3325t2

Document Sample
3325t2 Powered By Docstoc
					                                                              1

    1           UNITED STATES OF AMERICA

    2                   * * * * *

    3   DEPARTMENT OF HEALTH AND HUMAN SERVICES

    4                   + + + + +

    5         FOOD AND DRUG ADMINISTRATION

    6                   + + + + +

    7   CENTER FOR DRUG EVALUATION AND RESEARCH

    8                   + + + + +

    9      DIVISION OF ONCOLOGY DRUG PRODUCTS

   10                   + + + + +

   11                  54TH MEETING

   12                   + + + + +

   13                    FRIDAY,

   14               SEPTEMBER 19, 1997

   15                   + + + + +

   16         The   meeting   took    place     in   Versailles

Ballrooms I and II, Holiday Inn Hotel-Bethesda, 8120
   17

Wisconsin Avenue, Bethesda, Maryland at 8:30 a.m.,
   18

Janice J. Dutcher, M.D., Chairman, presiding.
   19

PRESENT:
   20

JANICE J. DUTCHER, M.D.
   21                                Chairman

JANNETTE O'NEILL-GONZALEZ
   22                                Executive Secretary

DAVID H. JOHNSON, M.D.
   23                                Member

JAMES KROOK, M.D.
   24                                Member

     A.
KIM 25 MARGOLIN, M.D.                Member
ROBERT OZOLS, M.D., Ph.D.
   26                                Member

RICHARD L. SCHILSKY, M.D.
   27                                Member
                                                    2

SANDRA SWAIN, M.D.
    1                         Member

DONALD W. NORTH FELT, M.D.,
    2

    3   F.A.C.P.              Guest Expert

DAVID M. ABOULAFIA, M.D.
    4                         Guest Expert

MICHAEL MARCO, B.A.
    5                         Patient Representative

DESMAR WALKES, M.D.
    6                         Consumer

Representative
    7

ROBERT DELAP, M.D.,Ph.D.
    8                         FDA Representative

JOHN 9JOHNSON, M.D.           FDA Representative

ROBERT JUSTICE, M.D.
   10                         FDA Representative

     KOBAYASHI, M.D.
KEN 11                        FDA Representative

ROBERT TEMPLE, M.D.
   12                         FDA Representative

SAMUEL BRODER, M.D.
   13                         Sponsor Representative

     DUTCHIN, Ph.D.
KEN 14                        Sponsor Representative

PARKASH GILL, M.D.
   15                         Sponsor Representative

GREGORY HARRIMAN, M.D.
   16                         Sponsor Representative

JOHN HOWES, Ph.D.
   17                         Sponsor Representative

GREGORY HARRIMAN, M.D.
   18                         Sponsor Representative

JOHN HOWES, M.D.
   19                         Sponsor Representative

MICHAEL BETTS
   20                         Patient Perspective

STEVEN CAROL
   21                         Patient Perspective

ERIC FLETCHER
   22                         Patient Perspective

GAVIN GRAY
   23                         Patient Perspective

DAVID GREEN
   24                         Patient Perspective

     MOLINA
JIM 25                        Patient Perspective
WILLIAM W. LI, M.D.
   26                         Public Comment

ALSO PRESENT:
   27
                                                 3

STEVE CARRIER, Ph.D.
    1

    2

    3

    4

    5

    6

    7

    8

    9

   10

   11

   12

   13

   14

   15

   16

   17

   18

   19

   20

   21

   22

   23     T-A-B-L-E    O-F   C-O-N-T-E-N-T-S

   24                                          Page

Call to Order and Opening Remarks
   25

Conflict of Interest Statement
   26

Open Public Meeting, Dr. Li
   27                            9
                                                           4

Applicant's Presentation
    1                               14

    2   Introduction, Dr. Howes 14

    3   Kaposi's Sarcoma, Dr. Broder

    4   Study Protocol, Dr. Gill

    5   Comparative Results, Dr. Harriman

    6         Dr. Duchin            25

    7   Patient Perspectives:

    8         Eric Fletcher         41

    9         Steven Carol          46

   10         Jim Molina            53

   11         Gavin Gray            61

   12         Michael Betts         61

Committee Questions
   13                      66

     Presentation
FDA 14                     101

   15   Ken Kobayashi      101

   16         Questions 122

   17   Donald Northfelt            134

Questions
   18         139

Adjourn
   19         146

   20

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

   22                                              8:47 a.m.

   23         CHAIRPERSON DUTCHER:        We are not going to

     started for about five more minutes.
get 24                                            We are

waiting for some handouts.
   25                            So feel free to get another
     of
cup 26 coffee.

   27         [Pause.]
                                                          5

    1       CHAIRPERSON DUTCHER:      Good morning.

This 2is the Oncology Drug Advisory Committee meeting.

My name is Janice Dutcher.
     3                        I'm an oncologist at Albert

Einstein Cancer Center in New York and I'd like to have
    4

the members of the committee introduce themselves and
    5

where they are from.
    6                  We can start at this end, please.

    7       DR. WALKES:      My name is Desmar Walkes.

I'm a family practitioner from Bastrop, Texas and the
    8

consumer rep substituting on the committee.
    9

   10       DR. OZOLS:    Bob Ozols, medical oncologist

from Fox Chase Cancer Center in Philadelphia.
   11

   12       DR. SWAIN:    Sandra Swain, medical

oncologist, Washington, D.C.
   13

   14       DR. SCHILSKY:      Rich Schilsky, medical

oncologist, University of Chicago.
   15

   16       LT. O'NEILL-GONZALEZ:      Jannette

O'Neill-Gonzalez, Executive Secretary for FDA and for
   17

     committee.
the 18

   19       DR. JOHNSON:      I'm David Johnson, medical

oncologist at Vanderbilt University.
   20

   21       DR. SIMON:    I'm Richard Simon,

biostatistician, National Cancer Institute.
   22

   23       DR. MARGOLIN:      Kim Margolin, medical

oncologist, City of Hope, Duarte, California.
   24

   25       DR. ABOULAFIA:      Dave Aboulafia, medical
oncologist and hematologist, Virginia Mason Clinic,
   26

Seattle, Washington.
   27
                                                           6

    1          DR. NORTHFELT:    Don Northfelt, I'm a needs

oncologist at University of California, San Diego and
    2

Pacific Oaks Medical Group.
    3

    4          DR. MARCO:    I'm Michael Marco, Director of

Opportunistic Diseases for the Treatment Action Group,
    5

New York.
    6

    7          DR. KROOK:     Jim Krook, medical

oncologist, Duluth, Minnesota.
    8

    9          DR. DELAP:     Bob DeLap, Oncology Drug

Division Director, FDA.
   10

   11          DR. JOHNSON:     John Johnson, Clinical Team

Leader, FDA.
   12

   13          DR. KOBAYASHI:     Ken Kobayashi, Medical

Officer, FDA.
   14

   15          DR. TEMPLE:     Bob Temple, Director of

Office of Drug Evaluation I.
   16

   17          LT. O'NEILL-GONZALEZ:     Good morning.

     going to be reading the Conflict of Interest
I'm 18

Statement.
   19

   20          The following announcement addresses

conflict of interest issues associated with this
   21

meeting and is made a part of the record to preclude
   22

even the appearance of a conflict.
   23                                     Based on this

Committee's agenda and information provided by the
   24

participants, the Agency has determined that all
   25

reported interests in firms regulated by the Center for
   26

Drug Evaluation and Research present no potential for
   27
                                                       7

a conflict of interest at this meeting with the
    1

following exceptions.
    2

    3       In accordance with 18 U.S.C. 208(b)(3),

full 4waivers have been granted to Dr. Sandra Swain and

Dr. Kim Margolin.
    5               A copy of this waiver statement may

be obtained by submitting a written request to the
    6

Agency's Freedom of Information Office, Room 12830 of
    7

the Parklawn Building.
    8

    9       In addition, we would like to disclose for

     record that Dr. Ozols and his employer, the Fox
the 10

Chase Cancer Center, have interests in Bristol-Myers
   11

Squibb and Pharmacia Upjohn, sponsors of competing
   12

products to Paxene which do not constitute financial
   13

interest in the particular matter within the meaning
   14

of 18 U.S.C. 208.
   15

   16       Notwithstanding these interests, it has

been determined that it is in the Agency's best
   17

interests to have Dr. Ozols participate fully in all
   18

matters concerning Ivax's Paxene.
   19

   20       With respect to FDA's invited guests, Dr.

Donald Northfelt has reported interest which we believe
   21

should be made public to allow the participants to
   22

objectively evaluate his comments.
   23                                 Dr. Northfelt

would like to disclose that in 1996 he received
   24

consulting and speakers fees from Sequus
   25

Pharmaceuticals.
   26

   27       In the event that the discussion involves
                                                        8

any other products or firms not already on the agenda
    1

for which an FDA participant has a financial interest,
    2

the participants are aware of the need to exclude
    3

themselves from such involvement and their exclusion
    4

will 5be noted for the record.

    6       With respect to all other participants, we

ask in the interest of fairness that they address any
    7

current or previous financial involvement with any firm
    8

whose products they might wish to comment on.
    9                                               Thank

you.
   10

   11       CHAIRPERSON DUTCHER:     Let me just

reiterate, as we discussed yesterday, we will have an
   12

open public hearing at this point in the meeting.
   13                                                  In

addition, additional time has been added to the
   14

sponsor's time for patients whom they would like to have
   15

speak on behalf of their drug to come forward.
   16                                              So that

will be later in the morning.
   17

   18       But for right now, we will have the open

public hearing and Dr. Li has asked to speak.
   19                                              Please

identify yourself and your constituency.
   20

   21       DR. LI:   Lieutenant O'Neill-Gonzales, Dr.

Dutcher, members of the Committee.
   22                                 Good morning and

thank you for the opportunity to come here to speak.
   23

   24       I'm Dr. William Li, medical director of the

Angiogenesis Foundation, a 501(c)(3) non-profit
   25

organization whose mission is to coordinate global
   26

efforts in bringing about angiogenesis-based
   27
                                                        9

therapies.
    1         Today I've come to this Oncologic Drugs

Advisory Committee meeting on Paxene or paclitaxel, to
    2

direct the Committee's attention to the angiogenesis
    3

inhibitory activity of paclitaxel, a property which we
    4

believe is under recognized.
    5                           The Committee should

consider that Paxene's antiangiogenic effects may
    6

contribute to its cytotoxic effect on tumor cells.
    7

    8         Paclitaxel is an effective cancer

chemotherapeutic agent that has been used to treat
    9

refractory ovarian cancer, metastatic breast cancer,
   10

advanced head and neck cancer, non-small cell lung
   11

cancer, and malignant melanoma.
   12                              Several clinical

trials suggest its effectiveness in regressing
   13

AIDS-associated Kaposi's sarcoma.
   14

   15         Paclitaxel has unique mechanisms of

action.
   16     The mechanism commonly cited is its binding

to the beta two subunit of tubulin.
   17                                  This prevents

depolymerization and promotes stabilization of
   18

microtubules.
   19            Because of this, paclitaxel inhibits

mitotic spindle formation, the G2 and M phase of the
   20

cell cycle, cell proliferation, cell motility and
   21

chemotaxis.
   22         This mechanism is thought to be directly

responsible for paclitaxel's anticancer effects.
   23

   24         There is, however, another mechanism by

which paclitaxel inhibits tumor growth.
   25                                      Paclitaxel
also inhibits angiogenesis, the process of new blood
   26

vessel formation.
   27
                                                        10

    1          Solid tumor growth is dependent upon

angiogenesis.
    2             Without a new blood supply, tumors are

restricted to a small size, less than two millimeters
    3

in diameter.
    4           Once angiogenesis is initiated by tumor

cells, new vessels bring in oxygen, nutrients and
    5

survival factors that allow for exponential tumor
    6

growth, invasion and metastases.
    7                                The concept of

antiangiogenesis, first proposed in the early '70s, is
    8

designed to inhibit this process and it's a new
    9

therapeutic modality being developed by pharmaceutical
   10

companies worldwide, and by the National Cancer
   11

Institute.
   12          We believe that paclitaxel's

antiangiogenic activity also contributes to its
   13

anti-tumor activity.
   14

   15          Paclitaxel inhibits angiogenesis by at

least three mechanisms.
   16                      It inhibits endothelial cell

proliferation.
   17              It inhibits endothelial cell

locomotion.
   18           And it inhibits protease production by

endothelial cells, including the production of
   19

collagenase, which is involved in dissolving the
   20

extracellular matrix surrounding growing new blood
   21

vessels.
   22

   23          Paclitaxel inhibits angiogenesis in

experimental systems such as the chicken
   24

chorioallantoic membrane and in vitro cultures of
   25

capillary endothelial cells.
   26                           Studies by Ernest Brahn

at UCLA also show that paclitaxel can inhibit
   27
                                                       11

angiogenesis in an animal model of collagen-induced
    1

arthritis.
    2         In companies like Bristol-Myers Squibb and

Angiotech Inc. have specifically referred to
    3

antiangiogenesis as one activity of paclitaxel.
    4

    5         How might this information influence the

Committee's views of Paxene?
    6

    7         First, Paxene's antiangiogenic activity

lends validity to its rationale for treating Kaposi's
    8

sarcoma.
    9       KS lesions are highly angiogenic, composed

of vascular-like spindle cells and they secrete at
   10

least six angiogenic cytokines, including basic
   11

fitroblast growth factor, vascular endothelial cell
   12

growth factor, platelet-derived growth factor,
   13

interleukin-6, transforming growth factor beta,
   14

GM-CSF, and also the HIV-Tat protein.
   15                                     Therefore,

antiangiogenesis is a rational approach to treating KS.
   16

   17         Second, because of its antiangiogenic

activity, Paxene may have promise for treating other
   18

angiogenesis-dependent diseases, including rheumatoid
   19

arthritis, diabetic retinopathy, psoriasis, and solid
   20

tumors.
   21      Further studies need to be conducted.   Until

such studies are completed, we believe that appropriate
   22

cautions for the off-label use of Paxene should be
   23

developed.
   24

   25         Third, there may be valuable lessons to be
learned from other angiogenesis-inhibitor drugs in the
   26

clinic, such as TNP-470, thalidomide, marimastat, and
   27
                                                      12

interferon-alpha.
    1                With these drugs, we are learning

that 2long-term therapy is needed for efficacy.    The

optimal biological dose may be lower than the maximal
    3

tolerated dose.
    4              And, that the detection of angiogenic

cytokines in blood, urine and cerebrospinal fluid may
    5

serve as useful surrogate markers to monitor therapy.
    6

    7        Fourth, if approval is given, during the

post-marketing surveillance period for Paxene, we
    8

encourage physicians to be alert for possible
    9

unanticipated, beneficial antiangiogenic effects such
   10

as the inhibition or stabilization of diabetic
   11

retinopathy or improvement in psoriasis in those Paxene
   12

treated AIDS patients with these co-morbid conditions.
   13

   14        There may also be unanticipated adverse

effects due to antiangiogenesis such as the inhibition
   15

of collateral formation in coronary artery disease or
   16

     delay of wound healing after surgery.
the 17

   18        In summary, we wish to emphasize to the

Committee that Paxene's effects include the inhibition
   19

of angiogenesis.
   20               This lends validity to its use for

treating Kaposi's sarcoma, opens up new avenues and
   21

potential applications of this drug, and it shows that
   22

this drug merits further specific examination for its
   23

effects as an antiangiogenic agent.
   24

   25        Thank you.
   26        CHAIRPERSON DUTCHER:    Thank you very

much.
   27   Is there anyone else in the audience who wishes
                                                         13

to speak at the open public hearing at this time?
    1                                                 [No

response.]
    2

    3        Then we are going to move ahead with the

applicant's presentation and I believe we have some
    4

handouts at this time.
    5                      I hope.   Okay.

    6        This is a discussion of NDA 20-826, Paxene

indicated for failure of first line or subsequent
    7

systemic chemotherapy for the treatment of advanced
    8

AIDS-related Kaposi's sarcoma.
    9                                Dr. John Howes is

going to begin the presentation.
   10

   11        DR. HOWES:    Ladies and gentlemen, members

of ODAC, good morning.
   12                      I'm John Howes with the

Regulatory Affairs Department of Baker-Norton
   13

Pharmaceuticals.
   14                 Today we will present data to

support the use of Paxene for the treatment of advanced
   15

AIDS related Kaposi's sarcoma in patients who failed
   16

first line and subsequent systemic chemotherapy.
   17

   18        Regrettably, Dr. Jerome Groopman, who was

scheduled to be the opening speaker, is unable to attend
   19

     meeting today.
the 20                 In his place on the agenda will be

     Samuel Broder, Senior Vice President for Research
Dr. 21

     Development at Ivax Baker-Norton Corporation.
and 22

   23        Since we do have a rather full agenda

today, I will now pass the podium to Dr. Broder.
   24

   25        DR. BRODER:     Thank you very much.
Kaposi's sarcoma is an angioproliferative tumor
   26

characterized historically by endothelial and spindle
   27
                                                       14

cell 1proliferation, angiogenesis, inflammatory cell

infiltration, and edema.
    2                       In 1994, a new herpes virus,

HHV-8 or KSHV, was discovered and found to be closely
    3

associated with this tumor and may play a role in its
    4

pathogenesis.
    5

    6        This tumor is one of the hallmarks of AIDS.

Slide 1 please.
    7               The inter-relationship between

immunodeficiency diseases and cancer generally, and
    8

between AIDS and Kaposi's sarcoma specifically, has
    9

been a very high priority of the National Cancer
   10

Institute and its viral cancer programs.
   11

   12        Clinical research done at the Institute

suggested that Kaposi's sarcoma is sensitive to
   13

paclitaxel, a natural product originally derived from
   14

     pacific yew.
the 15              This line of work is an extension of

about 30 years of research on paclitaxel by the National
   16

Cancer Institute.
   17

   18        Next slide please.    Paclitaxel, of

course, has effects on tubulin and the state of tubulin
   19

polymerization.
   20               But perhaps even more interesting, as

we heard in part, are newly described mechanisms of
   21

action for this agent.
   22                      Paclitaxel inhibits

angiogenesis and induces apoptosis by Bcl-2
   23

phosphorylation triggered by Raf-1 activation.
   24                                                It is

possible that these new mechanisms may be induced by
   25

lower plasma concentrations of paclitaxel than the
   26

effects on the microtubule system.
   27
                                                        15

    1         AIDS-related Kaposi's sarcoma frequently

can be an aggressive disease, often with extensive
    2

cutaneous lesions, but also involvement of the oral
    3

cavity and visceral organs.
    4                           AID-related KS can be

complicated by lymphedema.
    5                         Could I have the next slide

please?
    6     And this may involve the extremities, the face

or the genitalia.
    7

    8         Gastro-intestinal lesions may cause

bleeding, pain and obstruction and pulmonary lesions
    9

     be
may 10 associated with respiratory insufficiency or

death.
   11     Even in the absence of symptomatic visceral

disease or edema, Kaposi's sarcoma may have a serious
   12

impact on quality of life by causing disfigurements and
   13

social isolation or by serving as a visual reminder of
   14

an AIDS diagnosis.
   15

   16         When Kaposi's sarcoma lesions can be

covered or obscured by clothing, a patient's
   17

recognition that lesions are growing progressing is
   18

still a serious medical challenge.
   19

   20         Next slide please.   Although milder forms

of Kaposi's sarcoma in the context of AIDS with slow
   21

progression or without life threatening visceral
   22

involvement can be treated with local or intralesional
   23

therapies, the more serious, advanced forms, if left
   24

untreated, do not spontaneously resolve as a general
   25

rule, and require cytotoxic chemotherapy.
   26

   27         We believe that Kaposi's sarcoma and the
                                                         16

therapeutic challenges that this disease forces upon
    1

us will remain an important problem, notwithstanding
    2

the formidable advances that have been made in treating
     3

retro-viral diseases.
    4

    5          As is true in virtually all of oncology,

the status of prior chemotherapy is an important
    6

consideration.
    7             Efficacy results with patients naive

to chemotherapy should generally not be pooled with
    8

results in second or third-line therapy.
    9

   10          Since the early 1990s, the ABV regimen,

which consists of dixorubicin, bleomycin and
   11

vincristine, has been considered the standard of care.
   12

In evaluating individuals or in making comparisons
   13

between clinical trials, it is important to know
   14

whether the patients have been previously treated with
   15

doxorubicin.
   16           Moreover, in the past two years,

liposomal anthracyclines have been introduced, but for
   17

a variety of reasons, it is important not to lump these
    18

     therapies together indiscriminately.
two 19

   20          Next slide please.   DaunoXome, that is

liposomal daunorubicin, was approved as first-line
   21

treatment based on a prospective randomized trial
   22

comparing DaunoXome to ABV.
   23                          Although response rates

were similar, 23 percent for DaunoXome and 30 percent
   24

     ABV, there was significantly less alopecia and
for 25
neuropathy in the setting of DaunoXome.
   26

   27          Next slide please.   Doxil, that is
                                                       17

liposomal doxorubicin, was approved as second-line
    1

treatment of advanced AIDS Kaposi's sarcoma based on
    2

a 27 3percent response rate in 34 evaluable patients.

