, although it may translate into more morbidity

25   and discomfort for patients.

                               S A G CORP.
     202/797-2525               Washington, D.C.           Fax: 202/797-2525

1                     But unfortunately we don't live in an ideal

2    world,     and   there   are   practicalities      of   performing

3    clinical trials that we need to take into account when

4    forming our judgments about what is an acceptable loss.

5                     And this is what we are going to do for you

6    this afternoon hopefully, is that we are going to take

7    these three diseases, and try to go through them, and

8    show you some of the information that you can hang this

9    around.

10                    The first that we will talk about is acute

11   bacterial meningitis.          Well, the delta-1 for acute

12   bacterial meningitis, the magnitude of advantage over

13   placebo is well known in acute bacterial meningitis.

14                    There is data from the pre-antibiotic era,

15   and it is a very large benefit.          Therefore, the decision

16   should be based on that acceptable loss, and taking into

17   account the difficulty in doing trials, as well as the

18   fact that we may increase mortality by accepting drugs

19   that are less effective.

20                    The second indication that we will talk about

21   is hospital-acquired pneumonia.              And actually this is

22   a disease more of the modern era, where the magnitude

23   of the advantage over placebo is not as clear, and when

24   you actually try to hang a number on this, it becomes

25   quite difficult.

                               S A G CORP.
     202/797-2525              Washington, D.C.          Fax: 202/797-2525

1                   And then again you are still left with that

2    decision on what is an acceptable loss.        And then finally

3    after me, Dr. Thompson will go over acute bacterial

4    exacerbations of chronic bronchitis, where the advantage

5    over placebo is unclear, and may in fact be quite small.

6                   Or it may be different, depending upon which

7    subpopulation you are dealing with, and the decision on

8    acceptable loss here is not as critical, again because

9    we are not dealing with high mortality rates.

10                  So   let's   start     off    looking   at    these

11   components of delta for meningitis and hospital-acquired

12   pneumonia, and I have divided this up by asking several

13   important questions for each of the delta-1 and the

14   delta-2 components.

15                  For delta-1, one can ask the important

16   question of what is the magnitude of benefit of any

17   antibiotic therapy over placebo.            The second question

18   is, is the benefit of antimicrobial therapy in current

19   trials measured in the same way as in the original trials

20   showing that benefit.

21                  And the third question is, is the magnitude

22   of benefit of therapy over placebo, or the delta-1, large

23   enough that it should not effect the selection of the

24   overall delta for the clinical trial.

25                  In other words, we can skip the delta-1

                             S A G CORP.
     202/797-2525            Washington, D.C.          Fax: 202/797-2525

1    altogether and make a decision on the delta for the trial

2    based on delta-2.        The important question for delta-2

3    is what is an acceptable loss of efficacy compared to

4    accepted therapy in a serious disease, and there are two

5    sides to this coin.

6                    The first is the scientific considerations

7    of what happens to patients who fail treatment in various

8    patient        subsets     with     meningitis        or     hospital

9    acquired-pneumonia.

10                   And then what you heard a lot about this

11   morning are the practical considerations of the effects

12   of changing the delta on sample size as the efficacy rate

13   changes.

14                   Well,    let's      look      at   acute   bacterial

15   meningitis first, and try to figure out some information

16   about delta one, or the historical sensitivity to drug

17   effects in this disease.

18                   Clearly,    acute     bacterial     meningitis      was

19   highly lethal in the pre-antibiotic era.             The most common

20   organism before antibiotics was actually meningococcal

21   disease, which occurred in large outbreaks.

22                   And the overall mortality in these outbreaks

23   was somewhere between 70 and 90 percent without specific

24   therapy, and there are articles about the 1905-1906

25   meningococcal outbreak in New York City, which clearly

                                S A G CORP.
     202/797-2525               Washington, D.C.           Fax: 202/797-2525

1    defined this number for us.

2                   The other interesting point is that those

3    outbreaks occurred in mostly previously healthy young

4    people, who were in crowded conditions, and who then went

5    on to get ill.    So they did not have underlying serious

6    diseases.

7                   When       Flexner          first           studied

8    anti-meningococcal serum in this paper published in 1913,

9    it decreased the mortality in meningococcal meningitis

10   from 70 percent to 30 percent.        So, clearly a very large

11   mortality benefit, even with meningococcal serum.

12                  And then finally Schwenker published his

13   paper in 1937, which gave sulfanilamide, given both

14   subcutaneously and intrathecally to 11 patients, and this

15   reduced the mortality to 10 percent.

16                  And in this series, he treated 11 patients,

17   and 9 of those 11 patients survived.       One of the patients

18   who did die actually had bacterial eradication from his

19   spinal fluid, but went on to pass away anyway.

20                  What are some of the problems with this

21   historical data?      Well, we use different end points in

22   current clinical trials, and although mortality is one

23   of the end points that we still look at, we can argue

24   that sometimes that is not that high, and doesn't drive

25   the overall end points.

                             S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1                     For instance, in the trovafloxacin study that

2    was published in Pediatric and Infectious Diseases that

3    Dr. McCracken talked about this morning, the mortality

4    in each group was 2 percent and 3 percent, and clearly

5    different than what we saw in the pre-antibiotic era.

6                     So some of the end points that we look at

7    here, in addition to mortality, are developmental,

8    neurologic, and audiologic sequelae.                It is hard to get

9    a handle on what the effect of antibiotics is on these,

10   because if patients didn't get treated, they die.                       So

11   it is hard to tell.

12                    There is also different epidemiology today

13   than we saw in the past, and today pneumococcal meningitis

14   is the most common form of bacterial meningitis in the

15   United States, and that is even different from 10 years

16   ago in this country.

17                    And finally there are different populations.

18    In this study that was published a few years ago in the

19   New   England        Journal     of    Medicine,    it   compared      the

20   epidemiology of acute bacterial meningitis in 1995, to

21   the epidemiology in 1986, and showed that in 1986 that

22   the average age of a meningitis patient in the U.S. was

23   15 months.

24                    And the average age of a meningitis patient

25   in    1995     was   25   years,       a   huge   difference     in    the

                                 S A G CORP.
     202/797-2525                   Washington, D.C.          Fax: 202/797-2525

1    epidemiology, even over a short span of time.                 Now, let's

2    switch gears, and try to look at the historical data for

3    hospital acquired pneumonia.

4                        It is a much more difficult task, because

5    the clinical entity of hospital acquired pneumonia was

6    not described in the pre-antibiotic era.                     If we tried

7    to   look      at    some    of    the     organisms     implicated      in

8    hospital-acquired pneumonia, even though they aren't

9    acquired in the hospital in this pre-antibiotic data,

10   we can see that in the influenza outbreak in 1918, there

11   were a number of cases of post-influenza Staph aureus

12   pneumonia.

13                       And in one report, there were only two

14   spontaneous cures out of 151 cases on a military base

15   with Staph aureus pneumonia.             So, clearly a highly lethal

16   disease.

17                       There   were    very        few    reports    in    the

18   pre-antibiotic area of Gram-negative pneumoniaes, and

19   again part of the problem with these reports though is

20   how certain are we of the microbiologic diagnosis in these

21   case reports.

22                       So   really    there    is    no   way   to   compare

23   antibiotic therapy to placebo for hospital acquired

24   pneumonia, because these studies just don't exist.                       So

25   what we are left doing is trying to extrapolate data from

                                  S A G CORP.
     202/797-2525                 Washington, D.C.             Fax: 202/797-2525

1    the antibiotic era to see if we can find what the placebo

2    rate would be.

3                        Well, one way to try to do this is to compare

4    patients that get appropriate antibiotic therapy to

5    inappropriate antibiotic therapy, and I am going to

6    contrast these two studies to show you how difficult a

7    task this actually can be.

8                        If we look at this study by Celis that was

9    published in Chest, they looked at all-cause mortality

10   in patients that received appropriate antibiotics, versus

11   those who received inappropriate antibiotics.

12                       In this trial, appropriate antibiotics were

13   defined        as    an   organism       that     was   sensitive       to    the

14   antibiotics          that    the    patient       received.        And       again

15   obviously           you     can't    randomize          patients        to    get

16   inappropriate therapy, and so this is an observational

17   study.

18                       The all-cause mortality rate in patients that

19   received inappropriate therapy was 91.6 percent, and the

20   all-cause mortality in patients that received appropriate

21   therapy was 30.5 percent.                 So a 60 percent difference

22   between appropriate and inappropriate therapy.

23                       There is a lot of problems with this data,

24   however.            The   first     is    that     obviously       it    is    an

25   observational study, and the second is that the number

                                    S A G CORP.
     202/797-2525                   Washington, D.C.               Fax: 202/797-2525

1    of patients that received inappropriate therapy was very

2    small in this particular trial.

3                     So if we attempt to look at another study

4    that was done almost 10 years later, published by Alvarez

5    and Lerman in Intensive Care Medicine in 1996.                 These

6    people looked at this question in a slightly different

7    way, but it tremendously changes the numbers.

8                     They again looked at inappropriate versus

9    appropriate antibiotics, but this time they defined

10   inappropriate therapy as lack of clinical improvement,

11   or an organism that was not sensitive to the antibiotic

12   that the patient received.

13                    So there was more than one way to define

14   appropriate, versus inappropriate.             They also looked at

15   attributable mortality.        In other words, assuming that

16   the patient died, they died of pneumonia.

17                    Now, how one determines this isn't clear from

18   this paper, and it is not clear in any case how one would

19   decide what the patient died of.             So in this case, they

20   looked at the attributable mortality to hospital-acquired

21   pneumonia.

22                    And comparing appropriate to inappropriate

23   therapy.       If the patients received appropriate therapy,

24   the mortality rate was 16.2 percent, and if they received

25   inappropriate therapy, the mortality rate was 24.7

                               S A G CORP.
     202/797-2525              Washington, D.C.          Fax: 202/797-2525

1    percent.

2                      So only about an 8-1/2 percent difference

3    here.          Now,   again,     there     are    differences    in   the

4    populations between these two studies.                 The Celis study

5    enrolled only mechanically ventilated patients in the

6    ICU.

7                      The Alvarez and Lerma study enrolled patients

8    in the ICU, 60 percent of whom were on mechanical

9    ventilation, but the other 40 percent were not.                       This

10   is the kind of data that you have to deal with when you

11   are trying to decide what is the effect of antibiotics.

12                     And this is as good as it gets.               So it is

13   very difficult to find out.            Again, there is also problems

14   with this historical data.              There is a great difficulty

15   in     the     clinical   diagnosis          of   hospital-     acquired

16   pneumonia, and several studies that look at this show

17   that clinicians are only correct in their diagnosis of

18   hospital-acquired pneumonia, at least based on autopsy

19   studies, about 50 percent of the time.

20                     The problem with this is that patients get

21   enrolled in these studies that don't have the disease.

22    So you can't expect the antibiotics to have an effect

23   on someone that doesn't have an infection.

24                     Also, there has been a change in nosocomial

25   organisms over time, with a shift from GRAM-positive

                                    S A G CORP.
     202/797-2525                   Washington, D.C.         Fax: 202/797-2525

1    organisms back in the 1950s, with the introduction of

2    positive pressure ventilation, to GRAM-negatives and back

3    to GRAM-positives again today.

4                    There is also very different outcomes in

5    various patient populations.                 The mortality rate in

6    mechanically ventilated patients is much higher than that

7    in, say, ward patients or ICU patients who are not

8    ventilated.

9                    And again there is the problem of how do we

10   attribute      the    death     to    pneumonia   versus   all-cause

11   mortality, and even at autopsy, it can be difficult to

12   discern this information.

13                   And then finally we use clinical end-points

14   other than mortality in our current clinical trials;

15   things such as normalization of the white blood cell

16   count, and resolution of a chest radiograph, or resolution

17   of fever.

18                   So if we then go back to our original

19   questions, and again shifting gears back again to acute

20   bacterial meningitis, let's see if we can answer some

21   of these questions.

22                   For delta-1 for acute bacterial meningitis,

23   what is the magnitude of benefit of antibiotic therapy

24   over placebo.        Well, it appears that this is pretty clear,

25   and it is as large as 60 to 80 percent mortality benefit.

                                S A G CORP.
     202/797-2525                  Washington, D.C.        Fax: 202/797-2525


2                    But the magnitude of benefit on clinical

3    parameters,     such   as     auditory,         hearing,   neurologic,

4    developmental losses, is not as clear.                Is the benefit

5    of antimicrobial therapy in current trials measured in

6    the same way as in the original trials?

7                    Well, yes, and no.             We still use mortality

8    as an end-point, but we do use the other end-points of

9    auditory and neurologic developmental losses as well.

10                   And, thirdly, is the magnitude of benefit

11   of therapy over placebo large enough that it should not

12   affect the selection of the overall delta for a trial.

13    And the answer here appears to be yes, because again

14   the magnitude of the benefit is so large that you can

15   select the delta based on the considerations about

16   clinical loss.

17                   How about for hospital-acquired pneumonia

18   if we attempt to answer these same three questions.                  What

19   is the magnitude of benefit of antibiotic therapy over

20   placebo?       Much harder to answer than for bacterial

21   meningitis.

22                   And based on the two trials that I have

23   presented to you, the benefit can be anywhere from

24   8-1/2 percent to 60 percent, depending upon how, and in

25   whom this benefit is measured.

                                 S A G CORP.
     202/797-2525                Washington, D.C.            Fax: 202/797-2525

1                   And it is very unclear what the benefit of

2    antibiotics is on a resolution of clinical parameters,

3    such as fever, white count, and chest radiograph.                   The

4    second question is the benefit of antimicrobial therapy

5    in current trials measured in the same way as in the

6    original trials showing benefit?               Again, the answer is

7    yes and no.

8                   We still look at mortality, but again we are

9    looking at the resolution of those clinical parameters,

10   as well as part of the primary end points.

11   And then finally is the magnitude of benefit of therapy

12   over placebo large enough that it should not effect the

13   selection of the overall delta for the trial.

14                  Well, this is one of the things that we want

15   the Committee's help on today.                Given the problems in

16   looking at this trials, how is one to decide what the

17   acceptable     loss   is     given     some     of   the   practical

18   considerations as well.

19                  The other point that I want to make about

20   hospital-acquired pneumonia referable to some of the

21   discussions that went on this morning, is that there is

22   a clear difference about what the bacteriology means in

23   a   disease    like   acute     bacterial       meningitis,     versus

24   hospital-acquired pneumonia.

25                  And we talked a little bit this morning about

                                S A G CORP.
     202/797-2525               Washington, D.C.           Fax: 202/797-2525

1    using so-called hard end points of the microbiology of

2    some of these diseases.          Well, that may be appropriate

3    for acute bacterial meningitis, where you have sterile

4    body fluids, such as cerebral spinal fluid, where you

5    can measure an effect of the antibiotic.

6                   That     becomes        very      problematic        for

7    hospital-acquired pneumonia, and in fact a number of the

8    other respiratory indications, where the organism that

9    you isolate in the sputum may have absolutely nothing

10   to do with the patient's clinical course.

11                  And the flip side of that is that you can

12   find organisms in the patient's blood stream when their

13   sputum sterile.       So the microbiology in a disease like

14   hospital-acquired pneumonia becomes very difficult to

15   interpret.

16                  And we would like to hear what the committee

17   has to say about that as well.                Finally, for delta-2,

18   we need to talk about both the scientific and the practical

19   considerations of selecting delta-2.               Well, again this

20   is based on the consequences to patients of treatment

21   failure.

22                  In meningitis, there is a clear consequence

23   of treatment failure, and that is death. So there is a

24   clear mortality benefit of antibiotic therapy, and the

25   morbidity      here   is    developmental,        neurologic,       and

                                S A G CORP.
     202/797-2525               Washington, D.C.           Fax: 202/797-2525

1    audiologic sequelae.

2                   And again it is unclear what the magnitude

3    of benefit of antibiotics for those things actually is.

4     For hospital-acquired pneumonia, well, while there may

5    be a mortality difference as one of the consequences of

6    failure, although again the magnitude of that benefit

7    varies depending upon how and in whom that is measured.

8                   And also there can be a morbidity increase,

9    and clearly there are studies which show that patients

10   who      do    not   get     treated       appropriately       for

11   hospital-acquired pneumonia have an increased cost of

12   their hospital stay, and an increased duration of their

13   hospital stay as well.

14                  But again although we have that economic

15   information, there really is a lack of information on

16   the effect on the rate of clinical resolution of things

17   like the white count fever and chest radiograph.

18                  So finally, and you have heard a lot about

19   this this morning, and so I won't spend much time talking

20   about it, are the practical issues involved in selecting

21   delta.

22                  And the effect of the success rate on delta

23   you have heard a lot about this morning.          But there is

24   also something that goes into this beyond just sheer

25   economics, and that is how many patients actually have

                            S A G CORP.
     202/797-2525            Washington, D.C.         Fax: 202/797-2525

1    the disease.

2                   So we need to look at the epidemiology of

3    the   disease,   the   limitations         of   the   inclusion     and

4    exclusion criteria of a trial, and the inability of

5    patients to continue on randomized therapy in studies

6    of very severe diseases, where patients may not make it

7    to the end of treatment.

8                   You have seen this slide a couple of times

9    today, and I am not going to go through it in detail,

10   and I will just show you that what I really want to point

11   out is that you can see the relationship between delta

12   and success rate is not linear.

13                  As you tighten the delta the number of

14   patients required in a trial goes up rather steeply.

15   So let's talk about he epidemiology of the diseases and

16   what we know.

17                  And you heard a little bit about this from

18   Dr. McCracken this morning, and again this is based on

19   this information obtained from 248 cases of meningitis

20   acquired by the CDC and published in this New England

21   Journal paper in 1997 from data from 1995.

22                  Well, what we used to see in 1986 was that

23   haemophilus influenzae was the number one cause of

24   bacterial meningitis, and it occurred in children at an

25   average age of 15 months.

                             S A G CORP.
     202/797-2525            Washington, D.C.              Fax: 202/797-2525

1                   What   we    see    now    is   that   streptococcus

2    pneumoniae is the most common organism at one 1.1 cases

3    per     hundred-thousand          patients,     and     haemophilus

4    influenzae has dropped all the way down into a tie for

5    fourth place with listerial meningitis.

6                   Why is this important?           This is important

7    because the case fatality rates are obviously going to

8    influence the cure rate in the disease, and this varies

9    by organism.

10                  Haemophilus influenzae has a lower case

11   fatality rate than disease caused by streptococcus

12   pneumoniae.    If one were to do a trial in the United States

13   today, you would most likely get more streptococcus

14   pneumoniae isolates, but that would also mean that the

15   mortality would be higher.

16                  So if you compared a trial done today with

17   a trial done in the 1980s, the overall cure rate may be

18   lower now because you are having more strep pneumo cases

19   than you did haemophilus influenzae.

20                  This paper also estimated the number of cases

21   in the United States in 1986 and 1995 of acute bacterial

22   meningitis.    And it was estimated that there were about

23   13,000 cases in 1986, and now we are down to less than

24   6,000 cases in 1995.

25                  And Dr. McCracken mentioned this morning that

                                S A G CORP.
     202/797-2525               Washington, D.C.           Fax: 202/797-2525

1    another organism may come along to replace this, and this

2    study actually looked at the difference here, and it

3    really is due to the huge drop in haemophilus influenzae

4    Type B disease, and it has not been replaced by something

5    else, at least not to this point.

6                   So we have a shrinking number of cases in

7    this country as well.      Switching gears once again back

8    to   hospital-acquired     pneumonia.       Well,       just     like

9    everything else with this disease, it is unclear what

10   the epidemiology of this disease is.                It is not a

11   reportable illness.

12                  The    National         Nosocomial        Infection

13   Surveillance data estimates that there is about 250,000

14   cases per year in the United States, but this uses a

15   clinical definition of hospital acquired pneumonia.

16                  And even though hospital acquired pneumonia

17   may account for one percent of all patients entering the

18   hospital, and it is the second most common nosocomial

19   infection after urinary tract infections, and the most

20   common infection in the ICU, it still ends up being

21   relatively uncommon compared to some other diseases.

22                  And again these may not be entirely accurate,

23   because I pulled these from a number of different sources.

24    But I just wanted to put these as a framework for you

25   to see how things fall out.

                             S A G CORP.
     202/797-2525            Washington, D.C.            Fax: 202/797-2525

1                     Acute otitis media, 26 million cases a year;

2    acute          sinusitis,         23       million;         and       then

3    tonsillitis/pharyngitis, 21 million; community-acquired

4    pneumonia, about 4 million; and then we drop off down

5    here to 250,000 cases of hospital-acquired pneumonia;

6    10,000 cases of acute bacterial meningitis; and somewhere

7    less than that for acute bacterial endocarditis.

8                     So still these things are relatively uncommon

9    compared to some of the other ones.               Getting back to that

10   point about using bacteriologic end points.                    Again, it

11   depends upon what indication you are talking about.

12                    It may work for acute otitis media, and won't

13   work for acute sinusitis, because we don't get puncture

14   studies most of the time, although we do on occasion.

15                    It   won't       work      for    community-acquired

16   pneumonia, and it won't work for hospital-acquired

17   pneumonia.         But   it     may    work      for   acute   bacterial

18   meningitis.

19                    So it depends upon the indication whether

20   bacteriology is helpful to us or not.                    So some other

21   practical points.        The success rate in recent hospital

22   acquired pneumonia trials with piperacillin, tazobactam,

23   linezolid, ciprofloxacin, or trovafloxacin, have all been

24   in the 50 to 70 percent range.

25                    If one uses a smaller delta for those trials,

                                   S A G CORP.
     202/797-2525                  Washington, D.C.            Fax: 202/797-2525

1    the delta used in those trials was 20 percent by the way.

2     But if one would use a smaller delta than that, the sample

3    size would go up.

4                   However, the downside of accepting a larger

5    delta is that theoretically a new drug could then be as

6    much as 20 percent less effective than the comparator.

7     And if we are talking about a drug that already starts

8    off with a 50 percent cure rate, we are down to possibly

9    accepting a drug with a 30 percent cure rate.

10                  The other problem is that almost half of the

11   patients don't complete the trials, and you need to take

12   that into account when looking at the sample size.

13                  So if we just look again at the left side

14   of this graph, which you have seen many times, if we go

15   from a 20 percent delta, we go from a trial that needs

16   99 patients per arm -- and again this is assuming 80

17   percent power.

18                  But if we tighten it all the way down to a

19   5 percent delta, we are talking about fifteen hundred

20   patients per arm, or 3,000 patients in the study.          But

21   that is before you figure out that half of those people

22   drop out of the trial.      So you are talking about 6,000

23   patients per study here.

24                  So then some of the things that we need to

25   take into account for delta-2 to answer that question

                            S A G CORP.
     202/797-2525            Washington, D.C.     Fax: 202/797-2525

1    of what is an acceptable loss of efficacy compared to

2    accepted therapy in a serious disease.

3                   Well, the serious nature of meningitis and

4    hospital-acquired pneumonia would seem to call for a

5    selection of small deltas.         However, as we have seen,

6    smaller deltas would result in a larger sample size of

7    the trials, and one of the things that we would ask the

8    committee about today is whether this is practical given

9    what we know.

10                  But we need to balance this risk of accepting

11   drugs, which may be 20 percent less effective than

12   currently approved therapy.       And again if we are talking

13   about a 50 or 60 percent cure rate, 20 percent less than

14   that is a 30 or 40 percent cure rate.

15                  So the dilemma that we are left with here

16   today is to balance this risk to patients of accepting

17   larger deltas, especially in more severe diseases, versus

18   those realities of performing clinical trials.

19                  At this point, I will turn it over to Dr.

20   Susan Thompson, and she will talk to you about acute

21   bacterial exacerbations of chronic bronchitis.

22                  DR. THOMPSON:     Good afternoon.   I am going

23   to be speaking with you today about the selection of delta

24   in clinical trials of antimicrobial therapy for the

25   indication of acute exacerbation of chronic bronchitis.

                             S A G CORP.
     202/797-2525            Washington, D.C.       Fax: 202/797-2525


2                   The outline of what we are going to be talking

3    about today is given here.        First of all, we will give

4    a definition of the scope of the problem, and discuss

5    the selection of deltas specifically for AECB trials.

6                   Then we will spend most of our time reviewing

7    the trials available in the literature which our placebo

8    controlled for the indication of AECB, and discuss some

9    of the confounding issues and interpretation of those

10   trials.

11                  And we will give you some conclusions and

12   list for you what we feel are unresolved issues, and

13   alternatives    for   future    AECB       trials.       There     are

14   approximately 12 million cases of chronic bronchitis per

15   year in the United States.

16                  And it is the most common category of chronic

17   obstructive pulmonary disease.             Most cases of chronic

18   bronchitis are due to tobacco use, and most studies put

19   it in the range of 85 to 90 percent.             A few cases are

20   due to environmental pollutants, or such genetic factors

21   as alpha-1 antitrypsin deficiency.

22                  It is important to recall that AECB is a

23   distinct clinical entity from acute bronchitis.                  Acute

24   bronchitis is usually defined as sputum production in

25   the absence of underlying lung disease, and the vast

                             S A G CORP.
     202/797-2525            Washington, D.C.             Fax: 202/797-2525

1    majority of these cases have viral etiology as the cause.

2                     The Division of Anti-Infectives no longer

3    recognizes acute bronchitis as an indication for which

4    new drugs can apply.              Acute exacerbation of chronic

5    bronchitis accounts for 5 to 10 percent of all antibiotic

6    prescriptions in the United States.

7                     Currently, 17 antibiotics, plus or minus one,

8    carry the indication of acute exacerbation of chronic

9    bronchitis      and    are    labeled,        and     were     approved      via

10   non-inferiority trials.

11                    Some of the older antibiotics carry broader

12   indications which were granted at those times, including

13   either upper or lower respiratory tract infections.

14                    I have borrowed this slide from the CDC

15   basically to just give you an idea of the proportion which

16   bronchitis       represents        in    outpatient            antimicrobial

17   therapy usage in the United States.

