UNITED STATES OF AMERICA

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

         DEPARTMENT OF HEALTH AND HUMAN SERVICES

                 FOOD AND DRUG ADMINISTRATION

         CENTER FOR DRUG EVALUATION AND RESEARCH

                                   + + +

        MEDICAL IMAGING DRUGS ADVISORY COMMITTEE

                                   + + +

                                  MEETING

                                   + + +

                    Monday, February 9, 1998

                                   + + +



                 The Advisory Committee met in Versailles

Ballrooms I and II, Holiday Inn, 8120 Wisconsin Avenue,

Bethesda, Maryland, at 8:00 a.m., Ruth G. Ramsey, M.D.,

Chairperson, presiding.

PRESENT:

                 RUTH G. RAMSEY, M.D., Chairman

                 LEANDER B. MADOO, Executive Secretary

                 MARCO A. AMENDOLA, M.D., Member

                 PETER L. CHOYKE, M.D., Member

                 ROBERT W. JAHNKE, M.D., Member

                 JONATHAN M. LINKS, Ph.D., Member

                 LAURA L. BOLES PONTO, Ph.D., Member

                 CHARLES AUGUST, M.D., Consultant

PRESENT (Continued):

                 RALPH D'AGOSTINO, Ph.D., Consultant
                 RICHARD HAMMES, R.Ph. M.S. B.C.N.P,        Consultan

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

                 MARVIN KONSTAM, M.D., Consultant

                 CHARLES ROHDE Ph.D., Consultant

                 PATRICIA LOVE, M.D., FDA Representative

                 A. ERIC JONES, M.D., FDA Representative

                 LILIA TALARICO, M.D., FDA Representative

                 MICHAEL WELCH, Ph.D., FDA Representative

              MICHAEL BETTMAN, M.D., Sponsor
        Representative

              RICHARD T. DEAN, Ph.D., Sponsor
        Representative

              JEFFREY GINSBERG, M.D., Sponsor
        Representative

                 ALEXANDER GOTTSCHALK, M.D., Sponsor        Represent

                 JOHN LISTER-JAMES, Ph.D., Sponsor          Represent

                 J. KRIS PIPER, Sponsor Representative

              H. DIRK SOSTMAN, M.D., Sponsor
        Representative

                 RAYMOND TAILLEFER, M.D., Sponsor           Represent

                 ADEBAYO LANIYONU, Ph.D., FDA Reviewer

                 MAHBOOB SOBHAN, Ph.D., FDA Reviewer

            JOSEPH ZOLMAN, M.D., Ph.D., FDA Reviewer
ALSO PRESENT:


                 JOHN BALSER, Ph.D.

                 BOB CARETTA, M.D.

                 BOB HAGGERTY

                 KATHLEEN MADSEN, Ph.D.
                 CHRIS NICODEMUS, M.D.
                         S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
                 RICHARD WAHL, M.D.




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

                                                      PAGE

Introductions                4

Conflict of Interest Statement                  7

Open Public Hearing 10

Diatide Presentation:
      Introduction 11
      Diagnosing Deep Vein Thrombosis   14
      Scientific Rationale for AcuTect 22
      Clinical Study Overview 31
      Discussion of Training Example    38
      Efficacy Data from Pivotal Trials
            Richard T. Dean, Ph.D. 43, 53, 65, 72
            Jeffrey Ginsberg, M.D. 48, 67
            Alexander Gottschalk, M.D. 57
      Discussion of Clinical Experience 74
      Conclusion    83

FDA Presentation on Safety and Efficacy:

        Adebayo Laniyonu, Ph.D. 118
        Joseph Zolman, M.D., Ph.D.,             126
        A. Eric Jones, M.D.     130
        Mahboob sobhan, Ph.D.   137

Committee Consideration of Proposed Questions
      166




                              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                                                   (8:02 a.m.)

       3         CHAIRPERSON RAMSEY:           Good morning.      I'd

like to say good morning to everyone in the room, and
     4

thank 5you all very much for coming.

       6         I'm Ruth Ramsey, and I'll be chairing this

meeting this morning, and I think probably the best
     7

thing 8for us to do is just to go around the table, and

I will have everyone at the table introduce themselves
     9

and just briefly your role.
    10

      11         So we can start on the far end.            Dr. Jones,

      to
next 12 you if you could nudge -- just introduce

yourself.
    13

      14         DR. WELCH:       Yes.     I'm Mike Welch, Acting

Director, Division of Biometrics III, Office of
    15

Biostatistics.
    16

      17         DR. JONES:       My name is Eric Jones.          I'm

the clinical team leader in the Division of Medical
    18

Imaging Drug Products, FDA.
    19

      20         DR. LOVE:        Patricia Love, Division
Director of Medical Imaging, FDA.
    21

      22         DR. LINKS:       Jonathan Links, Johns Hopkins

University, a member of the Committee.
    23

      24         DR. PONTO:       Laura Ponto, MIDAC Committee

member, University of Iowa.
    25

      26         DR. CHOYKE:       Pete Choyke.         I'm a member of

the Committee.
    27                 I'm a radiologist at NIH.

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         MR. MADOO:        Leander Madoo, FDA.

       2         DR. D'AGOSTINO:         Ralph D'Agostino, Boston

University, biostatistician.
     3

       4         DR. HAMMES:        Richard Hammes, a nuclear

pharmacist and professor of pharmacy, University of
     5

Wisconsin, a member of the Committee.
     6

       7         DR. KASPER:        I'm Carl Kasper, a

hematologist, professor of medicine at the University
     8

of Southern California.
     9

      10         DR. JAHNKE:        Dr. Robert Jahnke.      I'm a

radiologist, member of the Committee from Albuquerque,
    11

New Mexico.
    12

      13         DR. AMENDOLA:        I'm Marco Amendola,

professor of radiology at the University of Miami and
    14

a member of the Committee.
    15

      16         DR. ROHDE:        I'm Chuck Rohde.      I'm a

biostatistician from Johns Hopkins University.
    17

      18         CHAIRPERSON RAMSEY:           Thank you very much,

and I'd just like to reintroduce Dr. Patricia Love who
     19

      just say a few words to us this morning.
will 20

      21         Thank you.

      22         DR. LOVE:        Thank you very much.

      23         I'd just like to also extend regrets from

Dr. Paul Botstein, who is unable to be with us this
    24

morning.
    25         She is out of town.

      26         Also, our Deputy Director, his wife just

delivered twins.
    27                    So he won't be with us today.
      28         You've met Dr. Welch.          Some of you who have
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
been in the meetings earlier, Dr. Nancy Smith was in
     1

this role.
     2           There's been some adjustments.            So Dr.

Welch 3is now our Acting Director for Biostatistics, as

was identified, and Dr. Mahboob Sobhan, whom you'll
     4

meet later, is the team leader for statistics.
     5

       6         We are looking forward to an exciting day

today.
     7     We have a number of very interesting issues to

discuss, but I'll save those other comments until after
     8

the open public session.
     9

      10         CHAIRPERSON RAMSEY:            Thank you, Dr. Love,

and welcome to everyone here on the Committee.
    11

      12         Our next agenda item is the open public

hearing, and at this time --
    13

      14         MR. MADOO:         Actually, I need to read the

conflict of interest statement and make a couple of
    15

meeting announcements.
    16

      17         First of all, welcome, Committee.            The

sponsor so kindly has provided us with desk copies of
    18

their presentations.
    19                            It should be in front of you with

a Boston clip on it.
    20

      21         If you examine your blue folders, you'll

     that we have the official meeting agenda.
note 22                                                      We also

     the
have 23 actual questions for the meeting.                You'll note

      the questions for the meeting are place in front
that 24

of you, and they're titled "Issues for Advisory
    25

Committee Discussion," and it's a three-page scenario,
    26

and it has essentially a couple of questions there
    27

terminating with approvability.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
       1         There's also a couple more items that have

been inserted in your folder.
     2                                    I was presented this

morning with a table that ostensibly relates to the
     3

division briefing document, and it looks like there's
     4

a correction.
     5                You might notice there are some data

points arrayed in a table, and it's titled "Number of
     6

Subject Enrolled in the AcuTect Clinical Studies."
     7

I'm sure the division will provide clarification on
     8

that when we reach that point.
     9

      10         There's also another item the division has

provided this morning to me, and it looks like it
    11

relates to aspects of their presentation.
    12

      13         Let me go ahead, please, and read the

conflict of interest statement for this meeting.
    14

      15         The following announcement addresses the

issue of conflict of interest with regard to this
    16

meeting and is made part of the record to preclude even
    17

the appearance of such at this meeting.
    18

      19         Based on the submitted agenda for the

meeting and all financial interests reported by
    20

Committee participants, it has been determined that all
    21

interests in firms regulated by the Center for Drug
    22

Evaluation and Research present no potential for an
    23

appearance of conflict of interest at this meeting with
    24

the following exceptions.
    25

      26         In accordance with 18 USC 2008(b)(3), full

waivers have been granted to Dr. Laura L. Boles Ponto
    27

and Dr. Marvin Konstam.
    28                            Copies of these waiver
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
statements may be obtained from the agency's Freedom
     1

of Information Office, Room 12A-30, Parklawn Building.
      2

       3         In the event that discussions involve any

other 4products or firms not already on the agenda for

which 5an FDA participant has a financial interest, the

participants are aware of the need to exclude
     6

themselves from such involvement, and their exclusion
     7

will be noted for the record.
     8

       9         With respect to all other participants, we

ask in the interest of fairness that they address any
    10

current or previous financial involvements with any
    11

firms whose products they may wish to comment upon.
    12

      13         And so let me stress we have a floor mic

out there, and as Dr. Ramsey will be chairing the
    14

meeting, we do have the opportunity during the open
    15

public hearing for people to come and address the
    16

Committee on germane issues relating to today's
    17

discussion.
    18             Please as you come to the mic specify your

name, affiliation, and if you were conveyed by a sponsor
    19

or otherwise.
    20

      21         That about entails my comments.        I notice

      Dr. Charles August arrived, and we're pleased to
that 22

      him, and I might note to Dr. Rohde that Dr. Young
have 23

      not be here today.
will 24                           So you might want to move up

one chair and be closer with your colleagues.
    25

      26         Thank you very much.

      27         CHAIRPERSON RAMSEY:         Thank you, Mr.
Madoo, and I certainly didn't mean to exclude you from
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
the program there, with apologies.
     1

       2         We'll next turn to the agenda item entitled

"Open 3Public Hearing," and at this time anyone is

welcome to step to the microphone.
     4

       5         (No response.)

       6         CHAIRPERSON RAMSEY:          Seeing no one coming

to the open microphone, we'll move on to the next item,
     7

which 8is the sponsor presentation by Diatide,

Incorporated, and I see on my agenda that the first
     9

speaker would be J. Kris Piper, Senior Director of
    10

Regulatory Affairs of Diatide.
    11

      12         MR. PIPER:       Good morning.        My name is Kris

Piper.
    13     I'm Senior Director of Regulatory Affairs at

Diatide.
    14

      15         On behalf of Diatide, I'd like to thank Dr.

      and members of the FDA and Dr. Ramsey and members
Love 16

of the Committee for giving us this opportunity today
    17

to come and talk to you about the new drug application
     18

for AcuTect.
    19

      20         AcuTect is a new radiopharmaceutical

diagnostic imaging agent with a proposed indication for
    21

scintigraphic imaging of acute venous thrombosis.
    22

      23         The clinical development of AcuTect began

in 1992, and we submitted the NDA in 1997.
     24                                                  This product

has been designated as a priority classification
    25

because currently there exists no imaging modality that
    26

can identify and distinguish acute venous thrombosis.
     27

In addition, as a Technetium labeled pharmaceutical,
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
AcuTect offers the potential for safety advantages over
     1

iodinated contrast venograms.
     2

       3         With us this morning, we have several

experts in the fields of radiology, nuclear medicine,
     4

and venous thrombosis to help us present our data on
     5

this product.
     6                Included is Dr. Bettman, Chief of

Cardiovascular and Interventional Radiology at
     7

Dartmouth; Dr. Ginsberg, Director of the
     8

Thromboembolism Unit at Hamilton Research Center; Dr.
     9

Gottschalk, professor of radiology at Michigan State.
    10

Unfortunately Dr. Gottschalk was not able to be with
    11

us personally today, but he wanted to convey his
    12

thoughts to the Committee and was able to provide us
    13

a videotape that we prepared yesterday, and we will be
     14

showing that later on in the program.
    15

      16         In addition, we have Dr. Sostman,

professor and Chairman of the Radiology Department at
    17

New York Hospital; and Dr. Raymond Taillefer, Chief of
     18

Nuclear Medicine at the Montreal Hospital and one of
    19

the clinical investigators in our pivotal studies.
    20

      21         Presenting for Diatide this morning are

Dr. Lister-James, our Senior Director of Research and
    22

Development, and Dr. Richard Dean, our CEO and Chief
    23

Scientific Officer.
    24

      25         The agenda that we will follow this morning

is slightly different than what you have that was
    26

prepared by Mr. Madoo.
    27                            First, Dr. Sostman will lead
off with a discussion of the clinical situation
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
involving diagnosis of DVT.
     1                                 Dr. Lister-James will

provide the scientific rationale and discuss the
     2

receptor binding properties and pharmacology of
     3

AcuTect.
     4         Following that, Dr. Dean will provide an

overview of the clinical study program, and then Dr.
     5

Lister-James will go through a training example of how
     6

we train the blind readers for the AcuTect scans.
     7

       8         Dr. Dean will then continue with a

discussion of the efficacy data from the pivotal trials
     9

of AcuTect, and he will be assisted by Dr. Ginsberg and
     10

Dr. Gottschalk.
    11

      12         Dr. Raymond Taillefer will then review his

experience with imaging with AcuTect.
    13                                                I might point

out that Dr. Taillefer has done several studies, well
    14

in excess of 40 case studies, using this product and
    15

is quite knowledgeable on it.
    16

      17         In conclusion, Dr. Bettman will have some

closing remarks, and Dr. Wyland, who's not shown on this
    18

slide, will also provide some remarks regarding his
    19

experience using AcuTect.
    20

      21         As you have seen in the briefing document

      we
that 22 provided and that FDA provided, today's

discussion is going to focus on the pivotal trials for
    23

      product.
this 24               These issues that have been raised by

FDA are what we intend to focus our presentation on.
    25

      26         In particular, we will discuss the

specific nature of AcuTect's receptor binding; the fact
    27

      AcuTect binds to platelet receptors and does not
that 28
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
bind to endothelial cell receptors; the ability of
     1

AcuTect to distinguish acute venous thrombosis from
     2

other 3causes of leg symptomatology; the fact that

AcuTect performs equally well in patients whether they
     4

are on heparin or other anticoagulants or not; the
     5

rationale for our pivotal trial design and the proposed
     6

efficacy criteria that we selected; the results of the
     7

primary and secondary analyses of efficacy in our
     8

pivotal studies; the decision and the rationale that
     9

we used in selecting the Hamilton Research Center to
    10

conduct a second blind read; and finally, why we believe
    11

      the data presented in the new application for
that 12

AcuTect support the proposed indication of this product
    13

      scintigraphic imagining agent for venous
as a 14

thrombosis.
    15

      16         With that I'd like to turn the podium over

to Dr. Sostman.
    17

      18         DR. SOSTMAN:        Good morning, ladies and

gentlemen.
    19           As you've already heard, my name is Dirk

Sostman.
    20         I'm professor and Chairman of Radiology at

Cornell Medical College and New York Hospital, and
    21

Diatide has asked me to appear as an independent expert,
    22

having spent many years working in this area, to
    23

indicate something of the context in which this product
    24

application is made.
    25

      26         This disorder, deep vein thrombosis, is a

highly prevalent one.
    27                            Estimates of the annual
incidence range as high as five million cases per year.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         It's associated with significant morbid

complications in the form of post phlebitic syndrome,
     2

but perhaps the most devastating complication is that
     3

of pulmonary embolism.
     4                              Approximately 30 percent of

deep vein thrombi which occur above the knee result in
     5

pulmonary embolism, and approximately 30 percent of
     6

pulmonary emboli are fatal in the absence of therapy.
     7

       8         Fortunately, effective therapy is

available in the form of anticoagulants.
     9                                                   However, the

problem is that anticoagulants themselves are
    10

associated with significant complications.
    11                                                      In the

Pioped study, for example, seven percent of patients
    12

who underwent anticoagulant therapy experienced
    13

significant bleeding complications, such as major
    14

falls in hemoglobin, bleeding into a joint, or bleeding
    15

      the brain.
into 16

      17         Accordingly, accurate diagnosis is

mandatory, and there are still significant limitations
    18

in diagnostic tests.
     19                           Clinical diagnosis is well known

to be nonspecific and insensitive, and imaging tests
    20

themselves remain with significant limitations.
    21

      22         Just to emphasize the importance of venous

imaging in this particular context of clinically
    23

suspected pulmonary embolism, this was really first
    24

underlined by a study from the Hamilton Group in the
    25

early '80s, published in the Annals of Internal
    26

Medicine.
    27

      28         Approximately 230 patients with
                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
clinically suspected pulmonary embolism were studied.
     1

A hundred of these had abnormal perfusion lung scans.
     2

Of these, 74 underwent pulmonary angiography and
     3

bilateral angiography, and 52 had venous
     4

thromboembolic disease, either pulmonary embolism
     5

alone, deep vein thrombosis alone, or the combination.
     6

       7         And in this study, patients with disease

requiring therapy, that is, either DVT or PE, were
     8

detected at rather similar rates by either imaging the
     9

lungs or by imaging the legs with bilateral venography.
    10

      11         The overall prevalence of disease in this

study was approximately 40 percent.
    12

      13         However, venography has fallen into some

disuse, and certainly bilateral venography is a very
    14

impractical test primarily because of the occurrence
    15

of complications.
    16                      Some of these have been reduced

since the publication of these series with the advent
    17

of nonionic contrast material, but there remain
    18

significant problems:
    19                            pain in a significant number of

patients; local inflammatory responses; extravasation
    20

of contrast material with the potential for soft tissue
     21

injury; the actual induction of the disease, DVT, by
    22

the test; and a variety of systemic complications of
    23

iodinated contrast materials, such as anaphylaxis or
    24

renal toxicity.
    25

      26         In addition, venography, although it's

considered the in vivo gold standard, is not without
    27

interpretive difficulties.
    28                                  Certainly, a well filled
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
venograph with no intraluminal filling defects is
     1

widely accepted as negative, and a case like this in
     2

which 3multiple filling defects are clearly outlined in

the calf and distal popliteal vein is widely accepted
     4

as a positive study, and there is little dispute about
      5

this. 6

       7         However, false negatives do occur, and

this is an example of a DVT which was originally
     8

considered as a negative and really resulting from
     9

vascular overlap.
    10                       Additional imaging did

demonstrate DVT in this patient.
    11

      12         Technical difficulties, such as

nonfilling particularly of the pelvic veins.
    13                                                       In this

case, the pelvic veins were poorly filled because of
    14

the presence of extensive bilateral iliac and caval
    15

thrombosis, which was demonstrated by bilateral direct
    16

pelvic venography.
    17

      18         Even with appropriate technique,

interpretive difficulties can occur.
    19                                                For example,

     patient with narrowing of the iliac vein.
this 20                                                      Is this

due to inherent neural thrombus or is it do to extrinsic
     21

compression?
    22              Additional imaging in this case, again,

demonstrated that this patient had pelvic DVT.
    23

      24         Perhaps the most challenging area for

conventional venography is the detection of acute
    25

thrombus in a patient with prior disease and the
    26

distinction of acute from chronic deep vein thrombosis.
    27

      28         As you can see, multiple collateral
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
pathways open up in this setting, and residual defects
     1

occur 2which can be difficult or impossible to

distinguish from acute DVT.
     3

       4         Although venography can be difficult to

interpret, it can be inconclusive, and it can be wrong,
     5

and it is not this that has led to its almost wholesale
     6

replacement by ultrasound.
     7

       8         However, I would point out that in my

opinion, the selection of a center which does have
     9

extensive current experience with venography as the
    10

     standard read is most appropriate, and I think the
gold 11

Hamilton Center is arguably the best one in this
    12

hemisphere for that role.
    13

      14         However, the replacement of venography by

ultrasound has occurred largely because of the fear of
    15

complications, and just to indicate to you how
    16

wholesale this replacement has been, when I was at Duke
    17

University, we reviewed approximately 300 patients who
    18

had venous imaging for the suspicion of pulmonary
    19

embolism.
    20          Of these 300 patients, a total of six

underwent contrast venography.
    21                                      The others were

managed with other imaging modalities which were
    22

noninvasive.
    23

      24         Chief among these is ultrasound.        It's an

excellent test, being both safe and cheap, and in
    25

appropriate settings, it's highly accurate.
    26                                                   For

example, in the thigh in the presence of clinically
    27

localizing findings, sensitivity and specificity are
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
in the 90s in almost all series.
     1

       2         However, ultrasound does have significant

diagnostic limitations.
     3                            Even in the thigh, in the

absence of clinically localizing findings, sensitivity
     4

has been reported as low as 38 percent, and in our review
      5

of the English language literature last year,
     6

approximately 1,800 published cases, the average
     7

sensitivity of ultrasound in this setting was
     8

approximately 65 percent.
     9

      10         Additionally, ultrasound is more

difficult in the calf and in the pelvis, and the calf
    11

sensitivities have been reported as low as 30 percent,
    12

specificities as low as 85 percent, and the average
    13

sensitivity in the calf in the absence of localizing
    14

findings in our review was 28 percent.
    15

      16         Accordingly, although ultrasound is an

excellent test and has been widely adopted, it is not
    17

the answer.
    18             Therefore, all the currently used

modalities for venous imaging have limitations,
    19

venography with complications and the iodinated
    20

contrast material; difficulties in distinguishing
    21

acute from chronic thrombosis; and difficulties in
    22

delineating the proximal extent of clot; ultrasound
    23

      reduced accuracy in the calf and pelvis;
with 24

significant limitations in distinguishing acute
    25

disease from chronic disease; and reduced sensitivity
    26

in patients without clinically localizing findings.
    27

      28         A few centers are using magnetic
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
resonance, but it is quite expensive, and its
     1

availability is limited, and significant experience is
     2

required because of the presence of flow artifacts
     3

which 4can look like thrombus.

       5         Against this background, for a number of

      investigators have sought a preferential hot spot
years 6

clot imaging agent, and this is an example of hot spot
     7

imaging with radiolabeled platelets, an agent which I
     8

personally wasted about two years of my life.
     9

      10         This was a good agent if you were willing

to accept that it was not accurate in the presence of
    11

anticoagulants and if you were willing to wait for
    12

several hours for imaging, and both of these
    13

limitations really precluded its widespread clinical
    14

adoption.
    15

      16         I've had the opportunity to review the

briefing document for the agent which you're asked to
    17

consider today, P280, and this is an example of a
    18

positive calf DVT with P280.
    19

      20         I was not involved in the development of

      agent or in the trials, but I have reviewed the
this 21

briefing document, and I would simply comment, if I may,
    22

      it
that 23 appears to me from this document, first, that

the Hamilton blind read is the appropriate gold
    24

standard, and, second, that the agent appears to be safe
    25

and effective, and my clinical impression is that it
    26

      potentially fill some important niches in the
will 27
clinical work-up of patients, such as the acute versus
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
chronic disease or post operative screening for a
     1

symptomatic DVT in high risk populations.
     2

       3          That concludes my remarks.           I'd like to

thank 4you for your attention and for the opportunity

to appear before you.
     5

       6          The next presentation will be Dr.

Lister-James from Diatide, who will discuss some of the
     7

preclinical and other issues.
     8

       9          DR. LISTER-JAMES:        Thank you, Dr.

Sostman.
    10

      11          Good morning, Dr. Ramsey, members of the

Committee, Dr. Love, members of the FDA.
    12

      13          In the next few minutes, I'm going to

review the scientific basis of the product AcuTect, and
    14

in particular, I'm going to address the following
    15

points:
    16         the need for this product; what is AcuTect;

why it should work; and how it works.
    17

      18          Now, the process of thrombosis, it's been

      established that thrombus biochemistry and a
well 19

disease state are interrelated, and in particular, the
    20

difference between acute venous thrombosis and chronic
    21

venous thrombosis is characterized more by differences
    22

in biochemistry than by differences in anatomy.
    23

      24          You just heard Dr. Sostman address some of

the difficulties inherent in anatomical imaging
    25

techniques.
    26              Approaches to imaging acute venous

thrombosis, there have been several approaches,
    27

including radiolabeled platelets, which Dr. Sostman
    28

                               S A G CORP.
202/797-2525    Washington, D.C.   Fax: 202/797-2525
just mentioned, but this procedure has limitations as
     1

he mentioned:
     2                inconvenient preparation, blood

clearance that's too long, and problems in sensitivity
     3

in the presence of anticoagulants.
     4

       5         I-125 Fibrinogen, which is a scanning

technique, not an imaging technique, which FDA approved
     6

at one time, was shown to be useful for the detection
     7

of acute venous thrombosis.
     8                                 This product has

limitations, not the least of which is it's been removed
     9

     the market because it's a blood product.
from 10                                                   It also

has slow blood clearance, requiring delayed scanning.
     11

      12         Radiolabeled antibodies have been

investigated for imaging DVT.
    13                                    An example of one of

those papers is shown in this slide.
    14

      15         Antibodies are large, complex molecules

      in
with 16 many cases slow blood clearance.              They also

carry with them the potential of an immune response.
    17

      18         And, therefore, there was an unmet need for

a rapidly clearing marker of acute venous thrombosis.
    19

AcuTect was designed to fulfill this unmet need.
    20

      21         AcuTect is a product which produces a

radiopharmaceutical, Technetium
    22                                       Tc99m apcitide.

Apcitide is a 13 amino acid synthetic peptide.
    23                                                      It

contains a binding region for the platelet GPIIb]IIIa
    24

receptor and a Technetium 99m complex.
    25

      26         The structure of the radiopharmaceutical

is shown here with a binding region in yellow on the
    27

      and a Technetium complex on the right.
left 28
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         The active binding region of AcuTect is an

analog of the arginyl-glycyl-aspartic acid sequence,
     2

also known as the RGD sequence, which is present four
     3

times 4on the molecule fibrinogen, and the RGD sequence

bides 5with the GPIIb]IIIa receptor on platelets.

       6         In AcuTect the arginine has been replaced

with a synthetic amino acid, and I'll come back to this
     7

point 8a little bit later because we believe this

modification is important in the receptor specificity
     9

of the agent.
    10

      11         So the active binding region is shown here

on the slide on the left.
    12                              The comparison is the RGD

sequence of the positively charged arginine,
    13

negatively charged aspartic acid, and on the right the
    14

binding region of AcuTect, a synthetic amino acid
    15

positively charged, negatively charged aspartic acid.
    16

      17         About the GPIIb]IIIa receptor, this

receptor is expressed only on platelets.
    18                                                It is not

expressed on endothelial cells.
    19                                      It is key in platelet

aggregation where it mediates the binding of fibrinogen
    20

platelets in the process of platelet aggregation.
    21                                                        It

     binds to fibrinogen when platelets are activated.
only 22

      23         This is shown schematically here.         These

are platelets that are bound to the extracellular
    24

matrix that are breaking the endothelium.
    25                                                 Actually

      adhesion to the extracellular matrix is not
this 26

GPIIb]IIIa receptor mediated, but what is mediated by
    27

      receptor is the aggregation of one platelet to
this 28
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
another through the molecule fibrinogen shown by the
     1

three 2blue dots here.

       3         Each platelet contains 50,000 GPIIb]IIIa

receptors expressed on its cell surface, which makes
     4

it one of the most highly expressed cell surface
     5

receptors, and in addition to binding the molecule
     6

fibrinogen, it also binds AcuTect.
     7

       8         This is a showing a little bit more detail

here the GPIIb]IIIa receptor on the surface of the
     9

platelet, normally binding fibrinogen, also binds the
    10

active binding region of AcuTect.
    11

      12         Why should AcuTect work?             It's because

platelets are involved in acute, but not chronic venous
    13

thrombosis, and AcuTect binds to platelets.
    14

      15         Going into a little bit more detail

regarding deep vein thrombosis, starting with the
    16

original which is normally felt to be regions of stasis
    17

in the lower limbs or breaks in the endothelium, coupled
     18

      the condition of hypercoaguability, initially
with 19

      believed to involve platelet deposition with
it's 20

subsequent incorporation of fibrin in red blood cells,
    21

and then propagation proximally with addition of
    22

additional platelets and fibrin.
    23

      24         This condition is a condition of acute

venous thrombosis, and thrombus may then go on to
    25

embolize or to organize as is shown schematically in
    26

      slide.
this 27            Platelet deposition in a venous valve
cusp, formation of the thrombus, propagation
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
proximally with addition of additional platelets and
     1

fibrin, and then the potential for embolization or to
     2

organization.
     3

       4         This condition here is the condition of

acute 5venous thrombosis, and this is the condition

which 6is most likely to embolize.             Once the thrombosis

becomes organized, it has much less chance of resulting
     7

in embolization.
     8

       9         The right-hand side then is the condition

of chronic thrombosis.
     10                           This is the condition of acute

thrombosis.
    11

      12         And so acute venous thrombosis has the

characteristics that it may or may not be occlusive.
    13

It often involves proximal extension of the initial
    14

thrombus.
    15          It is unorganized, fragile, and has a high

potential embolization.
    16

      17         Platelets are incorporated into the

thrombus in acute thrombosis where they're activated
    18

at the thrombus and where they express the GPIIb]IIIa
    19

receptor.
    20

      21         How AcuTect works?         Well, it binds to the

GPIIb]IIIa receptor on activated platelets.
    22                                                    It does

not bind to endothelial cell receptors.
    23                                                 It does not

     to
bind 24 red or white blood cells, and what is not bound

to the thrombus is cleared rapidly from the
    25

bloodstream.
    26

      27         Regarding the affinity of AcuTect for the
receptor, we have determined that the product inhibits
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
the binding of fibrinogen to the receptor with an IC-50
      1

of 1.8 nanomolar, indicating a high affinity of the
     2

product for the receptor.
     3

       4         And from the literature it has been shown

that fibrinogen has an inhibition constant for
     5

platelets of about 120 nanomolar, indicating that
     6

AcuTect has higher affinity for the receptor than its
     7

normal ligand fibrinogen.
     8

       9         Regarding receptor specificity, the

vitronectin receptor is expressed on platelets and
    10

endothelial cells, and the vitronectin receptor is
    11

receptor which is related to the GPIIb]IIIa receptor.
    12

In fact, it has a high degree of homology with the
    13

GPIIb]IIIa receptor, and if one was to expect any
    14

cross-reactivity of AcuTect with another receptor,
    15

     is
this 16 the one that one would expect it to cross-react

with.
    17

      18         We found that AcuTect does not bind to the

vitronectin receptor.
    19                            With concentrations as high as

1,000 nanomolar, it does not inhibit the binding of
    20

vitronectin to its receptor, and we believe that this
    21

modification or that this selectivity, the receptor
    22

selectivity, is based on the modification of the
    23

binding region of the agent.
    24

      25         We also looked at another assay to assess

the binding of AcuTect to the GPIIb]IIIa receptor using
     26

an assay of platelet aggregation since platelet
    27

aggregation is a GPIIb]IIIa receptor mediated -- sorry
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
-- dependent process, and we looked at the inhibition
     1

of ADP induced platelet aggregation in plasma.
     2                                                  This

is with human platelets.
     3

       4         The product inhibited the platelet

aggregation with an IC-50 of .38 micromolar, indicating
     5

specific binding to the GPIIb]IIIa receptor, and I
     6

should mention here that this is an in vitro assay to
     7

assess or to evaluate the binding of the agent to the
     8

receptor.
     9          The maximum theoretical possible

concentration of the product in vivo in a human does
    10

not reach concentrations high enough to cause any
    11

clinically significant platelet aggregation.
    12

      13         We also looked at -- we used that

particular assay of inhibition of platelet aggregation
    14

to evaluate the effect of anticoagulants on the binding
     15

of AcuTect to the receptor, and using blood from
    16

patients who had taken aspirin, we found no change in
    17

the ability of AcuTect to inhibit platelet aggregation,
     18

indicating that aspirin does not interfere with the
    19

binding of AcuTect to the receptor.
    20

      21         We also looked at the effect of heparin

where we conducted the assay in the presence of the
    22

therapeutically -- a normal therapeutic concentration
    23

of heparin, and again, we found no change in the
    24

inhibition of platelet aggregation by AcuTect when
    25

heparin was present, indicating no effect of heparin
    26

on the binding of AcuTect to the receptor.
    27

      28         And as you will see later on in the clinical
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
data, 1this is consistent with the clinical findings

that there was no effect on the ability of AcuTect to
     2

detect venous thrombosis whether anticoagulants were
     3

used or not.
     4

       5         We also looked at the binding of the

radiotracer to human platelets, and we found that the
     6

product bound three times greater to activated
     7

platelets than to resting platelets.
     8

       9         We also look at the in vivo thrombus update

in the dog model where an acute venous thrombus was
    10

induced in the femoral vein, and then we were able to
    11

obtain external images of the thrombus, and upon
    12

excision of the thrombus, obtained thrombus-to-blood
    13

ratios of four and thrombus-to-muscle ratios of 11,
    14

indicating the specific binding of AcuTect to acute
    15

venous thrombosis.
    16

      17         In terms of general pharmacology and

biodistribution, when Dr. Taillefer later on reviews
    18

the clinical cases, he'll talk a little bit about
    19

biodistribution, but one point I'd like to make here
    20

since the issue of immunogenicity was raised by the
    21

agency is that we conducted a study of this product in
    22

guinea pigs, which reached the peak doses over a
    23

two-week period followed by a challenge dose and saw
    24

no evidence of an immunogenic response.
    25

      26         And in addition, in a Phase I study of about

30 patients, we also saw no immune response, and this
    27

is what one would expect inasmuch as the product a
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
small, synthetic peptide injected intravenously in low
     1

concentration; would not expect an immune response from
     2

this sort of product, as distinct from monoclonal
     3

antibodies.
     4

       5         And so in summary, we conclude that based

on the data that I've just presented, that AcuTect
     6

should and does bind specifically to acute venous
     7

thrombi.
     8

       9         Now I'd like to turn the floor over to Dr.

Richard Dean, who will present the clinical findings.
    10

      11         DR. DEAN:        Good morning.         I'm Richard

Dean, and I will be leading a discussion and presenting
    12

an overview of the clinical studies to Diatide in my
    13

capacity as Chief Scientific Officer at Diatide.
    14

      15         I'll be assisted in the presentation by the

following individuals who have been previously
    16

introduced to you.
    17

      18         The clinical program consisted of a total

of 710 patients.
    19                    There were five Phase III studies

done. Two of those five Phase III studies constituted
    20

the pivotal studies for efficacy of this product.
    21

      22         Safety is indicated on this slide.              These

are adverse events occurring in more than one subject
    23

in the entire population of 710 patients.
    24                                                      As you can

      in
see, 25 each of these categories, the adverse events

      all causes was one percent or less.
from 26

      27         Additionally, we had the opportunity
through the pivotal studies of comparing the safety
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
directly to venography.
     1                            These data are indicated on

this slide, where we list treatment related adverse
     2

events associated with AcuTect or venography in these
     3

pivotal studies.
     4

       5         There's about 270 patients in each of these

populations.
     6              As you can see, categories that were

reported are listed here, and the difference between
     7

AcuTect for the total adverse events was statistically
     8

significant.
     9

      10         So we can say that compared to venography,

AcuTect is significantly safer.
    11

      12         Those constituted the major databases for

the safety of the product.
     13                              I'd now like to move on to

the efficacy of the product, and to do that I'd first
    14

      to
like 15 address the pivotal trial design.

      16         One of the key things for consideration is

the type of agents we're comparing.
     17                                         There is no active

agent that we can compare this to.
    18                                         So we are left with

comparing it to an anatomical imaging technique,
    19

venography, which is the gold standard.
    20

      21         These are the measures.              There are two

different types of measures that are performed.
    22                                                          So we

      to
have 23 be mindful of those as we proceed forward in

the study and design the study.
    24

      25         Additional information is shown on the

course of the disease here in this cartoon.
    26                                                      These are

the three stages.
     27                    You have a normal going to an acute.
A certain fraction of patients with acute disease will
     28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
go on 1to have a chronic condition as shown here, and

then a certain portion of these will go on to have an
     2

acute 3event on top of the chronic event.

       4         You can see how AcuTect is expected to

perform, picking up the acute clot, and you can see the
     5

anatomical test, how that is expected to perform, where
     6

this would be either venography or ultrasound, but in
     7

our particular case it was venography in the pivotal
     8

studies.
     9

      10         Herein lies part of the problem with this

particular disease, as was outlined with Dr. Sostman,
    11

inasmuch as the anatomical tests have difficulty in
    12

distinguishing these two conditions.
    13

      14         This cartoon here also highlights the

potential problem in comparing a biologically active
    15

or physiological test with an anatomical test.
     16                                               The CV

is contrast venography.
    17

      18         There may be some cases where a

nonocclusive clot may not be picked up by contrast
    19

venography for one reason or another, where AcuTect may
    20

      that up, and that would bias the study against
pick 21

AcuTect.
    22         Again, we just need to keep these things in

mind.
    23

      24         And the major thing that we believe would

      the study against AcuTect, of course, is old
bias 25

thrombi for which the anatomical test would indicate
    26

     a
it's 27 positive, but AcuTect would not be able to detect
acute disease.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         So with these limitations in mind, we

proceeded with the following staple data, and that is
     2

that AcuTect in venography will have the highest
     3

concordance in acute disease, and it was on that basis
     4

that we designed the entry criteria to capture that
     5

particular condition, and that is each patient was
     6

entered in the trial if he had the onset of symptoms
     7

within ten days or was ten days post surgery.
     8

       9         Each patient had both a venogram and an

AcuTect.
    10         So it was a within patient study.

      11         The efficacy criteria were decided upon

based on three criteria:
    12                             what was known about

interobserver agreement rates with contrast
    13

venography.
    14             That had to be taken into account.