    4         By contrast, the response rates reported

for paclitaxel, some of which we will discuss later,
    5

for second-line treatment of Kaposi's sarcoma have been
     6

higher.
    7      And this was in part discussed at the Advisory

Committee immediately preceding this current meeting
    8

in the June ODAC meeting.
    9

   10         For safety purposes, it is probably wise

to use all available patients.
   11                              But paclitaxel is not

an exception to the rule that for efficacy purposes it
    12

is important not to pool first and second-line patient
    13

data.
   14

   15         Also, because of the non-linearity of

paclitaxel pharmacokinetics, caution is in order when
   16

     extrapolates from one dosing level or apparent
one 17

dose-intensity to another.
   18                          We will touch upon these

points in our presentation.
   19

   20         Next slide please.   We believe that Paxene

makes an important contribution to the knowledge base
   21

     paclitaxel in second-line AIDS related Kaposi's
for 22

sarcoma.
   23       Our study included advanced patients who

frequently had failed second-line or third-line
   24

treatments.
   25          Specifically, many of the patients were
Doxil failures.
   26

   27         Another major point is that the study
                                                     18

presented today is the first prospective multicenter
    1

study of paclitaxel in advanced Kaposi's sarcoma, and
    2

as such may give a more realistic estimate of community
     3

based results.
    4

    5        We will also touch upon the concept that

perhaps in this tumor more than most there is an element
    6

of observer's subjectivity in making determinations of
     7

response.
    8

    9        We will also provide important information

on pharmacokinetics as well as information on
   10

co-administration with protease inhibitors.
   11                                           We

believe the latter is a very important set of
   12

information in that now there is a nearly universal use
   13

of this category of antiretroviral agent.
   14

   15        We believe that much of the information

that will be presented today is unavailable in any other
   16

form.
   17   For prescribers it is important to have as much

empirical data as possible on both the positive
   18

features and the limitation of paclitaxel.
   19

   20        Finally, we wish to thank the Chair and the

     and the members of this Committee for permitting
FDA 21

some of the patients who participated in this study to
   22

speak here today at the conclusion of our scientific
   23

presentation.
   24

   25        All clinical progress depends on the
willingness and courage of patients to enter studies
   26

on the safety and efficacy of new drugs.
   27
                                                           19

    1          Members of the Committee, members of the

audience, thank you very much.
    2                              I now would like to turn

the podium over to Dr. Gill who is the principal
    3

investigator of this study and he will provide some of
    4

the data related to efficacy results.
    5                                        Dr. Gill.

    6          DR. GILL:    Good morning.    Can you go to

the next slide.
    7              This Paxene study was conducted in

patients with advanced Kaposi's sarcoma. It was a
    8

prospective phase II trial in patients who had failed
    9

prior systemic cytotoxic chemotherapy.
   10                                         The trial was

conducted in nine U.S. sites and patients were enrolled
   11

between January '96 and April of '97.
   12

   13          Patients were eligible for this trial if

they had advanced disease defined by one or more of the
   14

following criterias:
   15                      multiple cutaneous lesions,

presence of visceral disease or symptomatic
   16

lymphoedema.
   17           Other eligible criterias included

failure of prior cytotoxic chemotherapy.
   18                                           Patients

were required to have KPS of 60 or above.
   19                                           And the use

of concomitant antiretroviral agents, including
   20

protease inhibitors, were allowed.
   21

   22          Primary study end points included best

response and time to progression.
   23                                   And secondary end

points were change in symptom distress scale and
   24

Karnofsky performance status.
   25                               Paxene
Pharmacokinetics were also performed in a subset of the
   26

patients and these data will be presented by Dr. Ken
   27
                                                       20

Duchin.
    1

    2        The response criteria used in this trial

were 3those defined and used by ACTG-Oncology committee

for the past several years.
    4                          Complete and partial

responses were required to be maintained for at least
    5

28 days.
    6

    7        The treatment regimen consisted of Paxene

given at a dose of 100 milligram per meter squared over
    8

three hours every two weeks after premedication with
    9

dexamethasone, cemetidine and diphenhydramine.
   10                                                One

does reduction was allowed to 75 milligrams for
   11

toxicity.
   12       In the event of more severe toxicity,

treatment was withheld until recovery.
   13                                      Use of G-CSF

     treatment of neutropenia was also permitted.
for 14

   15        Eighty nine patients were enrolled in

these nine sites through April of 1997 and two large
   16

accrual regents represent Boston and Los Angeles.
   17

   18        The patient demographics are outlined

here.
   19   The median CD4 count was low at 40 and a majority

of the patients had Karnofsky performance status
   20

between 70 and 80, 61 percent.
   21

   22        Antiretroviral therapy was taken by 71

percent of the patients at study entry, this included
   23

     of
use 24 protease inhibitors in 33 patients.      In

addition, a third of the patients were receiving
   25

therapy for CMV infection and 30 percent of the patients
   26

were receiving G-CSF.
   27
                                                        21

    1          The tumor assessment at baseline showed

mucocutaneous disease in all but two patients, facial
    2

disease in 42 patients, and oral disease in 40 percent.
    3

Tumor associated edema was also observed in nearly half
    4

of the patients and visceral disease was present in 42
    5

percent.
    6       Pulmonary involvement was the most common

site 7of visceral involvement.

    8          TIS staging system has been developed for

prognostic prediction for this disease and this
    9

accounts for three different areas, tumor burden,
   10

immune status and systemic illness.
   11                                   Poor prognostic

features for these include tumor associated edema,
   12

visceral involvement and extensive oral disease,
   13

immune status of CD-4 being less than 200 and the prior
   14

symptoms of opportunistic infections and the past
   15

history of these symptoms of low performance status.
   16

   17          Next slide please.   Utilizing these TIS

staging criteria, in this study two or more of these
   18

poor prognostic features were present in 90 percent of
   19

     cases.
the 20

   21          All patients had received prior cytotoxic

chemotherapy.
   22             Over a third of the patients had

received two or three prior cytotoxic chemotherapy
   23

regimens.
   24         Among these patients, 46 percent had

received liposomal daunorubicin and 30 percent had
   25

received liposomal doxorubicin.
   26

   27          A median of eight cycles of Paxene was
                                                     22

administered with a range of one to 27.
    1                                       Thirty four

patients remain on study after receiving ten cycles of
    2

therapy.
    3      The median dose intensity in this trial was

44 milligram per meter squared per week.
    4

    5        Response rates were assessed by intent to

treat analysis.
    6             Complete and partial responses were

observed in 46 percent with 95 percent confidence
    7

interval of 41 to 62.
    8                    These data represent the

independent review by Dr. Kaplan who was not an
    9

investigator in this trial.
   10

   11        This is a representative example of

responding patients.
   12                   A patient with advanced

cutaneous disease and extensive edema which was
   13

associated with pain and required use of crutches
   14

showed marked improvement after 19 cycles.
   15

   16        Looking at the impact of prior therapy and

outcome, patients who received one prior regimen had
   17

a response rate of 47 percent compared to 41 percent
   18

     those who received two or three prior regimens.
for 19

   20        The response rates in those who received

prior liposomal daunorubicin or liposomal doxorubicin
   21

were 51 percent and 33 percent respectively.
   22

   23        The impact of protease inhibitor use was

also examined.
   24             Twenty nine patients did not receive

     protease inhibitors during the trial.
any 25                                        The
response rate of 41 percent in this subgroup compared
   26

to the overall response rate of 46 percent suggests that
    27
                                                          23

protease inhibitors may not have a significant impact
    1

on the possibility or probability of response outcome.
    2

    3        The median time to response in this patient

population was 49 days.
    4                      And the duration of response

which was calculated from initiation of treatment has
    5

not been reached and would be in excess of 306 days.
    6

    7        Time to treatment failure for the study

population was 234 days.
    8

    9        I would now ask Dr. Harriman from

Baker-Norton to conduct the remainder of the
   10

presentation.
   11           Thank you.

   12        DR. HARRIMAN:    Thank you, Dr. Gill.    Good

morning ladies and gentlemen, members of ODAC and
   13

guests.
   14     My name is Gregory Harriman and I'm with

Baker-Norton Pharmaceuticals.
   15                             Before beginning my

presentation, I would like to have Dr. Ken Duchin from
   16

Baker-Norton get up and give a brief presentation of
   17

     pharmacokinetic studies.
the 18

   19        DR. DUCHIN:     Good morning.    Thank you

very much.
   20        We present data today on the

pharmacokinetics of paclitaxel in AIDS KS patients in
   21

     study just described by Dr. Gill.
the 22                                       It must be

recognized that these studies were very difficult to
   23

conduct given the demands on the patients' time and we
   24

     very grateful to the patients who participated in
are 25
this pharmacokinetic study.
   26

   27        Eleven patients from one site volunteered
                                                       24

for pharmacokinetic sampling.
    1                            These patients were

taking four to 20 concomitant medications, which
    2

included one or more reverse transcriptase inhibitors,
    3

imidazole antifungal and the protease inhibitor
    4

indinavir.
    5        The protease inhibitors are of particular

interest because paclitaxel and protease inhibitors
    6

are metabolized by cytochrome P453A and almost all of
    7

the marketed protease inhibitors carry a warning in
    8

their product label of potential interactions with
    9

concomitant medications that also utilize this
   10

metabolic pathway.
   11

   12        Serial plasma sampling, which involved

about 20 samples per patient, occurred over 51 hours
   13

during and after the three hour infusion of Paxene on
   14

     of
one 15 the cycles.

   16        Nine patients were studied on one cycle and

     patients were studied twice on two consecutive
two 17

cycles.
   18

   19        The next slide shows the mean plasma

concentration time curve for paclitaxel in the nine
   20

patients who were studied on one cycle.
   21

   22        Mean pharmacokinetic parameters are shown

in this slide.
   23            I wish to point out that peak plasma

concentrations (Cmax) averaged 1100 nanogram per mil
   24

or about 1.3 micromole and body clearance averaged
   25

approximately 27 liters per hour per meter squared.
   26

   27        A comparison of some of the
                                                       25

pharmacokinetic parameters obtained at the dose of 100
    1

milligrams per meter squared was made using a weighted
    2

analysis to values obtained from other Paxene studies
    3

in 37 patients with solid tumors who received a higher
    4

dose, 175 milligram per meter squared.
    5

    6         As noted on the left hand side of the slide,

a 75 7percent increase in administered dose was

accompanied by much greater increases in peak plasma
    8

paclitaxel levels and in areas under the curve to the
    9

last detectable concentration and to infinity.
   10                                                The

dash line would be the expected increase in these
   11

parameters if the drug obeyed linear kinetics.
   12                                               These

data demonstrate the nonlinearity of the
   13

pharmacokinetics of paclitaxel over the range of 100
   14

to 175 milligram per meter squared.
   15

   16         We also evaluated the pharmacokinetics of

Paxene in those patients taking indinavir and those who
   17

     not.
did 18      As noted here, there were no differences in

     average values for Cmax, body clearance, lima
the 19

distribution or elimination half life between these two
   20

groups.
   21

   22         In another two patients, paclitaxel

kinetics were obtained on two consecutive cycles, one
   23

in the absence of indinavir and the second after two
   24

weeks of indinavir therapy.
   25                          As shown here, the plasma
levels of paclitaxel were similar with and without
   26

indinavir, confirming that indinavir does not alter the
   27
                                                      26

disposition of paclitaxel.
    1

    2       Imidazole antifungal agents are known to

inhibit cytrochrom P450 enzymes and it was of interest
    3

to assess whether those patients taking imidazole
    4

antifungal, primarily fluconazole, had greater
    5

exposure to paclitaxel.
    6

    7       On this slide it is clear that there was

no indication that patients taking antifungal had
    8

higher Cmax values or reduced clearance values compared
    9

to those not taking these drugs.
   10

   11       In conclusion, these studies define for

     first time the pharmacokinetics of paclitaxel in
the 12

AIDS KS patients taking multiple HIV therapies.
   13

Paclitaxel displays nonlinear pharmacokinetics over
   14

    range of 100 to 175 milligram per meter squared when
the 15

administered over three hours and there was no
   16

appreciable interaction between paclitaxel and
   17

indinavir or the imidazole antifungal agents.
   18                                             Thank

you.
   19

   20       Now I would like to ask Dr. Harriman to come

back to the podium.
   21

   22       DR. HARRIMAN:    First, I would like to

summarize study results relating to quality of life and
   23

patient benefit.
   24              Then I will review the safety

results, including the safety of Paxene in patients on
   25

protease inhibitors.
   26                  Finally, I will provide some

conclusions regarding the efficacy of Paxene in the
   27
                                                      27

treatment of patients with advanced AIDS KS who have
    1

failed prior cytotoxic chemotherapy.
    2

    3        In this context, failed refers to patients

who progressed on or were intolerant of the
    4

chemotherapy.
    5           In many cases, these patients have

failed more than one cytotoxic chemotherapy regimen,
    6

including Doxil.
    7              Such patients are an important group

for whom the identification of effective treatment can
    8

be challenging.
    9

   10        Quality of life was assessed by a

prospectively-obtained patient-administered Symptom
   11

Distress Scale as well as by Karnofsky Performance
   12

Status and photographs.
   13                     The Symptom Distress Scale

contains 15 questions related to overall well-being,
   14

     example, outlook, concentration and fatigue; as
for 15

well as disease-related symptoms, for example,
   16

appearance, pain, mobility and breathing.
   17

   18        Each question uses a five-point

Likert-type format in which a score of one is the best
   19

possible score, meaning no distress, and a score of five
   20

is the worst possible score, meaning severe distress.
   21

     Symptom Distress Scale was to be administered at
The 22

baseline and every third cycle.
   23                             Internal consistency

     test-retest reliability estimates have indicated
and 24

    scale is reliable and the scale has been previously
the 25
validated.
   26

   27        Karnofsky Performance Status was to be
                                                       28

assessed at baseline and each cycle.
    1                                     Photographs of

marker lesions and other involved areas were to be
    2

obtained at baseline and every six weeks.
    3

    4          Shown here is the median total score of all

15 questions for patients at baseline and cycles four,
    5

seven and ten.
    6              There was a highly statistically

significant improvement in the median score at cycles
    7

four, seven and ten.
    8                    Very few patients were lost

between baseline and cycle four, indicating that the
    9

improvement seen at cycle four, at least, is unlikely
   10

     to
due 11 bias.

   12          Assessment of tumor responses can be

difficult and open to a certain amount of
   13

interpretation, as Dr. Broder mentioned before.
   14                                               Thus,

it is possible for a patient to not be scored as having
    15

a tumor response, despite having clear evidence of
   16

clinical benefit.
   17

   18          Shown here is a patient previously treated

with Doxil.
   19          He had extensive involvement of his foot

with tumor and a large ulcer.
   20                            The patient was informed

that he might have to have his foot amputated.
   21

Following treatment with Paxene, the patient had a very
   22

significant improvement in the tumor and ulcer on his
   23

foot.
   24   This patient was not scored as having a tumor

response in this protocol, although he clearly
   25

benefitted from his treatment.
   26                               This patient and

others are with us today and they hope to have an
   27
                                                      29

opportunity to tell us about their experience with
    1

Paxene.
    2

    3         This patient had extensive lesions of his

gums.
    4   He also had a very severe lesion on his chest.

While there were some differences of opinion as to
    5

whether he was a responder, he clearly has had
    6

improvement in his disease.
    7

    8         Shown here are median scores in patients

with 9facial lesions for questions relating to the

patients appearance at baseline and cycles four, seven
   10

     ten.
and 11      There was a statistically significant

improvement in this score at cycles four, seven and ten.
   12

Again, few patients were lost between baseline and
   13

cycle four, indicating that the improvement at cycle
   14

four, at least, was unlikely due to bias.
   15

   16         As can be seen, this patient had severe

disfiguring lesions and edema on his face.
   17                                          With

treatment, he had a marked improvement in the lesions
   18

     edema.
and 19

   20         This slide shows improvement in symptoms

such as pain and mobility related to lymphedema.
   21

Again, there was a statistically significant
   22

improvement in these symptoms at cycle four.
   23                                            While

improvement continued at cycles seven and ten, it was
   24

no longer statistically significant.
   25

   26         This patient had marked lymphedema in his

right leg which responded well to treatment, with
   27
                                                       30

maintained improvement to cycle 13 as shown here.
    1

    2          This patient had severely crusted lesions

with 3significant lymphedema in his left lower

extremity.
    4        The lymphedema showed definite

improvement at cycle three of treatment.
    5

    6          This slide shows improvement in symptoms

related to pulmonary disease which include breathing
    7

and cough.
     8       A statistically significant improvement in

the median score was seen at cycles four and seven.
    9

Although a similar magnitude of improvement was seen
   10

at cycle ten, this was not statistically significant.
   11

   12          This patient had severe pulmonary

involvement and had previously been treated with both
   13

DaunoXome and Doxil.
   14                   Of note, he was on oxygen prior

to treatment, but was able to discontinue this
   15

treatment following the Paxene treatment.
   16

   17          This patient also had pulmonary

involvement.
   18            At cycle 13 of treatment, pulmonary

lesions were significant improved, as demonstrated by
   19

a decrease in one of the pulmonary lesions seen on this
    20

     of
cut 21 the CT scan.     Free study and cycle 13.

   22          Forty-six percent of patients had

improvement in their Karnofsky Performance Status
   23

during treatment.
   24                 The improvement seen was

statistically significant.
   25                          The majority of remaining
patients had no change in their Karnofsky status and
   26

a few patients had worsening.
   27
                                                     31

    1       Thus, improvement in quality of life was

seen 2in patients treated with Paxene as judged by

improvement in symptoms, by Karnofsky Performance
    3

Status and by photographic improvement.
    4

    5       With regard to safety, frequent

hematologic and non-hematologic adverse event
    6

occurring in the 89 patients are summarized here.
    7                                                The

major toxicities were hematologic, including
    8

neutropenia and anemia.
    9                     Other frequently occurring

adverse events included asthenia, alopecia, nausea
   10

and/or vomiting, arthralgis and myalgias, peripheral
   11

neuropathy and rash.
   12

   13       Adverse events were also analyzed by

whether or not patients were on protease inhibitors as
   14

shown on this slide.
   15                  There was little difference in

    incidence of adverse events between the two groups
the 16

of patients and none of the differences were
   17

statistically significant.
   18

   19       There were a total of 70 opportunistic

infections in 30 patients during study representing 34
   20

percent of patients.
   21                  Of these opportunistic

infections, 17 which involved mycobacteria,
   22

pneumocystic, cryptococcus and CMV would be considered
   23

serious.
   24

   25       There were 11 deaths which occurred while
patients were on study.
   26                     Of these 11 deaths, the

investigators felt four were related to Paxene.
   27                                             Three
                                                     32

of these patients of sepsis with associated neutropenia
     1

and one patient died of congestive heart failure due
    2

to pulmonary hypertension.
    3

    4       We also have substantial safety data with

Paxene using different doses and schedules in patients
    5

who have other forms of cancer.
     6                            Shown here are adverse

events, which were included in the NDA, on not only
    7

AIDS-KS patients, but an additional 226 patients who
    8

received Paxene at either 140 milligrams per meter
    9

squared over 96 hours or 175 milligrams per meter
   10

squared over three hours.
   11                       Again, the major toxicities

were hematologic.
   12

   13       Next slide.     However, alopecia,

arthralgia/myalgia and peripheral neuropathy were also
   14

fairly common, although severe grades of these
   15

toxicities were not common.
   16                          Hypersensitivity

reactions were also relatively uncommon.
   17                                      We currently

have safety data on a total of over 500 patients.
   18

   19       In summary, while AIDS-KS patients are

potentially at increased risk because of their
   20

underlying disease and multiple concomitant
   21

medications, no unusual or unexpected toxicities were
   22

observed in AIDS-KS patients treated with Paxene.
   23

   24       Now, I would like to summarize the data

which has been presented by responding to the questions
   25

which were addressed by FDA to ODAC.
   26                                    First, Is the

Paxene study size of 89 patients adequate for approval
   27
                                                        33

of a 1drug for the use after failure of first line or

subsequent systemic chemotherapy for the treatment of
    2

AIDS-related Kaposi's sarcoma?
    3

    4          To answer this question, this study must

be put into perspective with respect to studies which
    5

lead 6to the approval of other drugs for similar

indications.
    7           As discussed, the study reported here

was a prospective, multicenter study enrolling 89
    8

patients, with two geographically distinct sites, Los
    9

Angeles and Boston, enrolling 25 or more patients each.
   10

It should be kept in mind that all 89 patients had failed
    11

prior cytotoxic chemotherapy and many failed two or
   12

more cytotoxic chemotherapies.
   13                             Thus, these patients,

by and large, represent a very refractory population.
   14

   15          In looking at the study sizes for other

drugs currently approved for second-line treatment of
   16

AIDS-KS, there were two studies which were the basis
   17

upon which Taxol was approved for this indication.
   18                                                One

study, which looked at dose and schedule of 135
   19

milligrams per meter squared every three weeks,
   20

enrolled 29 patients.
   21                    However, only 19 of these

patients had received prior systemic therapy, of which
   22

only seven evaluable patients had received cytotoxic
   23

chemotherapy.
   24            Moreover, only four of these had

received an anthracycline.
   25

   26          The second Taxol study used a dose and

schedule of 100 milligrams per meter squared every two
   27
                                                        34

weeks.
    1    In this study, 56 patients were enrolled.