18                    This slide is from 1992, although i suspect

19   that the proportions have not changed.                       Bronchitis, as

20   you    can     see,    represents       16.3         million     courses      of

21   antibiotics in the year of 1992, a significant proportion.

22                    It is important to note this slide was

23   presented      in     the    context     of      a    discussion       of    the

24   antimicrobial resistance, and clearly some of those

25   prescriptions that were written for bronchitis, as well

                                   S A G CORP.
     202/797-2525                  Washington, D.C.                 Fax: 202/797-2525

1    as some of these other diagnoses which are given for

2    outpatient or for respiratory infections, are given

3    sometimes      for   indications       which      don't       require

4    antibiotics.

5                   Moving then into a definition of acute

6    exacerbation of bronchitis, a fairly standard definition

7    of   chronic   bronchitis    itself        is   cough   and   sputum

8    production on most days for greater or equal to three

9    months in two consecutive years.

10                  And acute exacerbation of chronic bronchitis

11   is some combination of worsening dyspnea, increased

12   sputum volume, and/or increase in sputum purulence.

13                  The etiology is most commonly nontypable H.

14   flu, which usually encompasses 50 to 60 percent of the

15   isolates in most studies.        M. catarrhalis is 15 to 20

16   percent, and Strep pneumo is 15 to 20 percent.                    The

17   smaller number of atypicals has been found in various

18   studies.

19                  Moving then specifically to the issue of

20   selection of delta for clinical trials, I will reiterate

21   what you have heard many times today already.

22                  Delta-1 is the smallest effect size, if any,

23   that active drugs would be reliably expected to have

24   compared with placebo, and we will spend the majority

25   of our time on that for this indication.

                             S A G CORP.
     202/797-2525            Washington, D.C.            Fax: 202/797-2525

1                    Delta-2 is the largest clinically acceptable

2    lots in efficacy between the experimental drugs and the

3    active drugs, with the smaller of these two values

4    representing delta.

5                    For acute exacerbation of chronic bronchitis

6    then,     specifically    the     determination     of    delta-1

7    represents the estimation of the benefit, if any, of

8    active control over placebo.

9                    The determination of delta-2 for AECB is in

10   a sense relatively less pressing, in that AECB has a very

11   low mortality and morbidity, and for this indication than,

12   delta-2 is relatively large and certainly greater than

13   20 percent.

14                   Thus, for AECB, the smaller of the two values,

15   delta-1 would represent the delta for the studies.

16   Actually, I should have entitled this slide "Previous

17   FDA Guidance for AECB."

18                   The points to consider you are probably all

19   aware of.      From 1990, two recommended trials for AECB,

20   or one if the drug was submitted for CAP or HAP.                The

21   organisms we have already mentioned.

22                   And 10 to 20 percent was the usual delta for

23   AECB trials based on the efficacy rates which were usually

24   found.     The approach then to determine delta-1 for AECB

25   is essentially to review the results of the placebo

                              S A G CORP.
     202/797-2525             Washington, D.C.         Fax: 202/797-2525

1    controlled trials that are available to us from the

2    literature in an attempt to determine

3    delta-1.

4                   The two points that I think are important

5    to remember during our subsequent discussion is that,

6    first of all, in the past 40 years, less than eleven

7    hundred patients have been enrolled in randomized placebo

8    controlled trials of the antibiotic treatment of AECBs,

9    and none of those trials were of identical design.

10                  The second point that I want you to remember

11   is actually a list of caveats that many of these trials

12   share.     First of all is the uncertainty in the definition

13   of acute exacerbation.       The second and very important

14   caveat is the lack of consistent and a reproducible rating

15   system for severity of the presentation of disease.

16                  Third is a lack of standard outcome measures,

17   and you will see quickly that this becomes a problem in

18   interpretation of these trials.        And lastly, and probably

19   least    important,   is   the    role     for   non-physiologic

20   outcomes.

21                  I've chose to discuss in detail this trial,

22   which was published in the Annals of Internal Medicine

23   from the University of Manitoba in Winnipeg.                 It is

24   probably the most widely quoted placebo control trial

25   of AECB in the literature.

                             S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1                      These authors looked at 362 exacerbations

2    in   173       patients   with     AECB.        These     patients     were

3    randomized to receive either a placebo or antibiotics.

4     The antibiotics could be any one of Bactrim, amoxicillin,

5    or    doxycycline,         depending       on     the   investigator's

6    discretion.

7                      Patients    could      be     treated     also     for       a

8    subsequent exacerbation, in which case they received the

9    opposite treatment, placebo or antibiotics.                    Success in

10   this trial was defined as symptom resolution within 21

11   days, and of note most of these patients had -- excuse

12   me, all of them had a low

13   FEV-1.

14                     These authors did use a severity scale in

15   this trial, and it has been referred to as the Winnipeg

16   criteria.         Type-1     are   the     most    severely       affected

17   patients, and are patients who presented with cough,

18   increased sputum production, and purulence.

19                     Type-2 patients would have 2 or 3 of these

20   symptoms, and Type-3, only one, with one of the listed,

21   fairly non-specific, indicators of infection.

22   This chart basically goes through the results of the

23   trial, and I will walk you through it.

24                     On the left side of the slide are placebo

25   results, and on the right are antibiotic results, and

                                  S A G CORP.
     202/797-2525                 Washington, D.C.              Fax: 202/797-2525

1    the results are given in terms of either success or

2    deterioration.

3                    The numbers are given as percentages, with

4    the absolute numbers in parentheses.              I will direct your

5    first    to    the   overall    results      of   the    study,     which

6    demonstrated that 55 percent of patients who received

7    placebo had a successful outcome, and 68 percent of those

8    who had antibiotics had a successful outcome.

9                    The results were more impressive when it was

10   divided by the severity of the infection.               You will recall

11   that Type-1 were those more severely infected, and in

12   this    case    43    percent    who    received        placebo      were

13   successfully treated, versus almost 63 percent who

14   received the antibiotics.

15                   The other thing that I wanted to point out

16   to you on this slide was that the deteriorations tracked

17   in the direction that you might expect.                   Again, those

18   who were more severely infected at presentation had a

19   higher deterioration rate when they received placebo than

20   when they received antibiotics.

21                   The conclusions then that these authors

22   reached from the study were that antibiotic treatment

23   provided no benefits to Type-3, which were the least

24   severely affected, and could probably be justified in

25   Type-2, and demonstrated the greatest benefit in those

                               S A G CORP.
     202/797-2525              Washington, D.C.               Fax: 202/797-2525

1    with the most severe exacerbations.

2                   They also noted that a higher success rate

3    in the antibiotic treated groups may be less important

4    than the clinical deterioration.           They found in their

5    study that subgroups of individual symptoms were no more

6    predicted about the outcome than were the group that

7    constituted their severity scale.

8                   The caveats specific to this particular study

9    were first of all that no microbiology was done.               All

10   of the antibiotics used were assumed to be equally

11   effective.       It   was   of   course     conducted    in    the

12   pre-resistance era.

13                  Steroid use was not controlled, and there

14   were relatively small numbers of patients in the study.

15    Moving on then to I think another fairly well known study,

16   a meta-analysis conducted by SAINT and colleagues, which

17   was published in JAMA in 1995.

18                  This study was a meta-analysis of nine

19   placebo controlled trials of antibiotics in AECB.              And

20   it is important to recognize that these nine trials that

21   were included were actually out of 230 studies screened,

22   and that only those nine studies met their criteria.

23                  That criteria that they used was that the

24   study should be randomized, and there should be a

25   diagnosis of chronic bronchitis, and AECB, and at least

                             S A G CORP.
     202/797-2525            Washington, D.C.         Fax: 202/797-2525

1    a five day duration of follow-up, and data sufficient

2    to calculate an outcome size.

3                   Now, what they ended up doing, because there

4    were different outcome criteria used in the different

5    studies, was to calculate what they called an effect size,

6    which is a unitless measure of efficacy.

7                   The results were that when the trials were

8    combined, they yielded an overall effect size, which was

9    indicative of a small, but statistically significant

10   effect, favoring antibiotics over placebo.

11                  It is important to note, however, that the

12   breakdown of the nine trials was as follows, which were

13   that 3 of 9 sort of statistically significant benefit

14   of the antibiotics; and 3 of 9 showed a trend favoring

15   antibiotics; and 3 of 9 showed no difference from placebo.

16                  Because the authors realized that the effect

17   size would be a fairly confusing phenomena, they also

18   looked at the most commonly reported outcome measure,

19   which was the Peak Expiratory Flow Rate, and that was

20   reported by six of this nine trials.

21                  When they looked at those trials, they found

22   that 2 of 6 showed a trend or significant improvement

23   in Peak Expiratory Flow Rate favoring the antibiotics,

24   and the others obviously did not.

25                  The conclusion that these authors reached

                            S A G CORP.
     202/797-2525            Washington, D.C.     Fax: 202/797-2525

1    were that antibiotics yield a small, but statistically

2    significant, improvement compared with placebo that may

3    be clinically significant, especially in patients with

4    low baseline flow rates.

5                   The caveats in this particular meta-analysis

6    was what we have already mentioned.             That there were a

7    variety of outcome measures used.            In addition to Peak

8    Expiratory Flow Rate, the duration of the exacerbation,

9    the PaO2, symptom scores, or overall severity scores,

10   determined by a physician, were all used variously in

11   these studies.

12                  This placebo control trial by Allegra, et

13   al, was one of the ones that was not included in the same

14   meta-analysis because at the time their original results

15   were published in Italian.

16                  However,    they     published     a    more     recent

17   analysis that described their entire results, and I wanted

18   to present that to you today as another example of placebo

19   control trials.

20                  This   particular        trial    looked       at    the

21   amoxicillin/clavulinic acid versus placebo, both given

22   in a five day course.        And patients were greater than

23   40 years old had cough and sputum production, an FEV1

24   of less than 80 percent predicted and no patient received

25   steroids.

                               S A G CORP.
     202/797-2525              Washington, D.C.            Fax: 202/797-2525

1                     Of 761 patients screened, there were 369

2    exacerbations included in this trial, and the failure

3    rate was given here, which was 49.7 with placebo, and

4    13.6 also received antibiotics.

5                     The retrospective review, which constituted

6    the second paper, showed that those folks who presented

7    with low FEV-1, did worse with placebo.                   And they

8    concluded that those with severe function impairment,

9    and higher number of exacerbations, derived the greatest

10   benefit.

11                    I would like to present to you here not a

12   placebo control trial, but actually an evidence-based

13   clinical practice guideline put out by ACP and ASIM, and

14   ACCP jointly.

15                    What these authors did -- and it was published

16   in Annals of Internal Medicine in 2001, was to review

17   not only therapeutic interventions, but also modalities

18   of diagnostic testing for utility.

19                    In the review, the antibiotic treatment of

20   AECB, they included 11 randomized placebo controlled

21   trials.        These included the nine that we have already

22   mentioned that were included in the SAINT meta-analysis,

23   as well as two that had been published subsequently.

24                    In the review of these papers, these authors

25   concluded       that   antibiotics     are   beneficial    in    the

                               S A G CORP.
     202/797-2525              Washington, D.C.         Fax: 202/797-2525

1    treatment of patients with AECB.                    Patients with more

2    severe exacerbations are more likely to benefit from

3    antibiotics.

4                        I wanted to very briefly mention the placebo

5    control trial that involved antibiotic treatment of

6    patients with AECB.           This was published in the Lancet

7    in 2001, and involved a randomized placebo controlled

8    trial of ofloxaci, and 400 milligrams a day, versus a

9    placebo for 10 days.

10                       These 90 patients were sort of a unique group,

11   in that they did have AECB, but these are patients who

12   presented severely ill enough to imminently require

13   mechanical ventilation.            The authors fairly rigorously

14   excluded pneumonia, and they were allowed to receive

15   aminophylline, but not steroids.

16                       Given   the   extreme         presentation      of    the

17   patients, we see extreme results.                 The mortality actually

18   was 22 percent in patients who received placebo, and 4

19   percent in those who received ofloxacin, and the secondary

20   end point that was looked at was the requirement for more

21   antibiotics and which also showed the same trend.

22                       In addition, these folks had a decreased

23   duration       of    ventilation,      and       hospital   stay    in    the

24   ofloxacin group.            I would point out that again these

25   patients were severely ill, and really what we are seeing

                                  S A G CORP.
     202/797-2525                  Washington, D.C.              Fax: 202/797-2525

1    here is most likely a prevention of hospital-acquired

2    pneumonia, rather than treatment of AECB per se.

3                    What I would like to present here is actually

4    again not a placebo controlled trial, but a review of

5    the same.      The results that you will see here are from

6    an AHRQ evidence report or technology assessment.

7                    This particular document was prepared by the

8    Duke University Evidence-Based Practice Center.                     The

9    procedure      for   these    documents      is   that     the    EPCs

10   systematically       review     the    relevant       science-based

11   literature     on    their    assigned      topics,    and    conduct

12   additional analyses when appropriate.

13                   When this group of investigators examined

14   11 placebo controlled trials versus antibiotic treatment,

15   they included the 9 that we have discussed, and the two

16   subsequent trials that were in the Bach study, but not

17   in the meta-analysis.

18                   I wanted to very briefly mention one of those

19   two additional trials here, because I think it illustrates

20   one of the points that we are discussing.                    This as

21   conducted by Sachs, et al, and was published in 1995.

22                   And 71 outpatients who had TMP/SMX and

23   increasing AECB were treated with either trimethrin

24   sulfa, amoxicillin, or placebo.             All of these patients

25   received steroids.

                              S A G CORP.
     202/797-2525             Washington, D.C.             Fax: 202/797-2525

1                   There were no differences observed in the

2    recovery rates, changes in symptoms, or peak expiratory

3    flow rate, temperature, or sputum.         And the caveats to

4    interpretation of this study include the fact that the

5    roll of corticosteroids anti-inflammatory effect is

6    undefined.

7                   These patients did have relatively high peak

8    expiratory flow rates, and a low proportion of patients

9    with purulent sputum, implying that there were perhaps

10   not as ill as some patients in other studies had been.

11                  The conclusions that the AHRQ documents

12   reached was as follows.       Randomized control trials of

13   the antibiotic treatment of acute exacerbation of chronic

14   bronchitis show overall evidence of a relatively small

15   benefit in pulmonary function.

16                  These trials suggest that patients with more

17   evidence of bacterial infection, sputum purulents, and

18   more severe illness, worse peak expiratory flow rate,

19   benefit most from antibiotics.

20   However, this has not been conclusively demonstrated.

21                  Likewise, the hypothesed interaction between

22   corticosteroids and antibiotic use cannot be addressed

23   by existing trial data.       That concludes the review of

24   what is available to us in the literature regarding the

25   results of placebo controlled trials and the treatment

                            S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    of AECB.

2                    I would like to reiterate what I think are

3    some of the confounding issues in trying to reach a

4    definitive conclusion in that determination of delta-1.

5     First, there is the fact that concurrent effective

6    therapies      or   other    eogenous          factors   may    diminish

7    treatment group differences.

8                    And clearly you have seen in some of the

9    studies that systemic corticosteroids are one of those

10   factors, as well as inhaled, short-acting beta agonists

11   and bronchodilators, and oxygen therapy.

12                   All of those have been shown in independent

13   studies to have a treatment effect in AECB, and of course

14   cigarette smoking also is going to have that same effect.


16                   A very important point is the difficulty in

17   defining appropriate patient populations for study.

18   First is the issue which has been referred to in other

19   contexts of looking at bacteriologic end points.

20                   Clearly in AECBs that is not possible because

21   of the issue of sputum colonization with pathogens in

22   the COPD.      In addition, there has always been in various

23   studies the question of the unclear role of viruses,

24   atypical pathogens, environmental exposure, as well as

25   non-infectious problems in the causation of AECB.

                                 S A G CORP.
     202/797-2525                Washington, D.C.             Fax: 202/797-2525

1                   A very significant problem that remains to

2    be addressed is the fact that severity criteria for this

3    disease have yet to be validated.         The assumption that

4    the AECB severity can be judged by some combination of

5    presenting clinical features is intuitive, but is yet

6    to be confirmed by clinical studies.

7                   Just as an example to show potentially how

8    different populations of AECB can be constituted, what

9    you see here are representations of the study that I

10   mentioned to you from Winnipeg, as well as some data that

11   was extracted from an NDA, which came to us recently.

12                  What I wanted to point out was two things.

13    First of all, obviously these three criteria -- the FD-1,

14   the sputum volume, as well as severity symptoms, which

15   can be used or have attempted to be used to some degree

16   of prognostic prediction, were given here in this study,

17   but were not available to us for the NBA review.

18                  As well, I wanted to point out that the

19   patients here were significantly younger, and a much lower

20   percent of smokers, either current or past, which may

21   well affect the results given that the patient populations

22   would be significantly different.

23                  And I just wanted to very briefly mention

24   the old versus new antibiotics, and specifically we all

25   know that resistance is increasing, and that includes

                            S A G CORP.
     202/797-2525           Washington, D.C.         Fax: 202/797-2525

1    the pathogens that are presumed to be operant in AECB,

2    and most of the studies that we have reviewed were

3    conducted before the emergence of respiratory pathogens

4    that are resistant to multiple antibiotics.

5                   And having said that, however, I think it

6    is important to know that there has been no randomized

7    control trial which have showed the superiority of newer

8    broad spectrum antibiotics in this disease entity, and

9    there is no data to suggest increased failures with the

10   increase in antibiotic resistance.

11                  Having gone through this review of the

12   studies then can we determine delta-1, which is sort of

13   what we started out with in the beginning.   What we would

14   like to be able to do ideally would be to perform a

15   meta-analysis of the available literature, and then

16   calculate delta.

17                  The problems that we see in this approach

18   are, first of all, that the patient population in placebo

19   controlled trials that are available to us for review

20   was not uniform.

21                  Secondly, and probably one of the most

22   important things, is that the studies that were available

23   used very different designs, and very different end

24   points, none of which were ideal.

25                  The studies clearly had different outcomes,

                            S A G CORP.
     202/797-2525            Washington, D.C.     Fax: 202/797-2525

1    and some have shown a treatment effect and some did not,

2    and most of these studies were not recent.

3                   In conclusion then, in terms of the selection

4    of delta, the performance of a meta-analysis, with

5    subsequent     selection     of    delta,     would   not    yield        a

6    meaningful value due to the differences in study design,

7    including heterogeneous patient populations, and diverse

8    end points.

9                   A review of placebo controlled trials of

10   antibiotic treatment of AECB does not allow a definitive

11   estimation of the benefit of active control over placebo.


13                  Patients with more severe -- with a question

14   as to what that definition should be, a more severe

15   illness, may benefit most from antibiotics, but this has

16   not been conclusively demonstrated, nor have validated

17   severity criteria been demonstrated.

18                  What then are some options for what future

19   trials should represent.          Well, first of all, of course,

20   would be non-inferiority trials in all patients, which

21   is the current practice.        But I hope that I have presented

22   you data that convinces you that it is difficult to choose

23   an appropriate delta.

24                  Secondly, it would be placebo controlled

25   trials with an early escape option in all patients with

                             S A G CORP.
     202/797-2525               Washington, D.C.           Fax: 202/797-2525

1    AECB, or placebo controlled trials only in patients who

2    are perceived to be at low risk.

3                   For instance, mild to moderate Groups 2 and

4    3, and of course another possibility would be to do placebo

5    controlled trials in patients who have very severe

6    presentation.

7                   Another   option     would   be   non-inferiority

8    trials in severely ill-only AECB patients, with the

9    possibility     of   controlling      for   smoking    and    other

10   concurrent therapies, and understanding that we need to

11   have a reliable and reproducible definition of severe

12   AECB.

13                  You have already heard about the possibility

14   of three Arm studies involving a placebo, the new drug,

15   and/or the old drug.        And this would certainly be an

16   option here.

17                  Unresolved issues in AECB.         First of all,

18   are placebo controlled trials with an early escape option

19   acceptable in AECB studies, and a corollary of that is

20   should only patients with less severe disease be enrolled

21   in these trials.

22                  Secondly,   if    non-inferiority      trials     are

23   conducted in AECB, what should the delta be?           And lastly

24   should future AECB trials include only patients with

25   severe AECB.    Thank you for your attention.

                              S A G CORP.
     202/797-2525             Washington, D.C.          Fax: 202/797-2525

1                   CHAIRMAN RELLER:          Are there any questions

2    for Drs. Powers and Thompson?            Yes?

3                   DR. ROTSTEIN:       I would like Dr. Powers to

4    comment on hospital-acquired pneumonia and the use of

5    the clinical pneumonia severity index score that people

6    have used?

7                   There is a modified pneumonia severity index

8    score that people have used as criteria for entry into

9    nosocomial     pneumonia     trials,       and     also    to    gauge

10   improvement.     Could you comment on that?               You didn't

11   comment on that.

12                  And   also    the     use     of    quantification,

13   particularly endotracheal aspirates, looking at greater

14   than 10 to the 5th organisms per Ml.

15                  DR. POWERS:   Let me take your second question

16   first.     It becomes very problematic to validate the use

17   of BALs or bronchoscopic techniques.              There was a study

18   by Fagan that actually looks at people that had purulent

19   sputum, abnormal chest radiograph, and greater than 10

20   to the 3rd organisms.

21                  Versus   those      who     had    purulent      sputum,

22   abnormal chest radiograph, and negative cultures done

23   by that method.      And the mortality rate was 26 percent

24   in both groups.

25                  And so does that mean that there is no

                             S A G CORP.
     202/797-2525            Washington, D.C.              Fax: 202/797-2525

1    difference between those groups or does it mean that the

2    sensitivity of those bronchoscopic techniques is not very

3    good?

4                   Considering          that        those     bronchoscopic

5    techniques are not compared to any gold standard, that

6    becomes very problematic, trying to tell what those mean.


8                   When    I     looked     over     the    four    new   drug

9    applications for trovafloxacin and piperacillin, and

10   tazobactam, ciprofloxacin, and linezolid, I did not see

11   a use of that score that you are referring to, to try

12   to determine.

13                  So the question I was asked or is posing here

14   is that those may be useful.               I am not aware of them,

15   and I really can't comment.

16                  DR. ROTSTEIN:          One of the problems with those

17   trials is they use a conglomeration of patients, a

18   smorgasbord.          The      trovalfoxacin            study    excluded

19   ventilator-associated pneumonia patients.                  So you could

20   only be ventilated 48 hours or less.

21                  I was one of the investigators in that trial,

22   and I was one of the investigators in the linezolid trial

23   as    well,    and    that     included         ventilator-associated

24   pneumonia patients.          It was different.

25                  But     all      the      other     ones        have   been

                                  S A G CORP.
     202/797-2525                 Washington, D.C.             Fax: 202/797-2525

1    mild-to-moderate hospital-acquired pneumonia, and that

2    is why we have been unsuccessful in doing these trials.

3     The money is really ventilator-associated pneumonia

4    patients.

5                   DR. POWERS:      The question that comes up

6    though is whether a company would want to study hospital

7    acquired pneumonia in non-ventilated patients, and what

8    kind of advice would we give to those people, and I will

9    let the committee address that one as well.

10                  CHAIRMAN RELLER:      Dr. Archer.

11                  DR. ARCHER:   From a statistically challenged

12   person, namely me, I have a question.            Can you stratify

13   in a trial like an AECB trial, where there clearly are

14   different groups, can you stratify the patients going

15   into the trial and assign a different delta to different

16   strata within the same study, or is that a no-no?            I guess

17   that would be to the second person who presented the AECB.

18                  CHAIRMAN RELLER:      Dr. Thompson.

19                  DR.   THOMPSON:             I'm   probably        more

20   statistically challenged actually, but I guess the answer

21   to that is -- and I am going to start and let you guys

22   work on this.

23                  But clearly there are subgroups within AECB

24   that respond differently to bronchitis, and so whether

25   it is a practical matter to assign a different delta to

                            S A G CORP.
     202/797-2525            Washington, D.C.            Fax: 202/797-2525

1    different populations, I think that would be problematic

2    from a study design standpoint.

3                   And from a clinical standpoint, I would say

4    that we have yet to precisely identify them.                 So I think

5    that would be the problems that I see theoretically if

6    you could get around all of those issues, perhaps.

7                   But thus far there is not a set of validated

8    severity criteria that predict outcome.              I would say no.

9     And I think the other interesting thing that needs to

10   be further studied, and that I didn't present, is that

11   there is a suggestion in several studies that the best

12   predictor of prognosis is actually not the current

13   presentation, but rather history of cardiopulmonary

14   disease, as well as how many exacerbations they have had

15   in the past.

16                  And so it may well be that looking at those

17   factors might be more predictive, but I know that your

18   question is really delta, and I don't think that is

19   practical, and I will let my statistical colleague address

20   that.

21                  CHAIRMAN    RELLER:           Dr.   Temple,    and   Dr.

22   Fleming, if you have comments on this.

23                  DR.   TEMPLE:       Well, this is a complete

24   cop-out, but you could certainly do an all-comers trial

25   and stratify the population by the severity, and have

                               S A G CORP.
     202/797-2525              Washington, D.C.            Fax: 202/797-2525

1    different criteria for success in each of the strata.

2                   It would really be multiple trials, but in

3    a single environment.     You might even have a superiority

4    hypothesis in one, and a non-inferiority hypothesis in

5    the other, but it really wouldn't be one trial.

6                   Tom will have to tell you how you could do

7    that in a single end-point or not.

8                   DR. FLEMING:    After the break maybe?

9                   DR. BRITTAIN:    You might want to use or you

10   might want to base your delta on what proportion of people

11   you have in your trial in the three groups, and you could

12   think about it that way, and that would be one overall

13   analysis.

14                  But if you wanted to do it within each

15   category, then you would need a sample size, and you would

16   need a big sample size in that case.