      15         In addition, the prior experience that we

had with AcuTect was taken into account, as well as
    16

certain limitations that venography may have as you
    17

     seen previously on the slides comparing anatomical
have 18

tests to a physiological test.
    19

      20         So the target agreement rate we believed

would be possible a priori was 75 percent with a lower
    21

confidence limit of 60 percent.
    22                                       Now, that's not to

indicate that we believed that the agent is that
    23

accurate or not, but this is the prospective design that
    24

was agreed upon before proceeding with the trial.
    25

      26         The endpoints in the analyses are

indicated here.
    27                   As you would expect, the final
clinical diagnosis and the clinical venography reads
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
had very high agreement.
     1                              Those agreement rates were

close 2to 95 percent in each of the studies, Study A and

B.     3

       4         Priority efficacy endpoint is indicated is

indicated down here, which is a comparison of blind read
     5

AcuTect to blind read venography.
     6

       7         There were three different readers for

each of the AcuTect images, and there were three
     8

different readers for each of the venogram images.
     9                                                    The

venography readers were different than the AcuTect
    10

readers.
    11

      12         Now, the secondary endpoint was a

comparison of blind read AcuTect to the clinically
    13

interpreted venograms.
    14                            You expect here that with

additional clinical information, as has been reported
    15

in the literature, that you would have increased
    16

accuracy in assessing the disease.
    17

      18         Now, the venograms were evaluated as

follows.
    19         There was an institutional venogram

interpretation by a radiologist at the site, and then
    20

there were, again, three certified radiologists blind
    21

to the clinical information.
    22

      23         It's important to note right here that

there were no other selection criteria for these
    24

radiologists.
    25                It was assumed at this point -- and

these were all U.S. radiologists -- it was assumed at
    26

      point that a certified radiologist selected
this 27
randomly across the nation would be an appropriate gold
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
standard for this particular comparison.
     1

       2         Now, one of the questions you may be asked

today 3is is the institutional venography read an

appropriate gold standard.
     4                               We offer you the following

information, which would be in consideration of that
     5

question.
     6

       7         The percent of venograms that were

documented read prior to the AcuTect test are indicated
     8

here. 9 As you can see, they are in the 70 and 80 percent

region.
    10

      11         The way that was done is by indicating that

on the case report form.
    12                             So in many cases, the entry

was made and dated, or I would say in all of these cases
     13

the entry was made and dated on the case report forms
    14

prior to the performance of the AcuTect test.
    15

      16         In those cases for which that did not

happen, we followed up and documented by testimony that
    17

the venograms were read without prior knowledge of the
     18

AcuTect result.
    19

      20         That's not surprising because in most

institutions venography and nuclear medicine scans are
    21

      in
read 22 different locations within the institution.

      23         AcuTect images were evaluated as follows.

There was an institutional interpretation reported.
    24

      there were, as in the study indicated, three
Then 25

independent nuclear medicine physicians blind to the
    26

clinical information.
    27

      28         We had initiated the read for the database
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
using 1the combined time points.             The agency then

requested part way through that exercise that we
     2

conduct the read with both combined time points and each
     3

individual time point, blinding each individual time
     4

point 5to the particular patient to produce a much more

comprehensive data set.
     6

       7         We, as a matter of course, decided to

complete this read and report the information and read
     8

two as the requested study performed by the FDA.
     9

      10         I'd now like to reintroduce Dr. John

Lister-James, who will review with you the reader
    11

training for the interpretation of the AcuTect images.
    12

      13         DR. LISTER-JAMES:        If you would bear with

me for a second and let's take a couple of seconds for
     14

the computer to come up.
    15

      16         What I'm about to show you briefly is how

we trained our readers for the blind read of AcuTect
    17

scans.
    18     The purpose of this part of the presentation

is just to show you the reader training.
     19                                               Dr. Taillefer

later on in the program will review image
    20

characteristics and present case studies.
    21

      22         I'd also like to point out that the quality

of the images that you're about to see are not
    23

representative of what the readers saw since they were
    24

trained and read imagines on a large computer monitor.
    25

Unfortunately the only way to show you all the images
    26

at the same time is to use a projector, which doesn't
    27

do justice to the images.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         We started out by reviewing the venous

anatomy of the lower limbs, in particular, the
     2

difference between the deep veins and the superficial
     3

veins 4with the readers, and then went on to review the

blind 5read criteria with them, which included the

following:
     6

       7         That we were looking for linear central,

that's deep venous uptake; asymmetric when comparing
     8

similar segments, one leg to the other; and that the
     9

anterior views and posterior views were to be
    10

consistent with one another.
    11

      12         And when they were reading full image sets,

that's three time points, that the thrombus should be
    13

visible at more than one time point.
    14

      15         Now, we trained the readers on 20 images.

In the interest of time, I'm going to just show you three
     16

      and we do have some additional images.
now, 17                                               If any

member of the panel is interested in seeing additional
    18

studies, I can make those available at a break.
    19

      20         Just to orient you here, there's three sets

of images, three different time points, ten minutes,
    21

60 minutes, 120 minutes, and in this particular scanned
     22

      of
sets 23 images, they are presented as follows:

anterior pelvis, thigh, knee, calf, and posterior --
    24

there is no posterior pelvis -- posterior thigh, knee,
    25

and calf, and they're duplicated at each of the time
    26

points.
    27

      28         And I should mention here that these are
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
viewed as viewed by the gamma camera.
     1                                                So this is in

the anterior view the patient's right leg, left leg,
     2

right/left, right/left, right/left, and then on the
     3

posterior view right/left, right/left, right/left.
     4

       5         The readers were allowed to use different

gray scales and different color scales.
     6                                                  So either

black 7on white, as shown here, or white on black, and

they were allowed to use a contrast adjustment using
     8

this color bar, which I'm operating now, to adjust the
     9

contrast of the image looking for linear central,
    10

that's deep venous uptake, asymmetric from one leg to
    11

another.
    12

      13         Now, this is a negative case, and there's

no asymmetry in any of these images, indicating the
    14

absence of deep vein thrombosis, of acute venous
    15

thrombosis.
    16

      17         Turning to a positive image, I think you

may be able to see on this -- well, let me just reorient
     18

you here.
    19          In this particular set of imagines, the

anterior studies are on the left, and the posterior on
    20

the right for each of the time points.
    21

      22         And I think without any contrast

enhancement you can see that in the left calf of this
    23

individual at 60 minutes, you can see linear central
    24

uptake in the deep vein of the calf.
    25                                                You can see it

in the anterior view, in the posterior view,
    26

anterior/posterior, and also at 120 minutes,
    27

anterior/posterior.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         And this may be -- oops, let me just back

off here.
     2          Adjusting the contrast brings it up, makes

it a little bit easier to see.
      3                                     This is just adjusting

the contrast, black on white.
     4                                      It makes the images a

little easier to see.
     5

       6         And also if one should use a color scale

and a 7little bit of contrast adjustment, then you can

clearly see the thrombus in the calf of this individual.
     8

       9         Turning to another positive case, this one

     little bit more difficult to see than the previous
is a 10

one. 11It doesn't become immediately apparent as the

images first come up.
    12                            However, a little bit of

contrast adjustment, you will be able to see that there
    13

is asymmetric uptake in the right thigh of this patient
     14

versus the left thigh.
    15                             You can see it anterior and

posterior, and you can see here increased uptake in the
    16

right popliteal of this individual versus the left.
    17

      18         Now, I should mention here that you can

     see
also 19 a little bit of superficial venous uptake.             We

reviewed this with the readers to indicate to them that
    20

      should not read this as acute deep venous
they 21

thrombosis.
    22             So we read around that.

      23         Also, in some patients there's some uptake

around the knee.
    24                    This was also not considered to be

      venous uptake, and in some cases soft tissue
deep 25

uptake, again, not considered to be deep venous uptake.
    26

      27         What we were looking for is linear central,
that's deep venous uptake, asymmetric one leg to the
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
other, and again, I think you'll be able to see in this,
     1

again, using color here it makes it a little bit easier
     2

to read, looking at the asymmetry one leg to the other.
     3

       4         So I think I'll stop here, and as I say,

if you'd like to see any more, we can make those
     5

available at the break, and I'd like to turn the floor
     6

back to Dr. Dean.
     7

       8         DR. CHOYKE:        Can I ask a quick question

about 9the popliteal areas?             They're slightly warmer

because they're more superficial; is that?
    10

      11         DR. LISTER-JAMES:          If there is deep venous

uptake, they are noticeably different.
    12

      13         DR. CHOYKE:        No, even in the normal they

      slightly.
were 14

      15         DR. LISTER-JAMES:          Oh, yes, but in a

normal case you'll see no asymmetry that will be visible
    16

in both legs, whereas if there's thrombus there, then
    17

you see asymmetry.
    18

      19         DR. DEAN:        Thank you, John.

      20         I'd now like to review the results of the

trials with you, and the first thing I'm going to show
    21

are demographics.
    22

      23         Again, there was about 120 patients in each

of the arms, in the A study and the B study.
    24                                                      As you

can see from this table the demographics are highly
    25

consistent, and that'll be important later as we
    26

discuss the outcome of the efficacy trial.
    27

      28         And in addition, I'd like to show you what
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
the presenting signs and symptoms were.
      1                                               Study A is the

blue bar, and Study B is the yellow bars, and down on
     2

the bottom here, the percent of patients that present
     3

with these symptoms.
     4

       5         As you can see, again, highly consistent

set of presenting signs and symptoms, again, consistent
      6

with the expectations from out-patient studies in the
     7

literature.
     8

       9         One additional datum of note is the

clinical background of the study population.
    10                                                       These

are patients with prior thrombotic history.
     11                                                   These are

important because they can potentially bias the results
    12

against the product, but these are the approximate
    13

prevalences of disease expected, slightly higher
    14

prevalence of prior history in the B study.
    15

      16         So how did we do?      Here is a summary of the

efficacy results.
    17                      As you can see with the primary

endpoint, it was met in Study A and it was missed in
    18

Study B.
    19

      20         The secondary endpoints, which are a

comparison to institutional venography read, were met
    21

in both the A and the B study.
    22

      23         So herein is the problem, and when the

sponsor, the company, saw this data, the first question
    24

we had was:
    25             what's going on?

      26         And what I want to do now is take you

through our assessment of the data an dour findings and
    27

how we basically addressed the data at this point.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         The first thing we looked at was a

comparison of the agreement rates.
     2                                            This was the

combined data from both studies, and we're comparing
     3

to institutional venography result.
     4                                                Our thought was

that institutional venography, since it was the basis
     5

of the treatment decision on the patients, would be a
     6

good calibrating tool to understand what was happening
     7

in these studies.
     8

       9         Now, when we did this, we surprisingly

found out that the blind read venography data set's
    10

agreement with the institutional venogram
    11

interpretation was 63 percent, far lower than we would
    12

      expected.
have 13

      14         We know that there is some compromise in

the ability to get accuracy in a blind read because you
     15

do not have clinical information, but we did not think
     16

it would be that great.
    17

      18         Interestingly, the new test was agreeing

      institutional venography to a greater extent.
with 19

      20         When we pooled patients across both the A

and the B study, this difference was statistically
    21

significant.
    22

      23         This is the data set that let us know

exactly where the problem was and exactly what it was.
    24

These are the percent of patients that were interpreted
    25

as positive by the blind read venographers.
    26                                                       Here are

the individual readers down here.
     27                                       This is the majority
read, which could consist of either unanimous or two
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
to one, and by comparison is shown the institutional
     1

read over here.
     2

       3         Now, this is based on presenting signs and

symptoms and the type of population that was entered
     4

into the study.
     5                   The literature shows with multiple

references that you would expect about a 40 percent
     6

positive rate of disease within these types of
     7

patients.
     8

       9         You can see the institution calibrates

      with that, though it's slightly higher in the B
well 10

study, and this might be the effect of prior history,
    11

but noticeable is that A study is reasonably consistent
     12

      that, one reader slightly higher at 56 percent.
with 13

      14         But if you go to the B study herein is the

problem, is you have two readers that say that 94
    15

percent of the venograms were positive, and one say that
    16

83 percent of the venograms were positive.
    17

      18         When these two are taken into account to

determine the majority read, you get an interpretation
    19

      82
that 20 percent of the cases in that study were

positive, which is clearly wrong, and that was the
    21

problem.
    22

      23         So how do you address a problem like that?

Well, one analogy is, you know, you have serum samples
    24

and you're doing a clinical study and they're frozen
    25

and you send them out to the core lab and the core lab
     26

      back and it came back with funny numbers that you
came 27
didn't expect.
    28                 What do you do?       You go back and you
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
find a core lab that's well calibrated and you resubmit
      1

the frozen samples.
     2

       3         And what we did is we looked around the

literature, checked our network, and determined that
     4

the institution most likely to provide the gold
     5

standard was Hamilton, but before I do that, I just want
     6

to lead you through one little exercise to exemplify
     7

the problem.
     8

       9         If you took, going back here, this reader

here, Reader 1 in the B study and Reader 3 in the B study
    10

and asked them to compare themselves against each
    11

other, one is the gold standard comparing itself to the
    12

other as a new test.
    13                            You would come up with this

result, that there was a 63 percent agreement.
    14                                                     This

would not have met the confidence interval in our study,
    15

and it was not statistically significant.
    16

      17         So as I alluded to, what we did was we

selected the Hamilton Thrombosis Research Center with
    18

Dr. Jeff Ginsberg and Dr. Jack Hirsh to conduct a blind
     19

read. We were driven here by trying to find out truth
    20

because obviously we didn't have truth in the B study,
    21

as you can see.
    22

      23         These are the credentials of the Hamilton

Research Center.
    24                    It is a center that does a heavy

amount of investigation in venography and venography
    25

related studies.
    26                    Their venography reading criteria

      been validated in treatment outcome studies, and
have 27
     has
that 28 been reported, and their reading criteria also
                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
has been applied by this group in pivotal studies for
     1

FDA approved products, mostly recently Lovenox and
     2

Normiflo.
     3          These are low molecular weight heparin

products that are used for the treatment of DVT.
     4                                                     So

you can see why the institution was used for that
     5

purpose.
     6

       7         I'd now like to ask Dr. Jeff Ginsberg at

Hamilton to come up and comment on how the study was
     8

performed, the blind read, how Hamilton conducts reads,
     9

and what some of the problems can be if you don't have
     10

a standardized set of reading criteria, and what that
    11

can mean for the interpretation of venograms.
    12

      13         DR. GINSBERG:      Thank you, Dr. Dean.

      14         I suppose my task here is twofold.    One is

to convince you that not only are Canadians pretty good
     15

hockey players, but we also know how to interpret the
    16

venograms.
    17

      18         I'm a senior scientist at the Hamilton

Civic Hospital's Research Center and am in charge of
    19

clinical trials of venous thrombosis and have been for
    20

the last six to eight years, and over the last ten to
    21

15 years, we have been adjudicating venograms using a
    22

standardized technique for a variety of different
    23

treatment and prophylaxis studies, and as such, we
    24

continue to have and have had experience adjudicating
    25

anywhere between about three to 800 venograms per year.
    26

So we do have a fair bit of experience with it.
    27

      28         As was mentioned before, in the United
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
States I think what's happening is that the use of
     1

venography is falling off dramatically, and the routine
     2

use of venous ultrasonography has really replaced
     3

contrast venography as the usual test for the diagnosis
     4

of venous thrombosis, and that has really two effects.
     5

       6         One is that because there's only a minimum

number of venograms that's done in each institution,
     7

the institutions that perform these tasks are really
     8

losing some of their skills in the ability to adequately
     9

perform the test.
    10

      11         And secondly, it relates to the

interpretation of the test, and again, in an analogous
    12

fashion, the less number of tests that you do, the worse
    13

you are at interpreting the venograms, and in fact, we
     14

are seeing that across a number of different studies,
    15

particularly in prophylaxis in venous thrombosis that
    16

are run in the United States, and we've been asked to
    17

be the adjudication committee for a number of different
     18

multi-national trials.
    19

      20         Now, how do we adjudicate venograms?

Well, we read them and interpret them and call them into
    21

one of three classifications.
    22                                       In the first

classification, we call the result normal if all of the
    23

proximal veins, in other words, the external iliac,
    24

femoral and popliteal veins are seen and are normal,
    25

and as well if two of the three set of calf veins,
    26

namely, the posterior tibial and peroneal veins, are
    27

      and are normal.
seen 28                           If all of those veins are
                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
visualized and are normal, the contrast venogram is
     1

considered normal.
     2

       3         The other end of the spectrum is a venogram

that's diagnostic of venous thrombosis, and our
     4

criteria are very strict for those, and what we like
     5

to see is a constant or persistent intraluminal filling
      6

defect that's seen in two or more views, and that is
     7

the only criteria that we use for the diagnosis of
     8

venous thrombosis.
     9

      10         The third criteria is one that we call

indeterminate, and that occurs when any of the areas
    11

      I've cited previously is not well visualized or
that 12

not adequately visualized, and this has been a little
    13

bit of a bone of controversy, but if you think about
    14

the pathobiology, what might be accounting for lack of
     15

visualization of a venous segment?
    16                                            And there are

really three potential explanations.
    17

      18         One is that there could be a technical

problem.
    19         In other words, the radiologist injected the

lateral side of the foot and the medial veins are not
    20

being visualized because there's no contrast going up
    21

      side.
that 22

      23         The second possibility is that there could

be old disease, chronic disease that's not recanalised,
     24

and so that segment of venous thrombosis or old venous
     25

disease is not being visualized.
    26

      27         And then there's the third possibility,
and that is that there could be acute venous thrombosis
     28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
that's impeding flow.
     1

       2         In order to be conservative, we call these

venograms indeterminate because in our experience, the
     3

majority of these cases, in fact, do not represent acute
     4

venous thrombosis.
     5                       When we get our radiologist to put

a new 6needle in the center of the foot and reinject,

more often than not, we're able to visualize the veins
     7

completely, and we often seen normal venous flow.
     8

       9         So rather than calling that diagnostic of

venous thrombosis, which I think a lot of our American
    10

colleagues do, we interpret those venograms as being
    11

indeterminate.
    12

      13         Now, with regards to the process that was

carried on when we interpreted the venograms for
    14

Diatide, there are a couple of important things that
    15

are necessary to realize.
    16

      17         First is that we had absolutely no

information about the clinical status of the patients
    18

      we
that 19 were adjudicating, nor of the P280 or apcitide

results.
    20

      21         In addition, we were not informed that

there was any sort of a problem, in other words, that
    22

we were resolving a dispute or that there was any
    23

controversy about the initial interpretation.
    24                                                  All we

     was
knew 25 that we were adjudicating venograms for a study

for clinical use.
    26

      27         And the way it was done at a procedural
level is that two of the three experts that we had would
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
read the venograms simultaneously, and most of the time
      1

once we read the venograms, we would agree, and we would
      2

annotate the results on a mimeographed piece of paper
     3

of our interpretation.
     4

       5         About five percent of the time, there was

disagreement among the two reviewers, and in those
     6

situations what we would do is call in a third expert,
     7

and in that situation the majority would rule.
     8                                                        We

would 9have a discussion, and we would annotate the

results and adjudicate the results based on a majority
    10

decision.
    11

      12         Now, with any sort of study such as this,

the expectation based on literature review is that the
     13

prevalence of venous thrombosis should be somewhere
    14

between about 15 to 40 percent, and results in excess
    15

of that are really inconsistent with published data.
    16

      17         So I'll turn the floor back over to Dr.

Dean.
    18

      19         DR. DEAN:        Thank you, Dr. Ginsberg.

      20         Okay.      So that's what was done.       That's

how they did it, and now let's look at the outcome of
    21

a comparison of AcuTect to the Hamilton blind read.
    22

      23         This is the first data I'd like to show you,

which is a comparison of both of the blind reads to the
    24

institutionally read venograms.
    25                                        What you can see here

immediately is that now the venography test is agreeing
    26

      itself, that is, the blind read Hamilton
with 27
venography to the institutionally read venograms to a
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
much higher degree than the original blind read did with
      1

the institutional read.
     2

       3         This told us that our hypothesis was

consistent, that the gold standard that was applied
     4

here was flawed, and that Hamilton was more consistent
     5

with expectation.
     6

       7         Now, looking at the data further, I want

to show you this triangulation slide.
     8                                                It's like a

double triangulation slide.
     9

      10         Here we're comparing now the Hamilton

blind read again to the institutional reads.
    11                                                      You can

see the high rate of agreement, as you would expect.
    12

You would expect the institutions to be somewhat more
    13

accurate since they do have the clinical information
    14

on the patient and they have additional tests on the
    15

patient.
    16         So it's not inconsistent that you would

expect a slight drop when these were read blindly.
    17

      18         Over here is the problem.            This is the

problematic situation, which was the original blind
    19

      venograms, and you can see the comparison to
read 20

Hamilton right here is very poor.
    21                                          It is somewhat

better over here in the A study, as would have been
    22

reflected from the outcome of the efficacy analysis in
    23

the A study.
    24

      25         For those who are interested in

statistics, I'm not a statistician, but my
    26

statisticians tell me that the kappa between this
    27

comparison to this comparison is .6, and the kappa down
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
here is .2.
     1

       2         So this looks like we have now a calibrated

gold standard, and the question is:
     3                                           how does this now

compare to the AcuTect test?
     4

       5         This slide shows you the data for each

individual reader across both studies.
     6                                                This is the --

these 7are the AcuTect readers compared to the Hamilton

blind 8read data.

       9         I want to clarify a few things here.            We

      an
have 10 agreement rate.           The 60 percent line is going

across here.
    11              It's actually slightly higher than it

should be.
    12

      13         The aggregate really refers to a majority

of the independent readers.
     14                               So this could be all three

readers unanimous or it could be two out of the three
    15

readers to come up with this particular term,
    16

aggregate.
    17

      18         As you can see, both Study A and Study B

now are consistent as you would expect from the
    19

demographics and the presenting signs and symptoms.
    20

You can also see that AcuTect meets the efficacy
    21

criteria across both studies.
    22

      23         A star means that the null hypothesis has

      rejected, and the agent performs above the lower
been 24

confidence limit of 60 percent, and we have our
    25

statisticians here who can explain that in detail if
    26

      be.
need 27
      28         So these are the findings in this study,
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
and I 1now want to show you the summary of all three

venography reads and a comparison of AcuTect to all
     2

three 3venography reads.          That's indicated on this

slide.
     4

       5         Again, when we say "aggregate," we're

talking about a majority of the independent readers,
     6

and read one, of course, was the read that was initiated
      7

prior 8to the FDA's mandated read two, and as you can

see here, the original blind read failed to qualify
     9

AcuTect according to the efficacy criteria.
    10                                                    However,

      the Hamilton blind read and the institutional
both 11

blind read did result in AcuTect meeting the efficacy
    12

criteria in both Study B and Study A.
    13

      14         Now, before I get into the next slide,

which is the subset analysis, I would like to ask Dr.
    15

Alexander Gottschalk to comment on these findings in
    16

relation to his experience with thromboembolism
    17

primarily in the chest, which is the sequela of this
    18

disease, and the problems that were inherent in the
    19

Pioped study, how those were resolved, how those were
    20

addressed, and how they're similar to the situation
    21

      that we're addressing today with AcuTect.
here 22

      23         Dr. Gottschalk apologizes.           He would have

liked to have been here to address you personally, but
    24

as Mr. Piper indicated, he recorded his comments
    25

yesterday, and we have him on video.
    26

      27         So Dr. Gottschalk.
      28         DR. GOTTSCHALK (via videotape):           I was a
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
member of the Pioped Task Force.
     1                                        I was on the steering

committee, but also I was an active member -- the
     2

working group.
     3                 This, of course, worries about

embolism in the thorax and not in the legs, but I got
     4

interested in this receptor when I heard some of --
     5

       6         DR. DEAN:        If you'll bear with me for a

second we can just rewind this.
     7

       8         DR. GOTTSCHALK:         -- Gottschalk.    I'm

professor of radiology at -- good morning.
     9                                                    I'm Alex

Gottschalk.
    10             I'm professor of radiology at Michigan

State University.
    11

      12         I apologize to the Committee for not being

      to
able 13 come before you in person, but as many of you

know, we have a very active visiting professor program
    14

at Michigan State, and I have an eminent radiologist
    15

coming into town this morning, being Monday morning,
    16

and I cannot get back from Washington to take care of
    17

him before testifying, and therefore, I apologize, but
     18

      appear before you on videotape.
will 19

      20         I have had an active interest in venous

thromboembolism for about 30 years, primarily during
    21

my time at Yale when I was a member in the Pioped Task
     22

Force.
    23     I was on the steering committee, but also I was

an active member of the Nuclear Medicine Working Group.
     24

This, of course, worries about embolism in the thorax
    25

and not in the legs, but I got interested in this
    26

receptor when I heard some of the data presented by
    27

Raymond Taillefer, who I believe will be before you this
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
morning, and you will hear him present some of his data.
     1

       2         I was particularly interested because this

tracer shows as a hot spot area acute thromboembolism.
     3

That's a very important concept to me because both in
     4

the legs and both in the thorax the problem of chronic
     5

pulmonary embolism or chronic thromboembolism in the
     6

deep venous system is a difficult one.
     7

       8         I'm sure you are familiar with the fact

that ultrasonographers, as well as venographers, have
     9

trouble with the concept of chronic emboli or clot, and
    10

      result, a tracer that shows the acute clot as a
as a 11

hot spot is really a wonderful concept.
    12

      13         My old chief, Dick Greenspan, used to say

      in
and, 14 fact, presented at one time what he thought

would be a potential tracer for looking at clots in the
    15

lungs.
    16     It turned out not to be effective in anything

besides experimental animals, but he said if we ever
    17

found it, it would be the Holy Grail of imaging for
    18

emboli of all sorts, and I think this is certainly a
    19

potential step toward this with this particular tracer.
    20

      21         Now, like most of you, I read the proposal

principally within the last week in preparation for
    22

coming here to try to see what I could do to help the
    23

Diatide company with their presentation.
    24                                                As I read it,

I was impressed by the fact that the gold standard that
     25

      sought out was about four carats of gold and 20
they 26

carats of lead.
    27

      28         That did not surprise me because I have
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
been through this type of problem with the
     1

angiographers and pulmonary embolism trials.
     2

       3         In the trial of the 1970s, Dick Greenspan,

my old chief, I think one of the finest chest
     4

radiologists in the world and a pioneer in pulmonary
     5

angiography, came across the fact that the
     6

angiographers -- they were three and all good friends
     7

and worked together a lot -- had no trouble making the
     8

diagnosis of pulmonary embolism, but they had a fair
     9

amount of trouble, or at least more trouble than they
    10

would have liked, when they actually came to assessing
    11

the clot size in terms of how many segments were
    12

involved, and so forth.
    13

      14         As a result of this, when the Pioped trial

      into being, Dick knew he had to convene his
came 15

pulmonary angiography group and hold practice
    16

sessions, as well as discussions, of the criteria that
    17

      would use and how they would apply that.
they 18

      19         In particular, they were very rigorous in

terms of the criteria they would accept.
    20                                                You could see

      the clot.
only 21                The clot had to be visible either as

a mass within the vessel or impacted in a vessel such
    22

      the trailing end showed up.
that 23

      24         And what Dick was mostly concerned with was

the fact that in smaller vessels a vascular cut-off
    25

would not be considered emboli unless you could, in
    26

fact, see the trailing edge of the clot, and so he
    27

convened his group, and they practiced discussing this
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
and reading cases, not Pioped cases, but practice
     1

cases, in an effort to achieve some type of consensus.
     2

       3         When I read the data that was presented in

this trial and looked at the discrepancy between the
     4

original venographers of some 35 percent, I said to
     5

myself, "Well, I'll bet I know what happened."
     6

Venography, after all, is not quite as good a gold
     7

standard, if you will, as pulmonary angiography.
     8                                                        The

criteria are more loose.
     9                            You can see a clot.     You can

see a column of contrast cut off.
    10                                          You can use

collateral vessels, and so on, and I'll say my guess
    11

is that none of this group ever talked to each other,
    12

and probably one of them considered everything positive
     13

      clot, and another one considered, "Well, I'm going
to be14

      rigorous," and probably accept only visualization
to be15

of clots or certainly more difficult criteria for
    16

calling positivity, and as a result, they varied all
    17

      the map.
over 18

      19         I think it is to Diatide's credit that they

spotted a real problem fairly quickly, and as the
    20

literature points out, a series of DVT patients that
    21

     symptoms should really have only about a 40 percent
have 22

incidence of positive clots, and here they were running
    23

      80
with 24 percent incidence, and something didn't ring

right, and what didn't ring right was the fact that they
    25

had lead in the gold standard.
    26

      27         Now, I think it's fair to say, gee, should
      have been able to spot this ahead of time.
they 28                                                       Why
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
didn't they figure that out?
     1                                 Why didn't their advisory

group 2tell them that this kind of thing could happen?

       3         I think the answer to that is that you have

to have been there once to have an idea of how much
     4

trouble this can cause you, and I think Justice
     5

Greenspan, who is certainly as bright a person as I
     6

know, 7didn't recognize the fact that he and his two

other 8colleagues would run into some trouble worrying

about 9smaller vessels.           He didn't correct that until

he got to the second trial, which was the Pioped trial.
     10

      11         It isn't unreasonable to assume that the

company and their advisors, not having been there,
    12

would find this to be or not recognize this as a
    13

potential problem.
    14

      15         Now, having recognized it as a problem,

      do
what 16 you do?          Well, I think the answer is you try

to go somewhere where, in fact, people have a rigorous
     17

criteria.
    18          People have been there before.         People know

the difficulties with the technique of venography and
    19

are prepared to use the same criteria to interpret the
     20

venogram.
    21

      22         And they picked out a place that is

renowned for this type of study, and Jack Hirsh is
    23

certainly an international authority on venous
    24

embolism and DVT, and they were very fortunate, I think,
    25

in selecting the group at Hamilton, who are not only
    26

renowned in this ability, but also happen to be the
    27

place that the FDA has used for previous trials.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
       1         In short, they picked out a gold standard

that has something closer to 14 carats than four carats.
      2

It's very difficult to have a 24 carat gold standard.
     3

For example, in Pioped where angiography is considered
     4

to be one of the finest gold standards we have, you might
      5

be interested to know that the same angiographer
     6

reading 72 cases twice, unbeknownst to him, reading
     7

them over again, agreed with himself 89 percent of the
     8

time; 9that the angiographers in Pioped, using this same

concept that was used in this trial, and that is
    10

majority rules, the first two angiographers reading a
    11

      blindly by themselves agreed with the other
case 12

angiographer only 80 percent of the time, and 20 percent
    13

of the time they had to call in a third angiographer
    14

to get a majority rule.
    15

      16         It was possible to get three different

opinions because they used pulmonary embolism present,
    17

absent or indeterminate, and when that happened, they
    18

brought the case before the whole angiography working
    19

group.
    20     That happened about one percent of the time.

So that was not really a problem, and I don't believe
    21

there's any problem like that in this trial.
    22

      23         Therefore, it seems to me that it's

important to recognize, one, that the data from the
    24

trial, that is, the readings of the peptide, were never
    25

changed.
    26         They were the original readings that were

used, and it became clear looking at the data from just
    27

the history -- and it's well known how often DVT should
     28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
appear in a population of folks suspected of having DVT
     1

-- that there was something badly amiss with the
     2

interpretation that was being rendered.
     3

       4         In my view, it's totally explicable on the

fact that none of their readers, original blind
     5

readers, got together to discuss the criteria that they
     6

would 7use or even practice.

       8         That was my assumption, by the way, as I

read it.
     9         I would have bet that that had happened.   I

found out later when I talked to people that that, in
    10

fact, had happened, but I see no reason why that
    11

shouldn't have occurred.
    12

      13         For example, if you take a pulmonary

angiogram and say, "Well, I will use" -- I'm
    14

Angiographer 1 -- "I will use not only visualization
    15

of clot, but I will use perfusion deficit in the lungs
     16

      criteria for pulmonary emboli."
as a 17

      18         And Reader 2 says, "I am going to use not

      visualization of clot, but an occasional view of
only 19

the perfusion deficit if I think it is clearly the lobar
     20

or segmental."
    21

      22         And the third one says, "I will use only

visualization of the clot."
    23

      24         Then the precision of reading the

pulmonary angiogram would fall apart as well, and I
    25

think something like that had to happen with the three
    26

readers in the blind read because nobody ever -- they
    27

did not get together to figure out the criteria that
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
they used.
     1

       2         This was remedied, I believe, by using the

Hamilton Group where, in fact, they have practiced,
     3

where 4they have their criteria carefully established,

and where they did just exactly what I have proposed.
     5

       6         I would suggest that when you look at these

data, 7you simply throw away all the data from the blind

readings because I think the gold standard is flawed.
     8

It is 9loaded with lead, and I would look only at the

Hamilton data, which has no bearing on the readings that
    10

      made on the peptide, and I would consider those
were 11

two together, in which case I think you have a
    12

satisfactory trial.
    13

      14         DR. DEAN:        Thank you, Dr. Gottschalk.

      15         Okay.      I'm now going to, as we previewed

before, show you a subset analysis, and there's a lot
    16

of information on this slide, and let me walk you
    17

through this.
    18

      19         The subset analyses were performed on the

combined studies.
    20                      These are all AcuTect reads down

      versus the Hamilton blind read, which is now our
here 21

14 carat gold standard.
    22                             These are blind read AcuTect

studies, and these are the site's interpretations of
    23

AcuTect.
    24

      25         We have agreement rates, sensitivity, and

specificity.
    26              Blue is all the evaluable patients in the

study.
    27     Pink is the subset where we've removed patients
      a
with 28 prior history, which may confound the results
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
or bias the study against AcuTect, as we have seen, and
     1

red indicates those patients in the narrow window of
     2

within three days of onset of signs and symptoms, which
     3

would 4be the most narrow window we could get a

reasonable amount of patients to compare very close to
     5

the onset of disease.
     6

       7         I'd now like to ask Dr. Ginsberg to come

back up to the podium and comment on this.
     8

       9         DR. GINSBERG:      Thank you, again, Dr. Dean.

      10         I suppose I'd like to wax a little

philosophical, but as an individual who belongs to a
    11

group that sees about 1,200 patients with suspected DVT
    12

per year, the biggest nightmare that I have in medicine,
     13

anyway, relates to the patient with previous disease.
    14

      15         And as was exemplified by slides shown

previously, about a quarter of patients who present
    16

      a
have 17 history of previous disease, and in these

patients, the nightmare that we have is that we really
    18

don't have a gold standard, nor do we even have a very
    19

      test for the diagnosis or exclusion of venous
good 20

thrombosis.
    21

      22         And let me give you some examples.        Of 100

patients who present with suspected recurrent DVT or
    23

who have previous DVT, about 50 percent will develop
    24

      thrombotic syndrome or post phlebitic syndrome.
post 25

Those syndromes can be indistinguishable clinically
    26

      recurrent venous thrombosis.
from 27                                         So when those
patients present, the clinician is left with a
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
conundrum of knowing whether or not this is post
     1

phlebitic syndrome or new thrombosis.
     2

       3         That is compounded with the recent

observation that about 25 percent of patients who have
     4

prior 5thrombosis will develop recurrent thrombosis.

So not only is there a high prevalence of post phlebitic
      6

syndrome, a condition that's clinically
     7

indistinguishable from recurrence, but these patients
     8

are also susceptible to recurrence.
     9

      10         And the final sort of piece to the puzzle

is that once these patients have had venous thrombosis
     11

and their physicians are aware of that diagnosis, they
     12

often have a heightened awareness of the disease itself
    13

and will present themselves more frequently and in a
    14

      timely fashion than patients without previous
more 15

venous thrombosis.
    16

      17         So I think all of those factors underline

the frequency of the problem.
    18

      19         Now, as with any problem with venous

thrombosis, there's a danger in sending patients home
    20

who have the disease because we know that about half
    21

of them will come back with fatal or nonfatal pulmonary
     22

embolism.
    23          So we don't want to miss those who have

disease.
    24

      25         On the other hand, we don't want to over

diagnose because, as was pointed out previously, the
    26

treatment, which invariably is anticoagulant therapy,
    27

is associated with a significant incidence of adverse
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
experiences, about seven percent over one year and then
     1

about 2two percent per year, and then there's the

inconvenience of taking a pill every day and going for
     3

monitoring, and so on, and being labeled as somebody
     4

who is thrombophyliac.
     5

       6         So you don't want to over diagnose, and you

don't 7want to under diagnose.            What do we do currently?

       8         Well, what we do currently is a whole

mishmash of things, and in fast, whereas we have
     9

terrific diagnostic algorithms for virgin patients who
    10

      never had previous venous thrombosis, serial
have 11

ultrasound, serial IPG, venography, all of those are
    12

validated approaches.
    13                            There is no approach that is

currently available to the diagnosis of suspected
    14

recurrent DVT that has been validated by management
    15

trial.
    16

      17         The best test historically has been

radioactive fibrinogen uptake scanning, which has a
    18

number of limitations and is now no longer available.
    19

      20         Similar to apcitide, it's a

radiopharmaceutical and is a physiological test, but
    21

the down side with it is that it's derived from human
    22

products and has potential viral transmission and, as
    23

well, you have to wait 12 to 24 hours before you get
    24

an answer, and you don't want to do that in this disease.
     25

You want to make a diagnosis, get the treatment started,
     26

or send the patient home in a timely fashion.
    27

      28         Apcitide has the potential to overcome
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
both of those limitations.
     1                                It's not a human derived

product, and you can get an answer within two hours at
     2

the most.
     3

       4         Now, what do you and I care about

clinically?
     5             When you look at accuracy data and

agreement data, you say, "Well, that's very nice, but
     6

what we care about is can we make decisions based on
     7

the results of the accuracy indices."
     8

       9         And the most valuable characteristic of a

      for venous thrombosis is its sensitivity because
test 10

the sensitivity has a profound impact on the negative
    11

predictive value.
    12

      13         And our best estimate of sensitivity in

      study, and I think probably the red column
this 14

represents the best estimate because these are patients
    15

who presented within days of onset of symptoms, and keep
     16

in mind this is probably somewhat of a conservative
    17

estimate, in other words an under estimate of true
    18

sensitivity of apcitide.
    19

      20         This estimate of around 85 percent

sensitivity, and this includes both calf DVT and
    21

proximal DVT, is very consistent with tests, such as
    22

venous ultrasonography, which as was previously shown
    23

has a sensitivity of around 80 to 90 percent for the
    24

combination of calf and proximal DVT, and is certainly
    25

favorable when compared with impedance
    26

plethysmography, which has a sensitivity that's even
    27

lower than that, probably in the 75 to 80 percent range,
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
when calf DVT is pooled together with proximal vein
     1

thrombosis.
     2

       3         So this sensitivity is in the range of

something that I would consider to be extremely useful,
     4

particularly when we're so desperate in patients with
     5

previous disease, and we need all of the information
     6

that we can get.
     7

       8         So if you gave me this test tomorrow with

these 9accuracy indices, I would be happy to use it and

say this is probably as good a test as we've got in
    10

recurrent disease, and I may use it alone, but more
    11

likely I would use it in conjunction with other
    12

information, pretest probability, venous
    13

ultrasonography, and perhaps other tests that are
    14

available to me.
    15

      16         And that's what we're left with in patients

     previous disease.
with 17                           We often make a decision based

      a
upon 18 number of different test results.