However, only 40 of these patients had received prior
    2

systemic chemotherapy.
    3

    4        The approval of Doxil for second-line

therapy in AIDS-KS was based on 77 patients who had
    5

received prior combination chemotherapy.
    6                                        However,

only 734 of these patients were felt by the FDA to be

evaluable.
    8

    9        Thus, the Paxene study containing 89

patients and representing a refractory population of
   10

patients, is larger than any other study used to support
   11

approval of a drug for second-line or subsequent
   12

treatment of advanced AIDS-KS.
   13

   14        Next slide.   The second question was,

"Does the Paxene study show patient benefit based on
   15

     42
the 16 percent cutaneous tumor response rate, the

clinical benefits assessments and the quality of life
   17

assessments?"
   18

   19        As previously discussed, the overall tumor

response rate with Paxene was 46 percent.
   20                                       Patients had

advanced AIDS-KS as demonstrated by the large number
   21

of patients with disfiguring lesions, tumor related
   22

edema and visceral disease.
   23                          In addition, the vast

majority of these patients were poor risk by TIS
   24

staging.
   25      Moreover, as mentioned previously, these
patients were a very refractory population with respect
   26

to prior cytotoxic chemotherapy.
   27
                                                       35

    1          Thus, the 46 percent tumor response rate

should be viewed as highly significant.
    2                                     The fact that

patients had substantial response rates, even after
    3

failing Doxil, which until August 4th of this year was
    4

the only approved drug for second-line treatment of
    5

advanced AIDS-KS and the significant response rates in
    6

patients who have failed two or more prior cytotoxic
    7

therapies, should be viewed as evidence of substantial
    8

activity.
    9

   10          Time to progression and duration of

response with Paxene were also substantial given this
   11

patient population.
   12

   13          Moreover, patients demonstrated

improvement in quality of life based upon significant
   14

improvement in total Symptom Distress Scale scores, as
   15

well as improvement in symptoms related to facial
   16

lesions, lymphedema and pulmonary disease.
   17                                            This is

     first time that a prospective quality of life
the 18

assessment containing such a Symptom Distress Scale has
   19

been used in AIDS-KS patients.
   20                              Significant

improvements were also seen in Karnofsky Performance
   21

Status and evidence of improvement was documented by
   22

photographs.
   23

   24          In sum, the combination of high tumor

response rates, as well as improvements in quality of
   25

life measurements, provide substantial evidence in
   26

support of patient benefit.
   27
                                                      36

    1         The third question, "Is the Paxene safety

acceptable in view of the efficacy results and results
    2

available with alternative therapy?"
    3

    4         Efficacy results were just discussed.

With 5regard to safety, this slide shows the most

important or most common adverse events with Paxene in
    6

comparison to adverse events reported in AIDS-KS
    7

patients treated with Taxol and Doxil.
    8                                    The point here

is that Paxene exhibited no higher incidences of any
    9

of the toxicities seen with Taxol and in some cases the
    10

rate may be lower.
   11

   12         As discussed earlier, in this study a

substantial amount of safety experience was gained with
   13

     coadministration of protease inhibitors and
the 14

Paxene.
   15     No significant differences were seen in the

rates of major or common adverse events in these two
   16

groups of patients.
   17                  Furthermore, pharmacokinetic

studies were performed to assess the effects of
   18

protease inhibitors on the pharmacokinetics of
   19

paclitaxel.
   20

   21         Thus, while Paxene has some significant

toxicities, as expected with this cytotoxic drug, it's
   22

safety is no worse and in certain adverse events may
   23

be better than Taxol, which is currently approved for
   24

second-line treatment of AIDS-KS.
   25

   26         The fourth question, "Is the Paxene NDA

approvable for the indication of use after failure of
   27
                                                     37

first-line or subsequent systemic chemotherapy for the
    1

treatment of advanced AID-related Kaposi's sarcoma?
    2

    3        Paxene demonstrates a high tumor response

rate 4in patients, all of whom have failed at least one

or more cytotoxic chemotherapies.
     5                              Moreover, the tumor

response rate is similar to that of Taxol when used at
    6

the same dose and schedule of 100 milligrams per meter
    7

squared every two weeks and is higher than that of
    8

Doxil.
    9

   10        Importantly, Paxene demonstrates

substantial tumor response rates even in patients who
   11

have failed Doxil.
   12                In contrast, only one patient

previously receiving Doxil was treated with Taxol in
   13

registration-seeking studies.
   14

   15        In conclusion, Paxene induces tumor

responses as defined by ACTG criteria in 46 percent of
   16

patients with advanced AIDS-related KS who had failed
   17

first-line or subsequent systemic chemotherapy.
   18

Paxene improves quality of life, as assessed by a
   19

Symptom Distress Scale and Karnofsky Performance
   20

Status.
   21     Paxene is also safe in the treatment of

AIDS-related KS.
   22

   23        Paxene induces tumor responses in 33

percent of patients who have failed prior Doxil therapy
   24

     41
and 25 percent in patients who received at least two
prior cytotoxic chemotherapies.
   26                               Paxene is safe and

effective in patients on concomitant protease
   27
                                                      38

inhibitors.
    1

    2         The proposed indication is Paxene is

indicated after failure of first-line or subsequent
    3

chemotherapy, including liposomal doxorubicin, in
    4

patients with advanced AIDS-related Kaposi's sarcoma,
    5

and for relief of disease-related symptoms.
    6

Coadministration with protease inhibitors does not
    7

diminish the efficacy or alter the side effect profile
    8

of Paxene.
    9

   10         I would now like to provide an opportunity

     some of the patients who have been treated with
for 11

Paxene to come up and share their experiences with you.
   12

Thank you very much.
   13

   14         MR. FLETCHER:   Good morning, ladies and

gentlemen.
   15         My name is Eric Fletcher.   I am not being

financially rewarded for being here today.
   16                                           I'm here

     of
out 17 a heartfelt concern.

   18         Since I was 15 years old, I have worked as

a fashion model.
    19             This allowed me to move away from home

at 17 to support myself through college and to pay for
    20

it and I was a taxpaying citizen where I contributed
   21

to society in general.
   22                     This was until two years ago.

   23         In the fall of 1995, I was diagnosed with

AIDS.
   24   More devastating was the fact that I had

Kaposi's sarcoma, KS.
   25                    After an endoscopy to show that
    KS
the 26 was rampant throughout my insides, after a couple

of weeks lesions began to appear all over my body.
   27
                                                          39

    1          My world began to collapse.   I was 30 years

old. 2 I relied on my physical appearance as the basis

of my existence.
     3              This was my means of livelihood.     Why

was I being tortured?
    4                     I had been completely healthy

all my life.
    5           I was a vegetarian.    I didn't smoke.    I

never did drugs or alcohol and I was not promiscuous.
    6

I wanted to know why this was happening to me.
    7

    8          My doctors immediately started me on

chemotherapy.
    9             This scared me because I had seen the

faces of people on chemo and in my experience those
   10

people didn't have a long chance of survival.
   11

Reluctantly, I started a clinical trial of Donozone.
   12

I was concerned about hair loss, but I was assured that
    13

this would not be a side effect.
   14                                 This made a vain man

happy.
   15

   16          I remained on the study for about six

months.
   17      I experienced nausea, vomiting, sleep loss,

loss of appetite, subsequent weight loss and a host of
   18

other problems.
   19               My heart infraction rate became too

low. I couldn't tolerate the drug any longer.
   20                                                  Early

in 1996 I had to stop treatments.
   21

   22          My doctors decided to start me on ABV.      I

    told that I would definitely experience hair loss.
was 23

Around this time I started to experience edema, my
   24

features grew beyond recognition, my lesions grew
   25

worse.
   26     They became open ulcers and wounds.     I needed

my bandages cleaned and changed three times daily.
   27
                                                         40

    1        I went from 170 pounds down to 125 pounds.

I couldn't walk.
    2               I used a wheelchair because I didn't

have 3the strength to move, or to bathe, or to even go

to the toilet.
    4              Obviously the ABV wasn't working.

    5        Needless to say, I gave up hope.   I reached

a low in my life I had never known.
    6                                    I considered

suicide.
    7      I asked my primary care provider about

assisted suicide.
    8                 I started to give away my life

souvenirs and treasures.
    9                         I prepared myself and my

loved ones for me death, or they prepared me.
   10                                           They were

so tired of seeing me suffer that they said that if God
    11

     ready and if I wanted to, that I could give up.
was 12

   13        My hopes, my dreams were all gone.      I

considered myself a monster.
   14                           I couldn't look at myself

in the mirror.
    15            KS had taken away my pride, my dignity.

   16        In all my misery, however, the one thing

that I didn't lose was my spirit.
   17                                My soul is good and

joyously in all my darkness I attracted many wonderful
   18

people into my life.
   19                    Many doctors, nurses and the

support system.
   20

   21        One of those doctors highly recommended

that I try this new protocol.
   22                            I had no choice.   It was

either Paxene, ICU or death.
   23                            At this point, what was

there to lose?
   24              My hair?

   25        I started Paxene in June of 1996 along with
a triple antiretroviral protease inhibitor therapy.
    26                                                   I

    my
cut 27 hair really short so I wouldn't see it fall out.
                                                          41

Surprisingly, my hair never fell out.
    1                                      In actuality,

I never experienced any side effects.
    2

    3          My doctors told me I wouldn't see the

effects of the triple therapy for about three months
    4

to a 5year.   However, after my first cycle of Paxene,

I began to see and feel a positive difference.
    6

    7          I am now up to my 30th cycle.   Treatments

are every two weeks.
    8                   My lesions have faded.   Many are

barely noticeable.
    9                  My ulcers have healed.    I have

regained all my weight, plus some.
   10                                  I have regained --

I have my normal energy level.
    11                            I am even running three

miles a day.
   12

   13          More remarkably, my appearance has

improved so greatly that I am back to work as a fashion
   14

model headed for a career in television.
   15

   16          Now, here is my plea.   Paxene is not

political with me.
   17                  Nor is it a miracle drug.    It is

simply my life.
   18              It may not be a cure for this dreaded

disease, but it makes life a whole lot more manageable.
   19

It has given me the ability to once again look in the
   20

mirror to see what's really there, a person full of life
   21

    love and has given me the ability to share that joy.
and 22

   23          I hope you will immediately approve Paxene

so many other people will have a chance to once again
   24

have dignity and self worth.
   25                            But more importantly, as
only a person who has seen the face of death will ever
   26

know, the true miracle of this drug is its ability to
   27
                                                          42

allow one to appreciate every moment that they once
    1

again have been granted and to lead a more fulfilling
    2

and rewarding life.
    3

    4           I greatly urge you to immediately approve

Paxene for the treatment of KS.
    5                                A small company like

Baker-Norton cannot survive another couple of years,
    6

therefore they will have to discontinue operations and
    7

I will no longer have the drug.
    8                                Ultimately, the

promise of my future will be taken away again.
    9                                               Thank

you.
   10

   11           MR. CAROL:   Good morning.   My name is

Steve Carol and I'm here today at the invitation of
   12

Baker-Norton Pharmaceuticals.
   13                              Although I am being

compensated for my expenses, I am here today to invite
   14

     to
you 15 share in my enthusiasm about a discovery I

happened upon during this past year.
   16

   17           My wife and I were devastated when I was

diagnosed with Kaposi's sarcoma in 1993.
   18                                          Further

tests confirmed that I was HIV positive.
   19                                        At that time,

my doctors followed the approved therapy for KS which
   20

began with radiation treatments.
   21                                 Although tolerable,

     therapy did little more than slow the progress of
the 22

     disease.
the 23

   24           After that treatment came injections of

interferon and interlukin 2.
   25                             Again, that provided to
do little to improve my situation.
   26                                   Next came the

systemic chemotherapy treatments beginning with ABV,
   27
                                                    43

three drugs that were used in different combinations
    1

but with limited success.
    2                       I had little tolerance to

the drugs and would have to discontinue the use of them
    3

after two or three cycles of each combination.
    4

    5       Next came my participation in several

studies involving the use of liposomal chemotherapies,
    6

including Doxil and Donozone.
    7                            Once again, my

intolerance to the long-term use of the drugs caused
    8

my doctors to discontinue any further treatments.
    9

   10       By this time my weight had dropped from 200

pounds to about 128.
   11                  My hair was just beginning to

return after having been lost to the chemotherapies.
   12

Up to this time, my skin lesions had been confined to
   13

my feet, legs and arms.
   14                     But now I had several facial

lesions that were drawing much attention.
   15

   16       Another lesion had ulcerated on the bottom

of my foot and had left a very painful opening about
   17

     size of a quarter that you could place your little
the 18

finger into up to the first joint.
   19                                 This left me for

a year and a half either on crutches or confined to a
   20

wheelchair and unable to work.
   21

   22       My doctors told me that there was nothing

more that they could do for me and that the only thing
   23

left to consider was the amputation of my right leg.
   24

This was not a measure that would stop the cancer, but
   25

would end the every day threat of infection to a wound
   26

that would not heal.
   27
                                                       44

    1         The ulcer was very large and ominous.   My

wife 2could not bear to look at it, even from across the

room.
    3   Special nurses had to come to my home to clean

and treat the wound on a daily basis.
    4                                       A one- legged

man was sent to our home to talk to us about life after
    5

amputation.
    6

    7         Not being the kind of person that gives up

easily, I found out about Dr. Seville and the study he
    8

was conducting at the University of California-San
    9

Diego of a new treatment for KS.
   10                               Although skeptical,

I became part of the study and began treatment in
   11

December of 1996.
   12

   13         After several cycles I noticed a number of

things.
   14      First of all, I didn't feel sick to my stomach

    the
all 15 time.    The lesions on my face were disappearing

     the wound on the bottom of my foot had begun to
and 16

improve.
   17

   18         I had none of the intolerance to the

treatments that I had previously experienced, and for
   19

     first time in years, I began to feel good about
the 20

myself.
   21      I no longer woke up angry every morning just

because I woke up.
   22                 I no longer felt helpless against

something that was slowly taking my life.
   23                                        And although

there was some hair loss again, I thought that was a
   24

small price to pay for something that was obviously
   25

working so well.
   26

   27         I can now report to you that I walk without
                                                        45

the use of a cane or crutches and there have been no
    1

new lesions to report for many months.
     2                                    The tumors that

I do 3have are greatly diminished.   And I went back to

work 4last month.

    5        I and the others that are appearing before

you today represent not only ourselves, but thousands
    6

of others who suffer from this disease.
    7                                        We depend on

governing bodies such as yourself to help advance the
    8

use of such life saving drugs as Paxene and allow us
    9

to enjoy the same quality of life that each of you enjoy
    10

every day.
   11

   12        I am here today to ask you to grant approval

to the use of Paxene in the treatment of KS.
    13                                         Thank you.

   14        MR. GREEN:   Good morning.    My name is

David Green and I am a 47 year old executive chef.
   15                                                   I

tested positive for HIV in 1982 and remained
   16

asymptomatic until January of '94 at which time I found
   17

my first KS lesion on my lower back.
   18                                     In four months,

I had six lesions on my body.
   19

   20        At the time I was living in San Diego and

     doctors there said they were not aggressively
the 21

treating KS unless it was presenting a serious problem.
   22

Mine were not as yet.
   23

   24        Over the next year, I developed many more

lesions over my torso.
   25                     In June of '95, several of the
lesions became raised and three of them had started to
   26

weep.
   27   I still had no treatment.
                                                      46

    1        By October '95, the dressings on the

weeping lesions had to be changed at least three times
    2

a day.
     3   The lesions were becoming quite tender.   I also

noticed at this time a slight discoloration on the tip
    4

of my nose and a swollen spot on my upper gum.
    5

    6        At Scribbs Clinic in San Diego, I saw an

infectious disease specialist who sent me for
    7

consultations with both radiation and hematology,
    8

oncology departments.
    9                    At that time a lesion was also

found on my lung.
   10

   11        The recommendation was radiation to slow

     growth in my mouth and wait and see on the rest.
the 12

Also, perhaps I should consider moving back to Boston
   13

to be with my family and to get my affairs in order.
   14

   15        It took three months to wrap things up in

     Diego and get to Boston.
San 16                          In that time the lesions

in my mouth grew quite rapidly.
    17                             Now both my upper and

lower gums had turned purple and had grown to completely
   18

cover my teeth.
   19             My hard palate had also grown and the

only way I could eat was to put very small pieces of
   20

food in my mouth and try to swallow.
   21                                   It was painful.

   22        Another lesion the size of a marble

appeared under my right ear and I was now getting short
   23

of breath without much exertion.
   24

   25        On arrival in Boston, I was referred to Dr.
David Skadden at Mass General Hospital.
   26                                      He told me I

     a
had 27 few options.   We decided that I would first try
                                                         47

Doxil.
    1     It had just been approved and he felt that it

was the least toxic and a good place to start.
    2

    3          After only two treatments, the pain was

gone 4and after six I started to notice some changes in

the lesions.
    5            They were shrinking.   At about three

months into treatment I could see the tips of my teeth.
    6

The weeping lesions on my torso were beginning to dry
    7

up. 8

    9          Slow progress continued until June of '96

when I had a breakthrough.
   10                          One of the lesions on my

right thigh had flattened and become -- which had
   11

flattened became raised again.
   12                              It also became quite

tender.
   13      The lesion on the tip of my nose began to

darken as well.
   14               However, from the time I started

treatment, I had developed no new lesions.
   15

   16          It was at this point that Dr. Skadden and

I decided I should try -- should join the clinical trial
    17

    Paxene.
for 18         With only one treatment, the raised lesion

    again flat and with two the tip of my nose lightened.
was 19

I really looked forward to going to treatments.
   20

   21          The treatments themselves are very easy to

tolerate.
   22         The worst part is the length of time you

spend in the chair.
   23                  The side effects are minimal.     I

lost body hair, eyebrows and eyelashes.
   24                                        I do need

nupegen to keep my white count out but the dosage has
   25

been reduced.
   26             The other side effects, hiccups,

constipation and heartburn are not due directly to the
   27
                                                         48

Paxene, but rather the decadeon I'm given as a premed,
    1

and they are easily taken care of.
    2

    3         I feel so well these days that after

receiving treatment I walk a mile and a half to the
    4

Boston Living Center where I volunteer.
    5                                       I continue to

receive Paxene every two weeks for a year.
    6                                            My mouth

is now normal.
    7             I still have teeth and most

importantly, my sense of taste is still acute.
    8                                                  The

lesions on my torso are flat and dry and fading.
    9                                                   My

lungs are clear.
   10

   11         In July '97, I went on a three week cycle

with continued fading of the lesions.
   12                                      I am now on a

four week cycle and the lesions continue to fade.
   13

   14         I never thought that I would feel or look

so healthy again.
   15                 There are not enough good things

that can be said about Paxene.
   16                                It's a drug which I

believe should be made available to everyone.
   17

   18         MR. MOLINA:   Hello.    My name is Jim

Molina.
   19      I am not being compensated for being here

today.
   20     Baker-Norton Pharmaceuticals has paid for my

ticket since I was unable to afford one on my own.
   21

   22         I was diagnosed HIV positive on April 19,

1993.
   23    Upon my diagnosis, I asked the doctor if the spot

on my left shin had anything to do with the HIV.
    24                                                 "Oh,

it looks like a little KS, nothing to be alarmed about.
    25

We'll just monitor it and see if it changes" he replied.
   26

   27         Not knowing what KS was, I figured the
                                                        49

doctor knew what was best for me, so I went along with
    1

his advice.
    2           Later a chest x-ray was requested by my

doctor.
    3      The x-ray revealed a quarter size lesion in

the lower left lung and a cat scan was ordered.
    4                                                 This

revealed the same results as the x-ray.
    5                                         Next I had a

bronchoscopy.
    6             The test was inconclusive as the doctor

was unable to get to the area of my lung that was in
    7

question.
    8

    9          So the next move was to try a fine needle

biopsy or to remove the lower half left of my lung --
   10

     lower left half of my lung.
the 11                               I was uncomfortable

with the invasiveness involved in both of these
   12

procedures, so I chose to monitor the lesion regularly.
   13

   14          Time passed to about March of 1994.   It was

then that my doctor had pointed out some enlarged lymph
   15

nodes on my neck that I had thought had been there
   16

forever.
   17       My doctor insisted on a biopsy of the lymph

node. The biopsy revealed that I had Kaposi's sarcoma
   18

in my lymphatic system on April 4, 1994.
    19                                        This news was

devastating.
   20            I knew what cancer was, but I did not know

anybody who had KS.
   21                   I was still dealing with the HIV

diagnosis and trying to come up with a way to break the
   22

     news to my mother.
HIV 23

   24          I was hit with both barrels.   I had so much

to do.
    25    I thought I was going to die and I had to come
clean with my mother who had already lost her only other
   26

child in an alcohol-related accident.
   27                                      Let me tell you
                                                      50

that 1was one of the most difficult things I ever had

to do.
    2

    3          I am so fortunate to have the support of

my mother and my lover Phil.
    4                           I don't know how I would

have 5come through all of this without them.   Little did

I know that was just the tip of the iceberg compared
    6

to the battle ahead.
    7

    8          Within a period of about six months, my KS

had begun to spread.
    9                    Slowly at first, then all of a

sudden it went rampant.
   10                      I watched as my body changed.