17                  CHAIRMAN RELLER:      I think it is time for our

18   afternoon break, and we will reconvene at 2:45, 15

19   minutes.

20                  (Whereupon, at 2:34 p.m., the conference was

21   recessed and resumed at 2:53 p.m.)

22                  CHAIRMAN RELLER:            Before Dr. Goldberger

23   gives the charge to the Committee for discussion of the

24   questions, we want to have transitional comments in

25   response to the last query before the break having to

                             S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    do with stratification of patients in studies of acute

2    exacerbation     of    chronic    bronchitis,     and   what     the

3    appropriate statistical analyses would be, and Dr. Thomas

4    Fleming has some comments to make on that query.

5                    DR.   FLEMING:        Just very briefly.         The

6    question was asked if it would be at least possible to

7    entertain having a different margin in various strata

8    or subgroups.

9                    Thinking about it for a little bit, my sense

10   is, yes, it is.       Whether I would suggest that it is wise

11   or not is an entirely separate issue.            But if we used,

12   for example, the setting of acute exacerbation of chronic

13   bronchitis that we were just talking about, and if in

14   fact, just to simplify this discussion, one took it as

15   reasonably established that in less serious disease there

16   is no effective antibiotics on the end-points of interest,

17   and in more serious disease there is a 20 percent

18   improvement, then in less serious disease you might have

19   wanted to do a superiority trial using a margin of zero.

20                   And in more serious disease, you would have

21   allowed some margin.       Let's say it is in fact the fullest

22   margin that you might allow, which is a full 20 percent.

23    Then essentially one could aggregate the data from those

24   two   strata,    essentially     in    essence   looking   at    the

25   parameter of how much better are you than placebo.

                               S A G CORP.
     202/797-2525              Washington, D.C.         Fax: 202/797-2525

1                   So in the stratum of less serious disease,

2    you are just taking the estimated difference between the

3    experimental and the active comparator.        Whereas, in the

4    more serious disease, you are taking that difference.

5                   But then you are adding back what you think

6    the effect is against placebo.       You are rewarding an extra

7    20 percent in the stratum of more serious disease, thereby

8    doing an overall stratified analysis that gives you a

9    global estimate of how much you are better than placebo.

10                  So that is just one of, and I just wanted

11   to raise the fact that you could conceptually do it, and

12   there are probably other ways to do it, too.                     The

13   advisability of doing that is an entirely separate issue,

14   because you are really mixing apples and oranges here

15   a bit.

16                  And you are taking a superiority component

17   and you are taking a non-inferiority component, and you

18   are imputing the full 20 percent estimated benefit that

19   you think the active comparator antibiotic has in the

20   more serious disease stratum, and that may or may not

21   be the right thing to do.

22                  But it is at least conceptually possible

23   statistically     to   work    out     something     that     would

24   essentially allow a different margin essentially in

25   different strata.

                             S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1                     CHAIRMAN     RELLER:          Thank    you.         Dr.

2    Goldberger.

3                     DR. GOLDBERGER:        I actually almost started

4    to go into the questions, and so I will actually try to

5    keep    my     comments   brief.       We have heard a lot of

6    presentations this morning.

7                     We heard presentations from FDA staff on sort

8    of backgrounds for evolution of delta, and some of the

9    current concerns and issues from an FDA perspective.

10                    Certainly from our perspective on one hand,

11   while we recognize that there are real issues in some

12   of these indications, and the ability to do clinical

13   trials, and we also hope that we made the point that

14   talking about delta is not just a discussion of some arcane

15   statistical issue.

16                    It in fact does have relevance to actual

17   patient care and patient outcome.               We heard a lot of

18   prospectus from industry, IDSA, and academia.                  I think

19   industry certainly indicated a strong desire to work in

20   the development of new antimicrobial products.

21                    But I think they tried to make the case that

22   there are some real economic realities that they have

23   to live with, and in fact in other presentations industry

24   has been even more specific about what some of those

25   constraints are.

                                 S A G CORP.
     202/797-2525                Washington, D.C.           Fax: 202/797-2525

1                   And    that    they     would   like   to   see    some

2    approaches that would allow them to operate within those

3    constraints.       Take the Infectious Disease Society.

4                   They certainly showed a strong willingness

5    to help in any way that it could with this process, and

6    also I think expressed certainly a desire to provide as

7    much expertise as they certainly could.

8                   I    think    the    Infectious   Disease      Society

9    clearly is interested in their continuing to be an active

10   pipeline of new antimicrobial agents.                   I am sure,

11   although it didn't come out perhaps as strongly in their

12   comments, they are also interested in ensuring that

13   antimicrobial products that are out there, as well as

14   new ones, are used in a manner that sort of preserves

15   their useful life as long as is possible.

16                  We also then heard in the afternoon some

17   specific examples to help focus the discussion, dealing

18   with several different indications, and looking at how

19   much data we actually have in terms of thinking about

20   delta-1 and delta-2, keeping in mind that the delta-2

21   is ultimately a clinical judgment.

22                  One of the areas that we certainly heard a

23   lot about is the issue of bacterial meningitis, and it

24   is a very good example of some of the difficulties in

25   approaching this whole area.

                                 S A G CORP.
     202/797-2525                Washington, D.C.          Fax: 202/797-2525

1                   And that is that on one hand it is beyond

2    any question that the benefit of antimicrobial therapy

3    is enormous.    On the other hand, recognizing the severity

4    of failure, which can range from death to at least a

5    variety of developmental delays, hearing loss, et cetera,

6    we would like our new antimicrobials to work as close

7    as possible, at least to the same degree, if in fact not

8    better, than what is already out there.

9                   Yet at the same time, we recognize that to

10   do clinical trials like that probably has sample sizes

11   that are almost prohibitive.       Therefore, there was some

12   discussion about what would be the usefulness of focusing

13   more on PK/PD, animal models, and microbiologic end

14   points, as opposed to clinical success end points.

15                  This is clearly an area that needs further

16   discussion.    I think one of the issues that perhaps was

17   not entirely resolved was whether or not the bacteriologic

18   end point really captures all the information that we

19   need to see to be satisfied that the drug will be effective

20   clinically.

21                  Well, we have some questions which we will

22   get to in a second, and that we obviously would like some

23   discussion on.     We want to point out first that these

24   questions are meant sort of to introduce discussion,

25   depending upon the available time.

                            S A G CORP.
     202/797-2525           Washington, D.C.       Fax: 202/797-2525

1                   Certainly we would welcome other comments,

2    areas of interest that the committee would like to talk

3    about based on personal experience, and/or what has been

4    presented today.

5                   One issue in fact that would be nice to hear

6    some discussion about goes back to something that I just

7    mentioned a moment ago.

8                   Both in the meningitis discussion and in some

9    discussions at the break, I did hear the comment that

10   from an antibiotic perspective, we really should be

11   looking at what the drug does bacteriologically, rather

12   than clinical outcomes.

13                  And the question is how much weight should

14   we put on this approach, particularly in more severe

15   disease.       On one hand, obviously a major role of

16   antibiotics is of course to effect a bacteriologic cure.

17                  On the other hand, if we don't get the

18   requisite patient response, what are we supposed to do

19   with that type of situation.          And if there is time, we

20   would welcome some comments about that.           Leo, could you

21   put up the first question.

22                  The first area that we want to ask your

23   opinion about is using AECB as an example, please discuss

24   some of the different clinical trial design options in

25   infections     where   the   magnitude      of   the   benefit    of

                             S A G CORP.
     202/797-2525             Washington, D.C.          Fax: 202/797-2525

1    antimicrobial therapy over placebo remains uncertain.

2                    And we have several different options here,

3    and some placebo controlled trials, and three arm trials,

4    dose response trials, and as time permits, you might want

5    to expand this discussion to some to some other areas,

6    i.e., otitis media and sinusitis, where there have been

7    issues at times about the overall benefit of antimicrobial

8    therapy.

9                    From our perspective, beyond getting some

10   input about trial design, we are obviously interested

11   in   ensuring    that    our   approach      appears    to    be   most

12   appropriate, and whether that means the same approach

13   we have been using, or some modifications, we would like

14   to get the best possible data that we can.

15                   We also would like to think that given the

16   relatively limited amount of data there is about the

17   benefit of antimicrobial therapy in this indication, some

18   of the clinical trials that might be used to seek approval

19   might also provide some additional information on who

20   the patients are, and who really benefit from therapy.

21                   Because realistically there is a lot of

22   antimicrobial therapy used in bronchitis, and I think

23   there is little question that the use of antimicrobial

24   therapy, in addition to some degree of patient benefit,

25   probably       carries    with    it     some    development          of

                               S A G CORP.
     202/797-2525              Washington, D.C.             Fax: 202/797-2525

1    antimicrobial resistance.

2                   The question is are we getting the best

3    trade-off right now.     And if you could go to the second

4    question, Leo.

5                   And this is please discuss the implication

6    of choice of deltas in clinical trials for serious

7    infections.     Please consider in your discussion the

8    efficacy of a new drug compared to available therapy for

9    the indication e.g. HAP and meningitis.

10                  And basically the issues are smaller deltas

11   and the effect on sample size of clinical trials,

12   particularly when the infection is rare, and/or the

13   success rate is low.

14                  And larger deltas and the impact on patient

15   care if potentially less efficacious drugs are approved.

16                  And a simpler way I think of sort of summing

17   this up is that there is no such thing as a free lunch.

18    Either you spend the resources to be able to do larger

19   trials that give you more precise data, or there will

20   be on one hand some limitations on what you know about

21   the drugs.

22                  On the other hand, if the cost is too high,

23   the trials will never get done, and I think that this

24   is an area that we would like to hear all your comments

25   about.

                            S A G CORP.
     202/797-2525            Washington, D.C.     Fax: 202/797-2525

1                     It is a very difficult area, and it is a

2    problem for us, and clearly a problem for industry, and

3    whatever advice you can provide would be extremely useful.

4                     And finally the third question.              Please

5    discuss what other factors, characteristics, of a drug

6    product other than primary confidence interval results

7    could be included in a risk benefit analysis supporting

8    an FDA regulatory decision.

9                     And certainly to be included in this can be

10   safety         considerations,     PK/PD,     availability        of

11   alternative       therapies,     other   factors   as   you   think

12   appropriate.

13                    Traditionally, we have been more flexible

14   in situations where therapeutic options are limited, and

15   where the disease is severe, and the alternatives may

16   not be ideal, at least for some group of patients.

17                    We would clearly think that this should

18   continue to be the approach in the future, and in fact

19   I suspect there will be considerably more discussion about

20   this tomorrow when we talk about the development of drugs

21   for resistant indications.

22                    Nonetheless, even though we believe we have

23   some appreciation of the factors that are important in

24   these decisions, we think it would be useful to hear some

25   additional comment from the committee about factors that

                               S A G CORP.
     202/797-2525              Washington, D.C.         Fax: 202/797-2525

1    they would consider important with the degree of specifics

2    that people feel comfortable providing.               Thank you.

3                   CHAIRMAN    RELLER:           Let's    come    back    to

4    question one.    Discussion from the Committee, and by the

5    Committee, I would include the extended Committee, those

6    invited from IDSA, PhRMA, industry, and Members at all

7    of the tables, including the proximal ones.                  Jim.

8                   DR. LEGGETT:      I forget I was on the end again

9    once again, and so I might as well start.                 I spent my

10   time during the break trying to think about this.

11                  And regarding Issue Number 1, I think my

12   overall bottom line is I would favor anything but what

13   we are doing now, in terms of non-inferiority, among those

14   three items.

15                  I think in a trial ongoing with AECB, it is

16   going to be hard to restrict the categories since we don't

17   have any validated severity criteria.                And I think the

18   other thing about going forward and trying to include

19   everybody is the closer we can make the Phase III trial

20   to what is going to be generalized to outpatient use in

21   the future, the more likely we are going to get some data

22   that will help us.

23                  And I think we also know in that regard that

24   there is widespread antibiotic use as was just mentioned,

25   even with acute bronchitis, and the people that are going

                               S A G CORP.
     202/797-2525              Washington, D.C.             Fax: 202/797-2525

1    to     be      using   this      are     pulmonologists,         general

2    practitioners, and anybody but ID folks.

3                     I think we definitely have going forward in

4    these trials, we definitely have to account for steroid

5    use.    And if memory serves me well, in that Anthonisen

6    trial, they went back and you could look at the steroid

7    use, and that is what correlated with improvement in all

8    three of the subtypes.

9                     I think we could consider monitoring for

10   deterioration as a primary target end point, rather than,

11   quote, success/failure.           I don't think we should use a

12   microbiologic end point in AECB because the prevalence

13   of the, quote, pathogen recovery from the sputum is the

14   same, or even greater, when there is no exacerbation,

15   than when there are exacerbations.

16                    And the density, in terms of CFU per Ml in

17   the    sputum     is   no     different         in   exacerbations      or

18   non-exacerbations.          And to the extent that acute otitis

19   media and sinusitis are not diagnosed by puncture, and

20   so we don't have, quote, hard data, I think they need

21   to be treated the same as acute exacerbations of chronic

22   bronchitis due to the similar colonization problems and

23   the similar pathogens.

24                    And with the same similar high placebo

25   success rate.

                                 S A G CORP.
     202/797-2525                 Washington, D.C.            Fax: 202/797-2525

1                   CHAIRMAN RELLER:      Dr. Cross.

2                   DR. CROSS:    Well, I would agree that in a

3    situation like bronchitis, where we have a punitive

4    infection in a non-sterile site, I think that having a

5    bacteriologic cure would be extremely difficult.

6                   And I think based on the evidence presented,

7    that it seems certainly reasonable that a placebo in a

8    controlled trial still ought to be the norm from the point

9    of view that it is a less severe type of infection.

10                  We have the alternative of having the early

11   escape, which if properly designed would allow us to

12   identify those patients who are at the highest risk who

13   may benefit, as perhaps was indicated in the Canadian

14   study.

15            So I think that kind of design would allow us to

16   at least for the next study perhaps perspectively identify

17   criteria for folks who don't do well under the typical

18   placebo controlled trials.

19                  So I think that certainly given the natural

20   history of that process, I think we wouldn't be doing

21   the patients any undue harm, but still have the safety

22   valve to ensure that all patients are safely treated.

23                  CHAIRMAN RELLER:      Dr. Archer.

24                  DR. ARCHER:   I think with reference to AECB,

25   the patients that I see on the wards, I think one could

                            S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    establish criteria for the very non-severely ill, versus

2    those that are very severely ill, and either stratify

3    a study or divide them into two different groups.

4                    On   the    one   hand,        I    think   most    of    the

5    antibiotic use is really in the not very severely ill

6    patients,      and   that    is   probably          where   most    of    the

7    antibiotic resistance is generated as well.

8                    Whereas, studies may overpresent the more

9    severely ill patients.             So therefore I think it is

10   important to differentiate those groups, and doing a study

11   may actually help define how you can separate those two

12   groups out.

13                   And I would favor doing placebo control with

14   the not severely ill, and non-severely, non-placebo

15   control with some estimation of delta in the more severely

16   ill.

17                   And I think it is important in the severely

18   ill patients to include all current types of therapy that

19   are used for these patients who are deteriorating in their

20   pulmonary function, to include inhale steroids, systemic

21   steroids, all the nebulizer treatment, maximum therapy

22   in that group.

23                   Plus,      antibiotics         of    different      groups,

24   because that is what is done, and I think sometimes that

25   it is difficult to differentiate.                   One could maybe even

                                 S A G CORP.
     202/797-2525                Washington, D.C.                Fax: 202/797-2525

1    argue in some of those groups that that placebo control

2    is appropriate with everything else that is being done,

3    but I leave that to the pulmonologists.

4                    As far as other types of infections, I don't

5    see much acute otitis.          I really can't comment on that,

6    but I think that sinusitis is difficult to define, and

7    it   seems     like   more    microbiological     data     should     be

8    generated, in terms of punctures.

9                    Or possibly doing CT scans to try to define

10   who does and doesn't have sinusitis as a criteria for

11   study entry, because I think there is also a lot of

12   inappropriate use of antibiotics for poorly defined

13   sinusitis, and a lot of antibiotic resistance being

14   generated in that as well.             Let me see.     I guess those

15   are major comments.

16                   CHAIRMAN RELLER:          Dr. Ebert.

17                   DR. EBERT:       Well, it appears that there are

18   a variety of things that are going to impact the size

19   of the patient population in these studies, one of which

20   is the prevalence of the disease, and secondly, the impact

21   of therapy on outcome.

22                   And I think an acute exacerbation of chronic

23   bronchitis, both of these speak towards the use of a

24   large-scale study.           It should be an adequate patient

25   population, and also because we are not really clear on

                                  S A G CORP.
     202/797-2525                 Washington, D.C.          Fax: 202/797-2525

1    the impact of outcomes, a larger population should help

2    us in that way.

3                    I think if we want to go back to the basics,

4    it would be to do a very large scale study, and try to

5    validate subsets of patients who do in fact respond, and

6    who do not.

7                    If that in fact does not work, or if that

8    is not the tract that we want to take, certainly we have

9    talked     in   this   committee    about   enriching    patient

10   populations, or selecting out specific criteria for

11   entrance into the study to ensure that the populations

12   that we are treating are going to be at greater likelihood

13   of response.

14                   I also agree that a microbiologic response

15   is not likely to be a good end point for this particular

16   disease, which really leads us into the clinical response,

17   and the question I have there is really again the issue

18   of the timeliness of the assessment.

19                   And I don't recall hearing any discussion

20   of the time frame at which we are assessing clinical

21   response, and certainly with other disease states we have

22   talked about assessing patients at 28 days from the

23   beginning of enrollment in a study.

24                   And we have argued that that may in fact be

25   too long of a time.      So it may be that we need to look

                              S A G CORP.
     202/797-2525             Washington, D.C.        Fax: 202/797-2525

1    more closely at end-of-treatment as an assessment, rather

2    than some time point in the distant future.

3                   CHAIRMAN    RELLER:           Dr.   Ramirez     had        a

4    question, and then Dr. Patterson.

5                   DR. RAMIREZ:      Just a comment.          Just to add

6    a new factor to the complexity of the problem, is that

7    even though these factors are not well-defined in the

8    literature, when all different medical societies get

9    together to develop guidance for the management of

10   antibiotics     in   respiratory        tract      infections,       for

11   exacerbation    of   chronic      bronchitis,       and    nosocomial

12   pneumonia, and hospital-acquired pneumonia, the idea is

13   not to look at these diseases as a single disease.

14                  And we can clearly see, for instance, that

15   in community-acquired pneumonia, we all agree that there

16   are 3 or 4 groups of patients with pneumonia, and with

17   nosocomial pneumonia, there are at least 2 or 3, or 4

18   according to the society.

19                  And   in   acute      exacerbation         of   chronic

20   bronchitis, there seems to be that there are at least

21   three groups of patients.           And the classification of

22   patients mostly is based on the severity of the disease.

23                  And what we are trying to do is trying to

24   help the clinician in selecting empiric therapy based

25   on the likely resistant organisms causing the disease.

                               S A G CORP.
     202/797-2525              Washington, D.C.            Fax: 202/797-2525

1    And the problems that we are having is that we have

2    antibiotics that are approved for all community-acquired

3    pneumonia, and all acute situations in COPD.

4                   When in reality we know that the patient with

5    mild exacerbation, or I shouldn't say mild, but a patient

6    with low risk, for an acute exacerbation of low risk,

7    meaning that considering the three criteria considered

8    in the respiratory starts with FEV1, and considering the

9    prior use of steroids, we know that these patients

10   primarily are going to be infected with H. flu, and this

11   is one patient.

12                  And then the other end of the spectrum is

13   that we have the patient with the high release for

14   possibility for infection to due pseudomonas aeruginosa.

15                  Then the use of an antibiotic for acute

16   exacerbation of COPD, you probably need to contain the

17   patients within a risk factor for resistant organisms,

18   and trying to define again populations that we are

19   discussing here with otitis media, and trying to define

20   a patient that may have the resistant organism, or a

21   particular organism.

22                  I am trying to define antibiotic therapy more

23   specific for a particular group of patients.          I think

24   we all agree that if you have only one of the criteria,

25   you should not get antibiotics.

                             S A G CORP.
     202/797-2525            Washington, D.C.      Fax: 202/797-2525

1                   But with 2 and 3, and then the patient is

2    hospitalized, there is no question that we get the feeling

3    that    antibiotics    are     necessary.      I     think    that        a

4    stratification of the patient is critical in any one of

5    these clinical trials.

6                   CHAIRMAN      RELLER:         Thank     you.         Dr.

7    Patterson.

8                   DR. PATTERSON:      I would agree with Dr. Archer

9    that the placebo controlled trials with escape for the

10   Type II and III patients with AECV would seem appropriate.


12                  I would be more concerned about the placebo

13   controlled trials for the patient with the more severe

14   disease, and perhaps maybe there are a large number of

15   patients in this group, and that could be one place where

16   you could use a smaller delta to evaluate that.

17                  But I think also you could look at other

18   outcomes or endpoints like the duration of time between

19   exacerbations, and also not bacteriologic eradication,

20   but the flora that is present at the recurrence of the

21   exacerbation, and also to look at a comparison of therapy

22   with    symptoms,     versus    interval      pulse     therapy      or

23   prophylaxis, whatever you want to call that.

24                  And     looking        at     duration         between

25   exacerbations and also comparing susceptibilities of the

                               S A G CORP.
     202/797-2525              Washington, D.C.            Fax: 202/797-2525

1    flora at recurrence between those two groups, and would

2    you get less resistance with one group versus the other.

3                   Regarding   other     infections    like     otitis

4    media, I think that this has already been said today,

5    but I think that the double tap is of interest and that

6    bacterial eradication is an end point, although that is

7    difficult to do in this country.

8                   There are some centers that do that in other

9    countries, and that is of interest as an end point.             And

10   regarding clinical outcome as an endpoint, I think it

11   is another area where you could use a smaller delta because

12   of the large population of patients.

13                  CHAIRMAN RELLER:      Dr. Fink, please.

14                  DR. FINK:      Well, speaking as a pediatric

15   pulmonologist, I don't treat chronic bronchitis except

16   in cystic fibrosis, and where we do see it rarely, but

17   being familiar with the literature, I think there are

18   some complicating features that using AECB as an example

19   our important to point out.

20                  This   would    be    a     situation   in    which

21   international studies would in all likelihood be highly

22   flawed, and the reason for that statement is that in the

23   United States, we take cigarettes away when patients are

24   hospitalized.

25                  That is not done elsewhere in the world, and

                             S A G CORP.
     202/797-2525            Washington, D.C.          Fax: 202/797-2525

1    if you are going to deal with a controlled trial of chronic

2    bronchitis, whether or not the patient has access to

3    cigarettes or not is probably going to have a significant

4    effect on the response to treatment.

5                      We also blame a lot on H. flu.                 There is a

6    lot of newer data that says organisms such as RSV,

7    chlamydia, mycoplasma, which often with the exception

8    of RSV, and at least chlamydia and mycoplasma, often

9    respond to the same classes of antibiotics that are used

10   to treat H. flu.

11                     And that these organisms may be playing a

12   much greater role in exacerbations of chronic bronchitis

13   than is currently recognized.                So I think that part of

14   what    we     need   is    better    classification           of    chronic

15   bronchitis.       It isn't all the same.

16                     And from a clinical standpoint, probably

17   previous ICU admission is actually better than a scoring

18   system     for    disease    severity,          in   terms    of    risk    of

19   hospitalization.

20                     So I think part of what we really need in

21   chronic        bronchitis    is    better        classification,          more

22   comprehensive         studies     with   a      really   good       look    at

23   microbiology, including non-bacterial pathogens, and a

24   better understanding of the disease before we can really

25   design better trials.

                                  S A G CORP.
     202/797-2525                 Washington, D.C.                Fax: 202/797-2525

1                       CHAIRMAN RELLER:       Dr. Ramirez.

2                       DR. RAMIREZ:       I will agree, because we have

3    been saying that serious infections, that you need to

4    select the best therapy, and for this one, you need a

5    small      delta.          But        according    to    the      recent

6    identifications, patients with severe COPD has a higher

7    mortality than a patient with nosocomial pneumonia.

8                       And then we are going to be talking -- I mean,

9    if we are one of these patients with prior hospitalization

10   to an intensive care unit, that is another observation,

11   and there is a very high probability that this patient

12   is going to die during this hospitalization.                   And this

13   is the type of patient that we need to be sure that we

14   give the right antibiotics.

15                      CHAIRMAN RELLER:       Dr. Bennett.

16                      DR. BENNETT:       Several of us have commented

17   about placebo controlled trials with early escape, and

18   I am not certain that I really understand that.                It sounds

19   to me more like early discontinuation.

20                      But if my understanding is correct, there

21   are three things that we ought to take into account if

22   we   adopt     a    strategy     of    placebo    control   and    early

23   discontinuation.

24                      One is that you would have to make a double

25   blind.     Otherwise, you would have people with lack of

                                 S A G CORP.
     202/797-2525                Washington, D.C.            Fax: 202/797-2525

1    confidence in the experimental drugs and stopping the

2    drug for that reason.

3                   The other is that I think you would have to

4    have    very   rigid    criteria        as      best    you    could    for

5    discontinuation.        So      it   didn't      become       very   center

6    dependent on who wanted to stop the drug early, and

7    particularly    if     the    two    drugs      being    compared      were

8    different in their toxicity, for example, and that one

9    caused much more gastrointestinal distress.

10                  And you are now mixing two end points,

11   efficacy and discontinuation for toxicity.                      You would

12   probably be well advised to have a blinded data review

13   committee to look at all of the patients who had premature

14   discontinuation, or who escaped if you will because you

15   would want to see that there was some element of uniformity

16   between centers, and that the study definitions were

17   actually followed.

18                  And the last was I am concerned that early

19   discontinuation may not give one of the drugs a chance

20   to show its effect.          For example, if everyone got the

21   drug for 1, 2, or 3 days, you may not be convinced that

22   that was enough to actually give the drug a chance.