      19         Finally, a quick comment about the

specificity.
    20              Again, I think what this says is that the

specificity is around 70 percent, which doesn't provide
    21

us with a high enough positive predictive value to be
    22

diagnostic of venous thrombosis when the test is
    23

abnormal, but I think the important message is that the
    24

prevalence of a normal test is going to be high enough
    25

to make the test useful.
    26

      27         So in my opinion, if I was to have this test
tomorrow, what I would say is I would take patients with
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
previous disease, do the test, and if the test result
     1

is negative, I would be reasonably comfortable sending
     2

the patient home without anticoagulant therapy.
     3

       4         I turn the floor back over to Dr. Dean.

       5         DR. DEAN:        Thank you, Dr. Ginsberg.

       6         Now, one of the questions that was brought

up and was alluded to by Dr. Sostman was in regard to
     7

radiolabeled platelets.
     8                             Radiolabeled platelets

performed well, except in cases where the patient was
     9

undergoing anticoagulation.
    10                                   So the question is:    how

     AcuTect perform in the presence of anticoagulants?
does 11

      12         In this particular data chart, the

agreement rate was with the institutionally read
    13

venogram, and as you can see here, the data are
    14

consistent with there being no effective
    15

anticoagulants as indicated on the agreement rate of
    16

AcuTect.
    17

      18         Now, one of the things we would like to

address because I believe it will be the subject of
    19

discussion later is the risk of potential bias in,
    20

quote, unquote, post hoc analysis.
    21

      22         There are three points we'd like to make

here. One is that this is a methodological problem.
    23

     was
This 24 a search for truth.             We thought we had it.    It

was obvious we didn't, and we had to find it.
    25

      26         So one way or the other, this data wasn't

going to be useful until we found truth.
    27

      28         The second thing I want to bring your
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
attention to is that you saw that Hamilton produces the
     1

best measure of truth, and they were blinded to the
     2

clinical end AcuTect results.
     3

       4         In addition, as mentioned by Dr.

Gottschalk, the AcuTect images were not the subject of
     5

a retest.
     6          That's like you've collected the clinical

test sample, and the clinical test sample was collected
      7

according to protocol.
     8                            So the integrity of that is

maintained.
     9

      10         And the last point is that, of course, the

prevalence of the disease is consistent with the
    11

published results.
    12

      13         So in summary, what you've seen today is

      the blind read venography, as evidenced by Study
that 14

B, was flawed by an unexpectedly high positivity.
    15

Hamilton blind read validates the consistency of the
    16

study populations as is expected from the demographics
    17

in the presenting signs and symptoms, and the
    18

performance of AcuTect.
    19

      20         Based on the Hamilton blind read, AcuTect

would meet the efficacy criteria, and the Hamilton
    21

blind read, importantly, is a treatment validated
    22

reading criteria that has been used in pivotal studies
    23

for the FDA approval of Lovenox and Normiflo, again,
    24

products for the treatment of DVT.
    25

      26         So you can see how we would conclude that

AcuTect is safe and effective for the diagnosis of acute
    27

      vein thrombosis, venous thrombosis, and we would
deep 28
                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
ask that you consider and recommend approval for this
     1

indication.
     2

       3         Okay.      I would now like to introduce Dr.

Raymond Taillefer, who will present his findings in his
     4

clinical study with the agent.
     5                                    Dr. Taillefer has done

over 40 patients with AcuTect and will comment on the
     6

performance of AcuTect in his hands.
     7

       8         DR. TAILLEFER:       Thank you.

       9         Good morning.      Since my time is already

up, I'll be very brief.
    10

      11         (Laughter.0

      12         DR. TAILLEFER:       I'm Raymond Taillefer.

I'm professor of nuclear medicine and radiology, and
    13

I'm also the Director of Research and Nuclear Medicine
     14

at the Hospital Hotel Dieu de Montreal, and I, as
    15

pointed out by my colleague, I was involved as an active
    16

clinical investigator in that project, and I would like
    17

to share with you some data that we have and some images.
     18

      19         Before I will show you a few images,

detection of acute venous thrombosis with AcuTect, I
    20

would like to show you some data on the biodistribution
    21

of this compound which is relevant to what we can
    22

discuss as far as the imaging is concerned.
    23

      24         So the first thing that we should know

about this product is that the major pathway of
    25

excretion is through the kidneys, and in fact, close
    26

to 90 percent of the injected dose will be excreted
    27

through the kidneys over 24 hours after injection, and
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
about 150 percent after two hours following the

administration.
     2

       3         The hepatobiliary excretion will be

approximately six to ten percent over 24 hours, and
     4

obviously the organs which will show the excretion of
     5

this tracer will be significantly seen and very rapidly
      6

seen after the injection.
     7

       8         For those of you who are interested in

radiation dosimetry, the main effective dose
     9

equivalent is 0.034 grams per millicurie, which is
    10

basically similar to what we have in standard clinical
    11

nuclear medicine for different agents and different
    12

regular tracers that we use in daily practice.
    13

      14         The maximum organ absorbed dose will be the

urinary bladder wall with .22 rads per millicurie, and
    15

      is
this 16 why we can inject up to 25 millicuries per

patient.
    17

      18         And the estimated biological half-life of

AcuTect is 1.9 hours with a mean half-life in the plasma
    19

of approximately one to 1.7 hours.
    20

      21         Now I would like to show you whole body

distribution data performed in normal volunteers, and
    22

as you can see, these images are whole body images
    23

performed in the anterior view, posterior view, ten
    24

minutes after the injection of AcuTect, 60 minutes, and
    25

      four hours after the injection.
then 26

      27         So very soon, very early after the
injection of AcuTect, intravenous injection, we can see
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
that we have an increased uptake in the liver and also
     1

the kidneys, and in posterior view you can see the
     2

increased kidneys' activity and also urethral
     3

activity, and of course, bladder, urinary bladder
     4

increased uptake.
     5                      So this will be seen ten minutes

after 6the injection.

       7         Then 60 minutes later, you will start

seeing a slightly decreased liver activity and
     8

increased gall bladder retention and excretion and the
     9

     thing for the kidneys and bladders, which are very
same 10

      seen on the 60 minute images.
well 11

      12         Of course, because of the half-life in the

blood, which is approximately one to 1.5, 1.7 hours,
    13

we will see cardiac chambers.
    14                                    Here's the blood

activity in the cardiac area here which is normal, and
    15

      this uptake will slightly decrease and then over
then 16

at four hours after the injection this activity has
    17

significantly decreased, but we still have some gall
    18

bladder activity and also some kidney and urinary
    19

bladder uptake.
    20

      21         Now, if we pay attention the lower limbs

because this is the region of interest for us in
    22

clinical practice, then I did the same thing.
    23                                                   So we

      images performed at ten minutes after the
have 24

injection, 60 minutes, and two hours after the
    25

injection.
    26           We have the anterior thigh, anterior

knees, and anterior calf view, and the same thing for
    27

posterior pelvis, posterior knees, and poster calves.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         And as you can see we have, at ten minutes

after 2the injection, we still have some activity in the

blood 3pool, again, because of the half-life, and in

patients when we pay attention to the posterior knees
     4

area, 5we can see that there is a slight increased

uptake, linear uptake, responding to the popliteal
     6

vein, 7and this is a normal finding when we have

symmetrical uptake, and this is because the popliteal
     8

veins 9are more superficial, and this is why we can

clearly see them on the posterior view.
    10

      11         Also, in some patients we might see the

distal part of the popliteal vein and in some patients
    12

      we
also 13 can see the proximal part of the tibial and

peroneal veins.
    14

      15         At 60 minutes this activity in the

popliteal area will slightly decrease over time.
    16                                                       So

if we draw a sketch, a scheme from the activity from
    17

     popliteal region over time, you will see a decrease
this 18

     time of the activity, but in many patients we will
over 19

see a slightly increased uptake in the joint, which
    20

corresponds to a synovial uptake that we see with all
    21

types of antibodies and also different peptides, which
    22

is normal findings, and we must not confuse that with
    23

superficial or deep vein thrombosis.
    24

      25         But, again, as you can see, this activity

slightly decreases over time, and then at two hours
    26

after the injection we don't see anymore significant
    27

increased uptake in the popliteal region.
    28                                                So this is
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
a normal finding.
     1                      So we don't see any significant

increased uptake.
     2

       3         We always compare both limbs to each other,

and there is no activity also in the thighs.
     4

       5         Another negative case, so again it's very

important to always compare each side, and we do it
     6

systematically, both anterior and posterior views, in
     7

order 8to make sure that we are comparing exactly the

same segments of the veins.
     9

      10         So this is a case, an obviously positive

      in
case 11 a patient who had been treated with

anticoagulant therapy, with heparin for two days before
    12

the patient was enrolled in that study, and as you can
     13

see here, although the patient was under anticoagulant
     14

therapy, we can clearly see on these anterior views and
    15

posterior views performed 60 minutes after the
    16

injection, we can clearly see this increased uptake,
    17

which is quite linear, relatively intense, and
    18

corresponds to the pathway of deep vein.
    19

      20         In this case, these veins were the tibial

ones. So posterior and anterior tibial veins, which
    21

      a
show 22 very significantly increased uptake, and it's

      important, again, to compare to the other limb,
very 23

but also to make sure that this uptake corresponds to
    24

a deep vein and not to a superficial vein.
    25                                                In this

case, this is quite obvious, and again, this patient
    26

was under anticoagulant therapy for two days before
    27

getting demonstration with AcuTect.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         Another patient with also a deep vein

thrombosis involving in this case the right leg, which
     2

is well seen on the anterior view.
     3                                            In this case, the

images have been obtained tow hours after the
     4

injection.
     5           So this is not the deep vein thrombosis.

This corresponds to a urinary catheter.
      6                                                So this is why

it's very, very hard.
     7

       8         But then we don't see any significantly

increased uptake in the thigh, neither in the knees.
     9

      is
This 10 normal uptake in the knee joint, but here we

      this increased uptake corresponding to the deep
have 11

      thrombosis, which is very well giving aid to this
vein 12

patient.
    13

      14         Now, in some patients we can also see both

superficial and deep vein thrombosis at the same time,
    15

and this is an example.
     16                           Again, the same pattern:       ten

minute, 60 minute, and two hours after the injection,
    17

and if you pay attention to this image here, this is
    18

an image of the posterior calf obtained two hours after
     19

the injection.
    20                 We can see that there is a slight

increased uptake, a linear uptake, which corresponds
    21

      superficial vein, and in this case this patient
to a 22

had also superficial vein thrombosis, plus in the
    23

popliteal region we have this increased uptake
    24

corresponding to a deep vein thrombosis of both calf,
    25

popliteal region, and also the distal part of the right
    26

thigh.
    27

      28         So with this patient we had both
                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
superficial and deep vein thrombosis.
     1

       2         Now, as pointed out by my colleagues

previously, post phlebitic syndrome is a real clinical
     3

problem and a real puzzle in clinical practice, and this
     4

is a case of a patient who was admitted for recurrent
     5

episodes of possibly deep vein thrombosis.
     6                                                This

patient had a prior history of deep vein thrombosis on
     7

the right leg a few years before we did the study, and
     8

this patient was complaining of recurrent symptoms,
     9

especially a slight edema, and we did the study in this
    10

patient at ten minutes, 60 minutes, and two hours,
    11

exactly the same way we did for the previous patients.
    12

      13         And as you can see in this patient on the

anterior view, we have this slightly diffused increase
    14

uptake in the soft tissues on the right extremity that
    15

we can see on the posterior view, but at no time we are
     16

      to
able 17 recognize that there is a linear uptake

corresponding to a deep vein pathway.
    18

      19         So in this case when we have this kind of

slightly increased diffused uptake involving the soft
    20

tissue, we have two options.
    21                                 It can be either related

to venous insufficiency or lymph edema.
    22

      23         So in this case a follow-up study showed

      it
that 24 was not a recurrent episode of DVT, but just

a post phlebitic syndrome with inflammation, and this
    25

patient was treated with anti-inflammatory medication,
    26

but as detected here, we didn't see any significant
    27

signs of deep vein thrombosis, and this patient was not
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
treated for deep vein thrombosis, but just for
     1

inflammatory reaction.
     2

       3         The same thing in another patient with

similar history, but in this case we have similar uptake
     4

in all the segments.
     5                            So we cannot recognize any

increased uptake corresponding to a linear deep vein
     6

thrombosis, and this patient was treated for post
     7

phlebitic syndrome without deep vein thrombosis.
     8

       9         Also it's important as pointed out by Dr.

Ginsberg to detect previous -- not previous -- but acute
    10

      vein thrombosis in patients with post phlebitic
deep 11

syndrome, and this is a case of a patient having prior
    12

history of deep vein thrombosis.
    13                                         The patient came back

and now we can see that there is an increased uptake,
    14

linear uptake, corresponding to a deep vein thrombosis
    15

in the patient with recurrent symptoms of deep vein
    16

thrombosis.
    17

      18         DR. DEAN:         Thank you, Dr. Taillefer.

      19         And for a brief final comment I'd like to

introduce Dr. Michael Bettman, Chief of Cardiovascular
    20

Interventional Radiology at Dartmouth.
    21

      22         Thanks.

      23         DR. BETTMAN:         I appreciate the

opportunity to make some observations.
    24

      25         My involvement in this study has been

essentially nonexistent.
    26                                I have to confess to being

one of the blind readers in Study A, I believe.
    27                                                          I'm
      I
sure 28 was the one who was the most accurate.
                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
       1         (Laughter.)

       2         DR. BETTMAN:      But other than that, I have

had no involvement in this.
     3

       4         I would like to just comment really on the

nature of the disease and on the role of venography and
     5

of other diagnostic methods, and in my mind the
     6

necessity for the advantages of AcuTect.
     7

       8         First of all, as has been pointed out, deep

vein thrombosis and pulmonary emboli are very common
     9

disease entities.
    10                      They occur with great frequency,

and it's very clear from multiple studies that they are
     11

not diagnosable with any degree of accuracy from a
    12

clinical standpoint.
    13

      14         The clinical suspicion does play a very

important role, but it is only that.
    15                                            It is a suspicion

which should generate further tests.
    16

      17         The tests that are available for deep vein

thrombosis have been outlined to you.
    18                                                There have been

various radionuclide studies that have been tried over
    19

the years, none of which has really been entirely
    20

satisfactory, perhaps with the exception of the labeled
    21

fibrinogen studies, which had very high specificity,
    22

relatively low sensitivity -- I'm sorry -- very high
    23

sensitivity, very low or relatively low specificity,
    24

but is no longer available at any rate.
    25

      26         And venography has certainly been used for

a long time, as has ultrasound and impedance
    27

plethysmography and several others.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         What is the role of venography?                Well,

venography really ha fallen out of use with the advent
     2

particularly of ultrasound, and it's somewhat
     3

interesting because the accuracy of ultrasound overall
     4

has been shown in multiple tests to be somewhat
     5

fallible, at least compared to venography.
     6

       7         Nonetheless, ultrasound is noninvasive,

has really no complications other than its lack of
     8

accuracy and, therefore, has been widely utilized.
     9

The big difficulty with ultrasound or the big
    10

difficulties are, first, that it really is not
    11

particularly accurate below the knee, and, secondly,
    12

      it
that 13 is dependent on a degree of expertise.

      14         Venography is a diagnostic modality that

I think has fallen probably for good reasons.
    15                                                            It is

relatively invasive.
    16                            It's relatively expensive to

perform.
    17         It does have a discrete incidence of

complications, of unwanted complications, and it is
    18

      operator dependent.
also 19

      20         And I guess the question that I wanted to

address primarily is why was there the disagreement
    21

between the blind reading of venography, on the one
    22

hand, and, on the other side, the readings at Hamilton
    23

and the reading at the institutions.
    24

      25         The reasons, I think, are based in the

utilization of venography.
    26                                  Venography, as I said, is

dependent on a degree of experience.
    27                                                   It requires
assiduous attention to detail in order to be accurate,
    28

                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
and that means very careful needle placement, very
     1

careful fluoroscopic evaluation as the contrast is
     2

infused, very careful obtaining of films while there's
     3

good contrast filling.
     4

       5         In theory, it's not at all difficult, but

in practice if you don't do it with some frequency, it's
      6

really just not done well, and that, I think, leads to
     7

two problems.
     8

       9         One is in the performance and the other is

in the interpretation.
    10                            I think as these studies were

performed at the site in all likelihood a fair amount
    11

of information was gained from the clinical setting and
     12

      the fluoroscopic observation and not from the
from 13

films.
    14     Because of the lack of great utilization of

venography, I think it's likely that the films obtained
    15

at the different sites were really not entirely
    16

optimal.
    17         That's one side of the equation.

      18         The other side of the equation is why were

the blind readings not more accurate.
    19                                                 Why did they

agree to a greater extent with the readings at the site?
    20

      21         And I think the reasons, again, are related

to experience.
    22                 Since people are not doing venograms

      any great frequency, are not used to techniques,
with 23

and are not used to very careful evaluation, I think
    24

      it's logical to assume that the accuracy would be
that 25

somewhat lower than it would have been a few years ago
    26

      venograms were done with great frequency.
when 27
      28         So in summary, I think that AcuTect is a
                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
diagnostic test that is needed in this day and age for
     1

a common and important disease.
      2                                      I think that there was

clearly a disagreement between the blind readers and
     3

the on-site reading in the Hamilton readers.
     4                                                        I think

that is really clearly understandable and probably
     5

should not be heavily taken into consideration when
     6

considering the safety and efficacy of AcuTect.
     7

       8         Thanks for your attention.

       9         MR. PIPER:        That concludes our formal

presentation.
    10                Sorry for going a little bit beyond our

allotted time, but we'd certainly like to entertain
    11

questions if you have them now.
    12

      13         CHAIRPERSON RAMSEY:           Thank you very much.

      14         I think looking at the program and the time

allotted, I'd like to say let's take a break now, a 15
    15

minute break, and we will then begin again.
    16                                                   Let's see.

     five minutes after.
It's 17                             At 20 minutes after ten with

the question.
    18

      19         So if the committee could please hold their

questions, and we thank you very much for your
    20

presentation.
    21

      22                          (Whereupon, the foregoing

      23                          matter went off the record at

      24                          10:02 a.m. and went back on the

      25                          record at 10:22 a.m.)

      26         CHAIRPERSON RAMSEY:           We'll now be

starting the session for the Committee questions on the
    27

sponsor's presentation.
    28                              I want to thank the sponsor
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
for their presentation this morning and the Committee
     1

members, again, for all being here today.
     2

       3         And are you prepared?

       4         I know there are a few questions.            Dr.

Links, would you like to go first?
     5                                            You had a question

earlier.
     6

       7         DR. LINKS:        I have three related

questions.
     8           All in a way involve contrast venography.

The first is:
      9              what studies, if any, have ever been done

to determine the accuracy of venography in diagnosing
    10

DVT? 11

      12         The second is has venography been used in

the past as a, in quotes, gold standard to assess the
    13

diagnostic performance of any other test, for example,
    14

ultrasound?
    15

      16         And then the third question is:            if

venography is the gold standard in this particular
    17

trial and the indication is for acute venous
    18

thrombosis, what's the evidence that would tie the
    19

indication to the results of the clinical trial,
    20

specifically highlighting the word "acute"?
    21

      22         DR. DEAN:        Okay.    For the first part of

      question I'd like to ask Dr. Ginsberg to comment,
that 23

and that's in regard to the, as I understand it,
    24

validating venography as a true standard.
    25

      26         DR. GINSBERG:        Yeah.      I can actually try

and knock off the first two questions, if that's okay.
     27

      28         With regards to accuracy of venography,
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
that was tested in a prospective management study in
     1

which 2patients with a normal venogram were discharged

home and followed up.
     3                            So they presented with the

suspicion of DVT, had a venogram.
     4                                           If the test result

was normal, then they were followed up for I think it
     5

was six months to a year for the absence of venous
     6

thromboembolic events, and there were about 150
     7

patients who had such findings, and I think one percent
     8

returned with objectively confirmed venous
     9

thromboembolism.
    10

      11         So that supports the negative predictive

value of venography.
    12

      13         With respect to comparing other

noninvasive tests, there's an excellent study that was
    14

      by
done 15 Tom Lensing and published in the New England

Journal in 1989 in which what Dr. Lensing did was he
    16

systematically performed compression ultrasound and
    17

venography on all patients who presented with the
    18

suspicion of DVT and showed that the sensitivity of
    19

ultrasound for proximal vein thrombosis was over 90
    20

percent and of calf DVT was less than 50 percent, and
    21

      the specificity was around, I think, in the 96
that 22

percent range.
    23

      24         So certainly in that study, which was done

     single center, and incidentally, the Dutch group
in a 25

      very similar criteria to the ones that we used,
used 26

     we
that 27 have used and that we use in this study, I think
is the best evidence supporting the test.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         There's also a similar study done with IPG

and leg scanning comparing it with venography as a
     2

reference standard.
     3                            It was done very similarly and

validated that as a substitute for venography.
     4

       5         DR. LINKS:         Don't go away because as long

as you're up there, before we get to the other question
     6

I have a clarification on something you presented.
     7

       8         It was stated that there were potential

problems with both the performance and the
     9

interpretation of the venogram, and obviously you all
    10

could only address a reinterpretation, not a
    11

reperformance, so to speak.
    12

      13         What did you do to assess the technical

quality of the venogram, and did you throw out any
    14

patients because you said the venogram was not
    15

technically acceptable?
    16

      17         DR. GINSBERG:          Yes, we did.      So if the

venogram was unacceptable, for example, if the -- not
    18

     if
only 19 the veins were not visualized, but if there was

a very hazy film and we weren't able to get clear
    20

visualization of important areas of the deep veins, we
    21

did not -- we considered those inadequate or
    22

indeterminate.
    23

      24         DR. LINKS:          And were all indeterminate

reads for whatever reason thrown out of the study in
    25

the clinical results based on Hamilton?
    26

      27         DR. GINSBERG:          My understanding -- well,
sorry, Dean.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.       Fax: 202/797-2525
       1         DR. DEAN:        I should step in here because

Hamilton only knew -- Hamilton only saw films, and they
     2

recorded things on a piece of paper.
     3                                             So he doesn't know

what happened to the data.
     4

       5         (Laughter.)

       6         DR. DEAN:        So I'd like to ask a

statistician who was responsible for that to address
     7

that point.
     8

       9         DR. MADSEN:       Whatever was used as truth in

the Hamilton read case, yes, at all the regions, if the
     10

whole set of scans were determined to be indeterminate,
    11

      case was one of our unevaluable cases, but if an
that 12

individual region was indeterminate, that region
    13

wasn't included in the assessment.
    14

      15         DR. KONSTAM:        Sorry.     What was the number

of unevaluable cases at the end?
    16

      17         DR. MADSEN:       Altogether for each study we

had -- on the basis of indeterminate reads, we probably
     18

had ten or 11, I guess.
    19

      20         How many?        Nine, nine altogether.

      21         MR. MADOO:       Could you please provide your

name, please?
    22

      23         DR. MADSEN:        Sorry.     Kathleen Madsen.

      24         DR. LINKS:        So back to acute versus

chronic.
    25

      26         DR. DEAN:        Acute versus chronic.       I'm

going to as Dr. Sostman to comment on the data as it
    27

relates to the response.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         DR. SOSTMAN:        Well, first of all, I'm not

100 percent sure I understood the question.
     2                                                      So I'll

try to give you my answer as I interpret it.
     3

       4         In the first place, it's my understanding

from Dr. Ginsberg's presentation that the Hamilton
     5

readers used their criteria for acute DVT as a positive
     6

test, 7and he's shaking his head yes.

       8         Secondly, not actually related to the

venogram, but to the results of the apcitide study, if
     9

you remember that slide where they looked at the
    10

subgroup analysis and as the subgroup went to a more
    11

acute clinical presentation, that is, less than three
    12

      from the onset of signs and symptoms, the
days 13

sensitivity of the test went up, and that, I think, is
    14

quite consistent with the rationale for the test, which
    15

is binding of the agent to activated platelet
    16

receptors.
    17

      18         So that to me is what makes this

specifically an acute thrombus agent.
    19

      20         Does that answer your question?

      21         DR. LINKS:        Thank you.

      22         DR. DEAN:        Thank you, Dr. Sostman.

      23         CHAIRPERSON RAMSEY:           Questions?

      24         Could you just state your name and then ask

      question?
your 25

      26         DR. D'AGOSTINO:         Ralph D'Agostino.

      27         I have some questions go to the statistical
issues, and I'd like to ask, first of all, so that I
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
can understand the context of the studies where the 60
     1

percent comes from, and in your presentation you said
     2

75 percent, but the statistical analysis, if I
     3

understand it correctly, would have been really testing
     4

the 60 percent agreement rate.
     5

       6         Could you clarify why the 60 percent and

      really a one tailed test for the 60 percent rate?
is it 7

       8         DR. DEAN:        I'm going to ask my

statistician to come up and respond to that again,
     9

Kathleen Madsen.
    10

      11         DR. MADSEN:       I knew this was going to come

     You know, the 75 percent -- well, in the design
up. 12

piece studies, we were required to justify that the
    13

studies were being designed with adequate statistical
    14

power, 80 percent, and with adequate numbers of
    15

patients to establish this expected rate of 75 percent.
    16

      17         The approach we chose was to take a

confidence interval approach for establishing that
    18

studies were adequate in terms of power and sample size
    19

for establishing that the true rate was not less than
    20

75 percent by more than 15 percent.
    21                                                  That was the

confidence which translates to a 60 percent lower bound
    22

on the confidence interval.
    23

      24         One sided because we're not concerned that

it would be different from 75 percent in a positive way;
     25

on in the negative side.
    26                              So we saw it as a one sided

hypothesis test.
    27

      28         DR. D'AGOSTINO:         But it is then basically
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
a test of agreement equals 60 percent versus agreement
     1

greater than 60 percent that we're actually looking at,
     2

not 75?
     3

       4         DR. MADSEN:      Well, I think what we have in

this study design is -- yes, we're looking to estimate
     5

agreement rate, and what we're trying to establish is
     6

that the lower bound of the confidence interval, one
     7

side confidence interval, for that agreement rate is
     8

not less than 60 percent.
     9                              So it's tied to the lower

bound.
    10     It's not tied to the point estimate itself.

      11         DR. D'AGOSTINO:       I have two other

questions.
    12           One, in terms of interpreting the

statistics that we have before us, if I heard the
    13

discussion correctly or the presentation correctly,
    14

the analogy was if you find your assays are wrong, you
     15

go and get a better assay or you go and get a correct
    16

one so then you can believe your data.
    17                                                So that's one

way of looking at the data, that somehow or other we
    18

didn't have the correct answer to begin with.
    19

      20         Another way of looking at the data is that

I'm in the situation often where you run your
    21

statistical hypothesis test,
    22                                  you run your confidence

intervals, and you find out that you don't meet the
    23

criteria, that your study is not positive, and then you
    24

      "Gee, I wonder why.
say, 25                            The blind's broken.       I've

     my
done 26 primary analysis," and then you go and you say,

"My God, some of the individuals who were in the
    27

analysis really shouldn't belong there.
    28                                                 They were
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
protocol violators," and then I redo the analysis, and,
     1

lo and behold, the analysis is now positive.
     2

       3         And how do we as a group interpret the fact

that you did the primary analysis, it didn't work for
     4

you, and you did a secondary analysis and it did work?
     5

What are the levels of significance?
     6                                                  What is the

interpretation from a statistics point of view?
     7                                                           Leave

the clinical questions aside for the moment and let
     8

others address it.
     9

      10         But what is the statistics?               How do I

believe the second set of analyses?
    11

      12         DR. DEAN:        We're going to have another one

of our consultant statisticians, John Balser, address
     13

that.
    14

      15         DR. BALSER:        Is this working?

      16         Okay.      If I understand the question

correctly, the issue has essentially to do with what
    17

      of
kind 18 adjustments might one want to apply in a case

where we've done an additional analysis.
    19

      20         Typically that kind of adjustment is

required if, in fact, you've got more than one
    21

statistical test which is valid or potentially cases
    22

where you're looking at multiple endpoints and, you
    23

know, those kinds of situations.
    24

      25         DR. D'AGOSTINO:         I'm not asking that.

I'm asking:
    26             my analysis didn't work.              I redefine --

      say I'm redefining my data set, and now my analysis
I can27
      work.
does 28          How do I look at that?
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         DR. BALSER:      Yeah, I understand what

you're saying.
     2

       3         DR. D'AGOSTINO:       It's not multiple

testing.
     4

       5         DR. BALSER:      You're talking about

excluding certain data points from your analysis.
     6

       7         DR. D'AGOSTINO:       No, I'm talking about my

original analysis didn't work.
     8                                    I redefine my data, and

now it does work.
     9                     How do I -- I gave my -- the example

I gave was as an example of protocol violators, but it's
     10

the second look at the data, and that's the question
    11

I'm really asking.
    12

      13         How do I look at that in a statistics point

of view?
     14        What we'd like to see is you've done a study

and you get replication or you have two studies that
    15

replicate each other.
    16

      17         Here we have one study that was positive.

We have another study that was negative, but then it
    18

becomes positive if I redefine -- basically redefine
    19

my endpoint.
    20

      21         DR. BALSER:      We're not redefining our

endpoint.
    22          Our endpoint is still AcuTect as the test.

We are redefining, if you will, the gold standard.
     23                                                      The

      standard was an inappropriate standard to be
gold 24

using.
    25     It invalidates the test entirely, and that's

really the point.
    26

      27         We're not doing multiple testing.         We're
not saying that we're doing another test on the same
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
data and somehow, you know --
     1

       2         DR. D'AGOSTINO:       Well, but your gold

standard was by the blinded readers and now you have
     3

Hamilton reading.
     4                     We'll pick that up later on with the

FDA and maybe come back.
     5

       6         But let me ask one other question.       When

you give the analysis, you do it on the aggregate reads
     7

from each of the blind readers as opposed to individual
      8

readers.
     9         Was the protocol said to do the majority?

      10         DR. BALSER:      Are you talking about the

aggregate for the AcuTect readers or are you talking
    11

about the majority blind read?
    12

      13         DR. D'AGOSTINO:       In the analysis that was

presented looking at the AcuTect versus Hamilton versus
    14

the CV, it was only for the aggregate that I seem to
    15

recall you presenting as opposed to the individual
    16

readers.
    17

      18         DR. BALSER:      No, we actually did present

the individual reader results for AcuTect, each of the
     19

individual readers, as well as the aggregate.
    20

      21         DR. D'AGOSTINO:       What should I be looking

at? 22

      23         DR. BALSER:      Well, prospectively in the

protocol each individual reader was, in fact, to be
    24

looked at.
    25           I believe that aggregate came up somewhat

later, possibly in discussions as to how to simplify
    26

the presentation.
    27

      28         DR. D'AGOSTINO:       So I should look at the
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
individual readers.
     1                            So it's three out of six in the

first 2study and six out of six in the second.

       3         DR. BALSER:          For Hamilton I think it was

somewhat better than that, but essentially, yes, the
     4

individual readers should be looked at.
     5

       6         DR. D'AGOSTINO:            Thank you.

       7         CHAIRPERSON RAMSEY:              Other questions?

       8         Dr. Ponto.

       9         DR. PONTO:          I have two questions.         The

first is we've talked about we have an imperfect gold
    10

standard here, and that maybe our best gold standard
    11

would be outcome.
    12                      Did you look at outcome in any of

these patients?
    13

      14         And my reading of the documents, the

majority of these patients were treated as if they had
    15

      correct?
DVT, 16

      17         DR. DEAN:         Okay.     I'm going to refer that

to Dr. Nicodemus, who's our clinical operations head.
    18

      19         DR. NICODEMUS:            Yes.    Two questions.

The endpoint for the study was, in fact, the results
    20

of the comparison.
    21                       So an outcomes study was not

conducted with these patients formally.
    22

      23         The second question -- actually remind me

of the second question.
    24

      25         DR. PONTO:          The majority of the patients

      actually treated like they had DVT.
were 26

      27         DR. NICODEMUS:            Right.     The actual
patient treatment data reflects the treatment.
    28                                                             About
                              S A G CORP.
202/797-2525   Washington, D.C.       Fax: 202/797-2525
70 percent of the patients received some form of
     1

anticoagulation in the study.
     2                                     That reflects patients

receiving anticoagulation at the time of their
     3

diagnostic evaluation.
     4                             In some circumstances

patients were actually anticoagulated during the
     5

work-up as part of the rule out process.
     6

       7         The actual data for the number of patients

who received longstanding anticoagulation is not part
     8

of the analysis, but would be less than 70 percent.
      9                                                        It

would be probably closer to 50 percent.
    10

      11         DR. PONTO:       I have a second question.   The

FDA provided us with the individual AcuTect readings,
    12

and we've talked a lot about the agreement rates between
     13

the Hamilton read, the blind read, and all of that.
    14

      about the agreement between the AcuTect reads?
What 15

Can you give us some insight into that?
    16

      17         DR. DEAN:        Okay.    I'm going to ask one of

my statisticians to respond to that.
    18

      19         DR. MADSEN:       In terms of kappa statistics

among individual blind readers, they were pretty low.
    20

So if you just compared pairs of readers, you would see
     21

the kappa statistics that were, you know, generally
    22

      than .4, and that's considered a pretty low kappa
less 23

statistic.
    24

      25         So -- but there was -- we also provided a

measure of unanimity among readers, and I think you saw
    26

on the order of, you know, 60, 60 percent of the time
    27

      were unanimous in their readings of the images.
they 28
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         So the kappa statistic was low.

       2         CHAIRPERSON RAMSEY:           Dr. Konstam.

       3         DR. KONSTAM:        I'd like to ask three

questions.
     4           The first is of any of the sponsor's

speakers.
     5

       6         In acute venous thrombus, I assume we can

get to a point where the vein is actually occluded or
     7

nearly occluded or at least the flow is diminished, and
     8

that can be acute.
     9

      10         And I wonder what you feel that might --

how that might impact on the diagnostic ability of
    11

AcuTect vis-a-vis delivery to the thrombus or the
    12

entire thrombus.
    13                    Is that a problem?

      14         DR. DEAN:        That's a good question.

Having worked with both antibodies and small peptides,
    15

we have seen at least in the clinical images some
    16

differences.
    17              Whereas antibodies would often light up

the tip of a thrombus, these seem to diffuse right into
     18

the matrix quite readily so that we see the entire line
     19

light up, as you've seen the images.
    20

      21         So, you know, unless you strip out the

veins and actually look at a cross-section and
    22

everything, you can't get a definitive answer, but --
    23

      24         DR. KONSTAM:        Do you have any information

      your animal studies to shed light on this
from 25

particular question of what happens when the vein
    26

actually reaches a point of near occlusion?
    27

      28         DR. DEAN:        Sure.    Let me ask Dr.
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
Lister-James if he can respond to that.
     1

       2         DR. LISTER-JAMES:           In the animal studies

I don't think any of the animals that we studied had
     3

totally occlusive thrombi, and so we weren't able to
     4

address that, but I think, as Dr. Dean mentioned, we
     5

have a small, highly diffusible tracer, and even if that
      6

weren't to be the case, then one would be able to pick
     7

up the ends of the thrombus.
     8

       9         DR. KONSTAM:        Okay.     My second question

      gets back to this 60 percent or 75 percent figure
just 10

      Dr. D'Agostino was asking about.
that 11                                                 I mean, where

      that come from, either number, 60 percent or 70
does 12

percent, in terms of agreement?
    13                                         Is there some

precedent to that type of analysis and that level of
    14

accuracy as a gold standard?
    15

      16         DR. DEAN:        I'm going to defer to one of my

      members to address that.
team 17

      18         DR. KONSTAM:        I mean, 60 percent would

      ten percent better than a coin.
mean 19

      20         DR. DEAN:        Right, right.         One of the

things you have to -- and I'll allow Dr. Nicodemus to
    21

expound on this -- but one of the things you have to
    22

understand is there are going to be successive factors
    23

     lower the potential agreement rate between the two
that 24

tests, as you've seen.
    25                            So we felt that that, based on

prior history with this tracer, that 75 percent was a
    26

      target.
good 27
      28         Dr. Nicodemus, do you want to comment
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
further on that?
     1

       2         DR. NICODEMUS:        Right.      We have a number

of references from the literature, a Lensing paper, and
      3

others that we can provide you and are in the briefing
     4

documents in which the agreement rates between blind
     5

read venographers -- and this was the basis of this
     6

calculation.
     7              If you take venography, conduct the

reads 8blindly, and look at the agreement rates between

venographers, the agreement rates in those reference
     9

papers is on the order of about 75 percent.
    10

      11         That's similar with the agreement rate

      was seen, for example, between the institutional
that 12

      and the Hamilton read, 75 percent.
read 13                                                   So that was

the basis for that intended endpoint, and then, again,
     14

as Dr. Madsen commented, when one's looking for a 75
    15

percent agreement rate, one has to keep in mind the
    16

confidence interval that one gets, and it's the lower
    17

limit of the confidence interval that reflects the 60
    18

percent.
    19

      20         So really 70 percent, 75 percent agreement

would be what one would expect in the circumstance of
    21

a blind read exercise.
    22                            This is different from the

clinical exercise.
    23                       This is the constraints of reading

the information without information.
    24                                                 That was the

basis.
    25

      26         DR. KONSTAM:       Well, I guess I would just

comment that even the 75 percent figure in terms of
    27

inter-observer or inter-observer variability with
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
venography sounds very bad, and so it sounds like a bad
     1

state 2of affairs, and one would wonder whether, you

know, 3that is sufficient to seek in a new agent, but

I guess we'd have to think about that.
     4

       5         The third question, I'd like to ask Dr.

Ginsberg something.
     6

       7         I guess I hear you say that based on the

sensitivity figures that you see, and the number I
     8

remember is 70 percent range -- now, I know that in some
     9

subgroups it reached higher, but in the overall patient
    10

population, I think it was 70 percent or maybe slightly
    11

lower, and you commented that the sensitivity level
    12

      you saw might be adequate for you to say that if
that 13

      test was negative, you'd be willing to send the
this 14

patient home on the basis of that.
    15

      16         I think reflecting on that and reflecting

     on
also 17 the comment that the reason venography has some

value in terms of outcome is that there are outcome
    18

studies that have been done that show that patients with
    19

negative venograms are sent home and do okay by some
    20

standard.
    21

      22         I think based on those two points would you

support a prospective study in which you took AcuTect,
    23

you pulled out patients who were negative, and you sent
     24

      home and followed them and sort of confirmed in
them 25

an experimental trial that you were not harming
    26

patients by sending them home with a negative test?
    27

      28         DR. GINSBERG:      Yeah, it's an excellent
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
question.
     1          The sensitivity that I drew from the slide

was closer to about 85 percent, and it's really critical
      2

that we get up between 80 and 90 percent.
     3                                                 Otherwise

you're too low, and your negative predictive value
     4

falls 5to levels that are too low to reliably make

management decisions.
     6

       7         So what we're looking at really is this

sensitivity here, which is probably --
     8

       9         CHAIRPERSON RAMSEY:         Excuse me, Dr.

Ginsberg.
    10          You have to speak into the microphone so

      it
that 11 goes on record.           If you could use the pointer.

