1

DEPARTMENT OF HEALTH AND HUMAN SERVICES

FOOD AND DRUG ADMINISTRATION

CENTER FOR DRUG EVALUATION AND RESEARCH

JOINT MEETING OF THE

DERMATOLOGIC AND OPHTHALMIC DRUGS

ADVISORY COMMITTEE

and

DRUG SAFETY AND RISK MANAGEMENT ADVISORY COMMITTEE

Monday, July 12, 2004

8:00 a.m.

ACS Conference Room

5630 Fishers Lane

Rockville, Maryland

P A R T I C I P A N T S

DERMATOLOGIC AND OPHTHALMIC DRUGS ADVISORY

COMMITTEE

Robert S. Stern, M.D. - CHAIRMAN

Roselyn E. Epps, M.D.

Robert Katz, M.D.

Paula Knudson [Consumer Representative]

Sharon S. Raimer M.D.

Eileen W. Ringel, M.D.

Jimmy D. Schmidt, M.D.

Kimberly Littleton Topper, M.S., Executive

Secretary

DRUG SAFETY AND RISK MANAGEMENT ADVISORY COMMITTEE

Ruth S. Day, Ph.D.

Curt D. Furberg, M.D., Ph.D.

Jacqueline S. Garner, Ph.D., MPH

Eric S. Holmboe, M.D.

Arthur A. Levin, MPH [Consumer Representative]

Robyn S. Shapiro, J.D.

CONSULTANTS (VOTING)

Margaret Honein, Ph.D., MPH

Sarah Sellers, Pharm.D.

FDA PARTICIPANTS

Jonica Bull, M.D.

Denise Cook, M.D.

Tapash Ghosh, Ph.D.

Shiowjen Lee, Ph.D.

Jill Lindstrom, M.D.

Marilyn Pitts, Pharm.D.

Anne Trontell, M.D., MPH

Jonathan Wilkin, M.D.

Jiaqin Yao, Ph.D.

C O N T E N T S

Call to Order and Introductions

Robert Stern, M.D., Chairman, DODAC 5

Conflict of Interest Statement

Kimberly Littleton Topper, Executive

Secretary, DODAC 8

Welcome and Introduction

Jonica Bull, M.D. 11

Introduction and Overview of the Topic

Jonathan Wilkin, M.D. 13

Introduction to Psoriasis and the State of the

Armamentarium

Denise Cook, M.D. 14

Allergan NDA Presentation - Introduction:

Patricia Walker, M.D., Ph.D., Vice President,

Skin Care Pharmaceuticals, Allergan 38

Psoriasis: Disease Overview and Treatment Options

Alan Menter, M.D., Clinical Professor of

Dermatology, University of Texas, Southwestern 41

Tazarotene Pharmacology, Overview of Clinical

Development Program, Overview of Proposed

Risk-Management Program

Patricia Walker, M.D., Ph.D. 50

Risk Benefit Assessment

Alan Menter, M.D. 94

Discussion of Allergan Presentation 105

FDA Presentation 125

Toxicology Studies of Tazarotene

Jiaqin Yao, Ph.D. 131

Clinical Pharmacology and Biopharmaceutics 127

Tapash Ghosh, Ph.D. 139

C O N T E N T S (Continued)

Efficacy--Biostatistical Analysis of Pivotal

Studies

Shiowjen Lee, Ph.D. 156

Clinical Wrap-up

Denise Cook, M.D. 168

Evolution of Risk Management for Systemic Retinoids

Jill Lindstrom, M.D. 173

Risk Management Tools for Oral Tazarotene:

Context, Considerations and Experience

Ann Trontell, M.D., MPH 190

Open Public Hearing

Discussion and Questions

P R O C E E D I N G S

Call to Order and Introductions

DR. STERN: Good morning, everyone. I'm

Robert Stern, the Chairman of the Dermatologic and

Ophthalmologic Drugs Advisory Committee meeting.

And today we're here to consider the application of

oral tazarotene capsules for the treatment of

moderate to severe psoriasis, including

risk-management options to prevent fetal exposure.

I'd like to start the meeting by welcoming

everyone, and then beginning directly across with

me--if everyone would introduce themselves in terms

of their role at this meeting.

DR. HONEIN: I'm Peggy Honein. I'm an

epidemiologist with the CDC's Birth Defects group.

And I'm here as part of Drug Safety Committee.

DR. FURBERG: I'm Curt Furberg at Wake

Forest University. I'm a member of the Drug Safety

and Risk Management Advisory Committee.

DR. KATZ: Robert Katz. I'm a

dermatologist in private practice. Im a member of

the Drug Advisory Committee.

DR. KNUDSON: I'm Paul Knudson. I'm the

Consumer Representative on the Dermatology Advisory

Committee.

DR. SELLERS: I'm Sarah Sellers. I'm a

pharmacist and a drug-safety expert.

DR. SCHMIDT: I'm Jimmy Schmidt, private

practice in Houston, and I'm on the committee.

DR. RAIMER: Sharon Raimer, University of

Texas, Galveston. I'm on the Dermatology

Committee.

DR. EPPS: Roselyn Epps, Chief of

dermatology, Children's National Medical Center,

and member of the Dermatology Advisory Committee.

MS. SHAPIRO: Robyn Shapiro, Director of

the Bioethics Center at the Medical College of

Wisconsin, and I'm on the Drug Safety Committee.

DR. RINGEL: Eileen Ringel. I'm on the

Dermatological Advisory Committee. I'm a

dermatologist in private practice in Waterville,

Maine.

DR. STERN: And, again, I'm Rob Stern. I'm

a dermatologist from Boston.

MS. TOPPER: I'm Kimberly Topper. I'm the

Executive Secretary for this committee.

DR. GARDNER: Jacqueline Gardner,

University of Washington School of Pharmacy, on the

Drug Safety Committee.

DR. WILKERSON: Michael Wilkerson,

dermatologist and member of the DODAC committee.

DR. DAY: Ruth Day, Duke University. I

direct the medical cognition lab there, and a

member of the Drug Safety Committee.

DR. TRONTELL: Anne Trontell, Deputy

Director of the Office of Drug Safety in the Center

of Drugs at FDA.

DR. COOK: Denise Cook, I'm a Medical

Office in the Division of Dermatologic and Dental

Drug Products.

DR. WILKIN: Jonathan Wilkin, Director,

Division of Dermatologic and Dental Drug Products,

Center for Drugs.

DR. BULL: Good morning--Jonica Bull, the

Director of the Office of Drug Evaluation V.

DR. STERN: Thank you very much.

We'll now begin with Dr. Bull giving some

introductory--oh, we'll, now begin with conflict of

interest, from the person at my right, Ms. Topper.

Conflict of Interest Statement

MS. TOPPER: Thank you.

The following announcement addresses the

issue of conflict of interest with regard to this

meeting, and is made as part of the record to

preclude even the appearance of such at this

meeting.

Based on the submitted agenda for the

meeting, all financial interests reported by the

committee participants, it has been determined that

all interest in firms regulated by the Center for

Drug Evaluation and Research present no potential

for an appearance of a conflict of interest at this

meeting, with the following exceptions.

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

full waivers have bee granted to the following

participants: Dr. Michael Wilkerson, for his

speakers bureau activities for a competing firm,

which he receives less than $5,001 per year; Dr.

Curt Furberg, for his unrelated consulting for a

competing firm, which he receives less than $10,001

per year; Dr. Stern, for his unrelated consulting

for three competing firms, for which he receives

less than $10,001 per year, and from one firm, and

between $10,001 and $50,000 per year from the other

two firms; Dr. Ruth Day, for her unrelated

consulting for a competing firm, for which she has

greater than $50,000 pending.

In accordance with 21 U.S.C. 355(n)(4), an

amendment of the section of 505 of the Food and

Drug Modernization Act, waivers have been granted

for the following participants: Dr. Sharon Raimer

owns stock in two competing firms, worth between

$5,001 and $25,000 each; Dr. Sarah Sellers owns

stock in a competing firm worth between $5,001 and

$25,000. Because these stock interests fall below

the de minimis exemption allowed under 5 C.F.R.

2640.202(a)(2), a waiver under 18 U.S.C. 208 is not

required.

A copy of the waiver statements may be

obtained by submitting a written request to the

agency's freedom of information office, Room 12A-30

of the Parklawn Building.

There will be no industry representative

at today's meeting. As you may be aware, the FDA

has appointed industry representatives who

currently serve on each of these committees, but

both appointed industry representatives work with

the sponsors that are directly affected by the

matter before the joint committee.

In the event that the discussions involve

any other products or firms not already on the

agenda, for which an FDA participant has financial

interest, the participants are aware of the need to

exclude themselves from such involvement, and their

exclusion will be noted for the record.

With respect to all other participants, we

ask, in the interest of fairness, that they address

any current or previous financial involvement with

any firms they may wish to comment upon.

Thank you.

DR. STERN: Thank you very much.

Dr. Bull?

Welcome and Introduction

DR. BULL: Welcome. Our thanks to all of

you present who have taken time to be with us this

morning. Our thanks must include an acknowledgment

of the time the Advisory Committee members have

spent reviewing the background materials provided.

I would also like to extend my thanks to

an extraordinary group of scientists in the Center

for Drug Evaluation and Research, from the Division

of Dermatologic Drugs, the Office of Biostatistics,

the Office of Biopharmaceutics, who will be

presenting to you this morning. As well, I would

also like to acknowledge the work of the project

manager in the Division of Dermatologic Drugs,

Khalyani Bhatt, as well as a standing team from the

Executive Operations Office of Advisors and

Consultants, Ms. Kimberly Topper and Ms. Shalini

Jain.

The purpose of an advisory committee

meeting is to provide expert scientific advice and

recommendations to the agency regarding clinical

investigations and proposed marketing approval for

a drug product. Our focus for today's deliberation

is an application for oral tazarotene for the

treatment of moderate to severe psoriasis,

including risk management options to prevent fetal

exposure.

The mission of the Center for Drug

Evaluation and Research is to assure that safe and

effective drugs are available to the American

people. This means that we thoroughly assess the

adequacy of the clinical trial design and endpoints

for a proposed treatment--in this instance that of

psoriasis, as well as the adequacy of the trial

outcomes in support of the product's efficacy,

safety, and its overall risk-to-benefit.

This committee deliberated earlier this

year on another drug in this class of products, and

the continuing challenges faced in risk management

to ensure safe use and the optimal minimization of

adverse events, especially those related to fetal

exposure during a course of treatment. Our hope is

that the background materials and presentations

provided by the FDA and by Allergan will assist you

in responding to the agency questions, and provide

for a thorough and independent deliberation of the

important issues at hand.

We look forward to a productive and

informative day.

Thank you.

DR. STERN: Thank you.

Now, Dr. Wilkin will speak to us.

Introduction and Overview of the Topic

DR. WILKIN: Psoriasis is a very common

disorder. It's a chronic disorder, and it's a very

costly disorder, in terms of both monetary

expenses, and also in terms of the quality of life

of those patients who have psoriasis.

We'll have two speakers this morning: one

representing industry--Dr. Menter--and one from

FDA, Dr. Cook--who will describe the current

landscape available to dermatologists--the current

products in the armamentarium for psoriasis.

I think one of the pieces that will become

apparent is that there is no perfect drug. There

are products which have definite side effects;

other products which are very new, and we're still

going to be learning about their side effects. No

product has perfect efficacy.

And so this is the background against

which, I think, the committee needs to deliberate

in their recommendations for the particular product

today.

We do have a major focus on the

risk-management program to prevent fetal exposure,

but we must not lose sight that we're also thinking

about the overall balance between benefit and risk

for this product.

Thank you.

DR. STERN: Thank you very much.

And now Dr. Cook will talk to us a bit

about psoriasis.

Introduction to Psoriasis and the State

of the Armamentarium

DR. COOK:[Off mike.] [Inaudible.]

Sorry--can you hear me now?

We thought it appropriate, since people

were from varying backgrounds, to give a review on

psoriasis. I apologize for those who are

well-versed in the disease process.

[Slide.]

Psoriasis is a polygenic disease, and

varying triggering factors--for example, trauma,

infections or medications may elicit a psoriatic

phenotype in predisposed individuals.

Today, I'm going to speak on the

prevalence of psoriasis, the genetics and

pathogenesis; the clinical variants of psoriasis,

and the state of the armamentarium as it exists

today.

[Slide.]

Psoriasis occurs in approximately 2

percent of the world's population. The prevalence

in the United States may be as high as 4.6 percent.

Its highest incidence occurs in Caucasians. In

Africans, African Americans and Asians, the

incidence of psoriasis is somewhere between 0.4 and

0.7 percent.

[Slide.]

There is an equal frequency in males and

females. It occurs in a one-to-one ratio. It may

occur at any age from infancy to the 10

th decade of

life.

The first signs of psoriasis occurs in

females at a mean age of about 27 years, and in

males at 29 years.

[Slide.]

There are two general peaks of occurrence:

one at age 20 to 30 years, and one between 50 and

60 years.

Psoriasis in children is very low. The

incidence is between 0.5 and 1.1 percent in

children 16 years and younger, and the man age of

onset--when it does occur in children--is between 8

and 12.5 years.

[Slide.]

Two-thirds of patients who have the

disease have mild disease. One-third of patients

have moderate to severe disease.

Early onset--which is usually prior to age

15--is associated with more severe disease, and

these patients are more likely to have a positive

family history.

As mentioned earlier, this is a life-long

disease. The remitting and relapsing of the

disease entity is unpredictable. There have been

spontaneous remissions of up to five years reported

in approximately 5 percent of patients who suffer

from psoriasis.

[Slide.]

The genetics and pathogenesis of

psoriasis: there's a lot of information that

psoriasis is linked to the immune system, and that

the major histocompatibility complex where

psoriasis has been shown is on the short arm of

chromosome 6. It's also linked to many

histocompatibility antigens; the most common, and

the one with the highest risk of family history, is

HLA-Cw6. Other HLA antigens associated with

psoriasis include HLA-B13, -B17, -B37 and B216.

It's also felt that psoriasis may have a

t-lymphocyte-mediated mechanism associated with its

pathogenesis.

[Slide.]

Psoriasis is not just confined to the

skin, and there is evidence that this is a system

disease, and that it's from the Koebner Phenomenon,

which happens on normal skin, where patients may

have trauma, and then the lesions of psoriasis

appear. Patients also have been show to have an

elevated erythrocyte sedimentation rate; increased

uric acid levels may lead to gout; patients may

have mild anemia; elevated à

2-macroglobulin; they

may have elevated IgA levels; and they may also

have increased quantities of immune complexes.

[Slide.]

Psoriasis also may be associated with

arthropathy, and there is also an aggravation of

psoriasis by systemic facts--as I mentioned at the

beginning of the talk--and that could include

medications, focal infections, stress.

Psoriasis also comes in the form of

life-threatening disease. And there are two

variants of that that I'm going to speak about

later: erythrodermic psoriasis, and pustular

psoriasis.

Now I'm going to speak about the clinical

variants of psoriasis.

[Slide.]

The characteristic lesion of psoriasis is

a sharply demarcated erythematous plaque with

micaceous silvery white scale. This is supported

histopathologically by a thickening of the

epidermis; tortuous and dilated blood vessels; and

an inflammatory infiltrate, primarily of

lymphocytes.

[Slide.]

And here, from Bolognia--where all the

pictures that you're going to see--clinical

pictures that you're going to see--is from this

textbook of dermatology by Bolognia--and here we

have an erythematous plaque. You can see the

outline of the erythema; the elevation of the

plaque above the skin surface, and the thick

micaceous, silvery scale.

[Slide.]

The severity of the disease is usually

characterized by three cardinal signs of psoriasis:

plaque elevation, erythema and scale. Body surface

area also plays a part. Patients are very

concerned about how much of their skin surface is

covered by the disease. But in determining

severity, it could be very complex, because

different people see body surface area differently.

[Slide.]

The most common variant of psoriasis is

the chronic plaque psoriasis. The plaques may be

as large as 20 cm; psoriasis is usually a

symmetrical disease. The sites of predilection can

include the elbows, the knees, the presacrum,

scalp, the hands and the feet.

[Slide.]

I'm going to show you some pictures now of

chronic plaque psoriasis. Here you can see that

the disease is very symmetrical, and can involve a

decent part of the body surface area.

[Slide.]

This is a picture of psoriasis of the

feet.

[Slide.]

Now, chronic plaque psoriasis may be

widespread. It can cover up to 90 percent of the

body surface area. The genitalia can be involved

in up to 30 percent of patients. Most patients

also have nail changes which include nail pitting

and "oil spots." And sometimes the involvement of

the nail bed is very severe, with onychodystrophy

and loss of the nail plate.

[Slide.]

Here is a picture of widespread chronic

psoriasis. And I think all of us would agree that

this is probably a severe case of psoriasis.

[Slide.]

This is a picture of the genitalia with

psoriasis.

[Slide.]

Here is a picture of psoriasis of the

nail, with nail pitting and oil spot, where the

nail is--the nail plate is being separated from the

nail bed.

[Slide.]

And some more severe form of nail

psoriasis, with, again, oil spots, onychodystrophy,

and loss of the nail plate.

[Slide.]

Symptoms of psoriasis include pruritus,

pain. Patients who have widespread psoriasis

sometimes complain of excessive heat loss. Also,

patients hate the way the disease looks; sometimes

have low self-esteem, have feelings of being

socially outcast and really dislike the excessive

scaling.

[Slide.]

The next variant of psoriasis that I'm

going to speak about is guttate psoriasis. It's

characterized by numerous 0.5 to 1.5 cm papules and

plaques; usually has an early age of onset. It's

the most common form in children, often triggered

by streptococcal throat infection.

In children, the remissions may be

spontaneous. In adults it's often chronic.

[Slide.]

Here is a clinical presentation of guttate

psoriasis, with the small papules, and plaque here.

And this is a picture of someone who had

an eruption of guttate psoriasis after a sunburn.

[Slide.]

The life-threatening forms of psoriasis

are generalized pustular psoriasis and

erythrodermic psoriasis.

[Slide.]

Generalized pustular psoriasis is an

unusual manifestation of the disease. It can have

a gradual or an acute onset. It is characterized

by waves of pustules on erythematous skin after

short episodes of fever, from 39 to 40 degrees

centigrade. Patients may have weight loss, muscle

weakness, hypocalcemia, leukocytosis and an

elevated ESR.

[Slide.]

The cause is obscure, but we do know that

there are several triggering factors, and they

include: infection, pregnancy, lithium,

hypocalcemia secondary to hypoalbuminemia; irritant

contact dermatitis, and withdrawal of

gluccocorticosteroids, primarily systemic.

[Slide.]

And here is a clinical presentation of

pustular psoriasis. And you can see the erythema,

with the pustules scattered about.

[Slide.]

Erythrodermic psoriasis--in this disease,

which is also a life-threatening form of psoriasis,

the classic lesion of psoriasis is lost. The

entire skin surface becomes markedly erythematous,

with desquamative scaling. Often the only clues to

the underlying psoriasis are the nail changes, and

usually there's facial sparing in erythrodermic

psoriasis.

[Slide.]

Triggering factors may include systemic

infection, withdrawal of high potency topical or

oral steroids; withdrawal of methotrexate;

phototoxicity, and irritant contact dermatitis.

[Slide.]

Here is the clinical presentation of

erythrodermic psoriasis in a patient after

withdrawal of methotrexate.

[Slide.]

Now, I'm going to speak of the state of

the armamentarium of psoriasis. We're mainly going

to focus on moderate to severe psoriasis, since

that's the topic of the drug product under

consideration for today.

There is a wide range of therapies for

moderate to severe psoriasis. None induce a

permanent remission, and all have side effect that

can place limit on their use, and usually require

that patients are treated in a cyclical fashion.

[Slide.]

These therapies include topical

corticosteroids, topical vitamin D3 analogues,

topical retinoids, photochemotherapy, and systemic

therapies which may be oral or parenteral.

[Slide.]

Topical corticosteroids that are usually

used in moderate to severe psoriasis are those of

the high potency and super potent topical steroids.

These include the fluocinonide family,

betamethasone dipropionate cream, the clobetasol

priopionate family, diflorasone diacetate ointment,

and betamethasone dipropionate ointment.

[Slide.]

The side effects associated with use of

these drugs include skin atophy, burning and

stinging; and, systemically, suppression of the

hypothalamic-pituitary-adrenal axis. This may

occur after two weeks use with certain topical

corticosteroids. Usually those are the super

potent type.

[Slide.]

Topical vitamin D

3 analogues--the prototype

for this group is calcipotriene. There are three

formulations: cream, ointment and scalp solution.

The former two are approved for plaque psoriasis,

the latter for moderate to severe psoriasis of the

scalp.

[Slide.]

Side effects for topical vitamin D3

analogues are primarily cutaneous, and include

burning, stinging, pruritus, skin irritation and

tingling of the skin.

[Slide.]

The topical retinoids that are approved

for the treatment of plaque psoriasis are

tazarotene gel and cream. They are available in

two strengths: 0.05 percent, and 0.1 percent.

The side effects include pruritus,

burning/stinging, erythema, worsening of psoriasis,

irritation, skin pain. And there have been cases

of hypertriglyceridemia.

[Slide.]

Additional indicatiosn for topical

tazarotene in the 0.1 percent gel is approved for

the treatment of facial acne vulgaris of mild to

moderate severity. And the 0.1 percent cream is

also approved as an adjunctive agent for use in the

migitation of facial fine wrinkling, facial mottled

hyper-and hypopigmentation, and benign facial

lentigines in patients who use comprehensive skin

care and sunlight avoidance programs.

[Slide.]

Topical tazarotene--both products are

pregnancy category X. They are contraindicated in

women who are or may become pregnant. And there

are some requirements before and during therapy.

These include a negative pregnancy test two weeks

prior to initiation of therapy. Therapy must be

initiated during a normal menses. And women of

childbearing potential should us adequate birth

control.

[Slide.]

Now I'm going to speak on

photogemotherapy. There are two types of

phototherapy for the treatment of moderate to

severe psoriasisThese include ultraviolet B, or

UVB; and ultraviolet A plus psoralen, more commonly

known as PUVA.

[Slide.]

There are two types of UVB: broadband UBV

and narrowband UVB. The treatment is time

consuming. Patients usually must come two to three

visits per week for several months. And the side

effect is possibility of experiencing an acute

sunburn reaction.

[Slide.]

PUVA consists of ingestion of or topical

treatment with a psoralen followed by UVA. It is

usually reserved for severe, recalcitrant,

disabling psoriasis. This form of treatment for

psoriasis is also time-consuming. It usually

requires two to three visits per wekk, and at least

six weeks of treatment to get clerance.

There are several precautions that must be

taken for patients who are treated with PUVA.

Patients must be protected from further UV light

for 24 hours post treatment. And with oral

psoralen, they must have wrap-around UV-blocking

glasses for 24 hours post treatment.

[Slide.]

Side effects with oral psoralen include

nausea, dizziness and headache. Early side effects

with PUVA are pruritus, but late side effects

include skin damage, and the increased risk for

skin cancer, particularly squamous cell skin

cancer; and after maybe 200 to 250

treatments--which is really a long time--patients

may be at increased risk for melanoma.

[Slide.]

Contraindications to PUVA include patients

less than 12 years of age; patients with a history

of light sensitive disease states; patients with,

or with a history of melanoma; patients with

invasive squamous cell carcinoma; and patients with

aphakia.

[Slide.]

Now, the system therapies--these come in

two types: oral and parenteral. The oral therapies

are methotrexate, Neoral--or cyclosporine--and

Soriatane--acetretin. The parenteral therapy

includes, most recently approved biologics which

are Amevive, Raptiva and Enbrel. And I will

speak--as a prototype--on Amevive, which was first

approved.

Methotrexate is a folic acid antagonist,

usually reserved for severe, recalcitrant,

disabling psoriasis. Maximum improvement can be

expected after eight to 12 weeks.

[Slide.]

The contraindications for methotrexate

include nursing mothers, patients with alcoholism,

alcoholic liver disease, patients with other

chronic liver disease; patients with overt or

laboratory evidence of immunodeficiency syndromes,

and patients who have preexisting blood dyscrasias.

[Slide.]

This drug product is also a Category X.

It's contraindicated in pregnant women with

psoriasis, and pregnancy must be excluded in women

of childbearing potential, and pregnancy should be

avoided if either partner is receiving methotrexate

during and for a minimum of three months after

therapy for male patients and for at least one

ovulatory cycle after therapy for female patients.

[Slide.]

Side effects of methotrexate are numerous.

They include acute or chronic hepatotoxicity,

hepatic cirrhosis, leukopenia, thrombocytopenia,

anemia, stomatitis, nausea/volmitting, alopecia,

photosensitivity, burning of skin lesoins and,

rarely, interstitial pneumonitis.

[Slide.]

Multiple screening tests are necessary

before using methotrexate. There are also

recommendations for hepatic monitoring, which

include period liver function tests, including

serum albumin--although, I must say, liver function

tests are not a good screen with methotrexate for

hepatic damage. Therefore, there are

recommendations for liver biopsy which include

doing it pretherapy or shortly thereafter, also

after a cumulative dose of 1.5 grams, and after

each additional 1 to 1.5 grams of use.

[Slide.]

Neoral, or cyclosporine, is a potent

immunosuppressive. It is approved for adults that

are non-immunocompromised, with severe,

recalcitrant plaque psoriasis. Maximum efficacy is

achieved after about 16 weeks of therapy.

There are contraindications for use of

this drug, which include concomitant PUVA or UVB

therapy; using methotrexate or other

immunosuppressive agents; using coal tar or

radiation therapy. Patients with abnormal renal

function; patients with uncontrolled hypertension;

patients with malignancies and nursing mothers

cannot use this drug.

[Slide.]

There are many side effects for Neoral.

The highest ones are the possibility of

irreversible renal and onset of hypertension; then

headache, hypertriglyceridemia, hirsutism,

pareshesias, incluenza-like symptoms, nausea,

vomiting, diarrhea, lethary and arthralgia.

[Slide.]

Multiple screening tests--prescreening

tests--are needed for use of Neoral. And the tests

must continue throughout treatment, with dosage

adjustment as necessary to prevent end-organ

damage.

[Slide.]

Soriatane is the only oral retinoid that's

approved for psoriasis, and it's approved for the

treatment of severe psoriasis in adults. One can

see significant improvement with therapy after

eight weeks.

[Slide.]

Contraindications for use of Soriatane

include patients with severely impaired liver or

kidney function; patients with chronic abnormally

elevated blood lipid values; patients who are

taking methotrexate; and patients who use ethanol

when on therapy and for two months following

therapy in female patients.

[Slide.]

Soriatane is also a pregnancy Category X

drug product as it is a human teratogen. It's

contraindicated in pregnant females or those who

intend to become pregnant during therapy or anytime

up to three years post therapy.

[Slide.]

Side effects with Soriatane are those that

are usually associated with oral retinoid therapy,

and include chelitis, alopecia, skin peeling, dry

skin, pruritus, rhinitis, xeropthlamia, and

arthralgia.

[Slide.]

There are many laboratory abnormalities

also, and those include hypertriglyceridemia,

decreased HDL, hypercholesterolemia, elevat3d liver

function tests, elevated alkaline phosphatase,

hyperglycemia and elevated CPK. However hepatitis

and jaundice occurred in less that 1 percent of

patients in the clnical trials on Soriatane.

[Slide.]

Multiple prescreening tests also must be

used for Soriatane, and you must have continued

monitoring throughout therapy, with possible dosage

adjustment.

[Slide.]

The parenteral therapy, as I mentioned

before, are lately on the scene. And the one I'm

going to speak on is Amevive. It is an

immunosuppressive dimeric fusion protein. It's

made up of an extracellular CD2-binding portion of

the human leukocyte function antigen-3, which is

linked to the Fc portion of the human IgG1

molecule.

[Slide.]

Amevive is indicated for the treatment of

adult patients with moderate to severe chronic

plaque psoriasis. With 12 weeks of therapy, a

disease state of clear or almost clear was achieved

by 11 percent of patients via the intravenous

route, and 14 percent of patients via the

intramuscular route.

[Slide.]

The side effects with Amevive include a

dose-dependent reduction in circulating CD4 and CD8

T lymphocytes. Therefore this drug should not be

administered to patients with low CD4 counts. CD4

counts must be monitored before and weekly

throughout therapy.

[Slide.]

Side effects that have been associated

thus far with Amevive have been lymphopenia.

There's also been an increased risk of

malignancies, particularly skin cancer--or basal

cell carcinoma and squamous cell carcinoma--and an

increased risk for lymphoma.

There have been serious infections

requiring hospitalization. There is also a risk of

reactivation of chronic, latent infections, and of

hypersensitivity reactions.

[Slide.]

hopefully this has given you a good

background on the disease of psoriasis, and also

the state of the armamentarium for treating this

disease.

Thank you.

DR. STERN: Thank you very much for a very

nice presentation.

Could I ask two quick questions?

The first is: topical tazarotene, the

package labeling says that there should be a

pre-treatment pregnancy test in women who might be

or become pregnant. Is that the labeling for

topical tazarotene?

DR. COOK: Yes, what I had up there, it's

directly out of the label.

DR. STERN: Okay. And the second is: you

had, for acitretin that the indication is severe

psoriasis, not moderate to severe psoriasis.

DR. COOK: Yes, severe--

DR. STERN: It's severe.

DR. COOK: It's severe psoriasis.

DR. STERN: Okay. Thank you very much.

Thank you for a great presentation. It was very

clear. And I enjoyed it.

And now we will go on to the Allergan

presentation, with Dr. Patricia Walker, Vice

President of the Skin Care Pharmaceuticals

Division, giving the introduction for the sponsor's

application.

Allergan NDA Presentation

Introduction

DR. WALKER: Good morning.

Allergan is here today to seek approval

for our oral tazarotene gel formulation--gel

capsule formulation--for the treatment of moderate

to very severe psoriasis.

What I'd like to show you today is that

tazarotene is a retinoid, and as a retinoid, it

does have some unique pharmacology and receptor

activity. We've demonstrated efficacy in the

treatment of moderate to very severe psoriasis. We

feel our drug is differentiated from other

retinoids--and actually has an improved safety

profile relative to other drugs in this class.

Tazarotene is a teratogen--or a probable

teratogen--and we've developed a Risk Minimization

Action Plan around this.

[Slide.]

It is available in a topical formulation,

as you heard this morning from Dr. Cook. The gel

formulation was approved in 1997 for the treatment

of psoriasis, and for acne at that time. A cream

formulation was approved in 2000 for the treatment

of psoriasis; 2001 for the treatment of acne; and

then, later for the treatment of photodamage, or

signs and symptoms of photoaging.

At the time of the psoriasis cream

approval, we developed and worked with the

Derma-Dental Division to develop a new scoring

system, which we refer to as the "overall lesional

assessment. Later in the morning, in my talk, I'll

go over that assessment.

We started the oral tazarotene formulation

development in 1998, with Phase 2 studies. We

initiated the Phase 3 studies in 2001, and we filed

the NDA last November, in 2003.

Just to set the stage, we've studied many

patients with this drug. WE have nearly 1,700

patients studied with oral tazarotene, 901 of which

have been treated with at least 4.5 mg or higher.

[Slide.]

The introduction is from me, then you're

going to hear from Dr. Alan Menter, who's going to

give you a brief overview of the disease, and what

the unmet need is, and the treatment options.

I'll come back and share with you some of

the pharmacology of tazarotene; what the clinical

development's been; and our proposed risk

minimization action program.

And then Dr. Menter will wrap up with a

risk benefit assessment.

[Slide.]

Available to answer questions today are

several of my colleagues from Allergan in various

disciplines--

[Slide.]

--as well as some consultants who have

worked with us extensively on analyzing and looking

at our data.

At this time, I'd like to turn the podium

over to Dr. Menter to give you the disease overview

and treatment options.

Psoriasis: Disease Overview and Treatment Options

DR. MENTER: Thank you, Dr. Walker--Mr.

Chairman, colleagues, patients, ladies and

gentlemen.

My name is Alan Menter, and I'm a

clinician, practicing dermatologist in Dallas,

Texas. From a conflict of interest conflict of

interest point of view, I have consulted with

Allergan, and have been involved in clinical

research with Allergan, with oral tazarotene, as

well as with multiple other drugs related to

psoriasis. I do not own any stock in Allergan

corporation.

[Slide.]

As we've so eloquently heard this morning

from Dr. Cook, psoriasis is a common disease. It

is probably one of what we consider the autoimmune

diseases in all medical conditions. And I'm not

going to reiterate some of the things that Dr. Cook

mentioned in her excellent review, but just merely

highlight a few issues that I believe are important

when considering systemic therapy for psoriasis

patients.

[Slide.]

The prevalence, as she has mentioned, is

equal in male and females. And I think this is an

important issue when we come to talk about patients

who are candidates for systemic therapy, because I

believe at this stage, a number of

patients--particularly young females of

child-bearing potential--are currently excluded

from systemic therapy because of pregnancy issues.

And, as she mentioned, there are multiple

genes associated with psoriasis. And I think also

of importance is the fact that psoriasis is linked,

as a systemic disease, with other immune-mediated,

or autoimmune disease such as diabetes, lupus,

Crohn's--and there have been many genetic linkages

found in which psoriasis patients have other

diseases like diabetes, lupus and Crohn's disease.

We all recognize that psoriasis is a

condition that patients struggle with. And, as Dr.

Wilkins said, quality of life--that I'd like to

stress--is a major issue. This is not just a

physical problem that patients have to put up with,

they have to bear the emotional struggle that comes

with facing themselves on a day-to-day basis, their

loved ones, their peers, on a day-to-day basis with

this condition we call psoriasis; and itching, and

pain, and disfigurement are common. And patients

will tell you about the problems they have relating

to dealing with the day-to-day manifestations of

psoriasis.

[Slide.]

From a point of view of pathogenesis, I

think we don't recognize--as Dr. Wilkin also said,

and Dr. Cook mentioned--that psoriasis has to be

considered not a skin disease. This is a systemic

disease. And I think for too long we, as

clinicians, have really overlooked the systemic

nature of psoriasis. It is certainly a disease

that is driven by the immune system, by T cell

proliferations, the release of various

cytokines--chemicals that then induced this

hyperproliferation and the scale that we see

inherent in patients with psoriasis.

So I do believe that we must no longer

look at psoriasis as a pure skin disease; look at

it as a systemic disease like we do other

immune-mediated systemic diseases, like I

mentioned.

[Slide.]

It's a diverse disease. Eery patient with

psoriasis looks different. For those of us in

clinical practice who see psoriasis on a day-to-day

basis, psoriasis patients may, at first glance,

look similar. But there are nuances, there are

differences in expression of the disease. And even

within one individual patient, their disease may

change from discoid psoriasis, as Dr. Cook showed,

to pustular psoriasis, to erythrodermic psoriasis,

and back again to ordinary psoriasis.

[Slide.]

She showed pictures of genital

involvement. This leads to massive issues in

interpersonal relationships. And no longer can we

consider genital involvement, scalp involvement, as

mere nuisance issues. These are issues that really

do involve patients' interpersonal relationships,

at work and at home, on a day-to-day basis. And we

have to take cognizance of the fact that psoriasis

has a major burden on the quality of life. And for

those patients in the audience today, Im sure they

could tell you the issues that they deal with on a

day-to-day basis related to quality of life. It's

not just physical functioning, but the mental

functioning as well.

And I think when we consider retinoids,

it's interesting to note that there are retinoids

for non-dermatologic conditions, like leukemia and

cutaneous lymphoma. And psoriasis rarely has been

shown, in all aspects of quality of life, to impact

negatively, equally, if not worse, the mental and

physical aspects of a patient's life, with cancer

patients, arthritis patients, and diabetes

patients. So, again, stressing the quality of life

issues that are inherent in this.

And I've mentioned interpersonal

relationships, and I've mentioned work disability

as well.

And, as Dr. Wilkin says, this is a costly

disease, and it's not getting any cheaper as new

drugs become available to us. But I do not believe

we need to take a backseat to colleagues in other

areas of medicine who have expensive drugs

available to them to treat diseases like arthritis

and Crohn's disease.

[Slide.]

If one takes patients with various areas

of psoriasis--palmar/plantar psoriasis--a patient

who's--which is a fairly common area of

involvement--patients struggle with locations on

the palms, even though this may only affect a small

proportion of the body. Patients--who you can see

here--with palms and soles do not get by with

creams and ointments. They frequently need

systemic therapy to control their disease. So,

body surface area by itself should not be used as a

pure parameter for indication of systemic disease.

And the treatment has to be considered asa

life-long treatment. Psoriasis patients--as has

been mentioned by Dr. Cook and myself--patients

start early in life with psoriasis. The vast

majority of patients present before the age of 36.

So, for those of them who life a long life,

basically, they have to deal with this for the next

50, 60 years of their life. And treatment has to

be tailored accordingly. You cannot treat

psoriasis for three weeks, for three months, for

six months or for one year. Treatment is a

life-long treatment. And no cures are currently

present at the current time.

[Slide.]

So where are we with psoriasis therapies?

It's probably true to say that psoriasis is a very

under-treated disease, and there are various

reasons for this.