23                  So perhaps those of you who understand early

24   escape better than I do could explain how we would get

25   around these.

                                  S A G CORP.
     202/797-2525                 Washington, D.C.             Fax: 202/797-2525

1                   CHAIRMAN RELLER:      Dr. Fleming.

2                   DR. FLEMING:     I wanted to comment on just

3    that issue, and I don't know if you were commenting on

4    something else.     Well, I think you have raised a very

5    important issue, and I am struggling with this as well.


7                   I am not yet convinced that early escape would

8    work here, and in my thinking I am going back to Dr.

9    Thompson's slides, numbers 11 and 13.          On 13, she is

10   talking about success rates relative to what I understand

11   the primary success definition is given to be in Slide

12   11, which is symptoms resolved within 21 days.

13                  So if that is in fact is the primary end point,

14   I worry if early escape means dropping off the placebo

15   at some point before 21 days.        If it is dropping off the

16   placebo after 21 days, then I am not so concerned, and

17   here is my worry.

18                  The data on page 14 or 13, rather, is telling

19   us that eventually we should expect on placebo convergence

20   to a 55 percent success rate at 21 days.            At 21 days,

21   non-placebo, 55 percent will have resolution of symptoms.


23                  But suppose though at day 10 it is only half

24   that large, and I have no clue how rapidly this occurrence

25   of resolution of symptoms occurs, but let's say it is

                             S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    only half that large.

2                   So let's say it is about 30 percent.             There

3    are 70 percent who have not yet resolved, and if a number

4    of those people now escape placebo, and now you impute

5    failure automatically, you are going to underestimate

6    what the actual true success rate would have been on the

7    placebo.

8                   So if early escape means dropping off the

9    control arm prior to the time period at which you would

10   have achieved your full effect on the control arm, you

11   are going to have a bias underestimate of the success

12   rate on the control.

13                  On the other hand, if early escape means,

14   no, no, everybody will be on at least 21 days, and then

15   they can escape thereafter, then my concern is not

16   relevant.

17                  CHAIRMAN RELLER:        Dr. Temple.

18                  DR.    TEMPLE:      There is    a fairly narrow

19   experience     with   so-called      early   escape,     where     its

20   recurrence of symptoms like unstable angina is fairly

21   easy, and there have been trials that have been successful

22   using that.

23                  The reasons for doing it though are ethical,

24   and so you have to choose an escape provision that

25   satisfies your ethical needs.           And I don't know whether

                              S A G CORP.
     202/797-2525              Washington, D.C.           Fax: 202/797-2525

1    going 21 days satisfies your ethical needs or not.

2                   Intuitively,    I   would   say   somebody     gets

3    tremendously febrile and looks really sick, you get them

4    out, and start treating them, even though you don't really

5    know why that is happening, you just accept that.

6                   But that is really a clinical judgment.

7    clinicians have to sit down and say, okay, what scares

8    me, and what makes me worried about the fate of this

9    patient, and your obligation, and accompanying permission

10   to use a placebo where there is arguably at least standard

11   therapy, comes with some well-developed, mutually agreed

12   on criteria for what constitutes actions that would

13   protect the patient against going down the tubes.

14                  But in the absence of a lot of examples, it

15   is not easy to say what those are, and Dr. Bennett, who

16   doesn't understand this at all, raised all the right

17   questions, of course.

18                  But nobody really understands it.       There are

19   examples that are easy.       We have seen a withdrawal study

20   with -- never mind.     I am mixing two things.          We have

21   seen early escape associated with randomized withdrawal

22   studies, and that is probably the case where they have

23   been used most.

24                  And where people have looked at recurrence

25   of initial symptoms, and there have been cases where blood

                            S A G CORP.
     202/797-2525            Washington, D.C.         Fax: 202/797-2525

1    pressure over a certain point in non-responsive patients

2    who are being studied with a placebo got them out of the

3    trial and on to therapy.

4                   And you work it out on the spot, and I have

5    no doubt that these early escapes probably decrease the

6    apparent benefit of the drug.         It depends on why you leave

7    early.     But you pay that price for the ability to get

8    information in a setting where it is difficult to get

9    it.

10                  DR. FLEMING:        Or they could lead to an

11   exaggerated estimate effect if you are imputing failure

12   in the placebo, when in fact further follow-up of that

13   placebo patient would have led to a higher level of

14   success.

15                  My sense of interest in being able to do a

16   placebo controlled trial, I share that with others here

17   that it gives us in a real sense the truest way of

18   determining     whether     or    not        the   intervention      is

19   efficacious, is to do a head-to-head with the placebo.

20                  And if in fact we can reliably assess that

21   in short term follow-up in such a setting, the early escape

22   concept is appealing.       If in fact though we are not able

23   to follow the control patients adequately long through

24   the period in which we can get an unbiased assessment

25   of outcome, I think I would be more included to do a

                               S A G CORP.
     202/797-2525              Washington, D.C.            Fax: 202/797-2525

1    head-to-head comparison against a standard of care that

2    is largely anticipated to be relatively ineffective based

3    on what we are hearing from the data, at least in the

4    less ill patients, where you wouldn't have to escape.

5                   You could follow these people through 21 days

6    and really establish superiority.          So either doing a

7    head-to-head comparison against standard of care, or in

8    addition to standard of care, looking for superiority.

9                   And then if in fact we truly believe there

10   is interaction here indicating that there is adequate

11   data establishing the antibiotics are effective in those

12   patients that are more severely ill doing a separate

13   non-inferiority comparison in that population, those

14   approaches would be alternatives to early escape that

15   should also allow us to determine whether or not we have

16   truly added benefit relative to what is currently the

17   standard of care.

18                  CHAIRMAN RELLER:       Dr. Shlaes, Dr. Wittes,

19   and then Dr. Powers, and then we will have hands up again,

20   and we will get the next three.

21                  DR. SHLAES:    I just wanted to try to keep

22   this in prospective a little bit, at least for me.             So

23   I think that most drugs that are developed for AECB are

24   actually oral drugs that you would take as outpatients,

25   and so I don't think this is directed at those patients

                             S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    who just came out of the ICU and are coming back to the

2    hospital for another acute exacerbation, where they are

3    going to get admitted again.

4                   So I think it is really -- and to keep this

5    in perspective -- the outpatient setting.          The other

6    thing is that I think the 21 day evaluation was not 21

7    days of therapy.     It was just that that was the time,

8    and I think they pulled that number out of the air.

9                   I mean, I don't know why they picked 21 days

10   in that study, particularly if anyone knows, and maybe

11   Dr. Thompson knows why they picked 21 days in that study.

12    I don't know.

13                  But I think it was just a time when they could

14   bring patients back and get another FEV1 that was

15   realistic, but that is not 21 days of therapy.          So you

16   could have much shorter therapy, and withdrawal during

17   the shorter therapy, and still have a 21 day evaluation

18   for FEV1.

19                  And again I think the risk given outpatient

20   therapy, or early antibiotics, and hurting somebody with

21   a very severe disease would be small.

22                  DR. FLEMING:    By the way, I was assuming it

23   as you had indicated as well, that the end point is follow

24   everybody 21 days and find out what fraction resolved

25   their symptoms, which would be something that I would

                             S A G CORP.
     202/797-2525            Washington, D.C.      Fax: 202/797-2525

1    want to know whether somebody maintained therapy for 4

2    days, 8 days, or 21 days.

3                   And my concern is that if in fact natural

4    history would show resolution of symptoms, and the rate

5    increases as you follow people for a longer period of

6    time, such that 55 percent have resolved by 21 days, and

7    only 30 percent by 10 days, if we are pulling out in that

8    70 percent who haven't resolved by 10 days in the escape

9    clause, and hence impute non-success, then we are going

10   to have a final result of 30 percent success on an arm

11   that really should have had a 55 percent success rate.

12    That is the nature of the bias that I am concerned about.

13                  CHAIRMAN RELLER:      Dr. Wittes.

14                  DR. WITTES:    My comment has to do sort of

15   in general with this, with the valuable percentages which

16   have disturbed me today.

17                  And related -- and this is not unrelated to

18   the early escape, but it seems to me that in these AECB

19   trials, as in the others, I find this 35 to 50 percent

20   invaluable rate just too high.

21                  And somehow it seems to me that in order to

22   evaluate whether a therapy is working or not, there has

23   got to be a way of including end points for a higher

24   proportion of people.

25                  And in terms of early escape, and I fully

                            S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    agree with Tom, that the risk of this design in this sort

2    of situation, where you are evaluating 21 days as the

3    end point, if you have early escape designs, it may change

4    the end point.

5                   The end point may be time to more aggressive

6    therapy, or time to being able to be off it, or something

7    like that.     So that the design and the end point should

8    -- that the end point should help influence the way you

9    choose the end point.        It should not be locked into an

10   end point and then all designs say that.

11                  CHAIRMAN RELLER:      Dr. Powers.

12                  DR. POWERS:    I had a question for Dr. Fleming

13   that relates to something that Dr. Bennett said.              Often

14   times when we see people get discontinued from therapy,

15   it is hard for us to tell as medical reviewers why they

16   discontinued from therapy.

17                  And we used to get investigator comments or

18   a printout of handwritten or typed out as to what the

19   thinking of the investigator was at that point.           We don't

20   get that at all anymore, and so it is hard to tell why

21   they discontinued, and I often think that perhaps the

22   discontinuation is more of a measure of investigator

23   nervousness than it is of the patient actually doing

24   poorly.

25                  Would something like Dr. Bennett suggested

                             S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    firm rules for discontinuing patients address some of

2    the concerns that you raised about underestimating the

3    effect of placebo in those trials if you could at least

4    discern why the patients actually failed?                      Now, that

5    obviously brushes over the devil in the details of

6    determining what is a clinical failure in making those

7    rules, but would that address part of the problem that

8    you raised?

9                   DR. FLEMING:       Probably partially, but not

10   fully.     Just to follow the example that I was giving.

11   At 10 days, you have had 30 percent that have resolved

12   symptoms, and 70 percent haven't.              In that 70 percent,

13   of those 70 who haven't, eventually 25 will over the next

14   11 days if your criteria for escape are sufficiently

15   stringent that none of those would qualify, it would

16   resolve my concern.

17                  I kind of doubt though that you are going

18   to be that effective in being able to fully distinguish

19   who those 25 are from the other 45.                 And so I think it

20   would partially, but not fully, address the concern.

21                  CHAIRMAN RELLER:        Dr. Hardalo.

22                  DR. HARDALO:        I think you have actually

23   brought up some very important issues.                    First, as Dr.

24   Wittes     said,   the   evaluability        rate    is    one    of    the

25   challenges that industry has to deal with since we sponsor

                               S A G CORP.
     202/797-2525              Washington, D.C.                Fax: 202/797-2525

1    most of the clinical trials, and has a lot to do exactly

2    with investigator confidence.

3                   But it also has to do with the lack of clarity

4    that we see, and where we would want guidance from various

5    stakeholders, including IDSA, and the           American Thoracic

6    Society as to how do they define treatment failure.

7                   Is it failure to improve within the natural

8    history understood by them for that disease, or is it

9    clear    cut   deterioration    and        progression   based     on

10   objective criteria.

11                  That very much impacts exactly how can we

12   detail discontinuation rates.         But also it has a lot to

13   do with evaluability rates.         If there is no clear cut

14   objective criteria, what you have is patients coming off

15   the study for rather soft reasons, which makes them

16   unevaluable.

17                  You just simply don't have enough data with

18   your sample size to make any clear conclusions about the

19   efficacy of the drug, or the safety of the drug.

20                  In addition, there are a variety of factors,

21   not the least of which are the clinical practice.                  If

22   you are practicing in the United States, it is simply

23   impossible to have patients come back for daily visits

24   on an ambulatory basis.      It just is not going to happen

25   for most of the centers.

                             S A G CORP.
     202/797-2525            Washington, D.C.            Fax: 202/797-2525

1                      So you need to have a compromise as to what

2    is getting done in clinical practice, versus what is a

3    requirement for a clinical trial, so that you have good

4    quality data.

5                      And I think not the least of which is that

6    we also have to have assessments which are practical.

7    That although I really myself would like to have some

8    studies that require TAPS or quantitative cultures, in

9    reality, in managed care settings in the United States,

10   and   in       most   of   Western    Europe    and    Canada,     simply

11   microbiology has gone by the wayside because of the

12   emphasis on managed care that it ultimately does not

13   affect what is done to the patient in terms of the choice

14   of antibiotics.

15                     Therefore, the only microbiology data that

16   we do get is in the setting of clinical trials, and even

17   then it is going to be quite limited.                 So, yes, we would

18   love to discuss what would be relevant entry criteria,

19   and what would be relevant interim evaluability criteria

20   for discontinuation rules, and what would be relevant

21   end point data so that all of us can get the best quality

22   data from whatever sample size we agree upon.

23                     CHAIRMAN RELLER:         Yes?   Please, your name

24   and please comment.

25                     DR. TALBOT:        George Talbot, Barth.         Sorry.

                                  S A G CORP.
     202/797-2525                 Washington, D.C.            Fax: 202/797-2525

1     Hiding behind the water pitcher.

2                      CHAIRMAN RELLER:      If I put my glasses on and

3    I wouldn't need the introductions.              So help me out in

4    the afternoon.        Thanks, George.

5                      DR. TALBOT:   This is an awfully long way away

6    from you, and so I understand.          I have a general comment,

7    a big picture comment, as well as a specific suggestion.


9                      The big picture comment is that it is very

10   interesting to me to hear this committee talk about a

11   placebo controlled study design.              I think that that is

12   in some sense quite remarkable, and I would like to

13   compliment the FDA, and the FDA presenters for actually

14   presenting the group with the opportunity to break the

15   paradigm of clinical trials in this indication.

16                     I think that the opportunity this presents

17   for the community to learn about this disease and how

18   best to treat it is really quite remarkable.             So I think

19   it is a very good thing.        Now, the problem with breaking

20   the mode is that as you try to implement that, there may

21   be resistance to change.

22                     I could imagine resistance to change at the

23   level of IRBs, of Investigators, and of other concerned

24   groups.        So I think relative to some of the points that

25   have been made about violability, and about early escape

                               S A G CORP.
     202/797-2525               Washington, D.C.          Fax: 202/797-2525

1    designs, and so forth, that really it is incumbent to

2    take these discussions to a working group level so that

3    IDSA, and other groups of clinicians can offer the

4    specifics which allow these changes in design to be

5    implemented     safely,     appropriately,       and    with     the

6    confidence of the end users; that is, IRB's patients and

7    investigators.

8                   CHAIRMAN RELLER:        I would like to follow up

9    on Dr. Talbot's comments.          We heard earlier that some

10   of these patients who are marginal in terms of gas

11   exchange, and may be intubated, hospitalized, because

12   the acute exacerbation throws them over in terms of

13   respiratory pulmonary function.

14                  Would it be important if we are considering

15   placebo controlled trials to assure that those patients

16   don't have pneumonia with a negative chest radiograph,

17   so that we are really talking about acute exacerbations

18   of chronic bronchitis?

19                  And I was impressed in Dr. Thompson's review.

20    I am not at all convinced that if patients were -- we

21   had a randomized double-blinded control trial, with

22   appropriate supportive measures -- bronchodialators,

23   steroid use -- that we are at all confident that

24   antibiotics    contribute      much     or   anything   in     these

25   patients.

                               S A G CORP.
     202/797-2525              Washington, D.C.         Fax: 202/797-2525

1                   And if that be the case, I was also impressed

2    by this morning's discussion of all of the subtle,

3    sometimes      covert,     obtuse           pitfalls      in      these

4    non-inferiority trials.

5                   Wouldn't it possibly be much -- and the

6    dilemmas with the large number of patients, and the large

7    number of patients who were excluded because they can't

8    be a valuable.

9                   Would the practice of medicine be advanced

10   by just going ahead and demanding rigorous double-blind

11   placebo controlled trials for this entity as a more

12   efficient way to see whether or not a drug is effective

13   or not?

14                  And related to that is I am confused about

15   what it adds to have a placebo, a comparative agent, an

16   active control -- a new agent and an active control, and

17   a placebo, all in the same study, because it seems like

18   you are making things almost impossible to sort out when

19   you get into the discussion of deltas.

20                  Why not just do a placebo controlled trial

21   and get on with it?      Dr. Temple.

22                  DR. TEMPLE:    Let me partly answer that.              In

23   settings where you are convinced that certain drugs are

24   effective -- and depression would be a good example --

25   a three-arm study is an extremely informative study.

                              S A G CORP.
     202/797-2525             Washington, D.C.              Fax: 202/797-2525

1                   If you run the trial and your control agent

2    wins, and your new drug loses, you find another drug,

3    because you have learned what you needed to learn.             This

4    is a study that had assay sensitivity, and your drug could

5    not be shown effective in that study.

6                   If on the other hand both the control agent

7    and your drug fail, then the study couldn't distinguish

8    active from inactive drugs, and you don't have any reason

9    to be depressed.

10                  Now, here it is more complicated, because

11   from what I am understanding, nobody is entirely convinced

12   that any drugs actually work.              The only reason for

13   including -- there are two reasons for including the

14   active control.

15                  One is to -- and as Tom said and others did,

16   to see how the new drug actually compares with the other

17   drug in a setting where you establish assay sensitivity,

18   and that is not very important if you don't think they

19   work very well.

20                  The other is that in case that you really

21   in your heart believe this other drug works, this allows

22   you to distinguish from a setting in which you can't tell

23   anything from a setting in which you can tell things.

24                  So it can be an extremely informative design,

25   and that's why people in depression and hypertension,

                             S A G CORP.
     202/797-2525            Washington, D.C.          Fax: 202/797-2525

1    that is actually the standard test now.         Almost everybody

2    does it all the time.

3                   CHAIRMAN RELLER:        Dr. Glode.

4                   DR. GLODE:    I obviously cannot comment on

5    AECB as a pediatric infectious disease doctor, but I just

6    wanted to reiterate Dr. Talbot's points that have bothered

7    me, and that is the issue of the sort of standard of care.

8                   If   somebody      is       writing    12     million

9    prescriptions every year for this then patients and

10   doctors have some belief in antibiotics.               And so I am

11   very worried about the introduction of placebo controlled

12   trials relative to both the patient and the local IRBs,

13   and sort of the issue of if the FDA says it is fine, does

14   the world believe it, and are willing

15    to approve it.

16                  I think that is a big hurdle and that becomes

17   a big hurdle if people won't enter the trial, or if you

18   can't get it through your IRB.

19                  CHAIRMAN RELLER:        Dr. O'Fallon.

20                  DR. O'FALLON:    I think it is interesting that

21   -- well, I will just say my point.             We haven't really

22   made enough of a point that what is under the surface

23   of all of this is the overuse of antibiotics and what

24   we are concerned about is the coming disaster of overuse

25   of them.

                             S A G CORP.
     202/797-2525            Washington, D.C.             Fax: 202/797-2525

1                   So, in an issue like this one, or in a setting

2    like this, it may very well be that there are all these

3    prescriptions that are being written every year for

4    something that the drugs aren't helping, and we don't

5    have the data to prove that they either do or they do

6    not.    So there is an issue here to stave off this growing

7    wave of drug resistance.

8                   CHAIRMAN RELLER:      Drs. Ramirez, Cross, and

9    Chesney.

10                  DR. RAMIREZ:   I just have a question.     I have

11   no problem to do it with a patient with mild COPD, a

12   placebo-controlled trial, because I have not seen any

13   data in the literature that indicates that antibiotics

14   are better than placebo.

15                  And I am sure that I am not going to have

16   any problem to convince my IRB to say that if you have

17   a patient with COPD, which was described as just a clinical

18   entity.

19                  But if you have a patient with COPD with mild

20   exacerbations, and with just a couple of years of COPD,

21   and if that were more than 75 percent, then nothing is

22   going to happen to this patient if they don't take

23   antibiotics.

24                  And I am sure that at this moment I can

25   convince the patient that we are doing a trial to see

                             S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    if we can avoid giving you antibiotics and develop 5 years

2    down the road resistant organisms, and the patient is

3    going to be happy to be in the placebo arm.

4                   And then I have no problem, but the question

5    is that if I am in industry, and I come up with this new

6    antibiotic, who is going to pay for this study to test

7    my drug against the placebo?

8                   Everybody wants to test their drug against

9    the other drug, and to be sure that my drug is going to

10   be on the market.     I mean, how are you going to convince

11   the industry to do a study of a new antibiotic that is

12   going to be tested against a placebo?              Who is going to

13   pay for this?

14                  CHAIRMAN RELLER:       Let's continue around the

15   table, and we will get everybody, including Dr. Temple

16   and Dr. Nelson.      Alan.

17                  DR. CROSS:    I would like to just follow up

18   on a comment that Dr. Temple made about the three-arm

19   study, and about including an arm that has the, quote,

20   standard, drug.      In our last meeting on sepsis, a slide

21   was shown which the presenter made the point that there

22   were at least 4 or 5 drugs that in the first trial were

23   shown     to   be   effective,    which     upon    retrial      were

24   ineffective.

25                  And I am just wondering in the area of

                              S A G CORP.
     202/797-2525             Washington, D.C.            Fax: 202/797-2525

1    infectious     diseases      do    we    have    any       examples    of

2    antibiotics, which on repeated trials have had about the

3    same approximate point estimate of efficacy.

4                   And I guess a corollary to that is simply

5    the second point which you made earlier about assay

6    sensitivity,    and    can    we   measure      the    difference      in

7    effectiveness between drugs.

8                   But    at   least    from      what    we    heard     this

9    afternoon, there is even a more basic aspect of the issue

10   of sensitivity.       And that is diagnostic sensitivity,

11   especially when we talk about things like sinusitis or

12   bronchitis.

13                  And it appears that in the reviews that we

14   heard that there were various criteria for making a

15   diagnosis, such that it is really hard to even compare

16   most of these studies, even if you did have an answer

17   for my first question about reproducibility of results

18   in these specific areas.

19                  CHAIRMAN RELLER:         Dr. Chesney, and then Dr.

20   Temple.

21                  DR. CHESNEY:        Thank you, and I hope that I

22   can keep my thoughts organized here.                 But I would like

23   to echo a point that Gordon made, which is that

24   -- well, first of all, how did we get here.                 We got here

25   because colossal overuse of antibiotics by comparing one

                                S A G CORP.
     202/797-2525               Washington, D.C.             Fax: 202/797-2525

1    to another.

2                   And I think several points -- and number one

3    being, I don't think we know the natural history of a

4    lot of these diseases.    I don't think we know the natural

5    history of otitis media or sinusitis, or AECB, because

6    we began using antibiotics before people recognized my

7    second point, which is I think there are subsets.

8                   And very clear subsets within these groups,

9    and I think those of us in pediatrics could clearly

10   identify subsets of children who had acute otitis media,

11   and one of the big problems has been that they are all

12   just put together in these studies, and they don't

13   distinguish a two month old with a temperature of 106,

14   with an 8 year old with no temperature sometimes.

15                  And so I think that we really don't know the

16   natural history of what we are using the vast majority

17   of antibiotics for, and as that beautiful wheel diagram

18   from the CDC continues to demonstrate.

19                  So for me mild diseases is the real issue,

20   and I don't know how we are going to get some of these

21   answers without using placebo controlled studies.          And

22   I think a point that Dr. Talbot made that is so critical,

23   is to get the right players together.

24                  The people that are doing the double tap

25   studies on otitis media have some very well defined

                            S A G CORP.
     202/797-2525            Washington, D.C.     Fax: 202/797-2525

1    concerns and ideas about how to do these studies with

2    a very small number of patients, for example, for acute

3    otitis media, and we all heard Dr. Dagan a few months

4    ago.

5                   So I think getting the right people together

6    and looking at the issue of subsets, and readdressing

7    the whole issue of natural history for me are really the

8    big points.

9                   And determining what kind of delta to use,

10   or what kind of study to use, is obviously important.

11   But I think that is going to take a lot more discussion

12   within the smaller groups of right players if you will.

13    Thank you.

14                  CHAIRMAN RELLER:          Thank you.     Dr. Temple,

15   and Nelson, and then Metlay.

16                  DR. TEMPLE:       Presumably one of the reasons

17   studies come out differently is in fact the difference

18   in diagnosis, or the difference in the population that

19   got into a particular trial.

20                  If you had reason to believe that there was

21   an effective therapy, the effective therapy accompanying

22   the test drug helps you know whether this was a study

23   that got the right people into the trial or didn't.

24                  Now,   if    really      there   isn't     any    right

25   population, and we don't know whether any of this works,

                                S A G CORP.
     202/797-2525               Washington, D.C.           Fax: 202/797-2525

1    then that is a different question.               I just wanted to

2    comment on something that Dr. Talbot said.

3                   We have recently gotten through about a year

4    and a half in which many people, including the people

5    who wrote the Declaration of Helsinki, asserted that you

6    can't    use   placebo   controlled         trials   when   there    is

7    effective therapy, even for mildly symptomatic diseases,

8    a headache or something like that.

9                   So the discovery that FDA and the advisory

10   committee wants to have placebo controlled trials of

11   antibiotics for goodness sakes will draw attention.

12   There is on question about it.

13                  The answer I think lies in the very things

14   that you have been discussing.         You have real doubt about

15   whether people are being harmed or helped by this.

16                  You may be setting them up for a resistant

17   organism infections later that will take their lives.

18   So the case will be made on the credibility of those

19   assertions, and the lack of information about whether

20   there really is anything very effective.

21                  But it will draw tremendous interest.                      I

22   don't think there is any question about that from IRBs

23   and others who are very nervous these days about placebos.


25                  CHAIRMAN RELLER:       Dr. Nelson.

                              S A G CORP.
     202/797-2525             Washington, D.C.             Fax: 202/797-2525

1                      DR. NELSON:     I think actually a good lead-

2    in to my question as the Chair of an IRB, it is unclear

3    to me from a study design perspective that there is any

4    difference when you can't tell between the placebo and

5    an   active      control,   and     between    an   active     control

6    superiority trial and a placebo controlled superiority

7    trial.

