1

 

DEPARTMENT OF HEALTH AND HUMAN SERVICES

 

FOOD AND DRUG ADMINISTRATION

 

CENTER FOR DRUG EVALUATION AND RESEARCH

 

 

DERMATOLOGIC AND OPHTHALMIC DRUGS

 

ADVISORY COMMITTEE

 

 

 

 

 

 

Friday, August 27, 2004

 

8:00 a.m.

 

 

5630 Fishers Lane

Room 1056

Rockville, Maryland 20852

2

 

PARTICIPANTS

 

Jennifer A. Dunbar, M.D., Acting Chair

Kimberly Littleton Topper, M.S.

 

MEMBERS:

 

Paula L. Knudson

William Gates, M.D.

 

CONSULTANTS (VOTING):

 

Scott M. Steidl, M.D.

Jeffrey Lehmer, M.D.

Vernon Chinchilli, Ph.D.

 

CENTER FOR DEVICES AND RADIOLOGICAL HEALTH

OPHTHALMIC

DEVICES PANEL MEMBER (VOTING):

 

Jose S. Pulido, M.D., M.S.

 

PATIENT REPRESENTATIVE (VOTING):

 

Elaine King Miller, Ph.D.

 

INDUSTRY REPRESENTATIVE (NON-VOTING):

 

Peter A. Kresel, M.B.A.

 

FDA STAFF:

 

Jonca Bull, M.D.

Wiley A. Chambers, M.D.

Jennifer D. Harris, M.D.

3

 

C O N T E N T S

Call to Order, Jennifer A. Dunbar, M.D. 4

 

Conflict of Interest Statement,

Kimberly Littleton Topper, M.S. 5

 

Introduction, Wiley Cambers, M.D. 7

 

Eyetech Pharmaceuticals Presentation:

Introduction, David Guyer, M.D. 30

 

VEGF Overview and Macular Degeneration

Pathophysiology, Antony P. Adamis, M.D. 36

 

Pegaptanib Clinical Efficacy, David Guyer, M.D. 51

 

Pegaptanib Clinical Safety,

Anthony P. Adamis, M.D. 79

 

Pegaptanib Benefit/Risk Profile,

Donald J. D'Amico, M.D. 102

 

Committee Discussion 114

 

FDA Presentation, Jennifer D. Harris, M.D. 129

 

Committee Discussion 153

 

Open Public Hearing:

Daniel D. Garrett, Prevent Blindness America 192

 

Ellen Hofstadter 196

 

Nikolai Stevenson, Association for

Macular Diseases 198

 

Carl R. Augusto, American Foundation for

the Blind 201

 

Bruce P. Rosenthal, OD, FAAO Lighthouse

International 207

 

Bob Liss, OD 210

 

Committee Discussion 211

 

Questions 217

 

4

 

P R O C E E D I N G S

 

Call to Order

 

DR. DUNBAR: I would like to call the

 

Dermatologic and Ophthalmic Drugs Advisory

 

Committee meeting to order to review NDA 21-756,

 

for Macugen, and I would like the committee members

 

to introduce themselves. I am Jennifer Dunbar,

 

from Loma Linda, California, and I would like the

 

committee members, starting on my left, to

 

introduce themselves.

 

DR. GATES: I am William Gates, from

 

Nashville, Tennessee.

 

DR. LEHMER: I am Jeffrey Lehmer, from

 

Bakersfield, California.

 

DR. PULIDO: Jose Pulido, Rochester,

 

Minnesota.

 

DR. STEIDL: Scott Steidl. I am a retina

 

specialist from the University of Maryland, in

 

Baltimore.

 

MS. KNUDSON: Paula Knudson, with the

 

Texas Health Science Center, in Houston.

 

DR. CHINCHILLI: Vern Chinchilli, Penn

 

5

 

State Hershey Medical Center.

 

DR. BULL: Good morning, Jonca Bill,

 

Director of the Office of Drug Evaluation V, in the

 

Office of New Drugs here, at FDA.

 

DR. CHAMBERS: Wiley Chambers, Deputy

 

Director for the Division of Anti-Inflammatory,

 

Analgesic and Ophthalmic Drug Products.

 

DR. HARRIS: Jennifer Harris, medical

 

Officer, same division.

 

MR. KRESEL: Peter Kresel. I am the

 

industry representative, Irvine, California.

 

MS. TOPPER: Kimberly Topper, FDA, the

 

Executive Secretary for the committee.

 

DR. MILLER: Elaine King Miller, Amarillo,

 

Texas.

 

DR. DUNBAR: Now we will ask Ms. Topper to

 

read the conflict of interest statement.

 

Conflict of Interest Statement

 

MS. TOPPER: The following announcement

 

addresses the issue of conflict of interest with

 

regard to this meeting and is made a part of the

 

record to preclude even the appearance of such at

 

6

 

this meeting. Based on the submitted agenda for

 

the meeting and all financial interests reported by

 

the committee participants, it has been determined

 

that all interests in firms regulated by the Center

 

for Drug Evaluation and Research present no

 

potential for an appearance of conflict of interest

 

at this meeting with the following exceptions:

 

Dr. Jennifer Dunbar has been grated a

 

waiver under 18 U.S.C. 208(b)(3) and 21 U.S.C.

 

505(n) for her spouse's ownership of stock of the

 

sponsor. The stock is valued from between $25,001

 

and $50,000.

 

Dr. Jose Pulido has been grated a waiver

 

under 21 U.S.C. 505(n) for his children's ownership

 

of stock in the sponsor. The stock is valued from

 

$5,001 to $25,000.

 

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.

 

In the event that the discussions involve

 

any other products or firms not already on the

 

7

 

agenda for which an FDA participant has a financial

 

interest, the participants are aware of the need to

 

exclude themselves from such involvement and their

 

exclusion will be noted for the record.

 

We would also like to note that Dr. Peter

 

Kresel has been invited to participate as a

 

non-voting industry representative. Dr. Kresel is

 

employed by Allergan.

 

With respect to all other participants, we

 

ask in the interest of fairness that they address

 

any current or previous financial involvement with

 

any firm whose products they may wish to comment

 

upon. Thank you.

 

DR. DUNBAR: Now we will ask Dr. Chambers

 

to give an introduction of the issues that we will

 

review today.

 

Introduction

 

DR. CHAMBERS: Thank you, Dr. Dunbar. Let

 

me start with welcoming everybody. Good morning.

 

I want to particularly welcome the advisory

 

committee members, and the time that they have

 

taken both to review the material and to both

 

8

 

travel and attend today.

 

[Slide]

 

We are here today to discuss Macugen, and

 

this is the Dermatology and Ophthalmology Advisory

 

Committee meeting. Those of you who think you

 

should be some place else, we would welcome the

 

open seats if you want to give them up.

 

My name is Wiley Chambers. I am the

 

Deputy Director for the Division of

 

Anti-Inflammatory, Analgesic and Ophthalmologic

 

Drug Products, and it is our Division within the

 

Office of Drug Evaluation V that will be reviewing

 

this application today.

 

[Slide]

 

This application, unlike many others--or

 

at least the section that we will be reviewing

 

today, unlike many others, is part of the

 

continuous marketing application Pilot 1 NDA

 

submission which was part of PDUFA 3, which is the

 

Prescription Drug User Fee Act that was enacted

 

into law in 2002. This allowed for the

 

presubmission of individual modules in different

 

9

 

sections that would then be reviewed, and comments

 

given back. This would not be a final action but

 

it would be comments on a particular section, with

 

the goal of speeding ultimate approval of

 

particular applications by being able to give

 

interactive comments early on. The action on the

 

actual NDA will only be taken after all the modules

 

are submitted and reviewed.

 

[Slide]

 

Today's discussion is clinical only. We

 

are only dealing with the clinical section. We are

 

not dealing with the pharm. tox. section. We are

 

not dealing with the chemistry manufacturing

 

section. So, no one should expect that we will

 

take an action on this NDA today, tomorrow or the

 

next day because there are other modules which are

 

being reviewed in their own time course.

 

The expectation is that we will give

 

comments back to the sponsor of the application

 

within approximately six months of the time when

 

the module was submitted, and so we have scheduled

 

this meeting to deal with the clinical issues and

 

10

 

our clinical feedback. As you will hear later on,

 

we have particular questions that are geared toward

 

this application, but we are looking primarily to

 

see have we missed anything; are there other areas,

 

while we are still within the review period, that

 

we should be looking at further, or are there

 

issues that you think need to be further explored

 

before an application would be acted on one way or

 

the other?

 

[Slide]

 

I am going to spend some time today going

 

through basic clinical trial design issues for

 

products for macular degeneration in general.

 

[Slide]

 

The Division gives guidance as trials are

 

performed on a way to do a particular trial. We

 

don't believe there is a single method to do all

 

clinical trials. We have tried to give what we

 

think is a good way to do trials that will give

 

answers that we can then interpret. We clearly

 

recognize that there may be additional ways and

 

there may be reasons to have variance from what we

 

11

 

recommend. But just so that everybody is in the

 

same page, I am going to go through what we

 

generally recommend to sponsors of trials so you

 

know where there are potential differences, which

 

you may either agree with or disagree with, but

 

more for informational purposes.

 

We ask that trials be parallel on design

 

trials; randomized by person as opposed to

 

randomized by eye; double-masked, meaning at least

 

the investigator and the patient are masked to what

 

treatment they are receiving; and to try to

 

incorporate dose ranging within the study

 

development plan. That does not mean every trial

 

but it means that there be an exploration to dose

 

ranging.

 

[Slide]

 

The inclusion criteria for at least wet

 

macular degeneration, using that term as broad as

 

that is, is that we expect patients to have

 

choroidal neovascularization documented by fundus

 

photography and/or angiography. We expect there to

 

be specific observable features, including

 

12

 

membranes greater than a particular defined size

 

and with particular diagnostic features such as

 

leaking on fluorescein, such as leaking on

 

indocyanine green or ICG, but define a particular

 

population for which we could then label the

 

product.

 

[Slide]

 

We try to get the trials in total to be as

 

general as possible while still identifying a

 

population that the product works for. Patients

 

with concurrent ocular diseases that may be

 

associated with choroidal neovascularization we

 

think should be excluded to avoid any kind of

 

confounding issues. In this particular case that

 

generally means excluding people with presumed

 

ocular histoplasmosis and excluding high myopia,

 

primarily because these things can also cause

 

choroidal neovascularization and we want to try and

 

figure out which disease the product is working on.

 

[Slide]

 

We ask for replication. So, we want

 

safety and efficacy, supported by at least two

 

13

 

independent trials of at least two years duration.

 

We are looking for robustness in the findings. We

 

want independent trials, and to that extent we mean

 

geographically separate so that we know the product

 

does not just work in Washington, D.C. or does not

 

just work in Boston or one particular city where

 

the water supply is unique. These trials conducted

 

to date were each multicenter trials and so,

 

obviously, clearly meet that criteria.

 

Actually, before I go on let me say one

 

thing about the two-year trial. We have asked for

 

trials to go on for two years and we have had

 

discussions at this advisory committee before about

 

how long trials should go on for. We have

 

recognized that endpoints may be acceptable at a

 

one-year time point but we have asked that trials

 

continue on for two years. So, while you may not

 

hear two-year data, you can rest assured that the

 

trial will continue to go on for two years and we

 

will ultimately have that information which we will

 

factor into our decision. But we believe that,

 

because of the age of the population, one year is a

 

14

 

significant portion in the rest of their lives.

 

Consequently, if the product is showing benefit at

 

one year we believe we could potentially approve a

 

product and label it as working for however long it

 

works for, but we think that duration needs to be

 

at least one year, but have not been wedded to

 

anything more than that. If you end up disagreeing

 

with that, as with anything that I say today,

 

please feel free to make those comments to us.

 

[Slide]

 

The clinical trial program we think should

 

be able to identify adverse events that occur at

 

least at a one percent adverse reaction rate.

 

People may argue that one percent is too low, too

 

high. It is, for lack of a better figure, what we

 

have picked. That means you need at least 300

 

patients studied fully through that to be able to

 

determine that. We generally recommend at least

 

500 patients so that we are not dealing with,

 

"well, I've got 299" or "I've got 298" or "I've got

 

301." We know in this population, because of the

 

natural age and normal life span, people are not

 

15

 

going to necessarily survive through the

 

trial--just not related to the drug but related to

 

other reasons. So, we start out asking for people

 

to do trials of 500 patients or more.

 

We like the concentration to be studied

 

that is going to be marketed, we like

 

concentrations that are above what is going to be

 

marketed to be studied to try and exaggerate

 

potential adverse events so that we can get a

 

handle of potential adverse events that may occur,

 

even if they are not going to occur on the final

 

product that is approved, so that we have some idea

 

of what to look for. And, we would like the

 

frequency of dosing to be at least as frequent as

 

proposed for marketing. You will see in the trials

 

we discuss today dose-ranging studies that look at

 

different concentrations.

 

[Slide]

 

The duration, as I mentioned, should be at

 

least 24 months but, as I also said, the endpoint

 

could be as short as 12 months.

 

[Slide]

 

16

 

We do not require multicenter trials. It

 

is certainly easier to enroll larger number of

 

patients with multicenter trials. Our preference,

 

if a company is going to do a multicenter trial, is

 

that there be at least 10 patients per arm per

 

center. We have set that number so that we can

 

look at investigator interaction. Now, that is

 

frequently a difficult thing, to enroll that many

 

patients per arm per center, particularly if you

 

have a multi-arm study and you are doing dose

 

ranging. That dramatically increases the number of

 

patients you would have at a particular center.

 

You need to recognize that if we do not

 

have that many we are probably not going to be able

 

to look for investigator interaction at any one

 

particular center. We will do some other things to

 

look at that question but to get a true, you know,

 

is there one investigator that is disproportionate

 

to other investigators really requires more

 

patients than you will see in these particular

 

trials. This is not an uncommon problem that we

 

have. We don't have a solution. Generally, if you

 

17

 

are able to enroll a large number of patients at

 

any one center you probably wouldn't do a

 

multicenter trial. So, again, we welcome

 

suggestions on how to get around this.

 

[Slide]

 

Stratification is not necessary. If there

 

is a chance of imbalance in factors that someone

 

believes may influence the results, and in this

 

case there have been discussions about whether

 

occult versus classic potentially would influence

 

the results or whether baseline visual acuity would

 

potentially influence the results. We have

 

suggested that people stratify so that they have a

 

higher chance of having an equal distribution

 

between the individual groups--again, not required.

 

The hope is that randomization will take care of it

 

but stratification frequently helps.

 

[Slide]

 

Control agent--we have asked that at least

 

one of the clinical trials that is performed

 

demonstrates superiority to a control. We have not

 

defined what that control has to be. It could be

 

18

 

the vehicle; it could be a sham; it could be a

 

lower dose; it could be a different product. By

 

saying at least one trial has to demonstrate

 

superiority, that means we also potentially would

 

accept an equivalence trial. In today's discussion

 

we are going to deal primarily with superiority

 

trials but, recognize, we potentially would accept

 

either a superiority trial or an equivalence trial.

 

We prefer a vehicle control given our

 

druthers of different choices, and we prefer that

 

because it minimizes the bias. There is some

 

animal evidence--we are not aware of any human

 

evidence to date but there is some animal evidence

 

that mechanical manipulation may initiate

 

inflammatory mediators that may help the condition.

 

Consequently, by not having something that

 

simulates that same pathway, there may be some

 

influence going on by the way you deliver the

 

product, in this case the intravitreal injection,

 

that may be a positive effect. But there are

 

ethical issues, and I am sure we will probably get

 

into some of that, with giving vehicle controls.

 

19

 

One of the most common reasons cited for

 

not giving a vehicle control is the risk of

 

endophthalmitis. We recognize that there is a

 

theoretical risk of getting endophthalmitis in the

 

vehicle group. The clinical trials that were

 

performed here had cases of endophthalmitis that

 

were in the active control group.

 

I just want to be on record for stating

 

that, to the agency's knowledge, we have not had a

 

case of endophthalmitis in the vehicle control

 

group in any trial that has run that, and there

 

have been trials that have run it. So, we continue

 

to think it is not unethical to run a vehicle

 

control. Should we get an endophthalmitis case,

 

which I am not hoping for anyone, we may change

 

that opinion but at the present time we continue to

 

recommend vehicle controlled trials.

 

We do reluctantly accept sham controls,

 

but we have put a condition any time we have

 

accepted sham controls, and that has been that we

 

have wanted additional doses, in other words, more

 

than one dose tested to try to aid in the masking

 

20

 

of the trial. You will see that in the case of

 

these trials today there are multiple doses, in

 

addition to the sham, that is conducted. Again, we

 

recognize that having a sham increases the chance

 

of bias influence in the results, although just

 

having a sham does not necessarily create bias.

 

[Slide]

 

Dose ranging--we prefer to try and bracket

 

the dose that will ultimately be marketed, in other

 

words, study doses that are higher and study doses

 

that are lower than that which will be ultimately

 

marketed so we get a better understanding of the

 

drug product.

 

[Slide]

 

Efficacy has been discussed a lot. We

 

have a number of parameters that we readily accept

 

as being acceptable. We have other parameters

 

which we think may in the future be acceptable or

 

we will be willing to entertain if there is

 

validation, and validation does not necessarily

 

need to occur in this particular trial. The thing

 

that we readily accept as being important is a

 

21

 

change in visual function. So, our guidance to

 

people when we are having discussions about

 

clinical trials is that there be statistical

 

significance in clinical relevance in visual

 

function at more than one time point. By visual

 

function we mean visual acuity, visual fields or

 

color vision.

 

[Slide]

 

The evaluations we expect to be carried

 

out include, obviously, best corrected distance

 

visual acuity. By that, we generally mean using a

 

chart that has equal number of letters per line and

 

equal spacing between lines. The ETDRS is one type

 

of chart that meets that, and based on the

 

validation information that was conducted at a four

 

meter distance so that is our preferred both

 

distance and test but we are willing to recognize

 

other equivalent tests of best corrected distance

 

visual acuity.

 

We expect best corrected visual acuity to

 

be measured at every visit, and we expect those

 

visits to occur no less frequently than every three

 

22

 

months.

 

We expect to have dilated seven field

 

fundus photography sometime during the trial. We

 

expect to have fluorescein or indocyanine green

 

depending on what exactly is being studied during

 

the trial, and we have not specified exactly when

 

that has to be. We expect dilated ophthalmoscopy

 

to be performed both for evaluation and for safety

 

at every visit. We expect a dilated slit lamp exam

 

for the same reason. We expect to have endothelial

 

cell counts, not necessarily in every trial but

 

somewhere within the development plan, and have at

 

least one study that includes it at the beginning

 

and end of the trial, and the same thing standard

 

systemic clinical and laboratory evaluations.

 

[Slide]

 

Two meters versus four meters has been a

 

source of a lot of controversy. It is my

 

understanding it stems primarily from the

 

practicality of being able to have exam rooms that

 

are four meters. In my father's day and age, it

 

would have required 20 foot length and his exam

 

23

 

rooms were set up to do that. That is not the

 

current trend now. People use exam lanes that are

 

much shorter. But the subject has been studied.

 

It was the source of a lot of discussion in the

 

past, and there is a paper that set out four meters

 

as a standard that was published in Ophthalmology

 

in 1996 for exactly the purpose of discussing what

 

the best distance is.

 

It does not mean that you can't

 

theoretically correct. You know, two meters, four

 

meters--you can use the same distance and make the

 

charts smaller so you are looking at the same angle

 

that gets subtended. The issue is the variability

 

that occurs when measuring at two meters versus

 

four meters and the potential for any bias if the

 

patient is allowed to lean. Now, if we would strap

 

down or lock every patient into an exam seat and

 

never let them move at all, it probably wouldn't be

 

an issue but we don't do that. Just so people get

 

a feel, at a two meter distance 17 inches is equal

 

to one of one line. Those of you sitting in the

 

various seats, if you are leaning backward or

 

24

 

leaning forward, just sitting in your same seat can

 

easily do 17 inches. We don't have any reason to

 

believe that people are attempting to bias the

 

results or attempting to lean, and visual acuity is

 

a very common measure in ophthalmology so everybody

 

is aware to try to keep people from leaning or keep

 

that from influencing what goes on. But studies

 

have been done that show poor reliability at one

 

meter versus four meters. So, the assumption is

 

that there is also more variability at two meters

 

than there would be at four meters. The overall

 

impact on a particular trial is not known, and the

 

only way to know that for sure would be to do both

 

two meters and four meters, which we do not have

 

data to discuss today.

 

We think it is more significant for those

 

trials that have a feature that allows there to be

 

a potential in masking, such as sham. We think it

 

is more of an issue in an equivalence trial than it

 

is in a superiority trial. These trials that we

 

are talking of today are superiority trials; they

 

are not equivalence trials. But there are issues.

 

25

 

[Slide]

 

Our recommended endpoints to date have all

 

been, as I mentioned earlier, visual function. We

 

think at some point in the future we will end up

 

accepting anatomical changes but we have not yet

 

found anatomical changes that correlate directly

 

with visual function. So, currently we readily

 

accept doubling of the visual angle, which on the

 

ETDRS chart at four meters would be 15 letters or

 

more; a halving of the visual angle, in other

 

words, showing improvement in vision; a quadrupling

 

of the visual angle, which would be 30 letters or

 

more. These are all looking at percentage of

 

patients that have this particular finding because

 

we think a doubling of the visual angle is a

 

clinically significant difference that would not

 

occur within the variation of day-to-day visits.

 

[Slide]

 

We have also been willing to accept a

 

difference in the group mean. We do not know

 

exactly how much of a difference in group mean

 

would be clinically significant so for

 

26

 

consistency's sake we have said we will readily

 

accept a mean change of 15 letters. That does not

 

mean that something less than that may not be

 

statistically significant. We are just not ready

 

to accept without question anything less than 15

 

letters.

 

[Slide]

 

Let me just briefly talk about equivalence

 

trials just so you know the full scope of what we

 

have talked about with individual sponsors. We

 

believe it is possible to do comparison with an

 

active agent which has already demonstrated

 

repeated success. Visudyne is currently approved

 

for predominantly classic choroidal

 

neovascularization in atrial macular degeneration

 

and a couple of other things. So, for that

 

particular indication we would accept an

 

equivalence trials if one wanted to conduct it.

 

The way we have set up equivalence trials is that

 

we have asked that at least 50 percent of the

 

established treatment effect be preserved so that

 

95 percent confidence intervals be drawn around

 

27

 

those parameters to protect at least 50 percent of

 

the treatment effect. Again, it is not a

 

particular issue for this product but it may be an

 

issue for other products.

 

The analyses that we ask to be conducted

 

always include intent-to-treat with last

 

observation carried forward and per-protocol with

 

observed values only. We recognize these as

 

differences in the data available for analysis.

 

The intent-to-treat last observation carried

 

forward is the fullest data set we can obtain. It

 

is everybody that was randomized in the trial and

 

it is creating a value for everyone whether real of

 

extrapolated. A per-protocol analysis is the

 

minimal data set. It is only those patients that

 

fully met the protocol and only the values that we

 

have there.

 

We don't believe that either one of these

 

two analyses is the best analysis or is the most

 

proper or is the most representative. We think

 

they are extremes and we ask that both be conducted

 

and we look for differences between these two

 

28

 

analyses. If there are no differences between

 

these analyses we assume that, regardless of how

 

much inclusion/exclusion, your results are pretty

 

much the same and you can accept either one. If

 

there are differences we ask for additional

 

analyses to try and explore which one is likely to

 

be telling a better picture or why it is telling a

 

different picture.

 

Other analyses which you would have seen

 

in the briefing package include things like

 

worst-case analyses where we treat all dropouts in

 

the control as being successes and all dropouts in

 

the test product as being failures. This is not a

 

correct test. This is not an accurate test. We

 

are making assumptions in the worst direction to

 

look and see how robust the findings are. We don't

 

expect the product to win on a worst-case analysis,

 

but it does give us an idea of what the limits of

 

potential analysis results could be.

 

[Slide]

 

As a general rule, we ask for alphas to be

 

0.05. This is the common 5 percent for two-tailed.

 

29

 

In other words, p is less than 0.5. We ask for

 

power to detect a difference to be 80 percent or

 

greater, and we ask that any time anybody looks at

 

the data, any kind of look any time during the

 

evaluation that there be an adjustment in the

 

statistical plan, in other words, correction for

 

that alpha for any look that occurs. All of our

 

analyses that you see in any of our data sets will

 

include these features.

 

[Slide]

 

The last one I want to talk about is

 

pediatrics. There is an agency initiative to try

 

and include, when possible, pediatric patients in

 

the drug development of particular products. So, I

 

am covering it for completeness. In this particular

 

case, choroidal neovascularization is rarely seen

 

in pediatric populations and we have not asked the

 

sponsor of this application or any of the

 

applications that just deal with choroidal

 

neovascularization to include pediatric patients

 

because the population we don't think is relevant

 

in this particular case. But as a general rule we

 

30

 

do ask for pediatric patients to be included during

 

the development.

 

I am happy to take any questions and,

 

again, I thank everybody for your time, and look

 

forward to a fruitful discussion.

 

DR. DUNBAR: Are there any questions at

 

this point regarding Dr. Chambers' presentation?

 

If not, at this point then I would like to open the

 

forum for the sponsor, Eyetech Pharmaceuticals and

 

I will ask that the sponsor introduce each of their

 

speakers within their presentation.