First there were only visible lesions.
   11                                     Then I noticed

my ankles were beginning to swell, then my legs, then
   12

I couldn't squat anymore.
   13

   14          Now during all these changes the doctors

at Kaiser were going through the routine with the
   15

available drug therapies.
   16                         On September 13, 1995, my

oncologist prescribed interferon which I had no
   17

response to.
   18           The only thing it did for me was make me

feel like I had the flu after each injection.
   19                                              That

lasted for about two months.
   20                            So my oncologist wanted

to try radiation on my groin and upper thighs.
   21

   22          I was under the impression the radiation

     helping as my skin began to fall the new skin was
was 23

unscarred.
   24          Little did I know the radiation also

damaging my lymphatic system in my groin area.
   25                                               This
     obviously not the answer since swelling in my feet
was 26

     ankles began to increase with each day.
and 27
                                                        51

    1          Then in December of 1995 the oncologist

tried etopacide which was quickly added to the list of
    2

options that were not working.
    3                              And then vincristine,

vinblastine.
    4            As time passed, it was March of 1996 and

my doctors at Kaiser had to inform me that there were
    5

no other alternatives.
    6                       They had done all they could

for me at Kaiser.
    7                 What a cold day that was for me.

    8          I was suffering, swollen and beginning to

lose 9all use of my legs.   I can't even describe to you

    mental state that I was in.
the 10                            I still had yet to meet

another person who was going through this.
   11                                         I began to

hide from the public, so aware of my lesions and their
   12

ugliness.
   13       I was ready to give up.    I became obsessed

with my death and how it was going to happen and at that
   14

time I feared death.
   15                    I was left to lay on the couch

in constant pain, just waiting, waiting to die.
   16

   17          I had a lot of time to think and in my

thinking I began to pray for the strength to get me
   18

through each day and the guidance to get me to someone
   19

     could help me or even relate to this new disease
who 20

that was changing me in so many ways.
   21

   22          Then on April 15, 1996, my prayers were

answered.
   23       The latest addition of Positive Living had

an article on the cover about KS.
   24                                 This was the first

instance where I saw anything related specifically to
   25

Kaposi's sarcoma.
   26                 I read furiously and found myself

in the clinical trial section which I had never paid
   27
                                                        52

attention to before.
    1                    And then I realized that I had

everything to lose by not opening my eyes to the
    2

alternatives.
    3

    4          I found only one trial that I thought I was

qualified for since I was in such an advanced stage.
    5

So I 6called and spoke with Miki Ilaw Jacobson.      She

seemed so interested in meeting me.
    7                                  I was happy to have

someone respond to me in such a positive way.
    8

    9          Miki told me that all of the other drugs

I had tried -- they had tried and rejected, were nothing
    10

in comparison to the current trial for Paxene.
    11                                            She was

confident that she could help me and I felt I could trust
   12

     from the beginning.
her 13

   14          I met Miki the following day and I will

never forget that day.
   15                      Miki had restored my hope in

living.
   16     I had to wait two weeks before I could start

treatment and those two weeks proved to be the most
   17

challenging.
   18            It seemed like the KS knew what was

coming.
   19     I began to swell up and the new lesions were

coming faster than I thought possible.
   20                                       It was like a

game of beat the clock getting to infusion day.
   21

   22          By the time May 8th arrived, my day of

infusion, I had begun to give up.
   23                                  I was so depressed

I was pushing the people in my life away to prepare for
    24

my death.
   25       It must have taken me 45 minutes to get
myself from my car to the clinic.
   26                                 I could barely walk

     I
and 27 had to rest often on that endless journey to the
                                                      53

clinic.
    1     But I finally made it and I received my first

infusion with this new drug, Paxene.
    2

    3        The next day after the infusion, I was

amazed.
    4     I woke to legs that were relieved of much of

the pain and for the first time in a long time my legs
    5

had reduced in size.
    6                  I could even see the veins in my

feet.
    7   I was so happy.   I called everyone I knew and

I told them of my progress.
    8                          And so far with each

subsequent infusion, I continue to get better.
    9

   10        There have been times when other

circumstances have prevented me from getting my
   11

infusion.
   12       Every time this occurred, the KS began to

bloom again proving to me that I need this therapy
   13

continuously.
   14

   15        I am so happy to say that I'm feeling better

than I have in over a year.
   16                         The combination of Paxene

     the new antivirals I am on have changed my once
and 17

losing battle to a battle worth fighting.
   18                                         I know now

that I am no longer alone.
   19

   20        My suffering has changed to a will to fight

back.
   21   Paxene has given me time to reopen relationships

with those I once pushed away and I have been given a
   22

second chance to live.
   23

   24        For me the side effects have been minimal.

I began to lose most of my hair, but suddenly it grew
   25

back with a vengeance.
   26                     I began to have severe

heartburn after infusion, but we've learned that it can
   27
                                                           54

be controlled with prozac.
    1                             I also get the hiccups

after infusion, but that I can deal with myself.
    2

    3          I would like to thank Baker-Norton, Dr.

Parkash Gill and Miki Ilaw Jacobson for their support
    4

and all they have done for me to help me in my fight
    5

against KS.
    6          I honestly believe that without them and

my loved ones, I would not be here today to offer my
    7

testimonial.
    8

    9          So I and my family urge you to approve

Paxene, not only for use in people with KS but for their
   10

loved ones as well.
   11                      Thank you.

   12          CHAIRPERSON DUTCHER:     Thank you.   We

certainly do appreciate the input from the patients.
   13

    must release this was on the time that was allotted
You 14

to Baker-Norton.
   15                How many more speakers do you have?

Because you have reached your time limit.
   16                                            One more?

   17          [Brief discussion off mike.]

   18          CHAIRPERSON DUTCHER:     Can we finish in

five minutes?
   19             Okay.

   20          MR. GRAY:    Good morning.   I have prepared

quite an extensive presentation, but I will make it
   21

short.
   22    My name is Gavin Douglas Gray and I am here,

     my
and 23 expenses are being paid by Baker-Norton

Pharmaceuticals.
   24

   25          In December of 1992, I had a medical
examination and went back to find out that I was in fact
   26

     positive.
HIV 27            I dealt with the situation as best as
                                                        55

I could and as one best can given the facts.
    1

    2        A year and a half later I was diagnosed with

AIDS-related Kaposi's sarcoma and was forced to quit
    3

my job and go on disability and began receiving
    4

chemotherapy treatments with ABV which failed me after
    5

a couple of months.
     6                 I went on to interferon and failed

on that and went on to Doxil and remained on that for
    7

about six months.
    8                 After 48 treatments of Doxil, my KS

condition advanced to an even more malignant stage and
    9

I was at that point 25 pounds underweight, emotionally
    10

depleted with very little hope and as many others have
   11

said, just looking to my death as the last solution to
   12

my situation.
   13

   14        I was put on Donozone and I did not respond

to that, and I heard about Paxene through a friend whom
    15

I did not recognize at the time because he looked so
   16

wonderful.
   17        He looked like a whole new person.

   18        I went on to try Paxene reluctantly because

what else was I going to do?
   19                             Try it or be done with

it. 20I had an incredible response to Paxene.     My

lesions started to disappear.
   21                             The pain went away.   I

     able to eat again and started to get my level of
was 22

energy back to normal and today you are looking at a
   23

completely different man than I was prior to Paxene.
   24

   25        I ask all of you that if this drug has
brought me back from the edge of my grave, then it should
   26

also be allowed to help many others who cannot make it
   27
                                                          56

to a 1parochial study, who are in rural areas of this

country that should be receiving it.
    2                                      Approval of it

     must for them.
is a 3                It's in your hands to restore hope

and to give back the life that many of those people once
     4

had like I once did.
    5

    6         I am grateful for this drug.     I highly

recommend it.
    7           And it's much, much more tolerable than

any of the other drugs that I tried.
    8                                      And it works

unlike any of the others.
    9                         Thank you.

   10         MR. BETTS:   Hello.   My name is Michael

Betts.
   11     I'm a California resident currently receiving

Paxene treatment in combination with protease
   12

inhibitor treatment.
   13                   My travel expenses have been

paid for by Baker-Norton Pharmaceuticals Inc.
   14                                              That's

     only compensation that I am receiving.
the 15

   16         I am here today to urge your approval of

Paxene as a chemotherapy treatment against Kaposi's
   17

sarcoma and last year around this time I was actually
   18

planning a funeral.
   19                  I wasn't sure I was going to make

it. 20But I feel much better now.

   21         To my knowledge, I've been HIV positive for

approximately seven years and in April 1996, after
   22

noticing an irregular swelling in my right ankle, I was
   23

diagnosed with KS.
   24                 Between the months of April and

July of 1996, the swelling increased from being just
   25

my ankle to my entire right leg.
    26                              I am a fairly active

person.
   27      I run and exercise quite a bit.     And I was
                                                       57

really disturbed by this reduction in my personal
    1

mobility.
    2

    3         Along with the swelling caused by the

lymphedema, I had no energy, I had heat that kind of
    4

emanated from my leg and I had bumps that secreted an
    5

oozing pus almost constantly.
    6                             I noticed that when my

stress level increased or when I had an increase in
    7

physical activity during the course of the day, the
    8

swelling was more pronounced.
    9                             It was painful even to

wear socks.
   10

   11         My leg felt as though it was going to

explode from the pressure and it felt like it was
   12

filling up with a fluid that was just going to burst
   13

     of
out 14 me at some point.

   15         I was bloated most of the time and

uncomfortable.
   16             And on one occasion, my leg enlarged

so much during the course of the day that I couldn't
   17

take my pants and my boots off.
   18                              I had to go to sleep

that way until my leg went down.
   19                                I had a lot of

difficulty bending at both my knee and my ankle.
   20

   21         During the same period, my skin became

blotched and the swelling was noticeable through my
   22

clothing.
   23        The evidence of my conspicuous appearance

     medical condition made me feel depressed and
and 24

reclusive.
   25         I remember one really important event.   I
went to the supermarket one day and a woman and her child
   26

followed me through the entire market trying to guess
   27
                                                       58

what 1kind of affliction I had, what was causing my leg

to be so big that they could notice it through my
    2

clothes.
    3      And it caused me to isolate myself.

    4        I was so isolated and withdrawn that I

completely stopped attending family functions.
    5                                             I

stopped doing anything that required being in public.
    6

And my neighbors gave me the nickname the "vampire"
    7

because I only did things at night.
    8

    9        There is a level of humor I think you have

to retain in order to survive an illness and its
   10

treatment.
   11        But when my body started to change against

my will, it was devastating.
   12                           I had lost control.    I

     to
had 13 question whether I could walk to the store,

whether I'd wear short pants, whether I could take the
   14

stairs and the insensitivity of other people.
   15                                            I kept

a strong exterior, but I was withdrawing.
   16

   17        Initially, I was introduced to Donazol as

a chemotherapy treatment, but it wasn't effective for
   18

me. 19And in November 1996, I had my first chemotherapy

treatment with Paxene.
   20                     Since that first treatment

I've experienced minimal side effects.
   21                                     The side

effects I had included hair loss, numbness in my toes
   22

     hands, dry mouth, hiccups, sleeplessness.
and 23                                           But

then I also get a real good burst of energy the day
   24

after, so that's great.
   25

   26        In contrast to the side effects, I've had

Paxene therapy every two weeks for the last ten months,
   27
                                                       59

and have had great improvement in my condition.
     1                                             My leg

is almost back to its normal size and I have periodic
    2

swelling only as a result of excessive exertion.
    3                                                 The

KS has not spread and I've been told that the
    4

discoloration in my skin will correct itself in time.
    5

    6          I'm energetic and my quality of life has

greatly improved.
    7                 I feel more like myself than I have

in the last two years.
    8                      I walk my 90 pound dog two or

three times a day.
    9                 I still work full time.   I work out

with weights and I have a shameless appetite.
   10                                               I eat

everything.
   11

   12          I have begun again to think about the

future and thoughts about a job, hobbies, changing my
   13

job, hobbies and it's been great.
   14                                  It's true that I'm

     as
not 15 able bodied as I was two years ago and I hope

     that, but I'm not dead either.
for 16                                 And I would gladly

accept the minimal side effects which are lessening all
   17

     time to the alternative.
the 18

   19          I'm not so terminally ill that joy is gone.

I have hope.
    20          I'm a living and breathing testament that

medical strides are being made against this villain
   21

that we call HIV.
   22

   23          There is nothing worse than feeling like

your body is at war with itself and Paxene chemotherapy
   24

     made me feel like the calvary really is coming.
had 25                                                  I
strongly support the approval of the use of Paxene by
   26

     Food and Drug Administration so that its benefits
the 27
                                                        60

can reach others in need.
    1                        Thank you.

    2        DR. HARRIMAN:    We very much appreciate

ODAC 3providing an opportunity for patients to present

their stories.
    4            That concludes our presentation and we

will 5be happy to answer any questions.

    6        CHAIRPERSON DUTCHER:     Thank you and

thanks again to the patients that came to present their
    7

stories.
    8      The Committee really does appreciate your

comments and your input.
    9

   10        We now have time for members of the

Committee to ask questions of the sponsor.
   11                                          Who would

like to begin?
   12            Would consultants like to start?      Dr.

Swain?
   13

   14        DR. SWAIN:    Could you just discuss the

concomitant use of the protease inhibitors and the
   15

timing with your study and the patients and if that had
   16

     effect on responses that you saw.
any 17

   18        DR. HARRIMAN:   Right.   We had 32 -- sorry,

33 patients who were on protease inhibitors at the start
    19

of their treatment with Paxene.
   20                               We had a total of 62

patients who were on protease inhibitors at some time
   21

during their treatment with Paxene.
   22                                  Those -- the other

patients, other than the 33 that were on at the start
   23

of therapy, were begun on protease inhibitors at some
   24

time during their treatment with Paxene.
   25                                         We had
another 27 patients who were not on protease inhibitors
   26

at any time during their treatment with Paxene.
   27
                                                           61

    1         If I could have, if I could have back up

slide no. 158, please.
    2                      If you just look at, just break

the groups down into just two -- two groups.
    3                                                The

patients who never received protease inhibitor at any
    4

time 5and patients who were on protease inhibitors at

some 6time and look at tumor response rates, you can see

-- you can see that response rates were about 57 percent
     7

in patients who were on protease inhibitors and 41
    8

percent in patients who were not on protease
    9

inhibitors.
   10

   11         I'm sorry.    These patients were on

protease inhibitors during the entire ten cycles of
   12

treatment.
   13         So this excludes patients that were

started on protease inhibitors after they were begun
   14

on the protocol.
    15             And these patients were patients that

were never on protease inhibitors at any time.
   16

   17         So if you look at those two groups of

patients, you can see that response rates were roughly
   18

comparable.
   19         And I think that suggests probably that

at least in this situation, the Paxene is able to induce
    20

tumor response rates of similar magnitude regardless
   21

of whether patients were on protease inhibitors.
   22

   23         CHAIRPERSON DUTCHER:      Could you just

comment a little bit about the lymphedema response and
   24

     many patients had significant lymphedema and how
how 25
responses were assessed and the response rate?
   26

   27         DR. HARRIMAN:    Right.    There were a
                                                        62

couple of ways in which we tried to assess the effects
    1

on lymphedema.
    2              One of them I discussed earlier and

that 3is the improvement in symptoms related to

lymphedema in which we did see, based upon the Symptom
    4

Distress Scale questionnaire, improvements,
    5

significant improvements in the patients symptoms
    6

related to that.
    7

    8          In addition, we had photographs.   The

investigators were encouraged to take photographs of
    9

     patients with lymphedema and try and document any
the 10

improvements in that.
   11                     We've shown you some examples

of those patients.
   12                  The completeness with which

photographs were taken were not 100 percent so we don't
   13

have documentation in every case.
   14

   15          The third way in which we tried to assess

improvement was by trying to get measurements of
   16

circumference of the extremities at baseline and during
   17

treatment with Paxene.
   18                     The -- although we did see, in

that situation, what we believed to be some evidence
   19

of improvement, there were problems with getting
   20

complete measurements on a consistent basis in the
   21

patients and we did not feel that the data was complete
   22

enough that we could present a meaningful analysis in
   23

that regard.
   24

   25          DR. JOHNSON:   I'd like to ask you to go
back to the question Dr. Swain asked regarding protease
   26

inhibitors and actually reshow the slide you just
   27
                                                         63

showed us.
    1         Because I want to be sure I understand.

You have a total of 50 patients on that slide.
    2

    3         DR. HARRIMAN:    Correct.

    4         DR. JOHNSON:    You had 89 in the study.

    5         DR. HARRIMAN:    Correct.

    6         DR. JOHNSON:    So I would conclude from

that 739 patients were not on protease inhibitors when

they 8started on Paxene and at some point during the

course of receiving Paxene were started on protease
    9

inhibitors.
   10

   11         DR. HARRIMAN:    Yes.

   12         DR. JOHNSON:    You don't give us the

response data of those 39 patients.
   13

   14         DR. HARRIMAN:   Well, they were included in

     previous slide as a group.
the 15                             But let me show -- can

     go
you 16 to the previous slide we showed -- 157 -- no

that's not it.
   17

   18         DR. CARRIER:    I'm Steve Carrier, Director

of Biometrics at Baker-Norton.
   19                                 There is a little bit

of a competing risk thing going on here with the
   20

protease inhibitors start date and the response date
   21

    the Paxene.
for 22            The slide that you showed 21 patients

     were on protease inhibitor at the beginning of the
who 23

study and used protease inhibitors during the entire
   24

    cycles of the study during which by protocol we were
ten 25
determining best response, is a peer group in which we
   26

could look at response rates in the presence of protease
   27
                                                       64

inhibitor.
    1

    2        The other group of 29 did not receive any

protease inhibitor during that ten cycles and so we had
    3

a fairly peer comparison of response rates in groups
    4

that 5had the only difference being the presence of a

protease inhibitor.
    6                 The additional 39 patients began

protease inhibitor at some time during the study or
    7

previous to the study, but had not been on protease
    8

inhibitor the entire study.
    9

   10        So 33 patients began the study with

protease inhibitor, but 21 of them continued throughout
   11

basically the ten cycles.
   12                        Others changed, stopped,

paused, had breaks, new ones began.
   13                                  And those are

problematic as to when to -- to which group do you
   14

attribute the response?
   15                     Do you attribute it to

patients who respond early in Paxene and have not yet
   16

received protease inhibitor are fairly clear.
   17                                             But,

    you are conditioning your response on having after
now 18

protease inhibitor introduction on those who were
   19

unable to respond prior to the -- and there is no clear
   20

answer to that.
   21

   22        We have, however, attempted to -- we've had

a lot of discussions internally about this as you might
    23

guess, to look at this and so we've done some Cox
   24

regression analyses with the introduction of protease
   25

inhibitor as a time dependent co-variant in this model
   26

     we
and 27 wanted to know whether or not the introduction
                                                     65

of protease inhibitor increased or reduced the risk of
    1

an outcome variable.
    2                    And those variables were:

    3       time to response;

    4       time to progression of the disease where

us to follow-up or death without any knowledge of
    5

whether the Kaposi's sarcoma had advanced were censored
    6

as opposed to counted as events;
    7

    8       time to treatment failure where all loss

to follow-up, all deaths, and all progressive diseases
    9

were counted as events; and
   10

   11       mortality survival itself.

With the results that for time to response there was
   12

no significant effect on the response rate with the
   13

introduction of a protease inhibitor relative to not
   14

having a protease inhibitor on board.
   15

   16       The relative risk was about two with

confidence bounds of about .93 to 4.6 having protease
   17

inhibitor on board versus not having protease inhibitor
   18

on board.
   19

   20       For time to progressive disease, we didn't

really see a significant effect at all.
   21                                      The relative

risk was about 1.1 with confidence bounds of .35 to 3.3.
   22

However, when we finally get to time to treatment
   23

failure, which includes the mortalities now, the
   24

relative risk is down to .43 meaning a 57 percent
   25

reduction in treatment failure with the introduction
   26

of protease inhibitor.
    27                    Confidence bounds were .232 to
                                                       66

.797 1and a P value was 0.007.

    2         And finally the mortality where I think

this 3is consistent with everybody's expectations, the

relative risk is down to .266, the P value associated
    4

with 5is 0.0015 and confidence bounds are 0.23 to 0.80

with 6the risk being reduced by the introduction of

protease inhibitor or that -- over not having that
    7

protease inhibitor introduced into the patient
    8

population.
    9

   10         Thank you.

   11         DR. JOHNSON:    So do I understand from your

     regression analysis, are you -- did you just tell
Cox 12

us that the time to response was better --
   13

   14         DR. CARRIER:    The time to response was not

better.
   15     Whether you respond or not was not better.

     as
But 16 you begin to introduce the end points of life,

    mortality itself, then the introduction of protease
the 17

inhibitor reduced the risk of having the negative end
   18

point, a prolonged life, prolonged a time to -- before
   19

treatment failure occurred.
   20

   21         DR. ABOULAFIA:    Do you have any

information about viral loads on these patients who
   22

were recruited in these studies?
   23

   24         DR. HARRIMAN:    Yes, as part of the

protocol design, viral loads were not assessed and
   25

that's primarily because the onset of the study was at
   26

a time where that was being done less routinely.
   27                                                  We
                                                        67

do have some sporadic measures of viral loads and if
    1

we could just show some of these.
    2                                    If you could go to

-- okay, here is patient 856.
     3                           This is his viral loads.

At prestudy at cycle five and two measurements I guess
    4

at different times at cycle six.
    5

    6        Can we also see number 187 please.     Oh, I'm

sorry, here is another one, a patient whose viral loads
    7

were 8done at pre-study cycle nine and cycle 14.       And

190, 9and here is another patient whose viral loads were

done at cycle four, 13 and 16.
   10                              So that gives you just

a very sporadic information about viral loads.
   11                                                 But

again, that wasn't part of this protocol design and it
   12

     --
was 13 the protocol was undertaken primarily before

these were being done routinely.
   14

   15        DR. ABOULAFIA:     What would be interesting

to know, not so much what the effect of Paxene is on
   16

viral loads, but what the response rates are on patients
   17

     have non-detectable loads.
who 18                              Not using a protease

inhibitor is a surrogate marker --
   19

   20        CHAIRPERSON DUTCHER:     Use the microphone.

     need to use the microphone.
You 21

   22        DR. ABOULAFIA:     Sorry.    What I was saying

is it would be interesting to know what the effect, not
    23

of what Paxene is on viral loads per se, but the response
    24

rates of patients who had non-detectable viral loads
   25

versus those who had poorly controlled viral loads.
   26                                                   Do

     have any kind of data like that?
you 27
                                                      68

    1        DR. HARRIMAN:   Given the fact that we

really, as I said, have only sporadic measures of viral
    2

loads, we don't have any data that would substantively
    3

address your question.
    4                     What I -- just to try and get

at it indirectly though, what I would show you, if I
    5

could have slide 151 please.
    6

    7        One of the points to make in this is that

when 8patients respond to Paxene, and then this is

basically just a figure that's showing the percent of
    9

patients who were responders who responded at different
   10

cycles from zero through nine.
   11                             And, once patients are

begun on treatment, there is a fairly rapid increase
   12

in the number of responders.
   13                            It turns out the median

cycle of response is cycle three.
   14

   15        Now this is -- we feel at least some

evidence to suggest, given the fact that the patients
   16

were often begun on protease inhibitors at various
   17

times during treatment, that this very rapid increase
   18

in response suggests at least that the response we are
    19

seeing is primarily an effect of the paclitaxel, the
   20

Paxene itself, rather than at least not -- at least in
   21

part due to the Paxene and not entirely due to the
   22

introduction of protease inhibitors.
   23

   24        Moreover, if I could have slide 159 please,

this is a graph showing the rate of response in patients
   25

    were not using protease inhibitors.
who 26                                    And again, you

    a
see 27 pretty rapid response here as the patient receives
                                                        69

additional cycles.
    1

    2          If I could have the next slide please.

And, 3this is a slide of patients who were using protease

inhibitors and I think the two curves are fairly similar
    4

and again, I think it's indirect evidence but at least
    5

it suggests that the protease inhibitors are certainly
    6

not entirely, and we don't feel largely responsible for
     7

the responses that we are seeing.
    8

    9          DR. MARCO:   Can I do two follow-ups on the

protease inhibitor questions?
   10                              One, do you have a

breakdown by protease inhibitors?
   11                                    Somebody on hard

     sequenivere monotherapy versus somebody on triple
cap 12

therapy with indinavir is going to be different.
   13                                                  So,

if you could show us that.
   14

   15          And also I don't know if you really can

answer this, but in your NDA 6.8 tumor response by
   16

concomitant protease inhibitor use, that's a complete
   17

flip from what you are just showing now.
   18                                          Originally

     were telling us that patients on protease
you 19

inhibitors did worse, albeit not statistically
   20

significant, than patients not on protease inhibitors.
   21

What's the reason for the switch?
   22

   23          DR. HARRIMAN:   I don't think we ever said

in any documents that we thought that patients did
   24

worse.
   25    I think --
   26          DR. MARCO:   Not worse.    I said not

significant.
   27            But you say the success rate of 79.2
                                                        70

percent in patients not on protease inhibitors.
    1                                                And

you say a response rate of 50 percent on patients on
    2

protease inhibitors.
    3                   Even though it's not

statistically significant, these numbers on the slide
    4

are different.
    5

    6        DR. HARRIMAN:   Yes.   Okay.   Two points to

make.
    7   One, this is again, we did not feel those numbers

were 8not statistically significant.    We did not feel

that 9they were significant.

   10        Number two, part of the reason for the

confusion, the numbers that you saw were based on an
   11

analysis that we had done prior to getting our
   12

independent confirmation by Dr. Kaplan who reviewed all
   13

89 cases and did an assessment of tumor responses as
   14

well as cycle in which tumor responses occurred, and
   15

also the point at which progression occurred.
   16                                              We

reanalyzed our database using only Dr. Kaplan's
   17

independent assessment of our tumor responses and timed
   18

to progression and time to response.
   19

   20        So the numbers that I'm showing you here

today are based solely on Dr. Kaplan's analysis which
   21

I think accounts for the reasons there is a difference
    22

between that and the numbers that you see in the ODAC
   23

briefing document.
   24                 The analysis that we did with Dr.

Kaplan's numbers were actually done subsequent to the
   25

submission of that briefing document.
   26

   27        DR. MARGOLIN:     I have a few questions
                                                         71

related to the assessment of the durability of the
    1

responses and concern about long term therapy.
    2                                                The

first one is I think it's somewhat untraditional to
    3

assess the duration of response beginning of the onset
    4

of therapy rather than at the onset of some
    5

documentation of response.
    6                        But I'll just make that as

a rhetorical comment because obviously the FDA has
    7

looked at that question, I'm sure.
    8

    9        The question I have that I don't think was

in the data you presented was you gave a median time
   10

to treatment failure, I think of 234 days.
    11                                        But reasons

     off study -- I think it would be useful to see how
for 12

many patients went off study because they relapsed
   13

among responders and/or a Kaplan Meir plot of what's
   14

happening to the responders over time.
   15                                        Because a

duration of response not reached doesn't really tell
   16

us what's happening with at least some of these
   17

patients.
   18

   19        And then the related questions would be in

patients who responded but who had a brief response then
   20

relapsed, do you have any data about retreatment?
   21

   22        DR. HARRIMAN:   Okay, yes, very good

questions.
   23        In terms of discontinuations, if I could

have slide 128 please.
   24                    These are the reasons for

discontinuation, either in patients who received
   25

greater than two cycles or patients who -- greater or
   26

equal to those and patients who received less than two
   27
                                                          72

cycles of therapy.
    1                  The 15 patients who discontinued

treatment after two cycles of therapy, two were for
    2

death, two for toxicity, one for disease progression,
    3

two refused further treatment, and eight for various
    4

other reasons such as the patient moved or switched
    5

doctors and so forth.
    6

    7          In patients who discontinued therapy prior

to receiving two cycles of therapy, three of the
    8

patients were because of death, one was lost to
    9

follow-up, one refused further treatment and seven,
   10

again, were other, which was the various reasons that
   11

I indicated.
   12

   13          In terms of the Kaplan Meir, we don't have

that analysis but Steve, do you want to say anything
   14

in terms of the calculation of duration to response?
   15

   16          [Pause.]

   17          DR. OZOLS:   I have a question about

safety.
   18     You said the safety profiles in some instances

    be
may 19 better than Taxol.      Does that relate to possible

     of
use 20 protease inhibitors?       Do you have different

toxicity profiles for use with and without the
   21

inhibitors?
   22

   23          DR. HARRIMAN:    We feel that, first of all

    two products, Taxol and Paxene, although they both
the 24

contain the same active moiety, are different
   25

proprietary preparations with different formulations.
   26

Although we cannot address that specifically, it is at
   27
                                                          73

least a possibility that some differences in side
    1

effect profiles may be related to differences in
    2

formulation.
    3

    4          In terms of the possible role of protease

inhibitors, I think you know, it remains I think a
    5

question that cannot really be fully answered right
    6

now. 7 I think some of the side effects, adverse events,

the difference is that we observed would seem to be at
    8

least at first blush not likely attributable to the
    9

protease inhibitors, for example, differences in
   10

arthralgia, myalgia.
   11                       But I think, I really can't

comment further than that.
   12

   13          DR. OZOLS:    And then the other question in

your proposed indication for failure of first-line or
   14

subsequent chemotherapy, would that include Taxol?
   15

   16          DR. HARRIMAN:     There have been some

studies that have been done, and we actually have as
   17

an amendment to our protocol in patients who progressed
    18

on three hour infusions of paclitaxel to enter them into
    19

a protocol which uses 96 hour infusions of Paxene.
   20

There were small numbers, I guess Dr. Seville in his
   21

study when he was at the National Cancer Institute, had
   22

studied small numbers of patients that had progressed
   23

on three hour infusions and found some evidence of
   24

efficacy in those patients when they were switched over
   25

to 96 hour.
   26

   27          As you probably are aware, also, there are
                                                       74

studies ongoing looking at 96 hour infusions of
    1

paclitaxel in patients who have failed three hour
    2

infusions of paclitaxel or other chemotherapy.
    3                                               So, I

think that's a possible area that would be worth further
    4

evaluation.
    5

    6         DR. SCHILSKY:   I've just a couple of

questions.
    7         As I understand the spots criteria used in

the study, it's possible for patients who have visceral
     8

disease and cutaneous disease to be scored as a response
    9

just by virtue of improvement in the cutaneous disease.
   10

And, most of the examples of response that you showed
   11

us were patients who responded with their cutaneous
   12

disease.
   13

   14         Can you tell us something about what the

response is in visceral sites in patients who are
   15

getting this therapy?
   16

   17         DR. HARRIMAN:   Yes.   First of all, in many

of the patients, although some of them had evidence of
    18

visceral disease at the time they were entered into the
   19

study and the clinicians had indicated in the case
   20

report forms that the patients had various visceral
   21

disease, in order for a patient to be followed
   22

specifically for precise specific tumor response, they
   23

     to
had 24 have clear documentation, confirmation that

disease was present.
   25

   26         For example, in pulmonary disease they

would have had to have a bronchoscopy done prior to the
   27
                                                         75

study and documenting KS.
    1

    2       DR. SCHILSKY:     In your presentation you

said 3there were 37 patients who had visceral disease.

    4       DR. HARRIMAN:     Yes.   And, many of those

patients that's based upon the clinical diagnosis or
    5

the clinician's impression at study entry.
    6                                           We had

seven patients who were being followed for pulmonary
    7

disease specifically in whom attempts were made to
    8

document it, prestudy with bronchoscopies and so forth.
    9

     of
And 10 those seven patients that we had confirmation

of that, I believe it was five of those patients had
   11

evidence of response in their pulmonary disease.
   12

   13       DR. SCHILSKY:     I also have a question

about the pharmacokinetics.
   14                          I was just curious about

a couple of things.
   15                  One is that other

pharmacokinetics, or other PK studies of paclitaxel
   16

have suggested that the most relevant pharmaco-dynamic
   17

parameter is duration of exposure above the threshold
   18

concentration.
   19             I see that you didn't present data on

that particular parameter.
   20                         And I wonder if you even

    generate that since the patients were only studied
can 21

after 48 hours.
   22

   23       But it may be that the AUC and Cmax and so

on are not particularly relevant PK parameters given
   24

     way the drug seems to work.
the 25                               So do you have -- is
there any data on duration of exposure above seven
   26

threshold concentration?
   27
                                                         76

    1       DR. HARRIMAN:     Ken?

    2       DR. DUCHIN:     We didn't look at that

specifically because about half the patients in the PK
    3

analysis were taking nupegen at the time.
    4                                         So, we felt

that 5would confound the analysis.

    6       DR. SCHILSKY:     Why would that confound the

analysis if that's what the concentrations were?
    7

    8       DR. DUCHIN:     Because when we looked at

change on the neutrophil count --
    9

   10       DR. SCHILSKY:     I'm not asking you to

relate it to any clinical parameter, I just want to know
   11

if you have data on, you know, number of days or number
    12

of hours with a concentration above the threshold
   13

value.
   14

   15       DR. DUCHIN:     Oh, we have that, but we don't

have it today.
   16

   17       DR. SCHILSKY:     Okay.   I have one other

question before you go about the PK and the impact of
   18

    protease inhibitors.
the 19                      From the data that you showed

     it
us, 20 doesn't appear that there is any alteration in

     PK, which, I guess, is a little bit surprising to
the 21

me. 22But I wonder if -- I just want to be clear that

when the PK studies were done, was the only variable
   23

in a sense whether patients were getting protease
   24

inhibitors or not, or were patients also getting all
   25

of the other drugs that they were getting, plus or minus
    26

     protease inhibitors?
the 27                       Because it could be very
                                                       77

difficult to sort out the PK data and try to dissect
    1

out the impact of the protease inhibitors in the
    2

presence of multiple other drugs, others of which may
    3

have 4an influence on various cytochrome P450s.

    5       And so what you are looking at, I presume,

is a 6resultant effect and it is certainly conceivable

to me that effect may not actually reflect the actual
    7

impact of the protease inhibitors themselves.
    8                                               So do

you have any sort of more pure way of looking at the
    9

data?
   10

   11       DR. DUCHIN:   The purest way that we have

    in
are 12 those two patients where I presented, and clearly

     only change was the addition of indinavir.
the 13

   14       DR. SCHILSKY:    So all of the other

medicines that they were taking over that course of time
   15

remained constant?
   16

   17       DR. DUCHIN:   Yes, that's correct.

   18       DR. SCHILSKY:    Okay.

   19       CHAIRMAN DUTCHER:      Dr. Northfelt?

   20       DR. NORTHFELT:   Thank you.    Dr. Harriman,

I noticed that in the afternoon, after the FDA evaluator
    21

makes his presentation you are not offered any
   22

opportunity to rebut.
   23                   So I'd like to ask you to rebut

    statement that's made in the material I read.
one 24                                               And

I just want to tell you what my bias is in this so that
    25

     know where I'm coming from.
you 26

   27       As a clinician I don't think that the
                                                      78

objective tumor response criteria that we are using in
    1

these studies has any real value, especially in people
    2

with 3very advanced disease.   So I think really what we

need 4to understand is how these treatment impact on the

quality of life of these patients and what the clinical
    5

benefits are aside from the objective response
    6

criteria.
    7

    8       So I was very happy to see that you had done

a quality of life analysis in this study and that you
    9

    show some improvements in several areas.
did 10                                         But that

    not, that enthusiasm was not shared by the reviewer
was 11

from the FDA.
   12           So in part he says that results of the

analyses of the SDS components and the total SDS score
   13

should be interpreted with caution due to the lack of
   14

a control group in the study.
   15                            And then he goes on to

     for the same reason the impact of missing data
say 16

cannot be adequately assessed, thus no claims for
   17

improvement can be validly made.
   18                               And this is respect

to quality of life again.
   19

   20       Only statements pertaining to trends

toward improvement are supportable and he ends by
   21

saying the approval decision should be based only on
   22

clinical considerations of this application.
   23                                           So he is

essentially asking us to ignore all of the quality of
   24

life information that you have presented.
   25                                       And I'm sort
of crushed by that so I'd like you to get me back up
   26

again.
   27
                                                     79

    1        DR. HARRIMAN:   Okay, I'll see if I can do

that.
    2   First of all, again, I just want to emphasize

that 3this is, to our knowledge, the first time that an

attempt at taking a quality of life instrument, a
    4

Symptom Distress Scale, and studying it prospectively
    5

in advanced AIDS-KS patients with the attempt being to
    6

try and determine whether there is a feeling of
    7

improvement on the part of the patients of their
    8

symptoms.
    9

   10        Now, it's a fair statement to say short of

a head to head randomized comparison study, one can
   11

always argue that there could be a placebo effect in
   12

other things that would bias the patient in their
   13

responses.
   14        However, with respect to that, one point

that I tried to make during my presentation and I hope
   15

it was made, but in the figures that I showed -- let
   16

me go to the one on the facial -- the point I was trying
    17

to make is that when one looks at baseline median scores
    18

     at
and 19 least at cycle four where there is a highly

statistically significant improvement, the difference
   20

in the number of patients -- it was 29 that had
   21

evaluations at baseline and 27 at cycle four, so very
   22

    patients were lost between baseline and cycle four
few 23

in this case, although clearly patients get lost as the
    24

study goes on.
   25

   26        Now, I guess what I would argue is that

given the fact that you had very little loss in patients
   27
                                                         80

between baseline and cycle four, it's hard to argue that
    1

the worst patients are dropping out and you are only
    2

looking at the better patients that are still there at
    3

cycle four.
    4          So, for that reason I think one could argue

that 5this difference is probably real and meaningful.

    6          About all --

    7          DR. BRODER:    We understand and deeply

respect the FDA's review and we understand their
    8

comments.
    9         Our position is that we do not agree with

their assessment.
   10                 The prior attempts at these types

of assessments have been retrospective and essentially
    11

in effect an attempt to do quality of life assessments
    12

looking back in time, essentially after a study has been
   13

completed in many cases.
   14

   15          And so recognizing all of the potential

limitations that one might have, I guess the simple
   16

bottom line is that this was a prospective study with
   17

statistically significant results, at least at certain
   18

parameters and at certain time points.
   19                                        And it must

constitute an improvement over other previous attempts
   20

to make these quality of life assessments.
   21

   22          So with respect to the FDA on this specific

point, we disagree.
   23

   24          DR. SIMON:    I had a few questions, one on

pharmacokinetics.
   25                 Did you try to assess whether
Paxene affected the pharmacokinetics of the protease
   26

inhibitors?
   27
                                                      81

    1        DR. DUCHIN:    Yes.   In one study where we

had two patients that were done, we did look at
    2

indinavir levels in only a few samples.
    3                                      And they were

within the expected range for indinavir.
    4                                         But we did

not do a standard profile of indinavir concentrations.
    5

    6        DR. SIMON:    Do you have data relating the

objective tumor response to the symptomatic
    7

improvement on the Symptom Distress Scale for baseline
    8

to course seven?
     9             I think once you get beyond -- course

four -- I think once you get beyond course four, I guess
   10

my view is there is so many patients lost from
   11

evaluation that that data and those P values are not
   12

valid.
   13    But do you have a correlation of response,

objective response versus symptomatic improvement over
   14

     first four courses?
the 15

   16        DR. HARRIMAN:    Right.   We had done some

initial analyses in trying to correlate tumor response
   17

with improvement in the Symptom Distress Scale and,
   18

although I don't think we have that information here
   19

today with us, we did not see or rather we saw similar
   20

improvements in Symptom Distress Scale scores in
   21

patients who responded -- Steve?
   22

   23        DR. CAROL:    The data we have is based upon

     internal tumor response rate data and what we did
our 24

find was that even in non-responders there was
   25

reduction in the symptom distress score, the
   26

symptomatology in that first baseline to cycle four.
   27
                                                        82

And we couldn't distinguish it in that period from the
    1

drop, the median change we saw in the responding group.
    2

That 3was by our internal assessment of response.       We

haven't repeated that using the independent reviewer's
    4

assessment.
    5

    6         DR. SIMON:   One final question.    You

present in your application an analysis of Karnofsky
    7

performance data over time.
    8                         I guess I don't understand

how that's a valid analysis given that over time,
    9

particularly these patients taking protease
   10

inhibitors, their performance is going to improve and
   11

certainly when you take their last performance score
   12

     those who go off study early because they are not
and 13

responding to their HIV treatment are going to have
   14

lower -- not going to have improved performance scores
   15

     those who stay on longer because for a variety of
and 16

reasons are going to have improved performance.
   17                                              Maybe

as a result also of their anti-HIV treatment.
    18                                            I don't

    how
see 19 you can attribute that significant improvement

in Karnofsky performance, how you can attribute that
   20

to treatment with Paxene?
   21

   22         DR. HARRIMAN:   I don't -- I mean I take

your point and I'm not sure we are trying to argue that
   23

    entire improvement in Karnofsky performance status
the 24

is simply a consequence of treatment with Paxene.
   25

   26         However, I think what one can discern from

that data, I think, and this is, I think, an important
   27
                                                      83

piece of information to gain, is that certainly the
    1

treatment with Paxene is not causing a deleterious
    2

effect on the patients' Karnofsky performance status.
    3

Moreover, notwithstanding the improvement or the
    4

effects of -- possible effects of protease inhibitors
    5

and other variables on the results that we are seeing
    6

in Karnofsky performance status, I think in light of
    7

all of the other evidence that we have shown you, I think
     8

it's 9reasonable to conclude that perhaps at least some

of the improvement would be attributable to the study
   10

drug.
   11

   12        DR. SWAIN:   In the FDA document, it was

stated that there is 75 percent of the patients had a
   13

    week delay and about 40 percent had a two week delay
one 14

     you are recommending to give this drug every two
and 15

weeks, but it seems like most of the patients really
   16

didn't receive it every two weeks.
   17                                 Can you comment on

what's in there that there is no reason for that delay
   18

in a large number of patients and how as practitioners
    19

using this drug, it should be used?
   20

   21        DR. HARRIMAN:   The way the protocol was

designed is the patients were to have received the
   22

Paxene at two week intervals.
   23                           And that was adhered to

as much as possible during the first ten cycles of
   24

therapy.
   25

   26        It was, actually at the prompting of the

investigators, their feeling was that in every two week
   27
                                                      84

-- after ten cycles of therapy or after the patient has
    1

responded to the Paxene, that it's very inconvenient
    2

for the patients to come in every two week intervals
    3

to get the treatment and the hope was that one could
    4

increase the interval between cycles and still maintain
    5

responses.
    6        So we modified the protocol to allow for

intervals up to three to four weeks between cycles after
    7

ten cycles of therapy.
    8                     In fact, a number of patients

who had completed ten cycles of therapy went to that
    9

every three or four week schedule.
   10

   11        We don't have any really comprehensive

data in terms of being able to discuss whether patients
   12

that continue on an every two week regimen after cycle
   13

    or
ten 14 those that go to every three or four weeks whether

there is any difference in the time to progression or
   15

     rate of progression, because that wasn't really
the 16

part of the original protocol design.
   17                                   But I think it's

an interesting question that perhaps merits further
   18

exploration, and that is after the patient has had a
   19

tumor response, is it possible to increase the interval
   20

between cycles and still maintain responses.
   21

   22        DR. SWAIN:   And the second question.   Can

     discuss the hepatotoxicity with and without the
you 23

protease inhibitors that you saw?
   24

   25        DR. HARRIMAN:   Yes.   [Pause.]   We don't
have that data summarized, but we can certainly get that
   26

    you.
for 27     But unfortunately, we don't have it available
                                                      85

today.
    1

    2       What I can say is that there was certainly

no -- well, as you know patients on protease inhibitors,
     3

particularly indinavir, can have elevations in
    4

bilirubin and in some cases there appear to be a small
    5

number of patients on indinavir that can have
    6

concomitant increases in their transaminases as well
    7

as bilirubin.
    8           We did see some patients who were on

indinavir who had elevated bilirubin levels, but I
    9

don't have available at this point any data that
   10

specifically compare patients that were on and off
   11

protease inhibitors with regard to that.
   12                                        Sorry.

   13       DR. SCHILSKY:    Could I just follow up on

that for a moment because there is a substantial amount
   14

of data to suggest that patients with abnormal liver
   15

functions don't tolerate paclitaxel well and that even,
   16

     know what may be relatively trivial elevations of
you 17

transaminases may predispose patients to much more
   18

severe toxicity.
   19

   20       So it would seem to me that in

circumstances with patients who are taking a drug like
   21

indinavir which may case some hepatic toxicity that
   22

that certainly could place them at much greater risk
   23

of Paxene toxicity if the Paxene dose is not modified.
    24

     I
And 25 wonder if you thought about that and considered
    you
how 26 might deal with that issue in package labeling?

   27       DR. HARRIMAN:    Certainly the protocol
                                                        86

specified that if patients had significant elevations
    1

in their liver function tests, bilirubin above 1.5, a
    2

five 3fold or higher increase in atransaminases, that

that 4would be a criteria for dose modification of

paclitaxel.
    5          We can pull up the data in terms of dose

modifications.
    6             But what I can say in that regard is

that 7is certainly a potential concern that one has to

be aware of.
    8

    9          But again, I would just draw your

attention, at least in the broad sense, to the slide
   10

I showed comparing the safety -- the adverse events of
    11

patients on protease inhibitors and not on protease
   12

inhibitors.
   13          At least in that broad sense we are not

seeing any significant differences in terms of the
   14

toxicities.
   15          However, I agree with you, one would

probably need to look very carefully at the subset of
   16

patients where there are abnormal liver function tests
   17

in order to really be able to look at that.
   18

   19          CHAIRMAN DUTCHER:   One last question?

   20          DR. MARGOLIN:   I have actually two brief

questions that are not exactly related to each other.
   21

     is
One 22 sort of generic, not pertaining only to your

product, but in AIDS patients that are going to be --
   23

excuse me, HIV positive patients who are going to be
   24

receiving this drug at two to three week intervals of
   25

what looks like prolonged periods of time.
   26

   27          The question is whether the risk of
                                                      87

hypersensitivity reactions goes down sufficiently to
    1

consider tapering or perhaps even discontinuing the
    2

decadron.
    3        Because if you add up the amount of decadron

that 4is used in these patients who are already at

serious risk of OIs, it really gets to be quite a lot.
    5

And I wonder if you have data on the HSRs?
    6                                           I'll ask

my second question after you answer that.
    7

    8         DR. HARRIMAN:   Okay.   Yes, the within the

protocol there was expressed in the protocol certain
    9

doses of decadron that were to be used and its specified
   10

intervals.
   11         However, there was actually some

variability in terms of both the dose of decadron that
   12

     used and the schedule for when it was given among
was 13

    different investigators.
the 14                          And I think, I don't know

whether Dr. Gill would like to talk at all about this,
   15

because I believe he has some information in that
   16

regard.
   17

   18         But I do think that there is at least some

anecdotal evidence to suggest that as the therapy
   19

continues in patients who have not had any evidence of
   20

hypersensitivity reactions, one may be able to get by
   21

with lower doses.
   22               But to my knowledge, that's not been

formally studied.
   23

   24         DR. MARGOLIN:   The related question

actually, I think Dr. Gill will end up having to answer.
   25

     just curious whether there was an overlap in the
I'm 26

time frame of the accrual to this study and the other
   27
                                                          88

one at USC.
    1          I think I recall hearing one of the

patients mention the same protocol nurse that was at
    2

the last meeting, if I'm not mistaken.
    3                                        The reason I

ask that is because there is a question whether any
    4

selection factors or bias could have been introduced
    5

into 6which patients were put in which study at the same

institution.
    7

    8          DR. GILL:   Patient accrual in the first

trial ended in December and this trial began accrual
    9

in January.
    10         So there is no overlap.   And since I'm up

I can just say that the dose on decadron has been reduced
    11

in some patients down to four milligrams, but it's never
    12

been done in an organized way to give you a sense.
   13                                                    It

seems that you can go down to four.
   14                                    Can you go down

to zero is an important question and hasn't been
   15

addressed.
   16

   17          CHAIRMAN DUTCHER:    We have a couple of

more questions.
   18              Dr. Aboulafia?

   19          DR. ABOULAFIA:   Thank you.   Just as a

quick comment.
   20             There is a point in time where patients

achieve stabilization of their disease and they remain
   21

at that state.
   22             In terms of indications and how often

     give this drug, you are going to have to build in
you 23

     knowledge of what their HIV viral load is, and how
the 24

that reflects on their case load.
   25

   26          And what I mean by that is not everyone

needs to be maintained at two week dosing for the rest
   27
                                                     89

of their lives.
    1              And many of these patients who have

achieved an initial response and have a concomitant
    2

reduction viral load to nondetectable levels may well
    3

be able to go off chemotherapy or really go down to much
     4

less 5frequent dosings.

    6       And that's what I was trying to get at when

I was asking about the viral loads or if you have data
    7

on how many different antiviral combinations patients
    8

were 9-- had with them when they came into the studies.

Or alternatively, in the study how many times their
   10

antivirals were changed.
   11                        Those are the key things.

   12       It doesn't help me a lot to hear the data

of how many patients responded on protease inhibitors
   13

versus those that didn't if I don't have viral loads
   14

    CD4
and 15 counts to know really what their clinical state

was. Many of these patients are put on viral loads,
   16

it sounds like a fairly heavily pretreated group with
   17

a CF4 count of 30.
   18                And what that means is that some

of them are not going to respond to the protease
   19

inhibitors either and the fact that you are looking at
   20

groups that you had put those on doesn't mean, at least
   21

to me per se, that their viral loads are nondetectable.
    22

   23       DR. HARRIMAN:    I'm sorry -- I didn't hear

your last point.
   24

   25       DR. ABOULAFIA:    The fact that they are on
protease inhibitors doesn't allow me to infer that
   26

their viral loads are well controlled or nondetectable.
   27
                                                         90

    1        DR. HARRIMAN:   Right.   Yes, about all I

can say in response to your question is, as we all know,
     2

the changes that occurred in the management of HIV
    3

disease over the last two years has been very dramatic
    4

and the way the current standard of care and the current
     5

way in which we approach patients with HIV is very
    6

different than it was even a year ago when this
    7

protocol, or a year and a half ago when this protocol
    8

was begun.
    9

   10        I think clearly, you know, knowing a

patient's viral load is going to be very important in
   11

managing the patient and also assessing the relative
   12

need for other therapies for treating their concomitant
   13

illnesses such as Kaposi's sarcoma.
   14                                  I don't think we

     further address those questions at this time.
can 15

   16        CHAIRMAN DUTCHER:   Dr. Walkes?

   17        DR. WALKES:   You had mentioned that you

allowed for one dose reduction and you had also said
   18

that the curve was not linear over 100.
   19                                      Is it linear

below 100?
   20        And why do you have to reduce the dose?     Is

it because of things like hepatotoxicity?
   21                                          And one

other thing, if you do reduce the dose, is it still
   22

effective?
   23

   24        DR. HARRIMAN:   Those are good questions.

     reason for dose reduction was for protocol
The 25
specified toxicity.
   26                  Primarily it was for toxicities

associated with the paclitaxel, severe neutropenia,
   27
                                                       91

febrile neutropenia, grade the or higher
    1

hepatotoxicity or peripheral neuropathy, those types
    2

of things.
    3

    4        We had several patients who did have dose

reductions -- yes, we had nine patients who had dose
    5

reductions to 75 milligrams per meter squared because
    6

of toxicities.
    7             Some of those patients were able to go

subsequently back up to higher doses as their Kaposi's
    8

sarcoma improved.
    9                In other cases, they stayed at 75

milligrams per meter squared.
   10

   11        Because of the small number of patients,

we can't draw any definitive conclusions about the
   12

effectiveness of 75 milligrams except that we did have
   13

in at least a couple of cases, patients who were on 75
    14

milligrams per meter squared and were able to maintain
   15

their response.
   16

   17        Okay, actually, Eric Fletcher, the first

gentleman that got up to speak, had a dose reduction
   18

to 75 milligrams per meter squared.
   19                                   And he had a

response while he was on the 75 that he subsequently
   20

more recently had gone back up to 100 milligrams per
   21

meter squared.
   22

   23        CHAIRMAN DUTCHER:   Thank you.   I think we

     going to have to end the questioning right now and
are 24

take a break for ten minutes.
   25                            We will be back here at
11:15.
   26

   27        (Whereupon, the foregoing matter went off
                                                       92

    1       the record at 11:05 a.m. and went back on

    2       the record at 11:19 a.m.)

    3       CHAIRMAN DUTCHER:    We are now going to

begin the FDA presentation.
    4                         People will take their

seats please.
    5           Dr. Kobayashi?

    6       DR. KOBAYASHI:    Could I have the lights

down, please?
    7           Thank you.

    8       Dr. Dutcher, members of the Advisory

Committee, Dr. Temple, my colleagues in the FDA, ladies
    9

    gentlemen, today I will be presenting the clinical
and 10

portion of NDA20-826, Paxene for advanced
   11

AIDS-Kaposi's related sarcoma.
   12                            Before proceeding

further, I would like to acknowledge the many important
   13

contributions made by the members of the review team
   14

shown on this slide.
   15

   16       The indication proposed in the NDA and

under discussion today is for use after failure of
   17

first-line or subsequent systemic chemotherapy for the
   18

treatment of advanced AID-related Kaposi's sarcoma.
   19

     proposed dose and schedule is 100 milligrams per
The 20

meter squared intravenously over three hours every 14
   21

days.
   22

   23       The primary end point of the Paxene study

in this application is objective tumor response.
   24

Evidence of clinical benefit is being sought from the
   25

data on the following five domains, response of
   26

disfiguring facial and foot lesions by visual
   27
                                                      93

assessment, response of tumor associated edema by
    1

visual assessment, response of pulmonary lesions and
    2

change in performance status.
    3                           This in addition to

cutaneous tumor response data is being presented today
    4

to obtain approval of the Paxene in this indication.
    5

    6       The original NDA was submitted to the FDA

in June of 1994.
    7              The applicant initially proposed a

100 patient randomized controlled clinical trial in
    8

patients with AIDS-KS in July of 1995 and submitted the
    9

protocol for the current study in September of 1995.
   10

     applicant met with FDA on several occasions
The 11

following initiation of the study to discuss issues of
   12

    point definition and analysis.
end 13                               The NDA itself was

submitted in March 31st of 1997.
   14

   15       In a special considerations meeting with

    FDA on September 15, 1997, the applicant requested
the 16

that FDA consider a change in the indication to target
   17

third-line systemic therapy in patients previously
   18

treated with Doxil.
   19

   20       The applicant's pivotal study was

conducted between September 1995 and March 1997 and
   21

enrolled 89 patients with advanced AIDS-Kaposi's
   22

sarcoma in nine centers located in California,
   23

Massachusetts, New York and Florida.
   24                                   Literature

reports on three other studies were also included in
   25

     application, as shown on this slide here.
the 26

   27       All studies are single arm, open label
                                                        94

Phase II studies.
    1                The dose and schedule chosen for the

pivotal study was based on Study No. 139-281, conducted
    2

at the USC Norris Cancer Center and at Massachusetts
    3

General Hospital, both participating centers in the
    4

current study.
    5             A Brown University study enrolled only

four 6patients and used a markedly lower dose of

paclitaxel and will not be considered further in this
    7

presentation.
    8             Both studies 139-174 and 139-281 have

previously been presented to this Committee.
    9

   10          The only study, it should be pointed out,

using the applicant's formulation is the pivotal study.
   11

     other studies used the currently approved
The 12

formulation.
   13            It should also be noted that the

formulation used in the applicant's clinical study is
   14

     the same as the formulation which is intended for
not 15

marketing.
   16

   17          The study objectives were first to

determine response rate and median time to tumor
   18

progression for patients with advanced refractory
   19

AIDS-related Kaposi's sarcoma treated with a three hour
   20

infusion of Paxene at a dose of 100 milligrams per meter
   21

squared every 14 days.
   22                      Secondly, to determine the

toxicity profile of this dose and schedule.
   23                                            And

thirdly, to evaluate clinical benefit in this patient
   24

population.
   25

   26          Quality of life in the pivotal study was

also assessed using the Symptom Distress Scale.
   27
                                                        95

However, the applicant was advised that the FDA regards
    1

interpretation and reliability of quality of life data
    2

collected in single arm, open label studies as
    3

problematic.
    4

    5          As pointed out by the sponsor, eligible

patients have had to have failed at least one prior
    6

systemic chemotherapy regimen.
    7                             And acceptable

indications for treatment included one or more of the
    8

following:
    9          multiple, more than 25 mucocutaneous

lesions; visceral involvement -- initially symptomatic
   10

visceral involvement was required, however this was
   11

later changed to allow the simple fact of visceral
   12

involvement to qualify for entry; and finally
   13

symptomatic lymphedema.
   14

   15          Initially, at least five measurable

cutaneous lesions were required.
   16                               This was later

changed to specify that these lesions must be raised.
   17

Response was graded using a modification of the ACTG
   18

criteria initially described by Crown, et al.
   19

   20          In this system, a complete response in

accordance with standard oncologic practice requires
   21

     complete disappearance of any detectable residual
the 22

disease and this must persist for at least four weeks.
   23

Please also note that biopsy documentation of the
   24

absence of disease is required when flat lesions
   25

persist.
   26

   27          Partial response requires the absence of
                                                     96

any new lesions or edema and also any one of the
    1

following occurrences:
    2                     either a greater than 50

percent decrease in lesions counts that persist for at
    3

least four weeks; a greater than or equal to 50 percent
    4

decrease in the total area of the five marker lesions
    5

or complete flattening of at least 50 percent of all
    6

previously raised lesions.
    7                        Note that according to the

protocol, only the decrease in lesion count required
    8

28 day confirmation.
    9

   10       Criteria for progressive disease require

only the demonstration of new or progressing visceral
   11

disease, new or increasing tumor associated edema, a
   12

greater than 25 percent increase in the total lesion
   13

count, a greater than or equal to 25 percent increase
   14

in the total area of the marker lesions or a change in
    15

     character of at least 25 percent of all previously
the 16

flat lesions to raised.
   17

   18       Please note that unlike progression

criteria in other solid tumors in which a single new
   19

lesion would indicate progression, in this system a 25
   20

percent increase in the total lesion count is required.
   21

   22       I would like to point out two difficulties

with the current definitions, while at the same time
   23

acknowledging that a joint effort of the AIDS
   24

Malignancy Consortium, the National Cancer Institute
   25

     the FDA is currently underway to revise these
and 26

criteria.
   27
                                                      97

    1       First, the criteria did not explicitly

resolve the situation in which a patient progresses on
    2

one of the three response subscales prior to responding
     3

on either the same or a different subscale.
    4                                           And

second, the criteria did not clearly specify the method
    5

of calculating progression based on lesion flattening.
     6

It is important to emphasize the criterion in current
    7

use were applied to this application.
    8

    9       The protocol specified that overall

response was to be limited to the first ten cycles.
   10

However, after inspecting the data, it became clear
   11

that late responses on one of the three subscales, tumor
   12

lesion count, tumor size and nodule flattening,
   13

occurred in at least eight patients and therefore, a
   14

response was credited regardless of the time in which
   15

it would occur.
   16

   17       Although the protocol did not explicitly

state this, confirmation at four weeks was required for
   18

     partial responses not only for responders on the
all 19

total lesion counts.
   20                  This is in accordance with

standard oncologic practice and the applicant has
   21

accepted this modification in communications following
   22

distribution of the draft medical officer review.
   23

   24       Based on information previously supplied

by other investigators in this field, patients who
   25

progress on any subscale were deemed progressors,
   26

regardless of subsequent responses.
   27
                                                      98

    1        A total of 89 patients were enrolled with

a median Karnofsky performance status of 80 percent.
    2

In general, the study enrolled patients with advanced
    3

disease in that at least 80 percent of patients were
    4

poor 5risk on at least one of the prognostic staging

subscales.
    6        In tabulating the indications for

treatment, it can be seen that 80 percent of patients
    7

required treatment for multiple cutaneous lesions.
    8                                                 A

total of a third of the patients required treatment for
    9

visceral lesions either symptomatic or asymptomatic
   10

     approximately half had symptomatic lymphedema.
and 11

Exactly half had symptomatic lymphedema.
   12

   13        The patient population also fit the

description of refractory disease.
   14                                 Although the

median number of received prior chemotherapy regimens
   15

     one, there was a maximum of five, and there is a
was 16

significant percentage of patients who have had at
   17

least two prior regimens.
   18                       Between a third and a half

of the patient population had received and failed prior
    19

therapy with either Doxil or DaunoXome or both and the
   20

majority of patients had stopped their last systemic
   21

chemotherapy regimen prior to entry on this study
   22

because of either inability to tolerate treatment or
   23

because of progressive disease.
   24

   25        At least a third of patients had progressed
through their immediately preceding chemotherapy
   26

regimen.
   27
                                                        99

    1        For the Committee's reference, this

analysis is labeled the Per Protocol Analysis in the
    2

draft medical officer review previously circulated.
    3

After extensive review and discussion of further
    4

additional data submitted by the applicant, the primary
    5

FDA analysis concludes that this study shows a 42
    6

percent response rate using the previously cited
    7

interpretations of the protocol.
    8                                All responses were

partial and no complete responses were noted.
    9

   10        The median duration of response is 13 days,

although this has not been confirmed by our
   11

statistician.
   12            The time to response was 34 days and the

median time to progression calculated using 37 events
   13

     51
and 14 censored observations was 163 days.

   15        In accordance with the applicant's request

at the special considerations meeting, the issue of
   16

response in patients with prior Doxil therapy was
   17

examined.
   18       Twenty seven patients in this study have

previously received Doxil.
   19                          Thirteen as first-line

treatment, and 14 as second-line or greater.
   20

   21        Amongst the 13 patients receiving Doxil as

first-line treatment, there were three partial
   22

responders for a 23 percent response rate using the
   23

response categories assigned during the primary
   24

     analysis.
FDA 25
   26        In the 14 patients receiving Doxil as

second-line or greater treatment in which Paxene
   27
                                                     100

therefore would have constituted third-line or greater
    1

treatment, there were six responders for a 43 percent
    2

response rate.
    3

    4         This slide shows the areas of discrepancy

between the applicant and FDA in accounting for the
    5

responses.
    6        Please note this compares the revised FDA

primary analysis with the applicant's revised analysis
    7

which they have presented.
    8                         Let me bring that up for

you. 9 The major problems can be seen to be -- to occur

in the patients with responders, claimed responders who
    10

progressed prior to the actual observation of a
   11

response.
   12        Please also note that there was one

responder credited who could not be documented to have
   13

a greater than 50 percent decline on any of the response
    14

subscales.
   15        And please also note that the FDA review

upgraded three patients from the stable disease
   16

category to the partial response category.
   17

   18         The overall response rate of all enrolled

patients observed in the two major studies from the
   19

literature are shown here.
   20                        Although it should be noted

that data for these studies were not submitted to the
   21

     and only the published literature reports were
FDA 22

included.
   23

   24         In Dr. Gill's study, which was again as

noted the pilot for this study, there was a 59 percent
   25

overall response rate in all enrolled patients.
   26                                              In the

40 patients who had been previously treated, there was
    27
                                                     101

a 52 1percent response rate.

    2         While the publicly available results from

these studies appear encouraging, the Agency regards
    3

them 4as sufficiently different from the pivotal study

in both design and execution, that any formal, direct
    5

comparison would be inappropriate.
    6

    7         One issue that arose in discussions with

the applicant following the initial circulation of the
    8

draft medical officer review is the possibility of
    9

multiple valid interpretations of the progression
   10

criteria.
   11       Shown here is the clause in question.   With

     Committee's permission, I would like to take a few
the 12

minutes to present a short example that illustrates the
   13

difficulty.
   14

   15         This is an example selected from the

submitted database.
   16                  The ellipses here indicate data

that were excluded to ease a presentation which neither
   17

     nor detract from the point of this presentation.
add 18

Shown here are the cycle number, the day of therapy,
   19

    observed number of flat lesions at each time point,
the 20

     calculated number of flat lesions at each time
the 21

point, the observed number of raised lesions at each
   22

time point, and the change in the number of raised
   23

lesions from the previous cycle.
   24                               The line shown here

in magenta represents the nadir of the raised lesion
   25

count.
   26

   27         Based on extensive correspondence with the
                                                    102

applicant and the extensive and internal discussions,
    1

there appear to be at least five separate methods of
    2

determining progression.
    3                      This becomes important

because a patient's overall response integrates the
    4

outcomes on the three separate response subscales.
    5

Thus, for instance, an initial early progression based
    6

on lesion flattening would result in a patient being
    7

considered a progressor despite the occupance of a
    8

later response on the basis of tumor size or total
    9

lesion count.
   10

   11       Now Method 1A would use as the method of

determining the baseline for progression, the observed
   12

number of flat lesions at the nadir of raised lesion.
   13

In this patient, it would select a reference value of
   14

30 and 25 percent of that number, or seven new lesions
    15

would be required for the patient to progress.
   16

   17       Method 1B, which is the method used by the

applicant, models the number of flat lesions at the
   18

nadir of the raised lesion count.
   19                               In this patient, it

would select a reference value of 41 and then ten
   20

patients or ten lesions, excuse me, would be required
   21

     this patient to progress.
for 22

   23       Method No. 2 would use the observed number

of flat lesions in the cycle immediately prior to the
   24

nadir of the raised lesions as the baseline.
   25                                          In this
patient, it would select a reference value of 17 and
   26

four lesions would be required for progression.
   27
                                                    103

    1        Method No. 3 uses the number of raised

lesions that flatten by the nadir of the raised lesion
    2

count as the baseline against which progression is
    3

judged.
    4     In this patient, it would return a value of

33 as the reference, and therefore, eight new lesions
    5

would be required for progression.
    6

    7        Method No. 4, which was the method adopted

in the original FDA draft medical officer review,
    8

chooses the nadir of the raised lesion count as the
    9

reference value -- as the baseline.
   10                                  In this patient,

it would select a reference value of five and therefore
    11

only one lesion would be required for the patient to
   12

progress.
   13

   14        Each of these methods, although there are

multiple, has been applied by either the applicant or
   15

    of
one 16 several FDA reviewers in an informal survey taken

within our division.
   17                  Again, emphasizing that Method

1B was applied to this application, Method No. 4, it
   18

should be pointed out, most closely corresponds to the
   19

response criteria being developed currently in the
   20

NCI/FDA/ANC collaboration.
   21

   22        To repeat the earlier slide showing the

actual data, Method 1A would select a reference value
   23

of 30, Method 1B would select a reference value of 41,
    24

Method 2 would select a reference value of 17, Method
   25

3 choosing the number of flat -- lesions that flattened
    26

would choose this number here, 33.
   27                                 Method 4 which is
                                                     104

the number used in the original FDA review would choose
     1

a number of 5.
     2           And again, Method 1B is the applicant's

-- method applied by the applicant.
    3

    4       If one classifies this patient according

to each of these different methods, one comes up with
    5

these outcomes.
    6              And the point here is the extreme

variability in outcome resulting from these different
    7

methodologies.
    8             There is a ten fold variation in the

number of lesions that would be required for
    9

progression, a nearly four fold variation in the day
   10

     the day on which progression occurs, and
and 11

diametrically opposite response categorizations,
   12

depending for this response scale anyway, depending on
   13

    method chosen.
the 14               In fact, according to Method 1B the

patient would never have responded prior to the end of
   15

treatment and would have remained as a partial
   16

respondent throughout his entire course of treatment.
   17

   18       Presented here are the response rates from

    original review, shown for comparison and labelled
the 19

draft FDA analysis which again used Method 4 and showed
   20

a 35 percent response rate.
   21                          And the revised FDA

analysis which followed Method 1B, which more closely
   22

approximates the current practice.
   23                                The estimate shown

here, 42 percent, is our best estimate of the response
   24

rate from this study.
   25

   26       Also shown here for comparison are two

secondary analyses, and I apologize at this point for
   27
                                                         105

a typographical error in the handouts.
    1                                        The first

analysis, which is labeled as the relaxed FDA analysis,
    2

was carried out to account for the subjectivity which
    3

is inherent in the measurement and counting of Kaposi's
     4

sarcoma lesions and to all account for the clinical
    5

observation that initially confluent lesions may
    6

occasionally breakup and make the tracking of any
    7

individual lesion difficult.
    8

    9          If you adjust for these factors of observer

variability, one comes up with a response rate of 45
   10

percent.
   11      An analysis excluding five patients who were

ineligible for the study on the grounds of significant
   12

medical reasons, yields a response rate of 42 percent,
   13

that is 34 responders out of 81 patients.
   14                                           Both

calculations of this response rate are essentially
   15

identical to the 42 percent obtained in the revised
   16

primary analysis using Method 1B.
   17

   18          Moving on to the elements of clinical

benefit.
   19      Twenty five percent of 24 patients with

disfiguring facial lesions who had assessable
   20

photographs submitted showed improvement in their
   21

disfiguring facial lesions.
   22                           While nine percent of 11

patients with foot lesions who had assessable
   23

photographs submitted showed evidence of improvement,
   24

     12
and 25 percent of 48 patients who had lymphedema who
     assessable photographic evidence submitted had
had 26

improvement.
   27
                                                      106

    1       The submitted quality of life data is

weakened by the fact that it was collected in a single
    2

arm, 3open label study and therefore lacks comparator

to assess the extent or the impact of the extent and
    4

nature of the missing data.
    5                         For similar reasons, which

are outlined in more detail in the medical and
    6

statistical reviews, interpretations of analyses
    7

aggregating more than one subscale are also considered
    8

to be difficult.
    9

   10       Nevertheless, they made provide

additional helpful information in interpreting the
   11

response and clinical benefit data presented.
   12                                             This

slide depicts the result of a longitudinal data
   13

analysis performed by Dr. Koutsoukos, the statistical
   14

reviewer, on the mobility data using response
   15

assessments from the draft FDA analysis.
   16                                      He performed

a similar analysis to this using the response
   17

assessments from the applicant's initially submitted
   18

analysis and obtained essentially the same results.
   19

Therefore, only this will be shown.
   20

   21       On this scale, a decrease in score

represents an increase in an improvement in mobility
   22

    time in cycles is indicated here on the X axis.
and 23                                                As

     can see, there is no statistically significant
you 24

difference in the rate of improvement between
   25

non-responders which are indicated in the lower line
   26

     responders which are indicated by the top line,
and 27
                                                     107

although there is an improvement from baseline.
    1                                             Thus,

although an unadjusted analysis which pools all
    2

patients together, irregardless of response status
    3

does 4show a statistically significant overall

improvement in mobility over time, this improvement
    5

cannot be ascribed to differences between responders
    6

and non-responders.
    7

    8        For the sake of completeness,

statistically significant improvements over time were
    9

noted in the unadjusted analyses of Appearance No. 1
   10

which measures the worsening of appearance, mobility
   11

as shown here, breathing and Karnofsky performance
   12

status.
   13     However, analyses such as the one indicated

on this slide do not show any difference between
   14

responders and non-responders on any of the subscales.
   15

   16        This slide shows the response of patients

with pulmonary involvement that had evaluable data.
   17

Although the bottom line of 60 percent does appear
   18

impressive, it should be noted that it is drawn in five
   19

patients which represents a very small subset, 18
   20

percent to be precise, of the 28 patients with visceral
   21

disease who were enrolled in this study.
   22

   23        I should also note that the responses in

visceral disease all occurred in patients with
   24

pulmonary lesions.
   25                 This slide again depicts the work
of Dr. Koutsoukos on a performance status of data.
    26                                               It

should again be noted that there is no difference
   27
                                                    108

between responders indicated here, and non-responders
    1

indicated here.
    2

    3       Although there is again a significant

improvement over time from baseline, on this scale
    4

again an improvement, unlike the previous scale, is
    5

indicated by increase in the Y axis and again, time and
    6

cycles is indicated on the X axis.
    7                                And again, the same

comments made previously in reference to the mobility
    8

data 9also apply to this data.

   10       Turning to the safety analysis, the

applicant reported a total of 22 deaths, of which 11
   11

occurred at greater than or equal to 30 days beyond the
   12

last dose of study drug, and seven which were possibly
   13

related to Paxene.
   14

   15       These seven deaths were distributed in the

following manner:
   16                five of them -- five of the 22 were

attributed to cytopenia complicated by infection; one
   17

occurred as a result of a septic shock complicated by
   18

respiratory arrest; one occurred in a patient who had
   19

pulmonary hypertension with congestive heart failure
   20

     the total of seven deaths.
for 21

   22       This slide considers the occurrence of

opportunistic infections according to whether the
   23

event represented a new event, the continuation of an
   24

already established infection in that patient which
   25

became established prior to entry to study, or
   26

recurrence of a previous infection.
   27                                   There was one
                                                     109

patient in which such classification could not be made.
    1

    2         Although definitive conclusions cannot be

drawn from this study due to the lack of a randomized
    3

concurrent control arm, as a general statement, the
    4

instance of opportunistic infections does not appear
    5

unexpectedly high for this patient population and the
    6

profile does not show an unusual distribution of
    7

infectious organisms.
    8

    9         As expected, myelosuppression was

substantial with more than 80 percent of patients
   10

having either neutropenia, leukopenia or anemia.
   11

Approximately a third of patients had thrombocytopenia
   12

     there were 11 patients or 12 percent in whom their
and 13

neutropenia was complicated by febrile neutropenia
   14

which was defined as fever occurring during a period
   15

in which the neutrophil count was less than 1,000
   16

whether or not infection of a specific organism was
   17

documented.
   18

   19         The use of hematopoietic support was

liberal with 41 percent of patients requiring the use
   20

of supplemental PCSF and 25 -- a quarter of patients
   21

requiring either erythropoietin or red cell
   22

transfusions.
   23

   24         This study included a substantial number

of patients taking concomitant protease inhibitors,
   25

although again lack of a concurrently randomized
   26

control arm prohibits the drawing of definitive
   27
                                                     110

conclusions regarding the presence or absence of
    1

drug/drug interactions.
    2

    3         Known toxicities of protease inhibitors

include hyperbilirubinemia, diarrhea and renal calculi
    4

and there were six patients shown here -- or nine
    5

patients in whom an isolated elevated
    6

hyperbilirubinemia was observed as their only instance
    7

of hepatic toxicity.
    8                  In each of these patients the

time 9course was consistent with the hypothesis that

they represented the effect of protease inhibitors as
   10

opposed to Paxene toxicity.
   11

   12         Twenty nine patients, or 32 percent, had

arthralgia, myalgia, or severe arthritis which could
   13

    be
not 14 easily ascribed to a specific etiology apart from

     study drug and therefore the Agency adopted a
the 15

conservative position and ascribed the toxicity to the
   16

study drug.
   17         There were ten patients in whom

nephrotoxicity occurred.
   18                       Approximately a third of the

patients had neurotoxicity, 88 percent of patients had
   19

hepatotoxicity and there were three patients in which
   20

either frank malignancy or an unexplained generalized
   21

lymphadenopathy occurred.
   22

   23         In summary, the submitted Phase II study

of Paxene in patients with previously treated Kaposi's
    24

sarcoma should be considered an adequate and well
   25

controlled study of objective tumor response.
   26                                             The

objective response to Paxene in this patient population
   27
                                                      111

may be a clear demonstration of anti-tumor activity
    1

     the comparator in this case being the known natural
with 2

history that the tumors do not shrink without
    3

treatment.
    4        And the overall objective tumor response

rate 5was well documented at 42 percent of patients.

    6        However, proof of clinical benefit is less

clear with improvement in only 25 percent of patients
    7

with 8disfiguring facial lesions, nine percent of

patients with foot lesions, 12 percent of patients who
    9

     lymphedema and 60 percent of a very small subset
had 10

of patients with lung involvement.
   11

   12        The study was not adequate nor well

controlled to evaluate the secondary end points of time
   13

to progression, duration to response, or survival.
   14

Thank you.
   15

   16        CHAIRMAN DUTCHER:   Questions from the

Committee for the FDA?
   17                     Dr. Simon?

   18        DR. SIMON:   Could you say anything about

duration of response?
   19

   20        DR. KOBAYASHI:   Yeah, the overall

duration of response is 213 days.
   21                               Although the reason

it is not on the slide is we have not had time to have
   22

that confirmed by the statistician.
   23

   24        DR. SIMON:   That's the median --

   25        DR. KOBAYASHI:   That's the median
duration based on a Kaplan Meier analysis.
   26

   27        DR. SCHILSKY:    Ken, I had just two
                                                      112

questions.
    1        You mentioned right at the beginning that

the formulation which is proposed for marketing is
    2

different from the formulation which was actually
    3

studied as under the Phase II study.
    4                                  Could you comment

on that any further with respect to the FDA's level of
    5

comfort that the proposed formulation is actually
    6

equivalent to the formulation for which we have seen
    7

data.
    8

    9        DR. KOBAYASHI:   I think that involves some

proprietary considerations.
   10                           I think perhaps the

company would be, or our chemist would be, perhaps,
   11

better suited to answer that question.
   12                                     Or perhaps one

of my superiors.
   13

   14        DR. SCHILSKY:    I just think it's going to

be a little bit difficult for us to make a judgement
   15

about these data --
   16

   17        DR. KOBAYASHI:    I understand.

   18        DR. SCHILSKY:    -- if what we have been

spending the morning listening to is not even the drug
   19

that's being proposed for marketing.
   20

   21        DR. KOBAYASHI:    I understand.

   22        DR. DELAP:   Well, I think we are basically

satisfied that the data that you have seen a
   23

representative of the data that would be generated if
   24

    precise formulation to the market had been studied.
the 25
     I
And 26 don't know if the company wants to contribute

anything about any differences that there might have
   27
                                                    113

been, but there are bridging data that enable us to feel
    1

pretty secure that what we are looking at is the
    2

reality.
    3

    4       DR. SCHILSKY:    So if you are secure, then

I'm satisfied.
    5            And I guess my other question comes

      to
back 6 this issue of hepatotoxicity from proteases and

I was just wondering if in your review of the data
    7

whether you were able to sort of get into enough detail
    8

to figure out if a patient had, say, an elevated
    9

bilirubin attributable to protease inhibitors and was
   10

receiving Paxene at that point in time, did it appear
   11

that the patient had any greater toxicity from that
   12

cycle of Paxene during which the bilirubin was
   13

elevated?
   14

   15       DR. KOBAYASHI:    No, I did not.   We

conducted analysis comparing the toxicity according to
   16

whether or not the patients had received protease
   17

inhibitors or not and we broke that down, the hepatic
   18

component of that and tried to tease out whether this
   19

     isolated hyperbilirubinia due to protease
was 20

inhibitors or whether or not, or we didn't look at
   21

specifically the subset of patients who had elevated
   22

liver functions going into the study and whether or not
   23

they had any different toxicity experience.
   24                                           That's

certainly something that we will be looking at after
   25

this.
   26

   27       DR. NORTHFELT:    I have another question
                                                     114

related to protease inhibitor antiretroviral therapy.
    1

You mentioned in your closing statement that a good
    2

control for these data would be the experience that KS
    3

does 4not regress unless it's treated, I presume you

meant with chemotherapy.
    5

    6         Now, at the coffee break Dr. Aboulafia and

I were telling each other our fish stories about
    7

regression of KS under the influence of potent
    8

antiretroviral therapy with no chemotherapy.
    9                                            We both

    patients with pulmonary KS or lymphadenopathies KS
had 10

which is resolved substantially or completely, in a
   11

clinical sense, with no chemotherapy.
   12

   13         So, could you just reflect on that a little

    for
bit 14 us?    Because we've heard how we don't have very

good control on antiretroviral use here.
   15

   16         DR. KOBAYASHI:   I understand your point.

     point being that there is a second potential
The 17

medication being administered to these patients which
   18

could account for these responses.
   19                                  And how can we

reliably attribute the observed responses to Paxene as
   20

opposed to say the administered protease inhibitors or
   21

whatever?
   22

   23         I think that's an excellent question and

a very important issue.
   24                      And it highlights the

difficulty with interpreting, a couple difficulties
   25

actually.
   26        The first one is the simple off-the-cuff

highlights difficulty with interpreting data from
   27
                                                     115

single arm, non-randomized Phase II study in which
    1

there is not a concurrent control arm.
    2                                      It also

highlights the difficulty that the pace of medical
    3

progress is rapidly changing and we are talking about
    4

great improvements in our other -- in treatment for
    5

AIDS.
    6

    7       And so, how to factor that into designing

a study or looking forward to anticipating the next step
     8

in response to your question, how one would design that
     9

study given the realities of patient care in 1997 is
   10

a little bit more problematic and one in which I do not
    11

have a ready answer.
   12

   13       DR. MARCO:     Well, first I want to make a

comment that we just, we can't be saying protease
   14

inhibitor and thinking that they are all alike or that
   15

     regimens are alike.
all 16                     I mean, Donald, I'm sure you

treat your patients very well and you know exactly what
   17

to give them and what combinations.
    18                                But, listening to

some of these patients speak and the therapies that they
   19

were given for their KS, I mean it just shows how
   20

patients are not always treated properly.
   21                                         It's sort

of embarrassing.
   22

   23       What my question for you is, I'm having

trouble with the numbers as far as patients that were
   24

evaluable, i.e., if they had more than two cycles of
   25

therapy versus the patients that you talked about
   26

having protocol violations.
   27                          You originally say that
                                                      116

14 out of 89, 16 percent were protocol violations.
    1

Nine 2out of 89, ten percent lacked positive histologic

confirmation.
    3            How does that figure into your

percentages in your final lab?
    4

    5         DR. KOBAYASHI:   Right, after that draft

     out, we had communication with the sponsor and they
went 6

were 7able to supply us with biopsy reports for one

thing, and were able to satisfy us in several of the
    8

patients which appeared to be ineligible on the basis
    9

of the data which was initially submitted did in fact
   10

have sufficient data to support eligibility for the
   11

trials.
   12      That is part of the discrepancy in the

numbers.
   13

   14         DR. MARCO:   Okay, so -- I'll let you

finish.
   15      I'm sorry.