8                      So I guess I am asking to be educated that

9    if indeed physicians like me in an ICU who probably

10   reprobate in the use of broad spectrum antibiotics as

11   my patient is deteriorating, or families who are not going

12   to be willing to go into a placebo controlled trial or

13   patients.

14                     And from a study design perspective, is there

15   any difference between the active control superiority

16   and placebo controlled trial in this kind of setting to

17   where you can have your cake and eat it, too, on both

18   sides, and placing the issue of resistance and over-use

19   aside.

20                     I mean, I am finessing that issue at the

21   moment.        Is there?

22                     CHAIRMAN RELLER:       Dr. Temple.

23                     DR. TEMPLE:       Well, Tom referred to this

24   before, too.        If there were reasons to think that one

25   drug was actually superior to another, then go ahead and

                                 S A G CORP.
     202/797-2525                Washington, D.C.           Fax: 202/797-2525

1    do a superiority trial.           That always works and it is

2    interpretable.

3                    The question is whether there is any reason

4    to believe that that is true, and if it is not, then a

5    superiority trial can't work, won't work, and there is

6    not much point in it.

7                    And   your     only      choice     is   to     do       a

8    non-inferiority trial, which Dr. Thompson explained can't

9    be done.       And sometime else, namely a trial against

10   placebo, with appropriate are that people don't get hurt.


12                   DR. NELSON:      But, Bob, if the placebo and

13   the active control are not shown different in any studies

14   that have been performed, then what is the difference

15   in selecting the active control over the placebo in that

16   context?

17                   DR. TEMPLE:     No, I agree with you.         If there

18   is no reason to believe any of these things work, then

19   there is not much point in not just going ahead and doing

20   a placebo control trial, and only if you think that some

21   of them do work in the right setting is there a reason

22   to have that.

23                   CHAIRMAN     RELLER:          Did    I   understand

24   correctly, Dr. Nelson, that you are suggesting that why

25   not always have an active control if it really is

                                S A G CORP.
     202/797-2525               Washington, D.C.          Fax: 202/797-2525

1    tantamount to a placebo.      Is that what you are saying?

2                   DR. NELSON:   No, no, I wouldn't want to go

3    that far.

4                   CHAIRMAN RELLER:      Because if that be the

5    case, then if we could think of some examples, then we

6    would have examples of the very thing that initiated this

7    whole delta discussion.

8                   DR. NELSON:   Well, placing the mild issue

9    aside, if you want to carry this into a more severe disease

10   setting, it is unclear to me if the argument that you

11   can't determine a delta is based on the lack of difference

12   or reproducible difference between the placebo and an

13   active control in existing studies, it is unclear to me

14   that from a study design perspective there is any

15   difference then whether or not the control group is an

16   active agent, or the placebo agent, based on those prior

17   studies.

18                  And so if indeed you are arguing on a

19   feasibility that patients, families, and physicians,

20   would be more accepting of an active control from a study

21   design perspective alone, it is not clear to me there

22   is any advantage of the placebo group.

23                  That is the question that I am asking, as

24   much as wanting to be educated from that, so that your

25   feasibility would actually be improved by the active agent

                            S A G CORP.
     202/797-2525           Washington, D.C.       Fax: 202/797-2525

1    if you did a superiority trial.

2                   When I read E-9 and E-10, which I read to

3    be educated, I see a lot of discussion about a superiority

4    design is superior to the equivalence design.             So it is

5    unclear to me why that is constantly being sort of placed

6    aside.

7                   CHAIRMAN RELLER:      Dr. Temple.

8                   DR. TEMPLE:    I'm sorry to keep doing this,

9    but the distinction is between -- you can have an effective

10   drug for which you nonetheless can't describe a delta.


12                  We have thought about anti-depressants for

13   a long time, and about half of the satisfactorily designed

14   trials of drugs we know to be effective can't distinguish

15   drug from placebo because the diagnosis is different or

16   people get better.     Nobody knows why.

17                  But it is a fact, and which means that in

18   any given study that you can't know what the effect of

19   the active drug is, even though we are perfectly convinced

20   that those drugs work.

21                  And the situation here could be none of them

22   work at all, and none of them are known to work, and there

23   is no evidence of anything; or it could be that it is

24   study dependent.

25                  That is, that if you get just the right

                            S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    people, maybe it works then, and things like that.                Those

2    are reasons why you can simultaneously not write or not

3    design a delta, not identify a delta-one, but might find

4    it useful to include a putative active drug as a control.

5                    You would never need to do that, but it might

6    be informative, too.         But the differences between the

7    assurance of assay sensitivity in any given trial, and

8    the overall effectiveness of a drug.                  There are many

9    effective drugs for which you cannot design or describe

10   a delta, a delta-one.

11                   CHAIRMAN RELLER:        Dr. Fleming.

12                   DR. FLEMING:       I think what Dr. Nelson is

13   raising is a very important point, and if I am following

14   what he is suggesting here, is that it is in a setting

15   where standard of care is widely accepted, but thought

16   to have relatively little impact on the end point, either

17   favorably or unfavorably.

18                   And then is it ethically more comfortable

19   and easier to enroll in a robust fashion by randomizing

20   patients       to   that   standard       of   care     against      the

21   experimental, where you still have to show superiority.

22                   So you don't run into where we run into

23   troubles and if you are trying to show non-inferiority

24   there, it is not acceptable because there is no legitimate

25   margin.

                                S A G CORP.
     202/797-2525               Washington, D.C.            Fax: 202/797-2525

1                    But I think what you are saying is if you

2    are truly intending to show superiority, isn't that an

3    alternative approach to doing a placebo controlled trial

4    that might be more ethically acceptable, and might allow

5    for more rapid enrollment.

6                    And my own sense about this is in fact it

7    is, and it is not unlike the concept of doing dose

8    response, giving a low dose and a high dose, where you

9    are hoping that there is a gradient there such that the

10   high dose is much more effective than the low dose.

11                   And the risk to this approach is only if in

12   fact the active comparator really is more effective than

13   you think, and it is absorbing a fair amount of the

14   efficacy of the experimental; or if it is adverse, and

15   you are not recognizing that.

16                   Many examples of this exist.                 Just one

17   example of a trial that we were involved in, which was

18   looking at reducing maternal-to-child transmission of

19   HIV in developing countries, where the standard of care,

20   when we did this study a few years ago, was still placebo.

21                   And we designed a placebo controlled trial

22   against a short-course AZT regimen against a short-course

23   novarepine regime.      Ethics Boards eventually closed down

24   the   placebo    arm,   but   allowed       you   to   continue      the

25   short-course AZT, short-course novarepine comparison.

                              S A G CORP.
     202/797-2525             Washington, D.C.              Fax: 202/797-2525

1                   And the short-course novarepine have the

2    transmission rate of HIV relative to short-course AZT,

3    an example of what you are talking about.            Now, maybe

4    the actual effect of short-course novarepine is even more

5    than a halving, but it is sufficiently more potent that

6    we were able to show a difference in a trial where it

7    was judged ethical, because everybody was getting an

8    active intervention.

9                   So if in fact you believe that there is

10   considerable uncertainty about whether the standard of

11   care is effective, but it is widely accepted, and there

12   would be serious concerns about doing a placebo, you could

13   do a head-to-head superiority comparison against that

14   active comparator.

15                  And as long as it is relatively inert, in

16   terms of efficacy and risks, you would actually get an

17   informative sensitive answer to whether the experimental

18   therapy is effective.

19                  CHAIRMAN RELLER:      To continue the train of

20   discussion, we went a couple of circles.            Dr. Metlay,

21   your turn, and then we will come back to the floor table.

22                  DR. METLAY:   Thanks.       Well, first a comment

23   on this discussion, which is that I don't think the issue

24   is so much that these agents are ineffective, but that

25   they are effective on subsets of patients that we can't

                            S A G CORP.
     202/797-2525            Washington, D.C.          Fax: 202/797-2525

1    readily identify.

2                     And I think that is really the problem

3    practically speaking.        That said, I think that the idea

4    of a placebo controlled trial is very appealing.                  There

5    are some practical problems, and two of them have already

6    been sort of teased out a little bit.

7                     One of them is this issue when you said if

8    we could just exclude the patients who have pneumonia

9    from the AECB trials, and yet we are learning increasingly

10   so   that      that   distinction,    at      least   even   based    on

11   radiographic evidence, is problematic.

12                    And I think that one could argue that part

13   of the problem is, of course, that the way that we have

14   created these diagnosis based on some relatively arcane

15   tests now is really not the right way to guide therapy.

16                    But nevertheless we are sort of stuck with

17   them for the time being, and we are going to have to realize

18   these limitations as we start to think about actually

19   giving people placebos.

20                    The other issue is that I would agree that

21   we really have to use clinical outcomes as the measures

22   in these respiratory infections, and there I think we

23   have a disconnect that we are going to have to deal with,

24   in terms of enrollment, and IRB issues, and this escape

25   issue.

                                S A G CORP.
     202/797-2525               Washington, D.C.            Fax: 202/797-2525

1                     And that is this belief that in fact people

2    get better as they complete their therapy, when in fact

3    the observational data would suggest that people's course

4    of recovery is actually quite prolonged.

5                     And I am always sort of amazed by the clinical

6    trial data that suggests the proportion of people who

7    are better by seven days, when you go out and sort of

8    measure this in the real world if you will, and recognize

9    how long it takes for people to get better.

10                    And the consequence of that is that if you

11   are in a trial in which there is a placebo, most people

12   are   not      going   to   be   better in a shorter or even

13   intermediate period of time.

14                    And so I think there is going to be a lot

15   of emphasis on escape or switch.                It is going to be hard

16   to resist that, unless we sort of significantly change

17   the understanding of what we do know about the natural

18   history of the disease, and which is as I would say in

19   general that it is a lot longer than most people think.

20                    CHAIRMAN RELLER:         Dr. Chesney.

21                    DR. CHESNEY:       As Dr. Archer pointed out, I

22   also am statistically challenged, and I wanted to ask

23   Dr. Fleming that in your novarepine-AZT example, you

24   called that a superiority trial.                 How does that differ

25   from -- you know, this is very fundamental I'm sure, but

                                 S A G CORP.
     202/797-2525                 Washington, D.C.           Fax: 202/797-2525

1    how was that different from a non-inferiority comparison?

2                     DR. FLEMING:     Well, the analysis essentially

3    was looking at differences in transmission rates of HIV

4    maternal-to-child, and one of the primary end points was

5    at   six       weeks.     And     the    novarepine        reduced       the

6    transmission rate from -- I think it was from 21 percent

7    on AZT, to 11 percent on novarepine.

8                     By achieving statistical superiority, we

9    were able to conclude that single dose novarepine was

10   very effective, and at least provided that 50 percent

11   reduction,       possibly      more,    if     short-course       AZT    was

12   effective.

13                    If in fact those two rates had both been 11

14   percent, then the difficulty that we would have had is

15   we wouldn't have known whether they were equally effective

16   or equally ineffective.

17                    And that was the loss of not being able to

18   have the placebo arm in that trial.               So the only way that

19   study was able to conclusively establish benefit was by

20   having a superiority difference.

21                    If they had been the same, we would not have

22   known    if     they    were    equally        effective     or    equally

23   ineffective, because there was no predefined margin that

24   would have allowed us what short-course AZT did.

25                    If we had known that short course AZT halved

                                 S A G CORP.
     202/797-2525                Washington, D.C.               Fax: 202/797-2525

1    the transmission rate, and we saw comparable rates between

2    novarepine       and   AZT,   then    we     could   have     done       a

3    non-inferiority comparison.

4                     CHAIRMAN RELLER:      Mark.

5                     DR. GOLDBERGER:      Just a couple of things.

6    One is in terms of really thinking about the placebo issue

7    for AECB, it is probably worth you hearing where we are

8    in terms of what is actually being done in trials, and

9    i.e., the big trend in AECB, as it is in some other

10   infections now, is to shorten the duration of therapy.

11                    The last submissions to come into our office

12   I think, one is 5 days of therapy, and I think there may

13   be one, although with a loner half-life drug, as short

14   as 3 days of therapy.

15                    So in fact in terms of thinking about early

16   escape, we may be somewhat almost pass that if the duration

17   of active therapy is so short.          Perhaps the question may

18   have to be in some of these regimens can we say something

19   at the conclusion of the period of when the active drug

20   was given of, of active drug versus placebo, that would

21   sufficiently informative to help us in determining

22   whether that person on placebo ought to receive therapy.

23                    I mention that as an observation.          It is just

24   another issue as I see that Dr. Fleming is eager to

25   respond.       Well, it is good to see that at this late hour

                               S A G CORP.
     202/797-2525              Washington, D.C.           Fax: 202/797-2525

1    of the afternoon I have to say.

2                   There was some discussion about maybe end

3    points ought to be just keeping a person stable, and I

4    think that if we start thinking about that in at least

5    more severe disease, where at least there may be more

6    comfort antibiotics doing something, I think that is

7    something that may be worth talking about, or thinking

8    about a little bit.

9                   I mean, coming from the old school that the

10   goal of antibiotics and infectious diseases is really

11   to cure or very significantly mitigate infection, keeping

12   people stable makes me wonder a little bit.

13                  Plus, the illness that we are talking about

14   is acute exacerbation, and if exacerbation means getting

15   worse, you would like to think that something actually

16   could improve things.

17                  And with that, I yield the rest of my time

18   to Dr. Fleming, if that is okay.

19                  DR. FLEMING:   Well, just two quick thoughts.

20    First, I would like to distinguish between the time that

21   somebody is on a therapy and the time period over which

22   that administration could affect their outcome.

23                  Somebody might have been on therapy for 3

24   days, but the influence of that on their outcome might

25   not be fully known until some period of time beyond 3

                            S A G CORP.
     202/797-2525            Washington, D.C.      Fax: 202/797-2525

1    days.

2                   Secondly,   my    concern       in   many    clinical

3    settings with looking at end points that are very short

4    term, is that they may be missing the more global and

5    clinically relevant aspect here.

6                   And if we come back to here, and if we are

7    using, for example, 21 day periods for resolution of

8    symptoms, if we look over two days, we may get a relevant

9    comparison over two days, but that may only be the tip

10   of the iceberg of what really matters to patients.

11                  And I would argue that the clinical endpoints

12   as best possible should capture the essence of what

13   matters to patients, and so that factor should influence

14   as well how long we have to follow.

15                  DR. GOLDBERGER:      I would certainly say --

16   and if you don't mind my taking back the last little nibit

17   of my time, that I would certainly agree with you on the

18   second point.

19                  There is value I think in having some of these

20   longer term outcome measures.         Acutely, one might argue

21   that if in fact the duration of antimicrobial therapy

22   is so short that having the early escape at the end of

23   that, there is perhaps a little        less worry about giving

24   placebo for only several days.             That is what I am sort

25   of wondering about.

                             S A G CORP.
     202/797-2525            Washington, D.C.             Fax: 202/797-2525

1                   Does that pose as much of a problem when we

2    know that the active drug will be terminated at day 3

3    or day 5, and should we worry therefore as much about

4    the consequences of using placebo.

5                   CHAIRMAN RELLER:      I would be interested in

6    hearing perhaps additional comments from IDSA, Dr.

7    Talbot, and others, and from PhRMA, Dr. Shlaes, and others

8    about where appropriate, should there be -- and leaving

9    aside what indications that those might be.

10                  But should there be greater consideration

11   of the role of placebo controlled trials, looking at the

12   issues that Dr. Chesney pointed out, and I am impressed

13   as the discussions have gone on today that what started

14   out as an emphasis on one thing may be as bringing into

15   consideration that there are a lot of other issues that

16   may help us get to where we want to be, having to do with

17   what is the best way to assess efficacy, and recognize

18   safety in the approval and study of new antimicrobial

19   agents.

20                  Any comments, Dr. Shlaes, or Dr. Talbot, or

21   others?

22                  DR. SHLAES:   Well, I mean, I think we would

23   certainly be interested in placebo controlled trial

24   designs, assuming that they were ethical, and that we

25   could carry them out, and that people would accept them.

                            S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1                     I   think that we mentioned that in our

2    presentation this morning.                 So I think we are open to

3    that.     Obviously, they would have to allow us to carry

4    out   the      trials   in     a    way    that   provides    meaningful

5    information to all concerned.

6                     But we are certainly interested in looking

7    at placebo controlled trials, absolutely.

8                     DR. ANDRIOLE:             I would be interested in

9    looking at placebo controlled trials in this area, too,

10   because I think the other way to do it is to go with the

11   treatment control, and if you don't measure or don't feel

12   superiority, you are not going to get approval in that

13   area, but the other drug already has it.

14                    It is kind of setting up            a straw dog.        So

15   I think if it is really a question that that drug has

16   any effect, I think I would rather go to a placebo

17   controlled trial if it could get through an ethics

18   committee.

19                    CHAIRMAN RELLER:           Dr. Talbot.

20                    DR. TALBOT:             Yes, thank you.        There are

21   difficult economic questions perhaps for the development

22   of new drugs, but I think that the societal risk benefit

23   issue    requires       that       the    consideration    of    studies,

24   including placebos, be discussed not only today, but again

25   and again.

                                  S A G CORP.
     202/797-2525                   Washington, D.C.           Fax: 202/797-2525

1                     And that some solutions be reached so that

2    health care providers in the U.S. can be certain that

3    they are giving effective drugs and not creating a public

4    health risk, in terms of antimicrobial resistance.

5                     Now, I do have to add a disclaimer that I

6    am speaking for myself, and I am not sure that I am speaking

7    for IDSA.

8                     CHAIRMAN RELLER:      Dr. Sumaya.

9                     DR. SUMAYA:    One issue which I have heard,

10   but maybe not as strong, is where do we focus our energies?

11    Do we focus the energies toward looking at trials in

12   the mild, moderate group of patients, or should we focus

13   major energies on the severely ill?

14                    And can we do that altogether in one trial,

15   or do we have to separate that, or do it in stages, or

16   phases?        My prior experiences are that you go to the

17   severe, and then you go to the mild.

18                    In this case, I am not so sure about that,

19   because the mild brings in more things with overuse,

20   potential       resistance,    but    the    severe     deals      with

21   potentially greater mortality issues, and disease burden,

22   and complications.

23                    What I see is that the all-need criteria needs

24   area need to be much better defined, and criteria for

25   entry into any trial, for monitoring during the trial,

                               S A G CORP.
     202/797-2525              Washington, D.C.            Fax: 202/797-2525

1    and for the end points.

2                   So, obviously uniformity , clarity, and those

3    definitions across all those high areas would be very

4    important.     If the energies go towards the mild form,

5    mild to moderate, then I think a placebo control makes

6    very good sense.

7                   If we go more toward the severe forms, then

8    I think some type of comparison, perhaps the standard

9    care as Dr. Fleming had mentioned, versus a test drug,

10   would be the most appropriate.

11                  But again where do we focus the industry

12   focus?     Is it a mild to moderate issue, and/or the severe.

13                  CHAIRMAN RELLER:       Dr. Ramirez, and then Dr.

14   Leggett.

15                  DR. RAMIREZ:      Yes.       I think if we have an

16   infectious disease, and the infectious disease is caused

17   by bacteria, antibiotics will always be beneficial.

18                  Then the question is that we know that a

19   patient with mild COPD has bacteria in the airwave, but

20   we don't know if this is an infectious disease.            We don't

21   know if bacteria are part of this cycle or inflammatory

22   process.

23                  Then we are asking the industry to define

24   a clinical question.       Is a patient with a mild acute

25   exacerbation of COPD having an infectious disease, and

                             S A G CORP.
     202/797-2525             Washington, D.C.           Fax: 202/797-2525

1    are antibiotics necessary.

2                   And I think we are here as doctors, and the

3    only mention of this is the industry, and there is the

4    agency, and there are clinical investigators.             This is

5    a great question for clinical investigators.

6                   Do we need to use antibiotics in patients

7    with mild to acute exacerbations of COPD?             But I still

8    don't understand why I need to ask a drug company to

9    generate a new antibiotic and trying to test a basic

10   question to see if a person with a disease requires

11   antibiotics.

12                  I want to ask the industry do answer the

13   question if this person has an infectious disease.                    I

14   mean, this is not supposed to be the industry.            This is

15   supposed to be the clinical investigators answering the

16   question.

17                  Once we find that this is an infectious

18   disease, and the patient has a bacterial infectious

19   disease, then we decide to use a antibiotic.                     We

20   understand     that   acute   bronchitis    is   an    infectious

21   disease, and is caused by viruses.

22                  And we are not asking the industry to give

23   us antibiotics for acute bronchitis.        We just closed the

24   case.      The problem is that we don't know if mild

25   exacerbation of chronic bronchitis is still an infectious

                             S A G CORP.
     202/797-2525             Washington, D.C.         Fax: 202/797-2525

1    disease.

2                   CHAIRMAN RELLER:      Dr. Leggett.

3                   DR. LEGGETT:    One point that maybe I didn't

4    understand, but the most severe definition of the criteria

5    was cough and purulent sputum.             I mean, to me that is

6    not very severe.

7                   So in other words, I think that just talking

8    about the COPD patient in the ICU is three standard

9    deviations away from the first that I think of as even

10   having an AECV.    Maybe I didn't understand.

11                  CHAIRMAN RELLER:      Before moving to question

12   two, does anybody have anything additional they wish to

13   say about acute otitis media or acute sinusitis?                 Dr.

14   Chesney.

15                  DR. CHESNEY:    Just one quick thing.        I think

16   I terms of thinking of the natural history, we don't know

17   the natural history of resistant organisms in acute otitis

18   media and sinusitis.     And we have good reason to think

19   that it wouldn't be different.

20                  But I just wanted to make that point, that

21   we are dealing with new infections to some degree here

22   by very resistant organisms.

23                  CHAIRMAN RELLER:      Dr. Chesney, do you think

24   it is possible to assess efficacy of new or existing agents

25   against resistant pathogens, especially streptococcus

                            S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    pneumoniae, without tympanocentesis puncture studies

2    with sinusitis?

3                   DR. CHESNEY:    No.

4                   CHAIRMAN RELLER:      Dr. Talbot.

5                   DR. TALBOT:    I have one comment about the

6    Chairman's comments about AECB if I could to follow up

7    on Dr. Ramirez's point?

8                   CHAIRMAN RELLER:      Please.

9                   DR. TALBOT:    I think you raised a very good

10   point.     As Dr. Thompson mentioned the current conundrum

11   with AECB is that there is no generally accepted study

12   at this point that definitively proves that active

13   antibiotic therapy is better than no treatment, or

14   placebo.

15                  So let's say theoretically that such a study

16   was done that conformed to all appropriate statistical,

17   and clinical, and regulatory standards.        And Antibiotic

18   A was shown to in fact be superior.            Would that not

19   potentially obviate the need for successive placebo

20   controlled trials?

21                  Or would the committee think that AECB is

22   inherently more like depression with a lot of variability

23   in presentation and clinical course, such that even after

24   that first demonstration there would be a continued need

25   for placebo controlled trials?

                            S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1                   So is it just one that is needed, or does

2    there have to be a uniform and continuing inclusion of

3    placebo controlled trials?         And I don't know if that

4    answer is known at the moment.

5                   CHAIRMAN RELLER:      Thank you.    Dr. Ebert.

6                   DR. EBERT:   I just wanted to comment briefly

7    on the third part of the question on the dose response

8    trials, and I am a little bit unclear as to exactly how

9    that fits in here.

10                  But I think in general I would be somewhat

11   leery about using dose response trials as a measure of

12   efficacy without good pharmokinetic/pharmodynamic data

13   to form the basis for those clinical studies.

14                  And given what we have talked about so far,

15   and the possibilities of drugs being either equal to

16   placebo or not showing a clear definition, I would be

17   a little bit concerned that we may find in a dose ranging

18   that a, quote, subtherapeutic dose does in fact show some

19   clinical efficacy.

20                  And subsequently would just contribute to

21   a use of the drug at that dose, which might lead down

22   the line to resistance because of an in essence a

23   subtherapeutic dose.

24                  CHAIRMAN RELLER:      Dr. O'Fallon.

25                  DR. O'FALLON:    I am concerned about the fact

                            S A G CORP.
     202/797-2525            Washington, D.C.          Fax: 202/797-2525

1    that we really aren't talking much about the possibility

2    that a treatment can actually be damaging.

3                   And I am thinking not so much about a short

4    term basis, but rather say that a treatment clears things

5    out, but then it leaves a patient at risk to have a prompt

6    recurrence, or a more difficult recurrence, or something

7    of that sort.

8                   I mean, I don't know the diseases well enough,

9    but that in planning these studies, we should be open

10   to the idea that actually a treatment might be damaging.


12                  Now, the second thing is about the dose

13   response or something.       You know, they can do 3 and 4

14   arm studies, with one of them being a placebo, and the

15   others being 2 or 3 supposedly active, and ones that are

16   believed to be active therapies and that can be done in

17   one.

18                  That would probably require the industry to

19   cooperate, but they could get it done.            They could get

20   a lot more done with one study perhaps, one large one.

21                  CHAIRMAN RELLER:            Thank you.    The last

22   comment on question one.

23                  DR. FINK:    This relates probably to many

24   studies of lung disease.      I think you have to be careful

25   when we start talking about the value of PK/PD data.

                             S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    It is almost always blood levels, and penetration into

2    airways and the lung parenchyma itself may bear no

3    relationship to PK/PD data from the blood.

4                    So I think if we are really going to try and

5    use PK/PD data to extrapolate, you would have to talk

6    about doing it in experimental animals where you have

7    bleed them out, and then sacrificed them, and actually

8    measured tissue penetration and clearance from the lung

9    tissue itself, which has rarely, if ever, been done.


11                   CHAIRMAN RELLER:          Thank you.

12                   DR. GOLDBERGER:         Dr. Reller, would you want

13   to summarize or attempt to summarize what you have heard

14   as to question one?          It is always a big help for us.