If we look at the figures--that has been

mentioned by Dr. Cook and myself--of approximately

10 to 25 percent of patients; in the United States,

that means a minimum of 450,000 patients have

moderate to severe disease. Currently, only about

125,000 of those patients are being treated with

systemic therapy. So, hence, there are two-thirds

of the patients with moderate to severe psoriasis

are not being treated. And the question is: why?

And I think the reasons are shown here.

There are safety concerns with the drugs. It's

time-consuming to put up with the day-to-day issues

relating to these drugs. There's monitoring

involved in psoriasis; and, obviously the cost

issue, as well.

[Slide.]

So, I'm not going to review the

side-effect profile. All these drugs work well.

And I think we have to recognize that we're not

here to knock any individual product. We have

great drugs. We've had methotrexate available for

30 years. We've had retinoids available for nearly

20 years. The biologic drugs are new. But all of

these drugs have issues relating to them that make

for monitoring and make for difficulty in

day-to-day management of these patients.

[Slide.]

So, basically, in summary: psoriasis is

not a single disease. It is a very diverse

disease. It is a disease that has major problems

and quality of life issues.

And the other aspect we have to recognize

is that as patients age, they develop co-morbid

conditions. They develop liver problems which

precludes them from certain therapies such as

methotrexate. They may have compromised renal

function which precludes cyclosporine. And all

day--and my colleagues--my dermatology colleagues

in the audience--are faced with making choices of

drug therapies for patients who have co-morbid

conditions that will preclude certain drugs. So we

definitely need a full range of treatment.

And I do believe that it is important for

our psoriasis population that we have--as we do

have in other systemic diseases--a full range, and

comprehensive range of medications so that we can

choose, in conjunction with our patients, the

correct therapy for our patients.

Thank you.

DR. STERN: Thank you very much.

Oral Tazarotene - Pharmacology, Clinical

Development, Risk Minimization Plan

DR. WALKER: I'm now going to share with

you the pharmacology of tazarotene, and what we

think makes it unique; the clinical development

program; and the risk minimization plan that we are

proposing.

[Slide.]

Just to summarize the data that we have

for tazarotene as a molecule, it is a prodrug. It

actually has only one active metabolite, and that's

tazarotenic acid. It's what's known as a

third-generation retinoid, or acetylenic retinoid.

It's a locked molecule, which prevents

isomerization and non-specific binding, and it's a

receptor-selective molecule.

[Slide.]

As already mentioned, retinoids have been

on the market for a long time--both natural and

synthetic forms--for over 20 years. These are very

well known to dermatologists, but they're also used

outside of dermatology. There's isotretinoin,

altrans retinoic acid, etretinate--which has now

been replaced by acitretin-- and bexarotene.

Retinoids are known to be essential for

normal epithelial proliferation, differentiation,

and embryo-fetal development.

There are two types of retinoid receptors

that retinoids act through: the RAR receptors and

the RXR receptors. These receptors always occur as

a dimer. They can occur as either a hetero-dimer,

with the RAR binding with an RXR, or as a

homo-dimer, RXR-RXR receptors.

[Slide.]

There are also subtypes of these

receptors. The receptor combinations and subtypes

are important because there are different side

effects noted--and different biological effects

noted--with each subtype. And there's also

tissue-specific receptor expression.

[Slide.]

The current retinoid therapies used in

dermatology--primarily acitretin and

isotretinoin--are what are known as pan-agonists.

Acitretin is a pan-agonist for all three subtypes

of the RAR receptor. Isotretinoin and its

metabolites are pan-agonists for the three subtypes

of the RAR receptor, as well as the RXR receptor.

This is important because activation of

these subtypes are related to many of the side

effects that we heard about this morning from Dr.

Cook, such as hyperlipidemia, hepatotoxicity,

epistaxis, eye irritation and dryness--those side

effects are specifically associated with the RARà

subtype, as well as the RXR-receptor subtypes.

[Slide.]

This is a distinction for tazarotene.

Tazarotene is not a pan-agonist. Tazarotene has

specific receptor activation at á, to a much

higher level than at the RARà. There is no

activity at the RXR receptor.

This is important for treating skin

disease because skin disease specifically has a

receptor RAR in karotinocytes.

[Slide.]

It's a locked molecule. And the locked

molecule--if I can try to use the pointer here--the

locked molecule here is due to the triple bond

there. That prevents this molecule from flopping

around and giving non-specific binding.

This receptor selectivity that I've

described here we feel can enhance the therapeutic

effect--can enhance that effect by really

minimizing side effects that are unwanted, and thus

improve the safety profile.

[Slide.]

I'm now going to go on and share with you

the clinical development program.

We've done 12 Phase 1 studies in normal

healthy volunteers; one Phase 2 study in patients

with moderate to very severe plaque psoriasis, and

four Phase 3 studies, patients with moderate to

very severe plaque psoriasis.

[Slide.]

Our clinical Phase 1 studies in normal

healthy volunteers demonstrated that tazarotene

could be given as a single daily dose for all

patients. The dosing is not affected by the

patient's body weight, and not affected by whether

it's taken with or without food.

In in vitro studies, and some clinical

studies, we've demonstrated that there are no

expected drug-drug interactions. Tazarotene is

metabolized by the P450 enzyme system, specifically

CYP2C8 and the FMO.

The metabolism is not altered by alcohol

ingestion. And tazarotene has a very short

half-life of 7 to 12 hours.

[Slide.]

The efficacy data I'm going to share with

you now is based on the Phase 2 dose-ranging trial,

which is known as an 026P study; two Phase 3

pivotal trials, the 048P study, and the 049P. I'll

try to remind you whether it's a pivotal trial, and

what the number is; or two Phase 3 open-label

trials, which are the 050P and 052P.

[Slide.]

Tazarotene in the dose-ranging study--we

determined that tazarotene 4.5 mg per day as a

single dose would be an appropriate dose to take

into our Phase 3 trial. These results were based

on two stages of a dose escalation trial. The

first stage went from zero--or placebo--up to 1.1

mg per day. Then we did dosing escalation cohorts;

a 2.8 mg cohort together; a 4.2 mg cohort; and 6.3

mg.

We showed--and saw in the data, which I'm

going to show you in just a moment--that there was

really no clear dose response in doses ranging from

.4 mg up to 2.8 mg. We did see a nice clinical

response in the 4.2 and 6.3 mg dose groups.

[Slide.]

This is looking at the overal lesional

assessment; looking at patients who achieved a

"mild or less." I think you can see, with the kind

of orange colored bar, or peach colored bar, that

mild disease really--there was not a clear dose

response up to 2.8 mg, but you did see in the 4.2

and the 6.3 dosing groups that there was a nice

response, with at least 80 percent of the patients

achieving that "mild" disease.

Based upon these results, we chose the 4.5

mg dose to go into our Phase 3 trials.

[Slide.]

The Phase 3 trials looked at adult

patients, 21 years of age or older, with stable

plaque psoriasis on at last 10 percent of the their

body surace area in an overall lesional assessment

of at least 2, which was graded as a "moderate."

[Slide.]

So what is an "overall lesional

assessment?" I did mention in our introduction

that this measure was developed by Allergan, in

collaboration with the FDA, back in 1997, for a

cream development program. At that timethe

FDA--the Division asked us to work on developing a

scoring system which was clinically meaningful; a

scoring system which was static--didn't require

physician memory; and one which didn't require

physician's memory, was static, clinically

meaningful, and used all the signs and symptoms of

psoriasis.

So we worked and developed a scoring

system that took all the major signs--plaque

elevation, scaling, and erythema--they were on a

six-point scale, from none to very severe disease.

We used this in our cream development, and then

have used this trial subsequently in our oral

development. Physicians were trained on this

scoring system using a photo-numeric guideline.

[Slide.]

An example of some of the photos from that

guideline are shown on this slide. You can see,

it's a "0"--again--to "5" scoring system, which is

a six-point scale. "None" is no disease; "minimal"

is disease with a little bit of erythema. It

allowed a slight bit of scaling. A little more

scaling and erythema with "mild" disease. You have

a definite plaque at "moderate" disease, and then

the plaque and scaling really increased to "very

severe disease."

[Slide.]

The primary efficacy variable in these

trials were: patients had to enter with at least a

moderate disease; moderate, severe or very severe

disease. And to be considered a clinical success,

those patients had to reach a score of "none" or

"minimal." And that was the primary variable that

we looked at. This is a very stringent criteria.

So patients had to come in with moderate to very

severe, and they needed to be a "none" or "minimal"

to be a clinical success.

[Slide.]

We looked at other measures also. We had

a second co-primary variable, which was looking at

patients who had at least a two-grade change in

their overall lesional assessment. We looked at a

physician global response to treatment. We looked

at body surface area. And then we looked at each

individual sign of psoriasis--erythema, plaque and

scaling--and those were scored on a five-point

scale.

We looked at target lesions to see if

there were different lesions that responded or not

to psoriasis [sic]. We looked at elbows, knees,

scalp and trunk. And,again, we looked at hose in

terms of the specific signs of plaque, erythema and

scaling.

We looked at overall pruritus. We looked

specifically at scalp psoriasis; quality of life

indexes, as well as photographs.

[Slide.]

743 patients were evaluated in these

efficacy trials. 743 got the drug at least 12

weeks. We also did longer-term studies; the 52 and

50P trial. There were 261 patients who got oral

tazarotene at least 24 weeks; 153 for 48 weeks; and

101 patients for 52 weeks.

[Slide.]

In the Phase 3 pivotal trials--this is the

48 and 49P trials--the patients were randomized to

received 4.5 mg of tazarotene per day, versus

placebo, for 12 weeks. The visits were at weeks 1,

2, 4, 8 and 12. There was a post-treatment period

build into this trial--and, actually, all the

trials I'm going to talk about--which was also 12

weeks, and the patients were evaluated at weeks 16,

20 and 24.

[Slide.]

The demographics of the pivotal

trials--this is the 48, 49P trials--were that the

average was around 47 years of age; there were 60

to 80 percent males in the trial. This is

different than what the demographics of the disease

are, but is actually very consistent with a

systemic psoriasis trials.

The mean body surface area was quite high:

approximately 30 percent across all groups. And

the overall lesional assessment was 3.4--so

somewhere between a moderate and severe disease.

[Slide.]

Now, this is showing you the results of

the two pivotal studies, looking at the primary

efficacy variable of "none" or "minimal disease."

So this is this is that very stringent criteria.

The light blue bars on the bottom are the

placebo. The orange bars are the patients treated

with tazarotene. They're the orange lines.

What I think you can note, first off, by

just looking at this, the quick look is that the

two trials--same exact trial, different sites,

different patients--they very closely mimicked each

other. And I think you'll appreciate, as I go

through this data, that all the trials closely

mimicked each other. It makes my job a lot easier

when you don't have one trial that doesn't fit with

the others. All the trials mimicked each other.

So now looking at the orange lines as they

go up, you can see that at week 12, which is the

primary time point, between 15 to 20 percent of the

patients reached this efficacy level of "none" or

"minimal disease."

What's also interesting is that you look

at 16 weeks--which is the post-treatment

period--that's the darker side of the graph--you

can see that more patients in one trial--almost 25

percent of the patients--reached that criteria, and

the other group stayed about the same.

If you look through the post-treatment

period, you can see that the effect was relatively

sustained throughout the 12-week post-treatment

period. These results were statistically

significant as early as eight weeks.

As I've mentioned many times, the criteria

of "none" or "minimal disease" is a very stringent

criteria for success. Does that mean that only 20

percent of the patients improved with the disease?

What I want to show you here is: no,

that's 20 percent of the patients achieved that

stringent criteria, but more patients actually did

respond.

If you look at the two sides of the graph,

there's the tazarotene treated side, and there's

the placebo side. So it's tazarotene, placebo.

Patients entered the trial--predominantly

moderate overall lesional assessment. The yellow

bar are patients with severe psoriasis. The little

red bar at the top are patients with very severe

psoriasis. So this is the tazarotene group, at

baseline.

After 12 weeks of therapy, you can see

that the moderates are certainly decrease. The

"severes" are decreased, and the "very severe"

patients--although not many patients all

improved--and that this response was maintained in

the post-treatment period.

So where did these moderates and severes

go? Well, they go into the mild, none or minimal

caregories. And, again, this graph shows that that

response is somewhat sustained through the

treatment period.

If you look at placebo--well, your purple

and yellow bars, at the quick glance, don't change

much. And, certainly, your "very severe"--is the

little skinny red bar at the top--don't change at

all.

Body surface area--we did measure body

surface area. This is--the overall lesional

assessment is different than the POSI score. The

POSI score is a derived score, which includes body

surface area a part of the derivation of that

score. With our scoring system, the overall

lesional assessment is separate from the body

surface area.

[Slide.]

So here we show body surface area, and the

number of patients who actually had at least a 10

percent decrease. You can see, at week 12, between

30 to 40 percent of the patients had a 10 percent

decrease in their psoriasis. It peaks at weak

16--or four weeks off of treatment--and that effect

is relatively sustained throughout the treatment

period. And, again, is statistically significant

compared to placebo.

[Slide.]

We also looked at the measure of physician

global--response to global improvement. So this is

the physician saying, "How much better did this

patient get?"

What I have here is the dta divided by how

many patients the physicians felt got at least 50

percent better, and 75. The hash marks are the

total height of the bar; shows how many patients

got at least 50 percent better. And you can see at

week 12, at leaste 54 percent of the patients got

at least 50 percent better.

That was relatively sustained in the

post-treatment period, with 43 percent of the

patients getting better. I think you can see it is

statistically significant compared to placebo.

If you use a little more stringent

criteria of how many patients got at least 75

percent better--that's the solid bar--you can see

that at week 12, 30 percent of the patients got at

least 75 percent better, and that was sustained in

the post-treatment phase of 30 percent maintaining

that.

So, again, there's a very stringent

criteria of "none" or "minimal disease," which is a

very high bar--which we had approximately 20

percent of the patients achieve. But the majority

of patients do achieve some response of at least 50

percent or more improvement.

[Slide.]

These are just some clinical photographs

that show how the patients did in this trial. The

patient on the left is at baseline, and then his

response at week 12.

[Slide.]

These are some target lesions--the elbow,

on the top, and the knees on the bottom--at

baseline, and then the response at week 12.

[Slide.]

Another target plaque and elbow at

baseline, and then a really nice response, again,

at week 12.

[Slide.]

We had two long-term studies: the 052P

study and the 050P study. Although these were

primarily safety studies, I think toshow you just

one efficacy graph I think is helpful.

The 052P study was an extension study from

the privotal trial. So the 048, 049P

trial--patients who at 12 weeks either had

worsening disease or stayed the same were allowed

to enroll in an open-label trial.

Ninety-two patients enrolled that had

already been treated with 12 weeks of tazarotene,

versus 220 that had been treated the first 12 weeks

with placebo. They went into the six month trial:

12 weeks treatment with tazarotene for all

patients, and then another 12 week follow up.

So what we see in this group is that you

have a subset of patients who got 24 weeks of

therapy with tazarotene.

In contrast, the open-label study--the 50P

study--was a pure safety study. All patients got

tazarotene, 4.5 mg. They were dose for 52 weeks,

and then had a second 12 weeks post-treatment

response.

[Slide.]

The demographics of this trial were very

similar to the demographics I showed you for the

two pivotal trials. The mean age was, again,

around 47 years of age. The body surface area was

close to 30 percent, and the overall lesional

assessment score was close to 3.4.

[Slide.]

Now, this is a graph showing the primary

efficacy variable of "none" or "minimal" disease.

If you look at the yellow line first, the yellow

line are patients who were treated with placebo and

then enrolled into this open-label trial. So they

should respond like what we showed in the pivotal

trial. And they do.

At week 12, approximatley 20 percent of

those patients had reached the stringent criteria

of none or minimal disease, and that was relatively

maintained across the 12-week post-treatment

period.

What's interesting here--and the reason I

like to show this data--is that the orange line are

patients who didn't respond to the first 12 weeks

of therapy. And they didn't respond, and they had

to enter this trial with moderate or worse disease.

So they could not have improved. Some of those

patients went on to improve and to meet the

stringent criteria of an OLA of "none" or

"minimal;" in fact, apprxoiamtely 15 percent of

those patients did rech that criteria, and then

those patients had a sustaining of the effect in

the post-treatment phase.

So, it does suggest that 12 weeks may not

reach--all the patients may not have their maximal

response in 12 weeks.

[Slide.]

This is looking at the open-lablel study.

And, again, I think these results are very

consistent with what I've already shown you. These

are patients who all got 4.5 mg per day. If you

look at week 24 here, you have really pretty much

the peak efficacy effect. Approximately 20--a

little over 20 percent of the patients reach the

stringent criteria of "none" or "minimal" disease.

And it remains relatively stable throughout the

next 24 weeks of therapy. And the effect, again,

is somewhat sustained and maintained 12 weeks off

of therapy.

[Slide.]

We looked at many secondary measures.

Because of time constraints, I can't share all this

data with you. But we did show at weeks 12 and 24

that the results were statistically significant in

reducation of scaling, erythema and plaque. The

results were statistically significant even in

difficult-to-treat lesions, such as scalp, knees

and elbows. And the results were sustained t

hroughout the post-treatment period in all trials.

[Slide.]

At week 12, nearly 80 percent of the

patients were satisfied with their treatment, and

there wre statistically significant changes in

their quality of life scores. The improvement in

the quality of life using a specific psoriasis

index called a PQOL correlated with improvement of

OLA, and it correlated whether the improvement was

only one grade in OLA or higher.

[Slide.]

Just to summarize the efficacy:

aprpoxiamtely 20 percent of the patients achieved

"none" or "minimal" disease. Approximately 50

percent--or a little over 50 percent--had at least

moderate--or 50 percent or more clearing of their

disease. There was a significant improvement in

plaque elevation, erythema, scaling, and pruritus,

as well as a reducting in BSA.

[Slide.]

There was a maintenance of effect observed

follwoing discontinuation of drug. There was no

tachyphylaxis, and really, the majority of patients

did not have rebound.

A large percentage of the patients were

satisfied with the treatment and had an improvement

in their quality of life.

[Slide.]

Now, I'm going to spend some time going

over the safety data.

I'd first like to emphasize that we have

neaerly 1,700 patients who have been exposed to

tazarotene What I'm focusing on in the safety

presentation this morning, however, are patients

who got at least 4.5 mg, and really focusing on

patients who got 4.5 mg for at least 12 weeks;

that's 690 patients.

There are also patients who got the drug

at least 24 weeks--285; 48 weeks--153; and greater

than 52 weeks--101--greater than or equal to 52

weeks, 101.

[Slide.]

We looked at many measures for safety. We

looked at adverse efents; physical examinations,

vital signs, body weight. We di therapeutic drug

monitoring at selected sites. We did x-rays on the

spinal and ankle ligaments, looking

forcalcifciation or osteophyte formation. That was

done on all patients in all studies.

We did DEXZ scans, looking at bone

densitometry--again, all patients, all studies. We

did ophthalmologic evaluations, and specifically,

we did ERGs, only in the long-term study.

We did audiology evaluations, but focused

only on the long-term, one-year safety study. And

we did neuropsychiatric evaluations on all

patients, all studies.

[Slide.]

Oral tazarotene in the clnical trials was

very well tolerated. WE had a very low drop-out

rate due to adverse events. Less than 5 percent of

patients dropped out due to treatment-related

adverse events in the pivotal trials, or the 048,

049P trials.

More dropped out in the long-term

trials--the six-month trials--6.5 percent. And

almost 15 percent of the patients did drop out in

the open-label one-year study.

[Slide.]

We had very few serious adverse events in

the trial. The adverse events--there were only two

which were deemed to be treatment-related. And I'd

also like to point out that we did have one death

in this trial. The death was secondary to a

mechanical failure of a small aircraft, and not

thought to be related to the drug.

The two serious adverse events thought by

the investigators to be possible due to drug was,

one, for a patient who was hospitalized during the

post-treatment period for pain secondary to severe

ampullary stenosis. The other patient was

hospitalized due to hypertension, and it's

noteworthy that this patient did have a history of

hypertension. Both serious adverse events were

resolved.

[Slide.]

There were four pregnancies--or had been

four pregnancies with oral tazarotene. Only one of

those pregnancies was in the psoriasis trial--this

is the 050P, that's the one-year trial. That

pregnancy occurred eight weeks after the patient

had discontinued drug. And it's notable, because

that pregnancy was a result of non-consensual sex,

and the patient did choose an elective termination

following that.

The other three pregnancies were in the

acne Phase 2 trial, which is the 040P trial. One

of those resulted in elective termination by the

patient; one in a spontaneous termination or

miscarriage; and one was a healthy baby born. That

baby was exposed in utero to approximately 15 days

of drug, and was without any malformations. The

child is now 26 months old.

[Slide.]

Adverse events--we measured adverse events

in all the trials. I'm going to focus first on the

pivotal trial, because it has the placebo control

group. There were adverse events. These were

predominantly mild. The most common was

cheilitis--or chapped lips. It occurred in 65

percent of the patients. The next most common was

headache, and then dry skin; and, less commonly,

arthralgia, myalgia and back pain.

Note here, 2 percent of the patients had

hyperglycemia, versus placebo, but here were no

differences in laboratory values of hyperglycemia

relative to placebo. So these are ones that

investigators said were adverse events.

[Slide.]

What happens when we discontinue

treatment? So, here you have the same data from

the previous slide. And what I've shown you here

are only those which were statistically significant

between placebo and active.

If you see what happens post treatment,

you see that they actually all go down. The

cheilitis went from 65 to 48. It should be noted:

these are all adverse events that occurred at any

time during the post-treatment period. Headache

reduced in the post-treatment period to 4.7; dry

skin reduced' arthralgias, myalgias--showing that

all of these side effects that occurred during the

treatment period were reversing with

discontinuation of drug.

[Slide.]

What is important, I think, in this case

to show you what occurred with adverse events, it's

important with what we know about this class of

compounds, to say what didn't occur with our drug;

what didn't we observe.

We did not observe a difference between

tazarotene and placebo in alopecia. Alopecia is a

very common problem and a common reason for

discontinuing acitretin therapy. We didn't see any

endocrine abnormalities. Endocrine abnormalities

are common with bexarotene. We didn't see

depression, or differences in depression or

psychiatric evaluations between the tazarotene and

placebo groups, in terms of emotional lability or

depression.

We didn't see a difference in elevation of

liver function tests, and we didn't see any changes

in visual or auditory.

[Slide.]

So what happened in the long-term studies.

Here I'm showing you just the open-label data, so

there's no placebo. But you first have those

patients who were in the 052P extension study who

got placebo the first 12 weeks. So they should be

essentially--have the same AE profile as our

pivotal trials. Those who got it at least 24

weeks, and patients who got drug at least 52

weeks--or 52 weeks.

Again, cheilitis--very similar. It's the

most common side effect with this drug, but it is

notably mild. Dry skin--the second; arthralgia,

myalgia, headache--very similar to what we showed

in the previous slide. So what we also wanted to

look for is is there anything new that showed up,

and did they change over time?

I think, if you look here, you see that

arthralgia does appear to incrase with longer

duration of treatment, as does myalgia, and

possibly headache.

[Slide.]

Oh, let me go back one.

[Slide.]

Also, notably, we did see some alopecia

out a year; less than 8 percent--still

significantly less than observed with the other

retinoids, but higher than what I showed you in the

placebo trials.

[Slide.]

Liver function tests--this is an importnt

one to look at, especially considering this class

of drugs. I think these results are very

interesting.

First the ALT--transaminases--they were

higher--now we're looking at 12 weeks, 24 weeks,

and 52 weeks of therapy. They were higher in the

placebo group than any of the treatment groups.

AST--the other transaminase--was

relatively stable between the placebo-control trial

at 12 weeks, but does look possibly like it goes up

at 52 weeks--excuse me, at 24 weeks or 52 weeks.

But I think when you look at that you have to

remember there's no placebo group, and over a year,

at any one time, an individual laboratory value may

spike.

Similar, GGT; LDH we didn't see much;

bilirubin--direct, indirect, total. The only one

that we do see a really change was alkaline

phosphatase, which was elevated relative to the

placebo in the treatment groups at 12 weeks in the

52-week study--up as high as 14 percent.

[Slide.]

We looked at all laboratory values. I

showed you the ones that were statistically

significant. Looking now at all laboratory values,

in the tazarotene versus placebo trial, what else

do we see?

Well, creatinine phophokinase--higher in

the placebo group relative to the tazarotene group

This is also unique. I think if you think about

isotretinoin and elevations in creatinine

phosphokinase are a known problem.

Triglycerides--22.8 percent versus 16

percent. What does that mean? I'm going to go

into that a little more detailed in the next slide.

ALT--worse in the placebo versus tazarotene;

bilirubin worse in the placebo group versus

tazarotene.

So let's look at that triglyceride

elevation, which was statistically significant.

[Slide.]

We looked at the triglycerides in many

different ways. But I think that looking at it up

here, the way I presented it, is instructive. We

looked at patients who were elevated above

250mg/dL, and those who were elevated at greater

than 500mg/dL, assuming that around 500 would at

least be a definite trigger for a patient to either

leave the trial or to go on a second drug.

What you can see is that 30 percent of the

patients in the tazarotene-treated group were at

least--were above 250, versus 23 in the placebo

group. And that was statistically significant.

But if you look at the higher elevations,

you see that they were actually equal between the

two treatment groups, suggesting that there is some

elevation of triglycerides, but those elevations

are modest.

[Slide.]

We looked at the effects on bone. As I

mentioned earlier, we did DEXA scanning, to look at

bone mineral density in the lumbar spine, total

him, and htefemoral neck. And we did x-rays of the

spinal and ankle ligaments for calcification, and

looked for osteophyte formation.

[Slide.]

Focusing first on bone mineral density,

the data demonstrated that after 12 weeks of

treatment there were no differences in the median

percent change in bone mineral density in the spine

or the femur. In the hip, there appeared to be in

increase in bone mineral density, but that increase

was very small, and not--it was statistically

significant, but very unlikely not clinically

meaningful.

In longer-term studies, at 24 weeks and 52

weeks of therapy, we did show that there was a

change in the median percent of the bone mineral

density. But those changes were very small, and

they occurred only in the femoral neck and total

hip, and not in the spine.

We did see gains or losses--and there are

many ways to look at this data, and we can explore

this later this morning--many ways to look at the

data. But if you look at patients who had gains or

losses greater than 5 percent, we saw that in all

three areas. We did see that there were more

individual losses rather than gains in the total

hip and the femoral neck, and that there were no

differences for the spine. So the hip and the

femoral neck seem to be the key areas where there

were any changes at all.

[Slide.]

In this slide I'm showing you the mean

percent change in the bone mineral density, and

that these mean percent changes were very small.

The scoring system is a g/cm

2 And you can see

where the baseline screening visits were on average

there, and then what athey changed.

First of all, note the lumbar spine.

There were no statistically significant changes,

and they were very small. If you look at the total

hip, there were statistically significant changes

at week 24 and 52, but they were changes of .45

percent.

If you look at the femoral neck, there was

a change at 233k 24 of close to 1 percent--.92--and

at week 64, 1.27. But at week 52, nothing. So

these are decreases--small percentage decreases in

the bone mineral density.

[Slide.]

So, to summarize those findings, there

were median percent changse, but they were very

small. They occurred only in the femoral neck and

the total hip, but not the lumbar spine. We think

that these changes really are--could easily be

explained by differences and variance that you

would expect in the normal population.

There are individual gains and losses of

greater than 5 percent, but they are probably also

within the normal variation.

We also have data--our data demonstrates

that these changes are not associated with the

incidence of fractures. They're not associated

with the incidence of osteoporosis. They

don't--there doesn't seem to be an age association,

a gender association, or an association with

medications such as a history of systemic

corticosteroids.

[Slide.]

Now, let's look at the data for

hyperostosis. These were the x-rays. What we've

shown here--we looked a couple things. We looked

at what was the existing hyperostosis; so, really,

the prevalence, and how did that change through

time, as well as looking for changes in increases

in individuals.

Looking here first at the 050P study--this

is the one-year study--and we show that at

baseline--you know, between 50 to 58 percent of

there sites looked at--so, the cervical vertebrae,

the plantar ankle or the dorsal ankle had

pre-existing either ligament calcification or

osteophyte formations. That number seems high,

but it's probaly indicative of both psoriasis

patients, as well as age. AN dit is within the

reported numbers for that population.

But look at what happens at weeks 24 and

52, the numbers really don't change. The cervical

vertebrae go up slightly to 63. The plantar ankle

goes up and down at 54. The dorsal ankle stays

relatively the same.

So we didn't show that when you look

across at what you consider the prevalence for this

population, over one year they did not change.

[Slide.]

So did they actually worson? So they

didn't get more, but did the disease worsen? Here,

I've made a cut-off to show you the dta at greater

than a one-grade change, which would be more

significant than less than or equal to one.

You can see that at weeks 12 and 24, we

didn't see anything. At week 52, there were some

modest increases: the cervical spine, there was a

5.2 percent of the patients had an increase in

calcification or osteophyte formation. In the

plantar ankle there was a 1 percent. So there were

a fwe significant changes---- statistically

significant.

[Slide.]

I'm showing you--reviewed a lot of the

adverse events here. And I think what we feel that

this shows that there were some adverse events that

you would expect to see with a retinoid, but there

were other adverse events that you would expect to

see that we did not see, which points to, really,

our specificity ata the RAR á, .

What we dind't see was hepatotoxicity,

hypercholesterolemia, or changes in thyroid

function, which are RXR or RARà-mediated adverse

events.

What we did see are RAR á and adverse

events. We saw cheilits. We saw some arthralia,

some myalgia, some statistically significant

changes in hyperostosis and bone mineral density,

which may or may not be just due to normal

variation in this population.

[Slide.]

So what are we recommending for

monitoring, based on what I've shown you today?

Allergan is recommending that patients on

oral tazarotene do not need routine laboratory

evaluations, unless they are an at-risk population.

If thte patient has a pre-existing condition which

would result in elevated hypertriglyceridemia, they

would need to be monitored, such as patients with

diabetes or they start the drug with

hyperlipidemia.

We don't recommend routine bone mineral

density or x-rays for our patients. Again, unless

they have a pre-existing condition which put them

at risk, such as arthritis or osteoporosis, or a

propensity for osteoporosis.

We do recommend period monitoring for

patients who have a significant change in symptoms,

or are on this therapy long-term.

[Slide.]

And now I'd like to just turn my talk

towards the Risk Minimization Action Plan for oral

tazarotene.

[Slide.]

Oral tazarotene is a probable human

teratogen, and I'd use that qualifier because

technically speaking, until you see a teratogen, or

you see a malformed fetus, it is "probable." We

feel it's probable due to the class of drugs and

what we know.

Because we're looking at psoriasis__AND I

think this is something that was mentioned in the

introduction--you need to frame this ddrug and this

risk in the frame of what physicians already use

for treatment of psoriasis, and we commonly use two

other drugs which are teratogenic; specially,

methotrexate and acetretin.

Allergan feels that oral tazarotene is a

very viable and important treatment option for

women of childbearing potential. We feel that

because it has a short half life and would be

washed out of the body quickly, that it would be a

useful medication for this population. And because

of that, we are in support of having a Risk

Minimization Action Plan to protect the vulnerable

patients.

[Slide.]

The goals of our program are that no woman

who is pregnant shall be prescribed or dispensed

tazarotene. Women who are taking oral tazarotene

shall not become pregnant.

[Slide.]

Now, there's, I think, a little bit of

possibly confusion in what I'm going to show you

now, and what was proposed originally in our NDA.

And I've got some slight changes to what was in

your briefing package. I apologize for having

changes, but as many of you know, sitting around

this table, this has been an evolving process. And

I think this is a great opportunity for me to thank

Khalyani Bhatt, because she has arranged many, many

discussions between the Derm and Dental division,

the Drug Safety Group and Allergan as we've evolved

with this process. And she's been really terrific

at doing that.

We based our NDA, originally--which was

filed in November of 2003--we based that NDA on the

current isotretinoin program at the time, which was

the SMART program. We updated--and we've been

evolving--since the February 2004 meeting. We

wrote our briefing package with modifications of

the program that was in our NDA to account for the

changes in February. Since even submitting the

briefing package, we've had teleconferences and

actual meetings with the Division, and we've

actually modified it a little bit more. They are

slowly getting us to where they want us to be--is

what I like to say. We've had lots of discussions

with Dr. Wilkin.

So I'm going to highlight where the

differences are in the program.

[Slide.]

First of all, we're now recommending a

mandatory registry for all patients. And this is

something that we're interested in certainly

discussing with the committee. Our original

proposal was just for women of childbearing

potential; a targeted education program for all

patients; a mandatory registration and

certification for physicians and pharmacies; a

verification of all patients qualification at the

pharmacist through an interactive technology-based

system. This is the other difference--they're in

italics. Prior to this, the proposal, I believe,

in your briefing package was only for women of

childbearing potential. A laboratory-based

pregnancy test, which is hard linked between the

pregnancy testing and the drug dispensing.

[Slide.]

Managed access--this is really our main

difference between the recommendations that were

presented by the isotretinoin--for isotretinoin at

the February meeting--and that is a drug supply.

We're recommending for women of childbearing

potential that they get a one-month supply with no

refills. However, for patients who are males, or

women not of childbearing potential, we're

recommending that those patients be allowed up to

two refills. And this really is--the difference

between, say, an acne population and psoriasis.

We're also proposing a pregnancy exposure

registry, which is designed according to the FDA

guidance. It's a proactive study that will follow

up each pregnancy throughout its duration. We will

also look at program effectiveness metrics. We'll

look at pregnancy rate; knowledge, attitude and

behavioral assessments; process compliance

measures; and we'll do root cause analysis.

[Slide.]

As I've mentioned, our program really has

all the essential features of the isotretinoin

program, based on the recommendations of the

committee in February of 2004. But there are, I

think, important things that we need to consider,

and that is that that program is designed for an

acne population. Isotretinoin is prescribed for 20

weeks--you know, give or take some time, depending

on a clinician or a particular patient. And

monthly office visits for six months, while

burdensome, are not overly burdensome.

With psoriasis, which is a chronic

lifelong disease, requiring a patient to come in

every month is burdensome. It's burdensome to the

patient, to the physician, to the health care--you

know, health economics.

The majority of our patients are over 40

years of age--or that's the average age is above 40

for patients on systemic medications. So I think

we really need to consider this, because you could

have the unintended consequence of a physician not

prescribing tazarotene, or a patient not willing to

come in once a month for oral tazarotene and, in

turn, getting a drug which could possibly be less

safe for them, or not having any systemic therapy

when they need it.

[Slide.]

So we have a slightly customized program,

and we'd like to have discussion with this. And we

want to work with the agency to have a program that

protects a vulnerable population. We've very

behind that. But we'd also like a program that is

practical; one that could be implemented by

patients, by health care providers and pharmacists.

And we'd also like to propose, and discuss

today, whether a program for all oral systemic

retinoids--or even all teratogens used in

diseases--should have the same program to avoid

confusion in the marketplace.

Thank you.

I'm now going to turn the podium over to

Dr. Alan Menter to discuss the risk-benefit

assessment of oral tazarotene.

Oral Tazarotene Risk Benefit Assessment

DR. MENTER: Thank you, Dr. Walker.

[Slide.]

It's obvious, when confronted with

clinical research studies safety data that we as

clinicians, and you as a panel, have to make an

assessment as to whether the drug in

question--i.e., oral tazarotene--is worthy of usage

in our psoriasis armamentarium for patients with

moderate to severe psoriasis.

What I'd like to now do in the next six to

seven minutes is review the data and discuss this

issue relating to risk-benefit assessment. And I

really would like to wear my psoriasis advocacy

hat. I am--part of the work I do with the National

Psoriasis Foundation, who is represented here

today, is direct the advocacy group for the medical

advisory board. And my two colleagues, who are

here as consultants today, Dr. Krueger is immediate

past president of the medical advisory board for

the National Psoriasis Foundation, and Dr. Lebwohl

is currently the medical director. And we are

very, very involved in advocacy issues relating to

psoriasis patients and safe treatment for our

psoriasis patients.

So I think we've heard--both from Dr. Cook

and myself, and from Dr. Walker--that we are

dealing with a disease that has physical and

psycho-social implications; that is lifelong; and

we currently have good therapies for psoriasis but

we do have some limitations, and we certainly have

an underserved population of patients who have

moderate to severe psoriasis, as has been

discussed.

[Slide.]

Dermatologists have used retinoids for

many, many years--for decades. And, as I'll

discuss shortly, we are relatively comfortable with

the use of systemic retinoids for the diseases that

they are currently available for. However, I do

believe we now have a unique retinoid. And as has

been discussed by Dr. Walker, because of its unique

receptor selectivity, we believe that some of the

side effect profile that we've come to expect with

retinoids have been minimized, as has been

discussed in the clinical data shown by Dr. Walker.

We know that this drug has significant

improvement, both short-term and long-term, on the

clinical signs and symptoms of psoriasis. And the

vast majority of patients respond. Certainly--as

has been discussed--very few drugs clear patients,

short-term, long-term, on a long-term basis. And

we have a drug here that has a significant in the

vast majority of patients with psoriasis--dealing

with patients with monotherapy, with systemic

retinoids.

It is also important that dealing with a

lifelong disease that we do have a drug that does

not lose efficacy over time. And we now have

one-year data that shows that; that we do not lose

efficacy. And also, when the drugs are stopped for

whatever reason, that there's no risk of the

disease rebounding or producing the erythrodermic

form of psoriasis that Dr. Cook showed, which we

sometimes see with some of our other systemic

medications.

[Slide.]

you've seen multiple clinical slides of

the clinical effect of psoriasis, both from a

physical point of view as well as from an emotional

point of view.

[Slide.]

So, as I've mentioned, we've had retinoids

available for the past 20 years. And the current

retinoids that are available--etretinate is no

longer available. It's superseded by

acitretin--altrans retinoic acid--ARTRA--is a drug

that has recently been made available for the

treatment of promylocytic leukemia. And, of

interest for us dermatologists, that this is used

in conjunction with an old drug that dermatologists

have used for a long time--arsenic--to maintain

patients in control with a rare form of leukemia.

Bexarotene is available for cutaneous T

cell lymphoma. And isotretinoin, as we all know,

for acne.

However, the only drug in this group that

is approved for psoriasis is acitretin. And

because of the concerns of some of the safety

issues to retinoids and other drugs, we do believe

that the improved safety profile of oral tazarotene

does warrant consideration as a new systemic form

of therapy for psoriasis.

[Slide.]

So, what I'd like to do now is just

contrast the oral tazarotene--bring up a few key

points related to oral tazarotene, and how it does

compare with some of the other retinoids that I've

mentioned here now, particularly acitretin, a drug

that we all enjoy using and have used, as I said,

for many, many years, and very comfortable using

acitretin, as we will do in the future, as well.

However, the distinguishing features

relating to oral tazarotene I think are shown here

on the table. I think one of the most significant

features which opens up this drug to females of

childbearing potential is the short half-life of

the drug. The drug, as you can see, has a short

life of seven to 12 hours, and the majority of the

drug is eliminated within five days, contrasting

with the longer half-life of the other retinoids,

particularly, as is shown here, acitretin.

Ethanol--this may not be considered a big

issue, but when confronting patients in the clinic

on a day-to-day basis, and making choices for

therapy with out psoriasis patients, the question

of "can I drink socially?" "Can I have a

drink?"--they cannot with methotrexate. This is

not allowed with methotrexate. The label

specifically precludes social alcohol of any kind

with methotrexate therapy. And because of the

conversion of acitretin to atrentinate, and the fat

storage of this drug, alcohol is also precluded in

females of childbearing potential who utilize this

100

drug. And this may take two to three years for

elimination--hence, the three-year exclusion when

using acitretin, which will not be an issue at all

with oral tazarotene.

I mentioned earlier that as patients age

lipids becomce an issue with patients, and we're

frequently confronted with patients on

lipid-lowering agents as the population ages. And

I think the fact that we do not have this concern,

to a major degree, with acitretin [sic], again, is

I think, a significant improvement over other

retinoids that we currently have available to us.

[Slide.]

Liver toxicity--patients develop hepatitis

C, and we are frequently faced with issues relating

to patients with abnormal liver function tests.

Patients certainly have been shown, in the clinical

studies that Dr. Walker has discussed, to show

minimal changes--short-term or long-term--in liver

function tests between placebo and oral

tazarotene--as compared to one-third of patients

with acitretin.

101

The alopecia issue, I think, is a big

concern for us. Probably, if one had to ask me,

"What is the single most common cause for

discontinuing retinoids?"--other retinoids,

particular acitretin, in psoriasis patients--it's

alopecia, particular females, who certainly do not

enjoy losing their hair. And this has become a big

issue, and we have to tailor--drop the dose of

acitretin to minimize this concern, a mucocutaneous

side-effect concern. And the very fact that we

have minimal alopecia with oral tazarotene, I

belive, again, is a significant distinguishing

feature.

And, finally, the other mucocutaneous side

effects--outside of the cheilitis, the dry skin,

the pruritus and--certainly for the

ophthalmologists in the audience--in the panel--the

dry-eye syndromes and the problems we have with dry

eyes, in consultation with our colleagues in

ophthalmology, is of some concern with most of the

retinoids, but does not appear to be a significant

issue with tazarotene.

102

[Slide.]

And I think the most critical issue that's

obviously facing the panel today, and that has been

discussed in the risk minimization and risk

management plan as outlined by Dr. Walker and her

colleagues from Allergan, is the comprehensive

nature of the plan that has to be brought into

being, in order for us to be able to utilize

retinoid therapy in the future.

So, basically, females currently are

excluded from both methotrexate therapy, and all

females of childbearing potential--and, as I

mentioned earlier--are also excluded from acitretin

therapy. And I do believe this is a significant

proportion of the patient population: young females

of childbearing potential, who no longer have to be

excluded from retinoid therapy because of the

selective nature of this drug.

[Slide.]

In my final few slides, I do strongly

believe, again, as a patient advociate--which is

what i believ we all should be thinking of--is that

103

this drug has shown sustained clinical benefit of a

course of one year. And it's likely to be

continued, as clinical studies continue; that

ongoing therapy with this drug does show further

response. There has been a very high patient

acceptance for this drug, both because of its

clinical responsiveness, and because of its lack,

particularly, of mucocutaneous side effects and

alopecia. And I think the low ddrop-out rate--less

than 15 percent over a one-year period--I believe

is a very strong guide to the patient acceptance of

this drug, and is a very low rate as compared to

many other systemic agents.

[Slide.]

Discussing, again, the female issue

relating to it, we do believe--and I do believe, I

believe Allergan believes, my colleague

believes--that this is a drug that should be made

available for that population group who have

hitherto excluded from therapy; i.e., females of

childbearing potential. And with the risk

minimization action plan proposed, I do believe we

104

as clinicians, and the dermatologists in the

audience, will feel comfortable prescribing a

retinoid drug, bearing in mind that we have a great

deal of experience with the use of retinoids

previously, in psoriasis and other conditions.

And, finally, the point relating to

alcohol consumption, I believe I've touched on

previously.

[Slide.]

So, in summary, oral tazarotene does have

an improved clinical and adverse event profile over

other systemic retinoids.

The issues that concern us--namely, lipid

metabolism, hepatotoxicity, mucocutaneous toxicity,

alopecia--appear to be extremely low, and a

significant improvement over what we currently

have.

And some of the other issues that we need

to consider, obviously, are the bone mineral

density. And I think Dr. Walker has outlined these

issues to us.

[Slide.]

105

So, my final concluding slide, is that

based on what we've heard today, based on the

efficacy data and the safety data, and the risk

minimization action plan that has been proposed,

tazarotene capsules, I do believe--and I believe my

colleagues who are with me today believe--should be

made available, not just to a small select group of

patients, but to all patients who have moderate to

severe psoriasis; that a group of patients who I

believe currently is vastly underserved.

Thank you for your attention.

Discussion of Allergan Presentation

DR. STERN: I'd like to thank the sponsor,

and open to the panel for questions that are

directly relating to information presented by the

sponsor. We'll have lots of time in the afternoon

to discuss the global issues. But points of

clarification, additional data that might be

helpful.

So--Dr. Honein?

DR. HONEIN: Yes, I'd like to know the

denominator for the three pregnancies that occurred

106

in the Phase 2 acne trials, and the one pregnancy

that occurred in the psoriasis trials; and,

specifically, the denominator of reproductive-age

women.

DR. WALKER: Well, I'll be answering

questions from back here, if you'll give me one

minute.

The psoriasis trial, there were 263

patients who were enrolled in that trial, but how

many of those were women--80 percent of them were

males. So, roughly 20 percent. And I can get you

that exact number in a moment.

In the acne trial--that doesn't break down

the gender. We need the gender.

In the acne trial, that was the 049P, that

was a dose-ranging Phase 3 trial, and that was also

fewer women than men. Yes--I'm sorry--I don't have

the exact numbers for that trial, but there were, I

think, 183 subjects in the trial total, and fewer

than half were females, roughly 40 percent. But I

don't have that exact number.

I will point out that there was a

107

risk-management program piloted in the psoriasis

trials, but not in the acne trial.

DR. STERN: Dr. Ringel?

DR. RINGEL: I actually have three

questions.

The first has to do with the makeup of the

target groups for P 048 and 049. On page 31 of

the Allergan handout, we are told that

certain--that patients were on certain concomitant

medications, and certain concomitant medications

were excluded. Likewise, certain conditions were

excluded.

The first question is: I'd like to know

what were those conditions? What medications were

excluded? And what medications were specifically

allowed?

DR. WALKER: Patients could not be on

systemic medications which would affect their

psoriasis. And there were--could you bring up what

the--the full list of exclusion criteria in a

moment--but patients couldn't be on systemic

medications which would affect their psoriasis, and

108

they could not be--which would be, really,

primarily, corticosteroids or methotrexate,

acitretin. So any drug that would affect their

psoriasis, there was a washout period for systemic

retinoids which was longer--there was a washout

period for cyclosporine and for methotrexate in the

trials.

And you asked me another question?

DR. RINGEL: Umm--

DR. WALKER: Oh, and conditions.

DR. RINGEL: Conditions.

DR. WALKER: Their psoriasis could not be

rapidly increasing or decreasing--which does

somewhat eliminate, say, a guttate psoriasis,

because that often is rapidly changing. They

couldn't have a condition which would interfere

with their ability to do x-rays. So if they had,

say, a plate in their hip or their ankle which

would inhibit the ability to read the x-rays for

calcification, they were excluded.

If they had unstable psychiatric disease,

they would be excluded--or any condition that the

109

physician felt would make them unreliable or unable

to participate in the trial, they were excluded.

But that's it.

It was somewhat open-ended that the

physician could exclude people.

DR. RINGEL: How about alcohol-increased

liver function tests at baseline, or triglycerides

at baseline--alcohol abuse?

DR. WALKER: They could use alcohol in the

trial. That wasn't exclusion--

DR. RINGEL: How about if they overused

alcohol?

DR. WALKER: We didn't ask, one way or

another, about overuse. That--you know, what is

"overuse" can also be a little bit of a nebulous

thing. But, no, they were not excluded for alcohol

use, and we didn't take a definite history of

alcohol use.

Slide up, please.

The exclusions will be on the screen

there. I went over most of them.

Umm--I'm sorry, I'm going to have you--I

110

got off track on the alcohol use. You asked me

another question. Specifically, alcohol use

and--oh, triglycerides.

There was an exclusion for triglycerides

greater than 500 mg at baseline. They could have

elevated liver function tests or triglycerides if

the physician felt that they were stable. So, they

were either within normal limits, or the physician

felt that patient was stable to go on drug.

Patients were allowed to have hepatitis C

in the trial. They were allowed to have elevated

triglycerides below 500 and other labs.

For anyone here who's done extensive

psoriasis trials, if you don't allow some wiggle

room around labs, you'll never be able to recruit

patients, because they do have many co-morbid

conditions. They have, often, diabetes and other

things.

So, yes, we did allow that if they were

stable.

DR. RINGEL: And how about topicals? Were

all topicals allowed?

111

DR. WALKER: No. No topicals were allowed

except emollients. So there was no topical

tazarotene, dovinex or corticosteroids allowed.

On an occasional basis--and we do have

that data--some patients used emergency topical

steroids for, say, poison ivy or something like

that, for very short periods of time. And those

protocol deviations were all noted.

DR. RINGEL: The other major question I had

was that I don't understand how you applied OLA to

systemic medication. It makes a lot of sense to me

for a topical medication, because you can take a

target lesion and follow that lesion. But, as any

dermatologist on the panel will confirm, psoriasis

in the different body parts doesn't necessarily

resolve at the same rate. So you could have

wonderful clearing on the body, whereas no clearing

at all on the sacral area.

So I was wondering how--in other words,

when you described how the OLA was applied, which

only takes into account an individual plaque, it

seems to me, did you follow the worst plaque? Did

112

you take an average? Did you--how did you do it?

DR. WALKER: It's a clinical assessment,

and it's a clinical integration. And so,

essentially, the physician looks at the patient and

does--you know, not a numeric average, but an

average--you know, overall average based upon,

really, the worst lesion.

That can work, of course, for and against

you. If you take the worst lesion, it's certainly

harder to have a clinical success. But it is

driven by plaque, and it is an integration of the

entire body.

We did learn in the topicals that actually

there were patients in the topical trials who had

80 percent body surface area, where they applied

their tazarotene. Those were the higher patients.

But the scoring system worked there. We piloted it

in the Phase 2 trial and showed that it did work.

When we separated out plaque, erythema,

scaling, we separated out the target plaques, the

results were essentially the same for all groups,

which demonstrated to us that it did work that way.

113

It is a new measure, but it does have some

clinical relevance, and it did work in the trials.

DR. RINGEL: So if there's someone who had

complete clearing on the trunk, and no clearing at

all on the knees--which isn't unreasonable--someone

would kind of just have a gestalt of what it was

and--

DR. WALKER: No, they would not have

achieved clinical success as we set of "none" or

"minimal" disease. If they had complete clearing

everywhere, but then had severe plaque on the

elbows, they still would have an OLA of "severe."

They had to bring all of the plaques down.

DR. STERN: Could you just tell us what the

"intra" and "inter" rate of reliability was of the

score so we can get some idea, you know, really

what you mean by "gestalt," and how well tested

this is as a metric.

DR. WALKER: We did not do a specific test

to validate the scoring system separately, where we

looked specifically at inter-intra related

reliability. We do divide the scoring system out

114

in the trials by investigative site, and saw no

difference from site to site. But we formally did

not do that.

DR. STERN: And the rationale for not

looking at the test characteristics of this, in

terms of reliability, both inter and intra rate of

reliability for a non-conventional measure, where--

DR. WALKER: When we adopted that measure

for the oral systemic development of tazarotene,

the measure was no longer non-conventional to us,

because we had used it in the topical.

The scoring system has been used for other

systemic drugs. It was used in the Raptiva Phase 3

pivotal trials. It wasn't called an "overall

lesional assessment," but--yes.

VOICE: What did they call it?

DR. WALKER: An OLS--an "overall

lesional--"--I don't remember what the

"s"--"severity" score. So it was used--and

actually the results were very similar in their

trial to using the PASI score, in the sense that it

was effective for their drug also. And I think,

115

Dr. Lebwohl--do you have another question--

DR. LEBWOHL: To address Dr. Ringel's

comment: when we have looked, in previous studies,

at elbows and knees, which you'd expect to respond

more slowly compared to trunk, we didn't find a a

big difference. However, there are certainly areas

that clear more quickly, such as the face or

intratrcianous areas, that routinely clear more

quickly than other body sites. And I think the

word "integrated" was key here.

The assessment tool that was used here was

in response to, basically, dissatisfaction that was

expressed even at the FDA. In fact, I think, Dr.

Stern, the quote from you is: "PASI is passe," was

a quote I believe you said.

A difficulty with assessment tools--and

this one did seem to work. And I have seen a slide

of inter-investigator variation, or within one

investigator, variation. But to address directly

your comment, I can't recall a single patient who

had severe knees and mild trunk after treatment,

unless they started out that way.

116

DR. STERN: Dr. Wilkerson is next.

DR. WALKER: I'm sorry--can I answer Dr.

Honien--I hope I said your name right--her

question. I have the numbers. I'm sorry to

interrupt you.

In the 050P study, there were 83 females

out of 263 total patients. And in the 040P acne

trial, there were 81 femals out of 181 total

patients.

I'm ssorry for interrupting you.

DR. HONEIN: Those were reproductive age,

or all women?

DR. WALKER: All women. In the acne trial,

they were prdominantly of reproductive age. The

breakdown for reproductive age in 050P will be a

smaller subset.

DR. STERN: I'm sorry--Dr. Wilkerson,

please?

DR. WILKERSON: In regards to the alkaline

phosphatase, what was the fractionation of the

abnormal values?

DR. WALKER: We did not fractionate the

117

alkaline phosphatase for patients in this trial.

We did do fractionation in our 040P--our acne

trial--for bone and liver, and saw no differences.

But we did not specifically fractionate the

alkaline phosphatase in the 050P open-label trial.

DR. WILKERSON: So we're making an

assumtion that since liver function tests were not

abnormal, that the abnormal alkaline phosphatase

fraction is bounded?

DR. WALKER: That is the conservative

assumption that we have made.

DR. WILKERSON: I mean, I'm assuming that

you have serum placed aside that's been frozen,

or--

DR. WALKER: We have serum not sepcifically

saved aside for these patients. And I don't know

right now whether we could go back and

sub-fractionate those alkaline phosphatases for all

patients.

DR. WILKERSON: I mean, this is obviously--

outside of the pregnancy--you know, one of the big

concerns about this drug long-term. I mean, you

118

know, you're talking about a fairly inexpensive

laboratory test. I mean, the fact that your GDTs

also rose is sort of suggestive of a hepatic--I saw

those GDTs were up at one point also, which is

somewhat of a crossover.

So I think you have an ambiguous

laboratory situation that needs to be clarififed

for the long term.

DR. WALKER: Slide up, please?

DR. STERN: Dr. Levin?

DR. LEVIN: My questions relate to the

RiskMAP proposal, and I might--if you'd rather

follow this line and I'll come back later on.

DR. STERN: I think in the afternoon we'll

probably be spending a fair amount of time on that,

and we'd like to talk more about data presented in

the presentation.

Dr. Furberg?

DR. FURBERG: I had a similar question. I

was interested in the experience of the--you

piloted the risk management program, and would be

interested in knowing the experience; how was the

119

adherence with mandatory registration of patients,

registration of physicians, pregnancy testing and

so on.

But--if you want to answer now, later--

DR. STERN: I think probably more in the

afternoon for that.

Dr. Katz?

DR. KATZ: I have two questions: one, in

the mechanics of evaluating patients, did the same

investigator evaluate the improvement as evaluated

the side effects? Or was there--

DR. WALKER: It was the same investigator

who did that. They were not required to be

separate.

DR. KATZ: So this would not be put out as

a double-blind evaluation. It was--

DR. WALKER: Well, it was double--

DR. KATZ: --placebo controlled.

DR. WALKER: --yues, placebo controlled,

blinded in the sense that the investigator didn't

know what the patient was on--tazarotene or

placebo.

120

What your saying is would they know

because a patient had chapped lips, say, that they

were on active. And I don't know if you'd want

that now or in the afternoon, but we--

DR. KATZ: No, I just wanted to make sure

that that was made clear, that the investigator

evaluating the side effects also was evaluating the

improvement, negating any double-blind assertion.

DR. WALKER: The same investigator--

DR. KATZ: The other question is: did the

company have any expert in bone metabolism consult

regarding the concern that will come out in the

further discussion.

DR. WALKER: We've had several experts look

at the bone data with us, from statisticians who've

looked at the data normalized across populations.

We've got some experts with us here--

DR. KATZ: So will we be informed of their

considerations? At this meeting today?

DR. WALKER: Umm--well, we have a lot of

the data--I'm not exactly sure--do you mean, will

you be hearing--

121

DR. KATZ: My question is--

DR. WALKER: --from an expert that will--

DR. KATZ: Yes, will we be informed of what

their evaluation of--

DR. WALKER: Yes.

DR. KATZ: Thank you.

DR. STERN: I'd like to close this section

with a few questions of my own, in spite of Ms.

Topper says to me.

[Laughter.]

The first is: you talked about 79 percent

or 80 percent of people indicating--patients

indicating satisfaction. I don't think you

presented the--as I recall from the briefing

document--in fact 53 percent of the placebo people

had similar ones. So it was a 27 percent

difference, not some larger difference.

The second is: you talked about a

correlation between quality of life and the OLA

score, but I didn't hear how much the quality of

life instrument shifts were, and what, in fact,

that correlation was; whether it's an r-square, or

122

whatever measure you used. So I'd be very

intrersted in that.

And I guess the third is one that has to

do--aside from half-life comparisons between

tazarotene and acitretin which were featured in the

final presenter is--we have to remember that,

except for the half-life considerations, that

retinoid side effects are very much dose-related.

And we've only heard about one dose of tazarotene,

tazarotene in terms of efficacy and safety. And

the information on acitretin, I believe, comes from

the literature that deals with doses that are

literally more than an order of magnitude

different--some doses being as low as 10 mg in use,

and other doses being well in excess--up to 150 mg

in use.

So I think making comparisons other than

things related to half-life and the accumulation of

drug in pregnancy for all the other endpoints, it's

very hazardous to do without head-to-head

comparisons--and particularly, in the absence of

knowledge of where is this dose in efficacy or in

123

any other way relative to the safety data from the

competitive drugs.

So if you could--that's a comment, but if

you could answer my other questions.

DR. WALKER: I certainly--slide

please--satisfaction rates--you're correct, there

was a high placebo satisfaction rate. If we look

at patients who were extremely satisfied, very

satisfied or somewhat satisfied, the placebo is the

light blue bar, and the tazarotene-treated group is

the dark blue bar. And the difference

is--although, you know, and you clearly described

what the data was if you take all patients who were

satisfied. This is actually just breaking us down.

You can see they are statistically

significantly different from placebo for "very

satisfied," "somewhat satisfied," and "extremely

satisfied." However the differences aren't 80 or

90 percent--as you mentioned.

If you would put up the PQOl--put that

slide up, please?

[Slide.]

124

This is looking at the PQOL, and

correlating it to improvement--which I did mention,

as you rightly mentioned. If you look at the

placebo--this is looking at the score--and the

placebo group compared to patients whose

PQOLs--this is looking at change--all right?--or

reduction in PQOL score which is an improvement in

the quality of life--placebo had less than a

minus-1 improvement--.84. Patients who had a

one-grade improvement were 1.87; a two-grade

improvement, 2.43; a grade of "none" or "minimal,"

2.96.

So you can see that the patients had an

improvement of their PQOL score with any

improvement, and it was certainly greater as they

went to "none" or "minimal" disease.

DR. STERN: That wasn't quite my question.

My question is: what's the correlation between a

PQOL and an OLA score, rather than does PQOL go in

the same direction. And I think those are--you

know, essentially, a patient who gets an

improvement in their OLA score, are they going to

125

also say life is better in terms of impact of

psoriasis at an inter-individual.

So I'm really looking for some

non-parametric equivalent of an r-square.

DR. WALKER: We did not do that.

DR. STERN: I'd like to then--we're only

five minutes behind--[laughs]--which is pretty

good-

[Laughter.]

--and have us have a 15-minute break, and

we'll start very promptly at 10:20.

[Off the record.]

DR. STERN: We'll now be hearing from the

FDA, with Dr. Yao presenting the FDA's presentation

concerning toxicology studies of tazarotene.

FDA Presentation

Toxicology Studies of Tazarotene

DR. YAO: Good morning. I'm Jiaqin Yao

from FDA.

Today, I would like to talk about

toxicology study of tazarotene.

Tazarotene was previously developed for

126

topical use. This NDA submission is for oral

administration. So the sponsor has tested the

tazarotene by oral administration in rats, dogs and

monkeys for up to one year.

[Slide.]

The study showed that oral tazarotene has

a typical toxicity of other retinoids. The maximum

system exposure (AUC) to tazarotenic acid, which is

the major metabolite of tazarotene is almost as

similar of weight of small or the human systemic

exposure, which is in single dose 4.5 mg.

The primary target organ or system

included bone, liver, kidney, heart, thymus and

skin.

Tazarotene was tested to show that no

genotoxic effect, and in carcinogenic studies in

rates and mice showed that there's no significant

increase in tumor frequencies.

In the next couple slides I will focus on

the reproductive toxicity induced by tazarotene.

[Slide.]

Study has been done in male and female

127

rates at the dose 1mg/kg by oral. So AUC--that

means system exposure is three times over human

exposure AUC which is 4.5 mg/day. So that means

the step exposure is only about 30 percent of

human.

They show that the mating performance and

fertility is no change at this dose.

[Slide.]

As the dose incrases to 3 mg/kg per day,

the AUC is 0.7 times our human AUC by oral

tazarotene at 4.5 mg, it shows that there is a

sperm count and density decrease.

Studying the female rates at a dose of

2mg/kg per day by oral, AUC is 0.6 times human AUC,

the mating performance and fertility does not

change. But we see some development toxicity.

[Slide.]

Studiies have also been done in toxicity

in embryonic development--developmental toxicity.

The study has been show in female rats at 0.25mg/kg

by oral. The AUC is 0.2 times human AUC. We saw

some development delays, teratogenic effects, and

128

post-implantation loss.

[Slide.]

Another study in female rabbit, at 0.2

mg/kg per day by oral, the AUC is 4.7 times human

AUC, we see similar factor in those studies, just

using female rabbits.

Since this drug has been developed for

topoical studies, the sponsor has also done some

topical studies in the femal rats and female

rabbits, at a dose 0.25 mg/kg, theAUC is 0.2. We

saw that some fetal body weight decrease and

skeletal ossification decreased.

In the studies by topical, in female

rabbines, the dose is 0.25 mg/kg, AUC is 2.3 times

human, malformations are found in those studies.

[Slide.]

Another study is on the toxicology studies

in prenatal and postnatal development. In female

rabbis, they have done two studies. The first

study is 1mg/kg by oral, but the AUC is 1.1 human

AUC. We saw developmental bahavior delays.

Another sutdy used the same dose--1 mg/kg

129

by oral, the AUC is 0.4 times human AUC, we see

developmental delays.

[Slide.]

Another thing I would emphasize a little

bit about is this drug on the male reperoductive

system. As I mentioned before, at the dose of 1

mg/kg per day in male rats, AUC is 0.3, the mating

performance and the fertility does not change.

However, at a dose of 3mg/kg per day, the AUC is

0.7, we see some sperm count and density decrease

in male rates.

In general toxicology studies, the sponsor

has done one study on male dog, which is for nine

months. At 1mg/kg per day, by oral, the AUC is 1.9

times human AUC, we see testicular changes. As the

dose increases to 3 mg/kg per day by oral, AUC is

4.1 times human, the change is more than 1 mg/kg

dose.

[Slide.]

Based on the sponsosr proposal, the

maximum recommended human dose for tazarotene by

oral is 0.075/kg/day. For other drugs, such as

130

acitretin, it's 0.83mg/kg, and isotretinoin is

2.0mg/kg. So the daily dose is less than other

drugs.

However, when I checked the literature, we

find that compared with animal studies in rabbits

and rats, the lowest teratogenic dose unit is

mg/kg/day, is 0.2mg/kg/day in the rabbits. In

rats, it's 0.25 or 1, because the sponsor did two

studies. One is 0.25, one is 1.

Compared with acitretin, the lowest

teratogenic dose in rabbits is 10mg/kg, and for

rats is 150mg/kg. And I also compare with other

retinoids, we find that this drug product is--seems

is more important as teratogenic in rabbits, and

the rats.

Another thing we see those data, we can

see that the human is like most sensitive species

in teratogenic effect induced by those retinoids.

[Slide.]

So the conclusions from those data, we can

see hman may be the most sensitive species for

teratogenicity of the retinoids. So when we

131

consider about the drug dose, tazarotene is more

potent teratogen than other retinoids in rats and

rabbits, based on the mg/kg/day basis. And

tazarotene is a probably human tazarotene.

The next speaker will be clinical

pharmacology review by Dr. Ghosh.

Thank you.

Clinical Pharmacology and Biopharmaceutics

DR. GHOSH: Good morning. This is Tapash

Ghosh, from the Office of Clinical Pharmacology and

Biopharmaceutics, FDA.

My presentation will be to describe the

clinical pharmacology and biopharmaceutics aspects

of oral tazarotene.

[Slide.]

The focus of my presentation will be

pharmacokinetics of tazarotene and tazarotenic acid

in plasm; potential for drug-drug interaction; and

tazarotenic acid in semen.

[Slide.]

Tazarotene or tazarotenic acid have

multiple effects on keratinocyte differentiation

132

and proliferation, as well as on inflammatory

processes, which may conttribute to the

pathogenesis of psoriasis. Some of them include:

blocking of induction of epidermal ornithine

decarboxylase activity; suppression of expression

of MRP8, which is a marker of inflammation present

in the epidermis of subjects with psoriasis; and

inhibition of cornified envelope formation, whose

build-up is an element of the psoriatic scale

expression.

[Slide.]

This is a schematic of how tazarotene

works in our body. Once tazarotene is in the

systemic circulation, it undergoes fairly rapid

conversion to its active metabolite, which is the

tazarotenic acid, with the help of abundance of

acerisus enzyme present in our biological system.

Then the tazarotenic acid undergoes further

oxidation to tazarotenic acid sulfoxide, which,

maybe with the presence of the CYPS and FMO enzymes

present in the system. And then tazarotenic acid

sulfoxide may undergo further oxidation to

133

tazarotenic acid sulfonates.

Some portion of the tazarotene also

undergoes oxidation to tazarotene sulfoxide.

[Slide.]

Tazarotene is orally absorbed, as

approximately 90 percent of the oral level

tazarotene was recovered in pheresis and in urine

as primarily tazarotenic acid and its metabolites.

Tazarotene exposure increases fairly in a

dose-proportional manner, following oral tazarotene

from 3 mg to 6.3 mg.

[Slide.]

Tazarotenic acid is highly bound to plasma

proteins, with an unbound fraction of less than 1

percent.

Following intravenous dose, the apparent

volume of distribution of tazarotene and

tazarotenic acid was 3.55L/kg, and 0.75L/kg,

respectively.

[Slide.]

As I already described, in humans,

tazarotene is hydrolyzed quickly and extensively to

134

tazarotenic acid, which the primary active moiety

in the systemic circulation.

In vitro human metabolism studies

demonstrated that tazarotenic acid is metabloized

to inactive sulfoxide metabolite via CYP and/or FMO

enzymes in the liver.

[Slide.]

Fecal elimination is the predominant

elimination pathway, with 46.9 percent of the

administered oral dose eliminated in the feces as

tazarotenic acid. Approximately 19.2 percent of

the dose was excreted in the urine as inactive

sulfoxide metabolites of tazarotenic acid.

[Slide.]

Following IV administration, tazarotene

was measurable in plasma, and was eliminated from

the body with a mean terminal half-life of 6.2

hours.

Following IV administration, plalsma

tazarotenic acid concentration declined

bi-exponentially, with a mean terminal half-life of

13.8 hours.

135

[Slide.]

Following IV administration, the systemic

clearance of tazarotene was 2.23L/hour/kg.

Systemic exposure of the active

metabolite, tazarotenic acid, was 21.4 times that

of the parent compound.

[Slide.]

This is a profile of how tazarotenic acid

gets excreted from the system. An as I have

mentioned, this is a normal plot and this is the

semi-log plot, just to show the bi-exponential

decline of the tazarotenic acid.

The profiles are from day seven and day

13, which shows the profiles on those two days are

superimposable.

[Slide.]

As the previous speakers have already

mentioned, that tazarotene right now is already

approved in topical dosage forms. This table shows

the comparison of systemic exposure of tazarotenic

acid from different topical formulations.

Without going through each and every

136

formulation, I want you to concentrate on the last

two rows, where .1 percent gel was compared with

the 4.5 mg proposed capsule formulation. Here, if

we compare the systemic exposure in terms of AUC,

it is about one-fourth, and in terms of exposure of

Cmax, the topical exposure is about one-eighth,

compared to the oral exposure.

[Slide.]

However, the data obtained from topiocal

gel was during maximal usage condition, which may

not reflect the usual usage condition.

[Slide.]

In terms of drug-drug interaction, there

was no interaction found between tazarotenic acid

and Ortho-Novum 1/35, and between tazarotenic acid

and Ortho-Tri-Cyclen when given as oral tazarotene

dose of 6 mg.

However, based on the data, the potential

of drug-drug interactions involving CYP450s,

especially 2C8 and 2B6, may need to be further

explored.

[Slide.]

137

Now I'll be discussing the tazarotenic

acid in semen. Following once-dailing dosing of

tazarotene 4.5 capsules for two weeks in healthy

male subjects, more than 79 percent of semen

samples had tazarotenic acid concentration above

lower limit of quantitation, which is .1 ng/ml.

Median semen to plasma tazarotenic acid

concentration ratio at each predefined time point

from semen samples over the 72-hour period was

approximately 1 or less, except at six and nine

hours, where the ratio was greater than 1, as

dscribed in this following figure--

[Slide.]

--where the ratio, semen to plasma

tazarotenic acid concentration was profiled--again,

these are the sampling points. And, as we see, for

most of the time points, this is the body

presence--1--ratio 1. Most of the time points had

either 1 or less than 1, but at six and nine hours,

the ratio semen to plasma was greater than 1.

[Slide.]

The highest individual semen to plasma

138

tazarotenic acid concentration ratio 2as 2.8, and

occurred between 9 and 12 hours post dose.

The highest plus-or-minus standard

deviation, tazarotenic acid concentration observed

in semen was 44.4, + 22.2, observed in 16 subjects,

occurred at three hours post dose.

The highest individual tazarotenic acid

concentration observed in semen was 83.1ng/ml,

occured at three hours post dose, in comparison to

1.61ng/ml pleak plasma level from the same study.

[Slide.]

So, therefore, under worst case scenario,

assuming an ejaculate volume of 10 ml, the amoung

of drug transferred in semen would be 831 ng, which

is about 1/5,000th of a single 4.5 mg capsule dose.

The no-effect limit for teratogenicity of

tazarotene or tazarotenic acid is unknown in

humans.

[Slide.]

Fertilized egg may remain exposed to

tazarotenic acid in the semen following repeated

sexual encounters.

139

Finally, the risk to a fetus, if any,

while a male patient is taking the drug, or after

it is discontinued, cannot be ruled out.

This is the end of my presentation.

Thank you.

So now Dr. Cook is coming for the next

presentation

Clinical Safety

DR. COOK: Good morning again.

I've come to you this time to speak on the

clinical safety, from the FDA perspective, of oral

tazarotene as presented in NDA 21701. Some of the

presentation will be a repeat of Dr. Walker's

presentation earlier this monrning, but some of it

might be a little bit different.

[Slide.]

The safety data base, as you know, is

derived from the following four trials: two Phase 3

double-blind placebo controlled trials; and two

open-label Phase 3 trials.

[Slide.]

The duration of the trials were 12 weeks

140

of treatment in the double-blind placebo controlled

trial, with a 12-week follow-up; and there were two

open-label trials--as described earlier--one 12

weeks treatment with 12-week follow-up, and a

52-week trial with a 12-week follow-up.

[Slide.]

There wre 987 patients treated with

tazarotene, and 383 treated with placebo.

Tazarotene patients were treated with 4.5 mg once

daily numbered 831. There wre 640 patients, or 77

percent, treated for greater than or equal to 12

weeks; 31.4 percent wree treated for greater than

or equal to 24 weeks; 18.4 percent for grater than

or equal to 48 weeks; and 12.2 percent were treated

for 52 weeks.

[Slide.]

Discontinuations accounted for about 54

percent of the patients who discontinued from the

trials, either because of lack of efficacy or

adverse events. This is in the placebo-controlled

trials. And the discontinuations secondary to

adverse events in the placebo-controlled trials, it

141

was 3.4 percent.

In the short-term open-label trial, where

patients were taking a second course, versus those

patients who were only taking their first course of

tazarotene, discontinuations were 6.5 percent for

the second-course patients, and 3.2 percent for the

first-course patients. So there was a slightly

discontinuation rate for those patients who were

taking their second course of tazarotene. And, as

Dr. Walker mentioned earlier, there was a higher

incidence of discontinuation in the long-term,

open-label trial.

[Slide.]

Adverse events that led to

discontinuations in the long-term trial that

occurred for more than one patient included

arthralgia, myalgia, arthritis, back pain,

alopecia, dermatitis, joint disorder. There were

three patients who discontinued for abnormal liver

function tests; two for cheilitis, asthenia; two

for depression; and two for emotional lability.

[Slide.]

142

In the pivotal trials, overall, tazarotene

group had more adverse events than in the placebo

group: 90.2 percent versus 74.6 percent, and this

was statistically significant--although I must

mention that the trials were not actually powered

for safety.

And the significant adverse events that

are common to oral retinoids--mentioned

earlier--included cheilitis, dry skin, headache,

arthralgia. And this shows you the percentage of

patients who are on tazarotene--first--experienced

the event, versus those on placebo.

[Slide.]

Other significant adverse events in the

pivotal trial were myalgia, joint disorder, back

pain, nasal dryness, foot pain, rash and

dermatitis.

[Slide.]

Metabolic and Endocrine adverse events

that occurred in the trials: hypertriglyceridemia

did occur, and there was a significant difference

between tazarotene and placebo. And I will show

143

you a slide concerning that.

There was also one incident of

hypertriglyceridemia leading to pancreatitis, and

this occurred in one patient in the

placebo-controlled trials. The patient did come

into the trial with severe hypertriglyceridemia,

having a baseline value of about 6 mmol/L. The

data was presented in millimoles per liter. And

I'll talk about the conversion later--and then his

triglycerides continued to rise to about 13 mmol/L,

and then the patient was taken off the drug. And

about a week later, his triglycerides about 33

mmol/L and 62 mmol/L, and he was worked up--had

ERCP, and was found to have pancreatitis, probably

due to drug product.

The applicant found hyperglycemia, a

statistical difference.

There were also four cases of

hypothyroidism diagnosed; three in the

placebo-controlled trials, one in the long-term

trial, and all the patients were on tazarotene.

And I must say that even though when you looked at

144

TSH values in the placebo-controlled trials, you

really didn't see a difference between tazarotene

and placebo, as there were 2.9 percent of patients

who had elevated TSH on tazarotene, and there were

2.5 percent of patients who had elevated TSH on

placebo, yet the investigators did diagnose three

patients with hypothyroidism.

When I went back and looked at the case

report forms to see was there some other criterion

that had been used to diagnose these patients

outside of an elevated TSH, there had not been.

Elevated LFTs leading to

discontinuation--there was one patient in the

placebo-controlled trials, and this was a patient

who also had an elevated TSH and had been diagnosed

with hypothyroidism.

[Slide.]

Now, here is a table on elevated serum

triglycerides, and these are all patients in the

placebo-controlled trials--all comers. And what I

did was look through the line listings, and took

any patient who had a serum triglyceride that was

145

2.3mmol/L or greater, and 2.3 millimoles is about

200 mg/dL. And I base that on the fact that most

physicians would have some type of therapeutic

intervention once your triglycerides start to rise

about 200--even though we all know that even as

high as 130 would be considered mildly elevated,

but not necessitate treatment. And moderate--2.8

to 5.6 millimoles which--it's about 250 to 500

mg/dL, and greater than 5.61 is anything higher

than that.

And just for your information, if you're

greater than 11 mmol/L, then that's about greater

than 1,000 mg/dL. And there were maybe a couple of

patients who got up as high as 750 mg/dL.

And, at any rate, there are about 45.2

percent of patients in the 048 trial, and 42.3

percent in the 050--049--trial, that had elevated

serum triglycerides on at least two occasions

during treatment. And compared to placebo, this

was found to be statistically significant.

[Slide.]

In the post-treatment period for the

146

placebo-controlled trials, only cheilitis remained

a statistically significant event. Skin and

appendages as a whole was also considered

statistically significant, but that was driven

primarily by dry skin, which is not something

unexpected from an oral retinoid. And I will say

that the serum triglycerides returned to acceptable

range in about 55 percent of patients followed in

the post-treatment period; that is, those patients

who I just previously talked about who had elevated

triglycerides, if they were followed in the

post-treatment period, 55 percent of them did

return to an acceptable range.

[Slide.]

In the short-term open-label trial, there

were a few significant adverse events in patients

who were taking their second course of oral

tazarotene, compared to those taking their first

course; and that included arthralgias, back pain

and alopecia. And, again, the first

percentage--like 33.7 percent--is for the patients

taking their second course of tazarotene,

147

versus--for example, in the arthralgias--the 14.1

percent of patients who had arthralgia taking their

first course of tazarotene. And alopecia again

emerged as a new adverse event.

And significant laboratory abnormalities

in the short-term trial for patients who were

taking the second course of tazarotene included

elevated serum triglycerides, as in the

placebo-controlled trial, and also now elevated

alkaline phosphatase levels.

[Slide.]

In the long-term open safety trial, one or

more adverse events were reported for 98.9 percent

of the patients. And adverse events that occurred

for greater than 5 percent of the patients, again,

are cheilitis, arthralgia, myalgia, infection dry

skin--

[Slide.]

--back pain, headache, asthenia, pruritus,

foot pain, alopecia, leg pain and arthritis--

[Slide.]

along with paresthesia, flu syndrome,

148

nausea, joint disorder, insomnia, rhinitis and

bronchitis. And most of these events were reported

as mild in severity.

[Slide.]

Laboratory adverse events that occurred in

the long-term safety trial--which, again, is the

52-week trial--included hypertriglyceridemia,

again, at 41.1 percent, which sort of paralleled

the placebo-controlled trial. And, again, these

are patients who had elevation at least on two time

points during treatment, and their elevations were

greater than 200mg/dL; abnormal liver function

tests in 22.9 percent; elevated CPK in 16.3

percent; elevated alk-phos, 13.7 percent; and

isolated elevated SGPT and SGOT--because some

patients just had one or the other; abnormalities

in white blood cell counts, 5.7; and elevated GSH,

5.3 percent.

[Slide.]

And out of all of that, the higher

elevation--the alkaline phosphatase is something to

focus on a little bit. There was a higher elevation

149

compared to placebo-controlled trials in the

long-term study: 13.7 percent of patients versus

3.4 percent of patients in the placebo-controlled

trials, suggesting that the longer you're on the

drug, that there's some effect there to alkaline

phosphatase.

And, again, it remained elevated at the

end of the post-treatment period, which was a

12-week post treatment period, in 69.4 percent of

those patients who were abnormal.

Only 3.7 percent--or 2 out of 54 patients

who had elevated LFTs had abnormalities in their

liver function tests at the end of the

post-treatment period, suggesting that there may be

a bone origin for the elevation of the alkaline

phosphatase values.

[Slide.]

And this kind of takes us to the effects

that were found on bone metabolism in the clinical

trials. And I should preface this by saying we had

a consultant at the agency look at the data. And

there is a high drop-out rate, and missing data, in

150

the long-term trials. And so it's hard to make a

definitive conclusion. And more studies and a more

long-term look at its effect--at tazarotene's

effect over long term might really be necessary.

But what was found was that there was a

mean bone mineral density decrease over time for

the entire set of patients, with some having

decreases close to 30 percent. And over 10 percent

had significant decreases of greater than 5

percent. And four patients on tazarotene, and one

on placebo, had bone marrow density decreases

greater than 5 percent in the placebo-controlled

trials. And this occurred in men--all of them were

men, in the range of 40 years old to 69 years of

age. And one patient also had a decrease of 50

percent in bone marrow density, and that's still

being investigated.

But we feel that this might be significant because

men, who composed the bulk of the study--and there

were 68 percent of men in the long-term trial, and

over two-thirds of men in the placebo-controlled

trials--those patients usually lose bone mineral

151

density at about 0.5 to 0.75 percent per year, and

women usually lose about 1.5 to 2 percent per year.

That's a lose of either 1 to 3 percent over a 36

weeks supplies a greater than normal bone loss over

a year. And so, over a period of five to 10 years

of treatment, this really could be significant.

[Slide.]

In the long-term study, 5.3 percent of

patients had significant changes in calcification

and/or osteophyte scores of the cervical spine; 26

percent had worsening changes in hyperostosis and

ligament calcification with each vertebrae of the

cervical spine; and there was a significant

increase in ankle ligament osteophyte formation at

weeks 52 and 64.

[Slide.]

And this correlates to the musculoskeletal

adverse events that also increased in the long term

trial. And in the post-treatment period these

adverse events remained significant; arthralgia, at

19.7 percent; back pain, 17.8 percent; myalgia,

15.8 percent; and arthritis at 6.6 percent.

152

There were also six moderate fractures

occurring during the trials in patients on

tazarotene that were reported as "without known

cause," and we're still having that investigated.

[Slide.]

And other adverse events that occurred in

the post-treatment period of the oral tazarotene

trial are as expected with oral retinoids:

cheilitis, dry skin, asthenia and pruritus.

[Slide.]

Discontinuations due to neuropsychiatric

events--in the placebo-controlled trials, due to

emotional lability, there were three patients on

tazarotene and three patients on placebo who

discontinued. For depression, there was one

patient on tazarotene and none on placebo.

And in the open-label trials, two patients

discontinued for depression, five patients for

emotional lability, and one patient discontinued

secondary to a paranoid reaction.

[Slide.]

The conclusions that we can make from this

153

for neuropsychiatric events is that there is no

difference between tazarotene and placebo in the

controlled trials, for neuropsychiatric events.

However, due to the limitations of the metrics

employed and the statistical power, an association

cannot be ruled out, given the existing concerns

about such effects from other retinoids.

[Slide.]

And ophthalmology--we looked at--used

several metrics to looked for ophthalmologic

effects secondary to tazarotene. And those

employed were visual acuity, biomicroscopy and

opthalmoscopy, and no signal was detected.

[Slide.]

As stated earlier by Dr. Walker, there

were four pregnancies in the trials for oral

tazarotene; one in the psoriasis trial, and three

in the acne trials. There were two elective

abortions, one spontaneous abortion, and one term

delivery at 38 weeks. And, as she said, they have

later follow-up that, so far, this child appears

normal.

154

[Slide.]

Now I'm just going to switch over to the

efficacy portion, and just give you a brief

description of the trials--you've already heard

about this earlier today.

There were two Phase 3 efficacy trials,

which were identical in design. They were

multicentered, randomized, evidence-based and

placebo-controlled. The patients took either

tazarotene as a 4.5 capsule or its placebo once a

day for 12 weeks. And there was a 12-week

post-treatment follow-up.

[Slide.]

The key inclusion criteria included that

the patients had to be age 21 years or older. They

had to have a severity score of greater than or

equal to 3, which was "moderate" on the Overall

Lesional Scale.

They also had to a minimum surface area

involvement of 10 percent.

[Slide.]

Key exclusion criteria included:

155

spontaneously improving or rapidly deteriorating

plaque psoriasis; any patient with a previous

fracture, anomaly, or artifact of the ankles or

cervical spine; use of systemic retinoids within

eight weeks prior to study entry; use of systemic

medications known to affect bone within 12 months

prior to study entry.

[Slide.]

Also, patients with suicidal ideation were

excluded; females of childbearing potential who

were unable or unwilling to use two birth control

methods at the same time during the 28 days prior

to the week-zero visit and during the treatment and

post-treatment periods of the study; and also any

male who was unwilling to wear a condom when having

sexual intercourse with a female of childbearing

potential during the study was also excluded.

[Slide.]

The efficacy variables, as determined by

the FDA--and what we based our efficacy assessment

on--included: one primary efficacy variable, and

that was the Overall Lesional Assessment Score, and

156

this varied from "0" with no disease, to "5" with

very severe disease.

The secondary efficacy variables, which

were supportive of the primary, included the

clinical signs of psoriatic lesions, the erythema,

scale and plaque elevation, and overall global

response.

[Slide.]

Treatment success for the efficacy in

pivotal trials was defined as success being a score

of 0 or 1--meaning "none" or "minimal disease" on

the OLA scale at week 12--at end of treatment.

And now Dr. Shiowjen Lee is going to speak

to you about the efficacy results.

Efficacy-Biostatistical Analysis of Pivotal Studies

DR. LEE: Good morning. In the next 30

minutes I will be presenting you the biostatistical

analysis of pivotal studies.

[Slide.]

The efficacy evaluation of the two pivotal

studies included the following efficacy endpoints:

primary efficacy endpoint--treatment success--which

157

is defined as percentage of patients with Overall

Lesional Assessment--abbreviated as OLA--score of 0

or 1 at week 12, which has the disease of "none" or

"minimal."

The secondary efficacy endpoint included

clinical signs and symptoms, and overall global

response. The overall global response measures the

overall improvement from baseline about the disease

status. And Dr. Denise Cook will have comments

about this efficacy endpoint later.

And other efficacy end point, we are

particularly interested in the scalp and nail

psoriasis.

In this presentation, I will focus on the

primary efficacy endpoint and the scalp and nail

psoriasis end point.

[Slide.]

Presentation of the efficacy findings in

two pivotal studies are organized by the following.

First, I will present you the overall efficacy

findings of oral tazarotene versus placebo, and

next I will present to you the subgroup efficacy

158

results, in particular, by gender and by baseline

OLA severity score. Next, I will show you the

short-term efficacy of oral tazarotene in treating

the scalp and the nail psoriasis. And, next, I

will give you the relapse rate--it's referred to

the short-term relapse rate for the two pivotal

studies.

[Slide.]

This table gives you the baseline

demographics for the two studies; in particular, I

listed gender and OLA score.

For patients treated with tazarotene in

the first study, we enrolled 166 patents. And

among those 166 patients, 80 percent of them male,

and 20 percent female.

In the first study, enrolled 171 patients

in placebo group. And in the placebo group the

patients were 72 percent male and 28 percent

female.

For the second studies, the study

enrollment enrolled 182 patients in tazarotene, and

187 patients in placebo. For patients assigned

159

with tazarotene treatment, 65 percent male, and 35

percent female; and for placebo groups, 74 percent

male, and 26 percent female.

And I want to point out here, in the two

pivotal studies, the male patients accounted for

over two-thirds of study enrollment.

With respect to the baseline OLA score,

most patients had a moderate disease severity,

where they can enter the study. For example, in

the first study, patients with tazarotene group, we

have 60 percent patients had a moderate disease

severity. And for the second studies, there was 66

percent patients had a moderate disease severity.

And I want to point out here, in the first

study there were total of five patients in the very

sever OLA score at baseline. And two of them

treated with tazarotene. And in the second study

we have total 10 patients in the second study, and

four of them treated with tazarotene. And the

total patients with very severe disease severity at

baseline was 15 patients, and accounted for about

only 2 percent of patient enrollment.

160

[Slide.]

This table gives you the efficacy results

about the OLA score and the treatment success at

week 12.

For the first study, the treatment success

rate were 15.7 percent for tazarotene group, and

3.5 percent for placebo group. And for the second

study, the success rates were 18.7 percent for

tazarotene group, and 4.8 percent for placebo

group.

There are two--I would like to make a

point here. The first point is the treatment

success for both studies, they are under 20

percent. And, also, the most patients with

treatment success, they had a score, primary would

be 1. For example, in the first study, patients

treated with tazarotene, we have 22 out of 26

patients had a score of 1. For the second study,

we have 27 out of 34 patients had a score of 1.

[Slide.]

This slide gives you the subgroup results

of treatment success for the first study, primarily

161

by gender and by baseline OLA score. Recall, the

overall success rates were 15.7 percent for

tazarotene, and 3.5 percent for placebo.

For gender, female patients generally had

higher success rate than males. And here we have,

in tazarotene group, female success rate was 26

percent, and the male success rate was 13 percent.

With respect to the baseline OLA score,

generally the success rate decreases as the

baseline disease severity increases. For example

in tazarotene group we have 19 percent success rate

for patients with moderate disease severity, to 11

percent with severe disease severity at baseline,

and to 0 percent for patients with very severe

disease status at baseline.

[Slide.]

And this slide gives you the subgroup

results of treatment study. The overall success

rate for the second studies were 18.7 percent for

tazarotene, and 4.8 percent for placebo. And here,

again, are listed by gender and by baseline OLA

score. For female subjects, again, higher success

162

rate than male subjects. And here, I want to point

out, the treatment difference for female patients

in this study is about 9 percent. This is 24

percent minus 15 percent in 9 percent. And for

male subjects we had 16 percent in tazarotene

group--success rate--and the placebo group 1

percent. The treatment difference here was 15

percent.

And one might question about the efficacy

results driven by male subjects, because it has

larger size of treatment difference. Because,

again, over two-thirds of patient enrollment will

be males. We've done some sensitivity analysis for

this, and it's shown the tazarotene group still

superior to placebo, however, the bigger size

difference does have an impact of the significance

level of superiority.

With respect to the baseline OLA score,

disease severity increases then the success rate

decreases; from 20 percent for patients with

moderate disease, to the last category--very severe

disease--and with 0 percent success rate.

163

And I want to point out here, even

patients from the previous study, and for this

second study, none of the patients in tazarotene

group achieved treatment success for patients with

very severe disease status.

[Slide.]

The next I'm going to discuss with you

about the scalp and nail psoriasis. The evaluation

of scalp and nail psoriasis is based on short-term

efficacy of tazarotene treatment. And the severity

of each scalp, fingernail, toenail psoriasis was

evaluated based on a 5-point scale; 0 was "no

disease," and score of 4 was "very severe" disease.

[Slide.]

The efficacy results presents you--the

next slide--is based on the percentage of patients

having severity score of 0 at week 12.

This table gives you the results about the

scalp and nail psoriasis, with respect to the

patients with severity score of 0 at week 12.

As you can see, for the two studies, with

respect to the scalp psoriasis, there was 27

164

percent patients in tazarotene group had a severity

score of 0 at week 12. And for the placebo, there

was 7 percent in the first study, and 12 percent in

the second study. So, tazarotene shows some

efficacy in treating scalp psoriasis.

However, if you take a look at the

fingernail and the toenail, no patient in

tazarotene group had a severity score of 0 at week

12. However, you take a look at the toenail in

placebo group, there was one patient achieved

severity score of 0 at week 12, and there were two

patients in the second study achieved a severity

score of 0 at week 12 for fingernail psoriasis.

And I want to point out here is 12 weeks

might be too short for evaluating nail psoriasis,

as nail growth requires longer period of time to

grow.

[Slide.]

The results going to present you here is

baseline of patients who were treatment success at

week 22, and had a relapse during the 12 week

post-treatment.

165

Two definitions of relapse are considered

in this presentation. The first definition is

based on patients who fell back to baseline OLA

score or worse during the 12-week post-treatment

period.

The second definition is based on patients

whose achieved maximal improvement from baseline

was reduced by more than 40 percent during the 12

weeks post treatment.

[Slide.]

This slide gives you the results about the

relapse rate for the two pivotal studies. As you

recall, there were 26 patients in tazarotene group,

and six patients in placebo group had treatment

success at week 12 for the first study, and 34

patients in tazarotene group and nine patients in

placebo group achieved treatment success in the

second study.

Based on the first definition--"a"--which

includes the patients who fell back to the

overall--the "OLA score at baseline or worse, the

relapse rate for tazarotene group in the first

166

study was 23 percent, and it was 9 percent for the

second study.

Based on the definition "b," which

includes patients whose maximal improvement was

reduced by more than 50 percent, in tazarotene

group we have 35 percent relapse rate for the first

study, and 26 percent relapse rate for the second

study.

The next several slides, I'm going to give

you the summary of the statistical analysis in the

two pivotal studies--basically, just summarize the

results I just presented you.

[Slide.]

Oral tazarotene is statistically superior

to placebo regarding treatment success, however

success rates are below 20 percent for both

studies; one was 15.7 percent, the other one is

18.7 percent.

Female patients generally had higher

success rates than males. The male patients in the

two pivotal studies accounted for over two-thirds

of study enrollments.

167

Treatment success decreases as baseline

disease severity increases. It should be noted a

total of 15 patients, which accounted for about 2

percent of study enrollment, had a "very severe"

OLA score at baseline. None in oral tazarotene

group achieved treatment success at week 12.

Consequently, there is insufficient data to

evaluate the efficacy claim of "very severe" plaque

psoriasis.

[Slide.]

Oral tazarotene demonstrates short-term

efficacy in treating scalp psoriasis, but not nail

psoriasis. Two and one patients in placebo group

achieved severity score of 0 at week 12 in

fingernail and toenail psoriasis, respectively.

However, none was in oral tazarotene group.

[Slide.]

Based on definition (a) of relapse, which

is the patients who fell back to OLA score at

baseline or worse, the relapse rate was

approximately 15 percent for both studies. And

based on definition (b), which includes patients

168

whose maximal improvement from baseline was reduced

by more than 50 percent, the relapse rate was 30

percent for the two pivotal studies. And, again,

the relapse rate was measured based on patients who

had a treatment success at week 12, and they had

relapsed at the 12-week post-treatment period.

This is the end of my presentation.

Thank you.

Clinical Wrap Up

DR. COOK: Okay, I'm back [laughs]--I'm

back for, hopefully, the last time. And I'm just

going to try to give a clinical wrap-up. And,

basically, this is to give you other information on

the safety and efficacy of the chemical moiety

tazarotene.

[Slide.]

I'm going to speak about drug-use trends

of topical tazarotene, adverse events with topical

tazarotene, and we'll take a look at the efficacy

of topical tazarotene and, again, of oral

tazarotene.

[Slide.]

169

Drug-use trends for topical

tazarotene--the total number of tazarotene

prescriptions have been increasing. In 1999, there

were 226,000 prescriptions written, and in 2003,

there were 937,000 prescriptions written.

In 2003, most of the prescriptions were

prescribed by dermatologists--67 percent--with

family practitioners coming in at 7 percent. And

the most common diagnoses associated with the use

of topical tazarotene was 75 percent for acne, and

13 percent for psoriasis.

[Slide.]

58 percent of all of the prescriptions

were dispense to women. And we arbitrarily chose

12 to 44 years as the most common range for women

of childbearing potential, and 46.5 percent of all

claims are for women of childbearing potential.

[Slide.]

There have been 125 errors reports

associated with topical tazarotene--adverse event

reports. And the most common adverse events are

cutaneous: pruritus, rash, dermatitis exfoliative,

170

burning sensation, and erythema.

There are also some systemic adverse

events reported with the use of topical tazarotene.

There are three reports of elevated LFTs,

gastrointestinal problems; hot flashes and

perspiration; and one report of elevated

triglycerides.

The indications for which the prescription

had been written included 91 for psoriasis, eight

for acne, other skin 10, and for the others we

don't have the diagnosis.

There were more females--69 females and 52

males--for which these adverse events were

reported, and there were two hospitalizations, and

one congenital anomaly.

[Slide.]

The association with pregnancy

exposures--there have been 113 worldwide pregnancy

exposures-- and this data was actually obtained

from the sponsor.

There are 107 reports in the United

States, compared with six foreign reports. The age

171

range was 17 to 38 years, the median being 29

years. The indications for which these

prescriptions in these exposures had been written

were primarily acne--60; 20 for psoriasis; two for

facial wrinkling; and the others, we don't know.

And the outcomes for pregnancy, and fetal

outcomes: there were four spontaneous abortions,

two chromosomal abnormalities--trisomy 18 and

Cornelia de Lange--not associated with retinoid

malformations. And the others--107--there was

either no adverse outcome or unknown. And we don't

know how many of which of those two categories

compose the 107.

[Slide.]

And finally we're going to take a look at

the efficacy of the chemical moiety in the

treatment of plaque psoriasis.

In topical tazarotene, in the clinical

trials, the success was based on a greater than 75

percent improvement. And I have to preface this by

saying that we know that these are different

trials, these are different sets of patients. They

172

all had moderate to severe psoriasis.

With the 0.05 percent gel, the success

rate was 30 percent and 18 percent in the two

clinical trials. The 0.1 percent gel, the success

rate was 38 percent and 25 percent, compared with a

placebo effect of 13 percent and 10 percent.

If you look at oral tazarotene, which

we've been speaking on today, in the clinical

trials the success was "no" or "minimal disease,"

which is a more stringent criterion, and the

success rate was 15.2 percent, and 18.7 percent,

respectively, with a lower placebo effect.

Also, in the clinical trials on oral

tazarotene, a secondary efficacy parameter was the

global response. And a success of greater than or

equal to 75 percent improvement, or marked

improvement or better--which Dr. Walker showed you

earlier--showed that this success rate was 30.1

percent, and 30.8 percent, with a placebo effect of

8.2 percent and 9.1 percent.

So, in summary, in pre-clinical animal

studies, oral tazarotene, when compared milligram

173

for milligram to other systemic retinoids turns out

to be the most potent teratogen. Thus, as the

other retinoids are human teratogens, oral

tazarotene is most likely a potent human teratogen.

While it is true that oral tazarotene

showed efficacy as compared to placebo in the

treatment of moderate to severe psoriasis, given

the safety signals discussed today, in presence in

semen, its effects on lipid metabolism, bone

metabolism and thyroid and glucose metabolism, oral

tazarotene presents a complex risk-benefit

calculus.

We at the agency look forward to the

guidance of this advisory committee in analyzing

this complex drug product.

Thank you.

DR. STERN: I'd like to thank the FDA for

its very nice and lucid presentations, and we'll

now go on to Dr. Lindstrom, who will talk about the

evaluation of risk management for systemic

retinoids in a more generic way.

Evolution of Risk Management for Systemic Retinoids

174

DR. LINDSTROM: Good morning. I'd like to

discuss with you today the evolution of agency

thought regarding pregnancy prevention risk

management for systemic retinoids for skin

conditions.

[Slide.]

To accomplish this, I'll first describe

the current landscape in terms of retinoids on the

market for cutaneous conditions; the risks that

they present; and the risk-management tools

available to us. I'll then move on to describe the

historical development of risk management for

systemic retinoids for cutaneous conditions. And,

finally, I'll make a few summary remarks.

[Slide.]

There are four systemic retinoids approved

for the treatment of cutaneous conditions.

Isotretinoin is indicated for the treatment of

severe, recalcitrant nodular acne. The innovative

product was approved in 1982, and three generic

products were more recently introduced.

Etretinate, indicated for the treatment of

175

severe recalcitrant psoriasis was approved in 1986,

and this product was voluntarily removed from the

market in 2002.

Acitretin, with a similar indication, was

approved in 1996, and bexarotine, indicated for the

treatment of refractory cutaneous T cell lymphoma

was approved in 1999.

[Slide.]

Now, all of the approved systemic

retinoids are either recognized, or highly suspect

human teratogens, and all produce fetal

abnormalities in animals that are exposed in utero.

We have the most human data for

isotretinoin, which is recognized as a potent human

teratogen. It has a high frequency of adverse

outcomes in exposed pregnancies, with perhaps a

third of exposed pregnancies affected. The effects

are severe, including fetal wastage, structural

malformations in major organ systems, and impaired

function, such as neuropsychiatric delay.

Additionally, the window of vulnerability

is large. There's no recognized time period during

176

gestation when administration of systemic retinoids

would be considered safe.

[Slide.]

Considering that risk, what

risk-management tools are available to us? Well,

we broadly categorize our risk-management tools

into four groups: product labeling, such as the

package insert; targeted education--an example

would be patient brochures, reminder systems, such

as stickers or patient informed-consent forms; and

controlled distribution.

[Slide.]

I want to move now to describe the history

of pregnancy prevention risk-management efforts.

I'm going to use isotretinoin as a prototype, and

use three other oral retinoids as supplementary

examples. Before I do so, I want to acknowledge a

caveat that Dr. Walker has already discussed, and

that is that isotretinoin is indicated for the

treatment of severe acne. This is different than

the indication that oral tazarotene is pursuing.

And when isotretinoin is prescribed for severe

177

acne, it's prescribed as a circumscribed 20-week

course, and the majority of patients will

experience prolonged or permanent disease

remission--not all, but the majority--and will

likely require a single course of therapy.

However, the other retinoids that I will

use as supplementary examples--etretinate,

acitretin, and bexarotine--are prescribed in a

chronic, or chronic-intermittent fashion, and would

not be considered--or the use would be more similar

to that which might be expected for oral tazarotene

[Slide.]

Isotretinoin was approved--as I've

mentioned--for the treatment of severe recalcitrant

nodular acne in the early '80s. It was recognized

as a highly suspect teratogen, based on

abnormalities seen in animal studies. It received

a pregnancy category rating of X, and when it was

released to the market, risk management included

labeling, information regarding the potential risk

for teratogenicity was included in the

contraindications, warnings and precautions section

178

of the label.

[Slide.]

Despite this, the agency received the

first report of a human malformation following in

utero exposure to isotretinoin in 1983, and other

reports followed soon after.

In response, the labeling was

strengthened. First, the teratogenicity risk

information already included in the package insert

was highlighted by boldface type, and a boxed

warning was added at the beginning of the label.

In addition, targeted education tools included

"Dear Doctor" and "Dear Pharmacist letters, which

were sent out to health care providers to update

them on the human data as it accrued.

[Slide.]

But, again, additional exposures--in utero

exposures--occurred, and in 1988, the agency and

the sponsor both determined that strengthening of

the risk-management program--pregnancy prevention

risk-management program for acitretin was needed.

And after consultation and input from the advisory

179

committee, the Accutane Pregnancy Prevention

Program was introduced. It included tools from

three of the four categories of risk-management

tools.

The labeling was updated. The boxed

warning was updated to include additional

information regarding the timing and frequency of

pregnancy testing; the number and types of

contraception recommended; and the recommended

duration for their use.

Additionally, the package itself was

changed. The blister pack was introduced. The

"Avoid pregnancy" icon--the familiar red-circle

with the slash was introduced, and the boxed

warning, which had previously been just on the

package insert was now printed on the package

itself.

[Slide.]

Other components of the Accutane Pregnancy

Prevention Program included educational materials

for physicians--excuse me, for prescribers and

patients, as well as the introduction of a referral

180

and reimbursement program for contraceptive

counseling, as well as an informed consent form for

female patients.

Two other components of the Accutane

Pregnancy Prevention Program, not formal

risk-management tools in themselves, are the

patient survey and the prescriber survey, both of

which were introduced to assess the impact of the

program.

And what was the impact?

[Slide.]

In the first year after implementation, we

saw an initial rise in the number of reported

pregnancies--reported exposed pregnancies. This

was not surprising, as a new tool for

reporting--the voluntary patient survey--had been

introduced. However, in the subsequent decade,

this number--the number of reported exposed

pregnancies--leveled off and stayed relatively

constant.

Also during this time, the number of

patients treated with isotretinoin doubled. Now,

181

it might be tempting to conclude, having said that

the number of reported exposed pregnancies was

relatively constant, and the number of

prescriptions and the number of patients treated

was rising, it might be tempting to conclude that

the pregnancy rate was dropping.

This would be an erroneous conclusion

because the number of reported

pregnancies--pregnancy reporting is voluntary, and

the number of reported exposed pregnancies does not

necessarily represent all exposed pregnancies. And

we do know that adverse event reporting falls off

over time.

However, we can safely say--I think we can

safely conclude that the known total public health

burden of exposed pregnancies was not decreasing,

and the number of women at risk for exposure during

pregnancy was increasing.

[Slide.]

Because of these two issues, the

Dermatologic and Ophthalmic Drug Advisory Committee

was convened in 2000, and this advisory committee

182

recommended that the isotretinoin risk-management

plan at the time--the Accutane Pregnancy Prevention

Plan--be strengthened to include

registration--mandatory registration and controlled

distribution.

Now, there was a precedent for the

committee's recommendation, in that Thalidomide,

which was reintroduced to the U.S. market in 1999,

was introduced with a risk-management plan that

contained these elements of registration and

controlled distribution.

After the advisory committee, the agency

and the sponsor entered into intense negotiations,

and in 2002 the current risk-management plan was

implemented.

Now, the innovator sponsor entitled that

program the SMART program--the System to Management

Accutane-Related Teratogenicity. The generic

manufacturers which entered the market soon after

that, gave their risk-management plan--which was

identical in all of the essential

elements--different acronyms: SPIRIT, ALERT,

183

IMPART. In order to avoid confusion, I'm going to

refer to all of them as the "current

risk-management plan." Although it is being

revised, it is the risk-management program that is

in place at the present.

[Slide.]

Its components included elements for all

four categories of risk-management tools. The

labeling was updated, and a medication guide was

added. There were instruction guides for

prescribers and pharmacists; brochures for

patients.

[Slide.]

The patient informed-consent forms were

updated. There was a prescriber checklist to

assist prescribers in implementation of this

risk-management plan. Perhaps the hub of the plan

was the yellow qualification stickers. These

yellow stickers are applied by prescribers to

prescriptions when they write them for

isotretinoin. There's a place to write the date on

which the patient has been qualified.

184

Qualification entails ensuring that the patient has

had a recent negative pregnancy test, and that the

patient has agreed to use two forms of effective

contraception--unless that patient is abstinent,

post-menopausal, or male.

Also on the yellow qualification sticker

are reminders for pharmacists to dispense only a 30

day supply, not to give refills, and to fill the

prescription within seven days of the qualification

date.

Now, these qualification stickers are

obtained by prescribers by signing a letter of

understanding in which they attest that they

possess the relevant competencies necessary to

safely prescribe isotretinoin and that they agree

to fully utilize the risk-management program. The

yellow stickers are provided by the sponsor upon

receipt of the signed letter of understanding.

The voluntary patient survey and pharmacy

surveys--again, not risk-management tools in

themselves--were used to assess the impact of the

program. And at the time of approval of the

185

program, the sponsor was informed that the

effectiveness would be assessed at one year, and

performance benchmarks were set at a patient survey

enrollment of 60 percent, and qualification sticker

use approaching 100 percent.

[Slide.]

The one year metrics were presented to

this combined advisory committee in some detail in

February of this year. And I'm just going to

summarize them with a single slide.

The patient survey response rate was 36

percent. It failed to meet the sponsor-identified

benchmark of 60 percent.

Sicker use was high, exceeding 90 percent,

but it proved to be an unsatisfactory surrogate

endpoint, in that there was poor correlation

between sticker use and survey responses. For

instance, although, again, the sticker use was high

and above 90 percent of the stickers were correctly

filled out, a significant number of women on the

voluntary survey did not recall having received any

pregnancy test. And perhaps most importantly, the

186

number of reported exposed pregnancies was

unchanged after implementation of the current

risk-management plan from the previous year.

[Slide.]

And so, this combined advisory committee

recommended, in February of this year, that the

current risk-management plan be augmented, and

specifically, your recommendations included:

mandatory registration of all patients, both male

and female; registration of all pharmacies;

registration of all prescribers; and the

implementation of mandatory pregnancy registry.

And I want to inform you that the sponsors and the

agency are diligently working to implement these

recommendations.

[Slide.]

So, a quick review of the isotretinoin

risk-management program chronology: the drug was

approved in 1982. In 1988 the Accutane Pregnancy

Prevention Program was implemented. In 2002 the

current risk-management program was implemented.

And in 2004, this combined advisory committee

187

recommended mandatory registration, pregnancy

registry, and controlled distribution.

I want to discuss three other systemic

retinoids at this time.

[Slide.]

Etretinate was the second oral retinoid

approved in the United States. It was approved in

1986 for the treatment of severe recalcitrant

psoriasis. It was approved after isotretinoin

entered the market, but prior to the implementation

of the Accutane Pregnancy Prevention Program. And

the risk-management program for etretinate

included--was limited to labeling, with a boxed

warnings and warnings regarding the risk of

teratogenicity in the "Warnings, Contraindications

and Precautions" section of the label. And, as

I've mentioned previously, this drug was withdrawn

from the U.S. market in 2002.

[Slide.]

Acitretin--the third oral retinoid

approved for the treatment of a cutaneous

condition, was approved in 1996 for the indication

188

of severe psoriasis. There is a caveat in the

indications section, indicating that in women of

childbearing potential this drug should be used

only if they are unresponsive to other therapies,

or if other systemic therapies are contraindicated.

[Slide.]

Acitretin was approved, as I said, in

1996. This was after the implementation of the

Accutane Pregnancy Prevention Program, and hence

the sponsor was asked to implement a

risk-management program that was consistent with

the best practices at that time.

The sponsor labeled their risk-management

program for Soriatane, the "Soriatane Pregnancy

Prevention Program," and it contained elements

similar to the Accutane Pregnancy Prevention

Program: labeling, education and reminders.

Targretin--bexarotine--was approved in

1999, the fourth oral retinoid to treat a cutaneous

condition. It's indicated for the treatment of the

cutaneous manifestations of cutaneous T cell

lymphoma which is refractory to other system

189

therapies. And--again, approved in 1999, this

would have been after the implementation of the

Accutane Pregnancy Prevention Program, but before

the implementation of the current risk-management

program. And the risk-management program for

bexarotine is similar to that for--it's similar to

the Accutane and Soriatane pregnancy prevention

programs. It consists of labeling, targeted

education, and a limitation on the amount dispensed

to 30-day supply.

[Slide.]

I just want to conclude with a few summary

remarks.

All of these approved systemic retinoids

are known or highly suspect human teratogens and,

as such, present potential risks to the public

health that needs to be management.

Risk-management programs should incorporate current

best known practices, and these practices for

pregnancy prevention risk management have evolved

over time, and have progressively included elements

from the four categories of risk-management tools:

190

labeling, targeted education, reminder systems, and

controlled discrimination.

[Slide.]

And our current thinking regarding best

practices for pregnancy prevention risk management

for isotretinoin include the fact that per the

advice of this combined committee, the isotretinoin

current risk-management program needs to be

strengthened, and that elements--and that that

strengthened program should include the following

elements: mandatory registration of patients,

prescribers and pharmacies, as well as a mandatory

pregnancy registry.

And I now want to turn the microphone over

to Dr. Ann Trontell from the Office of Drug Safety,

who will discuss risk-management tools for oral

tazarotene.

Risk Management Tools for Oral Tazarotene: Context,

Considerations and Experience

DR. TRONTELL: Good morning.

I'm going to hope to set some context, as

well as describe considerations that FDA has taken

191

into account, as well as our experience in risk

management, and in particular in the context of

potential risk management tools for oral

tazarotene.

[Slide.]

I'm going to set the context in terms of

the recently issued draft guidances on risk

management, from the agency. I'll talk about our

experience with application of some of the tools of

risk management, and what we know of their

advantages and disadvantages.

I'll talk briefly about the isotretinoin

program. Dr. Lindstrom has already told you quite

a bit. I'll talk somewhat about he ongoing

negotiations. And then I will talk about the

options for risk management for tazarotene which,

in fact, have changed since these slides were

prepared.

[Slide.]

As many of you know, under the third

reauthorization of the Prescription Drug User Fee

192

Act, the agency was charged with developing three

interrelated guidances on the topic of risk

management. The first dealt with premarketing risk

assessment; the second, with pharmacovigilance and

pharmacoepidemiology--largely applied to the

post-marking setting; and the third document now

discusses what we term "risk minimization action

plans"--a term that you've heard this morning.

I'll use that, as well as "RiskMAPs"

interchangeably.

[Slide.]

In this guidance, a risk minimization

action plan--or RiskMAP--was defined as a strategic

safety program designed to meet specific goals and

objectives in minimizing known risks of drug

products. In this context, a RiskMAP is described

as a program that goes beyond what FDA usually does

in the approval of a drug product. Ordinarily FDA,

and with the sponsor, develop professional labeling

and then conduct routine postmarking

pharmacovigilance. And this largely constitutes

sufficient risk management for most marketed drug

193

products.

[Slide.]

For risk minimization action plans the

goals are described as targeting the achievement of

a health outcome related to the known risks of the

product. These goals would reflect the idea

outcome of the RiskMAP; that might be achievement

of a certain health outcome, or avoidance of an

undesirable health outcome.

FDA recommends that these be stated in

absolute terms to maximally reduce the risk. So,

in the case of teratogenicity risk reduction, the

goal might well be stated as "no fetal exposure

should occur.'

Foals are broken down into intermediate

steps that are also termed "objectives." And it's

in the context of objectives that we talk about

RiskMAP tools.

[Slide.]

These are processes or systems intended to

minimize known safety risks, and are designed to

target the achievement of at least one or more

194

objectives that serve the overall RiskMAP goal.

[Slide.]

In its draft guidance issued in May on

RiskMAPs, FDA set forth a number of different

considerations an how tools might be selected.

Each tool, ideally, should be adding value in

attaining the program goals. They should seek,

wherever possible, to use tools that have proven

effectiveness, either in other programs, or based

upon the scientific literature.

FDA also advocates that the tools chosen

be acceptable to a wide range of audiences, and

certainly those individuals who will participate in

the implementation of the plan, and low burden

should be a goal for that, as well.

The agency suggested that in selecting

risk-management tools, that one avoid unnecessary

limitations on product access, since that might

restrain or constrain benefits of the product, and

to similarly avoid the creation of multiple

customized tools, since this creates confusion as

well as burden on the health care system. And,

195

again, so far as one is able to anticipate

unanticipated consequences of a risk-management

program, those should be considered.

[Slide.]

In the draft guidance, FDA describes three

broad categories of tools that can be used for

purposes of risk minimization--some of these have

been described already by Dr. Lindstrom--the first

being targeted education and outreach; this

involving educational materials that go beyond the

professional labeling, and may be targeted to

health care practitioners or to patients.

The second category is what has been

termed "reminder systems." This may use tools such

as stickers, that were used in the isotretinoin

risk-management program, or informed consent. In

some instances limitations on product supply or

packaging has been put in place to try and guide

clinicians and patients in using products in the

most appropriate and safe ways.

The third category involves a somewhat

awkward terminology. It's somewhat equivalent to

196

the control distribution terminology that Dr.

Lindstrom mentioned in her talk.

In the draft guidance FDA refers to the

tool system involving limitations on distribution

as "performance-linked access systems." These are

programs that, in fact, do constrain availability

of the product to certain conditions' being met.

Often there's a selected group of individuals who

may be able to prescribe, dispense or use this

product. And often these are tied to mandatory

performance of some of the reminder systems

described in the previous category.

Some product examples may give you a

better understanding of these tool categories.

[Slide.]

In the area of target education and

outreach, we can't really give you a full list

because there are probably, at this point, several

hundred products that have either patient product

inserts, or medication guides--a much smaller

number in that category. Also, a number of

programs employ patient brochures, various forms of

197

continuing education for physicians and

pharmacists.

Reminder systems include the ones I list

here. Alosetron, like isotretinoin, uses a sticker

program that indicates that the clinician is

familiar with the disease [sic], its risks and how

to appropriately prescribe it to the patient in

light of certain safety considerations.

For the drug product lindane, the amount

of product now available to patients is limited to

one or two-ounce aliquots to reduce the risks of

individuals' using excessive amounts, or using it

repeatedly and exposing themselves to certain

toxicity. So the drug product abarelix, a product

used for advanced prostatic cancer, certain

constraints are placed on how that product is

prescribed; in particular, to restrict it to those

individuals whose disease warrants the risk of

anaphylactic reactions with it.

In the category "performance-linked access

systems," I have six products listed here.

Bosentan, dofetilide and mifepristone are programs

198

that, in fact, operate under a form of specialty

pharmacy distribution. There may be one, or

perhaps a few handful of pharmacies where

individuals are able to obtain the product. And I

will apologize--I mis-spoke. The products that are

under the specialty category include Bosentan,

mifepristone and xyrem.

For clozapine, dofetilide and thalidomide,

these are available through pharmacies that are

registered, as well as the registration processes

that extend to physicians and to patients. The

ones that appear with an asterisk are ones where

laboratory testing is part of what's required for

product access.

With regard to use and non-effectiveness

of the tools, we are still in the process of trying

to understand more how these tools work and which

are most effective.

[Slide.]

Targeted education and outreach, as I've

already said, it's been used most extensively, but

formal evaluation of the effectiveness of these

199

programs has been limited--though many obviously

believe the importance of education for all manners

of risk management.

Reminder systems--again, are relatively

limited in their number. They've been used

infrequently, and effectiveness to date has largely

been untested.

For the performance-linked access systems,

or ones that register various participants, these,

again, have been used sparingly, and typically

they've been applied to relatively small patient

populations where the therapeutic options for those

patients are limited. The registration process

that tracks physicians, patients or pharmacists, in

fact, has allowed good data capture in terms of

their effectiveness. So the effectiveness of this

small category of tools is high.

Let me walk through some of the

advantages, again, of these three major tool

categories.

[Slide.]

Targeted education and outreach has the

200

advantage that education is really a

motherhood-and-apple-pie issue. It's very hard for

anyone to discount its value, so it's generally

it's very high acceptability, and relatively easy

to implement in a variety of forums and media.

The benefit of education is it has no

effects on product access.

A disadvantage, however, is, in fact, it's

effectiveness is still largely unknown to us when

it's used in isolation of other tools. And, as was

described just recently in Dr. Lindstrom's talk,

when it was applied in the most early form of

isotretinoin in risk management, its effectiveness

was low, at least in terms of pregnancy outcomes

persisting.

[Slide.]

For reminder systems, advantages of these

programs are that they allow some remnants of

physician, pharmacist and patient autonomy. These

reminders are put in place to, in fact, make it

difficult [sic] for clinicians or patients to do

the night thing. They allow the opportunity for

201

ongoing education, and the reminder of individuals

on what is necessary to achieve use of the product.

And relative to the category of performance-linked

access systems, they're obviously less intrusive.

But we have to acknowledge, they imposed

time and monetary costs on the medical care system;

and that, again, their experience to date has been

largely limited. And as was discussed in the

February advisory committee, the experience of the

isotretinoin program showed high process compliance

with the program, but limited outcome

effectiveness, in that pregnancy exposures

persisted.

[Slide.]

For the performance-linked access systems,

advantages of this is that, in fact, it does

constrain use of the product to those conditions

where use is considered to be most safe. The

mandatory participation of such systems, in fact,

allows registration of participants and the ability

to better evaluate their performance. And, in

fact, this has led to our understanding of their

202

high performance.

The nature, not only for issues of

teratogenicity risk exposure in performance-linked

access systems, is because they do represent an

obstacle to the ready use of the drug product, they

generally tend to diminish overall utilization of

the drug product; and, in a sense, then if you

wanted to reduce exposure of females of

childbearing potential, that would be a secondary

benefit.

However, you may look at limitations on

access similarly as forms of disadvantages. And,

certainly, limitations in access may also present

limitations to patients' obtaining benefits from

the drug product. Of course, there are time and

financial burdens to such programs. And the risk

that is certainly known to the agency is that the

existence of such programs may prompt individuals

to try and seek the product in less burdensome

ways, and to try and obtain it illicitly, through

the internet or other measures.

Gain, the experience that has been

203

obtained to date in the area of pregnancy

prevention, through thalidomide, has largely been

limited to a small population of individuals who

are not of high fertility. So experience in its

extension to large numbers of young, fertile women

has not yet been done.

[Slide.]

In its draft guidance, FDA tries to set

forth when you might use different tools, and it's

somewhat of a circular argument: you use them when

you need them, and perhaps when a less severe

tool--or a less intrusive tool--has proven itself

to be ineffective.

Targeted education and outreach might be

used alone, or certainly in combination with other

tools. And in those instances where product

labeling and routine pharmacovigilance have shown

themselves to be insufficient. And the example

certainly pertains in the case of Accutane and the

development of the Accutane Pregnancy Prevention

Program.

[Slide.]

204

Reminder systems may be implemented at

such times as when targeted education and outreach

are insufficient, either based upon experience with

other drug products, or in the specific drug

product being addressed. And the example of this

system would be the development of what is the

SMART, SPIRIT, ALERT and so forth programs, or what

Dr. Lindstrom referred to as the "current

risk-management program" for isotretinoin.

[Slide.]

The performance-linked access systems are

probably the category where it may be a little

easier to define the products where these may be

merited, since these tools are, in fact, intrusive.

These are ones that we might expect would be used

largely for those products that have significant or

unique product benefits, but that have associated

unusual risks that may include irreversible

disability or death. Examples--in clozapine, the

risk is of agranulocytosis. For thalidomide,

again, teratogenicity and birth defects. And the

isotretinoin RiskMAP, as we've discussed, is now

205

under active development in that arena.

[Slide.]

Let me tell you a little bit more about

what's been in progress since we last met with this

committee in February. Performance-linked access

system is under active discussion and development

by the sponsors. The details still remain somewhat

undefined. They are under active development.

The plan is to have a centralized

clearinghouse that would involve all prescribers,

pharmacies and patients; and that this

clearinghouse would be configured to assure and

account for the performance of key safety features;

for example, pregnancy testing.

[Slide.]

the key safety features of the new

isotretinoin RiskMAP will include all three

categories of tools: targeted education and

outreach; reminders; and the linkage of access to

the product to the performance of certain

activities.

[Slide.]

206

For targeted education and outreach of all

participants, there will be a medication guide,

patient brochures, videos. And, again, education

will continue for health care practitioners,

including physicians and pharmacists.

Reminders will include informed consent

and attestation on the part of health care

practitioners. There is some discussion of ongoing

patient education and risk-factor screening as part

of this program.

And the linkage of access to the product

to the performance of key features goes back to the

clearinghouse, which involves all prescribers,

pharmacies and patients--pregnancy testing will be

required for the product to be prescribed,

dispensed, and for the patient to receive it.

The issue of the linkage of these separate

components, in fact, remains a challenge on two

grounds; first, on the technology, as also for the

potential concern of constraints on the product and

mechanisms for doing that that have been patented.

[Slide.]

207

now, these slides will actually be dated,

in light of the presentation you've heard this

morning. We had, at the time of these slides'

preparation, two earlier versions of the risk

minimization action plan proposed by Allergan.

[Slide.]

Let me jump ahead--the initial one

proposed was one that was similar to the existing

isotretinoin risk minimization action plan in place

and current, prior to February of this year, and

still currently in place. This plan, at that time,

exempted males, and females who were not of

childbearing potential.

The subsequent development was for--of a

program that had linkage of the product's access to

pregnancy testing, pharmacist validation of

pregnancy testing, and patients' having monthly

reeducation and assessment of their knowledge and

compliance with contraceptive practices for

prescriptions to be obtained. That one had

excluded males and females who were not of

childbearing potential. And, as we've heard this

208

morning, that has been since modified.

[Slide.]

So, in summary, systemic retinoid risk

management for the concern of teratogenicity has

evolved and has been largely informed by

isotretinoin over the past approximately 20 years.

It has migrated from labeling alone to the use of

targeted education and outreach, the use of

reminder system; and, now, the active development

of a performance-linked access system.[Slide.]

And the tazarotene program--again, as

you've heard this morning--is now configured to be

similar to what is being developed for the

isotretinoin program, with the performance-linked

access system; and, as the sponsors told us this

morning, applied to all patients.

Thank you.

DR. STERN: Thank you very much. Might I

ask one or two quick questions?

You've talked about isotretinoin. Is

there any parallel development for the other

now-approved retinoids? For acitretin, bexarotine,

209

in terms of a parallel system for

performance-linked access?

Because I think each time you've talked

about isotretinoin, and not mentioned those others

currently labeled.

DR. TRONTELL: I'll actually refer that

question to Dr. Wilkin.

DR. WILKIN: I can't speak regarding

bexarotine. That's not in our division.

We do have another systemic retinoid in

our division. There are no changes at this time in

the risk-management program, but we have

communicated our interested with the industry group

that owns that product, that we do want to have

this discussion; we'd like to know what the current

performance is, how successful it is, and think

about the need as to whether we need to upgrade

their risk-management program.

DR. STERN: Could I then ask a follow-on

question of someone who's an expert on risk

management--is, does one believe that when dealing

with agents that have similar risk profiles, and

210

the same dominant group of prescribers--although

the patients may vary in characteristics--that

programs that apply across the board are more

likely to be followed? Or are multiple programs

likely to be followed?

I think in your guidance there was one

little thing that said one of the things the FDA

wants to avoid is too much individual

customization. And I'm wondering if some of that

is on the basis of likelihood of good performance

when you have one-size-fits-all, rather than many

sizes for slightly different product.

DR. TRONTELL: At this point in time, in

fact, we--I'm not sure we have any risk-management

program that exactly duplicates another.

Certainly, it was based upon feedback we heard from

the practicing community of physicians, as well

as--and probably in particular--pharmacists.

Because the confusing array of manners in which a

product might be presented to a pharmacist was a

cause for concern.

So the idea was, in fact, to approve some

211

element of efficiency, memorability, and to

decrease confusion in its use.

DR. STERN: One last question before lunch.

DR. SHAPIRO: I just wonder if there's a

timeline about when the Accutane proposals might be

implemented, and when and how evaluation of the

impact might be evaluated?

DR. BULL: There is a timeline, but in

terms of it being a very, very complex negotiation,

involving multiple sponsors, I would say we hope to

have it as soon as possible. But it's

extraordinarily complex. So, I would say as soon

as all of the details can be attended to, there

will be something out there. But it's being

actively, very vigorously worked on.

DR. STERN: I'm sure it will be as soon as

possible, but the question is when that will be

[laughs.]

[Laughter.]

Dr. Wilkerson--and then we will close for

lunch.

DR. WILKERSON: I just had a point of

212

clarification with Dr. Trontell.

There was something you said about patent

issues. Could you just elaborate for us what--do

we have some restraints here, or something that we

don't know about?

DR. TRONTELL: We--as I think everyone is

aware, some of the discussion for isotretinoin was

largely framed based upon the successes of the

STEPS program that was put into place for

thalidomide. The issue of how features of that

program may or may not be applied to isotretinoin

has raised the potential question of patent

protection or infringement. And that is, again,

among the complex issues that are being sorted out

at the present time.

DR. STERN: Thank you.

We've ended almost on time, and I'd like

us to start promptly at one o'clock, and we'll

adjourn for lunch.

Thank you very much.

[Off the record.]

Open Public Hearing

213

DR. STERN: We've now come to the part of

the meeting that is the open public hearing. We

have three speakers who have indicated their

intention to speak. Before they speak, I must

read, exactly as written, the following.

Both the Food and Drug Administration and

the public believe in a transparent process for

information gathering and decision-making. To

ensure such transparency at the open public hearing

session of the advisory committee meeting, FDA

believes that it is important to understand the

context of an individual's presentation.

For this reason, the FDA encourages you,

the open public hearing speaker, at the beginning

of your written or oral statement to advise the

committee of any financial relationship that you

may have with the sponsors of any product in the

pharmaceutical category under discussion at today's

meetings. For example, this information may

include the sponsor's payment of your travel,

lodging or other expenses in connection with your

attendance at the meeting.

214

Likewise, FDA encourages you, at the

beginning of your statement, to advise the

committee if you do not have any such financial

relationships. If you choose not to address this

issue of financial relationships at the beginning

of your statement, it will not preclude you from

speaking.

I've also been informed that I may only

introduce the speakers according to number. It

sounds a little bit like a Seuss novel, but would

Mr.--would Person One please come?

MR. WHITE: Well, since I know Dr. Krueger,

the only other thing could be either Dr. Krueger or

Number One. So, I guess I'm Number One.

Ladies and gentlemen, thank you very much.

I'm delighted to be here this afternoon, and

appreciate the opportunity to make a few remarks

before this committee.

My name is Dale White and I am Vice

Chairman of the Board of Trustees of the National

Psoriasis Foundation. I am volunteering to be here

today on behalf of the Foundation and the community

215

it represents, to testify in support of the drug

oral tazarotene for the treatment of moderate to

severe psoriasis.

As the parent of a teenage with psoriasis,

I am excited about testifying today about the

urgent need psoriasis patients have for additional

treatment options. Thank you again for this

opportunity.

[Slide.]

By way of introduction, the National

Psoriasis Foundation is a leading nonprofit

organization fighting to improve the quality of

life of the more than 5 million Americans diagnosed

with psoriasis or psoriatic arthritis.

The Foundation was established in 1968 by

a grassroots network of patients and physicians.

Through education and advocacy, the

Foundation promotes awareness and understanding,

ensures access to treatment, and supports research

that will lead to effective management of this very

serious chronic disease--and, ultimately, a cure.

Each year the Foundation receives

216

financial support from tens of thousands of people,

and from 15 to 20 pharmaceutical companies. This

includes unrestricted support from Allergan, and

from it's competitors in the pharmaceutical field.

[Slide.]

There was a lot of discussion this morning

about the impact psoriasis can have on people. But

I'd like this afternoon to emphasis a few important

points about the disease.

First, of the more than five million

Americans who have psoriasis, an estimated 1.5

million have a moderate to severe form of the

disease.

The Foundation's national survey research

has shown that for 75 percent of these people, the

disease has a moderate to large impact on their

daily lives. For 26 percent of these folks, it

alters normal daily activities, and for 21 percent,

it stops them completely.

For 36 percent, it causes trouble with

sleep, and for another 40 percent, it affects how

they choose their clothing.

217

For every one, moderate to severe

psoriasis can profoundly affect one's work, family,

and personal relationships.

[Slide.]

Here are a few photographs that illustrate

how physically disabling and emotionally

devastating psoriasis can be, particularly for

people with moderate to severe cases.

[Slide.]

The Psoriasis Foundation believes there is

a need for more treatment options for people with

moderate to severe psoriasis. In addition to being

a very serious disease, psoriasis is a chronic

disease. It typically first strikes people between

the ages of 15 and 35, but it can affect anyone at

any age, including children.

Our research in 2001 showed that 78

percent people with moderate to severe psoriasis

were not using aggressive therapies because of

concerns about side affects and effectiveness. In

a recent national survey, more than one-third of

the patients said they were "very satisfied" with

218

the treatment they were receiving for psoriasis.

While there are several treatments

available for moderate to severe psoriasis, none of

these treatments work for everyone, or can be used

by everyone, or necessarily works the same over

time. An individual patient's psoriasis can change

in severity, and even if type over years, months,

or even weeks. Patients need and deserve choices

that meet their individual concerns about safety,

effectiveness, cost and access.

[Slide.]

The Psoriasis Foundation believes

psoriasis patients should have access to oral

tazarotene. It may offer many patients a reduction

in psoriasis symptoms, and thus an improved quality

of life. And its approval for the treatment of

moderate to severe psoriasis would give an option

to people who cannot use currently approved

therapies.

The Foundation supports a risk-management

program focused on women of childbearing potential

to minimize, to the greatest extent feasible, the

219

likelihood that a women will become pregnant while

taking this drug.

We hope Allergan and the Food and Drug

Administration adopt a risk-management program that

addresses this risk without imposing constraints

that would effectively limit access to oral

tazarotene for the many psoriasis patients whose

lives might be greatly enhanced by it.

Finally, as with any new medication, the long-term

side effects of oral tazarotene are unknown and

need further study.

[Slide.]

In closing, moderate to severe psoriasis

can dramatically and negatively affect a person's

quality of life. People with psoriasis need and

deserve more treatment options. And the Foundation

believes access to the new treatment, oral

tazarotene, is important and desirable.

By expanding the array of choices

available to treat this serious chronic disease, we

empower patients to choose the treatment that works

best for them.

220

I know that the quality of life for my son

and thousands upon thousands of people like him

will improve dramatically as a result.

Thank you.

DR. STERN: Thank you very much.

Could speaker number two please come to

the podium?

MS. FREEMAN: Before I introduce myself,

I'd like to thank the National Psoriasis Foundation

and the tens of thousands of people who support it

for giving me the opportunity to tell my story here

today and to represent the millions of people

suffering from psoriasis.

I also need to say that I do not have a

financial interest in the company that makes the

drug we are talking about here today.

It is ironic that I should be here, as

less than three weeks ago I was sitting in the

clinical trial research center, and I had just

filled out my monthly questionnaire. I closed the

folder and went to date it, and next to the day was

"Patient Number 1569." I looked at my new bottle

221

of pills, and again I was "Patient 1569," and I

thought, "Gee, here I am, I'm really just a

number."

So, today I want to introduce you to

Patient Number 1569. My name is Janey Freeman, and

I live in Yantis, Texas, which is two hours east of

Dallas.

I'm married. I have two children, and I

currently work as an office manager for a land

developer and an insurance agent. And Yantis is as

"country" as it sounds.

[Laughter.]

I was diagnosed with psoriasis when I was

20 years old. So for 34 years I have been injected

with steroids, wrapped in tar, put under lights,

and zapped with machines. I have used creams,

lotions--not to mention slept wrapped in

cellophane--and I have worn gloves and socks filled

with all kinds of creams and lotions.

I have washed my hair with tar shampoo,

and I've ruined a lot of towels and a lot of white

bathtubs, soaking in all kinds of products.

222

I have also taken methotrexate. This made

me sick two days out of the week. And about the

time I was feeling better, it was time for me to

take another dose.

I have had my blood drawn every six weeks

for 10 years. And I have had one liver biopsy.

Psoriasis is not usually life-threatening,

but some of the current treatments are. So, forced

to choose between a quality and a quantity, I chose

the quality.

What would my life be like had I not had

psoriasis? I'll give you some examples.

When I was younger, I thought about

modeling, but models don't normally have psoriasis.

And then I wanted to be a dentist, but at that time

my hands and my nails were really badly affected.

So I chose a field where I could sit at a desk and

be out of the public view.

I have missed a total of two years of work

due to the disease. I have turned down two

promotions. I missed my senior high school trip to

New Mexico because I was afraid I might flare.

223

I have never had a manicure or a pedicure,

and my best friend still cuts my hair.

From my early 30s to my mid-40s, I was

single. My psoriasis was then at its worst, and I

was too embarrassed to have an intimate

relationship.

My closet has always had two sets of

clothes: the clothes that I wear when my psoriasis

is bad, and then my "sometimes" or fun clothes that

I get to wear when my skin is okay.

I have had co-workers more out of my part

of the office. I have had nurses put on rubber

glovers for just a routine exam. And people have

even moved to another cashier after seeing my arms

or my elbows when I'm flared.

Psoriasis hurts. The lesions bleed. You

itch uncontrollably. It's embarrassing, it's

expensive, it's physically disfiguring, and

mentally exhausting.

I can't really tell you what my life would

have been like without psoriasis, but I can tell

you that Patient Number 1569 is better because of

224

oral tazarotene.

The medicine I am using is not perfect,

but I feel hopeful for the first time. My plaques

are clearing, and I'm at least 60 percent

improvement according to the research center, and I

have no new ones.

My scalp is clear and the itching is gone.

I've never been sick to my stomach, and the only

side effects I have experienced are mild joint pain

and the dry skin.

The first time I walked into my current

dermatologist's office was 10 years ago, and I was

crying. A few weeks ago, after my last exam, we

both were smiling, excited and hopeful.

I believe that his drug--oral

tazarotene--should be available to patients that it

might help. With continued research and dedicated

and caring physicians and staff, we can give hope

and options to those millions of people like me who

suffer every day of their life with psoriasis.

I thank you very much.

DR. STERN: Thank you.

225

And could we please have speaker number

three come to the podium?

MR. GORRE: Good afternoon. My name is

Tyrone Gorre. And I was born in Sacramento,

California, raised in Newcastle, which is about 45

minutes northeast, in the Sierra foothills--and

that's considered what is called "Gold Country."

I am a single, 45-year old fishing guide

and ranch hand, who is currently raising two

daughters--young adult daughters--who are seemingly

unaffected yet.

I want to say thanks to the National

Psoriasis Foundation for making this--giving me

this opportunity to speak to you guys today, and I

would like to say at this time that I have no

financial obligation or commitments or interests in

the company that makes this particular drug.

I've had psoriasis most of my life. I've

tried many different treatments over the years,

including UVA, B, psoralen with PUVA, most topical

steroids, Dovinex, and Tegison and methotrexate.

Through that period I really began to understand

226

what medical "practitioner" meant. That changed

for me.

My psoriasis really started getting bad

when I was about 25 years old. I had had it for a

long time before that, but it really became more

prevalent and widespread, with noticeable plaquing

on my face and all over my body. At this time I

was just a newlywed and beginning to have children.

Psoriasis had changed my life. I used to

be a life guard and wear shorts all the time. I

gradually changed my style of dress to basically go

into a coverup. I would always wear long pants and

long-sleeved shirts and even hats; collars,

turtlenecks--anything to hide this terrible

disease.

I was really embarrassed, because at that

point I had psoriasis on about 40 percent of my

body.

My daughter has recently told me that her

mother said one of the reasons that she divorced me

was because when I was young and married her I did

not have psoriasis, and while we were married I got

227

psoriasis and it became very bad.

She said that his was a big factor in my

divorce, and I really do believe it. It would be

terrible to go to sleep at night, and have the

woman I loved wake up covered in scabs and pieces

of skin on her.

There was few years when I simply played

cover-up, and hide my disease, because I had no

access to health-care coverage, or no insurances of

any form. So, I tended to worsen at that time.

Since I've had psoriasis pretty bad for

quite a long time, I had known a dermatologist who

was helping in research studies. At that time he

had asked me if I would participate in a study of a

new oral drug, tazarotene.

I thought about it seriously. At the

time, I had had a nephew who was beginning to show

signs of psoriasis. He was about the same age as I

was when I got psoriasis, or noticed psoriasis.

This was the extra motivation that I needed to join

the study. I was willing to do this because I

wanted to answer the question for myself, for my

228

family and all the people who I recognized as

having this disease.

During the first part of my study I was 99

percent sure that I was given a placebo. There

seemed to be zero effect. During the second part

of the study, I believe I received the drug. The

side effects that I noticed while taking the drug

were basically my feet peeled one time, and I

seemed to show some signs of achiness.

After about six weeks, I noticed the

thickness of my plaquing beginning to shrink

dramatically. And, man, my attitude was really

changing at that time. I was really excited.

Within about four months, I was 95 percent

clear. I had just small areas, and they were all

less than the size of a quarter. My research

clinic that I went to told me that I went from

having 25 percent to 4 percent. That was pretty

good improvement but, man, it felt like it was way

more than that. It was much more of an

improvement--in my mind.

At the point the study ended, I had only a

229

few patches of red skin that were very small. For

the first time in my life, I was virtually clear of

psoriasis.

Now, I wear flip-flops, open-toed shoes

and shorts. I even wear black shirts now. For the

first time in 20 years, I wore a black tuxedo with

a silk collar. That's absolutely amazing.

For the first time, I would go to northern

California and not get kicked out of the hot

springs. For the first time in my life, I went to

my health club and people didn't stare at me when I

went to the shower.

I used to feel bad about this. The amount

of mental pressure that is released is huge.

The confidence of not having psoriasis is

amazing. It brings amazing confidence back. It

was so rewarding to not to have to worry so much

about this problem.

There are three huge things that this

study has changed for me. One, with psoriasis you

cannot sleep at night. The constant itching just

absolutely drives you--and if you have a partner

230

would drive them--absolutely crazy, and it keeps

them up at night also.

Two, you are so self-conscious about

psoriasis that the psoriasis virtually eats your

confidence down to nothing. You do not even want

to be seen. So that's all changed now, too.

Three, the constant pain and stinging of

psoriasis is incredible. This consumes a major

part of your life. And now that has all gone.

Imagine this: take a mosquito bite--take

the mosquito bite and get it on your knuckle. Take

that itch that you feel, and multiply it times 20.

Then scratch that itch for five minutes, until you

break the skin and you make it bleed. And about

the time you start feeling that that itching is

gone, throw some salt on that wound and experience

the burning of psoriasis. That is what I would

feel up to a hundred times a day. And I've

experienced that for over 20 years.

With that sort of feeling a hundred times

a day, what kind of distractions from my life has

occurred, and what have I missed in my life?

231

I will live with psoriasis for the rest of

my life. I'm really glad that I got involved with

this study, because it gave me hope, and it should

give hope to other people with psoriasis, too.

After more than 20 years of struggling with

psoriasis, I feel like a new man after being on

this drug for the short term that I was.

This medication not only improved my

physical health, but it gave me back my

self-confidence.

One thing that I hope this committee will

do is make this drug available to patients who need

it. I'm just a common laborer--a working man. And

in most of the work that I do, I need tools to

accomplish the jobs that I'm asked to do. And I

really believe that there are not enough tools out

there--and I've used the tools to try to take care

of my psoriasis.

So I really hope that you guys have a

strong consideration for people who are suffering

like me.

This drug is definitely the least

232

threatening of all oral medications that I have

taken. It has worked for me, and I've done this

study for myself, for my family and for the

thousands of people just like me all over the

world.

Thank you very much for your time.

DR. STERN: Thank you.

Is there anyone who did not register who

would like to come and present at this, the open

public forum?

[No response.]

With no one so indicating, then we'll call

this the end of the open public forum and more on

to discussion and questions.

And, again, I'd like to thank the three

people who presented for taking the time and

traveling here to give us their feelings and

opinions.

Discussion and Questions

DR. STERN: Until two o'clock, I would

propose that we have the first part be questions of

clarification--essentially what we did after the

233

sponsor's presentation in the morning, but now the

questions could go to anyone: FDA, sponsors or, in

fact, any other committee member.

So, if anyone has questions in terms of

content, as opposed to really deliberation of the

questions that have been posed to the

committee--why don't we start with you, Dr. Honein?

DR. HONEIN: Yes, I just had a question of

clarification from FDA.

My recollection at the February meeting is

that we had recommended that pharmacists register

rather than the pharmacy. And what I saw presented

was the pharmacy in what was proposed for this.

So I was just wondering if I recalled that

wrong, or what our final recommendation was?

DR. BULL: I would say, given that the

overall program is not completely worked out, there

are still a lot of details remaining. And I think

some of the--is it on?

We're trying to look at what will be the

most efficient way to ensure the risk management,

and I think it may be premature to say exactly

234

which part. I think there was a lot of discussion

at the meeting--as you recollected--on the issue of

pharmacies and pharmacists. But I think that's an

element of detail that is still being evaluated as

what will be the most effective way to ensure that

the elements adequately address the risk.

DR. STERN: I had a question for FDA; a

little bit of a follow-up.

I know it's a very complex process, with

multiple drugs and multiple sponsors. But are

there other technical or external constraints that

might be considered in slowing--potentially slowing

the progress made toward developing a PLAS system?

DR. KWEDER: I'm Sandra Kweder. I was very

involved in your meeting in February. I'm the

Deputy Director of the Office of New Drugs.

And I had suggested to the group that I

follow-up on this question, simply because it's an

issue that's not only relevant at the working level

in the division, but something that the agency and

our office of chief counsel is concerned with, as

well.

235

Probably the main--the sponsors have been

working extremely well together, and this is really

unprecedented among a number of generic firms and

the innovator firm for isotretinoin. And I think

it's fair to say that.

But, without going into a lot of detail,

the patent issue appears to be potentially quite

large. There is a patent held by--there are

actually four patents held by Selgene Corporation

on risk management--the concept, basically, of any

risk-management program that links through one or

more computerized data bases--patients, pharmacists

who intend to distribute drugs, and physicians who

may be required to provide data to that system.

And the patents cover products where fetal toxicity

is of concern, as well as any potential adverse

event or contraindication.

So, these patents are available--you can

find them at the U.S. Patent Office website. But

this is the first time that we have encountered a

situation like this, and it--our lawyers are

studying this. But it does appear that it may pose

236

some obstacles to the implementation of a modified

program as you recommended, and as we have been

pursuing.

DR. STERN: Thank you.

Eileen?

DR. RINGEL: This may be a question for

Allergan--whoever wants to answer it is fine.

I was wondering if patients who had

erythrodermic or pustular psoriasis, that subgroup

was identified and, if so, what the data about

efficacy was for this medication?

DR. WALKER: I'll answer that for you.

We required the patients to have stable

plaque psoriasis, and did not do a subgroup--or did

not have patients actually enroll in the trial that

had erythrodermic or pustular psoriasis.

DR. STERN: Dr. Day?

DR. DAY: I just had a brief comment about

pharmacies versus pharmacists.

In the briefing document from the sponsor,

it does say "a representative;" "a pharmacy

representative." It does sound like one person

237

from each pharmacy--in the briefing document.

DR. WALKER: yes--I can comment on that.

We are proposing that a pharmacy be

registered, and that there is a representative from

each pharmacy who is responsible for training all

the pharmacists within that system--really, to

avoid problems such that if one pharmacist is

registered, they all aren't, one gets sick, is out,

that the drug couldn't be distributed. So the

entire staff would be registered by one key,

identifiable individual within that pharmacy.

DR. STERN: Dr. Katz.

DR. KATZ: A question for Dr. Walker.

Can the committee be privy to the expert

opinions of people with expertise in bone

metabolism?

DR. WALKER: They certainly can.

DR. KATZ: Can we--

DR. WALKER: Yes--

DR. KATZ: Thank you.

DR. WALKER: I have three different people

here that have analyzed the data, and a fourth

238

person who, unfortunately, can't be here. But we

do have the conclusions from that person. They've

had a family emergency.

We have someone who's an expert on both

the technology and interpretation of bone mineral

density data; on orthopedic changes--calcification,

osteophyte formation; and then an expert outside

statistician who's helped us look at the data to

see how these regressions from the mean, and are

they what you would expect in this population as

normal variance.

So it might be helpful for me if you tell

me--if you want me to bring them up individually,

or if you have questions that I can then field to

the appropriate person?

DR. KATZ: The main question is what they

would visualize as progression with the data that

we've seen, combining the hip demineralization, the

alkaline phosphatase--how that can be tied up. And

what one would expect for the future.

DR. WALKER: All right--this is such a

complex area. I have a lot of data in different

239

things. I'm going to start by having the

statistician, who's looked at the data that we

have, looked at the data from the 12-week study,

and then looked at the data from the open-label

050P study, and has done some statistical analysis.

Now, he's looking to see: is there a

variation that you would expect in the population?

So I think we'll start there. And then

I'd also like to show you data that Frederick

Bettingfield from Allergan will show you on: is

this actually regression to the mean?

And we'll go from there. I think this

could be fun.

DR. HELMS: Yes, I'm Ron Helms. I'm

Professor Emeritus of Biostatistics, University of

North Carlina, Chapel Hill. And I've been doing

this a long time--someone suggested I say.

Probably a statistician is the last person

you want to hear from, instead of the first. But

there is an important point here.

Can we have the slide up, please?

[Slide.]

240

This is data from femoral neck bone

mineral density evaluations. And I apologize from

the lightness of the slide.

This is in the 52-week study. The red x's

there are a scatter diagram. Along the bottom we

have baseline measurements, and on the vertical

axis we have the post-treatment or endo-treatment

measurements.

And the primary point I want to make with

this slide is that these data follow a bivariate

normal distribution very closely. The data are

closely approximated by a bivariate normal

distribution.

Now, that would be more meaningful to the

statisticians in the room than to others, perhaps.

The consequence of that is that: if we see a shift

from baseline to post-treatment over a period of a

year, and if this bivariate normal distribution

fits, then the shift shows up in the mean. It

doesn't show up in other kinds of features of the

distribution.

If you look on the bottom right-hand side

241

there, the man shift here was about minus .01; the

mean shifted in 102 patients, I believe it was,

from .943 to .933. And a similar median shift.

So, over a period of a year there was

about a 1 percent shift in the mean; a decrease in

bone mineral density.

While the slide is up I'll go ahead and

make another point. I hope this is okay.

There has been some discussion about

values that had decreased by more than 5 percent.

There's a line that's very difficult to see, right

along there--it's a green line--that's a 95 percent

line, and that is--points below that line are

points that were less than 95 percent at the end of

treatment--less than 95 percent of what they were

at the baseline. And there are some points in

there.

The fact that this is a bivariate normal

distribution actually means that that's not a very

good way at looking at these data. The ellipse

that you see there is called a 95-percent tolerance

region, and it's designed to capture 95 percent of

242

the data points. And in this case, it actually

captures 94 percent of them, which is a pretty good

fit.

But looking at points that are below the

line, where there was a reduction of 5 percent or

more, there were 11 percent of those points in the

data. If we were finding something going on other

than just normal trends, just noise, that appears

in these data points, we would expect to see more

than 11 percent. Actually, we would expect to see

13 percent, just on the basis of noise. We only

saw 11 percent.

So this is an indication--the conclusion

from that is that what we're seeing here is just

bivariate normal random variability.

Yes, sir?

DR. STERN: I wanted to ask you a question,

because to me, it's very different when you're

looking at sometimes rare idiosyncratic effects.

So I guess what I'd like to know is: what was the

power of this study to detect a 1 percent--1

percent of people having, in fact, 5 percent

243

reductions, and prove it statistically?

So, you've shown us everything fits in,

but I don't know the power of this and how you've

treated people without multiple observations.

So I would ask you--I'm interested in your

assuring me, with 80 percent confidence, that not

more than one in a hundred individuals has more

than a 5 percent reduction over a year. What's

your power to exclude that with a beta .8 and an

alpha of .05?

That's, to me what's relevant in a safety

study.

DR. LUE: I'm John LUE, biostatistics.

I've done some power calculation based on

a 1 percent background information, what it would

take to detect a 2 percent difference.

The power is about 11 percent.

DR. STERN: I'm sorry, is that a 2 percent

difference in mean? Or one in a hundred

individuals--

DR. LUE: One in a hundred.

244

DR. STERN: Okay.

So, in other words, there's an 89 percent

chance that a difference--one in a hundred people

could have more than a 2 percent reduction, and we

wouldn't have detected it in your study.

DR. LUE: I need to qualify that. This was

based on a sample side of average of 350 per

treatment group. So a basis--

DR. STERN: This is only about a 150

people.

Dr. LUE: Correct.

DR. STERN: So the power of this is

probably--

DR. LUE: Less.

DR. STERN: Less--like .05, .03--

DR. LUE: Right.

DR. STERN: --something in that--it's

probably not linear.

DR. LUE: Correct.

DR. STERN: Okay. Thank you.

DR. HELMS: Let me just clarify one point,

though.

245

There is a lot of--the substantial amount

of power here for--and this goes back to the point

I made about it being a bivariate normal

distribution. If it really is a bivariate

normal--and it fits very well--then the shift will

show up in the mean. Even if there is a small

subgroup, if it's a bivariate normal, the shift

will show up in the mean.

DR. STERN: It all depends on how small a

subgroup you'd be concerned about.

DR. HELMS: That's correct.

DR. STERN: And that's why I asked the

question. One in a hundred, to me, would be a

small number of people relative to those treated,

but a clinically very important endpoint, were it

true. And that's why I wanted to know the power.

So what I'm hearing is: there's not much

power at all, here.

DR. HELMS: That's correct. I mean, there

are 102 patients in the data.

DR. STERN: Thank you.

DR. WALKER: A couple more speakers for you

246

on the same topic.

DR. STERN: I'm wondering, if there's no

power to reject the null, whether we should spend a

lot of time. Because I have already an increasing

list of speakers.

So, why don't we go on--

DR. WALKER: Well, umm--

DR. STERN: --to the next thing?

DR. WALKER: I actually think, if you look

at this--and there's a lot of data in this field

that I do think is important to look at.

If you look at osteoporosis studies, for

instance, it's very common for patients to have a

reduction as great or greater than what we saw on

drugs.

So I do think it does lend credibility to

what we've seen, and makes it a little more

questionable.

We aren't saying that we don't know for

certain that you don't detect a signal. We just

think that the risk is minimal--not certain.

DR. STERN: I don't mean to be in any way

247

critical of the sponsor. I think we have to look

at what we're looking for. And to my mind, what

we're looking for--and perhaps other members of the

committee would disagree--is a relatively low

frequency event that could be idiosyncratic, and

then we have to regard what is the quantity and

quality of the data we have?

And I think I agree with you that we

can't--these data do not either tell us that this

drug is not bad for bones, or bad for bones, and

that we have to live with that uncertainty. And I

think we can look at these data five ways from

Sunday--one of the things I'm a little too prone to

do--and come to opposite conclusions.

But I think in the interest of time, we

should probably move on.

DR. WALKER: All right. I actually

appreciate what you're saying, and I think--I agree

with you. I think my colleagues would agree with

you.

The only thing to think about, when you

think about the drug, is that it is a class of

248

drugs that we do have 20 years' experience with.

And what we're seeing is nothing outside of what we

would expect with the class.

Thank you.

DR. STERN: Ms. Shapiro?

MS. SHAPIRO: This is really a quite

different topic. Okay.

I'm just wondering, from both industry and

the FDA, if you can help me understand the purpose

of, and impact of, the pregnancy test which, as I

understand it, would be a part of this proposed

risk-management program, as well--other than the

first one.

In other words, when you do them monthly

and you get a positive, is the purpose--are you

going to then give counseling to that person? Is

it going to be paid for by industry? Are you just

doing it so you can collect data?

What--it's going to be too late, maybe, to

prevent harm. So what's the purpose of it?

DR. WALKER: Do you want industry to start,

or FDA?

249

DR. STERN: Did you want to comment on

that, Dr.--

VOICE: [Off mike.] [Inaudible.]

DR. STERN: For some of us who've been

part of this process for the last 16 or 18 or 20

years, it's been a subject of debate. And we could

probably be here for about another week and not all

agree on what it is.

It's--everything have evolved in a way,

and there are varying opinions about which elements

are most and least effective. And I think this is

not the main purpose of our meeting today. And

I'm--

MS. SHAPIRO: But, you know, with all due

respect, if we're going to give advice on a

risk-management program, I'd like to hear some of

the proposed answers--in a nutshell.

DR. TRONTELL: Ahh--I can try and take the

first pass at this--Anne Trontell.

You know, the purpose of ongoing pregnancy

testing is, obviously, you would want to detect a

pregnancy early and inform the patient about the

250

exposure and options available to that person.

Education may have a secondary role, in terms of

reinforcing the importance of adhering to

contraceptive behavior.

And I would like--as a pointy-headed

epidemiologist, and someone heavily invested in

evaluating whether or not these various programs

have an impact--it also gives us very important

information to know whether or not the

interventions that are being addressed, in fact are

having their desired impact.

So, I think those three are probably what

we think is most important about it.

DR. WALKER: I'd like to add that,

actually, what's different in the program being

propose now, and what's different in the

recommendations for the new isotretinoin programs,

are that there is a response to that pregnancy

test.

You have a direct link--the pregnancy test

has to be negative before the drug can be

dispensed. Also, the educational materials--as you

251

heard this morning in the FDA presentation--the

educational materials, there is a performance that

the physician and the patient have to achieve that

will trigger dispensing of the drug. If the

patient doesn't understand the contraceptive

measures required, and that they need negative

pregnancy tests, then the drug, again, won't be

dispensed.

So it is different than the current SMART

program, in that there is--it's not just a

pregnancy test. There is actually a response to

that test.

MS. SHAPIRO: But if it's positive--it

could have been positive for 29 days.

DR. WALKER: Well, the entry into the

program requires at least two consecutive ones. But

then, next, if a woman of childbearing potential

gets it during her menstrual cycle then, in theory,

you should be within a two week period. And,

you're correct, it would just be stopping the drug

very early, rather than, certainly, preventing the

pregnancy.

252

DR. STERN: But, at least, for

isotretinoin, the available evidence is that there

is no safe period of exposure. There's a

literature going back to the late '80s or early

'90s based on the originator company's data that

showed that.

So, early stopping, once a pregnancy has

occurred in an exposed individual is--

DR. WALKER: Agreed.

DR. STERN: --a very serious event.

DR. WALKER: Agreed.

DR. STERN: Dr. Epps?

DR. EPPS: Sorry to return to the bone

topic just for a second.

There were--I guess the company didn't

comment on the fractures. I would like to know the

type of fractures, and characteristics of those

people.

DR. WALKER: Yes, we can show you those

fractures. And we did go through the case report

forms, and then go to the source documents at each

site, for each fracture.

253

And Dr. Beddingfield's going to share that

with you.

DR. BEDDINGFIELD: I'm Dr. Frederick

Beddingfield. I'm the medical director of skin

care at Allergan.

Slide up, please.

[Slide.]

We did go back and look at the fractures

that were noted. We looked at several possible

variables that could be associated with patients

who had the largest decreases in the study.

And, as you can see from this slide,

actually there were more than six fractures in the

study. But what's interesting to note is that the

age of the patient is typically young; typically

male; and the sites of the fractures were mostly

digits--and this has not been associated with

decreases in bone mineral density.

If I could now have slide S-87, please?

[Slide.]

We looked at what were the bone mineral

density changes in these patients, nonetheless.

254

Slide up.

[Slide.]

And what we found is that the bone mineral

density changes within changes, and across the

group, there's no consistent pattern, and there's

certainly--one can see there's as many increases as

decreases in bone mineral density. There's no

clear relationship at all to these patients. And

these were the six fractures mentioned, for which

there was bone mineral density data. There's no

clear relationship to the fractures and mineral

density changes.

If I could just take a second and have

slide S-183, please. There was mention in the

briefing package of a patient with a 50 percent

change in bone mineral density.

And--slide S-183, please?

[Slide.]

And just to--or, actually, slide S-183.

Thank you.

[Slide.]

Just to set the record straight, this is

255

the recording we're talking about. And this is

another recording from the femoral neck at the same

time. And these scans were of unacceptable

quality. I have further information on why they

were unacceptable if you need that. But those were

inaccurate results, and they were not accepted.

The patient did have some changes in bone

mineral density, but they were much, much smaller,

and nowhere near the same degree.

And if I could have slide S-206--there was

one other patient who was mentioned, with bone

mineral density change close to 30 percent.

Slide up, please.

[Pause.]

[Slide.]

Yes. And this is the changes in this

patient over time.

This is the value that we're referring to.

It's important to note that this patient was an

over 300-pound gentleman, which makes this a very

technically limiting study to perform. And most of

the values were nowhere near this range.

256

However, again, there were consistent

decreases in bone mineral density in this single

patient, but certainly not in the range of this,

over time.

DR. EPPS: Well, while you're handy, there

was a comment, I guess, at one time that, you know,

perhaps the gains or losses--depending upon which

we're looking at--were per natural progression.

Do you have normal progression?

DR. BEDDINGFIELD: Yes, we have normalized

the data.

DR. EPPS: Not normalized. I mean in the

normal population--not on drug.

DR. BEDDINGFIELD: Okay, what we did--and

I'd like to have our bone densitometrist speak to

this--but what we did was to use a t-score

evaluation, which is what the World Health

Organization recommends. And it normalizes the

data to the ideal adult male--young male bone

mineral density. And it helps put it into

perspective, because just because someone has a 5

percent change in bone mineral density doesn't tell

257

you where they end up. But the t-score does. It

lets you know if they're osteoporotic, osteopenic,

or normal.

DR. EPPS: Why weren't females used?

DR. BEDDINGFIELD: Well, that's something

to ask the World Health Organization, I suppose. I

wouldn't know the answer to that.

[Laughter.]

But that's a very good question [laughs].

Could I have slide S-196, please.

[Pause.]

Slide up, please?

[Slide.]

And this is a summary of the t-score

results--in the patients with 5 percent losses.

What we found is that these would be--the

patients with the worse losses in the study, that

we're specifically looking at here--44 percent of

them had normal bone mineral density throughout the

study, all measurements; 38 percent had osteopenia

at baseline, and they never became osteoporotic.

Just to put this in perspective,

258

osteopenia is 1 standard deviation from the mean;

osteoporotic is 2.5 standard deviations.

16 percent who were normal did develop

osteopenia. No patient in this group--not a single

one--developed osteoporosis. There was one patient

who started osteoporotic, and remained

osteoporotic.

Notwithstanding Dr. Stern's comments about

the power, which I certainly appreciate, I think

this is at least helpful information on the

patients with the most significant losses.

DR. STERN: I can't resist once more saying

that: these are one-year data for a chronic

disease, which we've heard earlier, has an average

duration in severely affected individuals of 40 to

60 years.

So, what we're looking for is some little

signal. If we had one patient who went from normal

bone density to osteoporosis in a year without some

other explanation, we would be--we probably

wouldn't be meeting here today.

So, we have to look for subtle signals.

259

And I don't find these one-year data--this

distribution--terribly reassuring.

[Pause.]

I'm sorry--Dr. Wilkerson.

DR. WILKERSON: I agree. I mean, we're

talking long-term disease here. We have a signal

from laboratory--just anecdotally, from years of

retinoid use, I've rarely seen elevated alkaline

phosphatase with the other products on the market.

So, obviously, there's something going on

with this drug. The sponsor didn't pursue any more

clarification of that. And, you know, as much as

we want to dance around this data here, we've got

another signal from hard-core laboratory, that is

reproducible, indicating a problem, you know, with

metabolism.

Now, is this drug going to be labeled an

and on-and-off type drug? Or is it going to be

labeled as a continuous administration? I mean,

what are you going for?

DR. WALKER: Our proposal is to have the

drug for chronic use. We studied it up to one

260

year.

We feel that the bone changes are minimal

and rare. I do fully agree that a rare event

cannot be picked up in clinical trials. This isn't

unique to this product, it's any product at the

time of approval, if it is a one in 100,000, or one

in 10,000 rate of occurrence, you will not pick it

up in most clinical development programs.

DR. WILKERSON: But, in all due respect, we

may be talking about one in a hundred, or one in

200.

DR. WALKER: But you won't--

DR. WILKERSON: Not that rare, rare event.

DR. WALKER: Okay. I agree.

DR. WILKERSON: The n of your study is so

small--particularly for the long-term

administration of this product. I mean, you've got

far too few people in your studies right now.

That's the problem.

And while we need some alternatives--and I

think that's what we're all here for, is to offer

patients alternatives, we also, as a clinician, we

261

need to offer them things that we are relatively

sure are safe for them to take long term--in five

years we don't discover that, my God, you've got 30

or 40 percent bone loss, you know, sitting here,

and you're 35 years old, and now you're

osteoporotic at age 40.

DR. WALKER: I absolutely agree with

everything you've said. But this drug is a new

drug, and this is a new indication for an oral

formulation. This is not a new class of drugs.

DR. WILKERSON: No, you're right. But we

have not--either we have missed the boat with the

other products on the market, and we have not

detected this, or, because the selectivity of this

particular drug, we are seeing a new side effect.

DR. WALKER: I--

DR. WILKERSON: This is not something that

has been worried about with other retinoids that

have been on the market. We knew about the DISH

syndrome. We knew about calcification on the

ligaments. But bone mineral loss has not at least

been on the radar screen of clinicians for use of

262

retinoids in the United States.

DR. WALKER: Well, there are published

reports of bone--

DR. WILKERSON: There may be--

DR. WALKER: --of bone mineral loss with

isotretinoin. If you look at the labeling for

isotretinoin and for acitretin, there is an

alkaline phosphatase that increases up to 30

percent.

So, I do--you know, agree with what you're

saying, but I take a little issue that these are

new or unique events with tazarotene that haven't

been observed in the class.

I do agree that it is, in general, not a

major concern to clinicians when they start

patients on acitretin or isotretinoin. However,

when I trained, I did check x-rays in patients on

acitretin for beyond--or, at that time, it was

etretinate--beyond one year. And we did follow

them routinely.

So, I do think that for chronic use, some

physicians feel differently. I might help if I had

263

one of our experts--Dr. Lebwohl, who's used a lot

of retinoids--comment on how he sees this drug

relative to others

DR. STERN: I think we're going to just run

out of time.

DR. WILKERSON: But, on the other--not to

beat this bag of bones and move on--the other issue

is: is this risk-management program that's being

proposed going to be restricted also by these

patent restrictions that we're talking about? Or

is your program exempt from that.

DR. WALKER: Well, I have not been involved

with the discussions for isotretinoin to be

familiar enough to comment on whether we will be.

However, if we are proposing and adopting

all the essential elements of isotretinoin, which

is what I've been telling you, it's very likely

that we may have the same restrictions.

But I would have to defer that question,

actually, to the agency for comment. We haven't

been involved in those discussions.

DR. WILKERSON: Could I hear from Dr.

264

Lebwohl.

DR. WALKER: Yes. Thank you.

DR. LEBWOHL: Sure. Mark

Lebwohl--dermatologist in New York.

First, I would say that the amount of data

regarding bone density that is presented here I

think compares very favorably to what has been

presented with the other retinoids that are

available on the market.

Now, we do have a long history of use and,

in fact, large numbers of patients who have been on

oral either etretinate or acitretin for many years.

And certainly in those patients I don't doubt that

there is an amount of mineral bone density loss

that can be found if you look closely, but it is

not clinical significant. We're not seeing

fractures--we're not seeing the kind of changes

that you see with systemic steroids.

So, while--you know, I think that the kind

of statement that is in the package insert for

those drugs is appropriate for those drugs, I think

it does put into perspective what we're seeing in

265

clinical practice both with this drug for, albeit a

shorter period of time, as well as with those drugs

over a long period of time, it's not clinical

significant.

And I'm not aware of any clinician

nowadays who routinely gets mineral bone densities

or x-rays in patients on oral retinoids for years.

You know, the person who actually knows

this better than I is Tom Fuerst, because he was

telling me about this with other drugs that are

actually used to treat osteoporosis.

DR. FUERST: My name is Tom Fuerst. I'm

trained as a medical physicist. I've worked in the

area of bone densitometry for the last 10 years,

using it to assess osteoporosis and fracture risk,

and monitor changes in bone mineral density, And I

just wanted to make a comment about the potential

long-term effects.

I don't have the answer to that. I don't

think the data in this room to answer that

question. But, in general, it's difficult to

extrapolate from shot-term data to longer term use.

266

The general trend, whether you're introducing an

agent that will increase bone mineral density to

treat low bone mass, or another agent that might

have a deleterious effect on bone mineral density,

in general the changes are larger in the first six

to 12 months, and not sustained afterward. There

may continue to be losses, but just a simple linear

extrapolation is difficult to do.

But, again, I don't have the data bout

long-term treatment with this drug, but just a

caution about extrapolation.

DR. STERN: Dr. Sellers?

DR. SELLERS. This is actually a very

basic question, and it has to do with efficacy.

When I read the literature on efficacy, I

don't see a significant difference between the oral

formulation and the topical formulation. And I was

wondering if you could comment on that?

DR. WALKER: I can. I was--first, I guess,

you want to phrase things that it's very difficult

to compare study to study. This assessment is the

same, but the population was very different.

267

The patients who came into the topical

trial did not need to be as severe as those that

came into the oral trial. Their average body

surface area was--around 10?--7 to 8 percent body

surface area. Their disease was less.

We also had, for our criteria for

efficacy, not a "none" or "minimal," but actually a

"mild," "none" or "minimal." And those trials,

very few patients actually made it to none or

minimal. And I'm not sure--in Dr. Cook's analysis,

I think you might not have meant "none" or

"minimal," but "mild." Because when you look at

our label for "none" or "minimal" the numbers are

actually smaller than what was presented today.

So I think it's difficult to compare. I

think large body surface areas are difficult. The

drug works, I should say, very well topically. But

it's really a different patient population.

Also, the topical drug is not appropriate

for intertrigenous areas because of erythema,

scaling and pruritus. It's difficult to put a gel

or a cream in the scalp. It doesn't help nails.

268

I will say although we didn't see nail

changes in the 12-week study, we did see positive

effects in the one-year study. You wouldn't expect

to see nail changes in a 12-week study based upon

the growth rate of nails.

So, there were other effects that you can

get long-term that you wouldn't get with a topical.

DR. STERN: Dr. Epps?

DR. EPPS: I have a question--just to

change organ systems--about the thyroid issue.

Were any of the patients symptomatic? What was the

outcome of some of those patients' alterations?

DR. WALKER: I'd like to share that data

with you. I think it's a complicated area, and

when you look at the data, it just shows you how

much variation, also, you get with thyroid within a

patient population.

And Dr. Beddingfield is going to share

that with us.

[Pause.]

DR. BEDDINGFIELD: Could I have slide

S-188, please? Slide up, please?

269

[Slide.]

This is the data that we have on thyroid

disease, and thyroid labs from all three studies;

the placebo-controlled trials, the six-month trial,

and the one-year trial.

And this is adverse events. So this is

hypothyroidism, as reported as an adverse even in

the placebo-controlled trial. There's no

difference between the tazarotene group and the

placebo group.

We also looked at patients with

percentages of TSH that were abnormal. And you can

see there's no difference between the two groups.

Thyroxin level--no significant difference

between the two groups--a trend for the placebo

group to have a higher rate of abnormalities.

And the similar pattern is seen throughout

the trial. You do have, here, a slightly higher

rate in the long-term treatment with tazarotene,

versus the 12-week treatment. This was not

statistically analyzed, but you can see the rates

of TSH abnormalities and thyroxin abnormalities

270

follow no consistent patterns there.

And then in the long-term study, comparing

the first six months of treatment to the second six

months, you don't see a spike up in the second six

months. And overall, the rate of adverse events

for hypothyroidism is quite low, and the comparable

rates of abnormal TSHs and thyroxine.

So, I really do not think we've seen a

signal here at all with respect to thyroid. And,

of course, this is quite different than what we've

seen with other retinoids.

DR. LEBWOHL: I just wanted to comment--the

reason that we looked at hypothyroidism in these

patients is because of a known retinoid effect in

causing central hypothyroidism, which is different

than what we're seeing here. With Targretin or

bexarotine, you see a drop in TSH, which then leads

to hypothyroidism, and the TSH remains low. So the

elevation of TSH is not a retinoid effect that

we're used to.

DR. STERN: Dr. Honein?

DR. HONEIN: Yes, I have a question about

271

any predictions that FDA olor the sponsor has about

off-label use, since in the trends you presented

for the topical version of this drug,

three-quarters of it is for acne.

And what I heard presented this morning is

sort of focused on the psoriasis population being

different. But if three-quarters of this drug is

going to be used for acne patients, I think that

would need to play into what sort of risk

management program is appropriate.

DR. WALKER: Topical use for acne has been

proven to be effective. For oral use, we did do a

Phase 2 dose-ranging study, and we showed some

efficacy. Whether it's as efficacious and would

meet the criteria for approval for nodulocystic

acne is not known. We're currently not pursuing

the Phase 3 program for the acne indication.

It's, I think, somewhat of a leap of faith

to feel that 75 percent of the oral would be used

for acne, since it hasn't been proved. It

certainly--not all systemic retinoids work for

acne. Isotretinoin is very unique. Acitretin,

272

etretinate--altrans retinoic

acid--bexarotine--three other systemic retinoids,

don't work for nodulocystic acne, and haven't been

proven.

So, I think that the logic is there, but

it's not--it is somewhat of a leap of faith.

Allergan won't promote this product for

off-label use. We won't encourage off-label use.

We won't do any of those things. We will promote

the product to be used on-label, which we're

requesting to be psoriasis.

Having said that, we do have a

risk-management program that does target the

vulnerable population, irrespective of the use of

the product. So we are protecting the patients to,

really, the same degree that the isotretinoin

program is protecting them. So in the event that

it is used off label, that population would be

protected.

And the next question, of course, is how

would you track where it was used, and we would

track that through known marketing data bases, such

273

the IMS data base, or automated data claims bases.

DR. STERN: But, the FDA, in their briefing

document, in fact, gave as illustrations a number

of studies that were presented as posters, which

showed various kinds of efficacy in acne for your

product, and I know you would not promote it as

such, but one--in fact, if one looks at the oral

retinoid use, there's at least an order of

magnitude difference in the number of people

exposed to isotretinoin than all other oral

retinoids combined, with acne versus all other

indications.

And the problem, to me, is that who is to

say that even without promotion, given that you

have a product that is topically used both ways,

that the average clinician won't think, "Oh,

tazarotene--something new. Works well topically

for both indications." And I guess my concern is:

in any off-label use, unless we have evidence--you

know, this is not the usual off-label use, but

rather, we have a drug with known substantial

risks, and then we have a drug with what might be

274

the dominant use--at least if it conformed to the

overall use pattern of retinoids--being in acne.

And I don't know what the benefits are in acne.

You know, if you could present me with data that

showed the benefits are equal to isotretinoin, and

you only use it for 20 weeks, and the remissions

are the same, I'd feel pretty good about--very good

about it, in terms of approving it now, and then

letting your NDA go forward.

But let's say this is a drug that doesn't

give remissions, and people use it in a different

way for acne, with more exposures. You know, it's

not the usual--"Well, they might not have gotten

approval." This is a class of drugs we're

extremely concerned about. And the psychology for

it, based on topical use, and based on some studies

that the company must have sponsored and saw fit to

have--let their individuals present at meetings,

has an acne claim, basically. You

know--psychologically.

And before you answer that question, I'd

like to pose a related question to the committee,

275

and I'll start it by showing my own ignorance.

Is there anyone else who's a dermatologist

on this committee that knew that the labeling for

topical tazarotene asked for a pregnancy test

within two weeks of starting? That's my first

question. Was there anyone else besides me who

didn't know about that?

Okay. I guess that means I don't have to

ask the second question: how many of you routinely

do it. [Laughs.]

DR. EPPS: And no representative has ever

told me that they should.

DR. STERN: And it's a very well sampled

product in the places where I practice. So it's

not like we haven't seen the folks--which I

think--so that really concerns me, too. You know--

DR. WALKER: Well, there's two very

complicated questions--or really more ideas that

that you've placed out there.

I'm going to start with the idea of the

acne, and the oral tazarotene.

I think you have to keep focused on the

276

fact that oral tazarotene works very well for

moderate to severe psoriasis. The acne issue is

out there. It is a retinoid. You can make that

leap of faith. But really, to restrict a drug for

severe psoriasis patients because you're concerned

about off-label use, I think is a bit

inappropriate. And the drug does work for

psoriasis, and that's why we're here today. I want

to remind you. I want you to think about what the

patients said that were here, and to keep focused

on all the work that we've done in the psoriasis

indication.

We have done some Phase 2 work for acne.

We presented that Phase 2 work at meetings. It

does suggest some efficacy. Whether it's close to

Accutane; whether you don't have relapse like with

Accutane, we haven't demonstrated.

You know, I don't think it's going to go

off and be an Accutane in the first year. We would

follow all that. If modifications needed to be

made because there was vast, you know, off-label

use, then I think that would be something that

277

could be discussed with this committee, with the

agency. But, you know, I really almost want to be

you: don't forget what we're really looking at.

Don't forget those pictures of those patients.

You know, this is for psoriasis. It does

work. We do have a very rigorous risk-management

program in place--or that we are going to put in

place, which is--it has all the components that he

isotretinoin program has. So we are tracking those

patients. We are protecting those patients,

because--we appreciate your concern, but we don't

think this should be restricted until we prove or

disprove that it works for a separate indication.

The other question that's out there is the

topical. It is clearly labeled. We do have all

our advertisements that have it. And the reason

you probably don't worry about it is that it isn't

a real risk. It is a very, very low systemic

absorption. We've had, I think, eight pregnancies.

I could show you that data. There's been no

retinoid-related effects.

If you look at the serum concentrations

278

with topical administration to the face for acne,

relative to the teratogenic levels, they're very,

very low. So, although it is a risk, the company

supports the X label. We don't advertise against

the X label. It's in all of our literature.

That--you know, the company isn't saying this, but

I think the fact that you're not that aware of it

is because it's not a real significant risk.

We aren't saying that with the oral form.

We're saying it is a significant risk. It is a

probably teratogen. And we are willing to go

beyond what any the retinoids out there are doing

right now. We're going beyond what the competitive

retinoid is for psoriasis, and we're going beyond

what bexarotine is, which is for cutaneous t-cell

lymphoma.

And just keep your focus--I'd like to say

on psoriasis and on the patients who need this

drug.

DR. STERN: Dr. Sellers.

DR. SELLERS: This issue is somewhat

problematic, though, because, in fact, all you

279

really need are the data from a Phase 2 trial to

establish your off-label market before a drug gets

approved for something else. And, in posters that

were cited were discussing efficacy, were

discussing safety, that were based on trials that

may not capture populations at risk for pregnancy

exposures, for some of the adverse events that we

discussed today.

So, I think although the company will not

be--quote-unquote--"promoting" the use, it's

already out there. And there's no way we'll be

able to control it, unless we continue with trials.

DR. WALKER: I will say that we did do

safety monitoring in that population, because the

risk of pregnancy, I really think we are covering.

But if you think about bone mineral

density, the x-rays--we did extensive studies in

that population. We looked at epiphyseal plate

closure, we looked at bone density, we looked at

osteophyte formation. We also did urinary markers

for bone resorption and absorption, as well as

fractionating--in that case, the alkaline

280

phosphatase. And we didn't see anything. That was

a six-month treatment.

So, for short-term treatments like what

you have with isotretinoin, you don't see the same

adverse event profile that you may see out beyond a

year.

DR. STERN: But you did see three out of

either 84 or 86 women enrolled in your clinical

trials become pregnant--with, presumably, since

these trials were relatively recent--presumably a

company risk-management strategy for the management

of the Phase 3 trials.

So--you know, that's not a great number.

DR. WALKER: The trial--although it has

essentially--it has a mandatory registration, since

the patients are in the trial--there wasn't the

extensive patient education. This was done several

years ago. There wasn't the mandatory patient

education. And then the reaction if the patient

didn't demonstrate appropriate education, which is

now being employed in the post-marketing of oral

tazarotene

281

So it is somewhat different, although I

agree, there were pregnancies and that concerned

us, and that's partly why we've modified our

program.

DR. STERN: Dr. Katz.

DR. KATZ: First I have a couple brief

comments.

Dr. Epps asked the question: how does the

decreased bone mineral density compare to what

would normally be found? In the FDA presentation

we were told less than 1 percent in males per year,

and here all the trends are greater than that.

Another comment to Dr. Sellers, when she

asked Dr. Walker whether the comparison with

topical and oral was similar. I would like to

emphasize, again: these were not double-blind

studies. And for the non-dermatologists around the

panel: topical, tazarotene, it gives irritation.

So it's unblinded immediately.

Nevertheless, all the studies you see are

entitled "double-blind" studies. You have to start

as double-blind studies, but anybody knows that

282

they weren't double-blind studies.

The other thing I would take issue

with--and there could be difference of opinion on

this--Dr. Walker, I don't know if you actually

treat patients--with the Tazorac, topically, but

you repeatedly state "it's very effective in

psoriasis." And that's all I do. I teach a half a

day, and otherwise I take care of patients, many of

whom have psoriasis. And it's not very effective

topically. You know, it may be in the rare person.

I haven't seen them in 10 years. But it may be.

But it's certainly not very effective.

And soon after it came out they were

suggesting its topical use with--I don't want to be

argumentative now, but topical use with topical

steroids. And it is very effective with the use of

high potency topical steroids, which work by

themselves.

Now, to get off from that, my own comments

are that these bone--not only bone density that

bothered me, but it's all in the same direction:

elevation of alkaline phosphatase, more retinoid

283

musculoskeletal symptoms. And it's all in that

direction. And it's only in a 52-week study. And

you worry about these patients long term.

It's unlike Accutane, where we treat

patients for 20 weeks or a little longer, and a

small number require--well, not so small number

require re-treatment, but it's for another 20

weeks. It's not with continued use.

So you really worry about an effect like

that.

And we can't use the argument that we have

20 years of experience with it, because we don't.

This is a different drug. It's a unique drug. It

only produces cheilitis in 65 percent of patients,

whereas Accutane, it's 100 percent of patients.

And so its unique effect on bone is not astounding.

And, in fact--in disagreeing with a colleague--in

the rare instances where we use Accutane long

term--which is rare, like in Darrier's disease--we

do check bone density after a year. We are very

concerned about that.

But this appears to have even more of

284

those effects.

DR. WALKER: I'd just like to comment.

Number one, I've learned about the

double-blind from you, and I made a not during

lunch. So thank you. I think you're absolutely

correct. I've been accused before having that

"very" in there, and I need to strike those from my

vocabulary. So that's another friendly reminder.

And also I think your point about the

bone--we understand what you're saying. I do feel

that it is within the range of what is reported in

the literature for the other products, but I

understand what you're saying and we appreciate

your opinions. Thank you.

DR. STERN: Dr. Day.

DR. DAY: The proposed brand name for the

oral is "Tazoral?" Is that the way you say it?

DR. WALKER: "Taz-oral."

DR. DAY: Well, I've gone around and asked

a bunch of people, and I've gotten lots of

different pronunciations, even from dermatologists.

So that could be a problem.

285

But there is a precedent with Tazorac out

there. Have I said that correctly?

DR. WALKER: You have said that correctly.

DR. DAY: All right. When you write both

of those--and if you happen to write in block

letters, the "C" and the "L" at the end could look

very similar.

The Drug Safety and Risk Management

Advisory Committee has been concerned with

confusion in drug names. And it is the same active

ingredient and so forth, but since there are

different indications for the oral and the

topical--and also different effectivenesses for the

different types of psoriasis and location for

psoriasis, could you comment on the implications of

a patient getting one, as opposed to the other, and

vice versa?

DR. WALKER: Yes, I think you bring up a

very important point. I think Tazorac is easy to

say, partly because a lot of us are used to saying.

The Tazoral--you know, I think that that has to go

through market testing and to see if patients

286

understand it, if they can distinguish it, if--you

know, there's specific testing to do for writing

it.

And, to be honest with you, I don't know

how much of that's been done. If Tazoral is not

the perfect name, then I think we can always work

on another name.

I feel it's important to have separate

names for the separate products, for a lot of the

reasons that have already been voiced here today.

Tazorac means acne to people. I mean, everyone

around the room says "Tazorac, acne--effective for

acne."

If you have a separate name that is

separated by the indication and by the severity of

the disease, I think it may help--although it is

the same active ingredient--I think a separate name

can help distinguish a very separate safety and

efficacy profile that this drug has.

DR. DAY: So, do I understand that the

market testing on this is not completed, in terms

of multiple pronunciations? I mean, I've gotten

287

"tayzer-ell," and all--

DR. WALKER: I will have to--I'm going to

ask one of my colleagues, because I am not in the

marketing group, I'm in the R&D group. So I'm

going to ask--and I'll answer that in just a

moment.

[Pause.]

It was initially tested. But because of

the comments and feedback that we got actually just

a couple of weeks ago, we are re-testing that to

look at it, taking into account what you've said.

DR. STERN: Dr. Gardner? And I hope I

pronounced your name correctly.

[Laughter.]

DR. GARDNER: Yes, you did, Dr. Sterm.

Thank you.

[Laughter.]

You know, I'm becoming increasingly

confused about what the risk-management program is.

Because as much as I have to confess to resisting

Dr. Walker's lecturing the committee about what we

ought to be focused on, the fact remains that if

288

you were to put a comprehensive risk-management

plan the way I think it's been described here, with

anybody who prescribes the drug has to be

registered; and anyone who gets it has to be

registered; and anyone who dispenses it has to be

registered--then it seems to me it almost doesn't

matter whether it's being prescribed for psoriasis

or acne. We would pick up on it if, in fact, the

risk-management plan--you can't get it unless you

go through this plan.

And even if that's true, then Dr. Day's

most recent comment is very, very important.

Because if someone is restricting Tazoral in this

way, and protecting everyone from harm--or,

theoretically, by it. Then if a prescription comes

through for Tazorac, and is mis-filled, then all

that protection goes out the window, because

someone mis-read what was written on a

prescription.

And so I guess I have two points: is the

plan--the risk-management plan as we understand

it--supposed to be comprehensive, regardless of

289

what the prescriber is thinking the indication is;

and, two, how to protect against a mis-reading of a

written prescription?

DR. WALKER: I want to thank you. I think

you've said what I tried to say earlier more

eloquently and efficiently.

Yes, the program--that's why we want a

separate name, to keep it separate. The

risk-management program would protect all

vulnerable patients regardless of the indication.

The indication could be tracked through marketing

data bases. So you could track it. All those

patients would be protected. And I think a name

helps with confusion at the pharmacy. It actually

helps the patients who need topical Tazorac don't

have that restricted because they get confused with

the program.

So what you've said is actually what the

company agrees on.

DR. GARDNER: One more thing--if you're

really registering everyone, then you shouldn't

have to track it through IMS data bases, because

290

you ought to have total coverage, shouldn't you?

DR. WALKER: Mm-hmm. Well, that doesn't do

indication, per se. Indication is not tracked as

part of the proposed registry for isotretinoin or

for the oral tazarotene formulation.

I have some other folks up here who'd like

to make a comment.

DR. KRUEGER: I'm Jerry Krueger, University

of Utah. Just on the Tazorac oral and the Tazorac

topical--you can't have a prescription be complete

if you just put down "Tazorac." The pharmacy will

call you up and ask you, "Do you want the gel or

the cream?" "Do you want .1 percent or .03

percent?"

So I don't quite see room for confusion.

DR. DAY: It can go both ways. It could be

"Tazoral--"

Dr. KRUEGER: So,

DR. DAY: --to Tazorac and vice versa.

And, furthermore, in electronic scrips, you might

say, "Well, then you don't have to read

handwriting." But there's a lot of scrolling down

291

and tapping, and you can mis-tap.

So this is a potential thing that is very

likely to happen. And we just need to think about

what the consequences would be, going in both

directions.

DR. KRUEGER: Yes, I can see some real

concern going in one direction. I don't see much

concern going the other. Thank you.

DR. STERN: Dr. Levin.

DR. LEVIN: A couple of questions for the

sponsor, and then back to the FDA, and where we are

with risk-management programs, because I think

that's key to how this afternoon proceeds.

Question to the sponsor: why aren't

you--or would you pursue a Phase 3 trial about the

use of this drug for acne, which would perhaps

answer the questions about whether this drug is

safety and efficacy for acne use, and might

reassure people who are concerned about its

off-label use. So that's a question. Don't you

see this as an issue, and if it is an issue, are

you willing to pursue it in the right way--in my

292

opinion--which is to go for an approved indication

using the approval process.

DR. WALKER: It's been a very difficult

process. We have met numerous times with the

agency on this. And I'm looking now at Khalyani

Bhatt, because she can verify we've asked her for

numerous meetings on this.

It is a very difficult trial to do. We

have had discussions--and maybe Dr. Wilkin wants to

add to what I'm saying--what we've been asked to do

is do a head-to-head comparison with isotretinoin

for five to six months, and then follow the

patients out for one year.

In order to have the power to do that,

that becomes an incredibly large study, if you add

x-rays and many things. It almost--and the

parameters that we've gone back and forth that we

need to do--becomes a study that is so onerous that

the company is not sure--with a lot of risk knowing

whether the drug works or not, based upon our Phase

2 data. You always take your data and you

extrapolate as to whether or not you feel you can

293

win.

At this point, we don't feel ready to do a

Phase 3. We don't feel that the Phase 3 designs

that we have discussed back and forth with the

agency--not saying what they've asked us to do is

wrong, just saying what we can accomplish as a, you

know, mid to small pharmaceutical company for a

dermatologic indication, that we can actually do

that.

So it's become a feasibility, economic,

and a very difficult study to do; to put all the

bells and whistles. I mean, you know, if you

imagine everything that's been in these trials; you

imagine every controversy that surrounds

isotretinoin, and you try to put that in a study to

look at comparability--which are much larger

studies; you multiply that times two, and you take

that out two years--because you're, at minimum, six

months enrollment--six to nine months, enrollment;

you've got another six-month trial; you've got a

year beyond that, follow-up, to look for relapse.

You need to win on relapse rate, not equivalence at

294

six months.

That just becomes a study that we're not

sure we could ever do.

DR. LEVIN: Okay. Another question: you

look at the Accutane risk-management program as

sort of the ceiling, and I would suggest you might

want to look at it as a floor. And let's go back

to the indication issue.

For example, one of the things that

perhaps could be built in here as a requirement for

ICD code-9s to be submitted, and actually not

filling a prescription where the coding--if ICD9

code's the appropriate code--isn't the appropriate

indication. I mean, that's a new way--that's a new

way of looking at risk management.

But, again, I don't think Accutane is

necessarily the ceiling here in how we manage risk.

It might be a floor or a mid-way point, and we

could add to it.

So, it's just a question--

DR. WALKER: Yes.

DR. LEVIN: If this is an issue that people

295

are concerned about, rather than relying on

marketing data, why not try to--and, again, I don't

know if this is, you know, feasible or not, but to

think about, talk about, building in a requirement

that patients only get through the system if they

have an appropriate indication that is entered into

the system, as an ICD9 code or whatever.

DR. WALKER: It's certainly feasible. I

think, for the practice of medicine in general,

it's not desirable. And I don't know if any

clinicians, either on the panel or in the room want

to comment on it.

You know, in dermatology--and I can see,

you know, that is--if it was determined to be such

a risk, it's certainly something that could be

applied. However, in dermatology, I know--I am, I

was asked before--I do still practice, but I think

in the opposite ratio of Dr. Katz in terms of time

spent doing research versus seeing patients.

In dermatology, a lot of our drugs don't

have approvals that are indicated. And I think

that doctors use drugs for many uses. So we

296

wouldn't--you know, I can't talk out of both sides

of my mouth--

DR. LEVIN: This isn't every drug. And one

of the challenges we have when we discuss a drug

which has some unusual risk profile is that we're

always operating with a knowledge that there is

diagnostic creep; and that a drug where we're

weighing the benefit and risk of toxicities versus

the benefit--and nobody is--you know, no one is

against providing relief to people who are

suffering some psoriasis that is refractory to

other treatments. Nobody's saying that.

But we have this equation we have to go

through.

Now, one part of the equation is very

slippery, because it's called--you know, the

diagnosis, or the indication for the drug. And we

sit here knowing that it's going to get used for

lots of other things besides that part of the scale

that we're sort of judging against the risk part.

So--

DR. WALKER: I agree. I think that you've

297

proposed something that could be considered. I'd

be interested in other people's opinions. It's

hard to argue that it shouldn't be considered.

DR. STERN: Might I suggest that there's at

least, in the PDR, one example of a drug where for

certain in what was once the dominant use of the

drug now has a black-box warning. If you look at

Allopurinol, it says very explicitly, "This drug is

not for the treatment of asymptomatic

hyperuricemia," the reason being, in fact, an

extraordinarily low-risk of Stevens-Johnson system

and hypersensitivity syndrome in association with

the drug.

And I guess that goes to my point: as

opposed to what I've heard is, we're not sure

enough that we can win the battle against Accutane

with respect to efficacy--and efficacy, to me,

means benefit in the acne indication. Maybe what

we need to hear from the company--I don't like

things that say, "Oh, let's restrict use." I'd

like to hear more from the company about how we're

going to educate physicians about, in fact, making

298

sure that they understand the proper use in areas

where, because of the risks of the drug, it's not a

good idea to use it, as opposed to a label and

things.

I'm wondering how willing--and how we

would then monitor, in fact, the follow-through of

the company in terms of making sure people

understand where we might be confident the

benefit-risk ratio is an appropriate one.

DR. WALKER: We plan to do that through two

ways--if you could bring this first slide up?

[Slide.]

Okay. We have designed a "What a

prescriber needs to know" brochure. We also have a

prescriber introduction letter; a prescriber

certification test--and that test will be part of

the same system of you prove that the prescriber

understands, just like you prove the patient

understands their risk, you have to prove that the

prescriber understands the risks and limitations of

the drug, as well as having a medication guide.

We also will have scientific meetings to

299

certify physicians to be into this registry. They

can't be in the system, they can't prescribe the

drug unless they've gone through the process.

So those physicians that go through the

process, part of that process will be educating

them that this is for psoriasis, what the labeled

usage are, what the risks of the drug are, as well

as the pregnancy risks. So all risks will be

discussed.

Now, Dr. Andrews would like to kind of go

through the flow chart of the risk minimization

action plan.

DR. ANDREWS: Thank you--

DR. STERN: I think we better not do that

now. That was not really responsive to my

question.

It's not "what's the usual stuff?" I was

sort of hoping to hear something new and different

that might really work. You know, these kinds of

things are things that the pre-1988 Accutane

interventions. I happen to believe that the 1988

Accutane interventions were better than we heard

300

today, but that's strictly editorial.

So I was hoping to hear something

imaginative, impactful, innovative and with, on

might think, a reasonable chance of success. And I

know--I could see Dr. Furberg having the same kind

of reaction to this

DR. FURBERG: I agree with you. And I

think that we need also some way of going after

people who are not complying. There should be--a

stick there, somewhere--not just pleading. Because

if you ignore this, there's no consequence.

DR. WALKER: Well--

DR. FURBERG: We need to set up a system so

that people do comply.

DR. WALKER: Well, I think I'm hearing two

things: one--I mean, I've got some things up there:

"What would be the roll out?" I don't think that

answers your question. There's nothing really

innovative up there.

I would love to hear--and we're very open

to any kind of innovative feedback that this group

wants to give us here, or in private.

301

But what you're saying about

consequences--this system's a little different than

what exists now, in that you register the

physicians. There is a feedback as to whether they

understand the system, and there's feedbacks that

the patients comply. There's feedbacks at the

pharmacy. And there's checks and stops all through

the entire system.

I won't go through it in detail, because I

think a lot of you already kind of know what that

flow chart is. But it's different in that the

consequences--if you don't admit that you

understand, you don't follow the system, your

patient won't get the drug. It will not be

dispensed to you.

DR. STERN: Dr. Honein.

DR. HONEIN: Yes, I just wanted to add a

suggestion to what Dr. Levin said: that we could

add indication to the registry. And I think we

suggested that in February for the isotretinoin

registry, not as a restriction, that if you didn't

enter the right code you don't get the drug, but as

302

a way to evaluate how the drug is being used over

time, and give us better information in the future.

And the second--to reply to Dr. Garner--I

think a major difference in what they're proposing

for risk management is that males and females who

are not of reproductive potential would be allowed

to have refills of this drug. And they're basing

that on the psoriasis population. And I think that

sort of deviation is not justified if the

population is going to be largely acne patients.

DR. STERN: Dr. Schmidt.

DR. SCHMIDT: I don't want anybody to feel

like I'm a diagnostic creep--

[Laughter.]

--but, you know, I really feel like we

shouldn't dictate how to practice medicine. And I

think a lot of these medications are going to be

used, you know, off-label.

And I, for one, at least think that

with--of course, you all also know my wife has

really bad psoriasis, and is on methotrexate and

Embrel for it. So I live with a women, you know,

303

who has psoriasis and still love her.

But I wanted to just say that I think we

ought to increase the amount that people don't need

prescriptions for this in males. I think seeing

somebody with psoriasis, you know, every couple of

months is maybe too much.

And then the other thing--and this may not

be the appropriate time for this, but I think we

need to think about this--is I want to make a

couple of clinical reflections on retinoids, and

how at least I treat people clinically, is I don't

think the retinoids are the greatest thing in the

world--you know, either Soriatane or Accutane, for

plaque-type psoriasis. I think the best place to

use the retinoids in psoriasis is the pustular-type

psoriasis, or the acropustulosis, on the palms and

soles, which can be devastating, painful, terrible,

horrible. And they work beautifully.

But you don't ever leave anybody on

anything, you know, for a long time. And when you

do, the beautiful thing about the--at least the

retinoids we have now, is you can decrease the dose

304

to where I have some patients who take one 10mg

Soriatane every three weeks, and it keeps them

clear.

So, a lot of these things you're not going

to be giving them a real high dose, no matter what

you're going to give them.

And then the other thing is, just like

with my wife, you rotate people off, like

methotrexate onto Embrel, onto something else, so

then you don't have to worry about, you know, these

side effects so much.

DR. STERN: Dr. Wilkerson?

DR. WILKERSON: AS a practicing

dermatologist, I would plead--although it sounds

like we've been siderailed by the patent office--in

terms of having what I would call the

"pan-retinoid" form. In other words, you know,

right now, as it potentially stands, we can be

scurrying to find five, six, seven, eight different

forms for the particular drug and/or generic

manufacturer, and/or sponsor making a drug. And

since the side effects and the requirements are so

305

similar, a unified form would simplify, and I think

would also increase compliance with things.

As far as the two to three month

additional refills, I've sat here and debated that

in my mind, also. As much as you want to see that

from the standpoint of the patient's lack of not

having to interact with the physician, it also yet

provides another point of error at the distribution

point, in terms of who gets--I mean, not to use any

gender or similar names, but there are names and

things where people could get refills that are

females because of, you know, because the wrong box

is checked, or an assumption's based upon a name

that would result in females getting it.

And I think just like with Accutane, where

it's monthly. It doesn't mean they have to have an

office visit, but they have to pick up that

prescription. To have two different systems, I

think, is just introducing a point of error in the

system well at the top which negates everything

else that we're trying to do here, which is to

reduce the rate of pregnancies.

306

DR. WALKER: I'd like to say that I think

you've highlighted a lot of the controversy very

well in what you've said.

As far as, though, distinguishing males

and females, you wouldn't need to do it by the

name. We've toyed with many different things.

Certainly, the sticker program has you--you know,

the physician fill out whether it's a male or

female. We've toyed with a specific prescription

pad that has male or females, and has, you know,

basically like a gray box on it for females, which

has the "no refills," and males are a different

color, so it is clearly outlined.

But the pharmacist would not distinguish

male or female by name, but rather by what is

checked within the system.

Of course, a female of childbearing

potential would also have the trigger of the

pregnancy test, which is connected with her name.

And each patient has a unique identified, which

would force the pharmacist, the

physician--everyone--to use that identifier and

307

check whether the pregnancy test has been ordered.

So there are safeguards for that aspect of

it.

The one-month versus three-months versus

more--it's very difficult--trust me, I have had the

same thought--one system. The trouble with one

system is they're not one patient type.

DR. WILKERSON: Right.

DR. WALKER: And this is a lifelong

disease. I think you're going to take patients

with psoriasis and you're going to put them on

methotrexate, cyclosporine, acitretin or other

biologics, because they can't come in--or afford

it. The health care system can't afford for

chronic cases to have them in there every month.

It's very expensive.

DR. WILKERSON: Well, I mean, not to micro

manage this down to that point, but, you know,

there are ways around that. What I'm just saying,

if you are designing a system, the less variability

there is at each decision point, the less likely

error is to slip in. And error will slip in,

308

regardless. But, just to make that easier.

And my other comment was: pharmacokinetics

on this drug--I can't believe that everyone in this

room, sitting here, has the same number of retinoid

receptors in their bodies. I mean, do we all have

the same number of retinoid receptors, and saturate

at the same pharmacodynamics--

DR. WALKER: Well, there's a range from

seven to 12 hours. So, no, everybody doesn't

metabolize the same way, but that--

DR. WILKERSON: No, I'm not talking about

metabolism, but in terms of once the receptor is

bound, and there's a signal sent, does everyone

have the same--in other words, it seems that

picking a 4.5 mg dose is rather simplistic when we

have 100-pound individuals versus 350-pound

individuals. Their volumes of distribution and

however else affects the pharmacodynamics of the

drug, and what this leads up to is perhaps, have

you correlated the weight of the individuals

against your bone data? In other words, you know,

are you seeing increased side effects in lower body

309

mass individuals perhaps?

DR. WALKER: We separated the bone--I'm

want to kind of go from your last question and move

forward. WE separated the bone data by gender and

age, and we didn't see anything. But I don't know

that we did it by body weight. They're shaking

their heads no.

DR. WILKERSON: What you were trying to

tell us before is there is no dose response.

DR. WALKER: Right. I am. And I'm going

to show you--

DR. WILKERSON: --[inaudible] defies the

laws of pharmacology.

DR. WALKER: I don't think it doesn't go

with the laws of pharmacology. What it says is the

drug is not lipophilic and it's not stored in fat;

that it's actually in the plasma volume.

And I'm going to have my

pharmacokineticist. He's got some very nice slides

that I think will help you see this.

It is not changed by weight.

DR. YU: I'm Dale Yu, pharmacokinetics.

310

What we've done is we looked at the

systemic drug exposure of tazarotenic acid, the

active ingredient, as a function of body weight in

the Phase 3 trials. And I will show you that data

now.

Slide up, please.

[Slide.]

Okay, if I can focus your attention on the

right panel of this plot, these are data from the

two Phase 3 trials. We looked at drug

concentrations from about 80 patients. The body

weight ranged from about 50 to 150 kg, and the

concentration on the vertical access, as you can

see, over a wide range.

What we were looking for is some kind of

trend, if there is a relationship between body

weight and concentration, we should see either

increasing or decreasing trend. And we were not

able to see that.

So our conclusion that the drug

concentration does not change as a function of body

weight, therefore we don't need to consider dosing

311

by body weight.

DR. WILKERSON: Well, I mean, you're just

looking at plasma concentrations here, right?

DR. YU: That's one of the things we looked

at.

DR. WILKERSON: But you're not looking at

efficacy--endpoint efficacy and expression of the

disease as a result of your pharmacologic action.

DR. LU: So your question is whether we

looked at efficacy as a function of body weight.

DR. WILKERSON: Well, dose response. I

mean, it can be this, but it can also be what dose

does it take to effect a particular response on the

disease process.

DR. WALKER: we do have that data, and

we'll share it with you.

DR. LU: John Lu, biostatistics, Allergan.

Slide up.

[Slide.]

A logistic regression was performed on

data combining 048P and 049P study. The dependent

variable was clinical success. And I regress these

312

covariates on clinical success. And we didn't find

that weight was a predictor of clinical success.

DR. STERN: And what was the fit of the

model?

DR. LUE: Umm--I didn't find a significant

difference. I felt it fit pretty good.

DR. STERN: But, we need to know how well,

in fact, each of these variables have any

individual relationship. So a lack of correlation,

in the absence of a model that has some predictive

value, either says that you're looking at all

randomness, or you've not chosen things in a way

that, in fact, you have the right independent

variables for the right dependent variables.

So I think--and your data sets--well, I

just think it's hard to interpret.

But we'd better stop, in terms of one

more-- DR. WILKERSON: One more comment. We've

heard all day that this drug is like all the other

retinoids. Well, all the other retinoids show a

dose response curve.

DR. WALKER: Well, that's--yes. That's

313

because of the distribution. The distribute out of

the plasma into the body fat.

DR. WILKERSON: Okay.

DR. WALKER: And tazarotene doesn't do

that. It is, you know, a--

DR. WILKERSON: Here today, gone

tomorrow--right?

DR. WALKER: Did you want to comment, Dr.

Helms?

DR. HELMS: Just a point on the model,

there.

Your point is a good point, but if you

look at the significance level for the treatment,

and the other factors that are significant, you see

that the model really will pick up something that's

important.

DR. WALKER: Any other questions?

DR. STERN: Dr. Ringel--and then we're

going to take a break.

DR. RINGEL: Okay, I'm going to give poor

Dr. Walker a break. This is going to be addressed

to Dr. Trontell. And I promise there really be a

314

question at the end of my little speech here.

[Laughter.]

I really feel as if we're just taking one

more step in opening up, you know, the Pandora's

box of teratogenicity here. With isotretinoin, you

give it to people for five months, and it has a

spectacular effect.

For acitretin, I think very few physicians

actually give it to women of childbearing

potential, and that's probably why we've been saved

from much of the teratogenicity of that drug.

On the other hand, we're talking about a

drug here--tazarotene--which is going to be used

indefinitely by women of childbearing potential.

And that seems very frightening. I don't know how

many women can use two forms of birth control and

guarantee forever that they will not be pregnant.

I think that's a very, very difficult task for

anyone.

On the other hand, you know, if this were

a really star-quality drug; if this were another

Accutane for psoriasis, and it were just wonderful,

315

you know, then maybe it would be worth it. But

it's not. At least not if it's like the other

retinoids that I know.

And on the other hand--a lot of hands

here--

[Laughter.]

--teratogenicity is very serious. It's a,

you know, extremely serious problem. So, you know,

you've got a risk-benefit issue.

I think Gloria Steinem--if she's here,

please cover your ears--I'm wondering--here's my

question: can we restrict this drug to males, and

women who are not of childbearing potential? I

think that that would solve a bunch of problems. I

know it's not politically correct, but I feel that

for the women and their future children, perhaps,

you know, we need to take different steps for a

different population.

There are drugs, for example--I mean, you

know, there are things like Propecia that are

indicated only for men. I mean, it has been done

in the past. Would another risk-management option

316

be to restrict this to males and women who cannot

have children?

DR. TRONTELL: Yeah, it's a tough question

to think about the mechanism, how you would go

about doing that.

If you tried to engineer some restricted

distribution that would somehow knock out those

individuals who are of childbearing potential,

you'd presumably enter into a lot of complexities

making that determination; is that an issue of, you

know, anatomy and physiology or of behavior that

defines your reproductive potential.

And I think the challenge gets at what was

discussed by the committee earlier about our

ability--or authority--to speak to off-label use of

medication. So there are some products that are

indicated for treatment of one gender versus the

other but, in fact, that doesn't prohibit

clinicians following reasonable practice patterns,

and a belief that usual and customary--forgive me

if not using the correct legal language--to employ

them.

317

So I'm thinking, in practice it might be

quite difficult to do what you suggest. In very

exceptional circumstances, where there may be a

unique indication, where the risks may warrant the

benefits to the patient using the drug, has FDA

ever even attempted to speak to the specific

indication for which a product should be used

safely.

DR. WALKER: Can I make one comment

before--

DR. STERN: Dr. Bull was going to--

DR. BULL: you know, I think you have to

go, also to--we try to make decisions based on

data. And I haven't heard the committee comment on

what I think is a disparity in the data base, and

Dr. Cook's earlier presentation. It may have also

been said in Allergan's--the disease has an

incidence that does not reflect a male

preponderance of the disease, but you're looking at

studies that had about 80 percent inclusion of

males.

Now, I don't know why that specifically

318

happened. But, in terms of the adequacy of the

data to speak to female patients is also, I think,

an issue we would welcome comments from the

committee based on, you know, just what you see

with regard to the demographics in the studies that

have been submitted.

DR. STERN: I take the liberty of

addressing that, as someone who's sort of followed

clinical research in psoriasis for a couple, three

decades.

And if you look at systemic agents,

phototherapy, PUVA, and you look at the clinical

trials published, one can predict that they'll be

approximately two-thirds male, in terms of

enrollment, across the United States and Europe,

and they'll have a mean age of between 44 and 48.

So, whatever it is that goes into who gets

enrolled, and trials of systemic agents,

phototherapy and PUVA, over the last 29 years,

there's a remarkable consistency about the profile

of those individuals in psoriasis trials, including

a fair number that are not part of NDA submissions,

319

but are just various kinds of clinicals done in

academic or other institutions.

So, I can't explain it, but it's sure

there.

Wilkin?

DR. WILKIN: I'd just like to comment on

Dr. Walker's earlier statement about actions with

FDA for the acne indication. And I think you

represented that quite fairly.

WE were ultimately interested in a

product--if it's a systemic retinoid and a

teratogen--that it not be inferior, efficacy-wise,

to isotretinoin. And the committee has picked up

on some of our concern that was in our portion of

the briefing document that went to the committee.

We referred to the posters from Allergan, presented

at the American Academy of Dermatology, and also

some of the articles that show up in the throw--the

journals that come for free--

[Laughter.]

DR. STERN: I think we call them

"non-peer-reviewed."

320

DR. WILKIN: Non peer-reviewed. That's

better than what I was actually going to say.

[Laughter.]

And this is actually the summary paragraph

from one of these: "Nodulocystic lesion count data

also show a statistically significant benefit for

both tazarotene 3 mg and 6 mg compared with

placebo. Allergan has completed Phase 2 studies in

the treatment of nodulocystic acne, and is

currently seeking a partner in the United States

and Europe for further clinical development."

So this is something that a lot of

dermatologists would get to read.

And I'll not read all of the posters, but

just one of them. It says: "The results in this

early Phase 2 trial show that once-daily oral

tazarotene, 3 mg or 6 mg, reduces the numbers of

both non-inflammatory and inflammatory lesions. It

efficacy against commidones is especially notable,

since these lesions are the precursors of

inflammatory lesions. Oral tazarotene could prove

to be an effective new therapy for patients with

321

acne, and further investigation is warranted."

So these are a whole series of posters

that were presented at the American Academy of

Dermatology.

DR. STERN: Thank you.

And why don't we take an eight-minute

break and come back at three o'clock.

[Off the record.]

DR. STERN: Back on the record.

We'll have one final question, from Dr.

Furberg, and then we'll move on to the questions.

Dr. Furberg.

DR. FURBERG: Well, I ha two brief comments

and a question.

One comment relates to bone mineral

density, which we are using a surrogate for

fractures. And my view is that there are no good

surrogates. And bone mineral density is certainly

not. And the fact that there was no good

relationship between bone mineral density and

fracture doesn't mean anything to me.

There are treatments that increase bone

322

density, at the same time they increase risk of

fractures. That's just an illustration of how weak

that is.

So if you're going--what I'd like to do is

to encourage you to use fractures as an outcome in

your future studies, and not rely too much on bone

density.

The second comments relates to the whole

issue about pregnancies. And I was surprised to

hear that there were 113 worldwide exposures. And

most of that information came from you. And for

107 of them, you have either "no adverse outcome"

or "unknown."

And that, to me, is terribly

disappointing; that you're lumping the two--you're

lumping the two, "no adverse outcome" and

"unknown." I mean, you should cut down the number

of--first of all, you should separate that, and the

unknown should be down to zero.

DR. WALKER: I'm not sure I understand; 102

what?

DR. FURBERG: Pregnancies.

323

DR. WALKER: We didn't have 102

pregnancies--

DR. STERN: Topical taz--that's what you

indicated.

DR. WALKER: We had eight pregnancies.

DR. STERN: No, in topical tazarotene.

DR. WALKER: You mean, yes--

DR. FURBERG: What I'm reacting to is the

fact that you are not doing more to eliminate the

unknowns in your data base. I mean, this is a

critical issue for us to know. And so I'm

expressing--this is a comment. I'm expressing my

unhappiness.

The question I have relates to

hyperglycemia. You had seven events. And I just

wonder whether the drug is causing diabetes--we

didn't hear anything about that. And, if so,

whether that should be part of the labeling.

And, second is: whether, if it has an

effect on glycemia, whether it should be used in

diabetics?

DR. WALKER: Okay.

324

DR. FURBERG: That's my question.

DR. WALKER: All right. Number one, the

pregnancies--those are spontaneous reports that

come in. WE actually track every report we can.

But sometimes they come in, not from a doctor, not

even from the person who's pregnant. So we track

them to the best of our ability, because they come

in spontaneously in the field from multiple

different sources.

So, I agree that it's unsatisfactory, but

with the systems that are in place for

surveillance, we always do the best we can do, and

we take those very seriously. But, I agree,

they're not to anyone's satisfaction.

The second point, being the

hyperglycemia--and I'd like to have a slide up so I

can address that.

[Slide.]

There were seven cases of hyperglycemia,

called an adverse event by the physician. That is

just what the physician determines to be an adverse

event. Those were in the double-blind studies--the

325

048, 049P studies. There were none in the placebo.

But what I think is more important to look

at is not the number of adverse events which were

called that by the physician, but what the

incidence of hyperglycemia was in the whole

population. Because I think it's just--the 2

percent is not really real.

If you look, the 14 percent of the

patients--or almost 15 percent--had elevations of

hyperglycemia in the tazarotene group; 18 percent

in the placebo group. So when you look across the

population, you really didn't see that.

DR. FURBERG: Well, I disagree with you on

that. I'm interested in the rare cases, where a

drug may induce diabetes. And I don't expect that

to happen in everyone. It's just a small group of

people that are susceptible. And you have seven to

zero. And I'm saying I'm not satisfied with you

classifying them as "hyperglycemia." I'd like to

know more about those. You should go back to the

case records.

Were those patients diagnosed as

326

diabetics?

DR. WALKER: They were--

DR. FURBERG: Started on treatment? That's

what I'm asking.

DR. WALKER: Actually, the patients were

diabetic. They were on treatment, often because

they come in for the labs fasting. They do not

take their hyperglycemic drugs.

The patients--if you look at the

scattergrams at all the time points, there are no

differences between the groups. And patients go up

and down.

So we didn't see any trend. We didn't see

any one-off's, we didn't see ones that popped up

high. So we did not see a signal. All we saw were

that some physicians--it turned out, seven

physicians--called the hyperglycemia an adverse

event. When you have a laboratory abnormality, the

investigator has a choice whether they call it an

adverse event or not. And in this case, they fell

into the tazarotene group.

We've looked very closely at the actual

327

lab data and don't see that.

DR. FURBERG: I wish you had presented

those seven cases so we could--we could take a look

at it.

DR. WALKER: Okay. And I don't have those

seven broken out.

DR. STERN: We're going to now end the

question period, and go on to the questions to the

committee.

So I'd like to thank Dr. Walker, because

unless there is something extremely pressing, I

think--you can now relax and enjoy--

DR. WALKER: I'll be very--you know, there

is one thing pressing. I think that there's been a

lot of discussion--

DR. STERN: I'm sorry. I'm going--I think

I'll have to take the--we only have two hours left,

and we have five questions, many of which have

subsets; some of which are just for discussion,

others of which are for a formal vote.

The first question, which is two parts,

the first part is:

328

"Based on the information from the

clinical studies conducted for tazarotene capsules,

is there adequate demonstration of effectiveness

for moderate to severe psoriasis?"

This is for discussion. And so I think

what I'd like to do is go around the table, for

people to make comments--or not, as they so

choose--as to whether they believe the clinical

studies presented here, and given to us, in fact

show that this is a--quote-unquote--"effective for

moderate to severe psoriasis."

And, perhaps--Dr. Honein, could we start

with you? And a "pass" is fine.

DR. HONEIN: I pass.

DR. STERN: Dr. Furberg?

DR. FURBERG: Yes, but effectiveness, for a

chronic condition, you don't show in 12 weeks. So

its effectiveness--I would call it short-term.

DR. KATZ: Not to parse words, like the

famous "is"--what does "is" mean?

[Laughter.]

But "effectiveness"--effectiveness or not?

329

Yes, I think it's been demonstrated that the drug

is more effective than placebo, for moderate to

severe psoriasis. Now, would you translate that to

quote somebody as saying "it's an effective drug?"

If you define effectiveness as better than placebo,

I would modify my response by saying, it's better

than placebo.

DR. STERN: That sounds like a low pass.

[Laughter.]

DR. KNUDSON: I'm going to echo what Dr.

Katz said. Yes, it is more effective than placebo,

but that's about the best I can say.

DR. STERN: Dr. Sellers?

DR. SELLERS: I have problems also with the

term "effectiveness." I would say that it showed

some efficacy. But with the background materials

that we were presented with other treatment

modalities, it does not represent a significant

advantage over currently marketed drugs.

DR. STERN: Dr. Schmidt?

DR. SCHMIDT: Yes, I agree. But--and, of

course, I'm looking at this through the filter of

330

my clinical experience. As I said before, I don't

think any of these retinoids are just real start

quality for plaque-type psoriasis. But I would

almost bet you a million dollars this is going to

be great, you know, for other forms of psoriasis.

And then the other thing that really is a

mitigating factor for me, is the short half-life

with this stuff. I mean, to me, to have a

retinoid, you know, that washes out of your system,

and you potentially can give it to younger people

for some of the sever acropustulotic-type

psoriasis, I fell like, you know, this is going to

be effective for us.

DR. STERN: But we haven't seen any data on

that.

DR. SCHMIDT: No, I said--

DR. STERN: Right.

DR. SCHMIDT: --I've got this clinical--

VOICE: Feeling?

DR. SCHMIDT: --feeling.

[Laughter.]

But I agree with what was said. What

331

we've been presented--

[Laughter.]

DR. STERN: Dr. Raimer.

DR. RAIMER: Yes, I think I have to go on

the side of being effective. If you can see at

least 20 percent of people clear or almost clear in

12 weeks, I think you have to call that--for all

the drugs we have for psoriasis, that's at least

somewhat effective.

DR. EPPS: I agree with the comments so

far. I agree I was a little disappointed with the

20 to 30 percent, but it has shown more than

placebo.

DR. HOLMBOE: I agree, I think it's mostly

an efficacy issue, not an effectiveness issue.

Also, I'd like to make a couple other points.

I think one caveat, in looking at this

data, is we don't know the reliability and validity

of this OLA tool. And that worries me a bit. And

I would have liked to have seen more information

about the tool used to score.

However, that having been said, if you

332

look at the number needed to treat, the number you

need to see effect is seven to eight. So you think

when you look at some of the things we use in other

conditions--putting aside for a moment our concerns

about teratogenicity--that's not a bad number to

treat. So I think it has shown some efficacy

related to placebo.

MS. SHAPIRO: I think Eric's last point is

compelling, and I will defer to the rest of you on

that.

DR. RINGEL: I basically agree. I think

without a valid--without an instrument that's been

validated, without knowing exactly how this OLA was

performed, and how reliable it is, it's very

difficult to say if this is an effective drug.

However, listening to the people

presenting their data, and knowing how retinoids

work in general, I would suspect that it probably

is an effective drug.

DR. STERN: I think the data suggests, at

best, modest effectiveness. In this clinical

development program of 700 patients, and we saw the

333

best photographs that--when I treat people with

narrow-band UVB, would make me wonder about whether

my lights were up to full power.

DR. GARDNER: I'll defer to my

dermatologist colleagues.

DR. WILKERSON: Yes, for efficacy. And it

fills--I think Dr. Schmidt's point is well taken,

and something that hasn't been talked a whole lot

about is that female bracket--who the other

retinoids are out of the question for--this

does--and I think that's the most compelling reason

to approve this drug is for that niche of females

who need a retinoid, who don't respond to any other

therapy, and want to get pregnant in the future.

Right now, that's really not a possibility with the

others on the market right now.

DR. DAY: Better than placebo; maybe better

but can't tell, given the current level of data.

DR. LEVIN: Yes, with all of the

limitations that have been expressed by others.

DR. STERN: Well, that was easy. [Laughs.]

Now to the second part:

334

"Is there adequate demonstration of

efficacy for 'very severe' psoriasis?"

And perhaps, Dr. Cook, can you quickly put

up your slide that broke out--just to remind people

that the proportion of individuals with "very

severe" at baseline who cleared was not

significantly, but was only in the placebo group,

is my recollection.

DR. COOK: Dr. Lee can put that up.

DR. STERN: Oh--sorry.

DR. COOK: No problem.

[Pause.]

[Slide.]

DR. LEE: Here's the subgroup results of

treatment success for the first study. You take a

look at the last row; basically break it down into

baseline, disease severity.

So the highlighted part--the very

severe--there were a total, in the first study,

only five patients enrolled. And the two patients

treated with tazarotene did not achieve treatment

success.

335

[Slide.]

This is the result for the second study.

And, again, the last row, a total of 10 patients

enrolled in this study. And four patients were

treated with tazarotene. And, again, none achieved

treatment success.

DR. STERN: Should we go around and start

with Dr. Levin?

DR. LEVIN: I would not see this as an

approved indication.

DR. DAY: And what question are you wanting

us to comment on at this time?

DR. STERN: Is there adequate demonstration

of efficacy for--quote-unquote--"very severe"

psoriasis. That's why I had that slide put back

up.

DR. DAY: Right--because the indication

is--for--I understand, but it does say "moderate

and very severe." So we're just commenting on

"very severe."

Yes.

DR. STERN: That there is demonstration of

336

efficacy for "very severe." They went 0 for 6, I

believe.

DR. DAY: I'm sorry. Why don't you

consider around this way, we'll come back.

DR. WILKERSON: yes, the data is obvious.

But I would point out that the metric here is--if

you take somebody with 90 percent psoriasis, and

they had to get almost to "clear" or almost clear,

in the real world that rarely happens. But that

patient may very well have improved, you know, 60

percent and still be happy with the results.

So, yes, based upon the data--but I don't

like the metric that we used in this particular

study. I don't think it's as flexible. Even a

POSI has its problems.

But, as far as the data and the endpoints,

the answer is no. But in a clinical sense, it

probably did make a lot of people a lot better.

They just didn't hit that magic clearing point,

which is probably unrealistic for somebody with

very severe psoriasis in most cases, for any

therapy.

337

DR. GARDNER: I think the answer to the

question is no, but I would have to defer to others

about the subtleties.

DR. STERN: I think the answer is no, and I

think the reason it's important is: if it were

labeled for "very severe" that would be a

relatively unique labeling for psoriasis product,

and it might well be that people went in with

expectations--both doctors and patients--of a level

of efficacy for the most severe cases, beyond which

is supported by the data.

So I think it's sort of part of keeping

the prescribing playing field as level as one can,

in terms of information that's in the insert. And

it certainly would restrict people from using it

when they thought, "Gee, this

person's--"--whatever, and might be susceptible.

DR. RINGEL: I don't think I have anything

to add. No, I don't think that it's justified by

the data.

MS. SHAPIRO: I agree with Dr. Stern's

comments.

338

DR. HOLMBOE: I agree with the above, and

even if it had shown some efficacy, there wouldn't

have been enough power here for me to be convinced

one way to the other, given there's only a total of

15 patients who had the "very severe" in both 048

and 049.

DR. EPPS: I agree also. It would be hard

to achieve minimal to no psoriasis within 12 weeks.

DR. RAIMER: I agree. No further comment.

DR. SCHMIDT: No.

DR. SELLERS: No.

DR. KNUDSON: Also no.

DR. KATZ: No.

DR. FURBERG: Yes.

DR. HONEIN: I agree, particularly with the

power comments, for trying to assess this.

DR. STERN: Gee, if it had gone 6-for-6,

versus 0-for-7, I would have taken it.

[Laughter.]

But I'm not a biostatistician.

Ahh--the next questions also for

discussion, are divided into two parts. They both

339

have to do with "has the safety profile for this

product been adequately assessed?" And the first

is:

"Please provide discussion of the clinical

and preclinical safety data, including comments on

bone and liver abnormalities, hyperlipidemia and

teratogenicity."

And the second is:

"Please discuss any potential issues

regarding long-term safety of oral tazarotene with

repeated use."

And I would ask everyone not to reiterate

their concerns, which are on the record, but really

to try to use the time to bring out something that

they don't think either they or someone else has

addressed for each of these two.

DR. HONEIN: I'm concerned about the

pregnancy rate that was observed in the clinical

trials. And I'm sort of not comforted by the

explanation that they didn't do a thorough

education campaign as part of the clinical trial,

because I don't understand that--and even three or

340

four years ago, how that would not be warranted.

And the other issue with respect to the

sort of off-label usage, just wondering if the plan

to have multiple dosages of this is going to make

it easier for off-label usage, or if there's

another rationale.

DR. FURBERG: For bone, there was a trend

for fractures, which I think is more important than

bone mineral density. So, inadequate power to

evaluate that possibility that this may be real.

For hyperlipidemia, we didn't have a good

discussion on that. We didn't get good

presentations on how many people had increases,

exceeded the treatment guideline goals.

And for teratogenicity, incomplete

ascertainment. We are missing information on known

cases.

And so, overall, I would say the safety

profile for the product is not adequately assessed.

DR. STERN: Dr. Katz.

DR. KATZ: The question being "Has it been

adequately assessed?"--and I think it has been

341

adequately assessed, given the time limit of 52

weeks, as well as could be.

However, the discussion of the safety data

that my comments on bone is very worrisome.

Everything's in the same direction: decreased bone

density, patients' having symptoms of

musculoskeletal symptoms, an alkaline

phosphatase--all in the same direction.

And these are the objective things. These

are not subject to double-blind , non-double-blind

things. These are the same.

Liver abnormalities--I think we are

reassured. And hyperlipidemia would not bother me.

First of all, it's a controllable problem. We

worry, as clinicians, about things that are not

controllable, like increasing osteoporosis that you

can't reverse. But hyperlipidemia, if you get

adequate lipid assessment--though the one case

bothered me. But that patient probably, in a

clinical practice, wouldn't have happened because

started with a high triglyceride and it kept going

up. And somehow the drug was continued. The

342

patient ended up with pancreatitis in the hospital.

I mean, I can't imagine that that would happen in

clinical practice. And so that wouldn't bother me.

And the teratogenicity is obvious bother.

And we spent time discussing the limited--the risk

management.

As far as the second part, potential

issues are bone--bone, which is paramount; and

obviously the pregnancy in patients--unlike the

Accutane, where we're concerned about this, people

using it for 20 weeks. People could be using it

for a lot longer than 20 weeks with this.

So that would be of great concern.

DR. KNUDSON: I'm so glad that Dr. Katz

comes before I do.

[Laughter.]

He said it just beautifully, and

articulated for me the things that I've been

thinking. Yes, I agree with him.

DR. STERN: Dr. Sellers.

DR. SELLERS: Yes. I just would reiterate

again the problem with power. When we design

343

efficacy trials, we don't really have the best

statistical sample for evaluating safety.

Also problematic is the length of

follow-up, and the demographics, again, because we

have more males in the study than females. And we

wonder to what extent we're excluding populations

that may be at risk for some of these observed

effects that we're seeing.

So I would say that the safety profile has

not been adequately assessed.

DR. STERN: Dr. Schmidt?

DR. SCHMIDT: I think there's no doubt

about the teratogenicity. As far as the

hyperlipidemia, I think that's been assessed.

Liver abnormalities, to me--and liver

chemistries--I hate to say it, but I don't do GGTs

and some of these other things. But I'm not going

to--I feel like they have assessed that.

And the only thing that's really

troublesome to me is this bone. I don't think

that's been, you know, really assessed. And it's

unfortunate that, you know, musculoskeletal pain,

344

and then bone pain--like in renal patients--to me

is different. You know, and I wish we had gotten

more information on that to make--but I agree, I

don't think we have the numbers for that.

DR. STERN: Dr. Raimer.

DR. RAIMER: Well, I think triglycerides

could be a potential problem. I don't necessarily

think the drug is potentially raising the

triglycerides too much, but 2 percent of the

placebo patients had triglycerides over 500 to

start with. So I think the company was not

planning to recommending any lab, but I think a

screening triglyceride to prevent people from

getting pancreatitis wouldn't be a bad idea.

And maybe considering if somebody's on the

drug for 12 months, recommending DEXA scans--at

least for a while, to follow bone densities, until

we're sure it's not going to be a problem--and

numbers of fractures.

DR. STERN: Dr. Epps?

DR. EPPS: I agree that the--I don't

believe it's been adequately assessed. The only

345

thing I would add to all the preceding comment is

regarding the hyperlipidemia. And, I guess, 41

percent was the number that I did see at one time.

And my concern, in young adults and, you

know--people, young people who are living a long

time with the disease, or with the potential to

take this medication, really cardiovascular and

long-term side effects, and whether or not those

were seen. And I didn't really hear much about

that.

DR. HOLMBOE: I agree with basically

everything's that been said previously. The only

thing I would add is that I would encourage the

sponsor to think about the way it's measuring some

of its adverse events--getting back to Dr.

Furberg's point of kind of these self-report

things, without better clinical definitions of

exactly what we're looking for, such as things like

diabetes and other endpoints.

MS. SHAPIRO: As a non-medical person I'd

again defer to you on this, although what seems

poignant to me, listening to you, is the length of

346

time that this drug is potentially going to be

taken, versus the drug we dealt with last time.

DR. RINGEL: Likewise, I agree with the

previous speakers. The only points that I have

that are different, first of all, has to do with

the issue of semen. I found--

DR. STERN: We'll get to that.

DR. RINGEL: Oh, I'm sorry.

DR. STERN: We have questions about that

later on.

[Discussion off mike.]

DR. RINGEL: I will save that.

In that case, briefly, the alkaline

phosphatase has been elevated and, intermittently,

AST and GTT are elevated. And, to my mind, it's

probably from bone, but it could be from

cholystasis, too. I think it's unlikely, given

what retinoids have done, which has usually been

hepatocellular problems. But I think that

fractionating that alkaline phosphatase makes a

whole lot of sense, and some sort of mandatory

Phase 4 study, you know, fractures--at the very

347

least--if not some other non-direct measurements of

bone metabolism.

DR. STERN: Dr. Garner?

DR. GARDNER: Nothing to add.

DR. WILKERSON: Well, as usual, I do.

[Laughter.]

Yes, I think the side effects of this drug

are manageable, but I think the package labeling

should include recommended or suggested guidelines.

I don't think this is a drug that should be put out

there on the market--I mean, when we look at

isotretinoin and how long it took to realize that

it caused depression and some of the other, you

know, adverse events. I mean those are one in 20,

one in 50,000-type events. We need to be vigilant

beyond spontaneous reporting.

So, I don't think patients ever get upset

about having some blood work drawn that protects

them. They get upset when things happen and we

haven't been vigilant--whether the package labeling

indicates so or not.

So I think all these things are--we've

348

been dealing with this for years, retinoid side

effects. They are easily dealt with with

appropriate monitoring.

DR. DAY: I agree with the previous

speakers. I wish more had been told to us about

the possibility of an on-and-off course of

treatment, because we don't know anything about the

very, very long use of this product.

DR. LEVIN: I guess I agree with the

previous comments. I just want to point out that

what's being said here really makes Question 6

extremely important, about additional studies, and

would they be needed, prior to after approval.

DR. STERN: We'll go on to one of two

questions that we've be asked to vote on. I guess

this is sort of generally the question at these

meetings, which is:

"Give the safety and efficacy information,

does the committee find a favorable balance of

risks and benefits which would support approval of

this drug?"

And I think that means "at this time," on

349

the basis of the information we have, since, as you

pointed out, there's Question 6.

And I'd like to take the opportunity to

make a comment: is that unlike the sponsor, I

believe that in deciding--one of the nice things

about being only a special government employee for

a day, is one has a greater latitude to think--as

that trite phrase is--outside the box, or outside

the box-labelings, in this case, and think about

what happens to the public health when a new

product--one's best estimate of what is likely to

happen to the public health when a product is

approved for a given indication.

And, therefore, I think we have to think

about how responsive we thought the sponsor was in

terms of some of us who had concerns about

off-label use, and minimization of that, in decided

about benefits and risks from a societal

perspective, and not as if this product would

always be used within the indication.

So that more global perspective--in the

absence of more reassurance, either about dynamite

350

efficacy for acne, or knowing about it very soon,

or an extremely robust program to make sure it's

not used widely in people at high risk for becoming

pregnant, where there are alternative drugs that

are at least, and perhaps more, effective for that

indication with similar risks makes me vote no.

Dr. Levin?

DR. LEVIN: I would vote no, too.

Abstain.

DR. WILKERSON: I would vote for approval

of this drug, with monitoring, and a Phase 4.

DR. GARDNER: I guess I would answer the

question, no, but want more discussion about

modification.

DR. RINGEL: This is a less toxic drug

Soriatane, but the issue of teratogenicity and its

effect of society, and everything we have learned

from Accutane over the last 20 years is so

compelling, that I would have to vote no at this

time--until that issue is settled.

MS. SHAPIRO: I guess I need clarity on the

question. Is this with a risk-management program

351

that I still don't completely understand--or not?

DR. STERN: Ahh--I think this is with the

risk-management program--

MS. SHAPIRO: That's been--

DR. STERN: --that one would presume, at

the individual level, was of equal efficacy to that

we hope will soon be in place for other retinoids.

I think, in my deciding on that, I thought

the risk per exposed individual of unintended

exposure during pregnancy, correcting for that

person's demographics, would be the same. And my

concern, really, is on the benefits side, where the

teratogenic risk is, in a population sense, is in

the acne population. And I don't know if the

benefit's equal.

So--and my concern is that it will be

largely--or to a large proportion--used with less

benefit, with equal risk, which is the logic behind

my vote.

MS. SHAPIRO: But if the risk-management

program were comprehensive enough, it might answer

some of your off-label concerns.

352

DR. STERN: No. If you have two drugs,

each of which has a risk your really worried

about--let's say, two drugs that have a one in a

thousand risk of instant death, and one of them

works--and the risk of death is proportional to how

long you're on it, or how many courses you have;

and one requires more courses or a longer period of

time--

MS. SHAPIRO: Mm-hmm.

DR. STERN: Even for that low risk, you

wouldn't want the other drug on the market.

MS. SHAPIRO: Right.

DR. STERN: And that's the logic behind my

vote.

MS. SHAPIRO: Okay. With that, I abstain.

DR. HOLMBOE: I don't feel I can answer the

question at this time until I know exactly what the

risk-management program is going to be, and what

additional studies are planned.

DR. EPPS: The question, given the safety

and efficacy information presented--my vote is no.

DR. RAIMER: I vote yes.

353

DR. SCHMIDT: I vote yes.

DR. SELLERS: Based on the overall public

health benefit and burden, I vote no.

DR. KNUDSON: I vote no, because there's

too much unknown.

DR. KATZ: Before my vote, I just want to

say that, obviously, when it comes to efficacy, it

works great--for some people. The need is

there--we agree. And biologically, it's exciting.

But it's not very efficient. With the

numbers we've seen, 17 percent--and then numbers

were given, "patient satisfaction," 75 percent.

Let's not forget about placebo of being 50 percent.

So, with all the problems involved, and the low

general efficacy, my vote would be no.

DR. FURBERG: My view is mixed. I think in

a highly selected group, with a risk-management

program, the drug may be okay. But there's a lot

of missing information that could open the

indications more broadly.

But I'm with Eric, that I think we need to

hear a little bit more about what would be the

354

risk-management program before I sort of finally

support approval for a very limited indication.

DR. HONEIN: Based on what's been presented

so far, I'd vote no.

DR. STERN: We need--Dr. Furberg, we need

"yes," "no" or "abstain."

DR. FURBERG: Abstain.

DR. STERN: Now we get on to a three-part

question, parts of which are for discussion, and

the third part for a vote.

"Allergan has submitted a

risk-minimization--"--and, actually, I'd like to

ask the FDA, since there has been substantial

progress and evaluation of risk-management program

since the original packet, whether they wanted to

perhaps modify this question, in terms of parts A

and B, since at least part B--it's my understanding

that the sponsor is now going to require--proposing

a requirement for all people to whom it's

prescribed, independent of gender or risk of being

pregnant. And there's been other evolution.

So do want to change A and B at all? Or

355

do you want us to still address them?

DR. BULL: Well, I think in terms of

general principles for a risk-management plan, I

think we've heard a fairly consistent message from

around the table that it would be--you'd like to

see a consistent program across all the retinoids;

and that the same kinds of risk management--given

that there's high probability that this one is an

equivalent teratogen to the other products that are

currently marketed--I guess if there are any

additional comments that are consistent with any

other concerns that you want to be sure the agency

attends to, maybe that might be the most useful to

us for A and B.

DR. STERN: Great. Do you want to start,

Dr. Levin?

DR. LEVIN: I--what am I voting on?

DR. STERN: As I understood Dr. Bull--and

please correct if I'm wrong--that there's been a

general sense, which is also my sense--that the

committee--or many members of the committee have

felt that a standardized program across teratogens

356

[sic], however it best evolve, would be applicable

to this agent, as well, if approved. And I guess

the question--the comments the FDA might like are

any disagreement with that general principle, and

any particular advice that's come out specifically

at this meeting, and not revisiting the February

meeting at all.

DR. LEVIN: I would agree that, both from

the perspective of reducing burden, and the

opportunity for error, that standardization is

sometimes a good thing.

I would encourage the agency to talk to

sponsors about the appropriateness of including

indication for tracking how drugs are used, and not

just relying on marketing data.

DR. DAY: Some standardization would be

very helpful for the reasons that Dr. Levin has

said.

I can't tell enough about the materials in

the briefing document about the risk-management

program. For example, there are different

brochures for males and females. And it could just

357

be that males have less information, so some things

are left out. But we haven't discussed the other

question, about the involvement in males, so this

gets a little bit circular, to decide all of these

things.

So, in general, a general program that

would be appropriate for all drugs in this class

would be nice, but sponsors should be allowed to

come forward with specific exceptions, given their

product--such as the half-life in the body, and so

on and so forth.

DR. STERN: Dr. Wilkerson?

DR. WILKERSON: Like I said before, I think

the pan-retinoid form, from the practitioner's

standpoint, is the best way to go with this, for

lack of confusion, and just everything else in the

market. And then--you know, it's hammering out the

details. But to have seven or eight different

renditions of managing these drugs, for

practitioners is just a nightmare.

As far as the indication being on the

scrip, my only concern about that is since we do

358

have to use drugs off-label many times, that it not

become a means of an insurance company denying

coverage for a drug that a patient needs because

it's not--quote--"the FDA indication," which is a

continual problem with insurance companies, is that

they tend to cling to these indications when, in

fact, we all know we use drugs off-label every day.

DR. GARDNER: I think I'd like to pass on

this until we get to the discussion of the male

involvement.

DR. STERN: Nothing to add?

DR. GARDNER: Nothing to add at this time.

DR. RINGEL: I'll pass, as well.

MS. SHAPIRO: Nothing to add.

DR. HOLMBOE: I would simply reiterate what

Dr. Furberg had said earlier about some sort of

feedback loop in this program. There's a lot of

monitoring activities that appear to built in, but

there doesn't appear to be any feedback, particular

to the practitioners.

There is good evidence now in the health

services research arena that feedback, particularly

359

when it's physician-specific, can change behavior.

And I think that would be really important to

include.

And I'd also like to see more active

involvement on the part of patients. There are

also techniques, such as getting patients to write

down a commitment to adhere to a program can also

increase compliance rates. And I think those

should be considered as well.

DR. EPPS: I like the idea of some kind of

standardized program. Currently, we have many

different pamphlets and folders and things to

distribute and have signed.

The other issue--although that's admirable

for patients to write things down, sometimes

they'll put down what they think you want to hear.

So, I mean, that's just being realistic. And I'm

sure the sponsor would go along with whatever FDA

recommended.

DR. RAIMER: I really don't have anything

to add. I'm still not sure why we include males in

a registry, but I do like the idea of having it

360

standardized, however we do it.

DR. SCHMIDT: I agree.

DR. SELLERS: I agree with the previous

comments, and I also strongly urge the use of

indications within the risk-management program,

because although clinicians and professionals are

generally aware that off-label use occurs all the

time, in many cases patients and consumers don't

realize that the drugs they're being treated with

have never been approved for the indications in

which they're being treated.

And so we'd like to have that information

available.

DR. KNUDSON: I'm extremely disappointed

that after the February meeting we're still hung up

on getting a standardized program. I thought we

had decided that there would be.

I would like to urge the FDA to please

find a way out of this patent problem. There must

be a way to solve that, and solve it fairly soon.

DR. KATZ: I have nothing to add.

DR. FURBERG: Nothing to add.

361

DR. HONEIN: Nothing specific to add, but

just to reiterate support for a standardize program

that would be easier for all elements to follow.

And I really do see a benefit of indication, and I

don't know that there's a way this can be part of

the registry without insurance companies' having

access to it. I would think privacy concerns might

be able to get around some of this. But I think it

would be useful, both for the feedback to

providers, and for evaluating the programs.

DR. STERN: thank you.

With the Executive Secretary's permission,

the next thing is for a vote. And it has to

do--it's the question of: "Are the scientific and

clinical uncertainties surrounding semen levels of

Tazorac--the metabolite of Tazorac--a factor to be

considered in tazarotene risk minimization?"

And I don't know--is it possible to do 5,

4 first, because that's really the factual basis.

MS. TOPPER: You may.

DR. STERN: So, why don't we do 5, which is

the discussion basis for then voting--I think.

362

So, Question 5, again, three parts, all

for discussion:

"How can FDA best address the potential

clinical relevance of high tazarotenic acid levels

in semen? Options might include: A) further

delineation of the potential risks (via

consultation with teratogenicity experts,

additional preclinical studies, etcetera); B)

informing clinicians and patients of the finding

and its uncertain clinical relevance; C)

recommending precautions (such as the use of

condoms) pending characterization of the potential

risk."

And we're asked to "Please comment on

whether further risk assessment should be done, and

whether any cautionary language or recommendations

should be made while additional risk assessment is

pending."

Who would like to--Dr. Honein.

DR. FURBERG: I'm for A, and I would like

to add "D"--additional studies. That's really what

we need. Right now we don't know the significance

363

of the changes.

DR. STERN: Dr. Katz?

DR. KATZ: Well, why wouldn't these

concerns be alleviated by just having a labeling,

just like females can't get pregnant for one month,

having males use the same precautions for a month

afterwards. So, in other words, having C cover

everything.

Is that not feasible, or am I missing the

point?

DR. STERN: I don't know. I mean, I--I'm

no surer about this than I think I'm hearing you

are about this.

DR. KATZ: It's got to be--half-life is

seven to 12 hours. Do we--is it appropriate to ask

if there's data on how--

DR. STERN: Dr. Wilkin?

DR. WILKIN: Yes, just--a point of

explanation on the query here.

One of the daily challenges that we have

at FDA is taking some evidence of a potential risk,

and then translating what we know into wording that

364

goes into labeling. And we're always trying to get

it just right. We don't want to over-warn, we

don't want to under-warn. If there's uncertainty,

we need to convey that. And even though it's a

daily task, it always seems to be something that we

need to spend a lot of time and a lot of thinking

on.

And I guess that's what we're asking you

to share that piece with us.

If we end up over-warning on this part,

what we might end up with is labeling that said,

you know, "You absolutely must--"--and if we don't

have other pieces in the labeling, then women who

otherwise had a pregnancy that was very highly

desired, they may have this concern and have a

termination based on really scary language.

So I think that's--you know, we're trying

to hear from the advisory committees, in this case,

you know, the direction that we should pursue here.

DR. KATZ: I appreciate that comment.

And, with that in mind, probably, either we

should--if they have the data, we should see that

365

again. I know it was very lucidly given to us by

the sponsor. But perhaps we should see that data.

I remember some astronomically small--one

out of--

DR. STERN: One in 5,000.

DR. KATZ: --5,000 number. But from other

toxicologic studies, can that ever be significant?

So I think we would need to have more

information before we answer that question--at

least I would. Maybe we could see the slide again,

if anybody else would like to.

DR. STERN: Why don't we see the slide that

showed the 1 in 5,000--an FDA slide.

DR. BROWN: My name is Paul Brown. I'm

acting pharm-tox supervisor in the Division of

Dermatologic and Dental Drugs. And we anticipated

this question might come up, so we actually have a

backup slide that addresses some of these issues.

[Slide.]

The slide you saw before had this

information in it, and I believe that the total

dose that could be achieved in semen--assuming

366

worst-case scenario--would be about 831 ng, which

about 1/5000th of a single 4.5 mg capsule dose.

To try to put that in perspective with the

teratogenicity studies, it's about 1/6000th of the

highest oral dose that was not teratogenic in

animals. So that would be a no-AL.

This last bullet just sort of goes over

some of the uncertainty about that information;

that, as Dr. Yao presented, the human may actually

be a more sensitive species than some other

animals.

So, looking at other retinoids, you see a

ratio that might reduce that 1-to-6,000 to

something else, and just using the numbers that are

available for the other retinoids, you might reduce

it to some of the numbers shown there.

That's just based, again, on the

literature and information with those other

retinoids. We don't know for tazarotenic acid.

The other uncertainty, of course, is that

the semen would be delivered by the vaginal route.

The teratogenicity studies are oral studies, so we

367

don't really know whether you could get greater

exposure to the fetus from the vaginal route than

through the oral route. There actually is some

evidence in literature that that is possible; that

you can get higher exposure in the uterus by the

vaginal route compared to other routes.

So, again, the margin of safety might be

reduced even more, and it's just uncertain at this

point.

DR. STERN: Am I correct that the greatest

concern would be, in fact, with exposure after

implantation, essentially; or after fertilization

by prior intercourse, as opposed to anything to

do--some of it might be mitigated by some kind of a

warning of using only--"If your partner becomes

pregnant, then use a condom--"--as contra-logical s

that may sound.

[Laughter.]

DR. BROWN: Yes, I think there's

two--[laughs]--I think there's two possible areas

where there could be risk; either it's exposure

initially, at conception--you know, what with the

368

tazarotenic acid in the semen at that point; or

repeat exposure later.

And there are fertility studies that

looked at teratogenicity when male animals were

treated only, and they didn't really see a

teratogenic signal at that point. So I think the

main focus has been exposure if there was repeated

exposure, once a female's pregnant.

DR. KATZ: Well, with that information, I

would think, until any further data comes up, then

we have to have some labeling as far as the male

goes.

DR. KNUDSON: Well, A, B and C all seem

very reasonable to me--plus D, as Dr. Furberg said.

But I will defer to others.

DR. SELLERS: It sounds like we have an

incomplete risk analysis, and that further animal

studies are necessary to understand what the dose

is at the target, and what the effect of exposure

duration may have.

DR. SCHMIDT: I think this is a real sticky

wicket, in the sense that one of those slides up

369

there said that this was like 10 cc of semen. And

I don't want to belabor that point, but all I can

say is, "What a man!"--you know?

[Laughter.]

But--seriously--you know, these are such

infinitesimal amounts, and I think that B,

informing clinicians and patients of

findings--[laughs]--this is--and findings, and its

uncertain clinical relevance would be the thing

that I would put down.

I think that when you have 11,000, six

times the lower the amount to produce in animals

and this other stuff, I think this is something you

are--you're going to create more problems.

And I know why we did this is because of

the problem with Accutane, and some of the

teratogenicity. But I thought that was not really

proven.

DR. RAIMER: I agree, at this point in time

we seem to have no idea of whether it's a problem

or not. But just--I mean, you'd think, logically,

it was in such small amounts that it's probably

370

not, but we're not sure.

So, I kind of hate to frighten people

unnecessarily, but I think we ought to put that

it's there, and that it's one-to-one with serum,

which would be in the labeling. But as far as

specific recommendations, I'm not sure we should

make them when we have no idea what we're doing.

DR. EPPS: I agree--more information is

needed. I think, in the industry information there

was something about a mucosal plug. Well, we don't

know whether there's mucosal absorption. Perhaps

some studies could be done in that way to find out

whether, you know, absorbed mucosally, because that

would be an issue.

And, I guess, the issue is: the continued

exposure, whether there's repeated exposure. It

may not be a one-time issue. It's an issue of

whether it's a cumulative effect, or whether it's a

one-time effect. We just don't know. And we need

more information.

DR. HOLMBOE: Just to argue for A, B and D.

MS. SHAPIRO: It sounds like we need more

371

information.

DR. RINGEL: One thing to take into account

is that we do recommend pregnancy tests for women

who are using topical tazarotene, just as a--you

know, just as round number, if somebody applies one

ml of tazarotene to their face, they've got about 1

mg topically. And if we're looking at semen--just

quick numbers here--it looks like it's about .0001

mg in the semen, which seems awfully small.

I guess what I would do is the same thing

I do in my office when I don't know what I'm

talking about, I usually just tell my patients, "I

really don't know what I'm talking about."

[Laughter.]

And we need to put down what we know in

the package insert, and leave it to the patient's

judgment.

DR. STERN: I think Dr. Ringel is

absolutely correct. And going back to what Dr.

Furberg--wouldn't it be nice and comforting if we

had follow-up on those hundred-plus pregnancies

during topical administration, where we'd have some

372

idea of whether there was an extraordinarily

low-level effect, and that would be very reassuring

to us.

And perhaps, in fact--if you'll pardon the

pun--the route to go in terms of deciding whether

this exposure was, in fact, one to be concerned

about, since we know with the widespread topical

use there are going to be a fair number of exposed

pregnancies with the topical tazarotene,

DR. GARDNER: I think I would vote in favor

of generating more information about these

subjects, and not in favor of anything that sounds

like more recommendations on things that we can't

monitor or enforce.

I know you don't want to talk about the

February meeting, but it does seem that we spent

two days trying to understand why the contraceptive

recommendations that were out there, in a known

arena, may be problematic; and then to add that

someone who's going to use this product for at

least 52 weeks should also always use condoms seems

to me to be an exercise in futility, especially on

373

as little data as we have.

So I would suggest that--I think someone

said "A, B and D"--whatever we decided D is--but

generating more information, or re-analyzing data

that are there to shed light.

Also, this doesn't exist in a vacuum. We

will be generating data from the risk-management

programs of the retinoids as we go along anyway,

and so this may appear as information we can use.

DR. WILKERSON: Well, my concern is--I

mean, there's--what?--an approximately a 1 percent

birth defect rate in the population? Is that

right, Dr. Honein?

DR. HONEIN: About 3 percent.

DR. WILKERSON: Oh, okay. So--2 to 3. As

far as I know, no one's ever died from wearing a

condom, except maybe unless they had an anaphylaxis

from the latex or something.

[Laughter.]

So--yeah, I don't see any problem in

putting in the package labeling, that this is

what's know, and the safe route is certainly to do

374

that.

I guess my other comment on this: I didn't

see anything as far as morphologies, or sperm

counts. I mean, I'm assuming, like the rest of the

retinoids, this did not have an issue as far as

reducing those counts, or the morphologic

appearance or functioning or anything of those

sperm.

VOICES: [Off mike.] [Inaudible.]

DR. WILKERSON: Oh, it's in the animal?

Did I miss that one sentence?

[Laughter.]

Okay. And it did not, right?

VOICES: [Off mike.] [Inaudible.]

DR. WILKERSON: Oh, it did.

VOICES: [Off mike.] [Inaudible.]

DR. WILKERSON: It's not a big deal? Well,

if you're a rat, it is.

[Laughter.]

So, yes--with cautions.

DR. DAY: A, B and D.

DR. LEVIN: A, B and D.

375

DR. STERN: Now we have to vote on 4.C--

which we've already talked about--formally.

So why don't we just go through. And I

think everyone's expressed their opinion.

Is that right? Do you need a formal vote

on that particular question?

MS. TOPPER: Yes.

DR. STERN: So this question deals with--it

has a lot of ambiguity, given the complexity of the

question. But people can say "yes," "no" or

"abstain."

"Are the scientific and clinical

uncertainties surrounding--

DR. BULL: Unless there are other opinions

that the committee would like to offer--

DR. STERN: Oh, great.

DR. BULL: --we are--I think you all have

given us, you know, very rich input on that

issue--unless there are further comments.

DR. WILKIN: Actually, several of the

members mentioned "D" as one of the options. So I

hope that comes back up again--5.D--when you talk

376

about 6; that is, what would the additional studies

be?

DR. STERN: So, with being granted a buy

for good behavior [laughs]--or perhaps other than

good behavior [laughs]--we'll now go on to Question

6, which is, again, a discussion question. And

that is:

"What additional studies are needed? Are

these studies needed before or after approval of

the product?"

And I think "studies" can be in the

broadest sense of the word.

Who would like to start with

recommendations?

DR. LEVIN: What was the result of the vote

on 3? Which was approval?

[Pause.]

MS. TOPPER: There were four abstentions,

three "yes" and nine "no."

DR. LEVIN: Umm--as one of the "no's" I

think I want to express what I really wanted to get

to was 6. I mean, the reason I voted no is because

377

I think that there's a lot we don't know that we

need to know.

And so, two concerns: there's a lot we

don't know that we need to know, and the devil's in

the details of how we manage the risk of this

particular product.

So, I think, in general--yes, there's a

lot more. And I think we've talked about all of

the missing pieces here. So I think the FDA has

heard--at least as far as I'm concerned--what needs

to be done.

DR. STERN: Yes.

DR. DAY: Before, one of the people on the

sponsor's side wanted to tell us about the

risk-management program and it wasn't an

appropriate time, but we haven't heard much about

it. And just having the list of all the things up

there doesn't tell us.

And there are a number of things that

might be in the works for the risk-management plan,

to see whether--for example, Dr. Walker talked

several times about the knowledge test, to prove

378

that the physician and the patient understand the

risks and so on. And I was wondering if there was

going to be comprehension testing of the messages

in advance, or not.

And so there are things like this, which

they already may have in their plan and so

therefore I wouldn't suggest anything--but they

might not. Or, they might have it in a way that

would be a good way to do it, or something else

might be needed.

So, in terms of general knowledge testing,

we don't have enough information yet.

DR. STERN: Dr. Wilkerson?

[Pause.]

I'm just going around, since I haven't

seen a lot of hands, to make sure that everyone

puts in their comments about additional studies.

DR. WILKERSON: Well, probably a couple

things. One, I think there is an adequate data

base to approve this drug. I think there is a need

in the market, and so I disagree.

You know, I think there are issues, as

379

there are always issues. We always need more

information. But we have to draw a line.

There is a critical need in the market for

a retinoid to treat certain female patients,

because the alternative that we have right now is

basically a bigger risk that this drug is, in terms

of pregnancy. And I think to delay this approval,

it is going to put a lot of people out there that

need therapy in harm's way.

But, in terms of additional studies,

certainly the male side of it I think is the issue,

and some of these other things.

But, yes, I think sometimes we subject

some of these drugs to more--far more scrutiny than

what the drugs that are already on the market have

been subjected to. And if we sat and picked those

apart also they would not be approved if we were

applying the same criteria.

DR. STERN: Dr. Gardner?

DR. GARDNER: I agree with Dr. Wilkerson.

I think that we tend to talk ourselves into a lot

of pre-approval requirements that may, in fact, not

380

be necessary or prudent.

I would like to encourage that, as the FDA

develops this--there are many questions that I have

about the program that I think are pending

discussions on the larger retinoid thing. For

example, we've talked a lot about how to make this

feasible to do, and what happens with the lab

tests--the lab-confirmed pregnancy tests--and some

of things that we haven't heard get resolved.

I'm sure that as this develops along,

working with the FDA, they will resolve them. If

we could define a list of things that probably

should be done sequentially with--or

sorry--concurrently with approving a drug, if the

dermatology community believes that it would a good

addition, I think we should focus on that and see

what needs to be done in order to supplement what

we already have as data.

DR. STERN: I think the study that's need

is one that--where the company robustly

demonstrates that they've figured out a way of

putting this drug into the market, that it will be

381

used, in fact, for its labeled indications, and not

for other indications that are more--that have a

prevalence of very high-risk individuals for

pregnancy; and that they have--that they

demonstrate that they have a way to do it, and that

they have an ongoing study, in terms of really

showing that that way that they're going to bring

the drug to market will, in fact, be effective in

having the drug used in the individuals for whom

the benefit-risk ratio is likely to be the best one

for a teratogen.

DR. RINGEL: One thing I'd want to make

sure is--that I haven't heard quite enough of--is

that this is a better drug for two-thirds of the

population than is Soriatane for the indication of

psoriasis. And I think that that's pretty clear.

And the problem is that it's a horrendous

drug for the other third of the population, who can

get pregnant. The one thing that needs to get

done, as far as I'm concerned, before approval, is

to solve that problem.

In terms of the other issues, I think

382

people are absolutely right; I mean, I'd like to

see mandatory adverse event reporting for all

medications, for serious adverse events. But,

unfortunately, that's not the policy of this

country.

I can't see singling out this particular

drug. I would love to see all of these things

followed up. But do I think this is worse than

other things that have come out on the market? I

don't think so. I really am concerned about the

pregnancy issue.

MS. SHAPIRO: I guess my biggest concern at

the moment is Dr. Stern's. So I would agree that

if we can figure that out, that would relieve a lot

of concerns.

DR. HOLMBOE: Let me just add that if it

does get in the marketplace, to make sure that the

outcome measures put in place from the

effectiveness analysis are robust, and that it will

really be approached as an observational registry

study to track a number of these metrics.

DR. EPPS: I would be interested in more

383

information and follow-up on the fractures; for

example, do people have multiple fractures, what

kind of trauma is associated or not associated.

I would be interested in following up on

some of the endocrine and other issues, whether

they be thyroid or triglycerides. Yes, they can be

treated, but if a significant number of people get

them, whether it's triglyceride of 500 or 750--I

mean those are high. I mean, serum looks cloudy at

that point. And that's something to be followed.

Also, perhaps, more testing with the

rabbits and mice, which seemed to have more data

regarding this drug, when compared to the other

drugs, as far as the lowest teratogenic dose.

Maybe that's a good basis for a comparison--doing

some trials regarding those. I guess it was in one

of the--oh--I guess it's Dr. Yao's information.

Those would be two animals, perhaps, where trials

could be done, and maybe some comparisons there for

teratogenicity.

DR. RAIMER: I think most of our concerns

have been expressed. I have a little bit of a

384

problem that we're basically not approving a drug,

in part because we're worried it could be used for

something that it hasn't been presented for. I'm

not sure that was our real purpose here.

So, I do have a concern that we're doing

that.

DR. SCHMIDT: I, too, think that we should

approve this drug, but I do think that if it is

approved, that there are some studies, like the

fractionating the alkaline phosphatase, the

repeated checks on triglycerides--not just to not

have any lab.

And then this bone thing, to follow that

out more than one year, on long-term studies.

But I really think that there is a niche

that we do not have a medication for, with some

aspects of psoriasis, that we really do need this

medication.

DR. SELLERS: I'd like to echo Dr. Stern's

comments, and also mention that without some of

these studies, we can't develop an adequate

risk-management program. The risk assessment for

385

vaginal dose of the drug is critical for informing

the risk-management program; also patient

populations in which the drug is going to be

used--and that is informed, whether we like it or

not--by the off-label use.

So I think that we need to look at the

effects, in these drugs, in some of the

sub-populations that weren't represented in the

data that was given to us.

DR. KNUDSEN: I have nothing to add. I

think it's all been said.

DR. KATZ: I agree with Dr. Raimer, that we

shouldn't be concerned with off-label use. After

all, some of the drugs of choice in the past for

certain diseases were used for years off-label;

like methotrexate for psoriasis, dapsone for

dermatitis hepatoformis. So that's the

responsibility of the physician and the patient,

who's informed, that it's off-label use.

My main concern is longer follow-up for

the bony abnormalities, and the, really,

efficacy--the low percentage of efficacy here. We

386

heard from two patients, in whom it was very

efficacious. But how about the other 17 out of

20--statistically--where it wasn't equally

efficacious, and they would be going through very

expensive treatment that also exposes the female

population to the obvious teratogenic effect.

DR. FURBERG: I abstain, because I still

have a problem coming down, and interpreting the

findings.

I think I would be in favor of approval

for a limited indication, if we could come up with

a strict management program--under that condition.

And if we do, I'd like to make it

conditional upon a thorough evaluation, and a re-

review in a about a year or two, to see whether the

management program worked.

We have had a lot of failures along the

way. But if we can set up a new, better system,

and we can show, after a year or two, that it's

working fine, then I'll be for opening the gates

more.

And while that is underway, I wish we

387

could also get some additional long-term follow-up

data on the patients we already have, or other ones

that you may enroll; information on--as we talked

about--failures; the metabolic effects on glucose,

and lipids and so on.

So--but I haven't really had--I'm not

getting a sense that we have come up with a strict

risk-management program that would be satisfactory

to me.

DR. HONEIN: Well, I think, despite the

difficulties from a public health perspective, the

most critical additional study to be done is to do

the comparison of this drug with Accutane for acne,

so that we know which one is better, and that data

is available and published. And if this is as good

or better, then we could develop a risk-management

program with that patient population in mind. If

the data is published that this is not as good,

then I think off-label use would probably be

minimal, and I'd have much less concern about this

being approved for psoriasis.

So I think that study is the absolutely

388

critical additional study that I'd like the

resources to be directed to.

DR. STERN: Thank you.

Dr. Wilkin, have we addressed all of the

issues posed to the committee? Or are there

additional questions you'd like to pose to us?

DR. WILKIN: I appreciate all the

discussion. I think you went beyond the

questions--

[Laughter.]

--you added additional thoughts that we

hadn't even thought to ask. And so you certainly

put an enormous amount of effort into reading the

briefing documents, and thinking all this through.

It will take us quite a while to go back

over the transcripts and digest all of this.

But we thank you very much.

DR. STERN: Thank you.

The meeting is adjourned.

[Whereupon, at 4:23 p.m., the meeting was

adjourned.]

- - -