   16         DR. KOBAYASHI:   And the other response is

that in discussing that slide I did think the disclaimer
   17

that five patients were exclusions on the basis of
   18

significant medical reasons.
   19                            Considering the nature

of this study, the physicians involved, and absolute
   20

lack of a biopsy report was not considered to be a
   21

significant medical reason, especially after we were
   22

able to get the documentation.
   23

   24         So these five patients that are excluded

here are patients who, for instance, one had a clearly
   25

elevated creatinine that should not have been elevated,
   26

or should not have entered on the study under the
   27
                                                     117

protocol criteria, that sort of thing.
    1

    2        DR. MARCO:     Okay, so five you basically

threw out.
    3

    4        DR. KOBAYASHI:    Right.

    5        CHAIRMAN DUTCHER:    Dr. Temple?

    6        DR. TEMPLE:    Ken, as part of the response

to Dr. Northfelt's question about the adequacy of the
    7

historical control that you have patients both on and
    8

not on protease inhibitors and look at response rates
    9

in both of them.
   10              Right?

   11        DR. KOBAYASHI:    I'm sorry, I was --

   12        DR. TEMPLE:    Isn't part of the answer to

     Northfelt's question about the adequacy of
Dr. 13

historical control always a very good question to ask
   14

in changing circumstances --
   15

   16        DR. KOBAYASHI:    Yes.

   17        DR. TEMPLE:     -- that you have patients

both on and off protease inhibitors.
   18                                    You are getting

smaller sample sizes, of course, by that time, but the
   19

response rates are not very different in those two
   20

groups?
   21

   22        DR. NORTHFELT:    Yes, my response to that

would be that there is protease inhibitors therapy and
   23

then there is protease inhibitor therapy.
   24                                       I mean there

     people who have viral loads of a half a million on
are 25
protease inhibitors and there are people who have viral
   26

loads of ten on protease inhibitors.
   27                                    And both of my
                                                          118

colleagues here have pointed out that without a real
    1

understanding of how well the protease inhibitor
    2

therapy is working, you can't know how much it confounds
    3

the observations of the chemotherapy.
    4

    5         DR. SIMON:    Yeah, but some of the patients

are not getting any protease inhibitors and they are
    6

responding, so --
    7

    8         DR. TEMPLE:    Yes, so it can't be all that

they 9are on protease inhibitors.

   10         DR. NORTHFELT:    Agreed.    But they may

have good immune system response to their HIV that keeps
   11

their viral load as low as any protease inhibitor
   12

treated patient in the next chair.
   13                                     So, we just don't

know enough about these patients I think.
   14

   15         DR. TEMPLE:    Concurrent controls are

better.
   16

   17         DR. BRODER:    May I respond.

   18         CHAIRMAN DUTCHER:    Sure.

   19         DR. BRODER:    I thank the Chair's

indulgence.
   20          We performed an examination of the

duration and the speed with which a response occurred.
   21

     there is a definite fund loading of response in
And 22

patients.
   23       It does not occur by chance or randomly

throughout the observation period.
   24                                     There is a slide

which was shown that could be presented again.
   25

   26         So there is a highly statistically

significant front loading of the responses juxtaposed
   27
                                                      119

to the administration of the Paxene.
    1                                    This makes it

exceedingly improbable with P values that our
    2

statistician could give you, that this is just
    3

occurring on a spontaneous basis across the observation
    4

period.
    5     But there was a front loading of the response

rates and I unfortunately can't show the slide, but we'd
    6

be happy to provide it to the Committee.
    7

    8        DR. ABOULAFIA:   Could you go back one

slide on your presentation?
    9                         And could you just comment

on this again, I'm not sure I understood your point
   10

here.
   11   It looks like you've taken into account disease,

visceral involvement and a fair number or moderate
   12

number have responded.
   13                     And I'm not sure I understood,

     Kobayashi, were you saying that a small number or
Dr. 14

a moderate number -- how did you put this data together
    15

in terms of a response?
   16

   17        DR. KOBAYASHI:   This slide is just -- was

intended only to bring back from the previous slide,
   18

repeat information from a couple of previous slides all
   19

in one place so that the improvement on the four domains
    20

of clinical benefit for which we could have reliable
   21

information and could be put in one place.
   22

   23        There is no real point to this slide, it's

simply -- or to this table -- it's simply there as a
   24

summary to aid and to deliberations we might want.
   25

   26        DR. MARCO:   In relation to that, the -- I'm

having trouble with the clinical benefit in completely
   27
                                                       120

understanding that the statistically significant
    1

betterment in appearance mobility and breathing, you
    2

agree with that, correct?
    3

    4       DR. KOBAYASHI:     Yes.

    5       DR. MARCO:    But, that's under sort of the

general well being versus these which are more specific
    6

to the lesions.
    7              How does this differ from what the

applicant has shown us?
    8

    9       DR. KOBAYASHI:     These were previously

defined as the domains on which we would be assessing
   10

    response to the patient.
the 11                         One of the problems with

looking at, with pooling different subscales from the
   12

quality of life data and perhaps our statistical
   13

reviewer could comment a little on this, is that there
   14

     a
are 15 little bit -- there are a substantial number of

correlations between them.
   16                          It's a little bit more

difficult to interpret.
   17

   18       So we felt that, in terms of the quality

of life data, that looking at a single response subscale
    19

would be better.
   20              As I say, in previous applications

with AIDS-KS, we've sort of considered these domains
   21

to be the ones of the areas of clinical benefit.
   22                                               And

I think Dr. Johnson had a comment he wanted to make.
   23

   24       DR. MARCO:    Well, no, I just, but

basically the sponsor showed us these beautiful graphs
   25

with these great P values and I mean either I'm getting
   26

it wrong or --
   27
                                                        121

    1        DR. JOHNSON:    I think the, you are talking

about two different things here.
    2                               This is lesions that

can be verified.
     3              In other words, we based our analysis

here 4based on photographs of these lesions.     I think

it's 5fairly objective.

    6        The quality of life data that the sponsor

presented and that you are thinking about is the
    7

patients' analysis of whether the patient has improved.
    8

And we have some difficulties with that for
    9

methodological reasons that have previously been
   10

described.
   11        But the slides that the sponsor showed were

based on the quality of life scales.
   12                                    These are based

on physical examination by the --
   13

   14        DR. MARCO:     But you are considering

physical examination, thus counting of lesions,
   15

clinical benefit.
   16

   17        DR. TEMPLE:     No, these are individual

patients who were thought to have had a persuasive
   18

improvement by photographs, sort of one by one.
   19                                                 It's

just different from an analysis scales or quality of
   20

life questionnaire.
   21

   22        It's not that they are inconsistent.     They

     just different ways of getting at the same kind of
are 23

thing.
   24    And in this setting, there is a certain feeling

that a response like that speaks for itself.
   25                                             If you
have lesions all over your face and then they are gone,
   26

it's sort of obvious that was a benefit.
   27                                         And that's
                                                       122

why these are -- there aren't that many of them, which
    1

is the point Ken made, but the ones that there are seem
    2

real.
    3

    4         DR. LI:   I would add that this is perhaps

the most conservative assessment because these are the
    5

ones 6where you can, as a dispassionate observer, Dr.

Kobayashi was able to look at these pictures and say
    7

     you know, in this patient clearly the facial lesion
yes, 8

got better.
    9          It's not to say that the lesions didn't

     better in some of the patients, just that he could
get 10

look at photographs and say I wasn't there, I didn't
   11

    the patient, but just looking at these photographs
see 12

I can verify that in this particular one.
   13                                            So I think

these are kind of the most conservative view of data,
   14

     they are not inconsistent with the other views of
but 15

     data.
the 16

   17         DR. MARCO:    I'm just having -- I

understand that.
   18               I just having trouble with the

semantics of it.
   19

   20         DR. KOBAYASHI:    Oh, I'm sorry.    I

misunderstood the question.
   21

   22         CHAIRMAN DUTCHER:    Any other questions,

comments?
   23        Thank you.    Okay, it's time to open the

discussion.
   24          Are there any other comments?      Should we

go directly to the questions?
   25

   26         DR. NORTHFELT:   Dr. Dutcher, could I just

make a couple of comments?
   27                          Thanks.   I just wanted to
                                                      123

make 1a couple of general comments about the nature of

the research that goes on in bringing these drugs to
    2

review like this.
    3                I want to kvetch a little bit more

about the response criteria, but I'll be very brief,
    4

I promise.
    5         And then I want to say something about

natural history.
    6               I hope this will be of some value to

the other members of the Committee.
     7                                I think that's why

David and I are here today.
    8

    9         First of all, the response criteria, I

think Dr. Kobayashi did a very excellent job of showing
   10

    difficult it is to interpret the response data that
how 11

     generated using these sets of response criteria.
are 12

And, that is not the fault of the sponsor of the study.
   13

These response criteria have been foisted on them and
   14

on the KS-afflicted community by us in the clinical
   15

science community that can't do a better job of defining
   16

what constitutes a response to therapy in this disease.
   17

     there has been a struggle going on for ten years
And 18

to try to create response criteria that actually
   19

expressed something meaningful about the way KS
   20

responds to treatment.
   21

   22         I think the clinical relevance of the

response criterias that are in common use, the clinical
   23

relevance of those is very dubious.
   24                                   I don't think

there is any reliable or reproducible relationship to
   25

anything clinically relevant using these response
   26

criteria.
   27        In other words, you can make the thing flat
                                                      124

but not help a guy, or a thing can stay bumpy but he
    1

can still be helped by the treatment.
    2                                     And these

response criteria do not express that.
    3

    4        Again, it is not the sponsor's fault.

They 5were stuck with these things and they were

struggling to the best of their ability, I think, to
    6

show 7us that the drug does something.    You know, but

they 8are very handicapped by this monster that we have

created in clinical science which is the response
    9

criteria.
   10

   11        Fortunately, thank God, there is a way out

of this eventually and then this Committee won't be
   12

burdened with this problem anymore.
   13                                   There is this

effort that was mentioned with the NCI and the FDA and
   14

     AIDS Malignancies Consortium to create some
the 15

meaningful response criteria.
   16                            I know there are people

in this room who are developing other new drugs for KS
    17

that they hope to bring to this Committee's attention
   18

some day.
   19       So, please avail yourselves of the efforts

that are being made by this Committee.
   20

   21        Dr. Murgo who is sitting here from the FDA

     is
who 22 very familiar with this is participating in

creating these criteria.
   23                       And I think it's going to be

a major advance in our ability to really understand how
    24

KS therapy works.
    25              I just want to read very quickly from
     abstract that describes this effort.
the 26                                        This was

presented at the AIDS Malignancies conference this
   27
                                                    125

spring.
    1     Dr. Feigel was the lead author and she said:

    2        "Evaluation of clinical

    3        benefit is complex in KS.

    4        The new criteria will focus on

    5        tumor specific symptoms,

    6        including evaluation both

    7        from the physician and

    8        patient perspectives.

    9        Categories considered

   10        significant include pain,

   11        edema, particularly

   12        extremities, scrotal and

   13        facial edema, facial and oral

   14        lesions, visceral-related

   15        symptoms and necrosis or

   16        ulceration of lesions."

So there you go, there is a nice list of actual,
   17

meaningful, clinical benefits that might derive from
   18

effective therapy.
   19

   20        And as soon as those criteria are developed

fully and put into place, we won't have to go through
   21

this all and more importantly, Dr. Kobayashi, Dr. Murgo
   22

     his colleagues won't have to go through the
and 23

difficulty of trying to extract something meaningful
   24

from these data.
   25

   26        I also want to comment about something that

appeared in the sponsor's information that we were
   27
                                                    126

provided.
    1       They tabulated response data from a number

of studies of KS therapy going back over the years, and
    2

I think it should be brought to the Committee's
    3

attention that KS is different now than it was five
    4

years ago, or ten years ago.
    5                          KS has changed profoundly

over 6the course of the ten years that I've been caring

for patients with this illness.
    7                               We can use data on

therapy for colon cancer from the 1980s.
    8                                        We can use

data 9on chemotherapy for breast cancer from the 1980s

to treat patients that we see today.
   10

   11        I don't believe that's possible with

Kaposi's sarcoma.
   12               The natural history of the disease

is changing before our eyes.
   13                           The therapies for

underlying HIV-related immune deficiency, as we have
   14

heard, are changing before our eyes.
   15                                   And so it's very

difficult, I think, to look back more than a couple of
   16

years and really think that you are understanding
   17

what's going on with this disease.
   18

   19        I also wanted to point out, finally, that

we heard comments from, I think, seven patients today
   20

     are on the study, and that's about ten percent of
who 21

     evaluable patients in the study, I think.
the 22                                            So we

     just before our eyes here, at least a ten percent
had 23

response rate I think.
   24                    And with that, I think a good

case could be made that this is an approvable drug.
   25

   26        You know, there are a lot of caveats that

we have been talking about all morning here, but we just
    27
                                                         127

heard from ten angels who are perched on our shoulders
    1

here 2this morning, telling us that something very

healthful happened in their lives.
    3                                   And I think of

everything that we've heard this morning, perhaps the
    4

sponsors should have highlighted their contribution to
    5

these patients.
    6              And I particularly want to thank them.

I think they have brought something very meaningful to
    7

the eyes and ears of the panel.
    8

    9          CHAIRMAN DUTCHER:    Thank you for your

comments.
   10         All right.    Should we go on to questions?

     right.
All 11

   12          This is question number one.    "Is Paxene

study size of 89 patients adequate for approval of a
   13

drugs for use after failure of a first-line or
   14

subsequent systemic chemotherapy for the treatment of
   15

advanced AIDS-related Kaposi's sarcoma?"
   16

   17          All those who feel that this is an adequate

well controlled study and that the data presented are
   18

sufficient for evaluation, please raise your hand.
   19

High.
   20

   21          One, two, three, four, five, six, seven,

eight, nine, ten, eleven.
   22

   23          Question number two.   If you look at your

summaries that were in the blue folder, you have
   24

question number two has several tables in it that
   25

reiterate the data analysis.
   26

   27          DR. MARCO:    Dr. Dutcher, can I make one
                                                        128

quick comment about question number one?
    1

    2        CHAIRMAN DUTCHER:   Sure.

    3        DR. MARCO:   If I might.    Granted the

applicant showed us that this is actually the largest
    4

patient pool for a study for second-line KS.
    5                                             And

that's great, but, and these studies are very difficult
    6

to do, especially because the instance of KS might go
    7

down.
    8   But I just want -- others in the room who are

developing drugs who are hoping to get their drug
    9

approved for KS, whether it be used for KS or possibly
   10

another cancer in the future, just being able to come
   11

to the FDA with such a small sample size to get your
   12

drug approved on the fast track, is problematic.
   13                                                   So

I think we need to start holding companies to a higher
    14

standard and for larger patient studies when they come
   15

to us in the future.
   16

   17        DR. SCHILSKY:   I wonder if we could ask for

clarification for our benefit from FDA, not so much
   18

about the sample size which I don't agree with anything
   19

     said, but about the study design.
you 20                                    It was my

recollection from one of the written documents that
   21

this submission is not able to be considered for
   22

accelerated approval.
   23

   24        If that's the case, I think we'd like to

be clear on what the regulatory issues are.
   25                                            Because
if it's not to be considered for accelerated approval,
    26

then does that put it -- do we need to be considering
   27
                                                      129

it with respect to whether there is appropriate
    1

comparator data, you know since we don't have a
    2

randomized trial.
    3               Maybe Bob you could clarify some of

those issues.
    4

    5        DR. TEMPLE:    Well, accelerated approval

refers to willingness to approve a drug on the basis
    6

of a 7surrogate end point that has nothing overt to do

with 8clinical benefit.    It was not our view here that

there was need for use of that consideration here
    9

because, as Ken showed you, there are at least 12 or
   10

13 people who had persuasive clinical benefit, and you
    11

heard probably some of those people on that slide
   12

talking here today.
   13

   14        So, despite its name accelerated approval,

it's not an advantage to be under accelerated approval.
   15

It means you don't have actual clinical benefit
   16

demonstrated.
   17           The feeling here was that in this case

there is.
   18

   19        The question is how much data you need and

whether this is an adequate and well controlled study,
   20

albeit historically controlled study, is the sort of
   21

thing we invite you to discuss.
   22                               Studies without

control groups, without concurrent control groups are
   23

     our favorite kind of study because we like easy
not 24

decisions.
   25        And every time you have a non-concurrently
controlled study you have various agonies about how
   26

plausible the control is.
   27                        And when the environment is
                                                        130

changing, there are even more such agonies.
    1

    2         But, accelerated or not accelerated

doesn't go to that question.
    3                             The requirement for

accelerated approval is adequate and well controlled
    4

studies that support the effect on the surrogate.
    5                                                 And

in this case, we are certainly mindful of the fact that
    6

we have information about paclitaxel and it's safety
    7

and things like that.
    8                    So we are looking at a new use

     different population of a drug and the size of the
in a 9

database one expects there at least related to safety
   10

might be different from what you would expect if you
   11

were working up a drug de novo that had never been in
   12

people before.
   13

   14         CHAIRMAN DUTCHER:    Okay.   Question

number two.
   15         Paxene study resulted in a 42 percent

objective response rate in 89 patients using protocol
   16

specified criteria.
   17                   In an analysis using only

eligible patients, the objective response rate was 46
   18

percent.
   19      You may refer to the tables.     The question

being asked "Does the Paxene study show patient benefit
   20

based on the 42 percent cutaneous tumor response, the
   21

clinical benefit assessments and the quality of life
   22

assessments?"
   23

   24         Any discussion?

   25         [No response.]
   26         CHAIRMAN DUTCHER:    Okay.   All those who

feel that the study does show patient benefit, please
   27
                                                          131

raise your hand.
    1                High.   One, two, three, four, five,

six, 2seven, eight, nine, ten, eleven.      The vote is 11

yes. 3

     4         Question number three.    "Is the Paxene

safety acceptable in view of the efficacy results and
    5

results available with alternative therapy?"
    6                                               All

those who would say yes, please raise your hand.
    7

     8         DR. SWAIN:    I'd just like to make one

comment.
    9       I would definitely like to see more of the

patitoxicity data looked at.
   10

    11         CHAIRMAN DUTCHER:    Okay.   With

clarification of that patitoxicity particularly in the
   12

situation of protease inhibitors, is the Paxene safety
   13

acceptable in view of the efficacy results?
   14                                              Yes?

One, two, three, four, five, six, seven, eight, nine,
   15

ten, eleven.
   16            The vote is eleven yes.

    17         Question number four.    "Is the Paxene NDA

approvable for the indication of use after failure of
   18

first-line or subsequent systemic chemotherapy for the
   19

treatment of advanced AIDS-related Kaposi's sarcoma?"
   20

     Ozols?
Dr. 21

    22         DR. OZOLS:    Well here I think you have to

address -- I mean, that's pretty broad.
   23                                         Three months

    we
ago 24 approved another drug.      So how does that relate

to Taxol?
    25        What about a patient who has received Taxol
already for this indication, for basically the same
   26

indication that has progressed or stopped responding?
   27
                                                             132

Are we saying that they should also be candidates for
    1

Paxene?
    2

    3           CHAIRMAN DUTCHER:       Dr. Johnson says no.

Okay.
    4

    5           DR. JOHNSON:    I think we thought that was

obvious.
    6

    7           DR. OZOLS:    Well, I mean are they the same

drug, are they different drugs?
    8                                     Are you going to say

they 9are different formula drugs and there is different

proprietary drugs, they may have different responses,
   10

toxicities?
   11            All that's been alluded to.

   12           Are you saying that this is identical to

Taxol?
   13

   14           DR. JOHNSON:     We are not saying.    That's

     to
yet 15 be determined.

   16           DR. TEMPLE:     That's not fundamentally

different from what you make of the situation whenever
   17

there are two manufacturers who make the same active
   18

moidient to two different drug products.
   19                                             Usually your

thought is if you failed on one thing, you wouldn't try
   20

     generic.
the 21

   22           DR. OZOLS:     Right.

   23           DR. TEMPLE:     If that were what the case

was. But we have, I must say we have not actually told
   24

people that for reasons John just alluded to.
   25                                                   We
thought that was fairly clear.
   26                                   Different formulation,

     know.
you 27       One package in liposomal, one not.        that's
                                                          133

a different question.
     1                     But usually one thinks that they

are pretty similar with respect to response rates.
    2                                                     Of

course, you have no data on that.
    3

    4         DR. OZOLS:    Right.

    5         CHAIRMAN DUTCHER:      Any other comment?

    6         [No response.]

    7         CHAIRMAN DUTCHER:      Okay.   Is Paxene

approvable for the indication of use after failure of
    8

first-line or subsequent systemic therapy for
    9

treatment of advanced AIDS-related Kaposi's sarcoma?
   10

     those who vote yes?
All 11                       One, two, three, four, five,

six, seven, eight, nine, ten, eleven.
   12                                         The vote is

eleven yes.
   13

   14         Any other comments?

   15         [No response.]

   16         CHAIRMAN DUTCHER:      Thank you very much.

     meeting is adjourned.
The 17

   18         (Whereupon, the above matter was concluded

at 12:22 p.m.)
   19

   20

   21

   22

   23

   24

   25

   26

   27
    134

1

2

3

4

5

6

7

8

9

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:10
posted:11/11/2011
language:English
pages:134