15                   CHAIRMAN RELLER:         Actually, Mark, I was gong

16   to propose that as the transition sentence and do next

17   number two.      We were asked to comment in relation to the

18   proposed       approach      for     selection        of      delta      and

19   non-inferiority (equivalence, clinical trials).

20                   And what I have got out of hearing all of

21   this    discussion      is   perhaps       as   important,        or   more

22   important, is the delineation in acute exacerbations of

23   chronic bronchitis.

24                   Those     patient      groups    or   definitions         of

25   disease -- and it may well be all such patients with acute

                                  S A G CORP.
     202/797-2525                 Washington, D.C.              Fax: 202/797-2525

1    exacerbations of chronic bronchitis -- is consideration

2    of placebo controlled trials, or superiority trial

3    design, and not the equivalence design in the first place,

4    and so rather than getting constrained by what should

5    the delta be, is to consider the nature of the trial design

6    in the first place in which patients are included.

7                     And secondly that for other respiratory tract

8    infections, especially acute otitis media, and sinusitis,

9    that smaller numbers of patients with knowing exactly

10   what you start out with, and what you end up with, with

11   the importance of emerging resistance, would be far more

12   useful     in    delineating      efficacy       of   new    compounds,

13   including       ones   for    resistant         organisms,     than   the

14   discussion of -- and not that it is not important.

15                    But again spending the emphasis on the delta

16   and power in non-inferiority equivalence trials.

17   Or to put it another way, that the precise entity being

18   studied, and what is the best trial design, and what would

19   be reasonable assurance of efficacy in the first place,

20   may be more productive than simple discussion of delta.


22                    Question      Number       2.        Please     discuss

23   implication of the choice of deltas in clinical trials

24   for serious infections, and include in our discussions

25   efficacy of new drug compared with currently available

                                  S A G CORP.
     202/797-2525                 Washington, D.C.           Fax: 202/797-2525

1    measurements        for    hospital-acquired            pneumonia       and

2    meningitis.

3                      And I think that these entities are so

4    different as has been amply pointed out that we should

5    consider them separately.            So let's take perhaps the more

6    -- well, let's just take meningitis first.

7                      Trial design for meningitis, where one of

8    the messages that we heard clearly from IDSA, and from

9    industry, from Dr. McCracken's presentation, is there

10   is a very serious clinical entity with grave consequences,

11   the number of patients involved is small, and some of

12   the design considerations with the non-inferiority trials

13   for a level of confidence looking at clinical outcomes

14   would    require     numbers      of    patients       that   are   either

15   clinically or economically, or both, not reasonable.

16                     So how do we assess with meningitis?                 What

17   is the most efficient approach to establish efficacy and

18   safety with new drug development?                  Comments from the

19   committee.        Dr. Bell.

20                     DR. BELL:     I think the best insight I heard

21   was your comment earlier about perhaps using different

22   deltas for different outcome variables if I heard that

23   right; microbiologic, clinical.

24                     You know, I am very concerned that when you

25   have    a      serious   infection      that     you   don't   want     the

                                 S A G CORP.
     202/797-2525                  Washington, D.C.            Fax: 202/797-2525

1    comparator drug to be much less effective than the

2    standard.

3                   CHAIRMAN RELLER:       Now, one way of perhaps

4    getting at this and zeroing in on it is that one could

5    talk about whether it is at 24 hours, or 48 hours, or

6    both, and is the committee, the ISDA, PhRMA, others, are

7    we in agreement that unless one can sterilize the CSF,

8    one doesn't have a drug for meningitis?

9                   DR. SHLAES:    Yes.    We actually talked about

10   this over lunch, and we were saying that if in fact you

11   had a drug that didn't do that, then probably you would

12   stop development pretty quickly.

13                  CHAIRMAN RELLER:       What beyond that -- and

14   the precise numbers, and timing, and how to assess that

15   could be -- those details could be worked out.

16                  I mean, it would require obviously not only

17   an initial diagnostic effort, and you would have to have

18   a    repeat    lumbar   puncture,          and   assure    adequate

19   microbiology that people would accept as being rigorous,

20   decent, and something akin to the tympanocentesis -- TCs

21   -- and sinus punctures.

22                  But having that, what in addition, in terms

23   of trial design, follow-up, numbers of patients, deltas,

24   would be wise to have?       Dr. Ramirez.

25                  DR. RAMIREZ:     In prior meetings, we were

                             S A G CORP.
     202/797-2525            Washington, D.C.            Fax: 202/797-2525

1    discussing     sometimes      the     lack      of     correlation      of

2    microbiological resistance and clinical deterioration.

3     And we always blame the consideration that we did new

4    composition in the lungs, and any antibiotic gets good

5    penetration in the lung.

6                   And in the presentation this morning, Dr.

7    McCracken mentioned that quinolones for meningitis is

8    going to be a reality, and the reality is because of

9    streptococcal pneumonia resistant to penicillin.

10                  We tend to agree that this is the area where

11   we are going to see the single failures, and our

12   pediatricians are telling us that they are failures with

13   cephalosporin,    and      there    has       been    some   delay    with

14   vancomycin.

15                  At least in our Children's Hospital now the

16   empiric therapies is cephalosporin, vancomycin, and

17   rifampin, until you prove that the pneumococci infection

18   has been resolved.

19                  Now,   we    have     the       quinolones,      and    the

20   quinolones are supposed to have good penetration and are

21   supposed to have good activity against the streptococcal

22   pneumonia.

23                  And to me this is the idea situation to prove

24   superiority.    I mean, you cannot be as bad at the third

25   generation     cepthalosporins,               again     in    resistant

                                S A G CORP.
     202/797-2525               Washington, D.C.              Fax: 202/797-2525

1    pneumococci, because otherwise, why try the quinolones.


3                   I mean, to me this type of trials is trying

4    to achieve superiority and resolve the problem of the

5    delta, and resolve the problem of the number of patients,

6    I think we should look for superiority in trials of

7    meningitis in pediatrics and looking for the pneumococci

8    resistance, because this is why we want to use the

9    quinolones in pediatrics.

10                  CHAIRMAN RELLER:      Dr. Archer.

11                  DR. ARCHER:    I would like to kind of raise

12   another issue.    As was brought up, as the rate incidence

13   of meningitis decreases in this country with vaccinations

14   and so forth, and in virtually all of the cases are

15   recruited from abroad, it may have increasingly less

16   relevance for what we do in this country, in terms of

17   practicing medicine.

18                  It may in fact be that bacterial endocarditis

19   might be a better example of a rare infection that meets

20   the same criteria that in fact meningitis does.

21                  That is, that you have got bacteriologic end

22   points, and you have got clinical end points that are

23   very clear.    It is a disease that is not decreasing in

24   this country and you probably could enroll enough patients

25   just in this country alone with our standard of care to

                             S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    affect a disease that would be relevant.

2                   That is, we will continue to see it; as

3    opposed to meningitis, which we hope will become less

4    and less relevant.    So as a paradigm, endocarditis might

5    actually be a better paradigm for this kind of delta

6    consideration than meningitis.

7                   And I wondered if anybody from industry had

8    any comments about that?

9                   CHAIRMAN RELLER:      Dr. Hardalo.

10                  DR. HARDALO:      Two points.    First, about

11   meningitis.    I think as Dr. McCracken adequately pointed

12   out, there are certain factors that are beyond the control

13   of the treating physician, not the least of which is the

14   duration of symptoms before the onset of effective

15   therapy.

16                  In order to prove superiority for any other

17   outcome other than bacterial eradication, we need to have

18   some clarity as to how do we standardize the populations

19   that we are studying so that when we look at differences

20   from the time of symptom onset to the time of initiation

21   and treatment that we can compare the end points.

22                  It would be useless to compare drugs if one

23   study population had a delay in treatment of four days,

24   and another study population had a delay of one day or

25   one hour.

                            S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1                     And you will never be able to do a reasonable

2    comparison of the superiority trial in that type of a

3    setting.       The second would be that although I would like

4    to believe that pneumococcal disease is going away in

5    the United States, he did show evidence that it clearly

6    is not, even with the advent of vaccines.

7                     The only thing that vaccines really have done

8    is reduced H. influenzae, but not necessarily taken care

9    of some of the other pneumococcal diseases.               So it will

10   have to be something that we do study in the United States,

11   as well as rely on data from our colleagues abroad.

12                    And I think there it really becomes again

13   an   issue      of   the   training      of    the   investigators,

14   understanding what is reasonable natural history of the

15   disease, and criteria for discontinuation, as well as

16   criteria defining failure.

17                    As was said, there are certain aspects of

18   the natural history that some investigators feel are

19   failure, but clearly are not.                 So I think it is a

20   standardization with input from the key stakeholders like

21   IDSA, and specialists in pediatrics, to understand what

22   should be the entry criteria, and getting into the study,

23   and what are the definitions for treatment, failure, or

24   progression, so that when you go to a superiority design,

25   we are all talking the same language, in terms of being

                                S A G CORP.
     202/797-2525               Washington, D.C.           Fax: 202/797-2525

1    able to determine efficacy and superiority.

2                   For endocarditis, I agree.             The time has come

3    that we need to look at the same types of cidal therapy

4    for determining drugs that are better than what we

5    currently have in the armamentarium.

6                   But    again      it     is     distinguishing         the

7    inflammatory     sequelae       of      disease       from     bacterial

8    eradication, and asking not only to demonstrate that you

9    have sterilized the blood stream, but somehow that

10   sterilization has some impact on the long term natural

11   history for that patient.

12                  And    picking     the    most       relevant    clinical

13   criteria, and the most relevant time points for that

14   determination.       And I don't think we have yet determined

15   what that may be, and that may be a subject for a workshop.

16                  CHAIRMAN RELLER:         A lot more is known about

17   this criterion.        I mean, clearly one would in the

18   endocarditis trial nowadays need patients who were

19   entered to have transesophageal echocardiograms so that

20   you could even out those who had valve ring asepses, and

21   those persons who came to surgery.

22                  And    in    addition,         you     know,    time    to

23   sterilization of blood, and follow-up afterwards.                     Dr.

24   Archer, along those lines, if you were to design such

25   a trial with four weeks and six weeks of therapy, and

                                S A G CORP.
     202/797-2525               Washington, D.C.             Fax: 202/797-2525

1    with all of those other things, and adequate training

2    of investigators, consistency in entry, to have some

3    reasonable assessment after therapy of cure, would you

4    not allow -- and sometimes what is done is the oral

5    suppressive therapy after a rigorous cidal regimen?

6                   DR. ARCHER:      Well, that certainly could be

7    part of any kind of a study.           I think that everything

8    is wide open.    I think Dr. McCracken's point though about

9    bacteriological eradication, versus sequelae, that are

10   irrelevant to the antibiotic.

11                  I mean, flipping emboli from a vegetation

12   after the vegetation is sterile or a valve leaflet

13   rupturing after the vegetation is sterile, are not really

14   necessarily indications of the efficacy of an antibiotic,

15   and now well it worked in sterilizing the disease.

16                  And so I think just like meningitis, I think

17   the    bacteriological    end    points,    which     are     easily

18   measurable in terms of sterilizing the vegetation, are

19   very good surrogate end points in endocarditis, just like

20   meningitis, and might be equally accessible to therapy

21   and therapeutic measurement, and delta calculations.

22                  And I think the issue of oral therapy, and

23   the issue of the length of therapy, there is a whole bunch

24   of things that need to be tackled, and with new antibiotics

25   coming out which are potentially more bactericidal, and

                              S A G CORP.
     202/797-2525             Washington, D.C.           Fax: 202/797-2525

1    might even shorten the course of therapy, I think is the

2    opportunity to do that now.

3                   CHAIRMAN RELLER:         Wouldn't one -- there

4    clearly are issues that affect clinical outcome that are

5    not -- they may be related, but they would not be a reason

6    to discount an effective drug for sterilization in the

7    spinal fluid, or a vegetation in the bacteremia associated

8    with endocarditis.

9                   But shouldn't those differences be evened

10   out if there were really good design and randomization

11   to treatment arms?    That is, those patients with delayed

12   therapy, and enrolled in an meningitis study, and those

13   who had embolic complications, or who came to surgery

14   because of either failure, or vegetation size, or emboli,

15   or whatever happened?

16                  But wouldn't that even out if you had proper

17   randomization?    Yes?    Dr. Glode.

18                  DR. GLODE:    But you have the example of that

19   in the study presented by Dr. McCracken, where again I

20   think you would have to prioritize whichever agency was

21   advising approval.

22                  You would have to prioritize those outcomes.

23    So if you look just at his example, then he had

24   bacteriologic success, and could have had a delta of 5

25   percent, and that would have flown, passed, right?

                              S A G CORP.
     202/797-2525             Washington, D.C.        Fax: 202/797-2525

1                   But on clinical success, which should have

2    again by what you just said, by randomizing people to

3    ceftriaxone, or trovafloxacin, you should have randomized

4    appropriately in the mean duration of symptoms prior to

5    therapy was the same, and the two groups, et cetera.

6                   So clinical success should have been the same

7    if you are assuming right that neurologic sequelae were

8    independent of antibiotic other than duration prior to

9    therapy.

10                  But it didn't, and so it fails the 15 percent

11   delta on clinical success.       So if you are the committee,

12   and it passes the 5 percent on microbiologic, but it fails

13   the 15 percent on clinical, then what are you left with?


15                  You have to say, well, I guess microbiologic

16   is more important, and I don't know why it came out

17   differently.    It should have come out the same.

18                  But do you see that by putting those extra

19   end points that you have to prioritize which ones are

20   more important to you than the other ones?

21                  CHAIRMAN RELLER:      I also got the impression

22   in his presentation that there were some questions about

23   the quality of the data collected at different sites,

24   and that's why I put the emphasis on proper randomization

25   and control, et cetera.

                             S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1                      Now, we have lots of hands.             This really

2    opened up the discussion, which is great.                  Let me try

3    to go in a reasonable order.                Dr. Nelson, Dr. Shlaes,

4    Dr. Talbot, and Dr. Wittes, and there will be others,

5    but that is a start.

6                      DR. NELSON:      Well, two quick comments.         What

7    I   took       away   from   the   fact     that   11   were    on    the

8    investigational agent and two were on the control agent,

9    is that there was an adequate block randomization by study

10   sites, and that they would have had to somehow control.

11                     But if there were 6 and 5, or 6 and 7, then

12   that might have fallen out as not being an issue.                     One

13   question again, and not that I would like to be educated

14   on,   but      this   notion    that    a   surrogate    criteria      of

15   bacteriological clearance, is there any evidence at all

16   that the relationship of a particular drug if it has a

17   different mechanism of action, of its cidal action, could

18   induce a different inflammatory response that could be

19   qualitatively different from patient to patient, to where

20   one   would      then   assume     no   relationship     between      the

21   surrogate marker of the bacteriological clearance, and

22   the eventual clinical outcome just based on the host

23   response?

24                     Is that possible, or is there any evidence

25   to suggest -- you know, same bug, different drug,

                                  S A G CORP.
     202/797-2525                 Washington, D.C.           Fax: 202/797-2525

1    different inflammatory response, depending on the drug?

2                      CHAIRMAN RELLER:         Well, you know, I think

3    there may be, and in one of the issues clearly there may

4    be differences in safety, going back to some very old

5    studies and some quite provocative titles, like "With

6    Endocarditis, Dead or Dead," and titles to early clinical

7    trials.        Dr. Shlaes.

8                      DR. SHLAES:       I just wanted to bring us back

9    to the reason, as you were trying to point out, I think,

10   as to how we got to microbiological end points for

11   meningitis, and now for endocarditis, at the beginning.


13                     And that was to make the trials doable with

14   diseases       that   are    very    severe,    but   have   very    low

15   incidence.        So it is clear to me from the discussions

16   this morning that the way we got there was to use surrogate

17   markers, such as microbiological efficacy, to allow you

18   to enroll a smaller number of patients, and you would

19   sacrifice therefore a number of the clinical end points

20   that you would normally use to be able to use the surrogate

21   end point, which you have confidence.

22                     And certainly in the case of meningitis, and

23   I think as Gordon Archer pointed out, probably in the

24   case of endocarditis, where you have confidence that the

25   microbiological eradication would be correlated with

                                  S A G CORP.
     202/797-2525                 Washington, D.C.          Fax: 202/797-2525

1    clinical outcome in some reasonable sense.

2                   So    I    think    that    that   is   still   a   very

3    reasonable approach to these diseases which are severe,

4    or where the incidence is small, and we must keep the

5    trial size small in order to actually be able to

6    practically carry out the trial.

7                   CHAIRMAN RELLER:           Or put another way, that

8    some of these entities to have smaller number of patients

9    that are well studied may provide more useful information

10   than a larger number of patients, where the quality of

11   recruitment, and the quality of follow-up, and the rigor

12   of the randomization, et cetera, is not there.

13                  Now, Dr. Talbot was next, and then Dr. Wittes,

14   and then we will get a fresh list.             I cannot handle more

15   than four at once.         Dr. Talbot.

16                  DR.       TALBOT:      Thank    you.     I   have     two

17   comments, one on behalf of Dr. Edwards, who sends his

18   regrets that he had to leave.             His comment was that IDSA

19   wishes to emphasize that its clinicians, even right now,

20   are limited in their therapeutic options for some very

21   serious illnesses, such as meningitis, endocarditis,

22   fungal diseases.

23                  So from a clinical perspective, and as front

24   line people in the battle against infections, I think

25   the IDSA membership feels that this is an acute problem

                                 S A G CORP.
     202/797-2525                Washington, D.C.           Fax: 202/797-2525

1    and that's why we are here.

2                   But certainly the IDSA would like to see some

3    meaningful progress today.       So, I have tried to distill

4    a little bit what I have heard.       The last time I did this,

5    Bill Craig told me that it was his job as Chairman, and

6    not my job as participant, but I am going to risk it anyway,

7    Barth, if you don't mind.

8                   You know, the issue with these serious

9    diseases, it is exactly as Dr. Shlaes mentioned.                   And

10   I think with serious illnesses, there are two questions

11   that are critical.

12                  First of all, do regulators and clinicians,

13   and pharmaceutical companies, want data on how drugs work

14   in these diseases.     The answer is yes.

15                  The   second   question     is   do     these     same

16   stakeholders want some certainty, statistical certainty,

17   about the results, and I think the answer is clearly yes,

18   and that has been adequately mentioned already today.

19                  So I think that there are potentially two

20   choices with a fallback position.          To allow the studies

21   to be done, one has to change the delta to widen it if

22   necessary, but that is for reasons that we have heard,

23   and not particularly appealing, given that these are

24   illnesses with severe morbidity and mortality.

25                  A second option is to change the end point,

                             S A G CORP.
     202/797-2525            Washington, D.C.             Fax: 202/797-2525

1    but use a strict delta.      That is what Dr. McCracken had

2    mentioned before is what you were saying Dr. Shlaes, and

3    I think given the state of advancement of anti-infective

4    drug development in a moment, that should be feasible

5    for things like meningitis, endocarditis, Dr. Archer,

6    and possibly others.

7                   That would allow you to have statistical

8    certain, but it would require that you have confidence

9    in that end point, and that is where workshop discussions

10   could generate a consensus about whether such end points

11   existed.

12                  Finally, if you had a situation where there

13   was no acceptable surrogate, you might be able to fall

14   back to the GC paradigm perhaps, where you said that if

15   you have a drug that gets 95 percent clinical efficacy

16   in a small subset, 80 to a hundred patients, in a serious

17   infection like meningitis, that is going to be good

18   enough.

19                  So I wondered if -- I hope that overview helps

20   focus the discussion with one hour to go.

21                  CHAIRMAN RELLER:       Thank you, George.      Dr.

22   Bell.

23                  DR. BELL:   I would like to come back to the

24   concept of different deltas for different types of end

25   points -- surrogate versus clinical outcome -- for a

                             S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    couple of reasons.

2                   One of them is that I think that no matter

3    what we or the FDA agree on the clinical community of

4    practicing physicians out there is going to be much more

5    comforted seeing clinical outcome data, than simply

6    surrogate data.

7                   And   to promote a drug based solely on

8    surrogate data might become problematic when there is

9    some    inevitable   reports    of   failures,   or   uncertain

10   successes.     They will want to see some evidence that

11   clinical outcome actually was better.

12                  I think the place where this has not been

13   the case has been in HIV, where as we were discussing

14   at the break, the viral load now is widely accepted as

15   the surrogate outcome for many good reasons.

16                  But the difference there is that this is a

17   uniformly fatal disease, and where there never was a cure.

18    And so people were happy to use the surrogate outcomes

19   to get the new drugs quicker.

20                  But as we start talking about diseases where

21   there are clinical cures, and it is just a matter of losing

22   the antibiotics, people are going to be very uncomfortable

23   no longer getting information on clinical cures.

24                  And I just wonder if the FDA could take --

25   I think it was you, and maybe it was Dr. McCracken, that

                             S A G CORP.
     202/797-2525            Washington, D.C.         Fax: 202/797-2525

1    different deltas -- well, maybe the delta for the

2    surrogate marker could be much narrower.

3                     And the delta for the clinical one could be

4    greater to deal with the patient accrual problem.                 But

5    that also eventually there would be something, and if

6    there was some paradoxical and unexpected effect for

7    reasons that we don't understand, this clinical outcome

8    really was worse, and at least there was some framework

9    in place to monitor that.

10                    CHAIRMAN RELLER:       David, I had brought that

11   up, and just to follow up your analogy of HIV infection,

12   maybe it is not so dissimilar.                I mean, if one has

13   bacterial meningitis, or bacterial endocarditis, with

14   staphylococcus aureus, and there is no sterilization of

15   the blood to vegetation or the CSF, I think there aren't

16   any cures either for practical purposes.

17                    But that does not mean to say that there

18   wouldn't be differences in therapy of drugs that can

19   sterilize the CSF, in terms of rapidity of doing that

20   sequelae       with   hearing,   et    cetera,   like   there     are

21   differences in the art therapies with tolerance, and side

22   effects, and other outcome measurements apart from

23   controlling viral replication, and viral load.

24                    So I think that has been brought up, and I

25   think it is one of the things that has come out of the

                               S A G CORP.
     202/797-2525               Washington, D.C.         Fax: 202/797-2525

1    discussions today that what the emphasis would be in

2    end-points, and Dr. Talbot has pointed that out as well,

3    could be indeed, or probably should be different with

4    the different clinical entities under study.

5                   And exactly what those criteria and their

6    prioritization is, Dr. Glode pointed out every one has

7    recognized at the outset of this meeting all these loose

8    ends.    They are not going to be tied up this afternoon.

9                   But the heteriogentity of the appropriate

10   responses I think is a message that is coming across very

11   clearly in today's discussions.            Dr. Patterson.

12                  DR. PATTERSON:      Well, I would agree that

13   especially in meningitis that you want to know about

14   clinical outcome, as well as bacterial eradication,

15   because for instance you could have an antibiotic that

16   is more rapidly cidal, and with increased cytokine

17   release, more cerebral edema, and it could be better at

18   bacteriologic eradication.

19                  But you might have a worse clinical outcome,

20   and so I think especially for meningitis that you are

21   also interested in clinical outcome, and I think that

22   Dr. McCracken suggestion that at the end of his talk to

23   continue the 300 patients, 20 percent delta, for clinical

24   outcome, is a good one.

25                  And   perhaps      then       for   bacteriologic

                            S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    eradication would be of interest and a smaller delta could

2    be used for that.

3                        CHAIRMAN RELLER:        Dr. Fink and Dr. Leggett.

4                        DR. FINK:       I was just concerned with Dr.

5    McCracken's         comment     that    I    am   not   sure   what     the

6    applicability of clinical outcome data in meningitis is

7    when you go overseas to populations where the patients

8    are malnourished, and where 30 percent were HIV infected.


10                       What is the meaning of clinical outcome in

11   that population, when it is so different from what is

12   treated in the United States, that an adverse clinical

13   outcome does not necessarily mean that the drug is bad.


15                       I am worried, because I think clinical

16   outcome is important, but I think if you are going to

17   do measures of clinical outcome that you would at least

18   have    to     do    it   in    a   population     that    has    similar

19   socio-economic status, similar societal status, to that

20   of the United States if you are going to use the results

21   here.

22                       CHAIRMAN RELLER:        Dr. Leggett.

23                       DR. LEGGETT:        I would like to echo the

24   comments of Dr. Patterson and Dr. Talbot before, who took

25   the words out of my mouth because you would not look this

                                    S A G CORP.
     202/797-2525                   Washington, D.C.           Fax: 202/797-2525

1    way.

2                   But I would like to point out that setting

3    a rigid delta for things that the drugs can control for,

4    and for things that the drugs cannot control for, seems

5    to me to be fundamentally different.

6                   If    we   are   talking        about   a    bacterial

7    eradication, whether it is endocarditis or meningitis,

8    up near 98 or 99 percent, you could sort of keep this

9    sliding scale, and whether you modify it in this sort

10   of modified Lewis criteria thing or not.

11                  But it seems that there is more noise to your

12   clinical outcomes, whether it is form embolic disease

13   or from cytokine release, that you have to leave room

14   for a larger delta, and for the practicality of doing

15   the studies.

16                  So to affix 10 percent and say that it is

17   10 percent, no matter what the cause of the difference

18   is, I don't think is going to help us down the road.

19                  CHAIRMAN RELLER:        Thank you.      Dr. Tally, and

20   then Dr. Hardalo.

21                  DR.   TALLY:       We    have    gone    through      the

22   rationale of studying endocarditis, and indeed we have

23   a proposal at the FDA right now that we have been talking

24   to them about.

25                  And Gordon is right.          There are a couple of

                               S A G CORP.
     202/797-2525              Washington, D.C.             Fax: 202/797-2525

1    drugs coming down the pike that have the characteristics

2    as defined in previous studies on endocarditis that mag

3    be suitable to treat endocarditis.