 

Eyetech Pharmaceuticals Presentation

 

Introduction

 

DR. DYER: Good morning.

 

[Slide]

 

Today we will discuss the first anti-VEGF

 

therapy for the eye and the first treatment to

 

target the underlying biology of neovascular

 

age-related macular degeneration. Pegaptanib

 

sodium achieved statistical significance for

 

clinically meaningful, prespecified primary

 

endpoint in replicate trials with strong supportive

 

31

 

data in secondary endpoints.

 

The efficacy was against usual care

 

controls, and this pharmacological agent also shows

 

a favorable safety profile and provides a treatment

 

benefit to many patients for whom no effective

 

therapy presently exists.

 

[Slide]

 

My name is David Guyer. I am from Eyetech

 

Pharmaceuticals. I previously was professor and

 

chairman of ophthalmology at the N.Y. School of

 

Medicine and a practicing ophthalmologist

 

specializing in macular degeneration.

 

Also speaking today will be Dr. Tony

 

Adamis, who was an ophthalmologist on the full-time

 

faculty at Harvard, and is now with Eyetech. He

 

ran the ocular angiogenesis laboratory as well.

 

Our risk/benefit section will be presented by Prof.

 

Don D'Amico, from Mass. Eye and Ear Infirmary at

 

Harvard.

 

[Slide]

 

Neovascular age-related macular

 

degeneration represents 90 percent of the severe

 

32

 

vision loss from this disease. Many patients note

 

a loss of independence and inability to read, to

 

ambulate and to recognize faces of their loved

 

ones. This occurs because when the disease forms

 

abnormal blood vessels that leak blood and fluid

 

waviness or blurred vision can be seen in the

 

central area that sometimes can lead to a scotoma

 

or blind area centrally that prevents them from

 

seeing straight ahead, and in up to a third of

 

patients clinical depression can be noted.

 

[Slide]

 

The devastating effects of this disease

 

were well summarized in a book by Henry Grunwald,

 

who was the former editor-in-chief of Time Magazine

 

and U.S. ambassador. In the book, "Twilight:

 

Losing Sight, Gaining Insight" Mr. Grunwald said,

 

"after a lifetime during which reading and writing

 

have been as natural and necessary as breathing, I

 

now feel the visual equivalent of struggling for

 

breath."

 

[Slide]

 

Macular degeneration represents a major

 

33

 

public health problem and urgent unmet medical

 

need. It is the most common cause of irreversible,

 

severe blindness in developed countries.

 

Ninety-five percent of retinal specialists believe

 

that macular degeneration represents an epidemic,

 

and there are 200,000 new cases a year in the

 

United States alone, and a prevalence of up to 1.6

 

million patients with active bleeding. Limited

 

treatments are available and 85 percent of retinal

 

specialists are dissatisfied with current treatment

 

options.

 

[Slide]

 

Macular degeneration represents a

 

progressive disease. Early on in the disease these

 

whitish-yellow spots, called drusen, occur and

 

patients can progress to the neovascular form of

 

the disease which is where pegaptanib is effective.

 

This is an angiogenic disorder and what happens is

 

abnormal blood vessels grow behind the retina where

 

they leak blood and fluid, as depicted here, and,

 

untreated, they lead to disciform scarring where

 

fibrovascular tissue destroys and replaces the

 

34

 

normal rods and cones in the retina. At this

 

point, usually moderate to severe visual loss is

 

noted.

 

[Slide]

 

Let's discuss the therapeutic options

 

available for patients with neovascular macular

 

degeneration. In the 1980s, the Macular

 

Photocoagulation Study Group showed the beneficial

 

roles of thermal laser photocoagulation. However,

 

very few patients are suitable for this treatment.

 

The treatment is most suitable when the abnormal

 

blood vessel, as seen here on a fluorescein

 

angiogram, is away from the center of the macula,

 

in what we call extrafoveal or juxtafoveal

 

location, because for patients where the blood

 

vessel is dead center or subfoveal the laser scar

 

itself can destroy the very tissue we are trying to

 

save. Unfortunately, most patients with

 

neovascular macular degeneration have subfoveal

 

disease where the blood vessel is dead center.

 

[Slide]

 

In the year 2000, photodynamic therapy, or

 

35

 

PDT, was FDA approved for patients with subfoveal

 

predominantly classic angiographic subtype. Thus,

 

for approximately three-quarters of patients with

 

neovascular macular degeneration there is no FDA

 

approved therapy, although there is off-label use,

 

with some limited CMS reimbursement, presently.

 

Today we will discuss the first anti-VEGF

 

therapy for the eye, a pharmacological treatment

 

that targets the protein VEGF that is responsible

 

for the hallmarks of all choroidal

 

neovascularization. Increased levels of VEGF lead

 

to neovascularization and increased permeability,

 

which lead to the clinical features of all

 

choroidal neovascularization, and pegaptanib blocks

 

VEGF.

 

[Slide]

 

VEGF is the common denominator for

 

neovascular macular degeneration. Numerous peer

 

reviewed papers have shown that for all

 

angiographic subtypes, by immunohistochemistry

 

staining, VEGF is present in both autopsy and

 

surgical specimens.

 

36

 

[Slide]

 

Pegaptanib sodium is a pegylated modified

 

oligonucleotide. It has a selective vascular

 

endothelial growth factor antagonist to isoform

 

165. Tony in just a few minutes. It is a sterile

 

aqueous solution in a single-use, pre-filled

 

syringe, which is important for safety reasons.

 

The recommended dose is 0.3 mg of intravitreous

 

injection administered once every 6 weeks.

 

[Slide]

 

We will show you today that pegaptanib met

 

a clinically meaningful primary efficacy endpoint

 

with statistical significance in replicate,

 

well-controlled clinical trials, with a favorable

 

safety profile.

 

[Slide]

 

I will now ask Tony Adamis to discuss a

 

VEGF overview and macular degeneration

 

pathophysiology.

 

VEGF Overview and Macular Degeneration

 

Pathophysiology

 

DR. ADAMIS: Thank you, David and good

 

37

 

morning.

 

[Slide]

 

In 1971 Judah Folkman first proposed the

 

targeting of a specific angiogenic factor as a way

 

to treat disease, and specifically a way to treat

 

cancer and ophthalmic disease.

 

[Slide]

 

It was in 2004, with the completion of

 

pivotal Phase III trials using Avastin which blocks

 

VEGF that this theory was in a definitive fashion

 

proven correct. This drug now was approved this

 

year as a first-line therapy for colon cancer. So,

 

we entered this era of biological anti-angiogenesis

 

therapy.

 

[Slide]

 

The target in that trial and in our trial

 

is VEGF, which is an acronym for vascular

 

endothelial growth factor. Prior to that it was

 

called vascular permeability factor. Unlike many

 

other growth factor names, these two are very

 

appropriate in the sense that they describe the

 

central biological functions of this protein. VEGF

 

38

 

makes vessels very leaky and VEGF makes vessels

 

grow. The leaky aspect of it was discovered in

 

1983 by Harold Dvorak and then the

 

neovascularization aspect or biology of VEGF was

 

discovered by Napoleon Ferrara, who has been a

 

leader in this area, and Dan Connolly, in 1989.

 

Since then, if one conducts a MEDLINE

 

search, there have been over 11,000 published peer

 

reviewed articles on VEGF. There is a large body

 

of knowledge concerning this growth factor. I show

 

you just one example of that here. This is the

 

protein structure of VEGF. We now can determine

 

very precise structure-functional relationships.

 

[Slide]

 

The disease we are here to discuss, as

 

David said, is age-related macular degeneration, a

 

very prevalent disease in our society and a very

 

complex one scientifically when one begins to study

 

it. We are beginning to unravel the earlier stages

 

of the disease, the stages where Bruch's membrane

 

is altered and gives you those yellow spots, the

 

drusen that David showed you in a clinical

 

39

 

photograph. We are also starting to understand the

 

complex interaction of the different cell layers

 

with the vasculature. But the area or the phase of

 

the disease, the late phase of the disease that we

 

are studying is the neovascular phase where vessels

 

begin to grow up towards the retina. These vessels

 

are abnormal and leaky, and they leak fluid and

 

lipid and they damage the photoreceptors which

 

sense light, and people lose vision and go blind.

 

This process, the angiogenic process, has been very

 

well studied.

 

[Slide]

 

As David said, the data indicate that it

 

is biologically plausible that blocking VEGF would

 

have a beneficial effect in this disease in a broad

 

population. When one looks at surgical specimens

 

or autopsy specimens of patients with the disease,

 

what is seen is that the common denominator is

 

VEGF. It is present in all angiographic subtypes

 

and it is present in all active stages of the

 

disease. So, therefore, the hypothesis that

 

blocking VEGF in neovascular MD would have a

 

40

 

broad-base effect has some broad biological

 

plausibility.

 

[Slide]

 

But those are not the only data that we

 

have. There is a large body of preclinical

 

evidence, roughly 15 years worth, which is

 

summarized on one slide here. Let me just briefly

 

walk you through it. In preclinical models of

 

vessel growth in the cornea, in the iris, in the

 

retina and in the choroid, if one gives a VEGF

 

inhibitor you can prevent the growth of vessels and

 

you can prevent the leak that is associated with

 

those vessels. So, VEGF seems to be required for

 

those processes.

 

Similarly, if one looks at those normal

 

tissues and now introduces VEGF into the system,

 

either by injecting the protein or genetically

 

over-expressing it, VEGF in and of itself is

 

sufficient to produce the neovascularization or

 

leak that can occur in these tissues.

 

Then, so that we have some context in

 

which to interpret those preclinical data, surgical

 

41

 

specimens and autopsy specimens from humans with

 

actual corneal neovascularization, iris

 

neovascularization, retinal and choroidal

 

neovascularization show that VEGF is expressed at

 

high levels in those tissues at the time when the

 

vessels are growing and leaking. So, the totality

 

of the data supports this approach of blocking VEGF

 

in specifically the disease under study today,

 

age-related macular degeneration.

 

[Slide]

 

It gets a little more complicated in the

 

sense that VEGF really refers to a family of

 

related molecules, and I want to talk about one

 

specifically, VEGF 165 which is the target of

 

pegaptanib.

 

[Slide]

 

We were faced with the paradox a few years

 

ago, as we looked at the accumulated data

 

concerning the role of VEGF in disease and in the

 

normal state. What we found was that VEGF is

 

required for the normal formation development of

 

vessels during development throughout the body. I

 

42

 

am just showing you here two examples. These are

 

the vessels of the normal colon and these,

 

obviously, are the normal vessels of the retina.

 

[Slide]

 

In the same molecule, VEGF was shown in a

 

number of definitive studies and laboratories

 

around the world that VEGF is required for the

 

abnormal vessels that can grow in the colon, and

 

this is colon carcinoma, and here is a case of

 

age-related macular degeneration. So, how is it

 

that the same protein can cause these vastly

 

different phenotypes, these different types of

 

vessels? One set of vessels are normal and they

 

don't leak and they behave appropriately; another

 

set looked very different and they behave very

 

differently.

 

[Slide]

 

Perhaps, we thought, some of that

 

complexity is encoded in these different isoforms.

 

Let me just explain what those are. There is one

 

VEGF gene but that gene encodes multiple

 

transcripts or mRNAs for VEGF that have different

 

43

 

sizes that translate into different proteins. So,

 

one of those major proteins or isoforms is VEGF

 

165, which just simply means that it is composed of

 

165 amino acids. Another major isoform, especially

 

in the eye, is VEGF 121. We asked the question

 

could it be that differential expression or

 

synthesis of these isoforms underlies the

 

complexity that we see in the vessels in the normal

 

and the diseased state?

 

[Slide]

 

So, in an experiment we conducted and

 

published last year, we studied the retinal

 

vessels. We studied the normal retinal vessels

 

that are developing as the retina forms and we

 

studied abnormal retinal vessels in a model of

 

retinopathy prematurity. This is a model where

 

vessels grow towards the vitreous and leak and are

 

distinctly abnormal.

 

What we saw was that when normal vessels

 

are developing the isoform expression of the two

 

major isoforms, 120 and 164 which are the rodent

 

counterparts to human 121 and 165, is roughly equal

 

44

 

during development. But rather strikingly, during

 

disease when disease vessels are growing there is a

 

shift to almost exclusive expression of the 164

 

isoform. So, it was an interesting association

 

that we saw of 164 with diseased vessels.

 

[Slide]

 

But to really get at the causality of 164

 

in the production of diseased vessels we conducted

 

the following experiment. In a model of abnormal

 

vessel growth we gave pegaptanib which blocks just

 

164 and compared it to a non-selective VEGF

 

inhibitor which blocks all the isoforms. We saw

 

that bpth were equally effective in preventing

 

abnormal vessel growth. Here is the control with

 

the abnormal vessels, and both are pretty good at

 

inhibiting that.

 

We also looked in a model of normal

 

retinal vessel development and, again, gave

 

pegaptanib and what we saw was essentially zero

 

inhibition of normal vessels. We did not affect

 

normal vessels. Whereas, the non-selective VEGF

 

inhibitor had a deleterious effect on these normal

 

45

 

vessels in the retina. So, the conclusion we made

 

was that VEGF 164 may be preferentially associated

 

with disease and targeting it gives you a much more

 

selective inhibition in that you are much less

 

likely to affect normal vessels in the developing

 

animal. But I will tell you that there has

 

subsequently been independent support of this,

 

specifically from UCSF, where this is also perhaps

 

true in the adult animal.

 

[Slide]

 

To be certain of our conclusion because we

 

used a reagent here, pegaptanib in particular, we

 

wanted to make sure this conclusion was robust.

 

So, we created animals genetically that where we

 

deleted specifically the 164 isoform and these

 

animals were able to make all the other types of

 

VEGFs. What we see here is that these animals have

 

completely normal retinas and normal retinal

 

vessels and they are no different than animals that

 

make all VEGF isoforms. In fact, these animals

 

grow up to a normal age. They can reproduce.

 

There are no abnormalities we can detect, even

 

46

 

though they cannot make any VEGF 164.

 

[Slide]

 

So, how was a drug made that specifically

 

blocked VEGF 164? Well, pegaptanib is an

 

oligonucleotide aptamer. It specifically is 28

 

nucleotide in life. Aptamers are molecules that

 

will fold in a very specific fashion. They have a

 

three-dimensional conformation such that they will

 

bind to the target protein of interest--in this

 

case it is VEGF--in a highly specific manner, and

 

in the case of pegaptanib with a very high

 

affinity. This binding occurs extracellularly.

 

The drug does not enter the cell. It is all

 

happening outside the cell, which is where VEGF is

 

residing. These features make it act very much

 

like an antibody but there are some important

 

distinctions, aside from it not being an antibody;

 

it is an oligonucleotide.

 

This class of molecules, in the published

 

literature and it has been our experience as well,

 

are quite non-immunogenic. In our preclinical and

 

in our clinical examination of pegaptanib we have

 

47

 

not seen a single instance when an antibody is

 

raised to it. And, as I alluded to, they have this

 

remarkable target specificity and this simply

 

attests to that.

 

[Slide]

 

This shows that pegaptanib is very

 

efficiently binding to human VEGF 165 and murine or

 

mouse VEGF 164, but there is no significant, or

 

essentially no binding to VEGF 121 or related

 

family member of placental growth factor.

 

[Slide]

 

So, what we would expect when pegaptanib

 

is administered to the eye is that you would have

 

selective VEGF inhibition of 165 which was

 

associated with pathology and in our animal model

 

spares the normal vasculature, and we would have

 

two very important biological responses as a

 

function of that blockade: vessel growth would be

 

inhibited, as would permeability, and the thinking

 

was this would translate to a better visual

 

outcome.

 

[Slide]

 

48

 

The last thing I would like to talk about

 

is how we chose our dose. This drug is

 

administered to the eye nine times a year, and

 

there are three doses that we chose.

 

[Slide]

 

Let me show you the data that we had in

 

hand when we were planning these trials. We knew

 

from our pharmacokinetic experiments that when

 

pegaptanib is given to the eye via intravitreous

 

injection it slowly exits the eye and it can be

 

measured in the plasma. Actually, the plasma

 

levels mirror the levels that one sees in the

 

vitreous. So, by sampling the blood you can infer

 

what is happening in the eye.

 

The other important thing that we learned

 

here is that when the drug exits the eye, at least

 

in this rabbit model, you have thousand-fold less

 

concentration in the plasma than you do in the eye.

 

In a more relevant primate model we saw that this

 

held up in the sense that it was 800 times less in

 

the plasma than it was in the eye.

 

[Slide]

 

49

 

We learned from those studies that the

 

half-life in the primate vitreous is approximately

 

four days. We also had data that we had collected

 

in tumor models and in a model of retinopathy

 

prematurity that when you give pegaptanib

 

intravenously the amount of pegaptanib that is

 

needed to inhibit the VEGF is about 1 ng/mL.

 

We also had another inhibitory

 

concentration that we had determined in vitro in

 

tissue culture in various assays of calcium

 

mobilization and endothelial cell proliferation.

 

The relevant concentration in tissue culture of

 

pegaptanib that was required to inhibit VEGF was

 

significantly lower. It was 0.01 mcg/mL or 10

 

ng/mL.

 

When we started out it was not entirely

 

clear which of these inhibitory concentrations

 

would be most relevant when you are injecting the

 

drug into the eye. So, we postulated that if this

 

is the most relevant inhibitory concentration, then

 

a 3 mg dose, given every 6 weeks would sufficient

 

block VEGF for the entire 6-week period. If, on

 

50

 

the other hand, this was the relevant

 

concentration, the 3 mg dose, the 1 mg dose and the

 

0.3 mg dose would actually all three be sufficient

 

to block VEGF for the entire 6-week period, and

 

perhaps that may translate to a plateau of the dose

 

response.

 

[Slide]

 

To summarize what I have just discussed,

 

VEGF appears to be an important control point for

 

neovascularization and vascular permeability, the

 

pathologies that lead to vision loss in age-related

 

macular degeneration. Pegaptanib specifically

 

targets the VEGF isoform VEGF 165, which we believe

 

is operative in disease. I have shown you data

 

from ROP but this has also been shown to be true in

 

choroidal neovascularization, diabetic retinopathy

 

and other conditions. And, pegaptanib dosing is

 

based on pharmacokinetic data which were collected

 

prior to the conduct of this study.

 

[Slide]

 

At this point, Dr. David Guyer will return

 

and David will talk to you about our clinical

 

51

 

efficacy data from the pivotal trials.

 

Pegaptanib Clinical Efficacy

 

[Slide]

 

DR. GUYER: In this section we will show

 

you that pegaptanib met a clinically meaningful

 

primary efficacy endpoint with statistical

 

significance in independent, well-controlled,

 

replicate trials, with a favorable safety profile.

 

[Slide]

 

The macular degeneration program consisted

 

of 6 trials, 1,281 patients and over 10,000

 

treatments at 117 sites in 21 countries. The dose

 

ranges that were studied ranged from 0.25 mg to 3

 

mg per eye.

 

[Slide]

 

These are the six trials. EOP1003 and

 

1004 are pivotal trials, sham-controlled,

 

double-masked, randomized trials. There were 622

 

patients in the predominantly ex-U.S. trial and 586

 

in trial 1004 in North America. The other four,

 

smaller trials were pharmacokinetic trials and

 

open-label single or multiple dosing trials with,

 

52

 

or without PDT, for the total exposed of 1,281.

 

[Slide]

 

The Phase I/II program showed that

 

pegaptanib appeared safe in all tested doses and

 

regimens with no dose-limiting toxicities. There

 

were no unexpected retinal or choroidal

 

abnormalities noted by angiography as read by an

 

independent reading center. As Tony mentioned,

 

these trials established the dosing regimen based

 

on pharmacokinetics.

 

[Slide]

 

The study objective of the pivotal trials

 

was to establish a safe and efficacious dose of

 

intravitreous pegaptanib sodium in patients with

 

subfoveal choroidal neovascularization secondary to

 

age-related disease.

 

[Slide]

 

The development of these pivotal studies

 

was done in conjunction with our expert advisory

 

panel, whose names are listed on this slide.

 

[Slide]

 

The study design was two randomization,

 

53

 

double-masked, sham-controlled, dose-ranging trials

 

of pegaptanib 0.3 mg, 1 mg and 3 mg and sham. The

 

treatment regimen was every 6 weeks and the

 

prespecified time point for the primary endpoint

 

was 54 weeks. PDT, photodynamic therapy, was

 

permitted per the FDA-approved label at the masked

 

investigator's discretion. Since shams could have

 

PDT, this represented a usual care control group.

 

[Slide]

 

Independent monitoring was done both by an

 

independent reading center that confirmed the

 

eligibility prior to randomization, and an

 

independent data safety monitoring committee, or

 

IDMC.

 

[Slide]

 

These were the members of the IDMC. It

 

was chaired by Prof. Alan Bird, who is here with us

 

today.

 

[Slide]

 

Because of the biology of neovascular

 

macular degeneration and the mechanism of action of

 

pegaptanib, we designed a trial with a very wide

 

54

 

range of inclusion criteria which included a broad

 

range of visual acuities, 20/40 to 20/320, and

 

broad angiographic criteria including all subfoveal

 

angiographic subtypes; lesion sizes up to and

 

including 12 total disc areas in size; greater than

 

or equal to 50 percent of the total lesion size

 

needed to be active choroidal neovascularization;

 

and for minimally classic and occult disease

 

subretinal hemorrhage and/or lipid and/or recent

 

change in vision was necessary for inclusion.

 

[Slide]

 

Ocular exclusion criteria included

 

previous subfoveal thermal laser therapy, and to

 

avoid older chronic cases any subfoveal scarring or

 

atrophy or greater than or equal to 25 percent of

 

the lesion being scarred or atrophic. Causes of

 

choroidal neovascularization other than age-related

 

diseases were excluded, and if a patient had recent

 

intraocular surgery or was thought to perhaps need

 

cataract surgery in the near future, they also were

 

excluded. Finally, no more than one prior PDT

 

treatment was allowed.

 

55

 

[Slide]

 

The general exclusion criteria included a

 

history or evidence of severe cardiac disease such

 

as myocardial infarction within the last 6 months,

 

ventricular tachyrhythmia or unstable angina;

 

evidence of peripheral vascular disease; or

 

clinically significant hepatic or renal

 

dysfunction; or a stroke within the last 12 months.

 

Our population, however, was very characteristic of

 

your typical elderly population in that 50 percent

 

of the patients had systemic hypertension; 25

 

percent were on statins; and 20 percent had

 

cardiovascular disease.

 

[Slide]

 

Stratification at randomization included

 

study center, a history of prior PDT use and

 

angiographic subtype.

 

[Slide]

 

Our primary efficacy endpoint, which was

 

prespecified, was the percent of patients losing

 

less than 15 letters from baseline to week 54, the

 

same endpoint that was used for marketing approval

 

56

 

of Visudyne.

 

This is an ETDRS chart where 5 letters

 

equal 1 line, and the 15-letter change or 3-line

 

change represents a doubling of the visual angle

 

which is a clinically meaningful change to an

 

individual patient.

 

[Slide]

 

Our primary endpoint used in

 

intent-to-treat, or ITT, population included

 

patients receiving at least one treatment and a

 

baseline visual acuity measurement. The last

 

observation carried forward, or LOCF, was used to

 

impute missing data. We will also discuss

 

supportive visual and angiographic endpoints, as

 

well as exploratory or subgroup analyses.

 

[Slide]

 

This table shows the various study visits.

 

Of note, a telephone safety check was done 3 days

 

after treatment. Tonometry or measurement of

 

intraocular pressure was done both before treatment

 

and 30 minutes after, and fundus photography and

 

fluorescein angiography was done at baseline and

 

57

 

weeks 30 and 54.

 

[Slide]

 

In order to preserve the integrity of the

 

masking there were two physicians involved in the

 

trial. One physician administered the study

 

treatment and the second physician was involved in

 

any patient assessments or decisions. Patients

 

were also masked in that the sham procedure was

 

identical to the active drug procedure except for

 

the actual penetration into the vitreous. This

 

meant that they had application of a lid speculum,

 

instillation of topical medications,

 

subconjunctival anesthetic, and pressure against

 

the globe using a needle-less syringe.

 

The visual acuity examiners were also

 

masked to both he treatment arm and also to

 

previous vision assessments, and the reading center

 

was not aware of the patient's treatment arm.

 

[Slide]

 

This slide represents the patient baseline

 

characteristics for both trials 1004 and 1003.

 

What we can see in each trial is that the active

 

58

 

doses and the sham are well balanced with respect

 

to sex, age, initial visual acuity, angiographic

 

subtype, prior use of PDT and lesion size. The

 

only difference between the two trials was that

 

there was slightly more prior PDT use in trial

 

1004. That was the North American trial, and that

 

was because Visudyne was approved and reimbursed

 

earlier in the United States than in Europe. Out

 

of 9 possible injections, on average all patients,

 

treated and sham, received 8.5 of the 9 injections,

 

and overall there was about a 10 percent rate of

 

discontinuation in the trial.

 

[Slide]

 

We prespecified to use a Hochberg

 

procedure to account for the multiple doses in this

 

pivotal trial. As per agreement with the FDA, it

 

was decided to unmask study 1004 first--that was

 

the trial that was recruited first, thus, the

 

results were available earlier--in order to

 

determine which doses to formally analyze in the

 

study trial study, 1003.

 

[Slide]

 

59

 

So, we proceeded to unmask the first

 

trial, study 1004, and we found for the 0.3 mg dose

 

67 percent of patients lost less than 15 letters

 

compared to 52 percent of sham. This hit our

 

Hochberg adjusted p value at 0.0031. Note that the

 

1 mg dose had a similar response rate, about 66

 

percent. The p value was 0.0273. The 3 mg

 

response rate was higher than the shams at 61

 

percent, however, it did not hit the necessary p

 

value.

 

[Slide]

 

For this reason, prior to unmasking the

 

second trial, it was prespecified to the FDA that

 

only the 0.3 mg and 1 mg doses would be formally

 

analyzed in the second trial. Then we proceeded to

 

unmask the second trial, study 1003.

 

[Slide]

 

This study showed replication of the

 

findings of the first trial study, 1004, in that 73

 

percent of the patients in the 0.3 mg dose,

 

compared to 59 percent of sham, lost less than 15

 

letters, again hitting our Hochberg adjusted p

 

60

 

value of 0.0105. Again, the response rate in the 1

 

mg group was similar at 75 percent and a p value of

 

0.0035, and the response rate in the 3 mg group was

 

69 percent. The p value you see here, 0.1252 was a

 

nominal p value because we decided, as we

 

mentioned, not to formally analyze it.

 

[Slide]

 

So, we can look at the combined data and

 

see that 70 percent of the 0.3 mg group, 71 percent

 

of the 1 mg group and 65 percent of the 3 mg group

 

lost less than 15 letters compared to 55 percent of

 

the shams, and for all of these active treatment

 

groups we had low nominal p values.

 

It is important to emphasize that for the

 

0.3 mg group we were able to show independent

 

replication in two trials of a statistically

 

significant effect in a prespecified clinically

 

significant primary endpoint.

 

[Slide]

 

I would like to turn now to some

 

supportive visual angiographic analyses. There are

 

a variety of ways of looking at various visual

 

61

 

outcomes that are standard for reassurance that the

 

treatment effect for showing the primary endpoint

 

is real. As we will present, all of these analyses

 

were in favor of pegaptanib which gives us

 

confidence in this treatment effect. Because the

 

independent trials had the same protocol and

 

demographics, and because we prespecified it in our

 

statistical plan, we will present these as pooled

 

data.

 

[Slide]

 

This graph shows the percent responders

 

over time. What we can see is that we were able to

 

show that the active treatment group had a

 

treatment effect over sham not only at our primary

 

endpoint at 54 weeks, but at every studied time

 

point the active treatment group did better than

 

the sham.

 

[Slide]

 

This is a graph of mean change in visual

 

acuity. Again, the active treatment group is here,

 

the sham or usual care group showing a progressive

 

decrease in vision, and the difference at 54 weeks

 

62

 

was approximately 50 percent in favor of the active

 

treatment group.

 

[Slide]

 

This treatment effect was early and

 

sustained, by as early as 6 weeks, which was the

 

first visit after the first injection the

 

pegaptanib groups had already distinguished

 

themselves from the controls and, as we can see

 

here, the 0.3 mg and the 1 mg group had done that

 

with the low nominal p value. This sustained

 

itself throughout the 54-week course of treatment.

 

[Slide]

 

Sham eyes were twice as likely to suffer

 

severe vision loss than actively treated patients,

 

as shown in this graph of percent of patients with

 

severe vision loss. We can see the sham controls

 

with severe vision loss compared to the

 

active-treated groups.

 

[Slide]

 

At week 54, again, there was a low nominal

 

p value for the 0.3 mg and 1 mg group compared to

 

sham, with progression to severe vision loss which

 

63

 

is 30 letters or 6 lines.

 

[Slide]

 

This also was seen for legal blindness in

 

one eye, which is 20/200 or worse. We again can

 

see that more sham eyes progressed to 20/200 vision

 

or worse compared to actively-treated groups.

 

[Slide]

 

Patients on pegaptanib were also more

 

likely to maintain and/or gain visual acuity. This

 

graph shows the prespecified endpoints of

 

maintaining or gaining vision that is greater than

 

or equal to zero lines gained, as well as greater

 

than or equal to 3 lines gained. These other two

 

endpoints were not prespecified but we can see

 

again in all cases a treatment effect for

 

maintaining or gaining vision compared to sham.

 

[Slide]

 

The next few slides will show the

 

distribution of visual acuity change at baseline

 

and compared to week 54. Let's first look for the

 

0.3 mg group. This was the range of visual

 

acuities at baseline. Yellow is the 0.3 mg group

 

64

 

and blue is the sham.

 

[Slide]

 

After 54 weeks in the trial we can see

 

that more patients in the 0.3 mg treated group than

 

sham had good visual acuities and more patients

 

with sham than treated patients had poorer visual

 

acuity. So, the shift in distribution was in favor

 

of our 0.3 mg group, and the p value for this was

 

less than 0.0001.

 

[Slide]

 

The same is true, as we can see here, for

 

the 1 mg group. This is the baseline visual acuity

 

distribution and at 54 weeks again we can see more

 

1 mg treated patients than sham having relatively

 

good visual acuities and more shams than treated

 

eyes having poorer vision. Again, this shift in

 

distribution is in favor of the 1 mg group had a p

 

value of less than 0.0001.

 

[Slide]

 

Finally, we can see that for the 3 mg

 

group also. Here is the baseline distribution and

 

at 54 weeks again more 3 mg patients had better

 

65

 

visual acuities than shams, and more shams had

 

poorer vision at the end of 54 weeks than the 3 mg

 

treated patients.

 

[Slide]

 

This is a graph of the cumulative

 

distribution function of vision. What it shows on

 

the bottom is the change in visual acuity up to

 

week 54 and the cumulative proportion on this axis.

 

This shows the robustness of the data as it uses

 

all of the data points for 54 weeks.

 

What we can see first is this S-shaped

 

curve. This is the blue sham patients. You can

 

see here, for example, at minus 15--that is minus

 

15 letters which was our primary endpoint, moderate

 

for vision loss, and we see minus 30 which, as we

 

talked about, represents severe vision loss, and we

 

can see the zero or higher time point which

 

represented maintaining vision. What we can see is

 

that, whether we are talking about preventing

 

vision, maintaining vision or gaining vision, there

 

has been a shift in distribution, a shift in the

 

distribution of the sham patients in all active

 

66

 

treatment arms to the right, suggesting benefit in

 

all areas. The area between the lines which

 

represents this improvement was highly

 

statistically significant for all three doses, for

 

the 0.3 mg dose less than 0.0001; the 1 mg dose

 

0.0001 again; and the 3 mg dose 0.0017.

 

[Slide]

 

I would like to now turn to the

 

exploratory or subgroup analyses.

 

[Slide]

 

It is important to emphasize that this

 

study was powered to test for statistical

 

significance in the overall study population, that

 

is, to test for the primary hypothesis or primary

 

endpoint of all subjects. Nevertheless, it is

 

important to explore various baseline

 

characteristics such as lesion composition, lesion

 

size, baseline vision, age, sex and pigmentation of

 

the iris.

 

[Slide]

 

Despite a reduced ability to draw

 

statistical conclusions because of decreased sample

 

67

 

size, in some cases as small as 18 patients,

 

multiple subgroup analyses which can both lead to

 

false positives and negatives--despite this no one

 

subgroup drove the overall effect, as we will show

 

you.

 

[Slide]

 

We will first look at the 0.3 mg and 1 mg

 

doses as was described in the FDA briefing book.

 

We have also analyzed and prepared the 3 mg dose

 

and if people are interested later we can show you

 

that. We will present this using pooled data

 

because it was prespecified and we will show the

 

individual trials after.

 

[Slide]

 

Here we can see for the pooled data at the

 

0.3 mg dose that in all cases of all patient

 

characteristics the 0.3 mg active treated group did

 

better than sham. This was for sex, age and,

 

consistent with the biology of this disease and the

 

mechanism of action of pegaptanib, for all

 

angiographic subtypes, predominantly classic,

 

minimally classic and occult, as seen here; also,

 

68

 

initial baseline visual acuity, size of the lesion,

 

race and pigmentation of the iris.

 

[Slide]

 

Here we can see for severe visual

 

loss--the first graph was moderate visual loss or

 

primary endpoint, but we can see that the

 

conclusions we made are supported by severe visual

 

loss, or 6-line loss, 30-letter loss in this graph.

 

The blue are the sham so all had more severe vision

 

loss than actively treated 0.3 mg group for all

 

patient characteristics. So, this supports our

 

primary analysis.

 

[Slide]

 

Turning to the 1 mg group, we can see the

 

same thing, that in all patient characteristics the

 

1 mg group did better than sham. Again, we can see

 

that this information is supported by severe vision

 

loss where, again, sham in all cases did worse than

 

the actively treated 1 mg dose.

 

[Slide]

 

Let's now turn to the individual trials.

 

Individual trials which are under-powered

 

69

 

inherently have more variability. Nevertheless, we

 

can make the same conclusion, that no one subgroup

 

drove the overall efficacy. Again, for trial 1004

 

with the 0.3 mg group we can see the very small Ns,

 

sample sizes, for some of these groups and, again,

 

we can see support for using severe visual loss as

 

another important clinical endpoint.

 

[Slide]

 

For trial 1003, with the 0.3 mg dose we

 

can see the same thing.

 

[Slide]

 

For the 1 mg dose, again we can see, in

 

trial 1004, that in all cases the treated groups

 

did better than the controls and this was supported

 

by the severe vision loss in 1004 again.

 

[Slide]

 

And, in trial 1003, again, for moderate

 

vision loss treated patients did better than the

 

blue shams and support with severe vision loss

 

where shams did worse than actively treated

 

patients for progression to severe vision loss.

 

[Slide]

 

70

 

In order to be sure there were no

 

important subgroup relationships, we also performed

 

a multiple logistic regression to identify any

 

potential factors either influencing the outcome or

 

modifying the treatment effect. Subgroups and

 

interactions of subgroups with treatment were

 

investigated.

 

[Slide]

 

These are some of the subgroups that we

 

evaluated, age, angiographic subtype, use of PDT,

 

sex, race, lesion size, status of

 

smoker/non-smoker, subretinal hemorrhage, the

 

fellow eye vision loss and lipid.

 

[Slide]

 

We found for the 0.3 mg dose that no

 

factors were identified as significant treatment

 

effect modifiers for 0.3 versus sham, and no

 

factors except treatment with pegaptanib were

 

identified as significantly influencing the

 

response, and this had a p value of 0.0003 in favor

 

of treatment.

 

[Slide]

 

71

 

For the 1 mg group we again found that no

 

factors were identified as significant treatment

 

effect modifiers versus sham, and for pegaptanib at

 

1 mg there was a relationship between treatment

 

with pegaptanib, again at 0.0001, and age which

 

favored patients with less than 75 years of age.

 

This is not to say that older patients did not do

 

better. It just said that there was a favor for

 

younger patients even both appear to respond.

 

[Slide]

 

What can we conclude from these

 

exploratory or subgroup analyses? First, we have

 

shown that the treatment benefit appears

 

well-distributed among a broad patient population.

 

Second, the efficacy is not consistently

 

concentrated in or absent from any particular

 

patient subgroup. No one subgroup drove the

 

overall efficacy.

 

[Slide]

 

The 0.3 mg dose represents the lowest

 

studied efficacious dose and it met its primary

 

efficacy endpoint with statistical significance in

 

72

 

independent replicate trials, as we have shown you.

 

The efficacy was substantiated in every clinically

 

meaningful endpoint tested. We have seen the

 

secondary endpoints. And, the 1 mg and 3 mg doses

 

show no additional benefits over 0.3 mg. Tony will

 

shortly show you that there was no safety

 

difference between 0.3 mg and 1 mg as well.

 

However, theoretically we all know that a lowest

 

dose yields the lowest systemic concentration. So,

 

the sponsor advisory board and independent data

 

monitoring committee endorsed the 0.3 mg dose as a

 

dose that should be selected.

 

[Slide]

 

I would like to turn now to angiographic

 

findings. We have mentioned to you that we believe

 

there are two mechanisms of action for pegaptanib,

 

anti-angiogenesis and anti-permeability. As I will

 

now show you, we have anatomical confirmation for

 

both mechanisms of action that support the visual

 

findings we have shown you today.

 

Let's first look at the anti-angiogenesis.

 

Here is a patient in the trial with predominantly

 

73

 

classic neovascularization that showed virtually

 

complete regression. The white large area is the

 

neovascularization. You can see it has almost

 

completely regressed after 54 weeks of treatment.

 

But this is one case. So, let's look at the whole

 

group.

 

[Slide]

 

What we can see is that there was a

 

decrease in the lesion size that had a low nominal

 

p value in favor of active treatment for the 0.3

 

and the 1 mg dose. So, we have anatomical

 

confirmation or support for anti-angiogenesis as a

 

mechanism of action that supports the visual

 

findings.

 

[Slide]

 

The second mechanism of action that we

 

described was anti-permeability. Here is another

 

patient in the trial that had significant cystoid

 

macular edema with neovascular disease. We can see

 

the cystoid-like patterns here. This is a sign of

 

a lot of permeability. After 54 weeks of treatment

 

we can see a great decrease in the permeability.

 

74

 

[Slide]

 

Again, we can show that leak size over

 

time was less for treated groups than for shams.

 

The p values here are noted.

 

[Slide]

 

In addition, we can look at the change in

 

leakage to week 54 as a sign of anti-permeability

 

action, and we can see that very similar to visual

 

distribution curves I showed you earlier, we can

 

see again that there was less leakage noted more

 

often in actively treated 0.3 mg patients than in

 

sham, and more leakage noted in shams than in

 

actively treated eyes. This change in distribution

 

had a low nominal p value of 0.0004. So, again we

 

have anatomical confirmation for anti-permeability

 

as an important mechanism of action that supports

 

the visual findings.

 

[Slide]

 

I would like to now turn to photodynamic

 

therapy, or PDT. I think it is first important to

 

have a historical perspective of the use of PDT in

 

this trial so you can understand some of the

 

75

 

challenges we faced when we were designing this

 

trial.

 

At the time of starting the trial PDT was

 

available primarily in the U.S., and there were

 

certainly ethical considerations that required that

 

PDT be permitted in patients with predominantly

 

classic disease. However, the PDT usage pattern

 

was not yet known.

 

[Slide]

 

So, what we decided to do was to create

 

very strict rules for the use of PDT in this trial.

 

What that meant was that patients had to have

 

predominantly classic disease and the masked

 

physician--remember, we had two physicians--the

 

masked physician determined if the patient was

 

eligible for PDT per the FDA label and then whether

 

that PDT was recommended for that individual

 

patient. If so, the treatment was administered per

 

the FDA label.

 

Now, to ensure that these strict rules

 

were being followed, we had a reading center review

 

the usage pattern and we found that 92 percent of

 

76

 

the time the reading center agreed with the

 

appropriate use of PDT in this trial.

 

[Slide]

 

PDT use could occur three ways: prior to

 

the study, at baseline, and post-baseline and,

 

actually, any combination of the three. It is

 

important to emphasize that overall the use of PDT

 

was extremely low. Three-quarters of patients were

 

never exposed to PDT in the study eye at any time

 

in the time trial.

 

[Slide]

 

Let's examine each one of these three

 

scenarios in detail. First let's talk about prior

 

PDT which was stratified and was balanced at

 

randomization. Also, notice the small numbers

 

again, emphasizing very little PDT use in the

 

trial, 18-29 eyes in the various subgroups, but it

 

was stratified and balanced.

 

[Slide]

 

Baseline PDT is the second scenario, and

 

the baseline PDT use was again very similar among

 

the groups. We can see here that for the active

 

77

 

treated groups 10-13 percent of patients had PDT at

 

baseline compared to 14 percent for shams and,

 

again, look at the very small numbers, 31 to 40

 

patients per subgroup.

 

[Slide]

 

Finally, let's talk about post-baseline

 

PDT use. Now, it is important to mention that a

 

meaningful analysis of potential post-baseline PDT

 

effects on efficacy is limited to the inherent bias

 

in the trial. What I mean by that is, remember,

 

the patients were never randomized to post-baseline

 

PDT use. In order to really assess the baseline

 

PDT use we would have had to design a trial

 

randomizing patients to PDT and baseline. That

 

wasn't this trial. As an example of this, what is

 

called the channeling bias, a patient with a poor

 

response might be the patient that would be

 

preferentially channeled to get PDT. What this

 

really means is that post-baseline PDT is an

 

outcome variable. So, for this reason, we must

 

treat post-baseline PDT in a different way, as I

 

will show you now.

 

78

 

[Slide]

 

We need to ask was there increased PDT use

 

in pegaptanib patients relative to sham that could

 

suggest that some of the pegaptanib efficacy was

 

derived from PDT?

 

[Slide]

 

The answer to this question was no. As we

 

can see, there was no higher use of post-baseline

 

PDT in active treated patients compared to sham.

 

[Slide]

 

The second important question about

 

post-baseline PDT use is was there an increase in

 

the average number of PDT treatments in pegaptanib

 

patients relative to sham?

 

[Slide]

 

Again the answer is no. As we can see

 

again, there was no higher post-baseline PDT use in

 

active treatment eyes compared to sham.

 

[Slide]

 

The third important question, which will

 

be addressed in detail in Tony's safety section, is

 

was there evidence of any adverse events with the

 

79

 

co-administration of photodynamic therapy and

 

pegaptanib that could lead to a drug-to-drug

 

interaction? The answer is no--more on that in

 

just a few minutes.

 

[Slide]

 

In summary, pegaptanib met a clinically

 

meaningful primary efficacy endpoint with

 

statistical significance in replicate, independent,

 

well-controlled clinical trials.

 

[Slide]

 

I will now ask Tony to come up and discuss

 

our clinical safety database.

 

Pegaptanib Clinical Safety

 

[Slide]

 

DR. ADAMIS: This is the entire safety

 

database. This includes the patients from the

 

earlier Phase I/II trials. What you see here is

 

that the total clinical experience to date includes

 

over 1,200 patients in over 10,000 treatments, of

 

which 7,500 are intravitreous injections that we

 

can monitor for the safety. There is a slight

 

imbalance that you will see in that there are more

 

80

 

patients receiving 3 mg than 1 mg of 0.3 mg. That

 

is because that was the dose that was used

 

throughout most of the Phase I/II program. In

 

addition, we gave doses of 0.25 mg and 2 mg in

 

those earlier programs as well.

 

[Slide]

 

The overall safety is shown here. As

 

regards any adverse events, you can see it is

 

balanced between all treatment arms and sham.

 

There is an imbalance in the serious adverse

 

events. These are largely injection related, and

 

we will talk about those in depth in a moment.

 

The discontinuations, you will note, due

 

to adverse events are low. They are one percent in

 

both the treated and the sham arms. Similarly, the

 

death rate is balanced to two percent.

 

[Slide]

 

Looking at the death rate just a little

 

more closely, we can see that there is no evidence

 

here of a dose response.

 

[Slide]

 

Let's look at the most frequent non-ocular

 

81

 

serious adverse events. This is a busy slide but

 

the thing to note here is, first, that there is

 

good balance between the treated and the sham arms

 

and, secondly, there is no clustering within a

 

system organ class. This is rather diffuse. These

 

conditions are age appropriate. The mean age of

 

this population is 77 years old that we studied.

 

These people had a number of concomitant illnesses.

 

Fifty percent of them had hypertension; 25 percent

 

were on statins; 20 percent had cardiac disease.

 

So, we believe it is representative of the

 

population.

 

[Slide]

 

We looked particularly for VEGF

 

inhibition-related adverse events as these have

 

been reported with other non-selective inhibitors

 

given intravenously at higher doses. We were happy

 

to see that there were no signals here. The most

 

sensitive signal, the one that has been picked up

 

with other non-selective inhibitors in smaller

 

trials than ours, less powered but nevertheless it

 

was evident, was hypertension. You can see here

 

82

 

that the rate of adverse events is 10 percent both

 

in the treated and in the sham arms--no signal

 

there for that very sensitive signal of VEGF

 

inhibition. Thromboembolic adverse events are

 

similarly balanced, as are ischemic coronary artery

 

disorders, heart failure and serious hemorrhagic

 

adverse events.

 

[Slide]

 

Why is it that we did not see any of these

 

VEGF inhibition-related phenomena? There is a

 

number of reasons. Some of these are theoretical,

 

some are real but in aggregate they provide I think

 

an argument. Pegaptanib is, as I said, selective

 

for VEGF 165 so the other major isoform 121 is

 

never blocked. So, all VEGF is never blocked with

 

pegaptanib, even if you gave it at very large

 

concentrations. It just does not bind to VEGF 121.

 

Secondly, the concentrations that we see

 

when we put 0.3 mg in the eye are many orders of

 

magnitude less in the plasma and those

 

concentrations are below the inhibitory

 

concentration that our models have told us both for

 

83

 

in vitro and in vivo inhibition of VEGF. So, we

 

believe that these are levels that are below the

 

ability of pegaptanib to affect VEGF levels in any

 

sort of substantive way.

 

Third, as I just said, there was an

 

absence of sensitive VEGF inhibition signals, the

 

most sensitive being hypertension which I showed

 

you but also in our 1006 trial, where we looked

 

carefully at proteinuria, again there is no

 

evidence that this drug is inducing proteinuria in

 

either our clinical population or in our

 

preclinical models.

 

Then, the report recently of

 

thromboembolic adverse events occurring in cancer

 

patients on chemotherapy and receiving Avastin--we

 

think there are a couple of very different things

 

about our population and that population that was

 

studied. Number one, cancer in and of itself

 

predisposes patients to thromboembolic phenomena.

 

They have indwelling catheters; they are bedridden;

 

and the cancer itself alters the clotting system.

 

Secondly, some chemotherapy has been shown to be

 

84

 

vascular toxic, to be prothrombotic. There is a

 

published literature on that.

 

So, add these two hits to the vasculature

 

and then block all VEGF to prevent the endothelium

 

from healing itself, one can have a theoretical

 

basis for understanding now why thromboembolic

 

phenomena may be more prevalent in a population

 

with cancer and chemotherapy. That is not age

 

related macular degeneration. This is a very

 

different population that is not, by and large, on

 

chemotherapy and do not have cancers.

 

[Slide]

 

Let's look at the ocular adverse events.

 

Again, this is a busy slide but we will talk about

 

these events in a little more detail. They are

 

listed here, those that occurred greater than or

 

equal to 10 percent of patients on either

 

pegaptanib or sham. You can see that there is a

 

slight imbalance in eye disorders, and we will talk

 

about these, and you see a number of various

 

adverse events listed here.

 

[Slide]

 

85

 

Let's talk about them in more detail,

 

number one that was listed on the previous slide

 

being eye pain. These patients receive nine

 

intravitreous injections over the course of a year.

 

It is rather remarkable actually that two-thirds of

 

them never reported a single instance of pain. Of

 

those patients, approximately the one third that

 

did report pain, it was mild or moderate in

 

character in 99 percent of them, and only one

 

patient exited this trial describing an adverse

 

event of pain.

 

The other important thing to note here is

 

that the eye pain in the sham arm, at 28 percent,

 

was significantly higher than what is seen in the

 

fellow eye, 2 percent. So, some of this mild pain

 

that these patients experienced--one conclusion you

 

can draw is that it may be due to the preparative

 

procedure prior to the injection of the drug. As

 

you recall, these patients have a speculum placed

 

in the eye. They have povidone-iodine scrub. They

 

have a subconjunctival anesthetic injection. These

 

things may have contributed to the lion share of

 

86

 

the reports of pain which, again, was mild. Then,

 

obviously, there is a difference here. The

 

remainder of it here can well be ascribed to the

 

actual intravitreous injection itself.

 

Of those patients who reported pain, it

 

was in a minority of their injections, two in both

 

the treated and the sham arms, and the median time

 

to resolution was two to three days which is the

 

time of the follow-up phone call.

 

[Slide]

 

With regard to vitreous floaters, there

 

was more than an imbalance here. It was 33 percent

 

in the treated arms versus 8 percent in the sham.

 

Again, there is a slight difference, 8 versus 1,

 

between the sham eye and the fellow eye so some of

 

this may be due to the preparative procedure but a

 

large portion of it, the majority of it, is very

 

likely due to the act of giving an intravitreous

 

injection itself. When giving a 90 mcL volume

 

injection into the eye, in the average human a

 

volume of 4 mL, you are displacing the vitreous and

 

it is perhaps not surprising that as a function of

 

87

 

that you are going to induce floater. These

 

floaters never were severe. All of them were

 

characterized as mild to moderate. No patient left

 

the trial because of floaters. It was in a

 

minority of injections, 1 to 2 injections, that

 

these were reported, if they ever were reported,

 

and the median time to resolution was 3 days in the

 

treated arms versus 7 days in the sham arms.

 

[Slide]

 

We looked at cataract very carefully. We

 

specifically looked at cataract in only the aphakic

 

eyes. One-third of these patients approximately

 

were pseudophakic. What we saw was that across all

 

treatment arms there was a slight imbalance, with

 

30 percent of the eyes having an adverse event of

 

cataract versus 26 in the sham arm. This slight

 

imbalance may be partially explained by the fact

 

that the phakic fellow eye also had a slight

 

imbalance, 17 percent in the treated versus 15

 

percent in the sham arms.

 

But we looked at this a little more in

 

depth. The type of cataract that one would expect

 

88

 

if this was due to a drug toxicity, the type that

 

has been amply described in the literature, is

 

posterior subcapsular cataract. So, when we looked

 

for that specific type of cataract grading, you can

 

see there is zero difference. It is 11 percent in

 

both the treated and the sham arms.

 

[Slide]

 

Nuclear cataract was similarly well

 

balanced. In fact, if you remove the eyes that

 

were vitrectomized, which we will talk about in a

 

minute, vitrectomy can cause a nuclear sclerotic

 

cataract to accelerate. This is 18 percent in both

 

arms and there is, indeed, a slight imbalance in

 

cortical of 18 versus 15 percent.

 

One piece of objective data we have is

 

that the vast majority of these patients came in at

 

baseline with cataract and only 3 patients

 

underwent elective cataract surgery over the 54

 

weeks of the trial in the treatment arms.

 

[Slide]

 

Anterior chamber inflammation was another

 

adverse event. You can see here that there is an

 

89

 

imbalance slightly with 14 percent occurring in

 

study eyes versus 6 percent in the sham eyes, and

 

there were zero reports in the fellow eyes. None

 

of these cases of anterior chamber inflammation

 

were characterized as severe. All of them were

 

mild to moderate and we believe they were largely

 

due to the active intravitreous injection and not

 

to the drug itself. The reports of inflammation

 

were all moderate and self-limiting and did not

 

increase during the course of the trial. In fact,

 

there was a slight trend to decrease, arguing that

 

there wasn't a sensitization to the molecule here,

 

in fact, supporting that this was due to the

 

injection itself. The median time to resolution

 

was 8-9 days, and no patient left the trial because

 

of inflammation.

 

[Slide]

 

We looked at interaction potentially with

 

PDT and specifically at ocular adverse events. You

 

can see here that the majority of patients did not

 

have the combination of PDT and pegaptanib, but of

 

those who did we looked very carefully at the event

 

90

 

rates and the important thing to consider here is

 

the event rate difference in the sham arms

 

plus/minus PDT, and does that difference change in

 

any sort of meaningful fashion when the PDT is

 

given together with pegaptanib. The answer is that

 

from these data there doesn't appear to be a

 

difference in those two measures. The same is true

 

with vitreous floaters. There is a slight

 

difference here and there is really no difference

 

here in the treatment arms.

 

[Slide]

 

But let's look at it another way. This

 

assessment is looking to see if there was a report

 

of an adverse event at any time during the 54

 

weeks. For instance, if the patient had PDT at

 

baseline but had an adverse event at 54 weeks it

 

would be captured and presented in these data. We

 

thought we would try to look at this a little more

 

carefully and see if there was a better temporal

 

relationship. So, now we are looking at data of

 

patients who had PDT plus/minus 2 weeks around an

 

injection of pegaptanib. These events may more

 

91

 

likely signify some sort of interaction and, again,

 

there are no alarming signals here.

 

When one looks at eye pain there is very

 

little difference here and there is very little

 

difference here between the sham and the treatment

 

arms. The same is true for corneal epithelium

 

disorders. For these two specific adverse events

 

one can postulate a mechanism as to why that is.

 

There is, you know, the povidone-iodine prep for

 

the injection which can affect the epithelium and

 

perhaps cause pain. On top of that is a near

 

temporal relationship the placement of a contact

 

lens for doing the PDT, and one could understand

 

why there might be a slight increase here. Again,

 

no patients dropped out because of any adverse

 

events related to a combination of PDT and the use

 

of pegaptanib.

 

[Slide]

 

Now let's concentrate a bit on ocular

 

serious adverse events. The three most common we

 

are going to discuss in detail here are

 

endophthalmitis, retinal detachment and traumatic

 

92

 

cataract. The ones below occurred at a very low

 

event rate. When the narratives in the cases were

 

looked at in depth there really did not appear to

 

be an association with the use of pegaptanib so we

 

will not discuss them further here unless you wish

 

to discuss it later in the question and answer

 

session.

 

Endophthalmitis occurred in 12 patients

 

over 54 weeks. That translates to a relative risk

 

of 1.3 percent of patients developing

 

endophthalmitis over the course of one year of

 

therapy. So that we could compare our rate to the

 

published literature this was converted to a per

 

injection rate of 0.16 percent. What we learned is

 

that the rate that we saw is not an outlier; it is

 

within the published norm and reported norm in

 

cases of endophthalmitis in patients receiving

 

intravitreous injected therapeutics.

 

As important as the rate is what happened

 

to these patients, what was the outcome. One

 

patient lost severe vision in this trial as a

 

function of their endophthalmitis, 1/12, which

 

93

 

translated to a rate of 0.1 percent over the course

 

of the year. Seventy-five percent of the patients

 

who developed endophthalmitis elected to stay in

 

the trial.

 

Traumatic cataract--you can see there were

 

five cases of it and there were five cases of

 

retinal detachment, of which three were

 

rhegmatogenous in nature.

 

[Slide]

 

I show you here the specific details of

 

all 12 cases of endophthalmitis. What you can see

 

here are the starting visions, the visions prior to

 

the event, and the change in vision from just prior

 

to the event which probably most accurately

 

captures the visual loss related to the

 

endophthalmitis itself. What you can see is the

 

one patient who lost 11 lines as severe vision

 

loss.

 

Let me just tell you anecdotally what

 

happened. It was a protocol violation. It turns

 

out this patient had an active lachrymal sac

 

infection prior to the development of the

 

94

 

endophthalmitis and the injection of the mediation,

 

and had an active lachrymal sac infection after the

 

event of endophthalmitis. The patient should never

 

have been enrolled because that was an exclusion

 

criterion.

 

The other patients, as you can see, were

 

treated aggressively and their visual outcomes tend

 

to be perhaps a bit better than what you would

 

expect for a case of endophthalmitis. In fact,

 

there are some patients here who gained one or two

 

lines of vision.

 

[Slide]

 

How were these patients diagnosed, and

 

were we able to identify the endophthalmitis

 

relatively early? This slide shows you exactly

 

what happened. Three patients were identified in

 

their follow-up phone call at days three-four post

 

injection. Eleven patients presented to their

 

physician's office with complaints, and this

 

happened between days two and five. Two patients

 

came in and were diagnosed in a routine follow-up.

 

The endophthalmitis cases I am describing here are

 

95

 

the 12 in the first year and the ones that have

 

occurred subsequent to that which I am going to

 

talk about.

 

[Slide]

 

We have been following the endophthalmitis

 

issue very carefully and I would like to provide

 

you with an update on where we are beyond the

 

54-week time period. As I just said, in the first

 

year 0.16 percent of injections, or 1.3 percent of

 

patients, developed endophthalmitis. In the

 

second, and now some patients have entered the

 

third year of this trial, there have been five

 

additional cases as of July 31st of this year, and

 

there has been one case in our Phase II diabetic

 

macular edema trial. So, if you look at the total

 

now, it is 18 cases of endophthalmitis with a

 

denominator of over 14,500 injections, and the rate

 

now is reduced somewhat to 0.12 percent per

 

injection.

 

In the first half of this trial when we

 

saw the case reports of endophthalmitis we convened

 

an expert panel of ophthalmologists and retinal

 

96

 

specialists who work in the endophthalmitis area

 

and we decided that we needed to heighten the

 

awareness of the need for strict adherence to an

 

aseptic protocol when one is giving an

 

intravitreous injection. In fact, there was a

 

letter sent to IRBs and a formal protocol

 

modification mandating the use of a sterile drape,

 

of a speculum, of the use of povidone-iodine. When

 

we did these things and we analyzed what the

 

potential effect could be, what we saw was that

 

prior to that protocol modification being adopted

 

at all sites between August of 2001 and May of 2003

 

the rate was 0.18 percent, and after that protocol

 

modification the rate has now fallen to 0.03

 

percent.

 

Can we ascribe the decrease in the rate to

 

the change in the protocol? Not necessarily.

 

There was more than one variable that was changing

 

here. At the same time that we instituted this

 

protocol modification and heightened awareness

 

about the aseptic technique there was a dramatic

 

uptake in the number of intravitreous injections

 

97

 

being given for off-label use in diabetic macular

 

edema with steroids, triamcinolone in particular.

 

So, the knowledge base and the experience of retina

 

physicians increased rather dramatically at the

 

same time that we saw a drop in our rates.

 

[Slide]

 

The visual outcome for the cataract cases

 

is shown to you here. For the one patient who lost

 

7 lines of vision, it was ascribed to progression

 

of macular degeneration. All of these patients, in

 

fact, had successful cataract surgery.

 

[Slide]

 

The visual outcome of the retinal

 

detachment cases is shown here. All of these were

 

successfully repaired and you can see the cases of

 

rhegmatogenous detachment which most likely were

 

injection related. The visual outcomes were quite

 

good.

 

[Slide]

 

Intraocular pressure was examined. As I

 

said earlier, it is not surprising if one injects

 

90 mcL into a 4 mL closed space that you will see a

 

98

 

transient rise in pressure. In fact, in

 

ophthalmology it is common with almost all

 

procedures that pressure spikes tend to occur.

 

Well, they occurred here and the transient rise in

 

mean intraocular pressure at the first prespecified

 

measurement, 30 minutes, was 2-4 mm across the

 

treatment arms.

 

It is important to note that the mean

 

intraocular pressure returned to pre-injection

 

levels one week following the injection, which was

 

the next visit, and that 90 percent of patients,

 

approximately 90 percent of patients, never had a

 

spike above the prespecified threshold of 35 mm and

 

any patient who did have a spike was not allowed to

 

leave the physician's office till the pressure was

 

below 30 mm.

 

Very importantly, there was no evidence of

 

a persistent increase in intraocular pressure over

 

one year. The drug did not seem to alter the

 

outflow of the eye in any way. In those patients

 

who did have a spike, the question was if you had a

 

spike was it because somehow the drug was altering

 

99

 

the outflow mechanisms, and if that was the case

 

you would expect to see an increased incidence

 

during the course of the trial as it progressed.

 

As the data show you here, that is not the case.

 

It doesn't appear to increase over time and, in

 

fact, may have been dropping slightly.

 

[Slide]

 

This slide simply shows the mean

 

intraocular pressure values over time for all three

 

treatment arms and sham, again giving us some

 

confidence that the drug is not inducing a rise in

 

chronic IOP.

 

[Slide]

 

We have a safety update for you regarding

 

angiography. Colored photographs and angiograms

 

were looked at in the independent reading center at

 

the University of Wisconsin. We have looked at up

 

to 97 percent now of our month 18 angiograms and 92

 

percent of our two-year angiograms to get a sense

 

of is there any evidence of cumulative toxicity.

 

The results are that there is no evidence

 

whatsoever of alterations in the normal retinal or

 

100

 

choroidal vasculature as a function of the drug

 

being in the eye now for up to two years, nothing

 

that deviated from the natural history of

 

age-related macular degeneration and no alterations

 

in the normal vessels.

 

[Slide]

 

The safety update, which was just

 

concluded in the past week by the independent data

 

monitoring committee, has reviewed 100 percent of

 

the patients through month 18 of this trial and 97

 

percent through month 24, and there have been no

 

deviations from sort of the pattern of adverse

 

events, the ones that we saw in the first year of

 

the trial. There have been no new safety concerns

 

except perhaps for a slight increase in the number

 

of retinal detachments. There were 6 that were

 

reported in the second year of this trial.

 

[Slide]

 

To summarize the non-ocular safety, there

 

was a very low discontinuation rate due to adverse

 

events. It was one percent and it was balanced in

 

the treated and the sham arms. Non-ocular serious

 

101

 

adverse events appeared to be similar in rate and

 

character between pegaptanib and sham, and the

 

mortality rate, as you saw, for the 77 year-old

 

population was similar between pegaptanib and sham.

 

[Slide]

 

As regards ocular safety, I think what we

 

can conclude is that the majority of the ocular

 

adverse events were judged to be procedure related.

 

They were transient and mild in character and

 

largely self-limiting. There was a low

 

discontinuation rate due to ocular adverse events

 

and the serious adverse events were infrequent.

 

They were rarely associated with severe vision loss

 

and were mostly procedure related. Finally, there

 

were mild transient and predictable, manageable

 

increases in intraocular pressure but no evidence

 

of a long-term rise in intraocular pressure.

 

[Slide]

 

At this point Prof. Don D'Amico, who is a

 

practicing retinal specialist at the Massachusetts

 

Eye and Ear Infirmary, will come and discuss the

 

risk/benefit profile for pegaptanib.

 

102

 

Pegaptanib Benefit/Risk Profile

 

DR. D'AMICO: Thank you, Dr. Adamis. Dr.

 

Dunbar, members of the advisory committee, ladies

 

and gentlemen, with your permission I would like to

 

introduce myself a little more fully and my

 

perspectives so that you can have the clearest

 

context in which to place my remarks.

 

[Slide]

 

With regard to this study, while it was in

 

progress I was invited to be a member of the safety

 

committee and later became its chair. At the

 

conclusion of the study I was asked to be a member

 

of the scientific advisory board. I perform a

 

virtually identical role for the Alcon Corporation,

 

chairing their safety committee in the evaluation

 

of their anecortave product. I also advise them on

 

surgical themes and instrumentation as well.

 

Finally, I am a consultant to the Iridex

 

Corporation serving as a member on the safety

 

committee for the transpupillary thermotherapy

 

trials and their PTAMD or laser for drusen trial.

 

I hold no equity in any of these companies nor any

 

103

 

of their competitors.

 

[Slide]

 

I would like to also share four

 

perspectives that will inevitably influence my

 

remarks and may be helpful to you also in your

 

evaluations. First, of course, I was a member of

 

the pegaptanib safety committee. Secondly, I have

 

had a career-long laboratory, as well as clinical,

 

interest in endophthalmitis and the effects of

 

administration of intravitreal medications. I am,

 

as introduced, an academic in the field of retinal

 

diseases and therapy. But perhaps most importantly

 

and most germane is that I have a very active

 

retinal practice at the Massachusetts Eye and Ear

 

Infirmary and care for many patients with macular

 

degeneration.

 

[Slide]

 

As has been said, neovascular AMD is quite

 

a source of human suffering. At the 20/40 level of

 

visual acuity driving privileges frequently become

 

impossible for a patient. At 20/80 or worse

 

difficulty is even present in trying to read large

 

104

 

print. And, 20/200 or worse is a commonly accepted

 

level of definition of legal blindness at which it

 

is difficult to recognize faces and independent

 

function is threatened.

 

[Slide]

 

How extant is this problem in the world

 

today? In a very careful meta-analysis of the most

 

comprehensive studies recently reported by the Eye

 

Diseases Prevalence Research Group, they looked at

 

studies in the United States, Western Europe and

 

Australia over an 11-year period.

 

[Slide]

 

Based on their analysis, it is the leading

 

causes of blindness in U.S. adults in patients aged

 

40 years or older. You see that slightly over half

 

are due to age-related macular degeneration.

 

[Slide]

 

They then applied their model to the U.S.

 

Census data for both 2000 and projected to the

 

future. In a morning filled with numbers, I will

 

spare you all the numbers here, but using a

 

definition of 20/200 or worse as blind and 20/40 or

 

105

 

worse as visually impaired, there are 3.3 million

 

Americans with visual impairment today. In the

 

future there will be approximately 5.5 million

 

American with visual impairment at some level,

 

again slightly over half, due to age-related

 

macular degeneration. So, it is clearly a problem.

 

[Slide]

 

As such, it merits our highest attention

 

as physicians, researchers, etc. to try to find

 

treatments and even cures. This slide is color

 

coded and it lists the candidate therapies for

 

neovascular subfoveal age-related macular

 

degeneration. Therapies which have demonstrated

 

effectiveness in replicate clinical trials are

 

shown in yellow. We have laser photocoagulation,

 

photodynamic therapy with Visudyne and the data you

 

have just heard on pegaptanib. The great majority

 

of interventions are listed in white, which

 

indicates ongoing study with various degrees of

 

promise, and it includes surgical options, as you

 

see here and a variety of other laser treatments,

 

as well as other pharmaceuticals, many of which are

 

106

 

nearing the end of their clinical trials. There

 

are also some abandoned therapies that were

 

ineffective and combination strategies, as you see

 

in the lower right, are becoming of increasing

 

interest.

 

[Slide]

 

Looking at the established therapies,

 

there are two. One is photocoagulation with

 

thermal laser which has been effective in

 

extrafoveal, juxtafoveal and subfoveal lesions.

 

However, in subfoveal lesions this therapy has been

 

abandoned due to the immediate destruction of

 

central vision following treatment and is no longer

 

in clinical use. Photodynamic therapy with Visudyne

 

is approved for subfoveal predominantly classic

 

lesions.

 

[Slide]

 

In addition, evolving clinical practice,

 

in a hope to provide improved care for patients

 

with macular degeneration, has led to a new

 

accommodation therapy which has become widespread.

 

That is the combination of a PDT treatment with

 

107

 

Visudyne in association with an intravitreous

 

induction of triamcinolone in the peri-PDT period.

 

This treatment has had some very promising early

 

pilot results but the literature is quite minimal

 

at present. Nevertheless, it has become a common

 

treatment in clinical practice.

 

[Slide]

 

Intravitreous injections are quite common

 

in my world as a retinal specialist. They were

 

employed and were actually the pathway to great

 

success in the therapy of endophthalmitis, and are

 

still continued widely in use for that indication.

 

We also utilize intravitreal injection as a

 

treatment of retinal detachment, as well as

 

administering agents for CMV retinitis. However,

 

there has been great expanded use recently in

 

office practice of intravitreal injections as

 

regards the use of triamcinolone acetodine for

 

diabetic macular edema, retinal vein occlusions,

 

uveitis, as I have just mentioned, in association

 

with photodynamic therapy.

 

[Slide]

 

108

 

Pegaptanib represents the potential for a

 

new approach, a pharmacotherapy, and what are the

 

advantages of pharmacotherapy? They are both

 

general and specific. In general, pharmacotherapy

 

offers the prospect of treatment at a molecular

 

level with improved targeting of the disease

 

process and, more importantly, limitation of the

 

collateral damage that invariably occurs with

 

larger scale interventions such as surgery or

 

laser.

 

Pegaptanib quite specifically is based on

 

very extensive basic science into the most widely

 

accepted, central disease processes in AMD, namely

 

neovascularization and leakage, with consistency

 

across multiple experimental models and studies.

 

[Slide]

 

As a member of the safety committee, we

 

looked for three specific areas in great detail.

 

One, were there any potential systemic side effects

 

from receiving an anti-VEGF medication? Secondly,

 

were there intraocular drug-related side effects

 

from this VEGF medication? Thirdly, were there any

 

109

 

mechanical side effects or complications from the

 

intravitreal injection procedure itself?

 

[Slide]

 

We did find serious ocular adverse events

 

related to the injection procedure. As you have

 

heard, there were 12 cases of endophthalmitis.

 

This incidence rate is quite comparable to that in

 

published series for intravitreal injection with

 

the other forms of intravitreal injection therapy

 

that I have mentioned previously. One of these

 

patients had severe visual loss. Nine of the

 

patients continued in the study and elected to

 

continue receiving study medication. Finally,

 

after protocol modifications, the incidence is

 

clearly trending downward.

 

There were five cases of retinal

 

detachment, which were repaired and some were

 

related to the underlying macular degeneration

 

itself. Traumatic cataract was seen in five cases

 

and all were surgically repaired without sequelae.

 

[Slide]

 

So, in these 22 serious ocular events, we

 

110

 

considered them in the context of 7,545

 

intravitreous injections performed in 1,190

 

patients by 117 centers worldwide, and many of

 

those centers had more than one injector. We felt

 

that this denominator indicated substantial safety

 

for this procedure.

 

We also found no evidence of systemic side

 

effects, no evidence of ocular drug-related side

 

effects, and the majority of other adverse events

 

were mild and transient within the eye. The

 

serious ocular adverse events were infrequent and

 

manageable. So, we concluded that there was a very

 

favorable safety profile that, in addition, may be

 

further improved by education and additional

 

training.

 

[Slide]

 

If we look at severe vision loss, again to

 

understand the context of these adverse events, if

 

a patient presented to the trial and received sham,

 

that is, usual care, there was a 22 percent risk

 

per year of suffering severe visual loss. When

 

they were enrolled in the pegaptanib group that

 

111

 

risk was reduced to 9.5 percent per year.

 

[Slide]

 

In the endophthalmitis, retinal detachment

 

and cataract serious ocular events that we saw, the

 

risk of severe vision loss, that is 6 or more lines

 

of vision, was 0.1 percent, indicating substantial

 

order of magnitude less risk from endophthalmitis

 

than from the real problem here which is the

 

macular degeneration itself.

 

[Slide]

 

Regarding efficacy, you have heard a

 

detailed presentation and I will just summarize.

 

There was significant reduction in moderate and

 

severe vision loss compared with sham. There was

 

promotion of vision stability and gain in a

 

proportion of patients. There was efficacy with

 

broad-based entry criteria including a range of

 

subfoveal neovascular AMD lesions. And, the

 

benefit of intravitreous pegaptanib therapy was

 

early and sustained.

 

[Slide]

 

As we have seen, in this slide baseline

 

112

 

visual acuity is on the left. Sham is indicated in

 

purple and pegaptanib in grey. At 54 weeks there

 

is a definite shift in the 0.3 pegaptanib group to

 

preservation of better vision on the left of this

 

chart compared to the visual acuities observed with

 

sham.

 

[Slide]

 

I am not a biostatistician but I will try,

 

for myself and for all of us, to place these

 

results in some kind of a wider context. What

 

could this mean? No one knows exactly how many new

 

subfoveal neovascular lesions occur a year, but

 

120,000 per year of new treatment-eligible patients

 

is probably a reasonable estimate. If those

 

patients were to behave similar to the gathered

 

group enrolled in this trial, we could make some

 

statements, and here they are:

 

Pegaptanib potentially prevents severe

 

vision loss, that is a loss of 6 or more lines of

 

vision, in 15,000 additional patients per year in

 

the United States compared with usual care, based

 

on a 57 percent reduction in the rate of severe

 

113

 

vision loss with pegaptanib.

 

[Slide]

 

Secondly, reaching a level of 20/200 or

 

worse within the treated eye, we could call that

 

blindness in the treated eye. Pegaptanib

 

potentially prevents treated-eye blindness in an

 

additional 22,800 patients per year in the U.S.,

 

again compared with usual care, based on a 38

 

percent reduction in the rate of treated-eye

 

blindness with pegaptanib.

 

[Slide]

 

In conclusion, from the perspectives

 

available to me and now available to you, I have

 

concluded that pegaptanib will have a significant

 

impact on AMD in regard to both individual patients

 

with AMD lesions that would become amenable to

 

treatment and, secondly, in its effects on visual

 

function and its preservation in the aging U.S.

 

population.

 

The positive results in this trial

 

indicate the beginning, and not the limit, of

 

pharmacotherapy for AMD. I agree with the

 

114

 

sponsor's recommendations that the benefits of

 

pegaptanib therapy for AMD strongly outweigh the

 

risks. Thank you.

 

Committee Discussion

 

DR. DUNBAR: Thank you to the sponsor and

 

Drs. Guyer, Adamis and D'Amico. At this point I

 

would like to open the floor for discussion and

 

questions for the sponsor from the committee

 

members, and ask that you will speak your name into

 

the microphone as you ask each question. Are there

 

any questions from the committee members? Dr.

 

Chinchilli?

 

DR. CHINCHILLI: Yes, I don't know much

 

about the disease and the patients that were

 

recruited for the two trials so, please, bear with

 

me. Could patients have AMD in both eyes? I mean,

 

roughly what proportion of patients that were in

 

the trials had that situation?

 

DR. GUYER: In general, for neovascular

 

age-related macular degeneration usually one eye

 

becomes active at a time. If the patient lives

 

long enough, they often will get it in the second

 

115

 

eye. In this particular trial the investigator

 

would choose--in a very few number of cases where

 

there would be active disease that was eligible for

 

the trial, the doctor would make that decision. If

 

we look at slide D-82--

 

[Slide]

 

--here we can see some of the baseline

 

characteristics. In two-thirds of the patients

 

this was the worse eye that was treated. Again, no

 

patient was treated in both eyes at the same time.

 

But in the lifetime of a patient there could be

 

some overlapping times where they have an active

 

lesion and the second one becomes active. Some

 

patients are fortunate enough not to get it in

 

their second eye but, unfortunately, if they live

 

long enough many will.

 

DR. CHINCHILLI: Thank you.

 

DR. DUNBAR: Dr. Pulido?

 

DR. PULIDO: A superb presentation; very

 

interesting results. Just two questions. Number

 

one, glaucoma was not an exclusion criterion in the

 

study. So, some of the patients had glaucoma and

 

116

 

AMD. Do you have any data as to the effect of

 

chronic injections on the small subgroup of

 

patients that had glaucoma?

 

DR. GUYER: I will let Tony answer that.

 

DR. ADAMIS: We were interested in that

 

question as well. Slide OS-31.

 

[Slide]

 

We looked specifically at patients with a

 

history of ocular hypertension and/or glaucoma, and

 

then followed their pressures throughout the entire

 

54 weeks. What we saw was that there was no change

 

in their intraocular pressure as a function of

 

treatment.

 

DR. PULIDO: The other question probably

 

is to you as well. There are some recent

 

articles--here is one from Nature, May: VEGF

 

delivery with retrograde transported Lentivector

 

prolongs survival in a mouse ALS model. Here is

 

another one: mural protection of ischemic brain by

 

VEGF is critically dependent on proper dosage.

 

Here is another one. So, we have gone under the

 

assumption that VEGF and VEGF 165 is specifically a

 

117

 

cytokine for angiogenesis, but there is more data

 

to show that there is an independent effect

 

directly on neural tissue, separate from its

 

angiogenic effect. ERG was not a part of this

 

trial. You did some ERGs on some dogs, from what I

 

saw here. I don't know how many, how long, etc.

 

So, considering the neuroprotective

 

effect, from your data--it is wonderful--that the

 

angiogenesis is important, critical to take care of

 

this significant problem in patients. But my

 

concern is the long-term chronic dosaging

 

considering that there is an independent effect of

 

VEGF as a neuroprotective agent.

 

DR. ADAMIS: As always happens in science,

 

what seems very straightforward becomes more

 

complex, and what you quote is absolutely correct.

 

I think that is Peter Carmeliet's paper in Nature.

 

But what has been learned in about the last five

 

years is that neural cells have VEGF receptors and

 

VEGF may be neuroprotective for certain tissues.

 

Certainly, in the ALS model that is the most

 

convincing story to date. Whether the effect is

 

118

 

direct or not is still being debated in the

 

scientific world, but it may well be direct because

 

of the VEGF receptor on the neural cells.

 

We were interested in this as well. Even

 

before we got into the scientific question as part

 

of our preclinical safety testing, there was a

 

9-month dog study where the dogs received 3 mg

 

injections every 2 weeks bilaterally. Then they

 

had ERGs done and there were no abnormalities seen

 

there. So, that gives us a little bit of comfort

 

but, more importantly, recently we examined this

 

issue and looked specifically at the isoform story.

 

We presented a paper at ARVO last spring where we

 

showed that in a model of retinal ischemia if one

 

gives a pan-isoform, non-selective VEGF inhibitor,

 

you can in fact induce some neural apoptosis. But

 

when we gave pegaptanib in the exact same setting

 

there was no induced apoptosis. So, again getting

 

at this thesis, the important thing with pegaptanib

 

I think is that you are sparing some VEGF to allow

 

it to have its physiological or perhaps these

 

rescue functions that can occur in the eye. So,

 

119

 

that gave us an additional measure of comfort that

 

we are not going to have neural toxicity.

 

DR. PULIDO: But the question still arises

 

have you done long-term ERG studies on these

 

patients?

 

DR. ADAMIS: Oh, I am sorry, no, we have

 

not done those in these patients.

 

DR. DUNBAR: Mr. Kresel?

 

MR. KRESEL: My disclaimer is that I am

 

not a statistician and so I am not sure if this is

 

even an appropriate way to ask this but I am going

 

to ask it anyway. You did a great job of looking

 

at endophthalmitis which, you know, obviously is

 

one of the things that people have concern about,

 

and referred to a decrease in patients that was

 

only five cases in years two and three. My

 

question is how many patients continued therapy

 

that far? So, did the number of patients decrease

 

and, therefore, the percent not go down? Because

 

what we saw is a cumulative number that, of course,

 

did go down.

 

DR. ADAMIS: It is a fair question. The

 

120

 

number of patients was decreased in the second

 

year. That is why the metric we used was on a per

 

injection basis. That accounts for any loss of

 

patients and those were the rates that I presented

 

to you today. So, on that basis it does go down.

 

Slide 129.

 

[Slide]

 

Just to show you the data, you can see

 

that prior to the amendment on a per injection

 

basis it was 0.18 percent, and then post the

 

amendment it was 0.03 percent but with that

 

additional confounding variable of a lot of

 

off-label steroid injections going on.

 

DR. DUNBAR: Dr. Gates?

 

DR. GATES: In the context of the cases of

 

endophthalmitis, could you expand on the initial

 

injection technique versus some of the changes that

 

you made secondarily? Because draping oftentimes

 

means different things to different people.

 

DR. ADAMIS: Correct. The details of the

 

injection procedure are on a slide but let me see

 

if I can recite them from memory for you, the

 

121

 

changes. There was a requirement for the

 

installation of an antibiotic drop or dilated

 

povidone-iodine prior to the amendment. Then,

 

after the amendment the drape that was specified is

 

a clear plastic one that adheres around the lids

 

and the lashes, and then the placement of the

 

speculum, and then also asking for a

 

povidone-iodine flush to be done, and then patients

 

received postoperative antibiotics. So, what we

 

tried to instill there was a sense of uniformity in

 

the procedure. There was more latitude prior to

 

that. Those were the changes, to the best of my

 

memory, that were instituted.

 

DR. DUNBAR: We have more than one-year

 

data, but would you anticipate that the patients

 

will continue every six-week intravitreal

 

injections for the rest of their lives?

 

DR. ADAMIS: That is an important

 

question. It is one of the questions we ask in the

 

second year of the design. We want to know,

 

obviously, about the safety in the second year and

 

then an important question was do people need to be

 

122

 

on this for the second year. So, the trial design

 

is one of randomized discontinuation to try to get

 

to an answer as regards that very important

 

question, do people need another nine injections?

 

DR. DUNBAR: Dr. Lehmer?

 

DR. LEHMER: I noticed in the data that

 

most of the p values were significant, at least in

 

the graphs and the tables, for the 0.3 mg and the 1

 

mg doses, and for the higher dose there was less

 

incidence, at least in the tables and graphs, of

 

statistically significant levels. Are there any

 

conclusions you have drawn about that? Is more not

 

better, etc.?

 

DR. GUYER: It is a great question and

 

obviously one we spent a great deal of time

 

analyzing. There really is no definite answer to

 

why the 3 mg, as you mentioned, perhaps appeared

 

not to do as well. Slide E-51, please.

 

[Slide]

 

There is one possible explanation that we

 

have looked at. This shows the mean change in

 

vision over time for each individual trial. On the

 

123

 

left is study 1004, on the right is study 1003.

 

What you can see is that in one of the trials,

 

1004, this is the 3 mg dose, this is the 0.3 mg and

 

1 mg dose, going head-to-head pretty throughout.

 

Of course, here is the usual care sham. It seemed

 

that the 3 mg dose in one of the trials didn't seem

 

to do quite as well as the other two active

 

treatments--still doing better than the sham. In

 

1003 you can see that actually all three doses

 

seemed to do equivalently.

 

So, one possibility is, you know, six

 

different events, three doses, two trials, one out

 

of six times by chance, it is possible that the 3

 

mg dose didn't do as well. Of course, as you

 

mentioned, all of these clinical parameters,

 

secondary parameters, etc., are all dependent on

 

the other. That is one explanation.

 

The thing that we do know, however, is

 

that the 0.3 mg dose, which represents the lowest

 

efficacious studied dose, clearly hit the primary

 

endpoint in replicate trials and showed consistent

 

behavior throughout the trials. Because of safety

 

124

 

issues, theoretical safety issues, we believe that

 

the 0.3 mg dose has met the requirements to be an

 

effective treatment here.

 

DR. DUNBAR: Dr. Miller?

 

DR. MILLER: Thank you. In terms of the

 

number of patients that have been in the trials,

 

are you comfortable or is the model sufficient

 

enough to tell us that there are no adverse risks

 

related to the population? For example, with Vioxx

 

we now know that after a period of time there are

 

now people that are coming up with cancer, that it

 

is causing cancer in some of them. Have you given

 

it to enough patients so that you would know if

 

there were rare cases where other problems would be

 

caused?

 

DR. ADAMIS: The population studied was

 

large in that it was 1,200 patients, large for an

 

ophthalmology trial. But for very rare events, and

 

this is a problem faced with all clinical trials,

 

that show up in patients on the order of one in

 

every 10,000 or so, you just don't have the power

 

in these types of trials to detect in a very

 

125

 

air-tight manner those signals.

 

That being said, with the power we have,

 

and we do have some significant power according to

 

the guidelines Dr. Chambers talked about earlier,

 

we were happy to see with all three doses that

 

there wasn't any evidence of toxicity, either

 

systemically or in the eye, related to the drug.

 

But we will never know with absolute certainty for

 

those very, very rare events.

 

DR. MILLER: Thank you.

 

DR. GUYER: Also as Dr. Chambers

 

mentioned, Dr. Miller, the fact that we had a

 

higher dose, 10 times higher than the dose that we

 

believe is the correct dose, gives us some comfort

 

that a dose 10 times higher has been studied in

 

many patients. So, that gives us more comfort than

 

in many other trials.

 

DR. MILLER: Thank you.

 

DR. DUNBAR: Ms. Knudson?

 

MS. KNUDSON: I was just curious about the

 

necessity for pregnancy tests and two forms of

 

birth control when your animal data indicated no

 

126

 

problems and you are dealing with a very elderly

 

population. What was the necessity for this?

 

DR. ADAMIS: That is the miracle of modern

 

medicine. There are people over age 50 having

 

babies. It happens rarely but, you know, in this

 

case one can't be overly cautious so that was the

 

reason for that.

 

MS. KNUDSON: Two forms of birth control?

 

DR. ADAMIS: That is the standard protocol

 

in clinical trials.

 

[Laughter]

 

MS. KNUDSON: Did you pay for them?

 

DR. ADAMIS: I don't know. I will find

 

out.

 

[Laughter]

 

DR. DUNBAR: Are there any additional

 

questions? Dr. Steidl?

 

DR. STEIDL: Thank you for a superb

 

presentation. I think I understand some of the

 

rationale behind the 15-letter vision loss, the

 

primary endpoint. I understand the comparison to

 

PDT. Most of my patients, when I say "you have

 

127

 

only lost two lines of vision; this is a success,"

 

you know, they are not too happy with that, nor to

 

they agree with me.

 

I guess one of the things that I really

 

wanted to know, there was, I guess as far as I

 

could see, only one paragraph devoted to quality of

 

life. Of course, if a patient has one bad eye they

 

may notice more in the treated eye than if the

 

other eye is 20/20, but I am just curious what your

 

feelings are, your comfort is with this. As

 

physicians, we often think it is good for the

 

patient but, you know, in terms of the patient's

 

perspective on this what have you gotten from your

 

trial?

 

DR. ADAMIS: Sure. First, I would also

 

mention that this difference, as we mentioned

 

earlier, is against a usual care control and

 

actually provides for approximately three-quarters

 

of patients the only positive one-year data. So,

 

we think that that is very, very important, in

 

addition to the fact that the primary endpoint was

 

supported by every secondary visual angiographic

 

128

 

endpoint that we saw. So, that gives us great

 

confidence in our endpoint. Slide number Q-2,

 

please.

 

[Slide]

 

We agree that it is very important to look

 

at quality of life measurements, and we did using

 

the NEI-VFQ25 which, as many of know, is a

 

validated measure. It was only measured in one of

 

the trials, in trial 1004 which was the North

 

America trial. Because validated foreign language

 

versions were not consistently available we did it

 

in just the one trial. For that reason, we were

 

significantly under-powered. We could not pool the

 

data. The results were not statistically

 

significant but there were trends that favored

 

pegaptanib treatment. As I said, it was

 

under-powered really to detect the small but

 

potentially meaningful differences between groups.

 

Slide Q-3.

 

[Slide]

 

We can see some of these differences. It

 

is important to mention that a 5 or more difference

 

129

 

in the LS mean is considered potentially to be

 

meaningful. So, anything between 0 and 5 is

 

probably not meaningful. What you can see here are

 

5 data points that hit that 5 level for the 0.3 mg

 

dose, and this has to do with color vision,

 

peripheral vision, distance vision, social

 

functioning and role limitations. So, these strong

 

trends, despite a very under-powered sample, give

 

us some confidence that the QOL, very much as the

 

angiography and the other secondary visual

 

endpoints, also supports the primary endpoint, and

 

we are getting significant benefit for these

 

patients, not, as you say, just measuring on an eye

 

chart, but actually benefit that is important to

 

them in the real world to help them get around.

 

DR. DUNBAR: Thank you very much. At this

 

point we will take a 15-minute break and begin

 

again at 10:30.

 

[Brief recess]

 

DR. DUNBAR: We will begin the agency

 

presentation by Dr. Jennifer Harris.

 

FDA Presentation

 

130

 

DR. HARRIS: Good morning.

 

[Slide]

 

I am Dr. Harris and I was the primary

 

medical reviewer for Macugen.

 

[Slide]

 

I am not going to repeat everything that

 

the sponsor has presented to you; I am just going

 

to try and bring up the salient points to try and

 

give you an idea of how we went through the

 

application and what we thought was important to

 

present this morning.

 

I will go briefly over the study design;

 

the efficacy results for each individual study so

 

you can see what replicated itself and what did not

 

replicate itself; conclusions about the efficacy; a

 

safety overview of the combined study, the pooled

 

study overview. There are a couple of specific

 

safety concerns that we want to talk about a little

 

bit more and the sponsor discussed a little bit

 

this morning but we just wanted to go over those

 

again. Then conclusions about the safety and then

 

we are going to briefly go through the questions

 

131

 

that we are going to pose to the advisory committee

 

and, of course, you will see them again after

 

lunch.

 

[Slide]

 

Again, there were two Phase III studies,

 

1003 which was an international study, and 1004

 

that was done predominantly in North America.

 

[Slide]

 

Both trials were randomized,

 

double-masked, sham-controlled as you have heard,

 

dose-ranging, multicenter trials. Within the

 

trials patients received intravitreous injections

 

of either 0.3, mg, 1 mg or 3 mg every 6 weeks for

 

54 weeks. These trials were actually 2 years in

 

duration. The data that we will be looking at

 

today is only from the first year of the trial. At

 

the 54-week time period these patients were

 

re-randomized.

 

[Slide]

 

This is just a little schematic, just to

 

show you where we are. We are at week 54 and this

 

is the data that you will be seeing. The two-year

 

132

 

data is probably sometime soon I think, this month

 

or next month. This is not the data you will see

 

today.

 

[Slide]

 

Subjects that were enrolled in these

 

trials were over the age of 50. They had subfoveal

 

choroidal neovascularization secondary to AMD. The

 

total lesion size was less than 12 disc areas, and

 

greater than 50 percent of the lesions had to be

 

active CNV. The best corrected visual acuity had

 

to be between 20/40 and 20/320. These patients, as

 

you have heard, were allowed to have PDT before

 

entering into the trial and they were also allowed

 

to have PDT during the trial. Prior to the trial

 

they could not have had anymore than one prior

 

photodynamic therapy treatment, and the patients

 

could not have had any previous subfoveal laser

 

treatment.

 

[Slide]

 

The primary efficacy endpoint, again, was

 

a proportion of patients who lost less than 15

 

lines of visual acuity from baseline at 54 weeks.

 

133

 

Those are considered responders. Secondary

 

efficacy endpoints were the proportion of patients

 

gaining greater than 15 letters of vision,

 

proportion of patients gaining more than zero

 

letters of vision, and a mean change in visual

 

acuity.

 

[Slide]

 

Just to give you an idea of the subject

 

disposition, there are approximately 612 patients

 

in the 1003 study that were randomized to

 

treatment. Approximately 53 percent of these

 

patients discontinued. As you can see, it was

 

pretty well distributed. The treatment groups were

 

consistent, with approximately 10 percent or so of

 

patients discontinuing in each of the treatment

 

groups.

 

[Slide]

 

For the second study, 1004, we see the

 

same thing. The distribution of patients enrolled

 

was approximately the same in each treatment group,

 

including sham and, again approximately 10 percent

 

or so of the patients discontinued therapy.

 

134

 

[Slide]

 

I am showing you this, not that I think

 

that you can probably read it but just to give you

 

an idea of who was enrolled and to show you really

 

that the groups were well balanced. They were very

 

well balanced between all three active treatment

 

arms, including the sham. The demographics of

 

patients that were enrolled in the 1003 trial were

 

consistent with patients who actually had the

 

disease.

 

I also wanted you to note down at the

 

bottom that patients with all subtypes of

 

neovascular AMD were enrolled. There was a

 

substantial number of patients with predominantly

 

classic and occult lesions that were enrolled in

 

the trial.

 

[Slide]

 

The same thing can be seen for study 1004

 

where the groups, again, were well balanced, were

 

representative of the population in which the

 

disease was seen and, again, all three subtypes of

 

neovascular AMD were represented.

 

135

 

[Slide]

 

Now I will go into the efficacy results.

 

Before we go to the efficacy results I want to just

 

put up this slide to show you how corrections were

 

made in the p value. As we went into the Phase III

 

trials we did not go into these trials with one

 

optimal dose and, therefore, you know if you have

 

one optimal dose, one time point, you look at the

 

0.05 value and you can determine whether the drug

 

works or not. We went into the Phase III trials

 

and we had three different doses so we had to find

 

a way to correct for that. A decision was made to

 

use the Hochberg procedure to actually control for

 

these multiple comparisons.

 

With the Hochberg procedure, each of the

 

treatment groups was compared to sham and if all

 

three of the p values were less than 0.05, then we

 

were considered to have three active doses. If

 

not, if two of the p values were less than 0.025,

 

then we had two active doses. Or, if one of the p

 

values was less than 0.0167, then we had one active

 

dose. If none of these criteria were met, then we

 

136

 

had no active doses. So, as you go through the

 

results you may see some 0.05 or even 0.025 and

 

that may or may not mean that that was actually

 

statistically significant.

 

[Slide]

 

This is the primary efficacy result for

 

study EOP1004. As you will see, at month 12 for

 

the 0.3 mg dose there was approximately 67 percent

 

treatment effect versus 53 percent of the sham

 

group. This was a statistically significant

 

result, with a p value of 0.016. Again, the actual

 

treatment effect is about 14 percent over sham.

 

The 1 mg group did show that there was a 67 percent

 

treatment effect versus 53 in sham. However, this

 

did not meet our pre-required p value.

 

[Slide]

 

For study 1003 we have similar results

 

and, again, the 0.3 mg group shows approximately a

 

73 percent treatment effect versus 60 percent of

 

sham, with a p value of 0.01. In this trial it was

 

also seen that the 1 mg group was also

 

statistically significant with a 75 percent

 

137

 

treatment effect versus 60 percent.

 

So, it appears that both the 0.3 mg and

 

the 1 mg group have approximately a 15 percent

 

treatment effect over sham, with the 0.3 mg

 

replicating its results in both trials and the 1 mg

 

dose did not replicate these results.

 

[Slide]

 

You have seen this graph before. This

 

shows you what was happening to the patients'

 

visual acuity in study 1004 throughout the first

 

year of the study. What we see is that all

 

patients continued to lose vision in all treatment

 

groups, including sham, throughout the first year

 

of the study. That being said, it does appear that

 

the patients in the sham group lose vision at a

 

higher rate than those in the other three active

 

treatment groups.

 

[Slide]

 

In study 1003 we see the same thing. All

 

patients continued to lose vision throughout the

 

first year of study on active treatment and in

 

sham, but those patients in the sham group appeared

 

138

 

to lose vision at a faster rate than those in the

 

Macugen treatment group.

 

[Slide]

 

We have a chart similar to the sponsor's

 

in that we looked at a subgroup analysis. The

 

reason why we do that is to make sure there isn't

 

one particular group that is actually driving the

 

results. As we see in this chart for study 1004,

 

if we look at all the subgroup analyses that were

 

done, the type of AMD, color of the irises, the

 

lesion size, baseline demographics and male/female,

 

what we see is that for each subgroup analysis the

 

0.3 mg group shows a higher response rate than the

 

sham group in each of the subgroups.

 

[Slide]

 

This was repeated in study 1003 where,

 

again, the 0.3 mg group shows a higher response

 

rate in all of the subgroup analyses over sham.

 

[Slide]

 

We also wanted to take a look at lesion

 

size, basically because of the proposed mechanism

 

of action of Macugen, and that is to inhibit

 

139

 

endothelial cell growth. So, we wanted to see

 

whether that was, indeed, happening. What we

 

noticed was that actually the total lesion size for

 

patients, as well as the total size of the CNV and

 

the total leak size, continues to increase for all

 

treatment groups. Even in patients receiving

 

Macugen, lesion size does increase but it does

 

appear that it increases to a lesser degree in the

 

0.3 mg group than in sham. However, it is noted

 

that it does increase in size.

 

[Slide]

 

The same thing was seen in the 1003 study

 

where, again, the total lesion size for all

 

treatment groups did increase in size, however, for

 

the 0.03 mg group it does seem to increase to a

 

lesser degree than in the sham group.

 

[Slide]

 

As you have heard, patients who entered

 

the trial were allowed to get photodynamic therapy,

 

which is an approved therapy for AMD. So, our

 

question became were we really seeing an effect of

 

Macugen or were we just really seeing the effect of

 

140

 

patients receiving an already approved therapy?

 

So, we took a further look at this and the first

 

chart you see here is the number of patients who

 

actually got on-study photodynamic therapy

 

treatment in study 1004. We see that approximately

 

the same amount of patients actually received

 

photodynamic therapy in all treatment groups,

 

including sham.

 

Another thing that we did note is that

 

while the protocol specified that only patients

 

with predominantly classic should have been allowed

 

to get photodynamic therapy, there were many people

 

who had occult or minimally classic CNV who also

 

received photodynamic therapy.

 

[Slide]

 

The same thing was seen in study 1003

 

where approximately the same amount of patients

 

across the treatment groups received photodynamic

 

therapy, with some occult patients and minimally

 

classic patients, again, receiving photodynamic

 

therapy. What was also interesting was that the

 

1003 study was an international study and you can

 

141

 

see that there were approximately half as many

 

patients who received PDT in the international

 

study versus the American study. That could be

 

based on practice patterns across the ocean.

 

[Slide]

 

We also wanted to look at not so much what

 

percentage of patients got photodynamic therapy but

 

were more patients in one group or the other

 

receiving more treatments? As we look at this

 

chart for study 1004, we are looking at the total

 

number of photodynamic therapy treatments. We see

 

that there is substantially less number of

 

treatments that were given in the 0.3 mg group

 

versus that given in the sham group.

 

[Slide]

 

For study 1003 the results are similar.

 

While there is not that big of a difference between

 

wham and the 0.3 mg group, the point is that there

 

were less photodynamic therapy treatments given in

 

the 0.3 mg treated group.

 

[Slide]

 

Lastly, we wanted to look at the results

 

142

 

and say, well, it looks as though the same

 

percentage of patients were receiving photodynamic

 

therapy; it looks as though the same number of

 

treatments were given. Well, did that make any

 

difference in terms of the responder analysis, the

 

primary efficacy endpoint?

 

So, what you are looking at here is the

 

responder analysis at month 12 for four different

 

groups, the first group being the group that

 

received no photodynamic therapy either before the

 

trial or during the trial. The second one are

 

those patients who only received pre-study PDT.

 

The third is a group that received on-study

 

photodynamic therapy only. The fourth group are

 

those patients who received pre-study and on-study

 

photodynamic therapy. The last line here is for

 

reference so you remember what the primary efficacy

 

results were for all patients that we just looked

 

at.

 

What we noted, which was good, is that the

 

majority of the patients who entered the trial

 

never had any confounding or problems with

 

143

 

photodynamic therapy, and that their results

 

actually were pretty consistent with the overall

 

results. In terms of the number of patients who

 

received photodynamic therapy either before or

 

during the trial, or both before and during the

 

trial, those numbers were so small that we really

 

can't make any conclusions about whether receiving

 

photodynamic therapy before or during the trial has

 

any effect on the efficacy results.

 

[Slide]

 

Similar results were seen for study 1003,

 

where we looked at the number of patients who

 

actually received photodynamic therapy. They are

 

extremely small and no conclusions can be drawn

 

from using concomitant PDT therapy. The results

 

for those patients who received no photodynamic

 

therapy, again, were consistent with the overall

 

efficacy results.

 

[Slide]

 

We were curious, I mean, our primary

 

efficacy endpoint is really those patients who lose

 

less than 15 lines of vision. We know, based on

 

144

 

the disease process, that patients will continue to

 

lose vision so those patients who lose less than 15

 

lines, that is probably a good thing for them. But

 

we wanted to know was there any possibility that

 

you could actually gain vision if you use this

 

drug. So, we looked at the number of patients who

 

gained greater than 15 letters of vision.

 

If you look at study 1004, actually there

 

is a statistically significant increase in patients

 

who actually gained vision in the 0.3 mg group and

 

the 1 mg group as compared to sham. However, those

 

results were not replicated in the 1003 study where

 

you see no statistically significant gain in

 

vision.

 

[Slide]

 

So, in terms of our efficacy conclusions,

 

we believe that Macugen 0.3 mg does reduce the risk

 

of vision loss in patients with neovascular

 

age-related macular degeneration. But keep in mind

 

that there is only approximately a 15 percent

 

treatment effect over sham, and that there is no

 

clinically meaningful increase in vision seen in

 

145

 

patients during the first year of using Macugen.

 

[Slide]

 

The sponsor has presented all of the

 

safety results. I am not going to go back through

 

all of them. I just want to say that we agree that

 

similar events were seen in all treatment groups

 

and no dose-dependent adverse events were seen.

 

Most of the events, we think, were related to the

 

act of giving an intraocular injection itself and

 

no so much to the drug. The majority of adverse

 

events, things like eye pain, superficial punctate

 

keratitis, floaters, iritis are those things that

 

we commonly seen with intraocular injections of any

 

drug.

 

[Slide]

 

But there are two safety concerns that we

 

want to talk about a little bit more. That is,

 

endophthalmitis again and also a little bit about

 

systemic VEGF inhibition and what that could mean.

 

[Slide]

 

In the database we had there were 16 cases

 

of endophthalmitis. What we heard this morning is

 

146

 

that actually there are 2 more cases. I guess

 

there is a total of 18 now. Of those 16 cases, all

 

of those 16 cases occurred in the pegaptanib sodium

 

treated patients, and none of the cases were in the

 

sham treated patients. All 16 cases occurred

 

within one week of injection.

 

[Slide]

 

So, I took a look at what kind of

 

organisms were actually coming out of the

 

endophthalmitis samples. We see that of the 16

 

cases, the overwhelming majority are those types of

 

organisms that are commonly seen around the lid or

 

around the ocular area--coagulase negative Staph.,

 

Staph. epi. There were about 6 cases that were

 

actually negative on the samples. So, it stood to

 

reason that maybe the problem was with the

 

injection procedure and the sponsor did take a look

 

at that and made some changes.

 

[Slide]

 

The original procedure called for the

 

patients to get 2-3 drops of 50 percent saline

 

diluted, 10 percent povidone-iodine or they could

 

147

 

receive 1 drop of topical antibiotic.

 

[Slide]

 

An amendment was made in the protocol

 

after I think 12 cases of endophthalmitis, and it

 

was changed so that patients would undergo a more

 

sterile preparation procedure, similar to most

 

intraocular surgeries, and patients would be

 

prepped and draped similar to intraocular surgery

 

and patients would receive either pre-injection

 

topical antibiotics for 3 days prior to injection

 

or 5 percent povidone-iodine flush immediately

 

prior to injection.

 

[Slide]

 

So, what happened to the endophthalmitis

 

cases? Well, we saw in the database that actually

 

13 of the 16 cases occurred before the protocol

 

amendment. Three of those 16 cases occurred within

 

3 months after the protocol amendment. This is

 

actually wrong now because I guess there have been

 

2 additional cases since that time. Based on the

 

data that we had, there had not been any new cases

 

of endophthalmitis 3 months after the protocol was

 

148

 

amended.

 

[Slide]

 

I just want to touch a little bit on

 

systemic VEGF and what that could or could not mean

 

in terms of this. Obviously, having VEGF is a good

 

thing in some instances and it is a bad thing. It

 

is a bad thing in the eye. We want to inhibit that

 

in cases like AMD. But we want it in the systemic

 

circulation, mainly because it plays an active role

 

in cardiac angiogenesis. This is important in

 

collateral blood vessel formation in patients with

 

myocardial ischemia. It is also an important

 

vasodilator and it helps to maintain coronary

 

artery blood flow and helps maintain patency of

 

coronary arteries.

 

[Slide]

 

So, what we did is we looked at the whole

 

database and we said, well, are there any events

 

within the database, the adverse event database,

 

that could possibly in any way be related to VEGF

 

being inhibited in the systemic circulation?

 

Of all the things that we came up with--

 

149

 

arrhythmia; atrial fibrillation which could be an

 

early indication of myocardial ischemia;

 

bradycardia; chest pain; coronary artery disease,

 

not just those cases where patients obviously came

 

into the study with a known diagnosis but those

 

patients who were diagnosed with coronary artery

 

disease during the trial; and myocardial

 

infarction--and we looked at the database and said,

 

well, is there a problem? Could we actually have

 

these systemic anti-VEGF effects based on the

 

intravitreal injections? What you see here on the

 

chart is that actually all the numbers are pretty

 

small across all the groups, and there is no real

 

indication that the intravitreal injection of

 

pegaptanib will have any systemic anti-VEGF

 

effects.

 

[Slide]

 

Just for completeness, in terms of the

 

death rate, there were approximately 25 patients

 

who did die during the study, approximately the

 

same in each study, and the majority of causes were

 

actually things like cardiovascular events,

 

150

 

malignancies and they were pretty typical of the

 

age of the population that we were studying. So,

 

we think those events were really due to the

 

population and not actually to the drug.

 

[Slide]

 

In terms of safety, similar events were

 

seen in all treatment groups. The most frequently

 

occurring adverse events related to the intraocular

 

injection itself and not to the drug. The risk of

 

endophthalmitis appears--and I have to emphasize

 

"appears" since there may be more cases that we

 

haven't seen--to be minimized by sterile technique

 

and there does not appear to be an apparent

 

increase in the risk associated with systemic

 

anti-VEGF activity.

 

[Slide]

 

We will just go over the questions

 

briefly. You are going to see the questions again

 

this afternoon but just so you can start thinking

 

about them. The questions that we would like to

 

have discussion about are, one, based on the

 

inclusion and exclusion criteria, are there

 

151

 

patients excluded from the studies that you believe

 

need to be studied?

 

Visual acuity measurements were conducted

 

using the ETDRS scale at 2 meters. The validity of

 

the ETDRS scale was established based on ratings at

 

4 meters. Are the visual acuity findings

 

sufficiently robust to overcome the potential bias

 

introduced by visual acuity measurements taken at 2

 

meters?

 

[Slide]

 

Has sufficient data been submitted to

 

evaluate the efficacy and safety profile of

 

pegaptanib sodium for the treatment of the

 

neovascular form of AMD? If not, what additional

 

data are needed?

 

Are additional analyses of the current

 

data needed to understand the efficacy or safety of

 

pegaptanib sodium for the treatment of the

 

neovascular form of AMD?

 

Has the concomitant use of PDT therapy

 

with pegaptanib been explored sufficiently? Are

 

there concerns with using this predictive

 

152

 

concomitantly with PDT?

 

[Slide]

 

Do the route and/or frequency of

 

administration of the drug raise any concerns that

 

are not addressed by the studies?

 

Endophthalmitis was observed in

 

approximately 2 percent of patients in the studies.

 

What is the optimal follow-up needed to minimize

 

the impact of potential endophthalmitis cases?

 

Are there any other adverse experiences

 

that are of particular concern for this product?

 

VEGF has been shown to be an important

 

component in the development of collateral vessels

 

in ischemic heart disease. Inhibition of VEGF in

 

the systemic circulation could present a

 

theoretical increased risk of symptomatic

 

cardiovascular disease in the target population of

 

elderly patients with AMD. Has the adverse event

 

profile of the two randomized Phase III trials

 

raised any concern over the possible systemic

 

effects of this therapy? Is there additional

 

monitoring that should be in place for patients on

 

153

 

pegaptanib sodium therapy? Thank you.

 

Committee Discussion

 

DR. DUNBAR: At this point I would like to

 

open the floor for questions for either the agency

 

or for the company. Dr. Pulido?

 

DR. PULIDO: Thank you. Two questions,

 

the first one is you said the treatment effect was

 

15 percent. That is because you took the 67 minus

 

50. Again, I am not a statistician; I am a

 

clinician--shouldn't it be the difference divided

 

by 50 to give you 25 percent as the treatment

 

effect? So, the delta of 15 over the baseline

 

which is 50?

 

DR. CHAMBERS: There are obviously lots of

 

different ways to look at it. What we have been

 

doing for ease of description is just to describe

 

what the percentage difference is between the two

 

different modes of therapy, and we thought that

 

easiest to be described as just a 15 percent

 

difference in the percentage of people who have

 

lost 3 lines of vision.

 

DR. PULIDO: The other question I had, and

 

154

 

maybe it would be better answered by the company,

 

when one looks at the serum levels, is that the

 

total amount of the drug that is being measured or

 

is that the unbound free form that is being

 

measured?

 

DR. ADAMIS: It is the total level.

 

DR. DUNBAR: Dr. Lehmer?

 

DR. LEHMER: Are there known levels for

 

VEGF or VEGF inhibition that are clinically

 

significant from the cardiovascular current

 

literature?

 

DR. ADAMIS: The short answer is no in

 

humans. The longer answer is that the most

 

sensitive signal of systemic VEGF inhibition is

 

hypertension. In the Avastin trials they picked it

 

up in their colon cancer, the renal study, their

 

lung cancer study, and some of those were much

 

smaller studies than ours and there was no evidence

 

of hypertension as a function of use of pegaptanib

 

in our study. So, I guess whatever that level

 

is--and it hasn't been determined--we are probably

 

well below that.

 

155

 

DR. DUNBAR: I have a question for the

 

agency about the duration of use of the drug. I

 

would like to know who will decide when to stop

 

therapy, the agency, the sponsor, or the patient's

 

physician? Is this something that will be

 

specified by the agency in relationship to the drug

 

approval process? Would it be included in the

 

labeling? Or, is this something that we won't know

 

for many years and would be addressed in further

 

labeling decisions?

 

DR. CHAMBERS: The most accurate answer is

 

that I think we will not know for a number of

 

years. The answer that everybody would like to

 

know is probably best studied by a 10-15-year study

 

of giving a particular product. We obviously run

 

into the difficulty of not having a therapy that is

 

potentially valuable available during the time that

 

we are doing that so we have chosen to take a path

 

where, if everything else looks good--and I will

 

repeat that decisions have not been made on this

 

particular product and there are lots of other

 

parameters that still need to be reviewed, but if

 

156

 

this product otherwise looks fine we would

 

potentially label it based on the information we

 

have available.

 

As you have heard, the sponsor presented

 

that as of their latest data safety monitoring

 

committee they have 90-some odd percent of the

 

information for the two years. To the extent that

 

we have two-year data, we will list two-year data.

 

If we don't, we will list one-year data and as more

 

data becomes available we intend to amend the label

 

to reflect what we know.

 

DR. BULL: I have one thing to add to

 

that. There is the opportunity for the committee

 

to make recommendations if you are uncomfortable

 

with the degree of follow-up, things such as Phase

 

IV commitments. I mean, there are a number of

 

options that can be systematically required of the

 

sponsor to do to look at the long term.

 

DR. GUYER: I think in answer to your

 

question, also clinical judgment of the

 

ophthalmologist will decide much of it until, as

 

Dr. Chambers mentioned, we do have the answer from

 

157

 

continuation of trials. If a physician sees a

 

patient that is, for example, scarred down and

 

realizes there is no further benefit of treatment,

 

we would expect that the physician would stop that

 

treatment, whenever that is and. Similarly, if the

 

physician sees active bleeding going on they might

 

continue it. So, I think a lot of it will be in

 

the clinical ophthalmologist's hands, at least in

 

the beginning.

 

DR. DUNBAR: That was my concern, that a

 

patient with a quiescent, scarred lesion was

 

vulnerable, worried about their blindness and might

 

subject themselves to very frequent injections for

 

a long period of time. Dr. Lehmer?

 

DR. LEHMER: We are certainly in the era

 

of implantable sustained release drug delivery

 

devices. At what point time-wise, if therapy is

 

determined to need to continue past a year or past

 

two years, should a recommendation for conversion

 

of this drug to an implantable device become

 

necessary?

 

DR. ADAMIS: It is an area that we are

 

158

 

very interested in, in the laboratory of the

 

sponsor. So, we are working on alternative

 

formulations to see if we can get an extended

 

release profile in implantables of that sort. I

 

think ultimately that may end up being an

 

improvement.

 

DR. DUNBAR: Ms. Knudson?

 

MS. KNUDSON: I just don't seem to

 

understand why the DSMB permitted the trial to

 

continue with the sham arm when at every point it

 

appears the sham arm is inferior to every drug dose

 

that was given. This is a disease, as I understand

 

it, which continually advances and one should treat

 

patients.

 

DR. ADAMIS: Yes, the data safety and

 

monitoring committee, their charge was to monitor

 

for safety. But the point you raise is a very

 

important one. So, in this randomized,

 

discontinuation trial design we actually allow

 

people who discontinue the drug and lose two lines

 

of vision to go back on so patients are not forced,

 

by and large except for a very small group, to stay

 

159

 

on sham for two years.

 

DR. DUNBAR: I have a question for Dr.

 

D'Amico. I was interested in Dr. Harris'

 

presentation that 6/16 endophthalmitis patients had

 

sterile endophthalmitis. I wonder, with your

 

experience with endophthalmitis, if you could tell

 

us do you think that these patients had infectious

 

endophthalmitis that were culture negative, or do

 

you think that that may be more of inflammatory

 

response?

 

DR. D'AMICO: Yes, in the trial, looking

 

for both of those things, that is inflammation

 

after injection or specifically infections, we

 

found really no evidence of a widespread

 

inflammatory effect at all. In studies of

 

endophthalmitis in general, for example after

 

cataract or other forms of ocular surgery,

 

invariably large studies always find that

 

approximately two-thirds will be culture positive

 

and one-third are, inexplicably, culture negative.

 

Now, what are those one-third? Well, some of them

 

will be organisms that have just not been

 

160

 

successfully collected by the culture technique.

 

Perhaps the specimen was too small; perhaps the

 

laboratory didn't plate it properly, or something

 

of that nature. Some of them may be fastidious

 

organisms that are difficult to culture. But

 

clinically I think we treat those cases as presumed

 

infectious. The patients had acute presentations

 

and they were invariably managed with TAP and

 

antibiotic injection. So, I think that they mirror

 

my clinical experience with endophthalmitis cases,

 

except somewhat for their outcomes which were

 

surprisingly somewhat better. They suggested

 

somewhat better visual outcomes than we might see

 

in clinical cases that, for example, would occur in

 

another context, after cataract or something like

 

that. Have I answered your question?

 

DR. DUNBAR: Thank you. Dr. Gates?

 

DR. GATES: Any conclusions as to that?

 

Is it a smaller bacterial load perhaps with this

 

injection?

 

DR. D'AMICO: Well, it is a new

 

phenomenon. Certainly, these patients were

 

161

 

extremely well followed and they included, you

 

know, contact with the patient and education to

 

inform patients about side effects, etc. So, the

 

patients were promptly detected, but it could be

 

that the load that is introduced in an intravitreal

 

injection is lower and, consequently, it has a less

 

fulminating presentation, but I don't know.

 

I will raise it because someone will, it

 

also may be that there is some interaction between

 

a VEGF medication and a profound inflammatory

 

infection in an eye. But that remains completely

 

speculative but it is something interesting, as a

 

scientific point of view, for further research.

 

DR. ADAMIS: Just to follow on Dr.

 

D'Amico's comments, there are data in the

 

laboratory now that VEGF can be pro-inflammatory,

 

and in models of ocular inflammation VEGF levels

 

come up and it is associated with the vitritis and

 

flare, and we have published, and others have, that

 

if you block VEGF in that sort of instance you can

 

decrease the inflammation as well as the leak. So,

 

it is speculation, as Dr. D'Amico said, but it is a

 

162

 

plausible hypothesis that it may be having somewhat

 

of an anti-inflammatory effect as well and you get

 

less standard damage that occurs when neutrophils

 

rush in in a case of endophthalmitis, but it is a

 

theory.

 

DR. DUNBAR: Dr. Chinchilli?

 

DR. CHINCHILLI: Yes, I have a question

 

for the agency. In the briefing document you

 

showed the results from the worst-case analyses. I

 

notice that in your presentation you really didn't

 

discuss that. Is there a reason you didn't present

 

them today? I mean, how do you feel about--well, I

 

will tell you that I think you shouldn't do them

 

but I was wondering why you didn't present them but

 

they are in the briefing document, or am I reading

 

too much into that?

 

DR. CHAMBERS: We do a large number of

 

analyses, which are neither shown in the briefing

 

document nor shown in the presentation, to try to

 

look at the robustness of the findings. We thought

 

it instructive to give what potentially is a bottom

 

lower limit and include it in the briefing document

 

163

 

just to try and frame people's idea of what the

 

magnitude could be of inclusion or exclusion of

 

different findings, but since it does not

 

necessarily represent an accurate finding we didn't

 

think, from a time perspective, that it was worth

 

continuing to present in a presentation.

 

DR. CHINCHILLI: Well, I think it is

 

highly inaccurate. I know you try to look at the

 

bounds but I think they are highly inaccurate

 

bounds. Later today--I don't know if you want to

 

get into this now, but I do have some

 

recommendations about analyses, endpoints and

 

things like that. So, I don't have to get into

 

that now.

 

DR. CHAMBERS: We don't disagree with you.

 

We don't think either of the analyses are

 

necessarily the most accurate; we could do

 

something in between.

 

DR. DUNBAR: Is there additional

 

discussion for the agency presentation at this

 

point in time?

 

[No response]

 

164

 

Now we have a decision about our agenda

 

because we have significantly more time with our

 

morning session than we expected. It is imperative

 

that we start the open public hearing as it is

 

scheduled at 1:00 p.m. so that the public can have

 

their voice in this matter. We have two options.

 

One is that we can take a longer lunch period and

 

then start the agenda for the afternoon as

 

previously published. Or, we can begin to answer

 

some of the FDA questions now and start our lunch

 

closer to the scheduled time and then have the

 

public hearing at 1:00 p.m.

 

So, let's begin to answer some of the FDA

 

questions now and then we will, of course, begin

 

the public hearing at 1:00 p.m.

 

DR. CHAMBERS: We would like to hear some

 

general discussion as opposed to just going through

 

the questions. So, that may be a better use of

 

some of the time this morning, just a general

 

discussion of the different topics that are on

 

there and then specifically go through questions

 

later.

 

165

 

DR. DUNBAR: Then we will start with Dr.

 

Chinchilli in terms of general discussion from the

 

committee.

 

DR. CHINCHILLI: Well, I mentioned this in

 

my previous question and I would like to talk about

 

the endpoint that is used and the analysis. I

 

don't quite understand why the analysis was done

 

this way, and then looking at the briefing

 

documents I see that this is the way the FDA

 

recommends the analysis be done. But there is

 

interest in less than 15-letter loss. I think it

 

would be better to reverse the definition, to look

 

at someone who fails, someone who is a treatment

 

failure who has 15 or more letter loss and then

 

look at the time to occurrence of that event. This

 

way you would better handle the dropouts and the

 

censoring that occurs.

 

Now, I realize the subjects in these

 

particular studies come into the study every six

 

weeks so you don't have a nice continuum for

 

determining when this treatment failure takes

 

place, but at least you can have more of a discrete

 

166

 

failure time process. It would just get away from

 

looking at these extreme cases, the worst-case

 

scenario that the agency likes to look at in terms

 

of bounding the results. It just seems to me that

 

that would be a better approach to the analysis,

 

that is, to reverse the definition and talk about

 

treatment failure and look at time until treatment

 

failure occurs, and doing time to event analyses.

 

That would be a much more accurate analysis, I

 

feel. I don't know how the agency feels about that

 

or if they would consider that. I don't know if

 

there is some reason I am missing that that is not

 

a good approach. And, maybe the company would like

 

to comment on that as well.

 

DR. CHAMBERS: We are certainly open to

 

looking at a number of different types of analyses

 

and different ways of doing it. The general

 

recording of visual acuities has been every three

 

months, not every six weeks. Consequently, you

 

have set fixed time points when you are getting the

 

information. So, time to event, when you are fixed

 

at every three months, we have not thought as being

 

167

 

particularly meaningful.

 

Whether you look at it on either side of

 

this coin, whether it be the people that improve or

 

the people that fail, we have generally thought as

 

being relatively similar. There are certain biases

 

that go in as far as the dropouts and which way

 

they are treated. Obviously, if you are assuming

 

that somebody is going to drop out and they never

 

get seen again, they don't get counted as a loss.

 

That accounts for some of the reason for doing a

 

number of the analyses that we do.

 

But, as I said, we do a large number of

 

different analyses looking at these things to try

 

and look for the robustness of the findings. In

 

this particular case, any way you look at it you

 

have very similar results. So, we did not stress

 

how it needed to be presented for this particular

 

case.

 

DR. CHINCHILLI: I agree. I mean, the

 

dropouts in these two trials was between 10-12

 

percent so that is not extreme. But I think you

 

are going to have trials where you may see more

 

168

 

dropouts, a higher rate than that, and all these

 

cases that you are proposing for analysis all

 

involve some form of data imputation. If you look

 

at the treatment failure approach and time to event

 

analysis, you know, you account for that censoring

 

and you are not imputing data the way you do in the

 

current methods. You know, I think I am getting

 

off the tangent here, but it just doesn't sit well

 

with me the way the outcome is constructed and all

 

these analyses are performed that involve some form

 

of data imputation. Again, I agree. I don't think

 

it makes a bit of difference with these two

 

particular trials here but in general it is not

 

really good methodology.

 

DR. CHAMBERS: We certainly are interested

 

in additional comments you have along that,

 

although I am not aware of any method that doesn't

 

have some type of bias and some type of assumptions

 

in the way it is presented, including the methods

 

you are discussing.

 

DR. DUNBAR: Dr. Lehmer?

 

DR. LEHMER: I just want to make sure I

 

169

 

understand correctly that, with regard to the

 

analyses, the intent-to-treat is what has been

 

presented, being the most inclusive; the

 

per-protocol analysis being the most exclusive. As

 

I understand from the briefing, when the two were

 

compared there were no significant differences and,

 

therefore, that is why we are using the

 

intent-to-treat because we want to be as inclusive

 

as possible to get the safety data. Is that a

 

correct interpretation of why we are using the

 

intent-to-treat analysis?

 

DR. ADAMIS: The safety data population is

 

even a little bit larger. Everybody was randomized

 

and received one treatment. The intent-to-treat

 

was the folks who had one baseline vision as well.

 

DR. GUYER: Can I have slide E-101, please?

 

Maybe we can just summarize this.

 

[Slide]

 

This shows the definition of the various

 

populations that we looked at, and it shows that

 

the all-randomized group were those that received

 

an actual randomization number. In this case it

 

170

 

was 1,208 and that represents the largest

 

number--as you said, one extreme. The safety group

 

received study drug, and that was 1,190, slightly

 

fewer. The intent-to-treat were patients, by the

 

sponsor's definition, that received study drug and

 

had an observed baseline vision. That was 1,186.

 

The per-protocol was all of the ITT patients that

 

had an observed post-baseline vision and no major

 

protocol violation. So, as you mentioned, it is a

 

much smaller group because they observed the

 

protocol perfectly and also had an observed time

 

point at week 54. That brings you down to 1,144.

 

Then you have a week 54 observed which are the

 

actual patients who received the study drug and had

 

a baseline vision and also a week 54 vision, and

 

that is 1,085.

 

[Slide]

 

Just to illustrate further maybe some of

 

the differences, E-102 shows again, starting with

 

the all-randomized where you start with 100 percent

 

of your population, going down to week 54 where you

 

get 92 percent of the data, at least for the 0.3

 

171

 

mg.

 

[Slide]

 

Finally, if we go to slide E-103, this

 

again shows just the two extremes, so to speak, the

 

all-randomized with an LOCF, which is in the FDA

 

briefing book, and the intent-to-treat where study

 

medication and baseline visual acuity occurred,

 

which is in our briefing book. Very importantly,

 

you can see that they are all the same. Slide

 

E-113.

 

[Slide]

 

We can see that on the left we have the

 

ITT population using an LOCF, which is in the

 

sponsor's briefing book, and we have the

 

all-randomized LOCF when there is a true ITT, in

 

the FDA briefing book. Then we have some of the

 

extremes, the per-protocol observed and the week 54

 

observed. Then you can see that we have very

 

robust data and that the sensitivity analysis and

 

different analyses all show, for the 0.3 mg dose, a

 

statistically significant change. So, any way you

 

look at it, either extreme, we see robustness of

 

172

 

the data.

 

DR. DUNBAR: Dr. Pulido?

 

DR. PULIDO: Ms. Chairwoman, your question

 

had been do we have general comments at this point,

 

and I would just like to state that I think the

 

data at this point looks very favorable. I would

 

say that my concerns about systemic complications,

 

from the data, appear very small.

 

My only concern is the long-term use and

 

the fact that there is the second aspect of VEGF

 

that only recently we are learning about, and I

 

would like to see some long-term follow-up using

 

ERGs and possibly visual fields in a small group of

 

these patients to make sure that there are no

 

long-term consequences of long-term use of this

 

drug. Otherwise, I am very impressed.

 

DR. DUNBAR: Are there any other general

 

comments from committee members? Dr. Steidl?

 

DR. STEIDL: You can correct my thinking

 

if I am wrong here, but it looked as though the

 

lesions continued in general to grow, maybe at a

 

slower rate, in the treated group. With the

 

173

 

half-life of, I guess, about four days and

 

effective vitreous concentrations that are weeks,

 

it would seem with that trend that it is quite

 

possible that this may be needed for a while beyond

 

the study time period. You know, somebody

 

mentioned the 0.16 percent per injection in

 

endophthalmitis rate. If you multiply that times

 

nine it gets pretty close to what was seen. I

 

don't know if it is valid to extrapolate that, but

 

then if you start thinking about doubling the time

 

and getting maybe to 3-4 percent, from my point of

 

view, it is getting pretty scary.

 

I don't tend to view those, from a retina

 

point of view, as sterile endophthalmitis because

 

in our lab we get a third to a half of clearly

 

infectious cases that don't come back positive. I

 

am wondering if that seems like a logical

 

assumption, that if this is to carry on we could

 

assume that the endophthalmitis rate would grow

 

proportionally.

 

DR. ADAMIS: Yes, I think your

 

interpretation of the data is correct and,

 

174

 

obviously, the cumulative risk increases as a

 

function of time. What our goal is, and we take

 

this responsibility seriously, is to make sure that

 

the injection procedure, which may be a modifiable

 

risk--that the risk gets down as low as possible.

 

We were fortunate in the second year after the

 

amendment to actually see that rate go down and,

 

subsequent to the amendment that occurred last May,

 

it is down to 0.03 percent per injection.

 

The other aspect of it that is equally

 

important is the visual outcome. That is, if this

 

event happens, are these patients being diagnosed

 

rapidly and being treated appropriately, and then

 

doing everything you can to preserve the vision as

 

a function of getting the infection. I think our

 

investigators did a rather superb job in the sense

 

that everybody was diagnosed within a week.

 

Everybody got intravitreal antibiotics. Over half

 

of them had vitrectomies and you saw the visual

 

outcomes. I will tell you that the visual outcomes

 

in the second year with the additional cases are

 

the same, if not better, than what you saw in the

 

175

 

data presented today. But your thesis is correct

 

that the more you use the drug, there obviously is

 

an incremental risk over time that increases.

 

DR. DUNBAR: Dr. Pulido?

 

DR. PULIDO: So, yes, there is a risk of

 

using intravitreal injections, but the alternative

 

is the present forms of treatment or systemic

 

medication that also increase the risk. It is a

 

small risk but I would rather take that risk than

 

give something that has systemic effects.

 

DR. ADAMIS: A point well taken. As Dr.

 

D'Amico said, I mean, it is important to take it in

 

the context of the potential benefit. So, the

 

reduction in severe vision loss is greater than 50

 

percent and the severe vision loss we saw as a

 

function of endophthalmitis was 0.1 percent. On

 

balance, at least in this first year, it looks like

 

the benefit outweighs the risk.

 

DR. DUNBAR: Dr. Lehmer?

 

DR. LEHMER: I agree with Dr. Pulido. The

 

data are very impressive. Along the lines of what

 

should be looked at in the future--the PDT impact.

 

176

 

Obviously, we are not able to really assess that

 

based on the numbers, but taking this information

 

forward, seeing what are the clinicians going to do

 

with this data, in other words, who do we apply PDT

 

to, what kind of population--a patient comes in

 

with macular degeneration, do we use Macugen? Do

 

we use PDT? Do we use both?

 

I suppose if patient recruitment were

 

going to start now we would see a much larger

 

percentage of the European community using PDT

 

since it has been approved there and there has been

 

some expanded use of PDT. So, I guess as far as a

 

future analysis--I don't know if that is already

 

under way--I would like to see more data on the

 

impact of PDT.

 

DR. GUYER: I think that is a very

 

important point. One of the things that was

 

important to us when we designed this trial was to

 

try to make it as much of a real-world trial as

 

possible. That is why we allowed photodynamic

 

therapy in it. Showing the data, we can't say a

 

lot about combination use or anything like that,

 

177

 

but I agree with you that certainly future trials

 

will be able to address those issues and it is

 

important.

 

DR. DUNBAR: Mr. Kresel?

 

MR. KRESEL: I guess being the pragmatic

 

industry representative, I will ask the question

 

the way I look at things, which is that we had a

 

lot of discussion about endophthalmitis and I think

 

you gave a really good answer as far as how the

 

patients were treated and how they were followed

 

up. But they were in a clinical trial where, you

 

know, they came back to see the physician at these

 

intervals. So, would you recommend in labeling

 

that kind of a follow-up so that those patients are

 

tracked and, in fact, appropriately diagnosed and

 

treated?

 

DR. ADAMIS: The optimal follow-up I think

 

still remains to be determined. One of the things

 

we have done is we have given grants to specialists

 

who are experts in this area to try to come up with

 

a preferred practice recommendation. The only

 

thing we can say is what we did and what the

 

178

 

results were. I think it is still an open question

 

as to which variables that we changed, and we

 

changed multiple and, as I said, the steroid

 

injections were taking place at the same time in

 

another population--which of those factors is the

 

most important still remains to be determined and I

 

think a lot more work needs to be done in that

 

area.

 

So as regards to what we will recommend,

 

it is still being decided. Until we hear back from

 

the experts we obviously will tell people what we

 

have been doing and the results that were

 

associated with that.

 

DR. GUYER: I also want to comment--many

 

of the retina people in the room know this--but in

 

the last three or four years there has been a

 

tremendous experience in the retinal world with the

 

use of off-label intravitreal steroids because

 

there is such an unmet medical need not only for

 

this disease, macular degeneration, but also for

 

diabetic macular edema. So, I actually think there

 

was a tremendous learning curve for retinal

 

179

 

physicians learning the best way to do intravitreal

 

injections. That occurred. We talked about the

 

protocol amendment and we hope that that had some

 

effect. But I think also equally important may be

 

that the retinal doctors had a very, very good

 

experience of the best way to practice intravitreal

 

injection administration.

 

As Tony mentioned, we did sponsor a

 

roundtable to try to get the thought leaders

 

together on the best way, and Dr. D'Amico was at

 

that and maybe he would like to comment on a few of

 

the findings from that that could help guide us.

 

DR. D'AMICO: Yes, under an educational

 

grant a roundtable was convened to look at the

 

growing use of intravitreal injections in

 

ophthalmic practice, and to try to assemble the

 

best available information on what we know about

 

how to make this procedure as safe as possible. In

 

this roundtable there were experts from the point

 

of view of infectious disease, from the point of

 

view of vitreal-retinal surgeons, people who deal

 

with antibiotic levels within the eye, and also

 

180

 

substantial representatives across industry who

 

have pharmaceuticals that are used by intravitreal

 

injection. While all I can tell you is that an

 

article is in preparation that will be ultimately

 

submitted to peer review literature, we have

 

initial plans to submit that article to the journal

 

Retina. It includes things such as the premise of

 

using povidone-iodine which emerged as an

 

incredibly important central aspect of using a lid

 

speculum. We were finding that, in many casual

 

surveys, people would do injections and allow the

 

lid margins, etc. to contaminate the needle, and

 

probably most importantly, to treat this as a

 

sterile intraocular procedure.

 

I was present. I was asked to be a part

 

of that committee and, if you wish, I have details

 

about who was there, etc. But I feel that this

 

document will be very valuable in helping the

 

evolution of the understanding of intravitreal

 

technique. So, it will become something that I

 

think we can go forward with. We can look at

 

various modifications now to make this safer and

 

181

 

safer.

 

But having participated in both the data

 

safety committee and also this panel, I am quite

 

convinced that the protocol modifications had a

 

very real effect on reducing the incidence of

 

endophthalmitis, and I am confident that incidence

 

can be kept low and probably be even further

 

reduced with appropriate education of both patients

 

and physicians, as well as appropriate training.

 

DR. DUNBAR: Dr. Pulido?

 

DR. PULIDO: Dr. D'Amico, there was a

 

recent article I believe in The American Journal of

 

Ophthalmology. It included people from Baskin

 

Palmer, looking at the incidence of endophthalmitis

 

following intravitreal triamcinolone injections,

 

and the incidence was double that of this, wasn't

 

it?

 

DR. D'AMICO: Correct. You know, it could

 

never have been known when these trials were begun,

 

but intravitreal injections have become quite

 

commonplace in retinal practice now with off-label

 

use of triamcinolone and the incidence which has

 

182

 

been reviewed shows that it is substantially

 

higher. Although I believe that that incidence, in

 

fairness, is decreasing as physicians treat the

 

injection technique with additional seriousness and

 

care. But, actually, a detailed review has been

 

made available to this review committee and showed

 

that the rate of endophthalmitis following

 

intravitreal injection with pegaptanib was well

 

within the range, and at the low end of the range,

 

for intravitreal medication administration across

 

the board.

 

DR. DUNBAR: Ms. Knudson?

 

MS. KNUDSON: The only thing that confuses

 

me a little is that you say that no patients

 

receiving the sham treatment had endophthalmitis.

 

Doesn't it seem that it is the drug then that was

 

causing it?

 

DR. D'AMICO: Well, the sham patients did

 

not receive the penetration of the eye with the

 

needle so that explains why it is that event which,

 

presumably, allows bacteria to gain entry into the

 

eye.

 

183

 

DR. DUNBAR: Recently several of the

 

comments reflected not so much concerns about the

 

statistical significance of the efficacy of the

 

drug but, rather, concerns for the future.

 

Previously Dr. Bull mentioned that the committee

 

can make Phase IV recommendations for plans for the

 

future, for future studies. What is the mechanism

 

for this? And, perhaps this is an appropriate time

 

for the committee to discuss some of these future

 

concerns.

 

DR. BULL: That would fundamentally fall

 

under recommendations for additional studies. If

 

these are data deficiencies that you might see as

 

impacting marketing of the product, it would argue

 

against whether or not you feel the data is

 

sufficient at this point in terms of the efficacy

 

assessment. If these are data needs that need to

 

be obtained in a systematic way, they can certainly

 

not hold up marketing of the product if you feel

 

there is sufficient efficacy in terms of what you

 

have seen.

 

We realize this is an incomplete data set

 

184

 

and I think that that needs to be kept in mind,

 

given the earliness of where we are in this

 

submission. In fact, there are modules in the NDA

 

that have not come in and have not been vetted by

 

the agency yet. So, I have to say that, you know,

 

we haven't seen the data, as has been mentioned, in

 

terms of the re-randomization. There are a number

 

of sort of outstanding assessments here that I

 

think certainly have significant implications for

 

further work. But I think things that need to be

 

looked at systematically certainly have the

 

potential of being addressed in Phase IV.

 

DR. DUNBAR: Dr. Pulido?

 

DR. PULIDO: Just for clarification, that

 

could be a postmarketing surveillance. For

 

instance, study ERG could be postmarketing,

 

following marketing approval surveillance in that

 

regard.

 

DR. BULL: You mean is post-approval?

 

DR. PULIDO: Yes.

 

DR. BULL: Potentially but, again, as I

 

said there is a huge caveat here that we are still

 

185

 

very early in the review of this application and

 

there are a number of other aspects, particularly

 

from chemistry manufacturing issues, that will need

 

to be addressed and other things that will impact

 

the totality of our assessment of the data.

 

DR. DUNBAR: Dr. Chambers and then Mr.

 

Kresel.

 

DR. CHAMBERS: Let me just clarify, the

 

range of different options includes additional

 

Phase III trials, additional Phase IV clinical

 

trials, as well as postmarketing commitments,

 

postmarketing monitoring. There is going to be a

 

certain amount of postmarketing monitoring that

 

automatically goes with any new drug product. But

 

what you are describing would probably more

 

accurately be done as actual controlled clinical

 

trials because you want, obviously, a baseline as

 

well as continued follow-up in order to look for

 

any potential changes. That is probably better

 

done with a control group and making sure you have

 

everybody in your trial.

 

DR. DUNBAR: Mr. Kresel?

 

186

 

MR. KRESEL: I guess my question isn't

 

really a question--well, it is but it is for the

 

rest of the committee because it is not one for me

 

to decide. But if I were in the sponsor's shoes,

 

and I have heard people commenting on how long can

 

we use this drug and what are the consequences, I

 

guess I would like to hear the committee weigh in

 

on how much follow-up post-approval the committee

 

thinks is appropriate, for planning purposes. That

 

is, you know, the sponsor is going to have two

 

years of data pretty soon. How much data does the

 

rest of the committee think is appropriate to

 

continue follow-up?

 

DR. DUNBAR: Dr. Pulido?

 

DR. PULIDO: Trying to answer your

 

question and Dr. Chambers' at the same time, I

 

don't know whether it is necessary to do a

 

randomized, controlled trial for the results of

 

ERG. One possibility is that there hasn't been any

 

change whatsoever so that if you take the patients

 

that already have been in the trial for a year and

 

do ERGs in a small group of them and compare it

 

187

 

even to the fellow eye and there is no difference,

 

well, that tells you volumes. That decreases the

 

chances of having to go ahead and do another

 

randomized, controlled trial and slow the

 

acceptance of this drug into the marketplace.

 

DR. DUNBAR: Dr. Chambers?

 

DR. CHAMBERS: Let me just ask don't you

 

think there is likely to be decreased ERG in

 

patients that had macular degeneration compared to

 

the other eye?

 

DR. DUNBAR: Dr. Pulido?

 

DR. PULIDO: Not necessarily because

 

macular degeneration is such a localized area that

 

is involved that the ERG overall may not be

 

affected. We know that macular disease does not

 

affect a large part of the ERG. So, my only

 

concern, again, is, is this affecting a broader

 

area of the retina than what we are measuring by

 

doing visual acuity measurements? If that is not

 

the case, I don't think we should delay it.

 

DR. CHAMBERS: I don't know that we are

 

talking necessarily about delaying it, but I guess

 

188

 

the question still in my mind is interpretation.

 

If you don't see anything, yes, that is great. If

 

you do see something, is that necessarily the drug

 

product or is that the disease going on? And you

 

don't know the answer to that.

 

DR. PULIDO: Then you would have to do the

 

trial you were considering.

 

DR. DUNBAR: Taking a step back to Mr.

 

Kresel's question, I would like to ask the other

 

committee members if there is any sense among the

 

committee to build a consensus of how long the

 

company should study the drug for the future after

 

this they finish this planned two-year period. Not

 

so much requesting additional data such as the

 

visual field and ERG that Dr. Pulido mentioned, but

 

just to continue the clinical trial, is there any

 

sense among the committee? Dr. Lehmer?

 

DR. LEHMER: In the PDT studies there was

 

a physical endpoint of no leakage. Is there a

 

similar endpoint with regard to this study looking

 

for that type of endpoint or stabilization of

 

vision? I think we have to have some kind of

 

189

 

clinically meaningful endpoint on which to base the

 

answer to that question of how long do we carry the

 

study for and, therefore, how long do clinicians

 

expect to carry on the treatment.

 

DR. GUYER: In the photodynamic studies

 

there was continued leakage. When they decided to

 

retreat they would do a fluorescein angiogram to

 

determine that. But over the course of the year,

 

similar to what we have seen, there was still

 

leakage occurring and that us the disease, and Tony

 

can perhaps give us some hypothesis for why.

 

So, for that reason, I think the two-year

 

data will be very, very important in the sense that

 

we will learn more about two years of therapy

 

versus one year of therapy. Until that data, as we

 

mentioned earlier, I think what is nice about the

 

eye is that you can look in and see the disease and

 

a patient who has significant disease with large,

 

scarred, poor vision obviously wouldn't be

 

necessarily a good candidate to continue treatment.

 

Someone that might not have any leakage, as you

 

say, could be used as a clinical endpoint for

 

190

 

perhaps stopping treatment, and people who are

 

actively bleeding would continue.

 

But it is important to say that really the

 

only recommendation we can make is this clinically

 

important finding is based on one year or 54 weeks

 

of treatment. So, we really can't say anything

 

more and it would be dangerous to try to speculate

 

that less treatment could give the same effects.

 

So, we believe that clinical judgment would be

 

very, very important in determining long-term

 

treatment.

 

DR. DUNBAR: Dr. Pulido?

 

DR. PULIDO: The other thing that I think

 

is important is the fact that even with one-year

 

follow-up--what was the mortality rate in this

 

group of patients? Wasn't it 10 percent or

 

something?

 

DR. ADAMIS: It was two percent in treated

 

and sham alike.

 

DR. PULIDO: Right, so I mean you are

 

already getting to a point where there is a certain

 

mortality in these elderly patients. To continue

 

191

 

it more than two years, I think you are going to

 

find a higher mortality rate and I don't know

 

whether we are going to find more than what we are

 

already finding.

 

DR. DUNBAR: Is there any additional

 

discussion at this time? If not, at this point

 

let's break for lunch and we will reconvene at 1:00

 

p.m. for the public discussion.

 

[Whereupon, at 11:45 a.m., the proceedings

 

were adjourned for lunch, to resume at 1:00 p.m.]

 

192

 

A F T E R N O N P R O C E E D I N G S

 

Open Public Hearing

 

DR. DUNBAR: We are beginning the

 

afternoon session of the Dermatologic and

 

Ophthalmic Drugs Advisory Committee on Macugen with

 

an open public hearing.

 

Both the Food and Drug Administration, the

 

FDA, 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, 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 products

 

in the pharmaceutical category under discussion at

 

today's meeting. For example, this financial

 

information may include the sponsor's payment of

 

your travel, lodging, or other expenses in

 

193

 

connection with your attendance at the meeting.

 

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.

 

At this point of the open public hearing,

 

I will ask speaker number one to please come to the

 

podium. Each speaker will have seven minutes to

 

present.

 

MR. GARRETT: Hi. My name is Dan Garrett

 

and I am with Prevent Blindness America. My

 

organization paid for my own travel to come here

 

today and I personally do not have a financial

 

relationship with any of the companies pertaining

 

to this drug.

 

I mentioned I am with Prevent Blindness

 

America. We are the second oldest voluntary health

 

agency in the country. We represent organizations

 

throughout the country that primarily focus their

 

efforts on screening, training, advocating,

 

194

 

researching and educating people on the importance

 

of good vision care. We also advocate for

 

increased research funding and increased funding to

 

the Centers for Disease Control in Washington, and

 

we try to impact public policy as it relates to

 

saving sight and vision loss.

 

The reason I am here today is not to

 

endorse this product but to encourage the committee

 

to make the right decision as it relates to the

 

science behind this drug. It might suggest that

 

this could prevent further vision loss for people

 

with AMD. That is why I am here today. My

 

organization does not endorse the product of

 

discussion today.

 

A few thoughts and figures, and I wasn't

 

here earlier today so forgive me if these are

 

repetitive. It is important to point out that

 

nearly 1.7 million Americans aged 40 and older have

 

AMD, and if nothing is done by the year 2030 the

 

number of blind and visually impaired could

 

possibly double. So, we are talking about a fairly

 

significant population. It is very important that

 

195

 

this committee consider this drug because it has

 

the potential to potentially stop vision loss.

 

Unfortunately, there is not a miracle drug out

 

there yet that prevents AMD but, hopefully, with

 

all the science and research that is going on that

 

will be in the near future for us.

 

Another interesting statistic, and this

 

could particularly hold for people with AMD because

 

they are the ones that have most low vision, vision

 

impairment is the cause of 18 percent of hip

 

fractures, and most people that have AMD are living

 

on their own and they have lost their central

 

vision so it is very difficult for them to navigate

 

their way around their home. If only one in five

 

of those hip fractures were prevented, more than

 

440 million dollars could be saved annually so that

 

is significant. So, any type of AMD drug that

 

could prevent further vision loss is certainly a

 

welcomed addition to the marketplace for patients.

 

My organization, again, advocates

 

advancements in treatment of AMD, and I just want

 

to say to the committee that I am sure you will

 

196

 

make the right decision on behalf of all the older

 

Americans in this country for the people that have

 

AMD. Anything that can prevent further vision loss

 

should be welcomed. That is all I have to say.

 

Thank you.

 

DR. DUNBAR: Thank you. At this point I

 

will ask speaker number two to come to the podium.

 

MS. HOFSTADTER: Good afternoon. I am

 

Ellen and I am 81 years old. I do not have any

 

financial ties with the drug company except my stay

 

in the hotel and my travel.

 

I was diagnosed with AMD two years ago. I

 

belong to an HMO. The HMO doctors checked me and

 

told me "you can go home; there's nothing we can do

 

for you." But I didn't take no for an answer. I

 

called the Jewish Eye Clinic and asked if there was

 

a doctor who could see me. The girl says, yes, and

 

in two weeks I have an appointment. I got Dr.

 

Schwartz. I had an eye test, an angiogram and he

 

looked my eye over and he said, "don't drive, but

 

do not sell your car because we might can help

 

you."

 

197

 

So, I took some lasers, some Visudyne in

 

my left eye but it didn't help. So, my left eye is

 

legally blind. Then I was approached by Dr.

 

Gonzalez who asked me if I would like to step into

 

a clinical trial with Macugen shots. Well, it was

 

very heavy for me because when I was a young girl I

 

was sent to Auschwitz and I was experimented on by

 

the infamous Dr. Mengele. So, I had really a

 

choice to make.

 

I didn't think long about it and I thought

 

I want my sight. So, I told them I would. So, I

 

got into the clinical trial and I got a Macugen

 

shot in my right eye. It sounds very scary but

 

really 20 minutes of discomfort is a small price to

 

pay. After the third shot I gained my sight back

 

to 20/20 and could read seven lines below. I had

 

altogether 12 shots and three weeks ago I had my

 

lost one and my sight is 20/20 in one eye.

 

And I really want to thank the researchers

 

who worked so hard to find a drug like Macugen to

 

help us for this dreadful disease. Thank you.

 

DR. DUNBAR: Thank you. Next I will ask

 

198

 

speaker number three to come to the podium.

 

MR. STEVENSON: My name is Nick Stevenson.

 

I am the president of the Association for Macular

 

Diseases. It is the only national support group

 

that is solely concerned with both the practical

 

and the emotional problems confronted by

 

individuals and families endeavoring to cope with

 

our particular type of eye disorder. To do that,

 

we publish a newsletter which advises our members

 

what is going on in the world of research, what is

 

not. There is an increasing number of scams and

 

frauds which are proliferating now not only in

 

numbers but in funding as well, and we maintain a

 

members hotline where members can always call in

 

and we can act as a link between them, their

 

problems and the problems that they may face.

 

I, myself, have been legally blind from

 

the wet type of macular degeneration for 26 years.

 

I have no financial interest in this pharmaceutical

 

company or actually any, except that they did pay

 

my travel and expenses to come down here. But what

 

I would like very much to emphasize for all of you

 

199

 

is something that many of you, I can understand,

 

have already experienced, how difficult and

 

understandably difficult it is for a fully-sighted

 

person to fully appreciate the enormous subtraction

 

from life that loss of vision represents, for some

 

far more than for others but, nonetheless, it is

 

not something that any of us foresaw in earlier

 

years of our lives. We may have thought of

 

disasters overtaking us, such as being struck by an

 

automobile or some disease attacking us in a way

 

that we found ourselves to be vulnerable. But the

 

loss of vision is something that few of us have

 

ever contemplated. We felt that there was a

 

warranty issued on our eyes and we had the full use

 

of our eyes for as long as we needed them. Then we

 

find that we don't and an entirely different set of

 

circumstances appears.

 

Now, it must be admitted that macular

 

degeneration varies widely in the degree of

 

severity with which it affects individuals. But

 

for those with more severe type, such as this drug

 

addresses, they have the problem of not recognizing

 

200

 

the faces of their friends, or their enemies if

 

they have them. Also, they are not able to drive

 

in a society where an automobile is as automatic a

 

feature as a horse once was out West, or even

 

almost an appendage of oneself--the automobile--is

 

taken away.

 

In addition to that, the inability to read

 

to varying degree, whatever it might be, is also a

 

very serious detraction from quality of life. That

 

blue sign over there; it is that entrance right

 

there past the blue sign--of course, you can see

 

it. And does this bus go to Amherst? Well, the

 

drive is too busy to answer you so he nods and you

 

don't see him nodding--these are not major events

 

but they have a cumulative effect and what is very

 

difficult for a great many of us to understand

 

fully, because we don't choose to, is that macular

 

degeneration is a progressive disease. As the

 

years go by; the eyes do get worse whether we have

 

the dry type or whether we have the wet type.

 

So, it seems to me high time that some

 

action was taken to try to avert the further

 

201

 

incidence of macular degeneration in its various

 

forms for the people who follow behind us. It has

 

been said of older people that, as they think of

 

their lives, the days grow longer and the years

 

grow shorter. So, as the years grow shorter, all

 

of us hope that somewhere--like Dr. Jonas Salk

 

finding the cure of polio back in 1954--something

 

may appear that will give us some surcease from the

 

anxiety, and the apprehensions, and the limitations

 

of macular degeneration. Thank you.

 

DR. DUNBAR: Thank you. Now I will ask

 

speaker number four to come to the podium.

 

MR. AUGUSTO: Good afternoon. I am Carl

 

Augusto, president and CEO of the American

 

Foundation for the Blind, an organization that is

 

dedicated to expanding the rights and opportunities

 

of people who are bind or visually impaired in this

 

country. Like Helen Keller before me who devoted

 

44 years of her life to the American Foundation for

 

the Blind and its causes, I am always grateful to

 

speak to governmental officials, corporate America

 

and the general public on how we can improve the

 

202

 

lives of people who are blind or visually impaired.

 

In my 30-plus years working as a

 

rehabilitation counselor and as an administrator in

 

agencies serving the blind and visually impaired, I

 

have seen first-hand the many difficulties faced by

 

those who are losing their vision as a result of

 

AMD, age-related macular degeneration. After

 

living most of one's life, relying heavily on the

 

sense of sight, not seeing well enough or seeing at

 

all can certainly turn the world upside down for

 

those people and their families. Add to that other

 

physical ailments, physical disabilities, personal

 

and social hardships that older people, many of

 

them, experience the emotional and the functional

 

adjustment to vision loss is very, very difficult.

 

Ordinary daily activities become

 

challenging, if not impossible. If you can imagine

 

not having the opportunity or not having the

 

ability to read the morning newspaper, to drive to

 

supermarket to get your groceries, to do your

 

personal business, to read your personal mail, to

 

cook for yourself--this is what is happening with

 

203

 

so many people losing their vision in this country.

 

Moreover, it is difficult to adjust emotionally and

 

functionally to a certain level of visual loss if

 

that vision worsens next month.

 

One of the first clients that I had as a

 

rehabilitation counselor was a gentleman suffering

 

from age-related macular degeneration. He was

 

about 50 years old and his deterioration rate was

 

steady over the course of time, and he was really

 

overwhelmed by this. His name was Jack. Jack had

 

lost confidence in his capabilities. He felt that

 

he couldn't do his job any longer. And, one of the

 

things he said to me was, "just when I think I'm

 

beginning to adjust, I lose more vision and the

 

despair sets in again."

 

Well, his visual loss forced him to retire

 

from his job long before he should have. It was a

 

financial hardship to his family. He was staring

 

at the walls every day and not feeling productive

 

at home. It took an emotional toll on the family.

 

His wife couldn't handle it any longer and she left

 

and now he was on my doorstep, wanting answers to

 

204

 

how to live independently.

 

I remember thinking that, gee, if I had

 

seen him a little earlier, or if the progression of

 

his sight loss was not as significant I might have

 

been able to help him realize that he could do his

 

job still using alternate techniques or technology.

 

But he lost his vision much too quickly and he did

 

give up.

 

Now, my blindness is caused by a recessive

 

gene disorder and it started when I was very young.

 

When I was eight years old I started losing my

 

vision and my loss was very gradual over the course

 

of time. I became totally blind at about age 45.

 

Some days I think I haven't reached 45 yet but that

 

is just a couple of years ago. But that gave me an

 

opportunity to learn the skills that I needed to

 

function independently at home and on the job. I

 

had an opportunity to tackle the emotional hurdles

 

that inevitably arise with severe vision loss, and

 

I truly believe I live a life that is as normal and

 

satisfying as anyone's.

 

Now, AMD is the leading cause of severe

 

205

 

visual loss in our country, and this visually

 

impaired population will continue to increase as

 

the baby-boomers reach old age. Simply stated, we

 

are outliving our eyes and delaying the effects of

 

AMD or stopping the effects of AMD would give

 

millions of people more time to adjust emotionally

 

and functionally, to locate rehabilitation

 

facilities, and to develop the skills that are so

 

critical in helping them to function independently.

 

If we can do this, any kind of slowing of the

 

deterioration would be a blessing.

 

There are services for people who are

 

blind or visually impaired. Low vision services

 

that are delivered by specially trained eye care

 

professionals enhance the remaining usefulness of

 

your vision when you do have vision remaining.

 

Other rehabilitation services are available from

 

private and public agencies throughout the country

 

to help you with personal management skills and

 

also vocational skills. And, assisted technology

 

is revolutionizing the way blind and visually

 

impaired people function.

 

206

 

However, there are two problems. Many

 

people with age-related macular degeneration and

 

other visual loss don't even have a clue that these

 

programs are available and they may not be in their

 

own communities. Secondly, we don't have the

 

funding in this country, federal or otherwise, to

 

support sufficient services to meet the growing

 

need for services for the increasing population of

 

blind and visually impaired people. So, anything

 

we can do to reduce the numbers of this population

 

would be helpful in that regard.

 

In closing, blind and visually impaired

 

people can live and work with dignity and success

 

alongside their sighted peers. People can adjust

 

and learn new skills and also to live

 

independently. But many of them need time to

 

develop. Many of them are not adjusting when their

 

vision continues to deteriorate, and without a

 

chance to learn to cope with vision loss gradually,

 

I am afraid that too many people will be like Jack

 

and will give up on themselves before they realize

 

that there is help out there that could help them.

 

207

 

Thank you.

 

DR. DUNBAR: Thank you. Now I want to

 

request that speaker number five come to the

 

podium.

 

DR. ROSENTHAL: I am Bruce Rosenthal,

 

Chief of Low Vision Programs at Lighthouse

 

International, New York City and Mount Sinai

 

Hospital. My organization paid my expenses.

 

However, in the past I have had an unrestricted

 

educational grant from Novartis for a booklet on

 

vision function.

 

Over 75 percent of the visually impaired

 

patients I have examined for the past 30 years have

 

a diagnosis of age-related macular degeneration. I

 

have been witness to how the devastating effects

 

that progression severe vision loss, especially

 

from the neovascular form of the disease, impact on

 

an individual's day-to-day activities. I have seen

 

how severe vision loss affects an individual's

 

quality of life, impacts on their independence,

 

lowers their self-esteem, and results in

 

depression. In fact, clinical studies have shown

 

208

 

that over 57 percent of people with retinal disease

 

have depression.

 

As a clinician, I am very concerned with

 

retaining visual function. Neovascular AMD has the

 

effect of destroying vital components of visual

 

function. We are all familiar with visual acuity,

 

as well as the importance of preserving it. But

 

other vital components of vision are also

 

irreparably destroyed by the effects of AMD. They

 

include contrast sensitivity, and in lay terms that

 

is how much a pattern must vary in contrast to be

 

seen, and has become increasingly recognized as an

 

important factor in influencing the quality of

 

life. We are also interested in retaining usable

 

visual field, color perception and stereo-acuity,

 

just to name a few.

 

The medical advances, as we all know, that

 

have taken place in the past 30 years have been few

 

and far between. However, thermal laser as well as

 

PDT have really helped to slow the progression and

 

maintain visual function, and one example that I

 

will give to you as a clinician is that the early

 

209

 

patients I was seeing with low vision would go from

 

20/800 down to light perception. My patients now

 

usually fall in the end stage between 20/200 and

 

20/400. Yet, serious vision loss continues despite

 

these interventions, as we know.

 

As Carl Augusto mentioned, we seem to have

 

an impact, however, with vision rehabilitation. As

 

a low vision clinician, I have seen that

 

individuals with AMD who have access to the latest

 

treatments benefit more from vision rehabilitation

 

services as well. These individuals have a greater

 

success rate in the use of low vision optical

 

prescriptive devices, absorptive lenses, as well as

 

high tech and electronic aids. These people can

 

continue to be employed, travel independently,

 

manage their own affairs, maintain their own

 

residence and perhaps even drive. Again, I

 

recommend that you consider the treatment that will

 

help preserve visual function and its benefits to

 

society.

 

DR. DUNBAR: Thank you. This concludes

 

the five members of the public that have registered

 

210

 

to speak at the open public hearing. However,

 

there are some additional members of the public

 

that have approached us requesting to speak and,

 

time permitting, they will be allowed to come to

 

the podium and give two-minute presentations. So,

 

I will ask if there are any other members of the

 

general public that wish to come forward at this

 

time. Thank you. We have someone coming forward.

 

DR. LISS: I am Bob Liss. I am an

 

ophthalmologist in practice, retinal diseases, in

 

Baltimore. I congratulate the sponsors and

 

certainly hope that this is approved.

 

I did want to comment that I am concerned

 

about the problem of endophthalmitis in terms of

 

the fact that the drug is very broadly applicable

 

drug that covers all subtypes of choroidal

 

neovascularization so it will be used much more

 

widely, and the people using it in the community,

 

whether they are retinal specialists or

 

ophthalmologists who are not retinal specialists,

 

because of the more broad range of the indications,

 

are selected different than the investigators. The

 

211

 

investigators, as much as the sponsors, have wanted

 

to have a real-world test of the trials. The

 

investigators are trained extensively and

 

controlled much better than the outside area. So,

 

I do think that control of complications,

 

particularly endophthalmitis, is important.

 

The second thing is a comment about the

 

quality of life. There was just a discussion about

 

contrast sensitivity and visual fields, along with

 

the early discussion about ERG and I think these

 

types of things should be included in future

 

evaluations. Thank you.

 

DR. DUNBAR: Thank you. Are there any

 

additional members of the public that wish to come

 

forward?

 

[No response]

 

Committee Discussion

 

At this point then we will open up for

 

general discussion among the committee members,

 

taking into account the presentations we have heard

 

from the public. Are there any comments at this

 

time? Dr. Lehmer?

 

212

 

DR. LEHMER: I was going to mention

 

earlier, and I was glad one of the public speakers,

 

Mr. Rosenthal, mentioned about contrast

 

sensitivity. A lot of my patients who have the

 

same level of visual acuity function very

 

differently on similar behavioral tasks in the

 

office and when we test their contrast sensitivity

 

it varies greatly. So, it seems like I would

 

second the motion of including that as a measure.

 

DR. DUNBAR: Dr. Chambers?

 

DR. CHAMBERS: The agency certainly agrees

 

they would like to be able to use contrast

 

sensitivity as a measure and certainly believe it

 

is a measure of visual function. The difficulty

 

with using contrast sensitivity in an assay is

 

figuring out which contrast sensitivity is the most

 

appropriate, and if you find a difference in one

 

frequency versus a different frequency what does it

 

mean? If you have any guidance on which

 

frequencies are more important than other

 

frequencies, we would love to hear those comments.

 

DR. DUNBAR: I am interested in the

 

213

 

comments about off-label use of the drug. I think

 

this is insightful because once the drug is

 

available to doctors--for example, would a doctor

 

perhaps instill it into the anterior chamber for a

 

patient with rubeosis? And, this is a conceivable

 

possibility. Do we know anything about endothelial

 

cell toxicity? This is a question actually for the

 

sponsor.

 

DR. ADAMIS: The question is an important

 

one. We did not look specifically at endothelial

 

cell counts. We didn't do any specular microscopy.

 

All we can report is that over the 54-week period

 

there did not appear to be an increased incidence

 

of corneal edema.

 

DR. DUNBAR: Is there any preclinical data

 

that might guide us about this question?

 

DR. ADAMIS: In the preclinical animal

 

studies there was no finding of corneal edema as a

 

function of the use, but in the animals as well, to

 

my knowledge, specular microscopy was not done.

 

DR. GUYER: Just as far as a comment on

 

other uses, the sponsor right now is presently

 

214

 

looking at other important diseases in trials. We

 

finished our Phase II program of diabetic macular

 

edema and actually, hopefully in the fall, we will

 

be talking with the agency about putting together a

 

Phase III program. As you mentioned, there are a

 

lot of conditions in the eye but today, you know,

 

we are specifically talking about the indication

 

for age-related macular degeneration.

 

DR. DUNBAR: As a pediatric

 

ophthalmologist, I am interested in retinopathy

 

prematurity. Do you have any comments about its

 

use in that situation?

 

DR. ADAMIS: Theoretically it is a drug

 

that I think may prove useful in retinopathy

 

prematurity but the data that I showed you is that,

 

you know, VEGF is required for normal vessel

 

formation and the conundrum has always been, well,

 

how can you block the bad vessels and leave the

 

good vessels alone? But it look like by targeting

 

165 we may be able to do that. So, that is

 

something we would consider doing in the context,

 

obviously at some point in the future, as a

 

215

 

clinical trial. We wouldn't recommend off-label

 

use at this point.

 

DR. GUYER: Also, in addition to

 

retinopathy prematurity to look at in the future,

 

and we mentioned the diabetic program also, we are

 

also presently in a Phase II program for retinal

 

vein occlusions and the macular edema that comes

 

from that. In fact, if we could just go to E-158

 

for a second, it just lists a couple of the trials,

 

if anyone is interested.

 

[Slide]

 

In addition to the diabetic program, we

 

presently are studying, as I said, retinal vein and

 

also we have a small program with Emily Chiu, of

 

the National Eye Institute, on von Hippel Lindau

 

tumors because of the increased permeability of

 

those lesions. We are considering, but have not

 

started yet, trials for pathological myopia and

 

histoplasmosis where, again, choroidal

 

neovascularization is associated; sickle cell

 

retinopathy; iris neovascularization, as was

 

earlier mentioned; and proliferative diabetic

 

216

 

retinopathy. Those are presently under

 

consideration.

 

DR. DUNBAR: Are there additional comments

 

from the committee at this time, especially

 

pertaining to the public hearing?

 

[No response]

 

Now I would like to shift our emphasis

 

once again to the general discussion that we began

 

this morning and see if there are any other

 

comments in general from the committee before we

 

move on to the questions. I will poll the

 

committee members one by one.

 

Dr. Chinchilli, do you have any additional

 

comments?

 

DR. CHINCHILLI: No, I do not.

 

DR. DUNBAR: Ms. Knudson?

 

MS. KNUDSON: No, I do not.

 

DR. DUNBAR: Dr. Steidl?

 

DR. STEIDL: No, I don't.

 

DR. DUNBAR: Dr. Pulido?

 

DR. PULIDO: No, I do not.

 

DR. DUNBAR: Dr. Lehmer?

 

217

 

DR. LEHMER: No, I don't.

 

DR. DUNBAR: Dr. Gates?

 

DR. GATES: None.

 

DR. DUNBAR: I have no additional

 

comments. Dr. Miller?

 

DR. MILLER: No.

 

DR. DUNBAR: And Mr. Kresel?

 

MR. KRESEL: No, I do not.

 

Questions

 

DR. DUNBAR: At this point then let's move

 

on to a discussion of the individual questions

 

posed by the FDA. I will read the individual

 

question and open up the question for general

 

disease and at the end of the discussion poll each

 

member.

 

The first question reads, has sufficient

 

data been submitted to evaluate the efficacy and

 

safety profile of pegaptanib sodium? Excuse me, I

 

was operating from an older list.

 

Back to question number one, based on the

 

inclusion/exclusion criteria, are there patients

 

excluded from the studies that you believe need to

 

218

 

be studied? Is there any general discussion about

 

the inclusion and exclusion criteria? I am going

 

to go ahead an poll each member. Dr. Chinchilli?

 

DR. CHINCHILLI: No, I don't have any

 

comments.

 

DR. DUNBAR: Ms. Knudson?

 

MS. KNUDSON: No, I don't have any

 

additional comments.

 

DR. DUNBAR: Dr. Steidl?

 

DR. CHAMBERS: Can I interrupt? Besides

 

saying you don't have any comments, if you think it

 

was appropriate--it is at least somebody saying you

 

think they were appropriate as opposed to just no

 

comments. Thank you.

 

DR. DUNBAR: Let's start back again with

 

Dr. Chinchilli.

 

DR. CHINCHILLI: Well, I am not that

 

familiar with ophthalmological clinical trials, but

 

the criteria seem appropriate to me.

 

DR. DUNBAR: And Ms. Knudson?

 

MS. KNUDSON: I think the criteria are

 

appropriate and in terms of sufficient data, my

 

219

 

only concern is what we have expressed before,

 

long-term use.

 

DR. DUNBAR: Dr. Steidl?

 

DR. STEIDL: I don't believe that there

 

were patients excluded that need to be studied.

 

DR. DUNBAR: Dr. Pulido?

 

DR. PULIDO: I agree with Dr. Chinchilli

 

and the other members so far.

 

DR. DUNBAR: Dr. Lehmer?

 

DR. LEHMER: I agree that the criteria

 

seem appropriate.

 

DR. DUNBAR: Dr. Gates?

 

DR. GATES: I am satisfied with the

 

inclusion/exclusion criteria.

 

DR. DUNBAR: I concur with the rest of the

 

committee. Dr. Miller?

 

DR. MILLER: I concur.

 

DR. DUNBAR: And Mr. Kresel?

 

MR. KRESEL: I agree with what the rest of

 

the committee has said.

 

DR. DUNBAR: We will move to question

 

number two, visual acuity measurements were

 

220

 

conducted using the ETDRS scale placed at 2 meters

 

from the patient. The validity of the ETDRS scale

 

was established based on readings at 4 meters. Are

 

the visual acuity findings sufficiently robust to

 

overcome the potential bias introduced by visual

 

acuity measurements at 2 meters? Dr. Chinchilli?

 

DR. CHINCHILLI: We haven't discussed this

 

although it was mentioned by the agency. You know,

 

the fact that there is a control group, the sham

 

group, and that you still see differences is

 

encouraging. The question is whether or not there

 

is some sort of interaction. I mean, would the

 

sham group not have a bias when it is done from 2

 

meters whereas the dosed groups would? You know, I

 

don't know if there is any logical explanation for

 

something hypothetical like that happening. It

 

doesn't seem like a major issue but I would like to

 

hear the ophthalmologists talk about this issue.

 

DR. DUNBAR: Then I will open this up for

 

general discussion before polling each individual

 

committee member. Dr. Lehmer?

 

DR. LEHMER: I was just going to say I

 

221

 

wanted to hear what the statisticians had to say

 

because when we are talking about robustness of

 

data, you know, I wouldn't know where to draw the

 

line on are these numbers robust enough to overcome

 

that difference. But I hear what you are saying,

 

that this is a comparison between groups that were

 

tested under the same conditions so my assumption

 

would be that the relative difference would still

 

hold up whether it is 2 meters or 4 meters.

 

DR. DUNBAR: Dr. Chambers?

 

DR. CHAMBERS: I will just clarify a

 

little bit. There are some differences in other

 

areas such as adverse events which might lead

 

someone to recognize which group they were in even

 

if they were not able to tell from the actual

 

procedures, such as some of the floaters, such as

 

some of the other many adverse reactions which may

 

lead them to, either appropriately or

 

inappropriately, believe they were in a group. The

 

concern is that there may be potential unmasking

 

because of some of the adverse events that then may

 

lead to differences, and the issue that there is

 

222

 

more variability with measurements at 2 meters

 

versus 4 meters, although we don't have good

 

quantitation on what that is.

 

DR. DUNBAR: Dr. Pulido?

 

DR. PULIDO: So, Dr. Chambers, is this a

 

possible way of getting around this problem? I

 

feel the data is good enough right now at 2 meters.

 

Because there is a concern, could future studies be

 

requested to be at 4 meters from the start for the

 

small chance that there may be a problem?

 

DR. CHAMBERS: It is the agency's

 

recommendation that they be at 4 meters to avoid

 

the issue even coming up. Were we talking about a

 

single letter we probably wouldn't be asking this

 

question either. We would say, well, that is

 

definitely within what the variation is. You may

 

choose to believe, well, it takes 16 meters before

 

you even get one line; this is a three-line change

 

so we think there is enough robustness in the

 

findings and robustness in differences in visual

 

acuities that, while we would have not like to have

 

had it, it is still okay. Or, you may say there

 

223

 

just is no way to go and tell and the agency needs

 

to deal with it as best they can.

 

DR. DUNBAR: Dr. Pulido?

 

DR. PULIDO: Though it would have been

 

nice if it had been done at 4 meters, there appears

 

to be enough robustness of the data that I accept

 

it at 2 meters. Is that a good paraphrase of the

 

way you had said it?

 

DR. CHAMBERS: I did not want to put words

 

in anyone's mouth. I was trying to put out

 

examples of the type of information we are looking

 

for in comments.

 

DR. PULIDO: That would have been the way

 

I would have said it without you having said it.

 

[Laughter]

 

DR. DUNBAR: I have a question for the

 

agency. Was this an agreed upon aspect of the

 

protocol prior to commencing the clinical trials,

 

or was this a point that came up in the analysis

 

later on?

 

DR. CHAMBERS: The agency, having had the

 

ETDRS done under an IND, is fully aware of how the

 

224

 

protocols were written for ETDRS and has always

 

assumed that if someone wrote ETDRS that they meant

 

that they would do visual acuities at 4 meters. We

 

have come to find out since that time that that is

 

not the interpretation necessarily in the whole

 

community and so there were clinical trials that

 

were started using the charts but moving them to

 

different distances and people continued to call it

 

ETDRS even though it does not meet the technical

 

protocol of ETDRS. In this particular case we were

 

aware of the difference after the trials had

 

started. To the extent we were aware of them

 

before the trial start, to the extent that we were

 

aware of them during the trials, we have made those

 

comments but in some cases we are aware that there

 

were trials that started before we were able to

 

comment on it. Then you would be caught with the

 

equal question of do you change the protocol in

 

midstream or do you run the protocol the way it was

 

started, even if you would have preferred to do it

 

a different way?

 

I will let the sponsor comment on their

 

225

 

own but it is my understanding the choice--and we

 

do fully understand it--is to continue the

 

protocol, at the point that you recognize there is

 

a difference, the way it was written so that you

 

don't raise further questions about, okay, you have

 

changed the protocol. What would have happened had

 

you not changed the protocol? So, we are left with

 

the data that we have. We obviously don't

 

encourage it in the future but this is what we

 

have.

 

DR. DUNBAR: I have a question for the

 

sponsor. Was every center done at 2 meters? Were

 

they all uniform throughout the protocols?

 

DR. GUYER: Could I have slide 14 up,

 

please?

 

[Slide]

 

First, yes, they were all standardized. I

 

think Dr. Chambers summarized very nicely in the

 

morning the difficulties with 2 meters versus 4

 

meters. When we started the trial our thought

 

process was, first, that historically other trials

 

were done at 2 meters, most of the other trials

 

226

 

were for this condition. Part of the reason was

 

that in order to be able to read all of the letters

 

on the chart, some patients would not be able to do

 

that at 4 meters. So, our thought was we could get

 

more patients to see at the baseline visions and at

 

week 54 on the chart and not have to move up to 1

 

meter.

 

But certainly the FDA has presented very,

 

very good information why 4 meters should be

 

considered as well. There is no perfect testing

 

distance. I think Dr. Chambers also, on his slide,

 

said it very well, that the key factor is if

 

masking is good and if you have some kind of rigid

 

way of making sure that the patient didn't lean or

 

move, then 2 meters is certainly a good testing

 

parameter. The real potential biases at 2 meters

 

have to do with two things. One is accommodation

 

which, obviously, in this population because of

 

presbyopia is not an issue. The second is the

 

leaning that Dr. Chambers mentioned.

 

Now, we have randomization which certainly

 

helps. So, we would hope that good randomization

 

227

 

and masking should be equal between sides. But we

 

also have some very important quality control

 

information.

 

[Slide]

 

We had very vigorous training and

 

monitoring of the visual acuity examiners before

 

the trial and during the trial. In fact, we had

 

over 450 audits performed in all of the centers

 

throughout the world. And, one of the questions

 

that was looked at was, was proper patient

 

positioning, such as leaning, prevented by the

 

acuity examiners? You can see that in these 469

 

audits, 98.3 percent of the examiners did use

 

proper patient positioning, which comforts us that

 

at least based upon this quality control we don't

 

believe that the patients were leaning forward.

 

We also have good evidence of proper

 

masking. All groups, the active groups as well as

 

the shams, all got 8.5 of the 9 injections. So,

 

that suggests that masking was good. Similarly for

 

discontinuation rates and reasons, which you can

 

see in the FDA briefing book.

 

228

 

[Slide]

 

Actually, when we did a trial for macular

 

degeneration a number of years ago we devised this

 

measuring stick which also must be used at every

 

examination. Here you can see a visual acuity

 

examiner to actually remind the visual acuity

 

examiner always to be sure that the patient is at

 

the right distance and that the patient doesn't

 

lean forward. This, I think combined with the

 

quality control, helps us.

 

Also, in Dr. Chambers' questions about

 

masking and floaters, which is a very good

 

question, we actually have looked to try to give us

 

some comfort that there was no difference in the

 

responder rate of patients who had floaters and

 

didn't have floaters.

 

[Slide]

 

This shows that whether the patients had

 

floaters or didn't have floaters we see an active

 

treatment effect for both. So, we tried to look at

 

the data from as many possibilities of potential

 

unmasking and did not see anything. So, we have

 

229

 

some comfort I think by the quality control and by

 

the good masking in the trial that 2 meters was

 

probably not an issue. But we certainly share with

 

the agency that in future trials 4 meters are

 

preferred. We wish more study centers, as Dr.

 

Chambers mentioned, had 4-meter testing which has

 

also been part of our thought process, that it is

 

difficult to get 117 centers with rooms that big.

 

But we are working in other trials to do 4-meter

 

testing after these discussions.

 

DR. DUNBAR: Thank you. Are there any

 

other general comments for discussion before

 

individual polling of the committee members? If

 

not, I will ask each committee member to answer the

 

question are the visual acuity findings

 

sufficiently robust to overcome the potential bias

 

introduced by visual acuity measurements at 2

 

meters? Dr. Chinchilli?

 

DR. CHINCHILLI: Yes, I believe the data

 

are reliable even though the measurements were

 

taken at 2 meters. I was comforted by some of

 

these quality control issues that the sponsor

 

230

 

addressed and was prepared to address.

 

DR. DUNBAR: Ms. Knudson?

 

MS. KNUDSON: I will echo what Dr.

 

Chinchilli said.

 

DR. DUNBAR: Dr. Steidl?

 

DR. STEIDL: Yes, given the significance,

 

the audits presented and randomization, I am

 

comfortable with them.

 

DR. DUNBAR: Dr. Pulido?

 

DR. PULIDO: I am comfortable with the

 

robustness of the data.

 

DR. DUNBAR: Dr. Lehmer?

 

DR. LEHMER: I am comfortable with the

 

robustness of the data.

 

DR. DUNBAR: Dr. Gates?

 

DR. GATES: I am also satisfied. In

 

examining patients on a day-in and day-out basis I

 

always ask them to lean forward for these different

 

tasks, and with this randomization not picking on

 

any particular segment of the patient population, I

 

know some will and some won't even if they are

 

physically able or not able. So, with this

 

231

 

randomization I am very satisfied with the

 

robustness.

 

DR. DUNBAR: I concur with the other

 

comments to this point. Dr. Miller?

 

DR. MILLER: Based on what Dr. Chambers

 

and also the sponsor has had to say, I concur.

 

DR. DUNBAR: And Mr. Kresel?

 

MR. KRESEL: I agree with the rest of the

 

committee.

 

DR. DUNBAR: We move to question number

 

three, has sufficient data been submitted to

 

evaluate the efficacy and safety profile of

 

pegaptanib sodium for the treatment of the