      12         DR. GINSBERG:       The pointer?     Okay.   We

don't have a lot of high tech in Canada.
    13

      14         (Laughter.)

      15         DR. GINSBERG:      The sensitivity that I was

referring to, and keep in mind that I think this is a
    16

conservative sensitivity, but this would be
    17

approximately 85 percent, and the reason I use this
    18

group of patients is that these are patients whose onset
    19

is less than three days, in whom the venograms are
    20

likely to be the most accurate.
    21

      22         When we include all patients, we've got the

background noise of the inaccuracy of interpretation
    23

of venography and the misinterpretation of venography.
     24

      25         In addition, the way the analysis was

conducted was one that would be a conservative
    26

sensitivity.
    27              So, in fact, a sensitivity of 85 percent
for both calf and proximal DVT would provide sufficient
     28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
impetus for me to do a clinical management study.
     1

       2          Now, could this be a stand alone test?         My

thought would be that I would take patients according
     3

to their pretest probability, which is an important
     4

predictor of post test probability.
     5                                             If it was low and

they had a normal P280, I'd send them home.
     6                                                     If it was

moderate and they had a normal P280, I'd probably also
     7

send them home, but if it was high and they had a normal
      8

P280, 9I'd probably use something else.

      10          DR. KONSTAM:      Well, I guess here's my

question.
    11           Taking that subgroup, say, as your new

hypothesis, would you support a prospective study in
    12

patients with onset of symptoms less than three days
    13

who had a negative test and go forward and follow them
     14

and watch outcomes or whatever follow-up you would
    15

design?
    16         Would that be a study that you'd like to see?

      17          DR. GINSBERG:      Unquestionably, and we've

     that with the D dimer assay, which has very similar
done 18

accuracy indices, and which we and others have shown
    19

can be used to manage patients.
    20

      21          CHAIRPERSON RAMSEY:         Thank you.

      22          I'd like to just say to the Committee we'd

      to
like 23 end the question session at 11 o'clock and go

forward with the next section.
    24                                      So with that in mind.

      25          DR. AMENDOLA:      I have a quick question.

If we take the Hamilton read as the gold standard --
    26

      27          CHAIRPERSON RAMSEY:         Could you state your
name, please, again for the record?
    28

                               S A G CORP.
202/797-2525    Washington, D.C.   Fax: 202/797-2525
       1         DR. AMENDOLA:        Dr. Amendola.

       2         CHAIRPERSON RAMSEY:           Thank you.

       3         DR. AMENDOLA:        If we take the Hamilton

read as the gold standard, were the two positive and
     4

the two negative in the predictive bodies of AcuTect
     5

calculated?
     6

       7         DR. DEAN:        Can Dr. Nicodemus?        Dr.

Nicodemus, can you address that question?
     8

       9         DR. NICODEMUS:         Actually can you repeat

     I
it? 10 wasn't quite certain of what you were saying.

      11         DR. AMENDOLA:        Right.      If we take the

Hamilton read as the gold standard, what were the true
    12

positive, true negative in predicted values of AcuTect
    13

calculated?
    14

      15         DR. NICODEMUS:        The true positive and true

negative in predictive values of AcuTect, actually the
    16

sensitivity slide that Dr. Ginsberg just showed was
    17

related to using Hamilton as the gold standard, and
    18

we'll see if we can -- in terms of true positives and
    19

      negatives from that, do we?
true 20

      21         DR. GINSBERG:        Basically sensitivity is a

surrogate for true positives.
    22                                      So the true positivity

      would be in the sort of mid-80s.
rate 23                                                 Specificity is

a surrogate for true negative rate, and so the
    24

specificity would be about 70 percent, and then false
    25

positives can be calculated based on extrapolations
    26

      those data.
from 27
      28         Is that the question you're asking?
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1          DR. AMENDOLA:      Right.      I want to have some

idea of what, you know, the accuracy of the test is and
     2

also the predictive value of the test if we take the
     3

Hamilton read as the gold standard.
     4

       5          DR. GINSBERG:      I see.     What you would need

to do, as you know, is to set up a two-by-two contingency
      6

      based on the prevalence.
table 7                                   I can tell you that with

the sensitivity of 85 or 90 percent and a prevalence
     8

of around 30 percent and a specificity of around 70
     9

percent, the negative predictive value would be in the
    10

range of 90 percent.
    11

      12          So, for example, with a prevalence of 30

percent and a sensitivity of -- were are we here?
    13

      14          Okay.      So these are the actual time

points.
    15         So with prevalences of around we saw between

26 and 30 percent, you can see the negative predictive
     16

value is slightly over 90 percent.
    17                                             The positive

predictive value, not surprisingly, is in the range of
    18

about 50 percent, and obviously as it's well know, as
    19

the prevalence falls and reaches more contemporary
    20

figures of, say, 15 percent -- and this is why I say
    21

I'm comfortable using this test in clinical management
     22

studies -- is that if the prevalence is more realistic,
    23

15 or 16 percent, the negative predictive value would
    24

be in the high 90s, which is as good as anything that
    25

we've got.
    26

      27          DR. AMENDOLA:      Thank you.
      28          CHAIRPERSON RAMSEY:         Dr. Choyke.
                               S A G CORP.
202/797-2525    Washington, D.C.   Fax: 202/797-2525
       1         DR. CHOYKE:        Pete Choyke.

       2         I think, you know, we'd all be happier if

there 3were matched pairs that everybody could agree

were negative and matched pairs where everybody could
     4

agree 5were positive, and what I'm wondering, and

recognizing that there will always be gray cases where
     6

people will disagree and that's sort of in the more
     7

subtle cases; I'm wondering whether it's possible from
     8

either the first blind read or the Hamilton read to
     9

identify a subset of patients who all the readers agreed
    10

     negative and all the readers agreed were positive
were 11

and look at how AcuTect did against those ends of the
    12

spectrum.
    13

      14         DR. DEAN:        Dr. Nicodemus.

      15         DR. NICODEMUS:         Yeah, I don't have that

specific analysis right now.
    16                                     I would point relative

to the Hamilton, of course, you know, there is a
    17

unanimity of interpretation as the old standard, and
    18

for the AcuTect scans, as we mentioned, there's about
    19

      percent unanimity rate, but the actual analysis
a 60 20

you're talking about I don't have available for you
    21

right now.
    22           I'm sorry.

      23         CHAIRPERSON RAMSEY:           One more question.

Dr. August.
    24

      25         DR. AUGUST:        As we listen to these

proceedings, there's a recurrence of three themes, and
    26

      is
that 27 the problems implicit with the fact that our
      standard is really not a gold standard, the
gold 28
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
problems relating to acute versus chronic
     1

thromboembolic disease, and then there is another one
     2

which 3we've talked less about, but is certainly here,

the interaction of the therapies that patients were on
     4

with imaging results.
     5

       6         And it occurred to me as I was reading

through the material prior to the meeting that with an
     7

appropriate animal model, one could really get a lot
     8

of insight into all of those three issues, and we've
     9

     that there are data that we've been presented from
seen 10

animal models.
    11                 Mostly they have to do with toxicity

and maybe pharmacokinetics, and I'm just curious to
    12

      whether you have such data or if you don't, why
know 13

don't we have it?
    14                      Is it because the animal models

really aren't relevant to the human situation or what?
    15

      16         But it seems to me that some of these --

     the approach to answering some of these questions
that 17

would be really admirably served by the use of an animal
    18

model.
    19

      20         DR. DEAN:        I'm going to ask Dr.

Lister-James to address that.
    21

      22         DR. LISTER-JAMES:          I think the approach

      we
that 23 took is we were most interested in the effect,

of course, on the interaction of the product with human
     24

platelets, and that's why we did the study in vitro,
    25

looking at the binding of the agent or the ability of
    26

the agent to inhibit platelet aggregation in the
    27

presence or absence of heparin as a direct measurement
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
of the effect of heparin on the ability of the product
     1

to bind platelets.
     2

       3         We did not do studies in dogs with or

without heparin.
     4                    I suppose one could do that study.

Of course, one has to take into account the fact that
     5

the cross-reactivity of the agent with dog platelets
     6

is less than with human platelets.
     7                                            So that does tend

to make it not quite as relevant as using human
     8

platelets.
     9

      10         So I think the real answer is that we chose

to use the human -- as close to human situation as we
    11

could.
    12

      13         In terms of getting at acute versus

chronic, that's particularly difficult to do in
    14

animals.
    15         As you're probably aware, dogs have a very

highly developed fibrinolytic system.
    16                                                 It's very

difficult to develop chronic thrombi in the dog model
    17

or, in fact, in other models, and so we could not think
     18

of a way to address dealing with that specificity issue
     19

in animals.
    20

      21         CHAIRPERSON RAMSEY:          Very brief, please,

Dr. Links.
    22

      23         DR. LINKS:       A question of clarification.

In looking through all of the data, obviously you would
     24

      us
like 25 to base everything on the Hamilton read.                   So

are we, therefore -- is it Tables 38 and 39 that you
    26

would like us to have as the take home message?
    27                                                        That's
on pages 64 and 65 of the briefing document.
    28                                                        I just
                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
want to make sure that the final take home message you
     1

want us to have is those tables and not some other
     2

tables.
     3

       4         DR. DEAN:        Okay.    Let me ask my medical

team here to respond to you on that as soon as they can
     5

confirm that.
     6

       7         CHAIRPERSON RAMSEY:           While everyone is

thumbing through, the next to follow, Dr. Patricia Love
     8

will introduce the FDA speakers.
     9                                           So, Patricia, you

could be prepared for that.
    10

      11         MR. MADOO:       Dr. Links, could you reiterate

the page numbers?
    12                      For Committee clarification,

apparently Dr. Links is referring to the sponsor
    13

briefing document, the blue binder.
    14

      15         DR. LINKS:        Right.     Pages 64 and 65,

Tables 38 and 39.
    16

      17         DR. NICODEMUS:        Yeah, I would comment that

it is actually our position that we do believe that the
     18

Hamilton is an appropriate gold standard, and those
    19

tables do appear to be appropriate tables.
    20

      21         We also would point out that the results

of Hamilton are consistent with the institutional site
     22

      as
read 23 a secondary and supportive analysis, and that

I wouldn't want you to discard the information relative
     24

to the institutional site read, which I believe is very
     25

consistent with the results of the Hamilton results as
    26

well, as a secondary endpoint.
    27

      28         CHAIRPERSON RAMSEY:           We have another
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
comment.
     1

       2         Please state your name first.

       3         DR. D'AGOSTINO:           Ralph D'Agostino.

       4         If we do that, then we're saying that we

don't 5buy the original protocol primary endpoint

because it was not the Hamilton.
     6

       7         MR. MADOO:        Do you have a comment, Dr.

Hammes?
     8

       9         DR. HAMMES:        Yeah, just one comment to

that, which is that the primary endpoint is the same.
    10

The point that we did make was that the true standard,
     11

which I think in a trial of this nature the issue is
    12

     is
what 13 clinical truth, and we clearly have identified

      clinical truth was being inaccurately diagnosed
that 14

using the prospective methodology.
    15                                            That, I think, has

      discussed in detail, and so I just would point to
been 16

      distinction, which we have reviewed.
that 17

      18         CHAIRPERSON RAMSEY:           Thank you.

      19         I'd like to thank the Committee and the

presenters for that session and now turn the podium over
    20

to Dr. Patricia Love, who will introduce the FDA
    21

panelists.
    22

      23         DR. LOVE:        Hello.    Just a couple of brief

comments before the review team presents their
    24

information.
    25

      26         First, I'd like to note that we've been

joined at the table by Dr. Lilia Talarico.
    27                                                      She's the
Division Director of Gastrointestinal and Hematologic
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
Products, where some of the therapeutic antiplatelet
     1

products that were mentioned earlier have been reviewed
     2

in the FDA.
     3

       4         Also, as often is the case when we're

coming to the end of an action, there is quite a dynamic
     5

that goes on between the sponsors and the FDA.
     6                                                We've

worked a great deal to try to make sure that the database
     7

that's presented to the Committee today is consistent.
     8

       9         However, as we listened this morning,

there were a couple of things that we noted that might
    10

      little bit different from what's in the
be a 11

application.
    12              We've talked to the sponsors about this

during the break, and they have agreed to submit the
    13

additional information, but for your reference two
    14

items might be of interest as you go through your
    15

proceedings today.
    16

      17         One is the amount of vitronectin binding.

There's a difference as you'll see from our presenters
    18

and the sponsor, a difference of either 100 or 1,000
    19

nanomolars.
    20             That might be a typographical error that

can be resolved.
    21

      22         Also, the Hamilton read information

prospective criteria is not in the existing submission,
    23

and the sponsor has agreed to amend that.
    24

      25         If someone could just turn on the slide

projector, the overhead there, please.
    26

      27         The only other point to make at this moment
is that the review team order of presentation is going
     28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
to be 1different from what is in your agenda.                Dr.

Laniyonu will present first, followed by Dr. Zolman,
     2

Dr. Jones, and Dr. Sobhan.
     3

       4         Thank you.

       5         CHAIRPERSON RAMSEY:         Dr. Laniyonu.

       6         DR. LANIYONU:      Thank you very much, Dr.

Love. 7

       8         Good morning.      Today I'll be presenting

the review team's pharmacology, toxicology,
     9

perspective of this submission, but before I go into
    10

the details of my talk, I would like to thank Diatide
    11

for the excellence of their submission.
     12                                               There was many

volumes that are well indexed, and it really
    13

facilitated our review process.
    14                                      Thank you very much.

      15         As I indicated, I was the review

pharmacologist on NDA 20-887, AcuTect.
    16                                                 In doing our

review process, we considered some key review issues
    17

      were unique to this product and some that we
that 18

encounter on a day-to-day basis in the division.
    19

      20         And these were the key questions that we

asked ourselves.
    21                    For receptor based agent, we wanted

to know the pharmacological basis of action of these
    22

products.
    23

      24         Secondly, we wanted to see whether Diatide

provided us with proof of concept and evaluates those
    25

concepts from a set of criteria that must be fulfilled
    26

by imaging that this interaction with receptor based
    27

      of
kind 28 them (phonetic).
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         Thirdly, we evaluated the experimental

evidence as presented by Diatide.
     2

       3         And finally, we considered the

pharmacology for and toxicology issues that arose from
     4

our review process.
     5

       6         As submitted by Diatide, the

pharmacological base of action of AcuTect includes the
     7

following, and this is really from the literature.
     8

       9         One, that fibrinogen binds to the

glycoprotein 2B3, which subsequently I'll be referring
    10

to as alpha-2, the third receptor, by the sequence
    11

argininyl-glycyl-aspartic acid, and this can also be
    12

called the RGD sequence.
    13

      14         Secondly, that if you synthesize peptides

containing the RGD sequence, that they're capable of
    15

binding to the receptor sites.
    16

      17         And finally, radiolabeled peptide.   With

continuing this sequence, it should be able to detect
    18

actual platelets in acute deep venous thrombosis.
    19

      20         Without going into the detailed mechanisms

of signaling by fibrinogen and other integral
    21

receptors, I would like to say that we actually agree
    22

      Diatide on these three bases.
with 23

      24         But what are the clinical implications of

an agent that acts via activation of -- that can only
    25

detect difference to both the true activation of
    26

platelets?
    27

      28         These are the clinical implications.
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
Just to go over my first point again, implicit to the
     1

proposed mechanism of action is the requirement for
     2

platelet activation.
     3                            So theoretically AcuTect will

bind with the platelets irrespective of the
     4

pathophysiological process of both the regions
     5

involved.
     6

       7         That leads to the difficulty in

distinguishing acute propagating thrombi from
     8

inflammatory actions requiring platelet activation.
     9

      10         And finally, you may have the discussion

      pressure activity (phonetic) and background
with 11

uptake processes.
    12

      13         For the proof of concept evaluation, we

wanted to know how does the affinity of apcitide for
    14

alpha 2, beta 3 receptors of platelet compare with the
    15

affinity of fibrinogen for the same receptor.
    16

      17         Thank you.

      18         Secondly, we wanted to see how selective

is apcitide for this receptor compared with the
    19

selectivity for the integral receptors sharing the
    20

common beta 3 subunit, for example, the alpha file
    21

(phonetic), beta 3 integral receptors of vitronectin,
    22

which is present on endothelial cell surfaces.
    23

      24         For the proof of concept studies, Diatide

submitted the following information.
    25                                                   They gave us

studies regarding the receptor binding properties of
    26

AcuTect, the binding of apcitide to human platelets,
    27

functional studies, and injury model of venous
    28

                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
thrombosis.
     1

       2         This is an in vitro receptor assay in which

we compare the in vitro concentrations, 50 IC-50 for
     3

fibrinogen receptor with that for vitronectin
     4

receptor, and as pointed out by Dr. Love, the figure
     5

indicated that this is 1,000 nanomolar.
     6                                                For the

solution that I reviewed, it was stated to be 100
     7

nanomolar.
     8

       9         So you have a suggestion in which apcitide

preferentially binds to fibrinogen receptors and less
    10

avidly with vitronectin receptors, suggesting of low
    11

cross-reactivity.
    12                      Whether this 100 or 1,000, I do

agree with this study that there is little
    13

cross-reactivity with vitronectin receptor sites.
    14

      15         Furthermore, they also show that

Technetium labeled apcitide binds specifically to
    16

washed platelets, and that can actually displace about
    17

77 percent of this binding by the process called
    18

bibapcitide, and that when you use a global stimulant,
    19

      as
such 20 adisen (phonetic) diphosphate to stimulate or

activate platelets, you have a threefold increase in
    21

binding.
    22

      23         So these studies demonstrated that you can

actually have in vitro binding to platelets.
    24                                                     So you

     two key concepts here.
have 25                               The first is that Diatide

has shown that AcuTect can bind to in vitro alpha 2,
    26

      3
beta 27 receptors, and secondly, you can actually
demonstrate the in vitro binding to activated platelets
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
by AcuTect.
     1

       2         So we need to ask:         what are the

consequences or the functional consequences of this
     3

receptor of this AcuTect-platelet interaction?
     4

       5         And the first one that you can actually

deduce from the proposed mechanism of action is that
     6

an agent such as AcuTect will actually inhibit in vitro
      7

platelet aggregations, and this are the peptides that
     8

are contained within the formulation when you give it,
     9

and all of them actually inhibits platelet aggregation,
     10

albeit by a different potency.
    11

      12         You have the bibapcitide which actually

contains two dimers of apcitide, BB (phonetic), B equal
    13

potent with P1007, which are two dimers, and the less
    14

potent is P1008, which has an individual concentration
    15

of about 700 nanomolars.
    16

      17         Furthermore, in ex vivo platelets

aggregation studies, in this case dogs were
    18

administered doses of AcuTect that correspond to either
    19

the maximum human dose, which is two microgram per kg
    20

or multiples of this 30x or 100x, and the percentage
    21

platelet inhibition was studied.
    22

      23         At the dose equivalent to the dose that a

50 kilogram person would obtain, there was no
    24

inhibition of platelets aggregation.
    25                                                As you increase

the concentration, 30-folds to 100-folds, you have the
     26

30 percent inhibition of platelet aggregation to 90
    27

percent inhibition of platelet aggregation, suggesting
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
that the dynamic activated platelet better interaction
      1

resulted in a measurable physiological response.
     2

       3         In this very vital study, bleeding time was

not systematically studied, and what is critically
     4

missing from this piece of information is that I do not
     5

have the dose or the concentration of AcuTect between
     6

1X and 30X, at which there was no inhibition of
     7

platelets aggregation.
     8                            So I really do not know the

safety margin between 1X and 30X for this study.
     9

      10         I believe one of the panel members

suggested that some of these studies can actually be
    11

accomplished through in vitro animal studies, and this
    12

is an example of such a study that might easily be
    13

accomplished.
    14

      15         Diatide presented data showing that

neither heparin or aspirin affected the
    16

anti-aggregatory effect of apcitide.
    17                                                 What was missing

was that there was no data to show whether heparin or
    18

aspirin will affect the binding of apcitide with the
    19

receptor.
    20

      21         This is important because one of the NDA

submissions, Diatide advanced the concept that maybe
    22

the concentration required for inhibition of platelet
     23

aggregation is far above that would normally be seen
    24

in the clinical setting.
    25

      26         While I agree with that, the functional

interaction with receptor is actually critical and
    27

important simply because the concentration that will
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
inhibit those receptors is invariably the same
     1

concentration range that would be used in clinical
     2

practice.
     3

       4         For the efficacy study, data used an injury

model 5of thrombosis, and as Dr. Lister has pointed out,

it's actually extremely difficult to get a chronic
     6

model.
     7     You can actually get a good, acute model of

venous thrombosis.
     8

       9         Using the canine venous thrombosis model,

we use a background entwined still embolization coil
    10

in the femoral vein.
    11                            It was established that the

negative control, Technetium labeled glucoheptonate,
    12

did not image thrombus, and that Technetium labeled
    13

      or
P280 14 apcitide provided good in vivo visualization

of thrombi, and I'm actually using the words as used
    15

by Diatide.
    16

      17         Finally, for the positive control, they

      Technetium labeled HMU PAO (phonetic) platelets,
used 18

and they felt that it gave excellent images of thrombi.
     19

      20         They went on to say that a clear advantage

of Technetium labeled P280 compared with the platelets
     21

is that there is rapid excretion.
    22                                             There is rapid

clearance of Technetium labeled in the body compared
    23

      platelet labeled cells, and you have a better
with 24

thrombus-to-background ratio.
    25

      26         This is a table adapted from Diatide's

submission, and it shows that for the glucoheptonate
    27

you have a thrombus-to-blood ratio of about two, and
    28

                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
for the Technetium labeled P280, you have a
     1

thrombus-to-blood ratio of about four, and for a
     2

Technetium labeled platelets it was about 5.4, again
     3

confirming Diatide's conclusion.
     4                                         Platelets labeled

Technetium seems to give better visualization of these
     5

thrombi.
     6

       7         The studies, therefore, demonstrated that

you can actually demonstrate binding of apcitide to the
      8

growing thrombus.
     9

      10         So all of these studies submitted by

Diatide demonstrated that apcitide preferentially
    11

binds to fibrinogen receptors.
    12

      13         Secondly, you have in vitro binding to

platelets, and as a consequence of these two effects,
    14

there's a dose related inhibition of platelets
    15

aggregations, and that in an animal model of thrombosis
    16

reflected uptake in thrombi.
    17

      18         I still have some lingering questions

though, and the first one is that as submitted by
    19

Diatide, I do not have a clear indication of what's the
    20

     for
NOEL 21 inhibition of platelet aggregation within the

clinical setting, and the NOEL is defined as the no
    22

observable effects level, that is, the dose of apcitide
    23

      will not affect platelets aggregation in the
that 24

clinical setting was absent.
    25

      26         And finally, what is the relationship of

the receptor binding of that of alpha 2, beta 3 -- to
    27

the proposed clinical use, and on that note I call on
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
Dr. Zolman to continue with the presentation.
     1

       2         DR. ZOLMAN:      Good morning, ladies and

gentlemen.
     3           I will present the safety evaluation

perspective on this drug.
     4

       5         My name is Joseph Zolman.            I am a medical

officer in the Division of Medical Imaging.
     6

       7         I reviewed the safety aspects of this drug

and concluded that the effect, the untoward effects of
     8

this drug are rather mild.
     9                                This was judged by the

evaluation of adverse drug events, their nature and
    10

frequency, as well as the effect of the drugs on vital
    11

signs and laboratory measurements.
    12

      13         Thus we are in general agreement with the

sponsor that the drugs are relatively safe.
    14                                                     However,

      agreement is preliminary based on the nature of
this 15

the safety database.
    16

      17         As you can see from this overhead, the

total of patients and normals enrolled is 714, and
    18

exposed 710.
    19              This is in agreement with the sponsor.

      20         Seventy-eight of these patients and

normals were exposed to early formulation and 632 to
    21

proposed for market formulation.
    22

      23         Adverse drug events were examined at 632

patients, vital signs at 450 patients, and labs were
    24

measured in 140 patients and normals.
    25

      26         However, the question and concern is not

in the total numbers of patients and normals involved,
     27

but with the depth of the observations.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         As we can see from here, on 169 patients

and normals were followed for adverse drug events for
     2

24 hours.
     3          Only 102 patients were followed for vital

signs 4for 24 hours, and labs were measured for 140

patients and normals at three and 24 hours.
     5

       6         Therefore, the investigation doesn't

provide sufficient amount of information, and the
     7

safety database is limited because of lack, of
     8

insufficiency of information beyond three hours.
     9

      10         As we can see from here, 169 patients were

followed for 24 hours for adverse drug events.
    11                                                           This

      very small number, 632 patients for the total of
is a 12

three hours.
    13              There were no deaths.             There was one

serious documented hypotension, and there were 34
    14

adverse drug events in the category of mild and moderate
    15

events.
    16

      17         The serious event related to a 34 year old

      five days after motorcycle accident.
male 18                                                     Following

the administration of the drug, the patient went from
    19

145 systolic pressure to 110 in 15 minutes, and then
    20

later to 70 in 60 minutes.
    21

      22         He was treated with fluid infusions and

recovered quickly.
    23

      24         The nature and number of mild and moderate

      events is depicted here.
drug 25                                  Essentially those were

few in frequency and mild in nature.
    26                                                 The numbers

reflect only those who were present more than one hour
    27

during the study.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         Of these presentation of the various

documents may reflect some degree of hypersensitivity.
     2

Even the serious case could be a potential case of
     3

hypersensitivity.
     4                      However, we don't have the data to

document this accurately and cannot assert it with any
     5

firmness.
     6

       7         The mild and moderate drug events could be

related to preexisting conditions, for example, pain.
     8

       9         Another safety concern is potential

immunogenicity of the product.
    10                                      The sponsor measured

IgG against the P246 and P1007.
    11                                      P246 is the peptide,

and P1007 is the fragment.
    12                                This was tested by ELISA

assay in samples taken at a baseline and 21 days after
    13

single dose of AcuTect.
    14

      15         The sponsor reported no significant change

in the measures.
     16                  All results were within two standard

deviations of mean of optical density for preinjection
    17

data.
    18

      19         This we consider a parameter information

because this is only one of potential parameters of
    20

immunoresponse which could be measured, and more
    21

definite data is needed, particular in reference to the
    22

parameters which could assess hypersensitivity.
    23

      24         As a summary, the review team agrees with

the sponsor in respect to safety data reporting.
    25                                                       We

are in agreement with the sponsor that the drug is
    26

relatively safe, but with this status this agreement
    27

is preliminary because only a limited number of
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
patients was monitored beyond three hours.
     1

       2         There is lack of information on the labs

so that they have more than three hours to change, such
     3

as creatinine and liver enzymes.
      4                                      There is lack of data

to assess potential hypersensitivity and lack of data
     5

pertaining to repeated dosing issues.
     6

       7         There is also lack of information on

bleeding time, which may relate to platelet aggregation
     8

in PT 9data.

      10         Thank you for your attention.            This is all

for the safety aspects of this drug.
    11                                                 Dr. Jones will

now continue with the efficacy evaluation.
    12

      13         DR. JONES:       Thank you, Dr. Zolman.

      14         Having surmounted that little problem of

technology we're ready to begin.
    15

      16         (Laughter.)

      17         CHAIRPERSON RAMSEY:          The hardest part of

the day is putting that microphone clip on.
    18

      19         DR. JONES:       Yes, and not piercing your

finger with the pointer.
    20

      21         (Laughter.)

      22         DR. JONES:       It adds to the excitement, I

think.
    23

      24         I'd like to start with basically

introducing some of the issues that I want to talk about
    25

      morning.
this 26               I hope to be very brief since so much

has already been said by the sponsor, and I don't want
     27

to be too repetitious.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
       1         However, I do want to repeat the claim for

AcuTect, and I do want to address some of the technical
     2

features of the image because they are very important
     3

to the agency to be able to support the claim for the
     4

drug. 5 It's important for us to have that kind of

information to validate the data that must go into the
     6

package insert.
     7

       8         The blinded read criteria, I would like to

remind the Committee what has already been presented
     9

by the sponsor.
    10                   I'll be quite brief about that.

      11         I also wish to talk about the case report

forms that the sponsor provided to the blind readers.
    12

There were two report forms that were to be filled out.
    13

      14         The data that was collected then would be

in my Point 4 of the relationship of the image findings
     15

to the proposed use and the issues that seem to arise
    16

      some of those results.
from 17

      18         The claim for AcuTect, as we all know by

now quite well, is that it is indicated for the
    19

scintigraphic imaging of acute venous thrombosis,
    20

emphasis on the word "acute," and that it's venous
    21

thrombosis.
    22             There's no mention of phlebitis in that

indication.
    23             It's a very distinctive, targeted claim.

      24         The technical features of the image.      The

technical features are very important to be established
    25

in Phase II since they are the hypotheses that are to
    26

be tested in Phase III to help us with the labeling,
    27

to truly help us establish the truth of what is being
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
seen by the readers.
     1                            They're very important.     They

should actually be descriptive of a manifestation of
     2

disease.
     3

       4         And having met those requirements, they

should be able to be easily incorporated into the
     5

package insert in support of the claim of the sponsor.
     6

       7         Reminding everyone again, ad nauseam

perhaps, about the blinded read criteria, the sponsor
     8

required that there be unilateral asymmetry; that the
     9

asymmetry might be in the iliac, thigh, popliteal, or
    10

     area; that the abnormality be seen on both anterior
calf 11

and posterior projections; and that the readers were
    12

allowed to adjust the contrast such that, as noted in
    13

Point 4 here, if asymmetry appears only after extreme
    14

contrast enhancement.
    15                            Then the image was to be called

positive if there's also a diffuse asymmetry.
    16                                                        It was

      called negative if there was no diffuse asymmetry.
to be17

      18         Now, the case report forms that were

provided to the readers are as follows.
    19                                                   I have to

apologize for this one.
    20                               It does not -- it did not

translate well electronically, but essentially what it
    21

allows the blinded reader to do is to record the site
    22

of positivity and whether or not the positivity is
    23

actually not seen or whether it's inconclusive or
    24

whether it's strongly positive.
    25

      26         Having made the determination that there's

actually a positive finding, the reader then went to
    27

the next case report form, which again didn't reproduce
     28

                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
well for me, and I apologize.
     1

       2         In this case report form, for all the

positive readings the blinded reader was to note the
     3

side of the abnormality, whether it was iliac in
     4

location, the thigh, the knee, or the calif.
     5                                                         The

intensity of uptake was to be recorded, whether it was
     6

slight, moderate, or highly intense.
     7                                                The shape of the

lesions, circular, linear, or irregular, was also to
     8

be recorded, and the extent of vascular involvement was
      9

the final feature.
    10

      11         That then took into account all of the

positive readings, positive images, that is.
    12

      13         Now, this left us with some limitation of

data. The case report form did not collect features
    14

of negative or indeterminate interpretations.
    15                                                          We

actually were not able to get true positive, true
    16

negative, false positive, false negative assessments.
    17

     would have been very helpful to our statisticians.
This 18

      19         The sponsor reported          the image findings

for the cases believed to be representative of acute
    20

thrombosis.
    21             The image findings for the negative cases

      not reported.
were 22                      It's unusual that all cases are

clearly positive in medical imaging studies.
    23                                                        What we

don't have is what's the break point between the
    24

negative and the positive image and what's the
    25

variation.
    26           What are the imaging endpoints that occur

in that region of interpretation?
    27

      28         Similarly, we don't have any data on
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
patients with phlebitis alone.
     1                                      What did the images

appear like with phlebitis?
     2                               Perhaps there is no reason

for concern with that since the sponsor indicates that
     3

there 4is likely to be no localization of activated

platelets in the presence of phlebitis, at least with
     5

AcuTect.
     6

       7         So this leaves the review team with some

questions.
     8           What are the relationships of the image

findings to the proposed use?
     9

      10         And as I've said, the question of phlebitis

versus thrombophlebitis to us remains still a murky
    11

area.
    12

      13         And the question of distinguishing acute

      chronic thrombosis is also not clear.
from 14

      15         The issue of anticoagulant therapy and its

influence on image has been discussed a bit already this
    16

morning, and I reintroduce our concern about that.
    17

      18         Repeat doses provide another concern.

      is
This 19 a diagnostic product.             Many imaging products

are used to assess baseline criteria of a disease, and
     20

following therapy they may be repeated again.
    21                                                    There

is the possibility that this product could be very
    22

useful in following therapy and may be needed to be
    23

repeated more than one.
    24                            I'm hypothesizing.

      25         That being the case, we've heard earlier

      Dr. Laniyonu that in the preclinical studies,
from 26

AcuTect tends to inhibit the aggregation of platelets.
    27

We also wonder about the immunogenicity, as Dr. Zolman
     28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
raised the issue.
     1                     If there's immunogenicity, is there

some possibility that this test they have created
     2

antibodies that may render it less useful or perhaps
     3

even some hazard introduced because of the induction
     4

of antibodies.
     5

       6         Regarding the collection of safety data,

we realize that 90 percent of the product is eliminated
      7

within the first 24 hours through the kidneys, and this
     8

may have caused the sponsor to have unnecessarily
     9

perhaps shortened the collection time of safety data.
    10

      11         However, the safety data is very important

to us, to be carrying it out particularly if any
    12

abnormal safety data occurs.
    13                                   We want to be able to

follow it until it returns to normal.
    14

      15         And the issue that has been the large one

today about the adequacy of the standard, I'm not going
    16

to say very much more about that, except that it has
    17

occurred to us that while venous contrast phlebography
    18

has been accepted by ourselves and the sponsor as a
    19

standard of truth, it really isn't actually a standard.
    20

      a
It's 21 comparator.          If we could have a standard as

someone mentioned, to actually get the clot and look
    22

at the histology, that would be ideal.
    23                                                It is

impossible.
    24

      25         We have a unique agent here under

discussion today that's a receptor, and receptor agents
    26

are going to introduce this problem in the future should
     27

any more come along, and I'm sure they will.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         So this is a big problem, and we need the

Committee's help with this particular issue of the
     2

standard.
     3

       4         Thank you very much.

       5         DR. SOBHAN:        That's the last time you have

to wait for that kind of a struggle with the microphone.
      6

       7         What I'm going to do this morning -- my name

is Mahboob Sobhan.
     8                       I am the division statistician on

AcuTect.
     9

      10         What I'm going to do is revisit some of the

features of efficacy.
    11                            I understand there are a lot of

questions came out from the panel members as to the
    12

consistency of the result, the comparator used, and
    13

     of
some 14 the endpoints like sensitivity and specificity.

      15         I'm going to skip some of the study

features because the sponsor has done a good job of walk
    16

you through those things.
    17                                So I'm going to skip some

of the study features.
     18                           I'll come to the endpoint, some

of the measured predictions.
     19                                   I will also skip because

Dr. Jones already explained some of those features.
    20

      21         My most focus should be on the results from

pivotal studies and then what role CVs play in this
    22

application, and then I'll finish with my conclusion,
    23

summary and conclusion.
    24

      25         This is just to revisit.               The comparator

is contrast venography or standard of truth, what we
    26

are studying this morning.
    27                                  External of the standard
of truth is no available, which is not possible probably
     28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
for venous thrombosis.
     1                              The objective is to detect and

characterize acute VT compared to contrast venography.
     2

       3         This is an idea you have seen this morning.

Procedure is measuring before and after.
     4                                                    I think it's

a matter of convenience rather than order, randomized
     5

order, and images are taken and three different time
     6

points.
     7

       8         To show efficacy, this is the -- as far as

protocol, this is the endpoints, the agreement rate,
     9

which is number of positives and negatives detected by
    10

     modalities and sensitivity and specificity, which
both 11

I put in the quote.
    12                            Quote means I call it pseudo

sensitivity.
    13              In other words, we don't have the real

truth, 22 carat gold.
    14                              So I put it in the quote.

      15         This is a little bit just to revisit some

of the mathematical or definitions of sensitivity,
    16

specificity in the real situation.
    17                                               If you had a gold

truth and you have a test agent that correlates with
    18

the gold, then you can define since you expect like
    19

this, and accuracy which is, you know, defined as this,
    20

and agreement rate used in this application is a
    21

surrogate for accuracy.
    22

      23         This will be the scenario for real

situation if you have the gold standard -- I mean the
    24

      standard of truth.
real 25                              As you can see, the sense and

expect (phonetic) is really -- the pivotalness
    26

(phonetic) is also the function of sense and expect,
    27

but we're not going to touch all those issues.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.       Fax: 202/797-2525
       1         The hypothesis as per protocol.           I want to

remind you this is as per protocol.
     2                                            The hypothesis was

to reject that the agreement rate of 60 percent is below
      3

-- I mean it's below 60 percent as opposed to more than
      4

60 percent, and they used this approach to demonstrate
     5

that the product works.
     6

       7         I have not seen anything about 75 percent

in the application as such.
     8

       9         I'm going to revisit this that was done.

Three things were done.
    10                            One is I didn't put it in here,

which is the blinded criteria they used to train the
    11

readers, and then this is how the score was done on the
    12

image.
    13     This was done even if the patients are

positive, which Dr. Jones showed you through the
    14

schematic diagram.
    15

      16         Let's focus on the blinded read.            There

are three types:
     17                  blinded read, per protocol, majority

     decision.
rule 18                For CV three readers, and majority was

the decision.
     19               For unblinded read at the institution,

unblinded meaning he or she had access to personal
    20

information, and this is done post hoc consensus.
    21                                                          The

AcuTect reads are done by three different readers, as
    22

     pointed out this point.
they 23                                  They're made at any time

point and collected by all time points.
    24

      25         Let's look at this.           This is from the

sponsor's submission.
    26                            First row is blinded read,

which is originally planned for protocol.
    27                                                   You can see
you have seen this number is 45 versus 82, which is like
     28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
twofold difference between Study A and B, and by
     1

unblinded it's more consistent to what decided in the
     2

application that the prevalence of venous thrombosis
     3

is 30 4to 40 percent.             It still is a little bit higher

in Study B, but after these two are done, the analyses
     5

are done, all the facts are known, this is what is
     6

retrospectively done, retrospective meaning after the
     7

facts.
     8     After all the studies are completed, this is

the analysis that they have done at Hamilton, and look
     9

at these numbers, 21 versus 33.
    10                                          Again, I agree 33

percent is very close to what the reference or the
    11

literature suggested, but here we have a problem also.
    12

     below, much below what we've seen by the other two
It's 13

methods on the first row and second row.
    14

      15         So the question is:           where to reject this,

why do we have to accept this, not mentioning other
    16

problems with the retrospective analysis?
    17                                                   Because we

see here almost 20 percent less than in Study A.
    18                                                           In

other words, Study B is going in the other direction
    19

      the Study B on Row 1.
than 20

      21         But for AcuTect readers it's pretty much

consistent, although it's still a little low side.
    22                                                          The

range here is for three readers.
    23                                          I am presenting read

      read two, read three results, 48 to 54 percent,
one, 24

meaning read one, read two, read three results.
    25

      26         And let's focus on read one.            That's the

      two, three.
one, 27                   The two was done on all time points,
so let's focus on read one.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
       1         This is the result they have submitted.

You recall confidence interval of the statistical
     2

approach.
     3          This is lower bound and upper bound.               The

solid 4bullet is the point estimate, in other words, the

agreement rate.
     5                   Let's focus on reader one to three.

       6         If you look at the upper panel, reader one,

reader two results, significant.
     7                                          Reader two is not

significant because the lower bound contains the point
     8

estimate, and if you go to the lower panel, which is
     9

Study B, none of them made it.
    10                                    All the null hypothesis

could not be exerted in any of the leader evaluation
    11

(phonetic).
    12

      13         In fact, the ideal situation is if we have

all the confidence intervals lying to the right of this
     14

point, the red line, here you have some negative
    15

results.
    16         So that's where the decision was made to do

Hamilton read.
    17

      18         Let me remind you and let me show you the

unblinded read result.
    19                            On the unblind result, which

is the reader has access to all patient information,
    20

there is virtually no change in the Study A.
    21                                                       All

readers are making, and then two readers are making as
    22

we have seen in the blind read.
    23                                        Here is some

improvement.
    24              Only according to one read, which is read

four, it is significant.
    25                              None of the other five was

significant.
    26              So still we're seeing some point

estimate falling beyond 60 percent, but nonetheless,
    27

there is statistical not significant.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
       1         Let's look at the Hamilton read.          Here we

have some inconsistency as far as Study A is concerned.
      2

You are seeing a little bit off here as opposed to blind
      3

and unblind read in Study A, but Study B, almost all
     4

made except one.
     5                    Five out of six are making it.        So

we can see the results of Hamilton makes in B, but it's
     6

not consistent with what we have seen as the protocol
     7

in both analyses, blind or unblind.
     8

       9         What are the implications?           Let's look at

      of
some 10 the agreements.           What are the agreements

between AcuTect read one versus blind?
    11                                                Blind, I'm

referring at the same time blind means general read.
    12

You can see in Study A the comparator statistics, which
     13

simply measures the agreement observed minus the chance
    14

agreement.
    15           Chance agreement means the mismatch

probability.
    16

      17         So if you look at this, it's still less than

.5, which is not good, as pointed out by the sponsor
    18

also, and in Study B you can see the magnitude is almost
    19

      than half.
less 20                  The AcuTect blinded versus unblind,

      still poor.
it's 21

      22         I didn't have the Hamilton read scores.

      couldn't calculate it, but as you can see here,
So I 23

     Study A and B, the agreement was really poor within
both 24

      methods.
both 25

      26         I heard a lot about the sensitivity and the

specificity of CV as well as ultrasound this morning.
    27

It was around 90 percent, reported as 90 percent, and
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
here is the sens. and spec. calculated.
     1                                                  Even though

we didn't have the real truth, the estimate, what I call
      2

sample sensitivity or pseudo sensitivity, here you can
     3

see AcuTect read one versus original read.
     4                                                    The range

is for three readers.
     5                            It goes from 60 to 76 in Study

A.   In Study A it's almost half, but blind read is
      6

better, but here it's much lower.
     7                                           So I don't see how

it is 8closer to the other two methods, in other words,

CV or 9ultrasound.

      10         So this has some implications for the use

of this product or even the labeling of the product.
    11

How is going to determine or say how this product works?
     12

      is
What 13 the sensitivity of this?                Is it good, as good

as what we have in the market or is it better or is it
    14

      effective?
less 15

      16         So what I'm showing here is the

sensitivity.
    17              Ideally what we like to see is the high

sens. and high spec., but in Study B you see the
    18

specificity is much higher because of what we have seen
    19

in the original contrast with the result.
    20

      21         So the message here is the sens. and the

spec. calculated based on the reference standard is
    22

below what we expect.
    23                            For a modality to be used in

practice, I think the practitioners would like to see
    24

as high as possible.
    25

      26         Now, the question came to us:            which one

are we to use?
    27                 We have discussed this.          The panel
members discussed this morning.
    28                                         This is what's done
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
in the application.
      1                       These are the desirable features.

       2         Prospectively planned, yes.            Unblind,

yes. 3Prospectively I forgot to mark it.

Independence, yes.
     4                       Yes.    Hamilton, no, because both

studies were interpreted by the same readers.
     5                                                        Blinded

to patient history in AcuTect scans, yes.
     6                                                   Unblinded,

they said yes.
     7                 I'm not sure.       This is Hamilton read,

blinded.
     8         Consistency in this area, I left it unblind

because the image criteria they used by the AcuTect
     9

readers and Hamilton readers were probably different.
    10

We have not seen it, so I can't comment on that.
    11

      12         The problems, although there is consensus

read, consensus read meaning you resolve the case, you
    13

know, whether it is positive, negative or whatever.
    14

The study results is dependent upon CV.
    15                                                  In B what we

      seen, the inconsistency.
have 16                                     That's what I'm

referring to.
    17

      18         So ideally the features or the advantage

point is still as per protocol analysis rather than both
    19

unblind or Hamilton, even though the unblind read is
    20

the most often practiced.
    21

      22         I read the sponsor -- the patient profiles

in both the studies were the same.
     23                                         Venographies rather

acted comparator other than the standard of truth.
    24                                                             As

I mentioned earlier, this has some implication on the
    25

sensitivity and the specificity.
    26                                           If it is a

comparator, it's viewed as a comparator, and sens. and
    27

spec. interpretation is rather difficult.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         Both the studies, there was agreement to

study 2with CV comparator.             Therefore, it could not be

determined.
     3

       4         The diagnosivity of blinded are different

in Study A and B.
     5                     That's what we have seen.         AcuTect

reads 6with CV in detecting more than 60 percent of

patients according to 50 percent of the blinded
     7

readers; similar results we have seen by both unblind
     8

and Hamilton ready in Study A, but not in Study B.
     9

      10         In B, AcuTect does not agree in any

reader's evaluation.
    11                            In the same study AcuTect do

agree two out of six, one blinded as CV and five out
    12

of six Hamilton read, and I think the sponsor has shown
     13

subgroup effects on agreement rate were not
    14

statistically significant in both the study and B.
    15

      16         So my conclusion was AcuTect NDA lacks one

of the requirements that we have in two adequate and
    17

     controlled trials.
well 18                             There is substantial evidence

      what we have here is shown in one study, not to
that 19

mention other acute versus thrombi question.
    20                                                       Study A

could be considered statistically adequate in support
    21

of the purported indication, and Study B is rather weak
     22

or negative.
    23              So we don't really support that that

should be considered.
    24

      25         That's all.         I conclude.

      26         CHAIRPERSON RAMSEY:            Thank you.

      27         And I'd like to thank all of the FDA
speakers for their efficient presentation, and we're
    28

                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
right 1back on time now, and according to our program

the next would be Committee questions on the FDA's
     2

presentations.
     3

       4         So I would like to ask if anyone on the

Committee has any questions at this time.
     5

       6         Yes, Dr. Hammes.

       7         And if all the questioners and responders

could 8please state your name first, it makes it a lot

easier for the record keeper.
     9

      10         DR. HAMMES:          Richard Hammes.

      11         A question for Dr. Laniyonu.              Trying to

get the biochemistry of this thing a little clearer in
     12

my mind, a decade or so ago I had the opportunity of
    13

doing some platelet aggregation work with a protein
    14

      was secreted by platelets called thrombospondin,
that 15

and as I recall, that was involved in the initial
    16

platelets sticking together, and it seems to me it also
    17

was involved in interactions with endothelium.
    18

      19         Is that the vitronectin receptor that you

      talking about?
were 20                           Are we talking about the same

receptors here with this product?
    21                                              Do you know?   Are

you at all familiar with that?
    22

      23         CHAIRPERSON RAMSEY:             Please use the

microphone.
    24

      25         DR. LANIYONU:            I'm not sure whether, you

know, the product that you worked with would be working
    26

on vitronectin receptors or not.
    27                                             The vitronectin
receptors are present on the endothelial cell line, and
    28

                              S A G CORP.
202/797-2525   Washington, D.C.       Fax: 202/797-2525
you do not actually have the glycoprotein 2B3 receptors
      1

on endothelial.
      2                 They are increased and situated on the

platelets.
     3

       4         What this product is showing is a high

degree of selectivity for the alpha 2, beta 3 receptors,
     5

and I 6do believe that the probability of

cross-reactivity with other integrin receptors will
     7

rather be low, but I'm not sure whether I can directly
     8

relate the receptors you studied before to the ones that
     9

      been proposed in this application.
have 10

      11         DR. HAMMES:      Another question.   Is there

any information on the time course of the availability
     12

of these receptors?
     13                       Do we know in the process of clot

formation when these receptors are no longer available?
    14

      15         DR. LANIYONU:      That's actually key and

critical questions because the platelet's involvement
    16

      thrombi formation or propagation is actually
with 17

limited to the initial stages of thrombi formation.
    18

Platelet formation is one of the earlier key steps
    19

required in the process, and so really you are dealing
    20

      a
with 21 product that would be uniquely sensitive to the

narrow time frame.
    22

      23         And the clinical implications of that is

that, at least to my knowledge, that if a patient is
    24

not available at the time period of the development for
     25

      diagnostic imaging, becomes extremely critical
this 26

because from my understanding of the concept, once that
    27

window of opportunity is missed, I do not really see
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
why, if we're just talking about this proposal, why
     1

those 2patients should be caught.

       3         CHAIRPERSON RAMSEY:          Yes?

       4         DR. ROHDE:       Yes.    Charles Rohde from

Johns 5Hopkins.

       6         I have a question about the presentation

of Dr. Sobhan.
     7                 Am I pronouncing your name correctly?

       8         The sponsor in the protocol clearly

indicates one sided confidence bounds, and yet your
     9

presentation gives 95 percent confidence intervals,
    10

and they are distinctly different, and I don't want the
     11

Committee to be misled by the fact that your lower
    12

bounds are a lot lower than theirs, as they should be.
    13

      14         DR. SOBHAN:      No, the protocol say they are

going to construct the confidence interval around the
    15

point estimate, and if the lower bound includes the
    16

point estimate, then they're going to reject the
    17

number.
    18

      19         DR. ROHDE:       Well, that is not what was

presented; is that correct?
    20                                   We need to get that

clarified.
    21

      22         Are there really 95 percent intervals and

you've reported just below the bounds or are they one
    23

sided 95 percent confidence --
    24

      25         CHAIRPERSON RAMSEY:          Please use the

microphone for a response.
    26

      27         DR. ROHDE:       It gets back to Professor
D'Agostino's question earlier about what's a one sided
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
or two sided.
     1

       2         CHAIRPERSON RAMSEY:           State your name.

       3         DR. BALSER:        John Balser, consultant to

Diatide.
     4

       5         CHAIRPERSON RAMSEY:           Thank you.

       6         DR. BALSER:        They are truly one sided

confidence intervals.
     7                            They are not two sided, and it

was prospectively stated in the protocol as such.
     8

       9         DR. SOBHAN:       Okay.     That's fine.   You are

looking at the lower bound only.
    10

      11         DR. ROHDE:        No, they --

      12         DR. BALSER:        No, they're one sided

confidence intervals.
    13                            We're not just looking at a

lower bound in a two sided confidence --
    14

      15         DR. SOBHAN:       Yes, but protocol, you wanted

to test -- you specifically stated if the lower bound
    16

includes the point estimate, they're going to reject.
    17

      18         DR. BALSER:        The one sided lower bound.

That's correct.
    19

      20         DR. SOBHAN:        Fine, but still what I'm

saying is that's what you intended to show.
    21

      22         DR. BALSER:       The problem is if you do a one

sided --
    23

      24         DR. SOBHAN:        You are concerned only with

the lower bound.
    25

      26         DR. BALSER:        No, we are concerned with a

lower bound of a one sided confidence interval.
    27

      28         DR. SOBHAN:        Right, but --
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         DR. BALSER:      That is not the same as the

lower 2bound of a two sided confidence interval.

       3         DR. SOBHAN:      Right.     Ideally you wanted

to see that you were on the right side of that.
     4                                                      Your

confidence interval should be on the right side.
     5

       6         DR. BALSER:      That's correct.

       7         DR. SOBHAN:      That's what you specified in

the protocol.
     8

       9         DR. BALSER:      That's correct, but --

      10         DR. SOBHAN:      We have talked about the

situation that the protocol states many, many times.
    11

      12         DR. BALSER:      I'm still not sure I

understand why you're concerned because it's clearly
    13

stated in the protocol it's a one side confidence bound.
    14

It is not a two sided.
    15

      16         DR. SOBHAN:      I am not disagreeing with the

conclusion, but you are saying -- I'm just showing it
    17

on the slide.
    18                That's what you have shown in the

application.
    19

      20         DR. BALSER:      No, you're showing on the

slide a two sided confidence interval.
    21

      22         DR. SOBHAN:      That's fine, but --

      23         DR. BALSER:      The calculation is quite

different.
    24

      25         CHAIRPERSON RAMSEY:         Dr. D'Agostino has a

comment.
    26

      27         DR. D'AGOSTINO:       There is a table
presented by the sponsor where they identify
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
significant results, and the count is still the same
     1

whether it's one sided or two sided.
     2

       3         DR. BALSER:       That's actually not quite

true, 4but --

       5         DR. D'AGOSTINO:        Well --

       6         DR. BALSER:       I mean, the institutional

read, 7actually there are two --

       8         DR. D'AGOSTINO:        Well, I'm talking -- I'm

sorry.
     9     I'm looking at the Hamilton read.

      10         DR. BALSER:       Yes, the Hamilton is fairly

robust to that issue.
    11

      12         DR. D'AGOSTINO:        Yeah, the Hamilton stays

the same.
    13

      14         DR. BALSER:       That's correct.

      15         DR. D'AGOSTINO:        It's an important

question in terms of our interpretation, and this is
    16

      I
what 17 was trying to get at.             I think we should

understand it.
    18

      19         DR. WELCH:       Yeah, Mike Welch here.

      20         I think the two sided confidence interval

would be more enlightened for our statistical policy.
    21

We typically for non-feriority studies look at two
    22

sided intervals, and I think we have, you know, an
    23

opportunity to do this and look at the data in that way,
    24

although the one sided was specified in the protocol.
    25

      26         CHAIRPERSON RAMSEY:          Yes, Dr. Choyke.

      27         DR. CHOYKE:      I'd just like to ask someone
      the FDA just to clarify in my own mind the
from 28
                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
requirement for two independent blinded studies.
     1                                                          I

mean, 2is that something that's written in stone or is

there 3wiggle room on that?           I mean, what's the wording

of your requirement?
     4

       5         DR. LOVE:        Okay.    That's a major issue.

There 6is an efficacy standard document that's out for

comment that talks about that particular issue.
     7                                                       There

are some specific examples state in there where the
     8

agency might accept one study in certain circumstances.
     9

      10         However, the agency's requirement says

"studies" for the Center for Drugs we're speaking of.
    11

Different centers may also have different
    12

requirements.
    13

      14         So one study is generally considered the

exception, but those are stated in that particular
    15

document.
    16

      17         CHAIRPERSON RAMSEY:           Dr. Links.

      18         DR. LINKS:        A couple of questions.

First, a follow-up to that.
    19

      20         It seems to me that we need clarification

on the definition of the word "independent" in terms
    21

of having pivotal studies because clearly this is a case
     22

where the AcuTect reads and the patient populations
    23

      totally independent, and the only thing that was
were 24

in any sense dependent was the same gold truth
    25

laboratory.
    26

      27         So I'm not convinced there aren't two
independent pivotal trials.
    28                                    I mean, my interpretation
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
would 1be that there are, in fact, two independent

trials.
     2         Now, whether or not you like the outcomes of

them is a different issue, but to me they're
     3

independent.
     4

       5          DR. LOVE:        Okay.    I think it depends on

how you tease apart the sentence and the words.
     6                                                      Okay?

There 7were two trials prospectively designed that were

independent, and if you look at the original blinded
     8

read, 9they are independent.               If you look at the

Hamilton blinded read, because the same three readers
    10

      Hamilton for both Study A and Study B, that's why
read 11

Dr. Sobhan is saying the Hamiltons reads are not
    12

independent.
    13

      14          DR. LINKS:        May I clarify that?

      15          CHAIRPERSON RAMSEY:           Please.

      16          DR. LINKS:        I could certainly understand

if you're going to introduce a new radiopharmaceutical
     17

you want at least two trials so you can be certain that
     18

across a spectrum of nuclear medicine physicians you'll
    19

get comparable diagnostic accuracy in your
    20

interpretations.
    21

      22          What I don't understand is the need to have

so-called independence in establishing the gold
    23

standard or the truth for the comparator.
    24

      25          DR. WELCH:        Yeah, Mike Welch again.     I

think there remains confusion on the differences
    26

between a comparator for active control arm and the gold
    27

standard.
    28           I think if you're talking about a true
                               S A G CORP.
202/797-2525    Washington, D.C.     Fax: 202/797-2525
standard of truth, that is, unquestionable, one can
     1

argue 2that can be used in two individual studies and

the results would be independent.
     3

       4         Here we all agree, I think, that the

comparator is error prone, and whatever false positive
     5

or false negative error rates it would have would
     6

certainly influence any biases in estimating
     7

sensitivity and specificity.
     8

       9         So judging this as a comparator, I think

we need to be comfortable that the comparator was judged
     10

independently in both studies.
    11

      12         DR. LINKS:       Got it.

      13         May I as one more question?

      14         CHAIRPERSON RAMSEY:          Certainly.

      15         DR. LINKS:       I'm still hung up on acute

versus chronic, and the reason I'm hung up on it is that
    16

in the past we've certainly recommended approval of
    17

radiopharmaceuticals that had even less agreement with
    18

a comparator, and we did so because they would still
    19

clinically impact patient care in a very positive way,
    20

and there were no competing imaging techniques.
    21

      22         And from a pharmacologic point of view, I

really want to know what is the actual evidence that
    23

exists, not the theoretical why it should work, but the
    24

actual evidence that this is an agent for acute VT
    25

because it seems to me the only real evidence is that
    26

there's a three times increased binding to activated
    27

platelets, and if I put on my PET brain receptor imaging
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
hat, a three-to-one specific to nonspecific ratio --
     1

I'm making a leap here perhaps -- would not be something
      2

to dance in the streets about in terms of saying this
     3

is a very specific agent.
     4

       5         And we are talking about receptor imaging

here, 6and I just don't see it.              I mean, am I missing

something?
     7

       8         DR. JONES:       Well, I'm in agreement with

your concern.
      9              The ability to distinguish acute versus

chronic, chronic wasn't looked at, and it remains a
    10

question, and certainly for a receptor based agent, you
    11

having that experience could speak to it better than
    12

I, but I had hoped for a better than three-to-one ratio.
     13

      14         So I don't disagree with you.

      15         CHAIRPERSON RAMSEY:          Are there any other

questions?
    16

      17         Dr. Ponto.

      18         DR. PONTO:       Yes.   I have some questions on

the final formulation.
    19                            It's my understanding that

there are three peptides in the final formulation.
    20

There is the active agent, and there's the P1007 and
    21

the P1008.
    22

      23         In the briefing documents it said that

those other two peptides do not label to Technetium or
    24

are not labeled by Technetium.
     25                                   Am I understanding that

there are the three peptides in the final product?
    26                                                       And

      evidence is there that Technetium does not label
what 27
to any of the others?
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
       1         DR. LOVE:          That's somewhat a proprietary

chemistry question.
     2                            I have to ask first the sponsor

if they want to address it first maybe.
     3

       4         MR. PIPER:          There is on the agenda a one

hour time for a closed session to address issues like
     5

this because they are getting into proprietary
     6

chemistry issues.
     7                      So we'd prefer to do that.

       8         CHAIRPERSON RAMSEY:             Please identify

yourself.
     9

      10         DR. HAGGERTY:          Bob Haggerty, Diatide.

      11         From the question posed earlier in

relation to --
    12

      13         CHAIRPERSON RAMSEY:             Could you please

activate the microphone?
    14

      15         DR. HAGGERTY:          Bob Haggerty, Diatide.

      16         In the question posed earlier for the

regulatory standpoint of two adequate and well
    17

controlled pivotal trials, I did want to just emphasize
    18

     there is a precedent within the agency of accepting
that 19

one adequate study for this type of approval in the
    20

past.
    21

      22         CHAIRPERSON RAMSEY:             Any other questions

or comments from the Committee members?
    23

      24         DR. D'AGOSTINO:           Could I ask one?

      25         CHAIRPERSON RAMSEY:             Dr. D'Agostino.

      26         DR. D'AGOSTINO:           I'd like to ask the FDA

the question that came up this morning in terms of
    27

shifting to the Hamilton.
    28                                  If you perform your study,
                              S A G CORP.
202/797-2525   Washington, D.C.       Fax: 202/797-2525
you run your analysis, and then you're overwhelmingly
     1

hit with the fact that the positive rates are too high
     2

by your blinded readers, and then you move to a new set
     3

of blinded readers, a new procedure for getting
     4

results, such as Hamilton, how do you respond to that?
     5

       6         I think that a good part of the sponsor's

argument rests on the fact of shifting to the Hamilton
     7

reading in terms of the efficacy of the studies.
     8                                                       Is

       a
there 9 response?          Is it because we found that in A that

the positive rates get so low that it throws a question
     10

      shifting or what's your response to it?
into 11

      12         DR. SOBHAN:      If I understood it clearly,

you are saying that how do we accept Study B or how --
     13

      14         DR. D'AGOSTINO:       How do you respond to the

     or
move 15 the concern that the data just on the face isn't

correct with the blinded readers in Study B?
    16                                                 That was

found after they did the analysis and so forth, but
    17

nonetheless it was found, and they shifted to the
    18

Hamilton.
    19          Now they have a new set of results which on

the surface look good, and if you never saw the blinded
     20

CV readings, you'd be probably impressed by it.
    21

      22         So how do you respond?

      23         DR. SOBHAN:      We have not seen until the

submission came in what was the -- how they have done
    24

Hamilton or what is the result of Hamilton.
    25                                                  As for

protocol, what we had is blind and unblinded read.
    26

      27         Now, when the submission came in, when we
saw Hamilton read, we knew the fact that it was done
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
retrospectively.
     1                    So ideally we would like to still see

our results as per protocol, not the retrospective
     2

analysis after the facts are known.
     3

       4         So our position -- I think my position, and

I'm not speaking for everybody else; my position is in
     5

the study design, in the clinical studies you should
     6

not rely on the results after the facts are known.
     7

       8         DR. D'AGOSTINO:         Thank you.

       9         DR. ROHDE:        Yes, Chuck Rohde from Johns

Hopkins.
    10

      11         If we pushed that line to its ultimate, if

the sponsor were to run another trial, then it's well
    12

known that there is no way they could ever show that
    13

      product worked because the statistical
this 14

significance is .06 now.
    15                               You run another study.    It

can't be reduced.
    16                      It's .06 plus the probability, et

cetera, et cetera, et cetera.
    17

      18         So it seems to me that the real issue is

the scientific issue.
    19                            Is the Hamilton study the one

to go from the standpoint of it correctly assesses
    20

what's going on with this group of patients?
    21                                                      That's

the issue, not these little statistical arguments about
     22

      or
this 23 that.

      24         So I think the rest of the Committee, if

      could focus on the scientific aspect, is that a
they 25

      study, you know, then we could not worry about
good 26

whether they're one sided or two sided or whatever.
    27

      28         CHAIRPERSON RAMSEY:           Is it four carat or
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
14 carat, huh?
     1                 That's really where we're at.

       2         Dr. Hammes.

       3         DR. HAMMES:      Richard Hammes.

       4         A comment on Dr. Links' point about ratios.

The parameters I've seen, it's clear to me that we're
     5

talking about much more than a three-to-one
     6

target/nontarget radio, and I would submit that it's
     7

probably related to the concentration of platelets in
     8

the clot also.
     9                 So we're dealing with the product of

the two in these concentrations, plus the increased
    10

affinity because those images understandably -- I'm
    11

      those are the best ones they had, but they are a
sure 12

      higher ratio.
much 13

      14         CHAIRPERSON RAMSEY:         Dr. D'Agostino.

      15         DR. D'AGOSTINO:       Just I think the sponsor

      have a way out if we recommend another study
does 16

because they don't have to combine the studies.
    17                                                         They

could do two positive studies.
    18                                      They could have

negative studies mixed in their pool; isn't that
    19

correct?
    20         It has to be two positive studies?

      21         CHAIRPERSON RAMSEY:         Any other questions

or comments from the Committee?
    22

      23         We will break for lunch.             I'm looking at

my watch, and it's 12:22.
    24                              We will not have a closed

session this afternoon.
    25                            We will go directly to the

discussion of the Committee questions for
    26

consideration, and that will begin, and I'd like to ask
    27

-- and, Mr. Madoo, you can make any corrections here
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       if
to me 1 you'd like -- that everybody be back at 1:25.

Thus we will begin promptly at 1:30 for this afternoon's
      2

deliberations.
     3

       4         Question?

       5         DR. PONTO:       I did have a question that I

was told they had the answer to.
     6

       7         DR. AUGUST:       Yes, can we bring that

question on?
     8

       9         CHAIRPERSON RAMSEY:          Mr. Madoo is asking

if you feel that's absolutely essential.
    10

      11         DR. PONTO:       I guess not.

      12         CHAIRPERSON RAMSEY:          Okay.    Mr. Madoo?

      13         MR. MADOO:       Yes. Just by way of noting,

there apparently is a buffet downstairs for 5.95, and
    14

      called the "soup and salad opera."
it's 15

      16         (Whereupon, at 12:20 p.m., the meeting was

recessed for lunch, to reconvene at 1:25 p.m., the same
    17

day.)
    18

      19

      20

      21

      22

      23

      24

      25

      26

      27

      28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
       1

       2

       3

       4

       5

       6

       7

       8

       9

      10

      11

      12




                              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                                                  (1:26 p.m.)

       3         CHAIRPERSON RAMSEY:         I'd like everyone to

take their seats if we could, please, and the Committee
      4

members to come to the table so we can begin promptly
     5

at 1:30.
     6

       7         We plan no moving through this cautiously,

but as efficiently as possible.
     8                                       We'll move forward.

I don't think it's quite two minutes.
      9                                               I guess we can't

really start early, right?
    10                              It's like closing the plane

      before the scheduled time.
door 11

      12         But there's a couple of things I want to

      about.
talk 13            We obviously have some important things,

including a vote at the end of the meeting, and there's
    14

a few things after talking to Dr. Love and others that
     15

we would like to clear up.
    16

      17         One, we want to make sure that everybody,

the statisticians, in particular, are in agreement or
    18

had their questions answered regarding the lower
    19

boundary of 60 percent or the other higher boundary for
    20

the inclusion or the final conclusion on the study.
     21                                                            So

we may want to reopen that again if people didn't
    22

understand exactly what that was.
    23

      24         And also to the manufacturers and to the

panelists why we accept the Hamilton data and analysis
    25

after the fact, if you will, if that's the right way
    26

of putting it, when this is not the way we usually
    27

approach a process.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         In other words, we looked at it.            We didn't

like it.
     2         We looked at it again, and if that's okay,

then why we're willing to accept that.
     3

       4         So I'd like to come back to those issues

later 5on during the course of the meeting.                 So you can

be thinking about it, please, while we go forward.
     6

       7         And now, Mr. Madoo, we are at the committee

consideration of agency proposed questions, and these
     8

questions are available on the desk if anybody didn't
     9

get a copy of them.
    10

      11         And, Dr. Love, you wanted to make a few

statements before we begin or no?
    12

      13         I think Mr. Madoo is going to give me the

charge, which is to read this into the record; is that
    14

correct?
    15

      16         MR. MADOO:       Yes, or if you would like, Dr.

Ramsey, at your discretion you might want to build up
    17

      particular question by having some more
to a 18

generalized discussion before we hard copy anything.
    19

As you wish.
    20

      21         CHAIRPERSON RAMSEY:          Well, I'll leave it

up to the discretion the Committee.
    22                                                 These questions

      are listed here, one through six, will have to be
that 23

      at
read 24 some point into the record, which I will do.

      25         Do we want to go right to those questions

or do we want to have other discussion prior to that?
    26

And are the statisticians here?
    27

      28         Perhaps, Dr. Love, maybe you can formulate
                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
the question again so that we can address it to
     1

everybody's satisfaction, if you would be willing to
     2

do that.
     3

       4         DR. LOVE:        Okay.    I suppose one of our

questions asked, after we finished the break I just
     5

wanted personally to be sure that the responses that
     6

Dr. Welch and Sobhan were giving were addressing the
     7

concerns of the panel statisticians.
     8                                                  From my hearing

of the conversation, I'm not sure we all walked away
     9

      the same feelings.
with 10                            That's the main question.

      11         So I'm just wondering from your

perspective has the question been answered.
    12

      13         DR. D'AGOSTINO:          Well, Ralph D'Agostino.

      14         I think in terms of the lower bound of the

confidence interval that I think it's fine.
    15                                                         The

protocol said one sided, and the 60 percent is the lower
    16

limit, I think, is clear, and I don't think it changes
    17

whichever you do.
    18

      19         With the Hamilton, it's quite a different

discussion, and let me just say what I'm worried about
    20

is that the Hamilton might be the ideal way of viewing
     21

these procedures.
    22                      The problem is it was not the

protocol specified.
    23

      24         And what we have before is, we have the

discussion is the Hamilton right to look at.
    25                                                          That's

number one, but then we also have how do we interpret
    26

our data that's before us because it's retrospective
    27

and it came out to be positive.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         But all the retrospective analyses that

aren't positive never made it here.
     2                                                 I mean it's

clearly here, and it clearly looks good.
     3                                                    If it didn't

look good, they never would have presented it.
     4

       5         So from a statistics point of view, there

really is no way, is what I was trying to get; there
     6

really is no way.
     7                     It's not a statistical adjustment.

There 8really is no way that I can see that one can talk

about 9making a statistical adjustment for the use of

the Hamilton.
    10

      11         It's a retrospective analysis.             We know we

      see retrospective analysis when they turn out to
only 12

be positive.
    13              So we have to grapple with the issue of

do we think the Hamilton is the right thing to do and
    14

do we want to ignore the fact that we can't put a
    15

statistics judgment really on this second study, and
    16

I think that's the way I would present it to the
    17

Committee in terms of the issues.
    18

      19         CHAIRPERSON RAMSEY:          Are there any other

comments?
    20

      21         DR. ROHDE:       Yeah, I would.          Chuck Rohde.

      22         I would agree with Ralph's comment.                 You

could look at the analysis of the second set of data
    23

      slightly different way.
in a 24                                   You could say what we

really had here are two response variables.
    25

      26         We had one response variable comparing to

the completely blind, and then we have another response
     27

variable comparing to the Hamilton data.
    28                                                      Either way
                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
there's still an issue of the fact that one was done
     1

after 2the fact and wasn't in the protocol.

       3         And the question is:             do the results

constitute good enough science to overcome that?
     4                                                             And

that's a matter that I can't judge as a statistician,
     5

      think the panel ought to satisfy itself that that
but I 6

science is good enough.
     7                              Then the decision is made on

that basis.
     8

       9         DR. LOVE:         I think those certainly were

the concerns from the review team's perspective as
    10

well, and the reason for a lot of discussion today.
    11

      12         CHAIRPERSON RAMSEY:            Thanks, Dr. Love.

      13         Well, you all can be thinking about that,

and while you do that then I think, Mr. Madoo, would
    14

it be appropriate at this time to read this into the
    15

record?
    16

      17         MR. MADOO:         Certainly.

      18         CHAIRPERSON RAMSEY:            Okay.    This, again,

are the questions, and the reason I'm reading them is
    19

     to
just 20 get them into the permanent record.                 So you'll

forgive me for that.
    21

      22         I'm going to skip the first paragraph and

start here with just number one.
    23

      24         Proof of concept relationship to the

proposed indication.
    25                            Implicit in AcuTect's proposed

use to detect acute venous thrombosis is the need for
    26

      oh, for -- thank you -- apcitide bind to activated
to --27
platelets and to preferentially distinguish activated
    28

                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
platelets from other cross-reacting binding sites in
     1

the endothelium.
     2                    Such distinctions affect AcuTect's

potential to affect the differential diagnosis of acute
     3

thrombosis, chronic thrombosis, phlebitis, and
     4

thrombophlebitis.
     5

       6         Also, activated platelets are found in

acute 7thrombosis and in the inflammatory process of

phlebitis.
     8

       9         (a)     Is there sufficient mechanism of

action information to confirm that apcitide binds
    10

preferentially to the alpha 2, beta 3 receptor, and that
    11

it can distinguish activated platelets from
    12

vitronectin receptors in the endothelium?
    13

      14         (b)     Is there sufficient mechanism of

action information to support the potential to
    15

differentiate acute thrombosis and acute phlebitis?
    16

      17         Two, AcuTect image technical features.

      18         The blinded reader instructions

identified specific image features found in the AcuTect
    19

positive images.
    20                    The case report forms recorded the

information if the images were positive.
    21                                                  Similar

information on the features of the negative images were
    22

not recorded.
    23

      24         Question:        Is there sufficient

information to describe the image features that can
    25

distinguish positive and negative results for acute
    26

venous thrombosis?
    27

      28         Three, standard of truth and efficacy
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
results.
     1

       2         The pivotal Phase III trials are designed

as agreement studies.
     3                             An external standard of truth,

example given, histopathology, is not available.
     4

Therefore, the assessment of the agreement depends upon
     5

the comparator imaging study and, as such, it is
     6

important for the results to be blinded.
     7

       8         (a)     Contrast venography results provided

the reference diagnosis.
     9                                 Contrast venography

interpretations are influenced by the reader's
    10

approach or similarity of the criteria used.
    11                                                      As such,

the results of the contract venography and the results
     12

of the primary outcome variable are dependent on which
     13

blinded read is used to determine the reference
    14

diagnosis.
    15

      16         The prospectively planned blinded read

preserves the independence of the two pivotal trials
    17

(280-32A and 32B).
    18                       The Hamilton read was

retrospectively performed after the original study
    19

results were known.
    20                            Also, the Hamilton read is not

independent across both studies.
    21                                           Neither blinded read

of the contrast venograms used prospectively
    22

standardized criteria to interpret the findings.
    23

      24         Question:          Which blinded read do you

recommend should be used to determine the contrast
    25

venography results, i.e., the prospectively planned
    26

blinded read or the Hamilton retrospective blinded read
    27

or neither?
    28

                              S A G CORP.
202/797-2525   Washington, D.C.       Fax: 202/797-2525
       1         (b)     As an agreement study, the target of

a 60 percent lower confidence bound reflects agreement
      2

of AcuTect with either positive or negative contrast
     3

venography results.
     4                        The trials do not have an external

standard of truth.
     5                       Therefore, a true sensitivity and

specificity assessment is not possible.
     6

       7         Similarly, an assessment of true false

positive and false negatives is not possible.
     8

Consequently, the implied clinical use as a screening
     9

modality, adjunct, alternative, or replacement for
    10

contrast venography is not clear.
    11

      12         Question:        Is there sufficient

information from the agreement of AcuTect and contrast
    13

venography results to develop labeling recommendations
    14

for clinical use?
    15

      16         (c)     Given the above considerations,

please respond to the following:
    17

      18         Number one, do you recommend accepting

Study 280-32A as one of the two pivotal studies to
    19

demonstrate the efficacy of AcuTect for scintigraphic
    20

imaging to detect acute venous thrombosis?
    21

      22         Two, do you recommend accepting Study

280-32B as one of the two pivotal studies to demonstrate
    23

the efficacy of AcuTect for scintigraphic imaging to
    24

detect acute venous thrombosis?
    25

      26         Four, safety.        For patients who received

the proposed for market formulation, the database
    27

provides the results of adverse event reporting in at
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
least 1632 patients up to three hours and up to 169

patients up to 24 hours.
     2                              It does not contain data on

creatinine or liver enzymes at the time points when
     3

changes are apt to be detected (if they occur).
     4                                                          The

in vitro data suggest that apcitide binding can inhibit
      5

platelet aggregation.
     6                            The potential clinical

manifestations were not tested with in vivo bleeding
     7

time measurements.
     8

       9         Question:        Is there sufficient

information to support the safety and reasonable
    10

labeling of AcuTect?
    11

      12         Approvability, Part 6.             In reference to

the considered information, please address the
    13

following.
    14           I think that's actually Part 5, but that's

okay.
    15

      16         (a)     Do you recommend AcuTect as

approvable for the scintigraphic imaging of acute
    17

venous thrombosis?
    18

      19         Question:        is there any other indication

      you recommend?
that 20

      21         Question:        If you do not recommend AcuTect

as approvable, are there other studies or trial designs
     22

     you
that 23 would recommend be completed before approval?

      24         Question:        If you recommend AcuTect as

approvable, are there other studies for efficacy or for
    25

safety that you would recommend as a Phase IV
    26

commitment?
    27

      28         MR. MADOO:        If I may interject now, when
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
Committee members are voting, the presumption is that
     1

your options are a vote for yes, a vote for no, or an
     2

abstain to any given question.
     3

       4         Thank you.

       5         CHAIRPERSON RAMSEY:           Dr. Love, could you

please repeat again the meaning of "approvable" as
     6

opposed to "approved"?
     7

       8         DR. LOVE:        In general term, approvable is

just before we do a final approval action.
     9                                                    We usually

interpret that to mean there is sufficient information
    10

in the application that we feel that any particular
    11

issue on which you're voting, approvable do not have
    12

to change if we need addition information.
    13

      14         The main reading for wording it as such

though is there may also be other outstanding issues
    15

maybe from CMC or other concerns that have to be sorted
    16

out. 17So the full direct approval decision may not be

made.
    18

      19         CHAIRPERSON RAMSEY:           Can anyone ask a

question?
    20

      21         MR. MADOO:       They're free to ask a question

as they see fit.
    22

      23         CHAIRPERSON RAMSEY:           Okay.    Panel

members, we are at freedom here.
    24                                           Dr. D'Agostino.

      25         DR. D'AGOSTINO:         Just for clarification,

are the consultants voting?
    26

      27         MR. MADOO:        Yes, yes, everyone around the
table, all 12 of you are voting.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         CHAIRPERSON RAMSEY:           Dr. Links.

       2         DR. LINKS:        Another process oriented

question.
     3          Are we going to just go through these

questions one by one and see if we're satisfied?
     4                                                          Are

we actually voting on each question or only the ultimate
      5

one dealing with approvability?
     6

       7         MR. MADOO:        I'll defer that response to

Dr. Love.
     8          Do you desire a discrete vote on each

question or do you desire a vote simply on the
     9

approvability question?
    10                             We could formulate Committee

consensus on the other questions.
    11

      12         DR. LOVE:        Right.    I was going to say

consensus would be okay.
    13                               Certainly if the Committee

wants to do that, that's fine.
    14                                       It doesn't have to be

a vote as long as we can have enough information to see
     15

how all the different issues are being addressed.
     16                                                        That

would be helpful to us.
    17

      18         CHAIRPERSON RAMSEY:           Dr. Ponto?

      19         DR. PONTO:       Dr. Love, we have in the past

      we've had applications before us had much more
when 20

information with respect to what the labeling is going
    21

to look like than we have on this one.
    22                                                  Part of our

approvability decision would be based on what the
    23

labeling is going to be.
    24

      25         For instance, the imaging protocol that

they're recommending is not explicitly stated.
    26                                                       Things

about the formulation are not explicitly stated.
    27

      28         So are we in a situation where we would
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
recommend that you then use your judgment with respect
     1

to these issues?
     2

       3         DR. LOVE:        Right, for labeling, yes.

       4         CHAIRPERSON RAMSEY:           Thank you, Dr.

Ponto.
     5     That was a good question.

       6         Any other comments or questions from the

Committee?
     7

       8         And then I guess I would propose that we

kind of look maybe a 1(a) and 1(b), the little actual
     9

questions and develop a consensus, and then after we
    10

do that move through all of these, and then move to the
     11

final questions.
    12                    Does that sound reasonable?

      13         All right.        Let's go then in order since

that's the way they're printed on the paper.
    14                                                    One (a),

is there sufficient mechanism of action information to
     15

confirm that apcitide binds preferentially to the A2/B3
    16

receptor and that it can distinguish activated
    17

platelets from vitronectin receptors in the
    18

endothelium?
    19

      20         Dr. Links.

      21         DR. LINKS:       At the risk of going overboard

on this acute versus chronic and acute versus anything
     22

      issue, I would like to pose a question.
else 23

      24         I like the way the introductory paragraph

under number one states it because it's much broader
    25

      the questions (a) and (b) ask, and it seems to me
than 26

      the real question is what would be needed as
that 27
evidence to say that this is an acute VT imaging agent,
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
and presumably the questions under (a) and (b) would
     1

be part of what's needed as evidence, but what I'm
     2

throwing open to everyone is are (a) and (b) sufficient
     3

evidence to label this an acute VT agent.
     4

       5         In other words, as an example, there's

certainly no data, clinical data, shown that in
     6

patients who have chronic DVT but no acute VT that the
     7

imaging results are negative.
     8                                     That's an example of

some additional evidence that might further the claim
     9

      this is an acute VT imaging agent.
that 10

      11         CHAIRPERSON RAMSEY:          Dr. Konstam.

      12         DR. KONSTAM:       But I'm not sure how

important that is because if, in fact, you agree that
    13

      accurate in the setting of detecting acute
it's 14

thrombus, then the issue is not -- I mean, interpreting
    15

it that way, the issue is not before us, what happens
    16

in chronic.
    17

      18         I mean we're going to get into questions

of what are we talking about by acute and chronic, but
     19

I guess they're asking for an indication for acute.
     20                                                       I

don't know if Dr. Love wants to comment on this, but
    21

it may not be -- from that perspective, I don't think
    22

      important whether or not it works in chronic or
it's 23

how it works in chronic.
    24

      25         DR. LINKS:       Just a question of

clarification then.
    26                        My own personal confusion is that

whenever you talk about the specificity of an imaging
    27

agent, you always have to say specificity for what, and
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
if it's acute VT, meaning it's VT versus no VT, that's
     1

different than it's specific for acute VT.
     2

       3         And since the indication is for acute VT,

I just want clarification on what we need as sufficient
      4

evidence.
     5

       6         CHAIRPERSON RAMSEY:           Would it be

appropriate to ask the manufacturers at this time, Mr.
     7

Madoo, for comments if they'd care to make any?
     8

       9         MR. MADOO:        If the Committee deems it

constructive.
    10

      11         CHAIRPERSON RAMSEY:           Yes.     I see yeses.

      12         Would the manufacturers care to make any

comments?
    13          Did they understand the question?

      14         DR. DEAN:        Yes, I think we understand the

question.
    15          It's a two-part question here, and I think

one relates to biology and one relates to the clinical
     16

manifestations, but I'm going to defer to the
    17

clinician, and I'm going to ask Dr. Ginsberg again if
    18

he can make a comment to that effect.
    19

      20         DR. GINSBERG:        Being a clinician, I would

ideally design a clinical trial to address that issue,
    21

and I think the way it would be designed would probably
     22

be as a randomized trial, but as one arm what you would
     23

do would be to do the AcuTect and see whether or not,
    24

if the results are positive, you can --
     25                                                 and these are

in patients with previous disease -- if the results are
     26

positive, then you would do a confirmatory venogram to
    27

      that there's acute thrombus there.
show 28
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1          If the results are negative, I would simply

withhold anticoagulant therapy and insure that those
     2

patients did well in clinical follow-up.
     3

       4          So I think that the clinical trial is

reasonably straightforward.
     5                                   The biology is a much

more problematic issue, but I would in a sense argue
     6

that the biology is secondary if the clinical trial
     7

dictates results that are consistent with good clinical
     8

practice.
     9

      10          DR. KONSTAM:       Can I follow up on that

though?
    11         I don't understand --

      12          CHAIRPERSON RAMSEY:          Dr. Konstam.

      13          DR. KONSTAM:       It's Mark Konstam.

      14          I guess I'm not sure how you're defining

acute venous thrombus.
    15                             I haven't heard anything that

clarifies that you have the ability clearly to make that
    16

distinction by venography, and there was variability
    17

in the duration of symptoms.
    18

      19          So, I mean, I think this is a very -- I mean,

I don't see how.
    20                     On what basis are you going to

distinguish acute versus chronic?
    21

      22          DR. GINSBERG:       Well, it's tricky in the

sense that you don't have a reference standard, but what
    23

we do think is the following:
     24                                   that if you have a fresh

intraluminal filling defect, that that's diagnostic of
    25

venous thrombosis.
    26

      27          So if you see that finding in the majority
of patients who have a positive AcuTect, then I think
    28

                               S A G CORP.
202/797-2525    Washington, D.C.    Fax: 202/797-2525
it's reasonable to assume that the positive predictive
     1

value 2is high and that this test is diagnostic of acute

recurrent venous thrombosis.
     3

       4         On the other hand, if the test result is

negative and the patient does fine in follow-up, I think
     5

as a corollary to that, you can assume that the negative
      6

predictive value for clinical events is high enough to
     7

avoid 8anticoagulant therapy.

       9         So it's a mixed trial.            You're looking at

an anatomical test and a clinical outcome test.
    10

      11         DR. LINKS:        Dr. Links again.

      12         That trial wasn't done, and that's the

conundrum.
    13           So help us out of the conundrum of the

specific indication that Diatide wishes to label the
    14

product with, and what evidence do you need, and do they
    15

      that evidence?
have 16

      17         DR. DEAN:        Don't go away because I may need

you. 18

      19         CHAIRPERSON RAMSEY:           Dr. Dean.

      20         DR. DEAN:        Yes.   If you recall, in our

discussions with the experts, this is obviously the
    21

conundrum, and everybody spots it right away.
    22                                                       Really

the best you can do here based on their advice to us
    23

is to take a patient for which this is the first time
    24

the disease has occurred in that patient, giving it a
    25

     high likelihood that it's acute, that there never
very 26

was a previously existing condition, and then have the
     27

entry criteria narrowed to a short window of time, like
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
ten days.
     1

       2         So that's how we framed the trial to be able

to capture the acute condition as best as possible,
     3

given 4all the difficulties associated with evaluating

a physiological type test.
     5

       6         Would you concur?

       7         DR. GINSBERG:        Yeah, I mean, I think that

makes 8good sense.          It removes the confound of --

       9         CHAIRPERSON RAMSEY:           Dr. Ginsberg.

      10         DR. GINSBERG:        -- previous disease and the

misinterpretation of previous disease.
    11

      12         I'm sorry.

      13         DR. DEAN:        She was just identifying you.

      14         CHAIRPERSON RAMSEY:           I'm just putting it

in the record so that we know who's giving these
    15

responses.
    16           Sorry.

      17         MR. MADOO:       I'd also like to note for the

record that the sponsor will be not voting, and they
    18

      not be contributing to the Committee consensus.
will 19

      20         (Laughter.)

      21         DR. DEAN:        thank you.

      22         CHAIRPERSON RAMSEY:           Did we answer (b)?

Is there sufficient mechanism of action information to
     23

support the potential to differentiate acute
    24

thrombosis and acute phlebitis?
    25

      26         Dr. Hammes.

      27         DR. HAMMES:        Richard Hammes.
      28         I think it's quite clear in terms of (a)
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
that there's strong data that the apcitide binds to the
      1

receptor in question.
     2                            Given though that in the

inflammatory process of phlebitis, as it states,
     3

platelets are also found, and if these platelets are
     4

activated, they also will bind with apcitide.
     5

       6         And I think the high incidence of false

positives that we've seen in the low specificity, it
     7

could 8well be a direct result of this.                 So given that,

I don't think you can say that it doesn't go to
     9

endothelium, which gets us right into Part (b).
    10

      11         I see no evidence that you can different

acute phlebitis from acute thrombosis in light of that
    12

data.
    13

      14         CHAIRPERSON RAMSEY:           Dr. Links.

      15         DR. LINKS:        A follow-up question to the

FDA. 16With respect to the labeling, I think that the

Diatide answers we just heard focused on sensitivity,
    17

and the comment that was just raised focused on
    18

specificity, and my question is:
    19                                           in guiding us for

approvability of an indication, is it an issue of both
    20

sensitivity and specificity?
    21

      22         And by specificity I don't mean a normal

group, but rather a group with a host of other diseases
    23

      may yield false positives in the context of the
that 24

specific indication, or are we looking at sensitivity
    25

and normalcy rate, so to speak?
    26

      27         DR. LOVE:        Some of that probably deals
      another set of the questions which talks about
with 28
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
labeling for proposed use.
     1                                    If you thought this was a

screening agent, you might be more concerned about one
     2

aspect.
     3         If you thought this was a replacement or an

alternative to contrast venography, then other issues
     4

become important.
     5                      If you think it's an adjunct, other

issues are important.
     6

       7          So I think that's all of what we'd like to

hear you discuss and think about when it comes to
     8

labeling other product.
     9                                Certainly we can label in

pharmacodynamic sections of the clinical pharmacology
    10

portion of a label.
    11                             We can put cross-reacting

information there, but it would also affect perhaps
    12

      of
some 13 the indications, and we have to think about

that.
    14

      15          CHAIRPERSON RAMSEY:             Comments?

      16          DR. CHOYKE:          Pete Choyke.

      17          I'd just like to clarify at least on (b).

You know, this isn't exactly pertaining to the
    18

mechanism per se, but certainly from clinically the
    19

images that were shown, you should be able to
    20

differentiate between acute thrombosis and acute and
    21

the missing word is "superficial phlebitis" because the
     22

distribution is going to be quite different.
    23

      24          So, you know, it doesn't get to the issue

of mechanism, but for me that's not so key because from
     25

a practical point of view, I think you will be able to
     26

differentiate that.
    27

      28          CHAIRPERSON RAMSEY:             That was the message
                               S A G CORP.
202/797-2525    Washington, D.C.       Fax: 202/797-2525
I got 1as well.

       2         Any other comments?          Dr. Kasper.

       3         DR. KASPER:      Yes, I had wanted to comment

that I don't think it's all that important to
     4

distinguish acute thrombosis and acute phlebitis.
     5                                                          I'm

not sure that there's any way that we can.
     6                                                 Even if one

has a 7normal venogram, that doesn't mean that there

isn't 8a little layer of thrombosis happening, and I

think 9that it is not bad that some phlebitis were

falsely positively diagnosed as thrombosis because the
    10

clinical reaction to that would probably be beneficial.
    11

      12         I don't think we can tell anyway.

      13         CHAIRPERSON RAMSEY:          Dr. Links.

      14         DR. LINKS:       In that regard, from a

clinical point of view could someone or a group of
    15

people please state the relevant clinical distinction
    16

and let's see what the evidence is for this agent based
     17

on that relevant clinical distinction or
    18

differentiation, whatever it is?
    19

      20         CHAIRPERSON RAMSEY:          Dr. Jahnke.

      21         DR. JAHNKE:       Yes, Dr. Jahnke.

      22         Well, certainly the treatment is

different.
    23           Phlebitis is not necessarily treated with

anticoagulation if there is no thrombosis.
    24                                                     So the

clinical treatment differs fundamentally.
    25

      26         If you feel that there is deep venous

thrombosis, heparin followed by cumidization
    27

(phonetic) for treatment.
    28                               If you feel it's an acute
                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
phlebitis, then rest, elevation, warm packs applied,
     1

et cetera.
     2           Perhaps anti-inflammatory agents are

indicated.
     3

       4         So there is a clinical -- I feel not as a

clinician; as a radiologist -- but there is a clinical
     5

difference in the treatment of those two entities.
     6

       7         CHAIRPERSON RAMSEY:             I think we heard that

they --
     8

       9         DR. JAHNKE:          Not that they -- they don't

usually coexist.
    10

      11         CHAIRPERSON RAMSEY:             Ruth Ramsey.

      12         And I heard that they can -- didn't we hear,

I should say, that they can all look the same
    13

clinically?
    14             Thus, the idea is this would attempt to

differentiate between them, not that I'm taking sides
    15

here.
    16

      17         Any other comments?

      18         (No response.)

      19         CHAIRPERSON RAMSEY:             Okay.    Let's move to

two. 20The question:              is there sufficient information

to describe the imaging features that can distinguish
    21

positive and negative results for acute venous
    22

thrombosis?
    23

      24         There we are.          We just talked about it.

      25         MR. MADOO:         Excuse me, Dr. Ramsey.       So did

we formulate any kind of consensus relative to these
    26

questions or are we going to be satisfied with just
    27

staccato comments or distinct comments?
    28

                              S A G CORP.
202/797-2525   Washington, D.C.       Fax: 202/797-2525
       1         CHAIRPERSON RAMSEY:           Dr. Love?

       2         DR. LOVE:        I guess what I've heard -- I'm

not sure that I hear a true consensus.
      3                                                 What I've heard

is that, yes, everyone seems to agree that there is
     4

preferential binding for the receptor, but there's
     5

still 6a potential for cross-reaction to the

vitronectin, and a difference of opinion on whether
     7

that is or isn't clinically relevant.
     8

       9         I've heard one feeling that it doesn't

matter, and perhaps we can handle it in labeling, and
    10

another opinion saying it does because it would affect
    11

treatment.
    12

      13         So I don't really hear a consensus on (b).

      14         MR. MADOO:       Does that, indeed, reflect the

Committee's position?
    15

      16         I guess no comment would imply that that's

the case?
    17

      18         DR. CHOYKE:        No, I didn't hear the latter

part.
    19

      20         CHAIRPERSON RAMSEY:           Dr. Choyke.

      21         DR. CHOYKE:        That there was some clinical

significance to the vitronectin reception.
    22                                                        I mean,

did anybody say that?
    23                            Because I missed it.

      24         DR. LOVE:        No, I don't mean the

vitronectin.
    25              I'm speaking now -- I'm sorry -- of the

acute phlebitis and acute thrombosis.
    26

      27         DR. CHOYKE:        Oh, oh, I see.
      28         DR. LOVE:        I'm sorry.
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1          DR. AMENDOLA:       This is Dr. Amendola.

       2          I think that if we put in (b) the

superficial phlebitis, my impression is that the agent
     3

does have the potential to differentiate thrombosis
     4

from phlebitis.
     5

       6          MR. MADOO:       So would you care, Dr. Ramsey,

to nutshell the Committee consensus for the record?
     7

       8          CHAIRPERSON RAMSEY:          Well, I would just

repeat what -- I'm sorry to mispronounce your name --
     9

but that there is sufficient data to support the
    10

potential to differentiate acute thrombosis and acute
    11

phlebitis, or to put it another way, is there
    12

sufficient, as they say here, is there sufficient data
    13

to support the potential to differentiate acute
    14

thrombosis and acute phlebitis?
    15

      16          Yes or no?       Do we want to vote on that?

      17          DR. KONSTAM:       Can I just comment on that?

      18          CHAIRPERSON RAMSEY:          Certainly.

      19          DR. KONSTAM:       We're whispering in the

corner.
    20         This is Marv Konstam.

      21          There's a consensus in the corner here that

maybe the question doesn't matter, and at least I'll
    22

      you my reason for thinking it doesn't matter.
give 23

      24          The only thing that really matters here is

if we're going to be able to identify a test that
    25

predicts outcome and that dictates management.
    26                                                       Now,

I think this could matter, the difference between acute
     27

phlebitis and acute thrombosis, but where I think it
    28

                               S A G CORP.
202/797-2525    Washington, D.C.    Fax: 202/797-2525
doesn't matter so much is because there's also the issue
     1

of chronic thrombosis and acute thrombosis.
     2

       3         We're going to get into a significant

degree of variation between the venogram and the scan
     4

results.
     5         We're going to try to presume from that that

that has some implication about outcome, and I think
     6

that's really where the gist of the discussion is going
     7

to lie.
     8

       9         I mean, I think if it were clear, if this

      the question, is it acute phlebitis versus acute
were 10

thrombosis, then maybe we could answer yes, but for me
    11

I don't think it matters.
    12

      13         CHAIRPERSON RAMSEY:           Any other comments?

      14         Dr. August.

      15         DR. AUGUST:        We spent most of the morning

hearing and seeing data that had to do with agreement
    16

and now we're being asked a totally different question.
     17

I don't think we were given enough information to allow
     18

us to answer Part (b) of that, to be perfectly honest
    19

      you, and I'm kind of confused at being asked in
with 20

the first place.
    21

      22         There was no algorithm.                There was no

scheme of positives for this test and negative for that
    23

      that would enable us to make that differential.
test 24

I think it's -- I personally think it's not a fair
    25

question to pose to the Committee at this time.
    26

      27         DR. LOVE:        Certainly we can understand and
accept that.
    28              This question was relating really to the
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
pharmacology information from the sponsor and from the
     1

agency in the beginning.
     2                              This is not from the pivotal

clinical trial.
     3                   It's the baseline information.

       4         CHAIRPERSON RAMSEY:           Do you have enough

response on the Committee?
     5

       6         DR. LOVE:        Yes.

       7         CHAIRPERSON RAMSEY:           Thank you.   You

saved 8me.

       9         Two, is there sufficient information to

describe the image features that can distinguish
    10

positive and negative results for acute venous
    11

thrombosis?
    12

      13         Comments?        Dr. Links.

      14         DR. LINKS:        There's some fantastic

nuclear medicine physicians in the audience who have
    15

experience with this agent.
    16                                   I guess I'd love to hear

     them because I suspect they've imaged patients who
from 17

are not part of the trial who may end up addressing some
     18

of these other issues that we've just been discussing.
     19

      20         So I wonder if Diatide would like to

identify a couple to just give us some very quick
    21

information that goes beyond what we've already had
    22

presented.
    23

      24         CHAIRPERSON RAMSEY:           I'm going to assume

      the Committee that that's all right with the
from 25

Committee to get the responses.
    26

      27         Please.
      28         MR. MADOO:        And, of course, with the
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
proviso that for conflict of interest purposes any
     1

opinions will have to be characterized as being a
     2

Diatide consultant or otherwise.
     3

       4         CHAIRPERSON RAMSEY:          Please identify

yourself.
     5

       6         MR. PIPER:       I'm Chris Piper.

       7         We have Dr. Bob Caretta.              He has been a

clinical investigator on AcuTect studies.
     8

       9         DR. CARETTA:        I'm Dr. Bob Caretta.         I'm

a community practitioner of nuclear medicine, and I've
     10

      involved in the Diatide trial, as well as other
been 11

trials, and I've used I125 and I123 fibrinogen
    12

extensively when they were both available in the late
    13

      and early '80s to look at deep vein thrombosis,
'70s 14

and I think this study is one of the best studies that
     15

we have to meet an unmet need, which is to find something
     16

      will show us an acutely forming thrombus in a
that 17

clinically suspect patient who is at high risk for the
    18

development of DVT and potentially pulmonary emboli.
    19

      20         The studies are relatively easy to ready

for a qualified nuclear medicine physician, and they
    21

require only planar imaging.
    22                                  They don't require three

dimensional or spec'ed imaging.
    23                                        They can be done

relatively quickly, within the first 60 minutes or
    24

sooner with a positive study, and unlike some of the
    25

other agents that I have worked on, i.e., prostacint
    26

(phonetic), the monoclonal antibody for prostate
    27

cancer, and oncosin, the colorectal imaging monoclonal
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
antibody, which require a high degree of training, a
     1

high level of skill, and a significant over read before
      2

you can feel comfortable in interpreting these images,
     3

I find that the AcuTect images are very, very easy to
     4

read. 5

       6         Plus they can be read in patients who are

trauma patients who have casts on their lower
     7

extremities because they have 140 kEV technetium gamma
     8

that comes through the calf.
     9                                       They can be read in

patients who are bandaged who come out of surgery, and
    10

     are certainly an adjunct test, not a replacement,
they 11

for Doppler ultrasound, but very, very useful when the
     12

ultrasound is either equivocal or negative,
    13

particularly in the calf.
    14

      15         We don't do in community practice contrast

venography anymore.
    16                            You heard from Dr. Sostman this

morning that at Duke they did approximately three to
    17

seven venograms out of 700 patients that they studied
    18

or so, and in community practice, no one does
    19

venography.
    20             It's all ultrasound, and we have a way now

of simply, rapidly, and effectively imaging forming
    21

thrombi.
    22

      23         MR. MADOO:          Excuse me, sir.      You're

attending this meeting on behalf of Diatide as a
    24

consultant?
    25

      26         DR. CARETTA:         I am a clinical investigator

for Diatide and here as a consultant, yes, sir.
    27

      28         CHAIRPERSON RAMSEY:             Dr. Ponto.
                              S A G CORP.
202/797-2525   Washington, D.C.       Fax: 202/797-2525
       1         DR. PONTO:        I think we're all in agreement

that there's this tremendous need that has to be met,
     2

and we've talked a lot about the disagreement in
     3

contrast venography, but I go back to the point that
     4

I made this morning.
     5

       6         In looking at the results that are

presented for the AcuTect product, if you look across
     7

the six readers, there's a lot of disagreement,
     8

especially on the B study, but even in the A study, the
     9

number of positive reads ran from 48 to 56.
    10                                                       The number

of positive reads on the B study went from 33 to 78.
    11

      12         Why is there such a disagreement between

the readers for the AcuTect images?
    13                                                  Anybody.

      14         MR. PIPER:        Chris Piper.

      15         We would like to respond to that.             Dr. Rich

Wahl, a consultant to Diatide.
    16

      17         DR. WAHL:        Yeah.   Richard Wahl, professor

of internal medicine and radiology at the University
    18

of Michigan and Director of Nuclear Medicine Section
    19

there.
    20

      21         I've been involved in a lot of trials of

new imaging agents, and I think perhaps what we need
    22

to reflect to are some of the comments Dr. Gottschalk
    23

      earlier this morning.
made 24

      25         Imaging methods and even, in fact,

histopathology are not perfectly reproducible from
    26

individual to individual.
    27                                With pulmonary
angiography, which has been in use probably for at least
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
30 years, the same individual looking at the same
     1

studies had about an 89 percent reproducibility rate.
     2

       3         When two individuals looking at similar

sets of studies were compared, comparability was about
     4

80 percent.
     5             Those are with a study that's been in

practice and in their practice for probably their
     6

entire careers, where they have a lot of experience with
     7

it.    8

       9         The situation here, as I understand it, is

     the
that 10 degree of concordance among readers was around

60 percent, which is lower than the 80 percent seen for
     11

pulmonary angiography, but I think you have to consider
    12

the 60 percent was based on a limited training set, a
    13

finite number of cases, with no prior experience with
    14

the methodology.
    15

      16         So as the readers would read more, it would

be expected that concordance rates and reproducibility
     17

would increase.
    18                   So I don't find the figures that

shocking or surprisingly low.
    19                                    I think that they're

actually pretty good, considering it's a brand new
    20

test, and the readers, even though there were three
    21

readers, there would have been limited experience among
    22

the readers.
    23

      24         So it's a new test, limited experience, a

majority concordance rate, and even with our most
    25

supposedly gold standard tests, we don't have perfect
    26

concordance, as we've seen with venography.
    27

      28         So I think that would be at least an
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
explanation.
     1

       2         DR. KONSTAM:      Marv Konstam.

       3         You know, I guess getting to the question,

I have a lot of trouble with this question because I
     4

don't 5see how we can answer in the affirmative that

there 6is sufficient information to describe the image

features that can distinguish positive and negative
     7

results, you know, until and unless we decide that we
     8

have some positive data and that we have sufficiently
     9

positive data against something that we consider
    10

approximating a gold standard.
    11

      12         And I don't know how -- I think we're going

to have to get to that question because that's going
    13

to be the key.
    14

      15         If the answer to that question is no, then

the answer to this question certainly is going to be
    16

     I
no. 17 think if the answer to that question is yes, then

I guess, yes, the sponsor did come forward with a set
    18

of criteria.
     19             I don't think we can tell, based on what

I understand from the FDA presentations.
     20                                               I'm not sure

we can tell to what extent the readers stuck to those
    21

criteria.
    22

      23         But, you know, if we agreed that we had some

positive results, then I suppose we could accept the
    24

sponsor's set of criteria since they trained their
    25

readers.
    26         That might be possible.

      27         But, I mean, I just think we have to first
decide whether we have a positive set of results or not.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         CHAIRPERSON RAMSEY:         Yes, D'Agostino.

       2         DR. D'AGOSTINO:       I guess I sit here with

some amazement that, you know, in other fields part of
     3

the judgment of whether or not you have a successful
     4

       is
trial 5 whether or not different readers could produce

the same result, and was it a lack of training and so
     6

forth?
     7     But I don't think that from what we've been told

that we have information that can distinguish the
     8

positives from the negatives.
     9                                    I don't think we have

information if you push it too far that we even know
    10

if we have positives, as you're saying, and I would
    11

presume that part of a clinical trial would, in fact,
    12

worry about the agreement of the raters and judge it
    13

on that.
    14

      15         And then once we say we have enough

agreement among the raters, then see where the
    16

positives and negatives differentiate, and I would
    17

think that the answer should be no here based on what
    18

we heard this morning.
    19

      20         CHAIRPERSON RAMSEY:         Dr. Hammes.

      21         DR. HAMMES:      Richard Hammes.

      22         I recall reading in the packet here

something about a region of interest analysis on these
    23

studies, but I didn't hear anything about that today.
    24

Is this some sort of quantitative analysis?
    25                                                     It would

be very helpful if it was, and if it wasn't, could it
    26

be done?
    27

      28         But quantitative data would answer a lot
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
of these.
     1

       2         CHAIRPERSON RAMSEY:           Comments from the

manufacturer?
     3

       4         DR. NICODEMUS:         Yeah, we have -- Dr.

Nicodemus from Diatide -- we have results here from one
     5

of the Phase II studies looking at region of interest
     6

ratios with different doses of radioactivity, and as
     7

you can see, looking at the 20 millicurie dose
     8

reactivity, which is what we are recommending, the
     9

region of interest ratio is 1.6.
    10

      11         I was wondering if Dr. Wahl would like to

comment on a ratio of 1.6 and the clinical significance
    12

of that.
    13

      14         DR. WAHL:        I didn't draw these regions of

interest, but I believe that these represent region of
    15

interest and symmetrical areas with the question of
    16

whether the clot is present or not.
    17

      18         Again, I'm Dr. Wahl from Michigan.

      19         So what you're looking for is between the

      value and 1., basically one.
mean 20                                           So as a physician

     you
when 21 look at these, you look at a number of things.

      22         Dr. Wyland has a lot of experience with

this, but looking only numerically would not be the
    23

typical approach to doing an interpretation of a scan.
    24

     think we know a lot about where clots occur in the
So I 25

legs, and looking to see if the hot spots are linear
    26

and in the expected anatomic location of veins is very
     27

reasonable, for starters.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         Similarly, if you're trying to see deep

venous thrombosis and you're looking at something
     2

superficial on the scan, that that logically wouldn't
     3

be deep venous thrombosis.
     4

       5         So interpretation involves prior

knowledge of disease processes, an then seeing if the
     6

scan pattern is consistent with where you know the
     7

pathophysiology would be expected to occur.
     8

       9         As far as these ratios, the better the

test, the higher the ratio between the affected and
    10

unaffected site.
    11                    So ratios, let's say, mean ratios of

1.3 to 1.4 would probably be useful depending on what
    12

the variance is.
    13                    We can see that the standard error

would be reasonably low.
    14                             A ratio of 1.6 at the 20

millicurie ratio would be a pretty substantial ratio.
    15

      16         As an example, when we do lung perfusion

activity ratios in trying to decide if a lung can be
    17

removed or not, our maximum differences are often in
    18

the 20 to 40 percent range between lungs, but on that
    19

      of
kind 20 data, we make decisions on which lung should

be surgically removed.
    21

      22         So this degree of difference is not

insubstantial, but I would think it would not be the
    23

      thing used to make a diagnosis.
only 24

      25         DR. KONSTAM:      No, I mean the issue of split

      function.
lung 26                I mean, there -- this is Marv Konstam

      mean, there you're looking for a physiologic
-- I 27
difference between two lungs.
    28                                    You're not attempting
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
to make an anatomic diagnosis.
     1

       2         I mean, this, I think, if I understand you

correctly, you're taking to suggest that there might
     3

be some degree of accuracy in making a diagnosis that
     4

there 5is pathology present.              You know, these ratios

seem awfully -- maybe I don't understand them enough,
     6

but they seem awfully low to me.
     7                                            They seem awfully

close 8to unity for ability to say with certainty that

you have or don't have pathology.
     9

      10         DR. WAHL:         Well, I think, as I indicated

earlier, that you wouldn't -- the typical nuclear
    11

physician would not use ratios alone, nor would a
    12

radiologist use a numerical Houndsfield unit in general
    13

to make a diagnosis.
     14                           Some of the visual findings have

to be consistent with the numerical values.
    15

      16         So I would think that these would be

adjunctive to the pattern, but if they are
    17

representative of an asymmetry between sides and the
    18

      that has the abnormal pattern also has 30 to 40
side 19

percent more accounts or 50 percent more accounts than
    20

the other symmetrical region of interest, that would
    21

certainly support there being -- obviously there's
    22

deposition of Technetium there and presumably by the
    23

pathophysiological mechanism.
    24

      25         Just not to belabor it, but other processes

we look at in nuclear medicine all the time, such as
    26

-- well, we do quite commonly sacroiliac joint uptake
    27

ratios looking for sacroiliatis, and the differences
    28

                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
between sides can be in the range of 25 to 30 percent,
     1

and depending on how many counts you get, the confidence
      2

intervals on those can be pretty tied.
     3

       4         DR. KONSTAM:        Are the background

subtracted numbers or not?
     5

       6         DR. WAHL:        I didn't personally draw the

regions, but I'm being told that that is correct, which
     7

would 8be, of course, the appropriate -- well, I think

that would be an appropriate way to do it.
     9

      10         DR. KONSTAM:        Right, but I defer to you or

to other nuclear medicine specialists here, but the
    11

ratios seem low, particularly if your background is
    12

subtracted.
    13

      14         DR. WAHL:        But I would say, again, that

they're not that low relative to other procedures we
    15

do and not interpreted in a vacuum.
    16

      17         CHAIRPERSON RAMSEY:           Dr. Links.

      18         DR. LINKS:       A question of clarification on

      whole issue of variability and how we as a
this 19

Committee want to deal with it.
    20

      21         It seems to me that historically if you

      at
look 22 validation of any new technique where you have

individual reader's data and then a consensus or
    23

aggregate, majority read, however you want to call it,
    24

     in
that 25 the early stages of technique development, the

first introduction of the technique, that the consensus
    26

or aggregate or majority read is always of higher
    27

accuracy than any of the individual readers alone.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         And typically in the literature that's the

numbers on which you initially judge the performance
     2

of the technique, and I'm just wondering here because
     3

this morning there was a specific question about
     4

whether or not we should focus on the individual
     5

reader's performance or the aggregate/majority
     6

performance.
     7

       8         And my own philosophy would say, hey, it's

a new 9technique.         The aggregate performance is must

representative of what, once it disseminates into the
    10

field and there's adequate training and use, what it
    11

      be.
will 12

      13         But I want some guidance.             What should we

be focusing on, individual readers, both performance
    14

and variability amongst them, or concentrate on the
    15

aggregate?
    16

      17         CHAIRPERSON RAMSEY:          I don't want to be

silly, but is that what we're talking about now?
    18                                                            Is

      in
that 19 this?

      20         DR. LINKS:       It's relevant to this because

it has to do with the variability in the AcuTect
    21

interpretation across readers, which has to do with
    22

      the criteria for how to interpret it and the
both 23

reliability of those criteria.
    24

      25         CHAIRPERSON RAMSEY:          Dr. Choyke.

      26         DR. CHOYKE:       Well, you know, my approach

to that would be that it's really the aggregate that's
     27

      important because it averages out all of these
most 28
                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
different points on an ROC curve basically.
     1

       2         But what it points out to me, I mean, when

you see numbers that are 60 percent agreement and that
     3

kind of thing, you know that what was missing here was
     4

a training set, you know, a real 20 patients where you
     5

were given feedback and then you could refine, and you
     6

were told what other people in the group did, and then
     7

you were able to refine your diagnostic criteria and
     8

converge on a number.
     9

      10         You know, clearly, there's got to be

training involved with this like any other test, and
    11

that's the missing element.
    12                                 I mean my impression is

      in
that 13 these tests they were never really given a

training set beyond what we saw and the ability to
    14

really get into what the group felt, you know, was the
    15

right answer.
    16                So that's why you have these

variabilities.
    17

      18         That's why I sort of focused on the

aggregate response.
    19

      20         CHAIRPERSON RAMSEY:         Dr. Kasper and then

Dr. D'Agostino.
    21

      22         DR. KASPER:      I think that we are now on

Roman numeral two, and I'd like to say that this is an
    23

anatomic diagnosis, and many things in radiology and
    24

in pathology depend on the experience of the
    25

radiologist/pathologist doing it.
    26                                         The more they do of

     particular area, perhaps the more experience, the
that 27
better the result is, and we don't really have and we
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
may never really have absolutes that we can use to
     1

describe the image features.
     2

       3         I think we have generalities to describe

the image features, and often that's where we are in
     4

certain radiologic and pathology situations.
     5                                                          So I

don't 6think we need perfection here or anything near

it, but general guidelines.
     7

       8         CHAIRPERSON RAMSEY:          Dr. D'Agostino.

       9         DR. D'AGOSTINO:        I don't know if you're

referring to me, but I certainly asked that question
    10

      morning about how many should we be looking at.
this 11

I expected the answer to come back:
    12                                                 look at the

aggregate and don't bother with the individuals.
    13                                                           I was

      of
kind 14 surprised.           I thought that was going to be an

issue that we could put to rest by that question, but
    15

it turned out that it went the other way.
    16

      17         I think the aggregate is clearly the right

thing to do, but I think the question that's here is
    18

      we
that 19 don't have information.               Even if we take the

aggregate, if there's negatives being stated, my sense
    20

is from the presentations we don't have a lot of
    21

information on the features that made it negative and
    22

so forth.
    23

      24         I think that's the question that's being

asked here, is it not?
    25                            It isn't so much that if we

believe it's positive, we believe it's a negative.
    26

There's information on the positives, but we've run out
    27

of information on the negatives, and that's the type
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
of question I thought we were responding to here.
     1

       2         CHAIRPERSON RAMSEY:           Dr. Jones.

       3         DR. JONES:        I wanted to -- is this live?

       4         I wanted to make comment with regard to

Peter 5Choyke's observation about a training session.

There 6were -- actually the company did have at least

20 cases presented to the trainees, but I don't think
     7

they assessed the trainees' response to see if they were
      8

uniform or if the actual consensus among the trainees
     9

was occurring.
    10                 There was a cadre of at least 20

patients in the training session.
    11

      12         CHAIRPERSON RAMSEY:           Dr. Love, have you?

      13         DR. LOVE:        Well, just commenting on the

      comment from Dr. D'Agostino was part of what's
last 14

behind this question, yes, the lack of the other
    15

information on the negative side and whether or not it's
    16

relevant at this point.
    17

      18         CHAIRPERSON RAMSEY:           I'm not sure we

answered that.
    19                 Did we answer it?          No.   Okay.

      20         Let's move on anyway.            All right.    We'll

go to 3(a).
    21             Now we're back to the Hamilton -- oh,

sorry.
    22

      23         DR. LOVE:        Actually, depending on what

     recommendations are in the long run, just a little
your 24

bit more on this one.
    25

      26         Are we to interpret your comments then as

saying the package insert or training sessions would
    27

be labeled in a manner that was consistent with the
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
information that was given during the training sessions
     1

to identify what is positive on an image, but what would
      2

be a negative finding?
     3                            That's the question.

       4         In other words, you can interpret --

       5         CHAIRPERSON RAMSEY:           I guess if it doesn't

meet the criteria for being positive.
     6

       7         DR. LOVE:        I mean it's the opposite.        It

means 8it isn't there.

       9         CHAIRPERSON RAMSEY:           That's right.

      10         DR. LOVE:        But I just want to make sure

      what I hear you saying is that if the
that 11

recommendations for interpreting positive are
    12

followed, then that would be sufficient to distinguish
    13

a positive or a negative.
    14                                Is that what the Committee

is saying?
    15

      16         CHAIRPERSON RAMSEY:           I would certainly

      other comments.
like 17

      18         CHAIRPERSON RAMSEY:           Dr. August.

      19         DR. CHOYKE:        Well, I heard during that

brief training session that they described the normal
    20

as symmetrical activity in the leg, and that seemed
    21

pretty good to me, and the images showed -- you could
    22

      see the faint glimpse of a femoral vein and the
just 23

other veins.
    24              If that activity was symmetrical, side

to side, that was a negative, and that seemed clear.
    25

      26         DR. LOVE:        Right.    There were four

criteria specified and also criteria for whether or not
    27

you had multiple times, and it's just the question that
     28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
Dr. D'Agostino was mentioning.
     1

       2         We have the information that says you read

it as 3positive if you see these things.                  Case report

forms 4identified information to confirm those items if

it was read as positive.
     5                                 You didn't have similar

information if it's read as negative.
     6

       7         So it's just pressing the point on how you

would 8want to see a package labeled if this is your

recommendation.
     9

      10         CHAIRPERSON RAMSEY:             Dr. August.

      11         DR. AUGUST:          Isn't it axiomatic that

something is negative if it lacks the criteria that make
    12

it positive?
    13              I mean this is not rocket science.

      14         (Laughter.)

      15         DR. KONSTAM:          Right.      This is Marv

Konstam.
    16

      17         And I understood for positivity, unless I

understood wrong, it required that all of the criteria
    18

be met, that is, asymmetric, central, what else?
     19                                                           What

      the other two?
were 20                           Linear.    What was the fourth?

      21         DR. LOVE:         And what you have to do if you

      the gain all the way up.
push 22

      23         DR. KONSTAM:          Oh, on two views, on two

views.
    24

      25         PARTICIPANT:          Two time points.

      26         DR. KONSTAM:          Oh, two time points, at two

      points.
time 27
      28         Okay.      So if it didn't meet any of those,
                              S A G CORP.
202/797-2525   Washington, D.C.       Fax: 202/797-2525
if there was one of those criteria that it didn't meet,
     1

then it's negative, right?
     2                                   Right.

       3         DR. D'AGOSTINO:          Can I ask a question?

       4         CHAIRPERSON RAMSEY:            Yes.

       5         DR. D'AGOSTINO:          There's so much

disagreement it must not be -- maybe it is rocket
     6

science, but not all of the negatives came out the same
     7

way. 8I mean some that were negative by one, declared

positive by another.
     9                            So you can't say that it was

obvious that you had a checklist that made it positive.
    10

There's something that made some people say negative
    11

and others say positive to the same thing.
    12

      13         DR. KONSTAM:         This is Marv Konstam again.

      14         I believe the reason we're struggling is

because these results are so marginal at best.
    15                                                         I mean,

as we get forward, if there were a clear gold standard
     16

and if we were at 90 percent agreement, then we wouldn't
     17

be debating about whether or not the people who were
    18

trained strictly followed the training.
    19                                                   We would know

      were trained.
they 20                      We would know what criteria they

      handed, and we would assume, well, they must have
were 21

followed it because they all got it right.
    22

      23         The reason we're struggling with this is

because a lot of times they didn't get it right, and
    24

we're not sure who got it right.
    25                                            So in that context

I don't think we know what's going on.
    26

      27         CHAIRPERSON RAMSEY:            Okay.    Thank you.
      28         Let's move on to 3(a), back to the Hamilton
                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
data and whether we can look back.
      1                                         I don't want to bias

my statement here, but let's see.
     2                                            I have a note.

       3         Why did we accept the Hamilton data post

hoc, as it were, when it is not the way we usually would
      4

look at data?
     5                And that's the question, and would we

accept then what we did to come to the final conclusion?
     6

       7         Is that what you're trying to --

       8         DR. LOVE:        No.   Depending upon your

results, that's just something I just wanted to make
     9

      we
sure 10 could hear, the issues and thoughts that are

behind your recommendations.
    11

      12         DR. LINKS:       That assumes we do accept it.

      13         DR. LOVE:        No, whichever.        I'm saying

if -- the reason obviously if the -- Biometrics and most
    14

of the offices generally do not -- are generally
    15

concerned about retrospective post hoc analyses, and
    16

generally if we were going to accept that, we'd need
    17

a very clear reason for why in order to make prospective
     18

policy decisions, and that's why I'm asking for
    19

clarification behind your recommendations, whichever
    20

      might be.
they 21

      22         CHAIRPERSON RAMSEY:           Comments?     Dr.

Links.
    23

      24         DR. LINKS:        This is maybe a semantic

quibble, but in an epidemiologic sense whether you
    25

collect the data and then analyze it or analyze it as
    26

you're collecting it, that's not the distinction
    27

between retrospective and prospective.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         So using the Hamilton data may be post hoc,

but it's actually no more retrospective than if you had
      2

used the blind data.
     3                            It's a quibble, but it's an

important point because we're pejoratively labeling
     4

the Hamilton as being retrospective, when in fact it's
     5

not epidemiologically.
     6

       7         My own personal opinion is that even though

I fully understand what Dr. Welch said earlier, it
     8

sounded very logical.
     9                            Philosophically I disagree.         I

don't think -- I think if you're comparing something,
    10

you compare to the best thing you have to compare it
    11

to, and there's no reason in the world why Diatide, in
     12

my opinion, should be penalized for variability in
    13

venography interpretation.
    14                                   You go to the best

interpreters you can find, and if there's only one group
    15

you can trust, use them for both studies A and B.
    16

      17         And so my own personal feeling is that in

the context of independence, that A and B are adequately
     18

independent, and as I say, I don't think they're
    19

retrospective.
    20

      21         CHAIRPERSON RAMSEY:            Dr. Konstam.

      22         DR. KONSTAM:         Marv Konstam.

      23         You know, I mean, I think I just would

continue Ralph's discussion of this earlier because I
    24

think he said what I feel about this.
    25

      26         I think you can take your pick, as far as

I'm concerned, about which way to go.
     27                                                  I think you can
stick to the prespecified analysis as it was described
    28

                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
in the protocol and stick with the first set of
     1

reviewers that the sponsors chose, which I consider the
     2

primary analysis of the study.
     3

       4         If you do that, you don't have a positive

study.
     5

       6         Now, in terms of what's right, I'm willing

to accept that retrospectively it makes more -- they
     7

should have chosen the Hamilton reviewers.
     8                                                 I would say

they should have done that to begin with.
      9                                               They didn't.

      what do we do?
Now, 10

      11         And I think the issue really is what Ralph

said. We don't know what to do with those because we
    12

don't know how to interpret them statistically.
    13                                                        I

would feel strongly that on a statistical basis, there
    14

is some unknown penalty that the study has to suffer
    15

     changing its analysis, and we don't know what that
from 16

     We
is. 17 don't know how to do that.

      18         And I think, you know, to me, again, I think

if it were 95 percent agreement I might not worry about
     19

     the
what 20 statistical test was, but we're talking about,

you know, marginal agreement, to begin with, or
    21

marginal level of acceptable agreement and an uncertain
    22

      standard.
gold 23

      24         So I think it's going to wind up accepting

a penalty that is going to put it into no man's land.
    25

      26         CHAIRPERSON RAMSEY:         Any other comments?

Dr. Choyke.
    27             Oh, sorry.     Dr. Jahnke.
      28         DR. JAHNKE:      Yeah, I think we mentioned it
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
-- again, Dr. Jahnke -- we mentioned it a few times,
     1

and the company pointed out that they did not
     2

capriciously decide to discard or not emphasize Study
     3

32A. 4It was because of the very high positive rate,

you know, the 80, 90 percent, 82 percent, which was much
      5

higher, double the expected rate of positivity of
     6

venography in a typical group of patients, which is
     7

stated to be in the 30 to 40 percent range.
     8

       9         So it was done with good motives at least

in some science mind.
    10

      11         CHAIRPERSON RAMSEY:         Dr. Choyke?

      12         DR. CHOYKE:      I'd just like to reiterate

      Dr. Links said about the issue of penalizing the
what 13

sponsor here for a problem with venography, which
    14

clearly has problems with reproducibility, all the
    15

problems we are condemning AcuTect for:
    16

reproducibility, interobserver variability, that kind
    17

of thing.
    18

      19         And, you know, I think that we have to be

careful about, you know, where do we go from here except
    20

to recommend some very elaborate outcomes study, which
     21

probably should be done down the line, but other than
    22

      we
that 23 don't have truth.           It's going to be very

difficult to get truth in this study, and repeating it
    24

won't get there.
    25                    Basically we're stuck with the data

      we
that 26 have, I think, and I think the best you can do

     it
with 27 is interpret it with the experts that have seen
the most.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
       1         CHAIRPERSON RAMSEY:         Dr. D'Agostino.

       2         DR. D'AGOSTINO:       Yeah.      I guess I'm not

sure why we're stuck with the data we have.
     3                                                    This is

      four million subject study that will take 27 years
not a 4

to perform.
     5             It's a study that can be replicated, and

I wouldn't want to replicate it.
      6                                     It can be a study that

can be done where you, in fact, have the readers trained
      7

appropriately at the beginning so that you don't run
     8

into this retrospective or post hoc, which is probably
     9

a better word for it, analysis that you have to
    10

interpret.
    11

      12         And, you know, we all have different

experiences, but the experiences that I see all the time
    13

is that we see the retrospective studies when they turn
     14

out to say what we want them to say, and we don't see
    15

all those retrospective studies that turn out to be
    16

negative.
    17

      18         I mean this was a re-analysis of the data,

and it turned out to be positive.
    19                                          I don't know, and

I said it before, but I don't know how to interpret if
     20

      could be reproduced one more time, and I think,
this 21

again, if the data were so striking that I would be
    22

willing to say, "God, what am I say?" but I don't think
    23

      data is so striking.
this 24                             I think it's marginal, and

I think that there are so many questions with trying
    25

to buy into the post hoc procedure that we shouldn't
    26

do it.
    27

      28         CHAIRPERSON RAMSEY:         I think we'll move on
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
to the next question.
     1

       2         Is there sufficient information from the

agreement of AcuTect and contrast venography results
     3

to develop labeling recommendations for clinical use?
     4

       5         That's a big leap, I know.             Dr. Love, do

you want to make any comments at this point?
     6

       7         DR. LOVE:        This question has to do with

some of the things you talked about earlier.
      8                                                    The false

positive/false negative agreement gives you some
     9

information about overall agreement in the diagnostic
    10

arena, but not necessarily the positive/positive,
    11

negative/negative issues that have been discussed.
    12

      13         So given those kinds of issues, do you feel

      the data would allow you to make recommendations
that 14

for use, screening?
    15

      16         CHAIRPERSON RAMSEY:           I guess we might

throw that back to you, if you've heard enough.
    17

      18         DR. LOVE:        Or you may want to -- you could

     look at this one after you answer the approvability
also 19

question.
    20

      21         CHAIRPERSON RAMSEY:           That's what I was

thinking.
    22          We might come back to some of these when we

      a
have 23 little more discussion.               So let's do that for

now. 24Let's skip that one.            I'll put it in the back of

our minds and go on to (c)(1).
    25

      26         Do you recommend accepting Study 280-32A

as one of the pivotal studies to demonstrate the
    27

efficacy of AcuTect for scintigraphic imaging to detect
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
acute 1venous thrombosis?

       2         And the corollary to that is:           do you

recommend accepting -- let's take those together --
     3

282-3B as one of the two pivotal studies?
     4

       5         Dr. Links.

       6         DR. LINKS:        A question of clarification.

If we do so, which standard are we using, the blind read,
      7

the clinical read, or the Hamilton read?
     8

       9         DR. LOVE:        Right.    That's Question (a).

Which one do you recommend?
    10                                 Maybe a little bit more --

I think I heard most people around the table say take
    11

the Hamilton read, but not necessarily everyone.
    12

      13         For us it would help us to get a clearer

answer on (a).
    14

      15         CHAIRPERSON RAMSEY:           I guess that's what

I heard, is to take the Hamilton read, but I would like
     16

comments from the committee members.
    17

      18         Dr. Kasper.

      19         DR. KASPER:        I think I'd agree with Dr.

Links that that seems to be -- that is the -- the
    20

Hamilton read is the read done by the people who are
    21

the very most expert, but I certainly would like to see
     22

in any publication all the reads because we've learned
     23

something from this.
    24

      25         We've learned the variability of readers,

and we've learned the degree of imperfection of the
    26

venography.
    27

      28         CHAIRPERSON RAMSEY:           Other comments?
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         Dr. Hammes.

       2         DR. HAMMES:      I think there's probably a

consensus to use the Hamilton read for the most part.
     3

What I see in this specific question here, if we throw
     4

out the first blinded read because it had 80 percent
     5

positive as one of the FDA reviewers brought up,
     6

shouldn't we throw out the A study because it only had
     7

20 percent in the Hamilton read?
     8                                       And that's a dilemma

to me 9that I haven't really reached a decision on.

      10         CHAIRPERSON RAMSEY:         Other comments?

      11         Dr. D'Agostino.

      12         DR. D'AGOSTINO:       It's not an upsetting

decision when you have a particular study and you have
    13

two reads that both give you the same result.
    14                                                    I just

--   when we say the Hamilton we say it because on
     15

absolute criteria we believe it's better than the
    16

procedure that was used in the study, but there's a
    17

merit in looking at Study A to say that it was designed
    18

      particular fashion, and it did on its own
in a 19

implementation come out with a positive result.
    20

      21         There is also a Hamilton read for that

study which doesn't contradict that result, but if we
    22

say that we accept A because of the Hamilton read, then
     23

      we're saying is that we are accepting the study
what 24

      deviated, for reasons that deviated from the
that 25

original protocol, and I'm not sure we have to buy into
    26

      for acceptance of A.
that 27
      28         I think in terms of what the next study
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
should look like, in terms of what we think is the better
     1

reading, we can say Hamilton, but we can take A on its
     2

own merits, I think.
     3                            Unless I'm missing something, A

was a 4positive study.

       5         CHAIRPERSON RAMSEY:            Dr. Konstam.

       6         DR. KONSTAM:         Marv Konstam.

       7         Yeah, Ralph.         I'm not sure I'm there

because the question, I think, as appropriately stated
     8

is:    do we consider one of two pivotal studies to
       9

demonstrate the efficacy of AcuTect for scintigraphic
    10

imaging to detect acute venous thrombus?
    11                                                   And I think

that's the question.
    12                            I think that's the right

question.
    13

      14         And in that context, I can't come to the

conclusion that it is.
    15                              I think I can accept the fact

     it
that 16 is a positive study based on its hypothesis and

based on a reasonable statistical analysis, but I think
    17

      all we're left with, you know, as clinicians at
that 18

the end of the day is that we've shown that there is
    19

at least 60 percent agreement with venography.
     20                                                        That's

      the study showed.
what 21

      22         Now, I mean, I agree with the comments that

we shouldn't penalize the company because of the
    23

problems of contrast venography.
    24                                            I don't want to

penalize anybody.
    25                      The question I'm left with is:

      do
what 26 we know from this study that is going to help

a clinician?
    27

      28         And I think if all we know from the study
                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
is that there's at least a 60 percent agreement with
     1

contrast venography, that's not helpful to me as a
     2

clinician.
     3

       4         DR. D'AGOSTINO:         Yeah, I was talking about

the Hamilton versus the non-Hamilton, and I guess I was
      5

glibly saying that we could take it as a positive study.
     6

I think it's a positive study for how it was designed.
     7

Whether or not it's a useful study is the question
     8

you're raising, which I thought we'd get to when you
     9

      about the approvability.
talk 10

      11         CHAIRPERSON RAMSEY:           Dr. Choyke.

      12         DR. CHOYKE:        Pete Choyke.

      13         I'd like to suggest that you ask for the

      from these two studies that has clear positives
data 14

and clear negatives, that is all readers agreed that
    15

the venograms were positive and the venograms were
    16

negative.
    17          It will be a small subset of your

population, but let's face it.
    18                                      If we have significant

disagreement with that data set, we're in big trouble.
    19

If we have significant agreement in that set, at least
     20

we know we're on the right track, and you know, your
    21

confidence about approval would be greatly enhanced,
    22

I think.
    23

      24         DR. LOVE:        What you're asking then is all

readers -- are you including the open venogram read as
    25

      or
well 26 the two blinded reads?

      27         DR. CHOYKE:        You could do it any way you
wanted to.
    28

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

       2         DR. CHOYKE:        But, you know, basically you

have a very nice data set now.
     3                                     You should have all the

data from 240 patients, and you have a zillion readers
     4

now. 5So you could really get unanimity of opinion.

       6         DR. LOVE:        Okay.    You're making a

recommendation.
     7                   That's why I'm --

       8         DR. CHOYKE:        Yes.

       9         DR. LOVE:        -- pressing it.       You're saying

      the data set and see where all venogram reads
take 10

agree, look to see whether the AcuTect read is the same.
    11

      12         Are you concerned about potential sample

     issues?
size 13             Let's say it turned out to be 20 patients.

I'm pressing on purpose so that I can understand what
    14

it is you want us to do.
    15

      16         DR. CHOYKE:        Well, that tells you

something if you only have 20 patients.
    17                                                  I mean I think

we're really in trouble here if out of all these reads
    18

      20
only 19 agreements are found.

      20         I mean I'm expecting something like 30

percent of the cases will agree or 40 percent.
    21                                                        I mean,

I sure hope the negatives will agree with each other
    22

      than that.
more 23

      24         So I don't really think you're going to

      20, only ten percent.
have 25

      26         DR. LOVE:        Right.

      27         DR. CHOYKE:        But you might.
      28         CHAIRPERSON RAMSEY:           Dr. Konstam
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
suggested possibly we might want to vote on (c)(1) and
     1

(2); is that correct?
     2

       3         And I would just ask Dr. Love if you feel

that you need a vote on those or do you have a flavor
     4

of -- 5

       6         DR. LOVE:          That would be -- this is an

important question.
     7                            A vote would be fine.

       8         CHAIRPERSON RAMSEY:             Dr. Ponto.

       9         DR. PONTO:          Isn't the question here

whether these are pivotal studies or not, whether we
    10

accept them as a study, not whether we're accepting the
    11

results as being positive or not?
    12

      13         DR. LOVE:         Oh, maybe this is a little bit

of our jargon here.
    14                            When we say do you accept it as

one of the two, yes, it implies a positive outcome.
     15                                                           So

it would be substantial.
    16

      17         DR. D'AGOSTINO:           Can I ask?

      18         CHAIRPERSON RAMSEY:             Yes.

      19         DR. D'AGOSTINO:           So then in answering

this, we have to get to the question of whether or not
    20

we think the study with the 60 percent and so forth made
     21

sense.
    22

      23         PARTICIPANT:          I don't want to vote.

      24         (Laughter.)

      25         DR. LINKS:          And another question of

clarification in that regard.
    26                                       The way the question is

worded it says "to detect."
    27                                     That to me sounds like a
sensitivity question.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.       Fax: 202/797-2525
       1         DR. LOVE:         That's not implied.

       2         DR. LINKS:         Okay.

       3         DR. LOVE:         It's basically for the

indication as proposed.
     4

       5         CHAIRPERSON RAMSEY:            Well, Committee, do

we want to vote?
     6                   Do you want to go through all of the

discussion and come back and vote on that?
     7

       8         I would actually prefer that.            It might --

as I sit here, I'm not perfectly clear in my own mind
     9

     I
what 10 would want to say, and it might clear it up, and

it might not.
    11                So let's go forward with the

understanding that we will come back and vote on these
    12

two. 13

      14         Four is safety.          Is there sufficient

information to support the safety and reasonable
    15

labeling of AcuTect?
    16

      17         Comments?

      18         DR. CHOYKE:         Pete Choyke.

      19         I think that this is pretty much as safe

as any drug that I've ever seen.
    20                                            So --

      21         CHAIRPERSON RAMSEY:            Well, that's a

pretty good comment.
    22                            Any other comments?

      23         Dr. D'Agostino.

      24         DR. D'AGOSTINO:          Do we -- and I'm

deferring the question, but I'm raising it -- do we have
    25

to worry about no information after the three hours
    26

basically?
    27           We have three hour information and then
      sort of global information that one day, but is
some 28
                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
there 1a concern?         I'm just asking that question.

       2         DR. KONSTAM:      Yeah, I guess I'd follow up

on that.
     3         I just wanted to ask Peter to follow up on

his comment because we heard the FDA safety reviewer
     4

saying he was not satisfied with the amount of safety
     5

data that he had.
     6

       7         So I'm concerned about that.         So could you

comment on why you don't agree with that?
     8

       9         DR. CHOYKE:      There were 169 patients who

      followed at 24 hours, and it was still less than
were 10

one percent side effects, and you know, I don't really
     11

      this for a fact, but I suspect that if you don't
know 12

see anything in three hours, if you see less than one
    13

percent in three hours, the chances that you'll start
    14

developing, especially from a peptide agent, things at
    15

24 hours when it's long excreted, I just don't see that
     16

      big concern.
as a 17

      18         CHAIRPERSON RAMSEY:         Any other comments?

      19         DR. KONSTAM:      Well, I mean, I just wonder

whether we ask Dr. Zolman, who raised this concern, what
    20

would he like to see in an expanded safety data set.
    21

      22         DR. ZOLMAN:      We would probably like to see

about 600 patients.
    23

      24         DR. KONSTAM:      Pardon me?

      25         DR. ZOLMAN:      We would probably like to see

      600 or 1,000 patients.
up to26                               Otherwise this would have

to be particularly treated as a different situation in
     27

labeling.
    28          In other words, it wouldn't be a standard
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
labeling.
     1

       2         DR. LOVE:        Numbers of sample sizes vary.

You know, there are a lot of different ways to approach
     3

it.    For repeat dosing the figures that Dr. Zolman
       4

mentioned are often quoted in ICH guidelines, but such
     5

don't 6exist for single doses, and we've certainly

approved products with smaller numbers than that
     7

certainly.
     8

       9         But I think the issue here is sometimes

      not so much what's the actual number.
it's 10                                                  You know,

there are tables you can look at to try to figure out
    11

how may patients do you need to try to be able to pick
     12

      adverse event with a certain degree of likelihood
up an13

      occurs, say, at one percent, two percent, .5
that 14

percent.
    15

      16         It's often very difficult in a single dose

trial to make those assessments.
    17

      18         Normally what we would like to see is at

least a larger data set that is monitored out to 24
    19

hours.
    20     We certainly balance that with pharmacokinetic

      where the excretion rates, whatever else might
data 21

      been seen in preclinical data.
have 22

      23         I think part of the concern is that some

things can't be detected at three hours, and what do
    24

we do about that?
    25

      26         CHAIRPERSON RAMSEY:           Dr. Hammes.

      27         DR. HAMMES:       Richard Hammes.
      28         I see three issues relative to the safety.
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
First 1and overriding in my mind is the comparison with

contrast.
     2          If we look at what's out there, this is so

much safer that there is no comparison.
     3

       4         The second point though is we are looking

at data, and this is a tracer and we need to remember
     5

that so that it's at subpharmacologic levels by
     6

definition, but there is data that shows platelet
     7

inhibition at 30 times the dose, and we don't really
     8

know at what dose that starts, and I think that's
     9

something that needs to be followed up on a close market
    10

basis at the least.
    11

      12         And then the third issue is the

immunogenicity, I guess, and the potential for multiple
    13

dosing in a sense, and an appropriate warning, you know,
    14

      is
if it15 approved to that effect, and further follow-up

studies, I think, are warranted.
    16

      17         DR. KONSTAM:      Marv Konstam.

      18         You know, I hear you, and you know, I just

     to
want 19 ask.       I mean, it is a tracer, but it's a peptide

tracer, and it's an RGD peptide, and I just would ask:
    20

I mean, are we satisfied that it is in such low
    21

concentrations that, you know, we're clear about its
    22

immunogenicity and any other adverse effects that it
    23

might have?
    24

      25         I don't know.      I mean, I'm uncomfortable

about it, you know, given Dr. Zolman's comments.
    26                                                   I'm

willing to be convinced that there's a reason to feel
    27

safe. I'm just not sure.
    28                              I'm not willing to go on
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
record saying, yes, I'm convinced it's safe.
     1

       2         CHAIRPERSON RAMSEY:          Could you please

state 3your name?

       4         DR. TALARICO:       Talarico.

       5         Some other RGD binders have been -- and we

don't 6know anything about this product at all.               So that

should be looked for, is another safety issue that
     7

should be seen, looked for in a larger number of
     8

patients.
     9

      10         Generally, probably it's not much of a

problem.
    11         It's a very small molecule, and in the

patients, they didn't find any occupiers (phonetic).
    12

So it's likely it's going to be a big problem, but other
     13

events will have to be in doubt.
    14

      15         CHAIRPERSON RAMSEY:          Dr. Hammes.

      16         DR. HAMMES:       Richard Hammes again.

      17         To address your concerns, the other thing

I look at is its very fast elimination.
     18                                                Given that fast

renal clearance, I think if you're going to see a
    19

reaction in all likelihood it would be when its
    20

concentration is high in the first two or three hours,
    21

and we're looking at 700 patients in that time frame.
    22

I feel comfortable with that.
    23

      24         CHAIRPERSON RAMSEY:          Dr. Ponto.

      25         DR. PONTO:       I would like to follow up

Dick's comment there.
    26

      27         The half-life is 1.9 hours.             Even the 169
patients were basically studied at 12 half-lives.
    28                                                             So
                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
we're 1looking at the time where the drug is essentially

eliminated from the body altogether, and they're not
     2

seeing anything substantial.
     3

       4         And as we said before, the alternative is

iodinated contrast, and so we're talking about drugs
     5

that are given in much higher quantities, physical
     6

quantities, as well as a much worse side effect profile.
     7

       8         CHAIRPERSON RAMSEY:            Dr. Jahnke.

       9         DR. JAHNKE:         Just a small disagreement.

I think the alternative we're realistically looking at
     10

is ultrasound, which is real safe.
    11

      12         CHAIRPERSON RAMSEY:            Good point.

      13         All right.         Let's move to Section 5, also

known as Section 6.
    14                        (a)    Do you recommend AcuTect as

approvable?
    15             This, as I understand it, doesn't mean

approved, but could be approvable in the future, for
    16

the scintigraphic imaging of acute venous thrombosis.
     17

      18         DR. LOVE:        Right, and I think just

clarifications.
    19                   Just thinking a we're moving some of

the questions in order, but perhaps think about the
    20

issue of one study, two study as you think about this
    21

because in a way, the answers to whatever this is, 3(c),
    22

     an
have 23 impact on this part of the question and whether

      one or two studies.
it's 24

      25         CHAIRPERSON RAMSEY:            Did everybody hear

that?
    26

      27         DR. D'AGOSTINO:          Excuse me?
      28         CHAIRPERSON RAMSEY:            Yes.
                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
       1         DR. D'AGOSTINO:         Does that mean that we

would 2have to have two positive studies before we could

say yes to this?
     3

       4         DR. LOVE:        Well, we would listen to your

comments and recommendations.
     5                                      As I said, there are

circumstances where we have taken one study, but we
     6

would 7need to understand the reasons why and in this

situation why is it an exception.
     8

       9         CHAIRPERSON RAMSEY:           Shall be vote?

      10         All in favor of recommending AcuTect as

approved for --
    11

      12         DR. KONSTAM:        I'm sorry to interrupt.            I

     wonder if now it wouldn't be worthwhile going back
just 13

and voting about the individual trials because I think
     14

in order to keep internal consistency for ourselves,
    15

I mean if we're going to vote that it's approvable, then
     16

I think under ordinary standards we'd have to be voting
     17

      we
that 18 have two positive pivotal trials, and so maybe

     would be an appropriate starting point to figuring
that 19

out whether it's approvable or not.
    20

      21         CHAIRPERSON RAMSEY:           Committee agree with

that?
    22

      23         PARTICIPANTS:        Yes.

      24         CHAIRPERSON RAMSEY:           All right.       Let's

     up
back 25 to 3(c), numbers one and two.                   Is there someone

on the Committee who would like to point out the key
    26

points of why one would or wouldn't accept 280-32A and
    27

the same for 280-32B?
    28

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

       2         DR. LINKS:       I'm going to be bold.       I'm

going 3to take Ralph's advice and try to get around all

of these issues of whether you even have to use the
     4

Hamilton read for both and propose that we accept both
     5

studies, the first with the original read and the second
     6

with the Hamilton read.
     7

       8         So A with the original read and B with the

Hamilton read, the justification for the substitution
     9

      being the aberrant initial venography results.
on B 10

      11         CHAIRPERSON RAMSEY:          Keeping in mind that

venography is four carat and something else might be
    12

14 carat, not that all of us who have done venograms
    13

     to
want 14 hear that, but there obviously are deficiencies

     the
with 15 technique.            Am I out of line with saying that?

      16         Any other comments?

      17         DR. KONSTAM:       Yeah, I guess I'll take the

other extreme, and you know, I just want to make clear
    18

      I
that 19 really empathize with the problem that the

sponsor is facing, which is that there is no adequate
    20

     standard for acute venous thrombus.
gold 21                                                So that's the

starting point, and that's a problem.
    22

      23         But I still come back to saying:            okay.

      do
What 24 we learn from the data?               And I don't really

consider either study pivotal in the sense that it makes
    25

clear to me or to the clinician that we have an effective
    26

agent for detecting acute venous thrombus.
    27

      28         And I say it, and I guess the strongest --
                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
I mean, I guess I'll come down on Study A, which is,
     1

I think, technically a positive study based on its
     2

hypothesis, but I don't believe that it's an acceptable
     3

clinical finding to help me out clinically, and that's
     4

because I don't learn anything by knowing that AcuTect
     5

      least 60 percent in conformity with the venogram.
is at 6

I just don't learn anything from that.
     7

       8         Now, I understand the problem.         So then

what do you do?
     9                  And I have some suggestions for what

to do.
     10    You know, I think, frankly -- I mean, we'll get

to it -- I mean, I think Dr. Ginsberg pointed to what
    11

we should do because he feels the data are supportive
    12

of going forward and doing some kind of a real
    13

prospective clinical trial with some outcomes.
    14

      15         All I'm saying is I think that if we want

to approve the -- if we think that there is something
    16

to approve here, I don't see it in these trials.
    17                                                         I

don't see how the trials help me say, yes, I should
    18

approve it for diagnosis.
    19

      20         DR. D'AGOSTINO:       Can I comment?

      21         CHAIRPERSON RAMSEY:         Dr. D'Agostino.

      22         DR. D'AGOSTINO:       I also want to follow

D'Agostino's advice --
    23

      24         (Laughter.)

      25         DR. D'AGOSTINO:       -- and come up with a

slightly different conclusion.
    26

      27         I do agree very much with what was just
said, but I also would say that in the first study they
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
put together a study with a particular set of particular
      1

criteria, and that the 60 percent -- and they designed
     2

it and implemented it, and it turned out to be positive.
      3

       4         I think from that trial -- I think going

into the second, which they were running simultaneously
      5

-- I think that it's unfortunate the way it turned out,
      6

but once you start going with the post hoc, I think you
     7

can no longer fall back on the interpretation of what
     8

you have in a strict fashion.
     9

      10         So I would say the second study didn't make

it, and I would say let the first study stand as a
    11

positive study, and from it learn the types of things
    12

you're saying, that it didn't really necessarily
    13

address the right question.
    14                                 Go on to design another

study that, in fact, has a better endpoint, has a better
    15

training period, uses the Hamilton or what have you for
    16

the gold standard.
    17

      18         But I think that there is merit in the first

study, and I think even though it may not be the ideal
    19

study, I think there is merit to call it a positive
    20

study, as long as there's a second study which then can
    21

be informed by it and designed and implemented
    22

correctly.
    23

      24         DR. AMENDOLA:      Dr. Amendola.

      25         I would really like to make a point here

because I don't like for the entire panel to understand
    26

      right now what is now in clinical practice is
that 27
nothing because I'm a practicing radiologist.
    28                                                  I do
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
ultrasound for DVTs, and really we study the calf down
     1

to the knees, and the calf is not really studied.
     2

       3         And let me tell you that we don't do

venography, contrast venography.
     4                                          It's not really in

standard practice.
     5                       It's really the exception to the

rule that we do contrast venography, and in fact, we
     6

are not really trained to read it.
     7

       8         But right now as we stand, what we do when

the ultrasound study is negative, what we do is we
     9

repeat the ultrasound studies three to five days later
    10

in the expectation that if there was a thrombus in the
     11

calf, the thrombus has extended to the thigh.
    12

      13         So really I would put forward here that

today we don't really study calf.
    14

      15         CHAIRPERSON RAMSEY:          Can I take the

prerogative of the chair to make a comment?
    16

      17         MR. MADOO:       Sure.

      18         CHAIRPERSON RAMSEY:          I've been in these

Committee meetings before where we approved various
    19

agents, and I hope you won't laugh at me, but as I sit
    20

      listening to the data, and I don't do a lot of
here 21

venography, but where the studies aren't perfect, the
    22

agent isn't perfect, the results aren't perfect, no
    23

      is
test 24 perfect, and yet here is something, and it is

something and perhaps something which I think I heard
    25

is probably helpful in the diagnosis of acute
    26

thrombosis, and perhaps something in the face of a life
    27

threatening illness is better than nothing, with the
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
recommendation that other studies be done to
     1

substantiate safety factors, efficacy, sensitivity,
     2

and specificity.
     3

       4         DR. D'AGOSTINO:       But you can do that

before or after approval.
     5                              I mean --

       6         CHAIRPERSON RAMSEY:         Well, as I read these

questions, we're not approving it now.
     7                                                Nothing that

we vote on here is approving it, but we're recommending
      8

for approvability, and therefore, that it could be --
     9

maybe Dr. Love could have better words for it -- and
    10

      with the recommendation to go forward with other
then 11

tests and to relook at some of the data as has been
    12

recommended.
    13

      14         DR. CHOYKE:      Okay.    This is often an issue

that's difficult to sort out when we were at this point.
    15

      of
Part 16 this depends upon whether you think there's

enough information to say that one or both of the
    17

studies or all of the studies, whatever, tell you that
    18

there's a definite answer, and the answer that you have
    19

so far from these studies is not apt to change if you
    20

      new study.
do a 21

      22         So, for example, if you thought that there

was enough information, but you needed more
    23

clarification before labeling could be developed and
    24

clarify how the product would be used, then you might
    25

say it's approvable pending completion of those other
    26

studies.
    27

      28         If you thought that another study might
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
change your decision, meaning if you have a Study A and
     1

       B
Study 2 and you weren't sure how they were going to come

out, if you thought a second study might not confirm
     3

your findings, if you thought that was a high
     4

likelihood, then you might recommended nonapproval
     5

because you don't really know what's going to happen
     6

in the long run.
     7

       8         So those are some other issues to deal

with. 9 I guess we sometimes will say a product is

approvable when we know that both clinical studies
    10

perhaps are adequate and acceptable, and we're just
    11

trying to sort out chemistry issues or something else.
    12

So we can certainly do that.
    13

      14         But if there's a one study/two study issue,

unless you're sure or unless we have been sure in the
    15

past, often we would maybe say it's not approvable
    16

unless it's just a point of clarification as was
    17

mentioned earlier, trying to clarify labeling.
    18

      19         CHAIRPERSON RAMSEY:           Dr. Konstam.

      20         DR. KONSTAM:        You know, I guess I just want

to debate this point.
    21                            You know, I hear the panelists

saying that, you know, we don't have anything right now
    22

and so let's go ahead.
     23                           I mean, I guess, not to be glib,

but I would say we could have done that before doing
    24

these two trials if we really want to say that.
    25

      26         I think we could probably have a starting

point by saying the only thing that really would matter
    27

      is
here 28 if we had a test that influenced treatment in
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
a rational way, in a way that we could anticipate that
     1

if you have a positive test and you follow Treatment
     2

A, the patient will do well.
     3                                    If you follow Treatment

B, the patient will not do well, the treatment here,
     4

I guess, being anticoagulation.
     5

       6         I am very, very, very far away from drawing

the conclusion from these data that we have that test.
     7

I just don't have it.
     8                            You know, I think -- let me put

it this way.
     9              I'm no closer to it.          I can imagine this

conclusion based on the preclinical data, and I must
    10

say that for myself the clinical trial data don't bring
     11

me any closer to it because we know that venography is
     12

imperfect, and now we have an agent that is at least,
    13

in the best analysis of these two studies, even with
    14

the Hamilton analysis in the second study, it's at least
     15

60 percent in agreement with venography.
    16

      17         I am nowhere near taking that result and

saying, "Now I know that these patients should be
    18

anticoagulated and that will save their life or reduce
    19

the incidence of pulmonary embolism."
    20                                                  I'm lost.

      21         And what I would suggest -- what I will be

recommending to the sponsor here is that we send them
    22

      to
back 23 do the study that Dr. Ginsberg, who I'm sorry

to see has left because I was going to ask him more about
     24

      exactly he would do; is now take the data that we
what 25

      and do a prospective clinical study, taking
have 26

patients with a negative scan.
    27                                      I think we have enough
wherewithal to do that, and then follow the patients
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
for six months and watch something that is important
     1

predefined clinical outcomes.
     2

       3         And then if we found that, then we'd have

something really important, and we would have done, I
     4

think, the entire medical community and their patients
     5

a service.
      6          I don't think we've done that at this point.

       7         CHAIRPERSON RAMSEY:          Dr. Links.

       8         DR. LINKS:       It seems to me that whether or

not what you're saying is the way to go at least in part
      9

depends on the context in which this agent would hit
    10

the market.
    11             If the context is as a replacement for

venography, it seems to me the type of study you would
    12

do is very different, more along the lines of what was
     13

done, than if the context in which it hits the market
    14

is that you assume that venography will be done at least
     15

a fraction of the time and this in some way should
    16

augment the information provided by venography rather
    17

      replace it.
than 18

      19         I personally don't have any problem

couching this in the context of a replacement for
    20

venography because then the present study design is,
    21

in fact, an appropriate study design.
    22                                                 The criterion

      60
of a 23 percent agreement may not be the right

threshold, but if all you're going to do is say it's
    24

a replacement, then all you have to do is show that it
     25

agrees.
    26

      27         DR. KONSTAM:       Well, I would counter by
saying, first of all, it is only 60 percent, okay, and
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
we've 1also heard repeatedly that nobody is going

venography anymore.
     2                            So why do we feel it's an

acceptable criterion for provability to say that
     3

something has at least a 60 percent agreement with
     4

venography?
     5

       6         I just don't get it.           I don't see it at all.

       7         DR. JAHNKE:          Dr. Jahnke.

       8         But that was the FDA's recommendation, I

believe.
     9

      10         DR. KONSTAM:          We don't need to live with

that.
    11

      12         DR. JAHNKE:          Right.     I know.

      13         DR. KONSTAM:          We understand that, and

that's unfortunate, but it's not our business.
    14

      15         DR. JAHNKE:         -- was backed into that, and

there's something we have not talked about much, and
    16

      clear we haven't talked about it much.
it's 17                                                       We keep

saying whether we should agree with the blinded read
    18

or the Hamilton setting, you know.
     19                                             The institutional

      did agree with the Hamilton study also, which is
read 20

the basis of the -- yes, it did.
     21                                       The institutional read

agreed with the Hamilton study, and that's what the
    22

basis of the clinical treatment was in this series.
    23

      24         CHAIRPERSON RAMSEY:             Right, because they

had clinical data.
    25                       I'm not even sure what order.

We'll just go around the table.
    26

      27         Go ahead, please.
      28         DR. AMENDOLA:          I was kind of surprised of
                              S A G CORP.
202/797-2525   Washington, D.C.       Fax: 202/797-2525
that fact, how the institutional read was much better,
     1

and I 2was wondering one of the reasons there was such

an improvement was because the data from the ultrasound
      3

studies were taken into account.
     4

       5         CHAIRPERSON RAMSEY:            Correct.      Other

data was taken into account.
     6

       7         DR. AMENDOLA:         Because I don't believe

that by notice of the clinical history that would
     8

explain the improvement in the results.
     9

      10         CHAIRPERSON RAMSEY:            Dr. Hammes.

      11         DR. HAMMES:         I'm real uncomfortable with

the 60 percent level.
     12                           You flip a coin and do just about

as good obviously.
    13                       The question is:            where is this

coming from?
    14              And it appears it's coming from the

venography rather than the apcitide.
    15

      16         And if you look at all the other supporting

studies, the institutional reads, the multiple human
    17

      they all support its value, and I think we need
use, 18

to keep that in the back of our mind.
    19

      20         If I'm a patient with suspected DVT right

      give me a Doppler or give me this study and if
now, 21

they're both negative, don't give me therapy and let
    22

me go home.
    23

      24         DR. KONSTAM:         What are the data that

support what you just said?
    25                                  What data are your drawing

      to
upon 26 conclude that?

      27         DR. HAMMES:         The data that -- first off,
venography I don't think would be a viable option for
    28

                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
me given the inconsistencies we've seen and the
     1

morbidity.
     2

       3          Secondly, Doppler is very good if you know

where 4the thrombus is to begin with, you know.             You've

      sore spot in your leg, and you can aim the Doppler
got a 5

at it, and you can find it.
     6                                  In the absence of those

localizing symptoms, Doppler doesn't find it.
     7                                                               We saw da

seven 8percent morbidity from therapy of

anticoagulation, I think we add something significant
     9

to the medical practice by making this tool available
    10

and at least screen out that portion of the patient
    11

population and with some significant benefit, also
    12

keeping in mind the relative safety of it.
    13

      14          CHAIRPERSON RAMSEY:         Dr. Choyke.

      15          DR. CHOYKE:      I'd just like to make two

points.
    16         One is that it's quite possible and quite

likely, given the magnitude of venous thrombosis as a
    17

problem in this country that if this agent was approved,
    18

     outcome studies such as the one you would envision
that 19

would be readily funded.
    20                              It's of such magnitude that

I think it would happen.
    21

      22          And I don't think -- I mean, I haven't been

involved with that many of these sessions, but I think
    23

the holding the sponsor to the standard of an outcome
    24

study is atypical.
    25                        It's not typically what's

required.
    26

      27          What's required is to show some degree of
efficacy for the agent, which I think if you believe
    28

                               S A G CORP.
202/797-2525    Washington, D.C.   Fax: 202/797-2525
the Hamilton read, you can show some degree of efficacy.
      1

It may not be the perfect drug.
     2                                         It likely isn't the

perfect drug, but I believe that outcome study that
     3

really should be done will be done in our current, you
     4

know, 5situation.

       6         DR. KONSTAM:        Well, I guess we can

recommend to the FDA that they require that the study
     7

be done either before approval or after approval; is
     8

that right?
     9

      10         DR. LOVE:        Yes, that's correct.

      11         CHAIRPERSON RAMSEY:           Dr. D'Agostino.

      12         DR. D'AGOSTINO:         We use vocabulary in

different ways.
    13                  When I sometimes use the term "outcome

study," it's a completely uncontrolled study that I'm
    14

      looking at practice.
just 15

      16         I think that what I'm talking about, a

clinical trial which may have a longer follow-up, but
    17

not in a typical effectiveness outcome study fashion.
    18

I don't know.
     19              Maybe you're referring to outcome study

      different fashion than I am.
in a 20

      21         DR. AMENDOLA:        I'd also like to make

clear, and I have an article here by Dr. Cronan, which
    22

is one of the experts in ultrasound of the DVTs.
    23                                                           Let

me read it to you.
    24

      25         "If clot is isolated to calf veins, it is

recognized that upward propagation, popliteal vein
    26

involvement occurs in approximately 20 percent of
    27

cases.
    28     Propagation of clot can be . . . if ultrasound
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
studies are performed at three to five days intervals."
     1

       2         The reason for this is because with

ultrasound we don't study the calf.
     3                                             Most institutions

do not study the calf.
      4                           So there is an area that we have

not -- and we are certainly not doing contrast
     5

venography for that episode.
     6

       7         CHAIRPERSON RAMSEY:           Dr. Ponto.

       8         DR. PONTO:        As Dr. Ramsey referred to

earlier, I've been involved in some of these decisions
     9

where the question is do we want to give the clinicians
    10

a new tool that they don't have already, and it's quite
     11

obvious that this area needs a new tool.
    12                                                  The question

     is
is: 13 this the right one or not?

      14         And that's what I'm grappling with, and do

these studies convince us that this is the right tool?
    15

      16         If the differences we saw in the agreement

rates could be attributed to the fact that the apcitide
    17

was telling us something that the venography was not,
    18

     I
then 19 would be more comfortable with giving people this

      to
tool 20 work with, but because they didn't give us any

outcome data, did not look to see who had a pulmonary
    21

embolism and who did not, I don't know if those
    22

differences are just because we cannot read the studies
    23

adequately or because it is a better tool.
    24

      25         And so I'm feeling sort of like Mark is over

here, that there's the need for another study that looks
    26

at outcome, that looks at a different predictive
    27

variable, maybe ultrasound, maybe not venography, but
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
something that says that the people with a positive
     1

study 2have a worse prognosis than the people with a

negative study.
     3

       4         CHAIRPERSON RAMSEY:           But there are also

clinical implications beyond just positive and
     5

negative study.
     6

       7         Dr. D'Agostino.

       8         DR. D'AGOSTINO:          Not to be a legalist, but

we're 9all talking or those who are talking about it a

study are talking about one study.
    10                                             The FDA wants two

studies, and this is my logic of saying Study A looks
    11

all right as long as it informs us about a very good,
    12

new study.
    13

      14         CHAIRPERSON RAMSEY:           Dr. Love.

      15         DR. LOVE:        Yes, the studies do not have to

be identical as long as they corroborate in some manner
     16

or another.
    17

      18         CHAIRPERSON RAMSEY:           Dr. Rohde.

      19         DR. ROHDE:        Yes.    Charles Rohde from

Johns Hopkins.
    20

      21         I'd like to hopefully clarify an issue

about the 60 percent.
    22                            The 60 percent is not the best

estimate of agreement of these two methods.
    23                                                      It is the

lowest value which is supported by the data.
    24                                                        The

actual numbers, estimates from the data, are in the 70
    25

percent range.
    26

      27         And if you put an upper confidence limit
on it, that would go very close to 80 percent.
     28                                                       So the
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
suggestion that we're talking about something that's
     1

about 2like flipping a coin is a little misguided.

       3         And the sponsor was told that this was the

criteria, and that was the criteria.
     4                                                It may be that

this study should have been run in some different way.
     5

What it sounds like to me, everyone is saying that there
      6

should have been a different outcome looked at and so
     7

forth, but what we have is something like 70 percent
     8

agreement.
     9

      10         Now, I'm not convinced that we cannot get

      information from this data than
more 11                                               we have.   For

example, we do have the original records, in which both
    12

readings were positive, in which both were negative,
    13

and some were positive and the other was negative.
    14

      15         We also have patient characteristics for

these data.
    16            We have the ability now to put in an effect

for differences between readers, and it would not
    17

surprise me that a really careful analysis would
    18

demonstrate that these two methods are absolutely
    19

equivalent.
    20

      21         That hasn't been done, but it probably

could be done with the right people and the right help,
    22

and it could be done probably very quickly.
    23

      24         So it strikes me that there's just about

as much doubt in my mind about the positive implications
     25

as there are about the negative implications.
    26                                                           It's

just, you know, we've gotten railroaded into looking
    27

at one specific issue, and I'm not sure if it's exactly
     28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
the right one.
     1

       2          CHAIRPERSON RAMSEY:          I'd like to bring us

back in order to move forward here, if we could vote.
     3

I'm not sure that we can or maybe it's inappropriate,
     4

but (c)(1), do you recommend accepting Study 280-32A
     5

as one of the two pivotal studies to demonstrate the
     6

efficacy of AcuTect for scintigraphic imaging to detect
     7

acute 8venous thrombosis, yes or no?

       9          So all those who would accept it, please

raise your hand.
    10

      11          (Show of hands.)

      12          MR. MADOO:       It looks like we have ten out

of 12.
     13    Could those who were not accepting raise their

hands so we can verify that?
    14

      15          (No response.)

      16          MR. MADOO:       Are any abstaining?

      17          MR. MADOO:       We have one extension.          It

looks like we're missing a vote.
    18

      19          DR. KONSTAM:       I voted no.

      20          MR. MADOO:       You voted no?        Okay.

      21          CHAIRPERSON RAMSEY:          I'll vote to accept.

      22          MR. MADOO:       Okay.    Dr. Ramsey will vote to

accept.
    23         So we have 11 accepting and one, Dr. Marvin

Konstam, no, not accepting.
    24

      25          CHAIRPERSON RAMSEY:          All right.       (c)(2)

Do you recommend accepting Study 280-32B as one of the
     26

two pivotal studies to demonstrate the efficacy of
    27

AcuTect for scintigraphic imaging to detect acute
    28

                               S A G CORP.
202/797-2525    Washington, D.C.    Fax: 202/797-2525
venous thrombosis?
     1                       Again, yes and no.

       2         All those who would accept it, say yes.

Raise 3your hand, yes.

       4         (Show of hands.)

       5         MR. MADOO:         It looks like seven.

       6         CHAIRPERSON RAMSEY:            All those opposed

raise 7your hand.

       8         (Show of hands.)

       9         MR. MADOO:         Five opposed.

      10         CHAIRPERSON RAMSEY:            All those

abstaining.
    11             That's 12.        Sorry.     I'll get out my

checkbook here.
    12

      13         DR. LOVE:         Excuse me.      Question.    Just

for sake of numbers, you were voting on the first
    14

question, but not the second?
    15

      16         MR. MADOO:         No, no.

      17         CHAIRPERSON RAMSEY:            I voted yes.

      18         DR. LOVE:        I'm sorry.      The second one then

is? 19

      20         CHAIRPERSON RAMSEY:            Seven to five.

      21         DR. LOVE:         Seven to five.        Thank you.

      22         CHAIRPERSON RAMSEY:            All right.     Let's go

      to
back 23 the last set of questions then.                   Do you

recommend AcuTect as approvable for the scintigraphic
    24

imaging of acute venous thrombosis?
    25                                               And this, again,

is not for approval.
    26                            It's just approvable.

      27         I think once we voted -- I guess you're
right.
    28     It does have to be -- you're right.                 You're
                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
right.
     1     You're right, but sometimes it's much more

overwhelming than others.
     2                               Sometimes they don't listen

to us.
     3

       4         DR. PONTO:       Point of clarification.         This

vote is based on the current status of the data,
     5

correct, not on any kind of reanalysis?
     6

       7         CHAIRPERSON RAMSEY:           Right.     That will

come up in the next -- I think in (c).
     8                                                  We'll make

recommendations.
     9

      10         Are we right, Dr. Love?

      11         DR. LOVE:        Yes.

      12         CHAIRPERSON RAMSEY:           So okay.    Yes is you

say yes to the approvability.
    13                                      No is you do not agree

      approvability.
with 14

      15         So all those who are in favor of

approvability for the scintigraphic imaging for acute
    16

venous thrombosis say yes; raise your hand yes.
    17

      18         (Show of hands.)

      19         MR. MADOO:        Seven.

      20         CHAIRPERSON RAMSEY:           All those no?

      21         (Show of hands.)

      22         MR. MADOO:        Four no.

      23         CHAIRPERSON RAMSEY:           Abstaining?

      24         MR. MADOO:       It looks like we're missing a

person.
    25

      26         CHAIRPERSON RAMSEY:           One abstention.

      27         MR. MADOO:        One abstention.
      28         CHAIRPERSON RAMSEY:           Okay.    Back to
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
discussion.
     1             Is there any other indication that you

recommend?
     2

       3         That's kind of a curve ball here, right.

Let's 4skip that for now.

       5         If you do not recommend AcuTect, but let's

just leave that as open, for open discussion again, as
     6

approvable, are there other studies or trial designs
     7

that you would recommend to be completed before
     8

approval?
     9

      10         We've heard some discussion of that

already.
    11         Looking at the data again, seeing if we could

get more out of it.
    12

      13         Dr. Kasper.

      14         DR. KASPER:      Well, other than looking at

the data again, perhaps given the discussion around the
     15

table that venograms are not done very much anymore
    16

except in a few places such as Hamilton, perhaps the
    17

FDA should reconsider its position that the comparison
     18

ought to be made with sonography, certainly for above
    19

the knee.
    20

      21         CHAIRPERSON RAMSEY:         Dr. Love is that

okay?
    22

      23         Any other comments?         Dr. August.

      24         DR. AUGUST:      I think we have a problem that

is worth going back to, and that is that one of the main
     25

issues with the failure of venography to be a true gold
    26

standard is that the image will be positive if there
    27

is an old, organized thrombus that's just hanging in
    28

                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
the vein, and the same is true for ultrasound, so far
     1

as I can tell.
     2

       3         So given that the AcuTect is going to

detect acute emerging thrombi and not the old ones,
     4

there's always going to be a real problem with
     5

discrepancy between presumably a higher number of
     6

images that are going to be positive by venography or
     7

ultrasonography and a lower number presumably that are
     8

going 9to be positive using AcuTect.

      10         And I think if we don't recommend that a

study be designed to take that into consideration and
    11

somehow get around it, then there will be
    12

dissatisfaction with the extent of agreement or
    13

whatever with every study that this Committee is asked
    14

to critique.
    15

      16         CHAIRPERSON RAMSEY:          Dr. Ponto.

      17         DR. PONTO:       I'd like to follow up on that

and recommend that outcomes be involved in any study
    18

in the future and reiterate what Dr. Kasper said, that
     19

we need to use the current technology that would be used
     20

in these patients, that being ultrasound.
    21

      22         I would also like to recommend that the

company institute the same type of a mechanism that has
    23

     done with some of the more recent drugs that we've
been 24

approved, that they have very rigorous training for
    25

their readers so that we would not have the disagreement
    26

      we
that 27 saw with the readers that we saw in this
particular study, both in a study context, as well as
    28

                              S A G CORP.
202/797-2525   Washington, D.C.    Fax: 202/797-2525
possibly in its clinical utility.
     1

       2         CHAIRPERSON RAMSEY:           Other comments?

       3         Dr. Love?        Oh.

       4         DR. TALARICO:          This is Lilia Talarico.

       5         I'd like to make a comment on the

differentiation between diagnosis of DVT or clinical,
     6

clinical DVT versus DVT that's going to be picked up
     7

for thromboprophylaxis, for example, in surgery,
     8

abdominal surgery, et cetera.
     9

      10         When you're dealing with

thromboprophylaxis, noninvasive tests are very poor,
    11

and venography must still play a role for diagnosis of
     12

DVT in thromboprophylaxis.
     13                                 So venography is not out.

      14         CHAIRPERSON RAMSEY:           We're not throwing

it out the window, but nobody wants to do it or have
    15

it. 16

      17         DR. KONSTAM:        No, I -- may I speak?

      18         CHAIRPERSON RAMSEY:           Yes.

      19         DR. KONSTAM:        Marv Konstam.

      20         I guess I've said this a couple of times,

but I guess this is a good place to say it again.
    21                                                             Of

course, I voted for not approvability.
    22                                                  So I'd like to

see this done before approval, but if we're going to
    23

      approvability, then I'd like to suggest that the
vote 24

FDA request a study, a Phase IV study.
    25

      26         And, you know, here's what I think.              I

mean, I think let's think of the implications of what
    27

we've approved.
    28                   I mean, we've approved an agent for
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
detection of acute venous thrombus with the presumption
     1

that that has an implication on therapy, and we don't
     2

know exactly how it's going to be used in the field.
     3

We really don't.
     4

       5         I mean, we've heard some comments about how

people think they might use it or would recommend using
     6

it, but I'm not sure that's going to make its way into
     7

the labeling.
     8

       9         I think what's going to happen is that the

agent is going to get out into the field, and it's going
    10

to be used variably.
    11                            Now, what I'd like to see is to

     what happens to a patient who has a negative study
know 12

and is sent home, and I think that this is a critical,
     13

important question because this is what is going to
    14

happen in the community, and I think that there's an
    15

obligation here to learn what happens when that
    16

happens, and I think it's also a great opportunity.
    17

      18         So I would design the study accordingly.

The specificity to my reading is clearly fairly low.
    19

So the advantage is that a negative study, sensitivity
     20

     little higher.
is a 21                       Take patients who have a negative

study, send them out without treatment, and follow them
    22

prospectively, and then the details of that can be
    23

worked out in terms of the duration of follow-up and
    24

the outcomes that we want to follow.
    25

      26         But I would urge very strongly that the

company be required to do such a study.
    27

      28         CHAIRPERSON RAMSEY:            Thank you.
                              S A G CORP.
202/797-2525   Washington, D.C.      Fax: 202/797-2525
       1         Dr. Hammes.      Sorry.

       2         DR. HAMMES:      Yeah, Richard Hammes.

       3         I concern with Dr. Ponto that we need some

outcomes data as part of this inevitably.
     4                                                I would also

suggest that in future studies -- and I'll have to defer
     5

to our radiology colleagues -- but it seems to me that
     6

if you could direct ultrasound with the nuclear study,
     7

ultrasound ought to be able to confirm the presence of
     8

absence of a clot, and that may be a viable approach
     9

to get a better gold standard at least in the positive
     10

results.
    11

      12         DR. AMENDOLA:      Dr. Amendola.

      13         I think that that is a very logical

question, and I think that ultrasound should be used
    14

as the comparison, not venography because venography,
     15

      is
one, 16 not, as we heard before, it's not a gold

standard, and, second, it's not used, but ultrasound
    17

is used every day.
    18

      19         CHAIRPERSON RAMSEY:         Any other comments?

Yes. 20

      21         DR. CHOYKE:      I'd just like to point out

      the company started with ultrasound, but it was
that 22

recommended by the agency that they shift to contrast
    23

venography.
    24             So, you know --

      25         CHAIRPERSON RAMSEY:         Yes, we heard that

earlier.
    26

      27         DR. CHOYKE:      -- that's a little unfair.
      28         CHAIRPERSON RAMSEY:         They're responding
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
to our request, right?
     1                            Well, the FDA's request.

       2         DR. LOVE:        The change from ultrasound to

contrast venography, yes, was after we talked with them
     3

about 4the issues of the calf and pelvis, and we were

talking about you don't know prospectively where the
     5

patient is going to have the abnormality, and that was
     6

the rationale behind changing to the contrast
     7

venography.
     8

       9         The sponsor did do a reasonably large size

study.
    10     I think it was 100-and some odd patients, 200,

in that study.
     11                That study was completed from a safety

perspective and analysis, and some of that immuno --
    12

no, the immunogenicity data was not that one.
    13

      14         But at any rate, they did do a study, and

there was some analysis done by Dr. Sobhan just to try
    15

to look to see whether there was any difference in the
     16

results just in terms of percentages of positive or
    17

negative with the ultrasound or the contrast
    18

venography, and there wasn't much difference.
    19

      20         These are all different data sets,

different studies, but the results weren't appreciably
    21

different.
    22

      23         CHAIRPERSON RAMSEY:           Dr. Links.

      24         DR. LINKS:       A radical comment.        It seems

to me that half of our discussion has been the result
    25

of not a poor study, but a poor study design relative
    26

to the indication, and I, for one, all through the day
     27

      been somewhat frustrated that in a sense the
have 28
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
studies that we have before us are not the studies you
     1

would 2do to specifically address the proposed

indication, but they sound like they were certainly the
     3

studies that ultimately the company and the FDA
     4

together decided were the studies to be done.
     5

       6         And I'm just wondering if we're the group

that's supposed to grapple with recommendations after
     7

the fact, shouldn't we have a shot at grappling with
     8

study 9design issues at least some of the time before

the Phase II trials start.
    10

      11         DR. LOVE:        Yes, you may, and we would

certainly love if you would look at this.
    12                                                  If there is

a recommendation for a new protocol, we'd love to bring
     13

it back to the Committee.
    14

      15         DR. KONSTAM:       Yeah, but you know, wouldn't

it have been a loud statement to make that as a comment
     16

about the -- sorry, but you know, I agree.
    17                                                  I agree with

      you're saying, and I think that really that
what 18

statement becomes loud if you say, "You know what?
    19

      data set doesn't really support approvability.
This 20

     is
This 21 the study," and forget what the FDA recommended

to years ago or whatever it is.
    22

      23         But, you know, sorry to keep -- but I agree

     you.
with 24         I think it is appropriate for us to say what

we think is good criteria for approving or not approving
     25

an agent like this.
    26

      27         CHAIRPERSON RAMSEY:           Having been involved
     number of studies, you go into them thinking that
in a 28
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
this is the right thing to do, and then when you look
     1

back and say that really isn't what I wanted to do after
      2

all, but I think when you started you thought it was,
     3

and you went in with every good intent, and I think it's
      4

just after you get done that you realize that it didn't
     5

give you the answer that you were actually looking for.
      6

       7         So I'm not absolutely positive it's poor

design, although it might be.
     8

       9         But that being said, let's move on to (d),

and I think we pretty well covered (d).
    10

      11         DR. LOVE:        Right.    Could I ask a question

      on
back 12 (c)?        (c) says do you recommend that this is

      before approval, and so what you've recommended
done 13

is approvability.
     14                    Do you -- on this study that you're

talking about, or studies, whatever it might be, are
    15

you recommending that that's done before it is approved
     16

or after it is approved, meaning in Phase IV or
    17

beforehand?
    18

      19         CHAIRPERSON RAMSEY:           I think we're going

to have to vote on this one.
    20

      21         Do you recommend AcuTect as approvable?

If you do -- oh, if you do not.
     22                                      Well, let's say if we

     If
do. 23 we do recommend, and we did, would you like

other studies or trial designs to be completed before
    24

approval?
    25

      26         DR. AMENDOLA:        I think we have to decide

which studies.
    27

      28         CHAIRPERSON RAMSEY:           And then which one.
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         DR. LOVE:        Well, I don't necessarily --

I'm not really asking for which, but basically you've
     2

recommended some outcome studies, and I've heard
     3

different perspectives on whether or not you need to
     4

know that before labeling can be developed versus
     5

after, basically before you know how to use the product
     6

versus after you know how to use the product.
     7                                                     So that's

why I'm asking that question.
     8

       9         CHAIRPERSON RAMSEY:           Oh, boy, that's a

     one
hard 10 because outcomes are sometimes ponderable, not

always obtainable, and there were other things that I
    11

think the Committee asked, and that is to remassage the
    12

      that's available already and bring that forth to
data 13

      at
look 14 it again.

      15         And Dr. August and then Dr. Links.

      16         DR. AUGUST:       Charles August.

      17         I do think that they ought to respond to

the issue of safety with longer -- with a longer time
    18

period of observation and larger numbers, and I think
    19

that's quite possible, and I think that it may well be
    20

     the
that 21 immunogenicity issue could be settled by maybe

another draw, a month, six weeks down the line.
    22

      23         CHAIRPERSON RAMSEY:           Dr. Ponto -- no,

somebody else.
    24                 Oh, Dr. Links.         I'm sorry.

      25         DR. LINKS:       Since the majority already

voted for approvability, perhaps a way out of this
    26

particular conundrum is to suggest that the outcomes
    27

trial, which I think all of us, including those who
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
voted 1for approvability, would like to see be done as

a Phase IV with the long term approval conditional on
     2

accomplishing that Phase IV trial within -- in other
     3

words, approval to be withdrawn if the Phase IV isn't
     4

accomplished within some time frame that the FDA sets.
     5

       6          DR. LOVE:        There are some regulations that

allow 7that.        Normally they are very specific

statements for accelerated approval that hasn't been
     8

accomplished thus far.
     9                              So we can take that under

advisement, but I don't know if there's a regulatory
    10

mechanism to complete get us out of that.
    11

      12          DR. D'AGOSTINO:         Why don't we just

reconsider what we did before and ask them to have this
    13

      before approvable?
done 14                              I mean, I think that the

studies are very important, and we just don't have all
    15

of this information.
    16

      17          DR. LINKS:        Question.      How long would

      a
such 18 trial take?

      19          DR. D'AGOSTINO:         Let's let the company

worry about it.
    20

      21          DR. LINKS:        No, I'm asking a clinical

question.
    22           How long is adequate follow-up?

      23          DR. D'AGOSTINO:         Oh, oh, oh, you're

talking about how long the outcome period.
    24

      25          DR. LINKS:        Yeah, right.

      26          DR. D'AGOSTINO:         Some people say in six

months.
    27         I don't know if that would be idea, but six
months sounds reasonable.
    28

                               S A G CORP.
202/797-2525    Washington, D.C.     Fax: 202/797-2525
       1          DR. KONSTAM:        I'd be satisfied with six

months.
     2         I don't know what other people think.            I mean

if you -- if somebody came in with a questioned
     3

diagnosis of DVT and went home with no therapy, what
     4

would 5we consider a reasonable period of follow-up to

know that we didn't do any harm?
      6                                       Six months to me seems

pretty reasonable.
     7                        Maybe less, maybe three months is

reasonable.
     8              I don't know.       Certainly no more than six

months.
     9         I don't think we'd need --

      10          DR. AMENDOLA:        Probably three months.

      11          DR. KONSTAM:        Three months might be okay,

somewhere in that range.
    12

      13          CHAIRPERSON RAMSEY:           Any other comments?

Dr. August.
    14

      15          DR. AUGUST:       I think that the likelihood

is good that if and when this is on the market it's going
     16

to be used repeatedly at least in a subset of patients
     17

who may have a chronic problem, and I'm curious to know
     18

whether my view is shared by the people who actually
    19

      care of these patients and whether or not we
take 20

shouldn't anticipate that issue with some suggestions
    21

for guidelines that might be given by the FDA to the
    22

company.
    23

      24          CHAIRPERSON RAMSEY:           Dr. Love.

      25          DR. LOVE:        I guess I interpreted that as

a question to the panel.
    26

      27          CHAIRPERSON RAMSEY:           Well, I think
someone recommended 600 patients for follow-up, and you
    28

                               S A G CORP.
202/797-2525    Washington, D.C.     Fax: 202/797-2525
said it could be done with less.
     1                                          So I think we could

maybe 2defer to the FDA to decide on a number that would

be necessary for safety.
     3

       4         DR. LOVE:        Maybe I misunderstood your

point.
     5

       6         DR. AUGUST:        My question was quite a

different one, and that is that even with the safety
     7

record that it now enjoys, I think the temptation will
     8

be great in a subset of patients who have chronic
     9

problems to use this technique over and over again, and
    10

yet everything that we've heard about today has been
    11

the results of a single study.
    12

      13         And my question really is:             should we

anticipate?
    14             I would like some guidance from the

physicians, the clinicians who take care of these
    15

patients as to whether my surmise is correct.
    16

      17         And then if it is, are there some

anticipatory suggestions or guidelines that the FDA
    18

could make in that regard, the simplest being, I
    19

suppose, to emphasize in the labeling that this study
    20

and the safety and efficacy data that we have pertain
    21

      to
only 22 patients who were studied once, and we can't

guarantee, for example, that if they were injected
    23

repeatedly with this polypeptide that there would not
    24

      immune reaction and there might not be anaphylaxis
be an25

on the second or third or whatever exposure.
    26

      27         And you could probably come up with some
other things as well.
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
       1         DR. LOVE:        Okay.   Yes, certainly there is

a history of putting such warnings or comments in the
     2

labeling if there's a limited safety database.
     3

       4         I guess what I also heard you asking though

is are there some recommendations or guidelines for
     5

repeat dose studies, and that's where I thought you were
     6

asking the other panelists if there were some things
     7

that you wanted to recommend on how that might be
     8

studied; is that correct?
     9                               Is that what you're asking?

      10         DR. AUGUST:        Well, what I asked was if

there were clinicians who would comment on the
    11

likelihood that there would be patients who would be
    12

treated over and over again, and then if there were,
    13

      should we produce or should we recommend to you
then 14

     you
that 15 create some guidelines for that repeated usage

so that perhaps people can be aware that the data that
     16

we have and on which we recommended approvability was
    17

limited.
    18

      19         DR. AMENDOLA:        So the issue is the repeat

doses, and as far as we can determine there is no safety
    20

issue, no knowledge about the safety of that.
    21

      22         DR. LOVE:        Right, no knowledge.

      23         DR. AMENDOLA:        So now we have this three

or five days at least.
    24

      25         CHAIRPERSON RAMSEY:           Dr. Konstam.

      26         DR. KONSTAM:        You know, I'd just like to

      you know, I attend on the cardiology wards, and
say, 27
there are patients that are going to come in with
    28

                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
questions of deep venous thrombosis, and after this
     1

drug is approved, based on the data set that we have
     2

now, I am not going to know how to use it.
     3

       4          I've heard many suggestions about how to

use it.
     5         Some of them seem cogent, but they're not

really supported by the data set.
     6                                           I think the

suggestion was that based on the level of sensitivity
     7

that we see in a subset of patients who presented with
     8

symptoms within the last three days, we might be safe
     9

accepting that as a solitary test and sending the
    10

patient out.
    11

      12          I think that the data that I see show that

that's a reasonable hypothesis, but I'm not totally
    13

convinced about that.
    14

      15          And I also don't know what to do about a

positive test.
    16                  I'm not sure whether we see anything

in the data set that tells us how to handle a positive
     17

test. Is a positive test a trigger to do a venogram,
    18

which isn't commonly done?
    19                                 Is it a trigger to do an

ultrasound?
    20

      21          Again, I mean, we could come up with

recommendations, but I guess we need to.
    22                                                 Someone needs

to come up with recommendations about how to handle
    23

these different contingencies.
    24

      25          I for one do not see guidance in the data

set about how to deal with these questions.
    26

      27          CHAIRPERSON RAMSEY:         Dr. Jahnke.
      28          DR. JAHNKE:      Therefore, you agree with the
                               S A G CORP.
202/797-2525    Washington, D.C.   Fax: 202/797-2525
FDA that this agent is not approvable for the detection
      1

of thrombosis, I guess.
     2                             I mean, that was Dr. Jones'

conclusion.
     3             That was mine also.

       4         DR. KONSTAM:        Yeah.

       5         DR. JAHNKE:        And my question, and it may

not be proper to answer this, but you, of course, don't
     6

have to agree with the opinion of the Advisory Committee
      7

in taking it into consideration, I assume.
     8

       9         DR. LOVE:        Yes.   Just a process note.

Basically when the preliminary reviewers make a
    10

recommendation before something has come to the
    11

committee, that's basically the review team
    12

recommendation.
    13                   After we listen and consider all of

the points that you've recommended, then the final
    14

action is taken.
    15

      16         So we will very strongly consider

everything that's been said here.
    17                                          There are times that

the agency has agreed with -- this across the board.
    18

We're not just talking about this Committee -- there
    19

are times when the agency agrees with a recommendation.
     20

There are times when it does not.
    21                                            I think it's

appropriate that if we do not follow your
    22

recommendations, that we would communicate with you
    23

about what the issues were and why if we are not.
    24

      25         What I'm hearing is a lot of different sets

of perspectives from the Committee on all of these
    26

issues.
    27

      28         CHAIRPERSON RAMSEY:           I'd like to also go
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
to 5(b).
     1         Are there any other indications that you

would 2recommend?

       3         I'm not sure that's appropriate.           Dr.

Links 4said no.        I'm not sure if that's appropriate

under 5our topics for today, but it's there so I wanted

to address it.
     6

       7         Dr. Jahnke.

       8         DR. JAHNKE:      Going back to something Dr.

Love said earlier that we somewhat skipped around,
     9

addressing the issue of straining versus definitive
    10

evaluation.
    11             I don't think we have addressed that

adequately, have we?
    12

      13         CHAIRPERSON RAMSEY:         I don't think we --

      14         DR. JAHNKE:      Some of it goes to what Dr.

Konstam said.
    15

      16         CHAIRPERSON RAMSEY:         It's been mentioned.

      17         DR. JAHNKE:      I mean, should this be used

     screening exam if you have a low or moderate level
as a 18

of confidence?
    19

      20         CHAIRPERSON RAMSEY:         Low prob.

      21         DR. JAHNKE:      Versus a definitive

examination if you have a high level of suspicion.
    22

      23         CHAIRPERSON RAMSEY:         I think that

dovetails in with saying that we need more studies, that
    24

we need to look at it more.
     25                              So at the present it would

probably be a screening, but I guess I can't actually
    26

answer that.
    27

      28         Any other comments from any other panel
                              S A G CORP.
202/797-2525   Washington, D.C.   Fax: 202/797-2525
members?
     1

       2         (No response.)

       3         CHAIRPERSON RAMSEY:           Mr. Madoo, is there

anything else?
     4

       5         MR. MADOO:       No, I guess we're done, right,

Dr. Love?
     6

       7         CHAIRPERSON RAMSEY:           I was just going to

ask Dr. Love if there was anything else she wanted us
     8

to -- 9

      10         DR. LOVE:        I think you're certainly

covered.
    11         I'd like to take a moment just to thank you

      much for your detailed consideration of this.
very 12

These are a lot of important issues.
    13                                                  Certainly we've

dealt with issues that surround receptors in general.
    14

      of
Some 15 these issues are things that the Committee

discussed with the guidance document, looking at
    16

physiologic or biochemical issues, and here you have
    17

an anatomic standard of truth.
    18

      19         So these are issues that are going to be

important in the long run.
    20

      21         You've also dealt with issues about

agreement studies, and that's going to be a
    22

prospectively active issue that we'll have to continue
    23

to address.
    24

      25         But I thank you very much for your comment.

      26         CHAIRPERSON RAMSEY:           Thank you.

      27         I'd like to also thank all of the panel
members for coming.
    28                        I think this has been one of the
                              S A G CORP.
202/797-2525   Washington, D.C.     Fax: 202/797-2525
more interesting panels where we really dug at some
     1

issues.
     2         I want to thank all of you for taking the time

to come here, and thank everybody in the audience and
     3

the presenters, as well.
     4

       5          (Whereupon, at 3:44 p.m., the meeting was

concluded.)
     6

       7

       8

       9

      10

      11

      12

      13

      14

      15

      16

      17

      18

      19




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

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:7
posted:11/18/2011
language:English
pages:242