4                   And particularly the new endocarditis that

5    is now representing approximately 30 to 35 percent, and

6    that is staph aureus.   So when you have appropriate models

7    and blood levels, and the initial data to support that

8    you can go into that, then we had been in discussion to

9    look at this.

10                  Now, these are difficult infections, and you

11   need to be in special hospitals, and where you can do

12   the transesophageal to apply the new criteria, and to

13   who does have endocarditis.

14                  But again we have heard around the table that

15   the treatment of this disease is multi-factorial, because

16   you need to have cardiac surgery there, because that is

17   part of the treatment of staph aureus and endocarditis.

18                  And that is not drug driven, and it may be

19   needed initially when the patient presents.        It should

20   randomize out, but again what David brought out, and I

21   think it has been brought out today, these difficult

22   diseases, and that are difficult to study, which have

23   these very hard bacteriological end points, can be studied

24   in a prospective manner, and not to get hung up on the

25   real delta in the beginning.

                             S A G CORP.
     202/797-2525            Washington, D.C.      Fax: 202/797-2525

1                    But to look prospectively and looking very

2    carefully, and I think the two responses that I think

3    you need in endocarditis is to initially bring the

4    endocarditis under control, and sterilize the blood.

5                    That is very hard, and I think if you have

6    not done that in a certain period of time, it is clear

7    cut.    It is a failure and the new drug is either going

8    to be equal standard of care therapy rate now or it is

9    not.

10                   And I think we can come to that when we develop

11   that data.      The second evaluation does take in these other

12   factors, and the one with the long follow-up is the relapse

13   rate that comes afterwards, and was the drug effective.


15                   And I think you need a good number of patients

16   to say that, but I don't think you need the 500 patient

17   studies.       I think you can do it with a smaller number

18   of proven cases of endocarditis.             And that is the

19   discussion that we are in now, and I think we could be

20   moving forward to try and answer some of these questions.

21                   But I think this is the one where you really

22   have to be in dialogue with the regulatory agency and

23   be in dialogue with your investigators, prospectively

24   monitoring very closely to make sure that you don't get

25   in trouble because of the high or deleterious effect of

                              S A G CORP.
     202/797-2525              Washington, D.C.      Fax: 202/797-2525

1    a failure rate is usually in this one severe morbidity

2    and death.

3                   CHAIRMAN RELLER:     Dr. Tally, you raised some

4    very important points, and I wanted to ask do you think

5    it is important to emphasize that this quality of

6    investigator,    the   centers     where   patients     would     be

7    recruited and enrolled, would have the capacity to take

8    care of these patients properly.

9                   And   with   endocarditis,     as     Dr.     Archer

10   mentioned earlier, these are studies -- I mean, they would

11   not be exclusive to the United States, but the United

12   States, and Western Europe -- I mean, these require --

13   I mean, a standard of care that we would accept requires

14   a sophisticated center where to study fewer patients well

15   may provide better answers than missing data that people

16   aren't going to be able to evaluate at the end of the

17   day.

18                  DR. TALLY:     Well, I think if you stick to

19   institutions that are approved for cardiac surgery, and

20   can do valve replacement, the you already are at a level

21   of care that is a higher standard, I think, then routine

22   hospital care in the United States.

23                  CHAIRMAN RELLER:      Dr. Hardalo.

24                  DR. HARDALO:     I think all of these things

25   point out the need for developing consensus on exactly

                             S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    -- within clinical outcome, we have heard multiple end

2    points.        The hierarchy of those end points from those

3    which are most directly related to anti-bacterial or

4    antimicrobial efficacy, and down to those which are more

5    related to anti-inflammatory treatments or other sequelae

6    of the disease.

7                     In endocarditis, we have heard embolism,

8    immune complex disease, other sequelae which have little

9    or nothing to do with the anti-bacterial clearance of

10   the infection, and that has a lot to do with the duration

11   of   disease      and   prior   underlying    history   for    that

12   particular patient.

13                    Indeed, the need for cardiac surgery may not

14   necessarily have anything to do with antibacterial

15   therapy.        It may have to do with other host factors.

16   For meningitis, clearly there is a difference in terms

17   of when you do your clinical outcome, but in what kinds

18   of patients.

19                    I am sure as the pediatricians in the group

20   can say, that getting an auditory test on a 2 year old,

21   and trying to get a reasonable indicator of whether you

22   have auditory sequelae, is quite different than trying

23   to get one on an eight year old.

24                    And trying to interpret that as you follow

25   the patient over six weeks, and six months, can lead a

                                S A G CORP.
     202/797-2525               Washington, D.C.       Fax: 202/797-2525

1    certain amount of noise in interpreting the results.

2    And so you have to have some consensus on how much noise

3    you are going to allow based on the populations you have

4    tried to study.

5                     Certainly the efforts by the industry as the

6    information becomes much more critical, and as these

7    patient populations become much smaller, is to really

8    go     through     extensive        efforts          to   qualify      your

9    investigators.

10                    It is no longer the standard just to take

11   all-comers who want to do critical investigations.                       We

12   have been held to an increasingly high standard in good

13   clinical practices for exactly this reason.

14                    We want to believe the data at the end of

15   the day that we have put so much into developing the

16   protocol, and there is so much resting on this in terms

17   of delivering good quality data to our clinicians.

18                    CHAIRMAN RELLER:         Thank you.       Dr. Ramirez,

19   and then Dr. Ebert.

20                    DR. RAMIREZ:       I just would like to emphasize

21   what you just mentioned.            This is critical.           Plenty of

22   the discussions of the patients at the end of the trial

23   will    have     data   to      evaluate        is    because     of    the

24   investigators.

25                    And really I can summarize the meningitis

                              S A G CORP.
     202/797-2525                 Washington, D.C.             Fax: 202/797-2525

1    presentation by Dr. McCracken, and say that he has a

2    problem with an investigator.        There was a bias against

3    the quinolones, and every patient that was on quinolones

4    was a failure.

5                   I mean, there was not a problem of the

6    assignment of the trial, and we don't need to increase

7    the delta.     We just need to change the investigator.

8    But the study is supposed to be blind, and how come

9    investigators are going to know that my patient with

10   meningitis was getting quinolones, or is getting the

11   standard therapy?

12                  But essentially we just need to have good

13   investigators.    I think that in this regard really we

14   don't need to blame the FDA.       We just need to blame the

15   industry and with an intention to get patients in

16   empirical trials.

17                  I mean, I liked what you just mentioned,

18   highest standards for investigator, but what we see in

19   different universities around the country is that the

20   clinical trials are no longer there.

21                  The clinical trials go to the very busy

22   private physician, who has a nurse running around and

23   drawing everybody.    We can't even say that these are bad

24   investigators when there are no investigations to begin

25   with.

                            S A G CORP.
     202/797-2525           Washington, D.C.        Fax: 202/797-2525

1                     And another thing is that I don't think we

2    need    to     travel    all    over     the      country    to   find    bad

3    investigators.          I mean, we can do it at the center trials

4    here.    And why is it that we are having such a poor quality

5    in our research?           It is probably because we are not

6    selecting good investigators.

7                     DR. HARDALO:          I would really want to argue

8    with that.       Part of the reason is that when you have to

9    do a trial of 2,000 patients in the United States,

10   especially if these patients can have no prior antibiotic

11   therapy,       and   especially        you   want    to     get   resistant

12   pathogens, you are not going to find them in the United

13   States or in many areas of Western Europe.

14                    And that has been shown in time after time

15   when you look at the trials that are enrolled.                        Again,

16   we would love to work with United States centers, but

17   some of the realities of making a trial feasible requires

18   us to go outside of the country.

19                    And     you     are     absolutely         right.        The

20   investigator selection issue, it is a monitoring issue,

21   and we can do what we can in real life.                              But the

22   investigators are clinicians and who also have their

23   obligations to do trials according to good clinical

24   practices.

25                    CHAIRMAN RELLER:            Dr. Ebert.

                                    S A G CORP.
     202/797-2525                   Washington, D.C.             Fax: 202/797-2525

1                   CHAIRMAN RELLER:      This is a comment I think

2    more of surrogate outcomes in general, but certainly the

3    examples that were used regarding microbiology I think

4    were very compelling.

5                   But I think something that as we start to

6    develop surrogate outcomes for other diseases and try

7    to use those in lieu of clinical outcome, we need to keep

8    in mind that as we try to reduce the delta for the use

9    of these clinical outcomes, or excuse me, these surrogate

10   outcomes, we need to be sure that those surrogate outcomes

11   are achieved at a fairly high level.

12                  In other words, a very high percentage.        For

13   example, the sterilization of nearly a hundred percent.

14    If the frequency at which these surrogate outcomes is

15   achieved is at a lower level or similar to clinical

16   outcomes with regard to the frequency, I don't think we

17   have really accomplished anything, and using the small

18   delta is just going to drive up the sample size again.


20                  CHAIRMAN RELLER:     Dr. Metlay and then we will

21   -- we included infective endocarditis, which was not in

22   the question, but I think some very important points have

23   been raised related thereto for future drug development.

24                  And then we need to have any comments, if

25   there be any, for hospital-acquired pneumonia before

                            S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    going to the final, but shorter, third question before

2    concluding at 5:30.     Dr. Metlay.

3                   DR. METLAY:    I guess what I am struggling

4    with to some extent is to what degree do these surrogate

5    end points, bacteriological eradication, really are a

6    solution, or just an occasional exception to the rule.


8                   One of the insights, for example, in the last

9    couple of years, and perhaps relevant in the treatment

10   of community-acquired pneumonia, is that therapy within

11   8 hours saves lives.

12                  It seems plausible to me that if we were

13   measuring bacterial eradication at 24 hours, for example,

14   or even 48 hours, that we would fail to detect benefits

15   of some therapies, or some strategies within that kind

16   of a window, because our measure is not sensitive enough.


18                  It is not inherently the case that bacterial

19   eradication is a more sensitive measure for the efficacy

20   of the drug given that in the end what we are interested

21   in are patient outcomes.

22                  So I think that there are lots of applications

23   of meningitis, and in some ways like an ideal one, but

24   I think how well that would generalize and get you to

25   a lot of other solutions is not clear to me at all.

                             S A G CORP.
     202/797-2525            Washington, D.C.      Fax: 202/797-2525

1                   CHAIRMAN RELLER:          Dr. Cross.

2                   DR. CROSS:        Well, just as a follow-up to

3    that, in certain disease processes, especially infections

4    with bacteria in sterile sites, a prerequisite is that

5    you have to clear the site of infection.

6                   In the case of pneumonia, it is a lot more

7    complicated pathophysiology of which the clearance of

8    bacteria perhaps is only a small point.             But I would agree

9    with Steve's comments that if we do have a surrogate end

10   point -- and so far the only surrogate end points that

11   I have heard have been bacterial clearance from sterile

12   sites has to be very high.

13                  But to reemphasize a point that Jan made,

14   perhaps there ought to be serious consideration given

15   to the clinical outcomes in the situation where the

16   antibiotic itself does have an effect on the inflammatory

17   response.

18                  And    Dr.    McCracken         mentioned     about    the

19   inflammatory response in meningitis, but also there has

20   been perhaps more made out of it than it ought to be.

21                  But people have tried to compare differences

22   in, for example, ceptazam (phonetic) versus enepenam

23   (phonetic),    both    of    which     can     clear   the    blood    of

24   GRAM-negatives very rapidly, but one of which may liberate

25   in the process of that killing a pro inflammatory agent

                                 S A G CORP.
     202/797-2525                Washington, D.C.           Fax: 202/797-2525

1    more than the other.

2                     So in that situation, I think on the one hand

3    we can have a small delta for the clearance of the

4    bacteria, which is a prerequisite, but on the other hand,

5    I think we still ought to allow for some potential

6    differences from a difference which may arise not as a

7    result of the pathophysiology of the disease which we

8    might not know anything about. but because of the

9    mechanism by which that antibiotic may work.


11                    CHAIRMAN RELLER:            Dr. Chesney.

12                    DR. CHESNEY:           Just two quick comments.                     I

13   think    George      made       the    comment       this    morning         that

14   sterilization of the middle ear correlates very well with

15   clinical outcome, and I think that is something that we

16   have just learned in the last few years.

17                    The other thing is that I just wanted put

18   a   little     bit   of     a     plug     in      here    for     quality      of

19   investigators.       In terms of the NIH having put so much

20   money     into    the     PPRUs,         which       are    the       Pediatric

21   Pharmacokinetic Research Unit, that some of you may not

22   know about.

23                    But these are wonderful research units --

24   I think there are 13 in the country -- that have been

25   set up exclusively to study drugs in children and to

                                     S A G CORP.
     202/797-2525                    Washington, D.C.                 Fax: 202/797-2525

1    maintain that, and set the standard for that kind of

2    quality.       So I think as pediatricians that we would like

3    to thank them at every opportunity that we get.

4                     CHAIRMAN    RELLER:           Contributions   to    the

5    discussion for hospital-acquired pneumonia.                Dr. Archer.

6                     DR. ARCHER:     I would like to start this off

7    again.     As a comment about the dichotomous nature of these

8    infections, I think somebody mentioned it earlier, but

9    that hospital-acquired pneumonia is an excellent example.


11                    For instance, there is a population of

12   hospital acquired pneumonia, and people in the VA know

13   about this very well, and in the extended care facilities,

14   patients who develop pneumonia while in the hospital and

15   who don't make it to the ICU, and don't get ventilated.

16                    And   post-operative           patients    developed

17   hospital-acquired pneumonia, and those are very different

18   than the hospital acquired pneumonia patients who are

19   ventilator dependent.

20                    And I think the bacteriology is different,

21   and so I think you could also argue that you could have

22   different populations of hospital-acquired pneumonia

23   patients, some of whom may do better than others as well.

24                    And I don't know that those have been well

25   separated out in studies, at least the studies that I

                                 S A G CORP.
     202/797-2525                Washington, D.C.           Fax: 202/797-2525

1    have seen.      And a second comment about hospital acquired

2    pneumonia, particularly those in intensive care units,

3    is that it is way too easy to get bacteriology as they

4    are suctioning patients out in Q-5 minutes I think, and

5    a lot of these people are -- and there is bacteria

6    everywhere, and they are cultured frequently.

7                    And I think this is a slippery slope.             If

8    you include these in studies, then you have to have some

9    measure of eradication of the bacteria that are within

10   the spectrum of the drug that you are using.

11                   And I think that is very difficult with

12   hospital-acquired pneumonia, because as has been said,

13   the presence of bacteria don't often correlate, and nor

14   do I think the eradication correlates very well.

15                   And I have not seen a lot of study design

16   where attention is paid to the effect of the drug on the

17   bacteriology of the pneumonia, or the organisms that are

18   recovered from the sucrate.

19                   CHAIRMAN RELLER:      The guidelines that were

20   published in the collaborative effort with FDA and IDSA

21   in 1992 were a giant step forward from the former days

22   of   lower     respiratory   tract    infections   when    it    was

23   delineated         as     community-acquired           pneumonia,

24   hospital-acquired pneumonia.

25                   What I do not recall, and maybe a further

                              S A G CORP.
     202/797-2525             Washington, D.C.          Fax: 202/797-2525

1    distinction is necessary and an important message to send

2    from    this   committee    to    the    next    iteration      is     the

3    separation in hospital acquired pneumonia into those

4    patients who are intubated and those who are not.

5                    I don't think that currently exists in the

6    hospital-acquired pneumonia guidelines.                Correct me if

7    I am wrong.

8                    DR.    ROTSTEIN:              Well,    there      is        a

9    differentiation that the ATS had based on organisms, the

10   types of organisms that people would have, and whether

11   they were admitted to the ICU with hypotension, et cetera.

12    So the ATS does differentiate somewhat.

13                   CHAIRMAN RELLER:        Right, the ATS, but in the

14   guidelines,      the   points    to     consider      documents,       Dr.

15   Albrecht,      currently   the    agency       does   not   make     that

16   distinction in clinical trial design?

17                   DR. ALBRECHT:       It is correct that we don't

18   have a separate guidance for ventilator-assisted, or

19   associated pneumonia.         I think there is mention of it,

20   but not a separation at this point.

21                   CHAIRMAN RELLER:        Because maybe we would

22   -- you know, in addition to, and apart from the delta,

23   if the committee thinks that is an important distinction

24   to make, in terms of evaluation of clinical outcome, or

25   bacteriologic outcome -- I mean, outcomes, whatever the

                                S A G CORP.
     202/797-2525               Washington, D.C.             Fax: 202/797-2525

1    end points are, we should get that point across clearly.

2     Dr. Ramirez.

3                   DR. RAMIREZ:     My opinion is that there is

4    a significant difference.           I mean, pneumonia is a

5    continuation of disease from community ambulatory care,

6    to the patient who is going to be in intensive care unit

7    and on a ventilator.

8                   And there is definitely a continuation of

9    the disease in study, after study, after study indicated,

10   that early nosocomial pneumonia -- and how you define

11   early in different investigations is defined differently

12   in days.

13                  But 5 days, or 7 days, whatever is the

14   definition of early, early nosocomial pneumonia, you look

15   at the pathogens, and they are exactly the same pathogens

16   that communicate community pneumonia.

17                  The patient is in the hospital for X-amount

18   of days, and develops nosocomial pneumonia, and at least

19   in our hospital guidelines, we don't use anti-nosocomial

20   regimen, because these patients are going to have H. flu,

21   streptococcal pneumonia.

22                  These people don't have the time in the

23   hospital to be colonized with the nosocomial resistant

24   pathogens.     In early nosocomial pneumonia, in any studies

25   from Europe -- and in our intensive care unit, we have

                             S A G CORP.
     202/797-2525            Washington, D.C.      Fax: 202/797-2525

1    a trauma unit.

2                   And if you go to the unit, you are on a

3    ventilator.    You develop pneumonia, and you have early

4    nosocomial     pneumonia,        bronchial,       or     haemophilus

5    influenzae, number one.

6                   If you are smoker, you have early nosocomial

7    pneumonia, and you don't need -- there is no question

8    that nosocomial pneumonia is a single disease.                    It is

9    different.

10                  Now,    here      you    have     multiple        medical

11   co-morbidities, and you are in the unit, and you have

12   been in the hospital for 2 weeks.              There is no question

13   this patient is going to be colonized with whatever

14   organisms is living in your hospital.

15                  And    to    me    the     distinction       of    early

16   development     of    nosocomial        pneumonia      versus     other

17   organisms, these are two different pathologies.                    This

18   is one person with a community organism versus another

19   person.

20                  And another thing I would like to say since

21   I have the microphone is that in the delta-1 question

22   in nosocomial pneumonia, and the two studies that were

23   presented, one was 90 percent mortality with placebo,

24   and the other with 10 percent mortality, if we don't have

25   data for one disease, I think we have to look at similar

                                S A G CORP.
     202/797-2525               Washington, D.C.            Fax: 202/797-2525

1    diseases and translate the data.

2                   We   know      that      in     community-contacted

3    pneumonia and the pre-antibiotic era that you have

4    bacteremia pneumococcal pneumonia, and there was 80

5    percent mortality.

6                   And then intuitively, I would agree with the

7    90 percent mortality.         With nosocomial pneumonia, you

8    don't use antibiotics.             If we know that you have

9    nosocomial pneumonia, and you don't use any antibiotics,

10   then I would not say that only 10 percent benefit for

11   antibiotics.

12                  It would be more towards 80 or 90 percent

13   benefit, or probably even 100 percent benefit with

14   antibiotics compared to placebo.              Then I would resolve

15   the delta-1 question with this.

16                  Now, the delta-2 question is the question

17   that we have been discussing, and the problem with

18   nosocomial pneumonia for delta-2 is that the problem is

19   not a problem with the drug.          It is the problem with the

20   clinical diagnosis.

21                  In any clinical trial, approximately 50

22   percent of the patients don't have nosocomial pneumonia.

23    And then this is the problem, because 50 percent of the

24   patients, it doesn't matter whatever you use, they are

25   just going to have the natural cause or the ARDS, or

                            S A G CORP.
     202/797-2525               Washington, D.C.          Fax: 202/797-2525

1    whatever other disease they have that we call pneumonia,

2    because we don't have any better way to make the diagnosis,

3    and the delta-2, I don't know how to resolve the problem.

4                   CHAIRMAN RELLER:     Dr. Patterson and then Dr.

5    Leggett.

6                   DR. PATTERSON:     Okay. I would like to come

7    back to a point that Dr. Powers made in his presentation

8    that looking at overall mortality I think is not the right

9    outcome in hospital-acquired pneumonia, because there

10   are a lot of other things obviously that these people

11   die from.

12                  And so l think looking at attributable

13   mortality, although that is sometimes difficult to tease

14   out, would be a much more important outcome to look at.

15    But overall mortality, I think wouldn't be the right

16   outcome.

17                  And then also based on a Fagan study that

18   showed improvement in outcome in people who were diagnosed

19   with the associated pneumonia with a protected specimen

20   brush, versus those who were empirically treated based

21   on what was in their sputum and sort of the traditional

22   way of diagnosing it, what are the critical care people

23   think about using the protected specimen brush with

24   quantitative culture more in the setting of diagnosing

25   and studying ventilator-associated pneumonia?

                            S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1                     CHAIRMAN RELLER:          Yes, please?

2                     DR. ROTSTEIN:         Just another comment about

3    pneumonia, and hospital acquired pneumonia.                      This is one

4    area that we really could look at resistance, because

5    this is where resistance occurs.

6                     These       people     are       often     on     multiple

7    antibiotics over prolonged periods of time, and this is

8    where we see our resistant organism.                 So any trial that

9    does look at this really should look at resistance issues

10   as well.

11                    CHAIRMAN RELLER:           Thank you.       Dr. Shlaes,

12   and then Dr. Leggett.

13                    DR. SHLAES:          Actually, I just wanted to

14   comment that I think that this particular area of

15   hospital-acquired pneumonia is the most difficult of the

16   areas that the committee is considering, and that the

17   FDA is considering.

18                    And because of the heterogeneity of the

19   population included in this umbrella, and in addition,

20   actually       the    CDC   is   thinking        about    changing    their

21   definitions, in terms of what is community-acquired and

22   what is hospital-acquired.

23                    I am hoping that the CDC is talking to the

24   FDA about their considerations, and that may help in fact

25   in   helping         us   dissect   out    these     two    populations.

                                   S A G CORP.
     202/797-2525                  Washington, D.C.             Fax: 202/797-2525

1    Actually, there are probably 3 or 4 populations, in

2    nosocomial pneumonia.

3                   And it may be that some of those things that

4    we have been calling nosocomial pneumonia are actually

5    community acquired pneumonia, and would fit better in

6    the new CDC definitions when they come out.

7                   And that may be an easier way for us to start

8    teasing this apart a little bit.           So I really think this

9    is a challenging area, and this is going to require

10   stakeholders that are not just industry, and IDSA, and

11   FDA, but is actually going to require some help from CDC,

12   and perhaps others, to just figure out some of these

13   definitions.

14                  DR. RAMIREZ:    The CDC is going to use it after

15   seven days to nosocomial?

16                  DR. SHLAES:    I don't know what they are going

17   to do.    David is here, and maybe he can tell us what they

18   are going to do.        But they are reconsidering their

19   definitions        of         community-acquired,            versus

20   hospital-acquired infection in general.

21                  CHAIRMAN RELLER:     I think there are multiple

22   manuscripts from different places under review, and that

23   the data aren't in yet.           But basically health care

24   associated infections may look more like nosocomial

25   infections than community-acquired in the strict sense.

                             S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1                     And    the     proportions         shifted,          and     not

2    everybody who comes in from the community has not had

3    recent association or be it extended care.                        But I think

4    the issues are very important.

5                     And that the definition of community acquired

6    pneumonia and hospital-acquired pneumonia will need some

7    redefinition, and including the ventilator, and those

8    complicated by the need for ventilatory assistance fall

9    into a different category, in terms of expected response,

10   and distribution of pathogens.                   Yes?

11                    DR.     CROSS:              I          think       that       in

12   hospital-acquired pneumonia the bacteriology in this will

13   be a real bear and has to be really clearly defined.

14   I think as has been said that the bacteriology of

15   ventilator-associated pneumonia is quite different.

16                    But the other thing to consider, especially

17   as      we       talk     about           hospital-acquired,                  and

18   community-acquired, is the rather extensive, and very

19   formidable data from 20 years ago looking at the role

20   of underlying illness, in terms of colonization with

21   GRAM-negative criteria.

22                    For example, on day one of entry into the

23   ICU, J. Sanford and Reiner showed about a quarter of the

24   patients       are   already      colonized         with    GRAM-negative

25   bacteria.

                                   S A G CORP.
     202/797-2525                  Washington, D.C.                  Fax: 202/797-2525

1                   Similarly, the classic studies of Valenti

2    showed     that     the   likelihood        of    colonization      with

3    GRAM-negative bacteria, even people walking in off the

4    street, is a function of their underlying health status.

5                   Therefore, it isn't a simple breakdown to

6    say that people who are in the less than 48 hours, or

7    96 hours, will have a certain amount of or certain types

8    of bacteria, in the absence of actually defining those

9    critical factors which have already been well-defined

10   in terms of health status, and bacteriology.

11                  CHAIRMAN RELLER:          So you are getting at the

12   importance of this attributable mortality issue as well.

13    Dr. Ramirez, and then we will have a comment from the

14   back.    Go ahead.

15                  DR. RAMIREZ:         I just want to clarify that

16   when I mentioned the early versus late -- and I totally

17   agree with the GRAM-negatives -- is that people can come

18   from home with klebsiella, E. coli, and they have

19   multi-medical co-mobilities.

20                  But the multi-resistant pseudomonas, you are

21   going to get in the hospital.                  Another thing is that

22   sometimes when we see studies done for the drug companies,

23   they want to test this particular drug against the others.

24                  We    have     seen     in      ciprofloxacin     versus

25   emipenam, and in all the latest studies of nosocomial

                                 S A G CORP.
     202/797-2525                Washington, D.C.           Fax: 202/797-2525

1    pneumonia.       But in reality, what I see happening in

2    critical       care   units    is     that       the   patient     may   have

3    ventilator-associated               pneumonia,         and    we     suspect

4    pseudomonas, and the tendency is to use combination

5    therapy.

6                     And the problem that I sometimes discuss with

7    industry is that we don't want to use antibiotics.                               I

8    just want my antibiotic.             But we are using more and more

9    combination       therapy      in    an   attempt       to    prevent      the

10   development of resistance and improved outcome.

11                    Wouldn't it be more realistic to do studies

12   of combination therapy based on ventilator-associated

13   pneumonia, and with the more severe form of nosocomial

14   pneumonia?

15                    CHAIRMAN RELLER:            I think there are big

16   differences in terms of Western Europe and the United

17   States and those who believe in the importance of

18   quantitative cultures from bronchoscopy specimens.                               I

19   know that in our own center there are brushers and

20   non-brushers, believers and non-believers.

21                    And I think that one of the messages that

22   comes across is before a discussion of deltas, that one

23   has to spend considerably more time in delineating what

24   it is that we are talking about with hospital-acquired

25   pneumonias as a prelude to a meaningful discussion of

                                   S A G CORP.
     202/797-2525                  Washington, D.C.               Fax: 202/797-2525

1    what kind of equivalence or non-inferiority trials, and

2    what the numbers should be.

3                     I need some help from those who wish to make

4    comments who are not seated around the table with their

5    nameplates.       So, please introduce yourself, and then

6    comment.

7                     DR. SCHENTAG:       Hi, I'm Jerry Schentag from

8    the University of Buffalo.             I am presenting the triad

9    of people who harass you folks with PK/PD type comments,

10   but I am the only one here today.

11                    So I felt obligated to speak, and I think

12   on this nosocomial pneumonia thing, if you do a multiple

13   logistic       regression    analysis,         and   include   all    the

14   clinical factors that you can dig up on nosocomial

15   pneumonia patients, and you add to it the activity of

16   the antibiotic.

17                    And then you plot that against how long it

18   takes to kill the bacteria -- and not whether or not you

19   kill it, but how long it takes to kill that bacteria on

20   serial culturing.

21                    And if you do the serial culturing, you can

22   get about 80 percent of the variance in the relationship

23   killing that organism over time just from the antibiotic

24   activity, leaving about 20 percent of the remaining

25   variance in that logistic regression to be explained by

                                 S A G CORP.
     202/797-2525                Washington, D.C.            Fax: 202/797-2525

1    the other factors.

2                   Now, I agree with you that this is not an

3    easy scenario to assign which one the pathogen is when

4    there is lots of organisms, and there is lots of drugs,

5    but it is relatively easy to assign an outcome to that

6    organism, which I do believe from studying this now for

7    quite a few years of multiple different antibiotics, and

8    we looked at maybe 15 or 20 antibiotics this way over

9    the last 10 or 15 years.

10                  And    I     do    believe       that   you       could    show

11   differences between concentrations to activity ratios

12   of each of those drugs, which makes sense.                         In other

13   words, it is the activity of the drug that determines

14   the microbial outcome.

15                  What I don't know is whether it always

16   determines whether you perceive the surrogate end point

17   of cure to follow that or not.           And ventilator-associated

18   pneumoniaes, it probably does reasonably well.

19                  In         the         non-ventilator             associated

20   pneumoniaes,    it     is       probably    like       a   lot    of     other

21   pneumoniaes; cures don't always follow eradication of

22   the organism.       There are other factors that aren't quite

23   so closely linked.

24                  But    cure       is   nonetheless          the   surrogate,

25   because the effect of the antibiotic is on the bacteria.

                                 S A G CORP.
     202/797-2525                 Washington, D.C.               Fax: 202/797-2525

1     Now, have we been able to find any evidence of endotoxin

2    storm or any of those other things contributing to

3    outcome?

4                   Well, my submission on that point is that

5    we have tried awful hard with the sepsis drugs, and we

6    haven't been able to show much of an additive benefit,

7    and we just managed to find a small one not so long ago,

8    but it is by and large not a dramatic effect if it is

9    there.

10                  Most people would look at all of those trials

11   and agree with that.      So I guess my comment is that you

12   shouldn't reject microbial end points so easily as

13   surrogates, given that they can almost always show

14   superiority with very small numbers of patients in each

15   group    between   two   antibiotics,       or   in   fact    between

16   combinations of one handful of drugs, versus the other

17   handful of drugs when you want to start looking at that

18   as cumulative activity, just assuming additivity.

19                  Thank you for letting me make that comment.

20    I had to get that off my chest.             Jerry Schentag from

21   Buffalo, okay?     Just in case.

22                  CHAIRMAN RELLER:        For the record, thanks,

23   Jerry.     Dr. Bennett, and then we need to move.            We could

24   use Dr. Schenag's comments as a transition to question

25   number three.      Dr. Bennett.

                              S A G CORP.
     202/797-2525             Washington, D.C.             Fax: 202/797-2525

1                   DR. BENNETT:   I wanted to give Art Goldberger

2    my two cents about deltas since we spent the morning

3    talking about deltas, and very early of the afternoon.

4                   But what I think I have learned is that 10

5    percent is not for everybody, and not for every trial,

6    and not for every indication.       So that a 10 percent delta

7    as a receipt in general is too inflexible.

8                   But I have also heard that the STEP function

9    is also inflexible in a different way, and not very useful.

10    So what I am taking home from this meeting is that you

11   are going to have to come up with guidelines that are

12   specific for indications, and maybe even have some

13   protocol definitions built in.

14                  And then you will be able to get deltas.

15   So your goal of having us bless a given delta, I just

16   don't hear that.     And that is why I think we are not

17   talking about it.

18                  DR. GOLDBERGER:     That in fact wasn't really

19   the goal of the meeting.        I don't think we recognized

20   upon reflection that a fixed delta for everything was

21   necessarily the best way to proceed, which is why I during

22   my introductory remarks made some points about this as

23   being the beginning of a process, rather than the goal

24   of coming up with a judgment at the end of the day.

25                  So we would agree with your comment, that

                            S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    I think it would be very difficult to squeeze in everything

2    under a single delta.

3                   DR. BENNETT:         The only reason that I made

4    the remark the way I did was that I had the impression

5    from conversations in the hall that that is what the FDA

6    had been doing; that is, using a 10 percent delta for

7    many different indications across the board.

8                   And     that   was    raising    some     appropriate

9    hackles, but that apparently was not correct.

10                  DR. GOLDBERGER:        It is fair to say that there

11   was at one point what I would describe as a communication

12   breakdown,     which    hopefully      we    have   satisfactorily

13   rectified with regards to that.

14                  I wouldn't want to say that those people who

15   were upset were upset entirely based only on their

16   imagination, because I don't think that is a fair

17   statement.

18                  But I think we recognized that this was in

19   fact not the preferred way to proceed, which was the reason

20   for trying to get as broad an input as possible, for

21   instance, at today's meeting.

22                  DR. BENNETT:      Thank you for clarifying.

23                  CHAIRMAN RELLER:         Question Number 3.           The

24   FDA announced that they were not going to slavishly follow

25   the STEP wise.       What they were going to do was perhaps

                               S A G CORP.
     202/797-2525              Washington, D.C.             Fax: 202/797-2525

1    prematurely anticipated.

2                    But what Dr. Bennett just summarized, is

3    where I think the parties at this meeting are fairly

4    concluding, is the reality that there must be a diversity

5    in what goes into a fair assessment, and realistic

6    assessment, of efficacy balanced off with safety of

7    anti-infective compounds, and that will be different by

8    different indications, and other very important issues

9    need to be addressed explicitly.

10                   And in some cases, objective end points; and

11   in   others,        a   redefinition     of      what   constitutes       the

12   appropriate         study    populations          with,    for     example,

13   hospital-acquired pneumonia.

14                   Question       Number      3.      Discuss       any   other

15   factors or characteristics of a drug product other than

16   the confidence intervals, the deltas, that could be

17   included       in       risk-benefit     analysis         supporting      FDA

18   regulatory decisions.

19                   Now, actually, these things have already come

20   up in the discussion.            So, it would be in addition to

21   what has already been said.

22                   Any comments about safety considerations,

23   PK/PD considerations, and the availability of alternative

24   therapies in this balance of safety and efficacy which

25   is the fundamental basis for regulatory approval?                         So,

                                   S A G CORP.
     202/797-2525                  Washington, D.C.              Fax: 202/797-2525

1    Dr. Fink, Dr. Glode, Dr. Shlaes.

2                   DR. FINK:    I am not going to address safety

3    considerations, but I think the one thing that is

4    glaringly      missing     from       that    list    is     patient

5    acceptability.

6                   Ease of administration, perceived burden of

7    the administration of the drug; is it once a day, four

8    times a day; does it give you an upset stomach.              I can't

9    get a parent to give amoxicillin when it gives their child

10   diarrhea.

11                  So I think you have to really look at what

12   is going on outside of the controlled clinical trial that

13   affects real world adherence to use of the drug in an

14   appropriate manner.      And that that needs to be very high

15   on the list of alternate considerations.

16                  CHAIRMAN RELLER:        Dr. Glode.

17                  DR.   GLODE:       I    will   need   possibly      Dr.

18   Fleming's comments on this as well, having both served

19   on the Vaccine Advisory Committee, and dealing with from

20   the perspective of safety, and how many children do you

21   need in a trial to assure safety.

22                  So I guess I don't know the answer to the

23   question of before a new antibiotic comes to a Phase III

24   efficacy and safety trial, in Phase I and Phase II, how

25   many hundreds or thousands of individuals have been

                              S A G CORP.
     202/797-2525             Washington, D.C.            Fax: 202/797-2525

1    studied for safety?

2                   Because   if,    for    example,   you   take    the

3    meningitis example, where you might use as an end point

4    bacteriologic sterilization of the spinal fluid so you

5    can use very small numbers of people.

6                   Then, you know, you compromise your ability

7    to look at safety issues it seems to me.           Now, if they

8    have already been looked at, but it is so detrimental

9    to everyone concerned, starting with the patients, when

10   a drug is withdrawn from the market after approval due

11   to an adverse event that was not recognized during the

12   preclinical trials.

13                  And I was wondering if anybody has gone back

14   and looked at the last 10 drugs removed and sort of asked

15   the question were they adequately studied in the first

16   place?

17                  Well, by the time that a new antibiotic gets

18   to Phase III, is there some approximate number of patients

19   who have received it to assure safety, or are we relying

20   on a Phase III study?

21                  CHAIRMAN RELLER:         Comments from the FDA?

22   In some of these past meetings, I think there has been

23   considerable discussion on some events that the numbers

24   simply can't preclude knowing until a drug is approved.

25                  I know that these issues came up in the

                              S A G CORP.
     202/797-2525             Washington, D.C.         Fax: 202/797-2525

1    electrophysiologic effects QT intervals, and arrhythmias

2    with fluoroquinolones, that there may be some effects

3    that are simply not knowable until actually put into

4    clinical practice.        Comments, Dr. Goldberger, or others?

5                     DR. BENNETT:     We could to use your example,

6    electrophysiologic effects.           You can do a dose escalation

7    study with a drug, 10 or 12 patients per arm, with careful

8    monitoring of QT and establish whether the drug has some

9    effect on QT.

10                    But absent an enormous prolongation, the

11   chances of seeing anything in a clinical trial database

12   of 5,000 people are essentially zero.             You are up there

13   probably       needing   tens   of    thousands    of   people     in

14   post-marketing databases to see anything, if in fact there

15   is any type of signal, just to use that as an example.

16    Bob may want to add some other things.

17                    DR. TEMPLE:     If I understood the question,

18   the question was how much do you know at the end of Phase

19   II.    There is an unfortunate idea that you know a great

20   deal about safety at the end of Phase II, and that is

21   just completely wrong.

22                    If you are lucky, you will have a few hundred

23   patients.       Well, you only know a very little bit about

24   safety from that.        Phase III, which in antibiotic terms,

25   given multiple indications, will typically have several

                                S A G CORP.
     202/797-2525               Washington, D.C.         Fax: 202/797-2525

1    thousand people, gives you much more assurance about

2    events up to the order of one in a thousand, or something

3    like that.

4                   But what Mark was describing is how we use

5    surrogates for toxicity in fact, a drug that prolongs

6    the QT interval a lot probably won't be approved unless

7    it does something really spectacular.

8                   A drug that causes certain kinds of liver

9    test abnormalities probably won't be approved because

10   we   believe    certain    findings          that   are     not     lethal

11   themselves, predict ultimate lethality.

12                  So that all of those things go on, and

13   nonetheless, some slip through, and have to be taken away

14   later.     But you only know a very little bit at the end

15   of Phase II because you just can't find out that much

16   in a couple of hundred people about real events.

17                  CHAIRMAN RELLER:        Dr. Shlaes.

18                  DR. SHLAES:      I just want to make a comment

19   that PK/PD.    I mean, in the anti-bacterial realm, we have

20   had very good animal models, which have been very

21   predictive of general success in the clinic for a very

22   long time.

23                  We have had proposed guidance, I believe,

24   that came from your predecessor, Dr. Reller, who was Dr.

25   Craig, suggesting that PK/PD be used much more in

                               S A G CORP.
     202/797-2525              Washington, D.C.                Fax: 202/797-2525

1    consideration of approval for certain indications.

2                   I think we are going to talk more about this

3    tomorrow.      But    we   have     known     about   this    in    the

4    anti-bacterial realm a lot longer than the HIV people

5    have known about it.

6                   And yet they have -- and as a matter of fact,

7    I am not sure how much PK/PD they have compared to what

8    we have, in terms of our confidence and ability to predict

9    success.

10                  Yet, they are using it much more routinely

11   compared to us.      So I think it is about time that we had

12   a little confidence in the predictability of these animal

13   models, and our ability to do PK/PD to get antibiotics

14   approved, especially for those indications which are

15   difficult because of low patient populations.                    And I

16   think we are going to talk more about that tomorrow.

17   Thank you.

18                  CHAIRMAN RELLER:         Dr. Metlay.

19                  DR. METLAY:      Well, I guess I would just add

20   as an extension to that that the whole issue of the impact

21   of the agents on microflora, oral and icteric microflora,

22   I think really very much that we had a lot more data on

23   the impact of cross different drugs, and we have been

24   in cross-classes.

25                  Because I think in the end that a lot of our

                                S A G CORP.
     202/797-2525               Washington, D.C.           Fax: 202/797-2525

1    indications and recommendations are going to ultimately

2    come down to those kinds of considerations.         So that we

3    could be better minimizing the impact on resistance

4    emergents.

5                   And I know that is the theme for tomorrow,

6    but it seems to be quite integral in this discussion as

7    well, and I am trying to understand whether there are

8    new compounds out there that really is value added.

9                   CHAIRMAN RELLER:      Dr. Cross and Dr. Shlaes.

10                  DR. CROSS:   I would just like to follow up

11   on Dr. Shlaes' comment, and just ask as a matter of

12   information, how good are the animal models for lots of

13   the things that we look at?

14                  For example, in the sepsis field, it is

15   accepted that there is no one good model which is

16   predictive of any therapy consensus.             I know from

17   personal work in animal models, for example, that there

18   are very few animals of staff orates.

19                  And that certain organisms, like klebsiella,

20   are not pathogenic in mouth-to-mouth except for one type.

21    So it would be very hard to test the drug for ESBL, for

22   example.

23                  So as a point of information, how good are

24   the animal models, in terms of the PK/PD?       Well, I think

25   as you know, you can carry out and do Bill Craig's model,

                            S A G CORP.
     202/797-2525            Washington, D.C.        Fax: 202/797-2525

1    which is the thigh infection model, and get I think very

2    good information on the critical pharmokinetic parameter

3    based on blood levels.

4                   So whether it is AUC, and whether it is peak,

5    and whether it time above MIC, and then you can use that

6    to make predictions, knowing PK and people about what

7    the efficacy will be under various circumstances.

8                   And in fact, Jerry Schentag, and Bill Craig,

9    and others, have carried out studies on people, and you

10   do see very good correlation between the PK/PD predictions

11   that you get from an animal model like the thigh model,

12   and what you see in people.

13                  Sometimes you have to do additional studies

14   on people, and as somebody brought up earlier the issue

15   of drug concentrations in the lung, and in the ELF.            Those

16   studies can now be done in people, and you can get very

17   good PK/PD information in people.

18                  And frequently this does correlate in what

19   you see in analysts.    So I think that is on example where

20   those correlations work quite in predicting the kind of

21   doses that you might have to use, and the kind of

22   concentrations that you might have to achieve in people.

23                  CHAIRMAN RELLER:      Dr. Ramirez, and then Dr.

24   Soreth.

25                  DR. RAMIREZ:    Yes.        Regarding my wish list

                             S A G CORP.
     202/797-2525            Washington, D.C.            Fax: 202/797-2525

1    for risk-benefit analysis in clinical trials, I would

2    like to add a better determination of cost of treatment,

3    because at this moment when we have a new antibiotic on

4    the market, the only thing we know is that it is going

5    to be less effective as the old antibiotic for the

6    management of the particular infection that this is.

7                   And then when we are on the P&T committee

8    trying to define what is the most cost effective therapy,

9    if one antibiotic costs $30 and the other costs $25, the

10   one that is most cost effective is the one that costs

11   $25.

12                  And this is because clinical drugs do not

13   allow us to define what is the most cost effective regime.

14    And I think that matches perfectly with the discussion

15   of looking at other outcomes besides clinical outcome.

16                  I think we need to be looking at other

17   outcomes for costs, and for acute exacerbation of chronic

18   bronchitis was already mentioned, and the time that the

19   patient takes to return to work, and these types of issues

20   need to be in the protocol.

21                  For community-acquired pneumonia, there are

22   large studies which indicated more or less (inaudible),

23   and probably we know the time to (inaudible), and we can

24   define in the hospital/patient time to switch therapy,

25   because we know that switched therapies are associated

                            S A G CORP.
     202/797-2525            Washington, D.C.    Fax: 202/797-2525

1    with early hospital discharge.

2                   And then I don't care too much if the two

3    antibiotics cure the patient the same at 30 days.                    If

4    the antibiotics decrease the length of a stay for two

5    days, this is going to be the most cost effective,

6    regardless of the cost for the antibiotic.

7                   And   for     nosocomial-acquired          pneumonia,

8    issues such as time of exacerbation of days in the

9    intensive care unit, because a decrease of one day in

10   the intensive care unit is going to be definitely the

11   most cost effective antibiotic for nosocomial-acquired

12   pneumonia.

13                  And I would like to see incorporated more

14   outcomes that are going to help us physicians when we

15   are admitting the P&T and try to define ways that are

16   the most cost effective antibiotics incorporated in the

17   clinical trials.

18                  CHAIRMAN RELLER:         Dr. Soreth.

19                  DR. SORETH:      I just wanted to make a comment

20   on safety considerations that Dr. Glode had raised.                       I

21   think in addition to clinical trials fundamentally not

22   being powered to tell us much about           or elucidate much

23   about uncommon adverse events, we also have to recognize

24   that in the clinical trial setting we are studying

25   patients under ideal conditions.

                                S A G CORP.
     202/797-2525               Washington, D.C.           Fax: 202/797-2525

1                     And that the amount of information that we

2    might have in the development program about the use of

3    concomitant       medications,        about      underlying    co-morbid

4    conditions, disease states that affect drug metabolism,

5    and excretion, and so forth, can be quite limited.

6                     And once a drug is on the market, and

7    thousands, and hundreds of thousands of patients are

8    exposed under less than ideal conditions, under real

9    conditions -- concomitant meds, states of hydration

10   varying widely -- only then do we really understand the

11   full    safety     or   toxicity       profile      of   a    drug,    but

12   unfortunately not at the time of an action.

13                    CHAIRMAN RELLER:          I would like to thank those

14   attending -- yes?

15                    DR. YUH:      Can I make two comments?

16                    CHAIRMAN RELLER:          Please.

17                    DR. YUH:      I think we are getting close.            My

18   name    is     Lianng   Yuh,    and    I    am    representing     PhRMA.

19   Actually, I am speaking for myself.                 I think the sense

20   of urgency is that we would like to know of any interim

21   solutions before we come up with any real good guidance

22   on antibiotic development, because a lot of the companies

23   have experience with different guidance, and I think it

24   has been there about -- longer than a year now.

25                    So we need some interim solutions before we

                                 S A G CORP.
     202/797-2525                Washington, D.C.             Fax: 202/797-2525

1    have a better solution.        I agree with Dr. Goldberger that

2    we need to welcome different indications, different

3    special cases, to come up with better solutions.                  But

4    interim solutions are important to us.

5                     Secondly, I would say that any designs we

6    are discussing, hopefully we can also address the concerns

7    from other regions, and not just the United States or

8    North    America,    because    we    tried   to   harmonize      our

9    experiments.

10                    There is a word they say, that patients are

11   waiting.       There is a sense of urgency and that we have

12   to move forward.      Thank you.

13                    CHAIRMAN RELLER:      Dr. Fleming.

14                    DR. FLEMING:        Are you still soliciting

15   responses to Issue 3?       A resounding yes, I think.

16                    CHAIRMAN RELLER:      For you, yes.

17                    DR. FLEMING:    Well, I will be brief.          I am

18   actually kind of folding my answers to Issue 2 and Issue

19   3 as well.       When I think of the factors that should be

20   considered, I think this is a little bit just stating

21   the obvious.

22                    But I think it is still worth stating, and

23   that is that I am assuming that this question is written

24   with the understanding that in many, if not most, cases

25   the primary confidence interval we are talking about here

                              S A G CORP.
     202/797-2525              Washington, D.C.          Fax: 202/797-2525

1    is on the primary end point, which I would hope would

2    usually be a direct measure of clinical benefit.

3                    And in that context, then certainly other

4    factors that should be considered are secondary measures

5    of clinical benefit, such as hospitalization.                    And

6    mortality      results,   safety,      tolerability,    drug-drug

7    interactions, will weigh in, as will as we have already

8    heard    convenience,     acceptability      of   administration

9    measures.

10                   And I had mentioned this morning in my

11   presentation      that    when    defining   margins,      if    one

12   anticipates substantial differences in issues relating

13   to safety, tolerability, drug-drug interactions, or

14   convenience, those issues in fact could influence the

15   actual final choice of the margin.

16                   External results from interventions that are

17   members of the same class are certainly factors that would

18   be considered.      And I mention last, not because it is

19   the least, but because I want to address it separately,

20   are measures of biological activity.

21                   And I have no concern about the fact that

22   clearly they are, such as bacterial eradication, measures

23   that influence your overall sense of strength of evidence

24   of effects having been established.

25                   My concern arises in those settings that we

                               S A G CORP.
     202/797-2525              Washington, D.C.         Fax: 202/797-2525

1    advocate their use in lieu of understanding results about

2    efficacy directly, or results about clinical end points

3    directly;      i.e.,   as   a    surrogate     marker   that   is       a

4    replacement end point.

5                    Just as a reminder of these classical complex

6    issues, one has to understand the disease process well

7    enough to be confident that this specific measure that

8    you have is really in essence fully capturing the

9    mechanism by which the disease process influences the

10   end point.

11                   And furthermore, one has to be confident that

12   there aren't significant unintended mechanisms of action,

13   anti-inflammatory activities, or other factors, that

14   could influence the critical end points that are not being

15   captured by this marker.

16                   So we run into some fairly complex issues.

17    We have mentioned specifically in question number two

18   that for the specific setting of meningitis the use of

19   the marker because of the fact that there is a quite clear

20   understanding of the biological mechanisms here, and

21   could be an appropriate replacement for a cure end point.

22                   Let me just mention that it is not completely

23   obvious thought that that gets you a very low sample size.

24    In HIV, when we are using viral load, we are looking

25   for differences that are easy to quantitate that are very

                                 S A G CORP.
     202/797-2525                Washington, D.C.        Fax: 202/797-2525

1    large in magnitude, and that allows us to get a much

2    smaller sample size.

3                   I think that Dr. McCracken was mentioning

4    this morning that with standard therapies that we might

5    be able to achieve 99 percent bacterial eradication, and

6    we should be able to with this marker be able to clearly

7    see differences.

8                   Well, if we wanted to discern the difference

9    between 99 and 98, that would take about 6,000 patients.

10    So that is no easy answer here.            If on the other hand,

11   we were trying to discern the difference between 99

12   percent bacterial eradication versus 93 percent, then

13   we are down to around 250 people.

14                  So my question here isn't so much whether

15   bacterial eradication is an important thing, but how much

16   can we fall away from 99 before we care, and that is a

17   critical question to find out whether use of that marker

18   truly will give you a much smaller sample size.

19                  CHAIRMAN RELLER:            Thank you, Tom.       Dr.

20   Goldberger, we have tried to have forthright comments

21   on all of the questions that you posed, and a rigorous

22   discussion, which I think has taken place.

23                  And I would like to in closing thank Dr.

24   Shlaes and the Pharmaceutical Research Manufacturers

25   Association, his colleagues, industry, and Dr. Tally,

                            S A G CORP.
     202/797-2525            Washington, D.C.           Fax: 202/797-2525

1    and Dr. Talbot, and other members representing the IDSA,

2    as well as of course all of the members of the committee,

3    including those who were added to the committee for

4    discussions from the pediatric subcommittee and other

5    advisory committees with expertise relevant to the

6    discussions today.

7                   So thanks to all, and we will reconvene for

8    Phase II tomorrow morning at eight o'clock with discussion

9    of the development of drugs for emerging resistance.

10                  (Whereupon, at 5:47 p.m., the meeting was

11   adjourned, to resume at 8:00 a.m., on February 20th,

12   2002.)














                            S A G CORP.
     202/797-2525            Washington, D.C.    Fax: 202/797-2525




                   S A G CORP.
    202/797-2525   Washington, D.C.   Fax: 202/797-2525

Shared By: