1
DEPARTMENT OF HEALTH AND HUMAN
SERVICES
FOOD AND DRUG
ADMINISTRATION
FDA SCIENCE BOARD ADVISORY
COMMITTEE
Thursday, April 22, 2004
8:00 a.m.
Advisors and Consultants Staff
Conference Room
5630 Fishers Lane
Rockville, Maryland
2
PARTICIPANTS
Kenneth I. Shine, M.D., Chair
Jan. N. Johannessen, Ph.D., Executive
Secretary
BOARD MEMBERS:
Gail H. Cassell, Ph.D.
Josephine Grima, Ph.D., Consumer
Representative
Susan Harlander, Ph.D.
Cato T. Laurencin, M.D., Ph.D.
Cecil B. Pickett, Ph.D.
F. Xavier Pi-Sunyer, M.D., M.P.H.
Jose C. Principe, Ph.D.
Jim E. Riviere, D.V.M., Ph.D.
Allen D. Roses, M.D.
Katherine M.J. Swanson, Ph.D.
John L. Thomas, Ph.D.
OFFICE OF THE COMMISSIONER:
Lester M. Crawford, D.V.M., Ph.D.,
Norris E. Alderson, Ph.D.
FDA:
Daniel A. Casciano, Ph.D.
David W. Feigel, Jr., M.D., M.P.H.
Kathy Carbone, M.D.
John Marzilli
Robert E. Brackett, Ph.D.
Steven Galson, M.D., M.P.H.
3
C O N T E N T S
Call to Order, Kenneth I. Shine, M.D.,
Chair, 5
Welcome and Opening Remarks,
Lester M. Crawford, D.V.M., Ph.D.,
Acting
Commissioner of Food and Drugs 9
Overview of the FDA Initiative on
Obesity,
Robert E. Brackett, Ph.D., CFSAN,
FDA 19
Obesity - Therapeutics, David G. Orloff,
M.D.,
CDER, FDA 27
Obesity - Research, David W.K. Acheson,
M.D.,
CFSAN, FDA 47
Highlights of the Obesity Working Group
Report,
Alan Rulis, Ph.D., CFSAN, FDA 76
Questions and Discussion with the
Board/Presenters 96
Update on ORA Peer Review Process,
John R. Marzilli, Deputy Associate
Commissioner for Regulatory Affairs,
FDA 129
John J. Specchio, Ph.D., ORA Science
Advisor 135
Open Public Hearing:
Arthur Frank, M.D., George Washington
University 150
Elizabeth Jacobson, Ph.D.,
AdvaMed 158
Richard Atkinson, M.D., American
Obesity Association 170
Introduction to Critical Path, Janet
Woodcock,
M.D., Acting Deputy Commissioner for
Operations, FDA 175
Perspective on Anti-Infectives and
Vaccines,
Gail H. Cassell, Ph.D., Eli Lilly and
Company 220
Perspective on Chronic Disease
Therapies,
Robert M. Califf, M.D., Duke
University 241
4
C O N T E N T S
Drug Formulation and Development, and
Tissue
Engineering Issues, Robert S. Langer,
Sc.D., MIT 268
Overview of Opportunities - Drugs,
Robert Temple,
M.D., CDER, FDA 298
Overview of Opportunities - Devices,
Larry G.
Kessler, Sc.D., CDRH, FDA 324
Overview of Opportunities - Biologics,
Jesse Goodman, M.D., M.P.H., CBER,
FDA 339
Questions and Discussion with Board -
Recommendations 363
5
1 P R O C E E D I N G S
2 Call to Order
3 DR. SHINE:
Good morning, ladies and
4
gentlemen. We will come to
order. Welcome to this
5
meeting of the FDA Science Board Advisory
6
Committee. I am Ken Shine. Mark McClellan, when
7
he was Commissioner, asked me if I would chair this
8
committee then he split!
9 [Laughter]
10 I was a little anxious because our
11
colleague, Dr. Crawford, who is the Acting
12
Commissioner, was scheduled perhaps to be in Japan
13
but it turns out that he is able to be here so I
14
don't feel quite so deserted.
But we are pleased
15
to be able to welcome you to this meeting which
16
will be focused on two major and extremely
17
important issues for all of us, the issues relating
18
to the epidemic of obesity, which is a worldwide
19
epidemic of extraordinary proportions in terms of
20
all of the implications of that epidemic; and then
21
an examination of the Critical Path in terms of the
22
necessity to find ways to bring new products to
6
1
market in a timely and cost effective way.
2 We have a very distinguished advisory
3 committee and before we
ask Dr. Crawford to make
4
his introductory comments, perhaps we could go
5
around, starting with Dr. Pickett, and ask you to
6
identify yourself and then just a sentence or so
7
about your area of interest or perspective that you
8
bring to the committee so that folks have a sense
9
of your approach to things. Dr.
Pickett?
10 DR. PICKETT:
Yes, good morning. I am
11
Cecil Pickett. I am President of
Research and
12
Development for the Schering-Plough Corporation,
13
which is the pharmaceutical arm of Schering-Plough.
14 I
obviously have a fundamental interest in drug
15
discovery and drug development and how we can bring
16
innovative new therapies to patients.
17 DR. PRINCIPE:
Good morning. My name is
18
Jose Principe. I am
Distinguished Professor of
19
Electrical and Biomedical Engineering at the
20
University of Florida, and my expertise resides in
21
electrical systems and machine and learning
22
algorithms.
7
1 DR. RIVIERE:
Hello. I am Jim Riviere. I
2
am Distinguished Professor of Pharmacology, North
3
Carolina State University, and have expertise in
4
pharmacology and toxicology.
5 DR. GRIMA: Hi.
My name is Josephine
6
Grima. I am the Director of
Research and
7
Legislative Affairs for the National Marfan
8
Foundation, and I am the consumer representative.
9 DR. THOMAS:
Good morning. I am John
10
Thomas, Vice President, retired, and Professor
11
Emeritus of Pharmacology and Toxicology, University
12
of Texas Health Science Center at San Antonio.
13 DR. SWANSON:
Good morning. I am Katie
14
Swanson. I am with the President
of KMJ Swanson
15
Food Safety, a food safety consulting firm.
16 DR. LAURENCIN:
I am Cato Laurencin. I am
17 a
Lillian Pratt Distinguished Professor and Chair
18
of Orthopaedic Surgery, and university professor at
19
the University of Virginia. My
areas of expertise
20
are biomaterials, tissue engineering and
21
nanotechnology.
22 DR. PI-SUNYER:
I am Xavier Pi-Sunyer. I
8
1
am Professor of Medicine at Columbia University
2
College of Physicians and Surgeons, and I am
3
Director of the Division of Endocrinology and the
4
Obesity Research Center at St. Lukes Roosevelt
5
Hospital. My area of interest is
diabetes and
6
obesity.
7 DR. HARLANDER:
Hello. My name is Susan
8
Harlander. I am president of my
consulting firm,
9
called BIOrational Consultants. My
background is
10
in food science and nutrition and I work primarily
11
in the areas of genetically modified foods and
12
drugs.
13 DR. ROSES: I
am Allen Roses. I am Senior
14
Vice President in Genetics Research,
15
GlaxoSmithKline. I spent 27
years at Duke and the
16
final 20 of those I was Chairman of Neurology. My
17
expertise is in neurology and genetics, medicine
18
and a variety of other.
19 DR. SHINE:
Thank you. Have I missed
20 anybody
on the committee? We will be hearing
from
21
some of our colleagues at the FDA shortly. I am
22
Ken Shine. I serve currently as
the Executive Vice
9
1
Chancellor of Health Affairs at the University of
2
Texas System, after two terms at the Institute of
3
Medicine. I am interested in
health policy. As a
4
cardiologist, I also recommend the food for
5
breakfast--
6 [Laughter]
7 --and hope that in the interest of
8
controlling obesity we will all have heart healthy
9
lunches. So, we will see!
10 I think we are very fortunate that Les
11
Crawford can serve as the interim Commissioner. I
12
have known Dr. Crawford for many years and his
13
leadership in this organization has been manifest
14
over and over again. He brings
wonderful
15
experience and perspective and we will ask him to
16
make a few remarks. Les?
17
Welcome and Opening
Remarks
18 DR. CRAWFORD:
Well, thanks very much, Dr.
19
Shine, and thank you very much for agreeing to be
20
chairman. It is one of the
lasting legacies of the
21
great Mark McClellan that you are here.
You cannot
22
be replaced until he comes back--
10
1 [Laughter]
2 --it is almost like a biblical thing.
3 DR. SHINE: Now
I am worried!
4 DR. CRAWFORD:
In any case, I want to
5
welcome you as the chairman and also we have four
6
new members, all of whom I have known before except
7
Dr. Allen Roses, and I am particularly grateful for
8
you to agree to serve on this committee. We expect
9
great things from you and you will have a fun time.
10 I
think your particular area of expertise is very
11
much needed by the committee and I hope it is an
12
enjoyable and fruitful thing for you.
13 I have known Dr. Susan Harlander for a
14
long time, and she was been nominated for this
15
committee and turned us down one time before but
16
she can run but she can't hide.
She is right here
17
today, and thank you for that.
18 Xavier Pi-Sunyer and I served on the same
19
committee, another committee which met only one
20
day, Dr. Pi-Sunyer, and went away, some ten years
21
ago. This one is enduring. So, thank you also.
22 Gail Cassell--I think Gail is not here yet
11
1
but she is coming. Also, she is
very well
2
recommended because she comes from the University
3
of Alabama, Birmingham. She and
I speak the same
4
language and communicate often.
So, we are
5
grateful to her also.
6 We have had, as I mentioned earlier, the
7
transfer actually of Mark McClellan to the Center
8
for Medicare and Medicaid Services.
That is a blow
9
for us but it is a strengthening of our Department
10
of Health and Human Services, which is our umbrella
11
organization under Tommy Thompson.
So, Dr.
12
McClellan is still available to us and we see him
13
fleetingly but he is very much interested in what
14
is happening here, and we will continue to stay in
15
touch throughout this administration and beyond.
16 We remain focused on his strategic plan
17
that has now become our strategic plan and also on
18
the five goals where were developed during that
19
time, which were shared and worked out to some
20
extent in concert with this committee.
Those have
21
stood the test of time. All the
work is not
22
completed but we believe those benchmarks that were
12
1
established by the committee or commission on the
2
strategic plan which meets monthly in order to do a
3
check of how well we are doing--we think all of
4
that work will be accomplished within this
5
presidential administration. At
that time we will
6
put a ribbon around it, put it in a box and serve
7
it up to the American people. I
think it is a job
8
already well done but a job not quite completed.
9 We are focusing on our mission of
10
protecting and advancing public health.
We face
11
many new challenges and opportunities.
We are
12
proceeding with your help through the most
13
difficult part, which is bringing science to bear
14
on the regulatory process.
15 This session of the FDA Science Advisory
16
Board is scheduled, actually, at a very opportune
17
time. The agency has undertaken
a number of major
18
new initiatives, two of which you will be hearing
19
about in much greater detail during the remainder
20
of the day.
21 On March 12 of last year Secretary
22
Thompson released a new FDA report outlining the
13
1
agency's strategy for combating the major public
2
health problem of obesity, actually initiated it on
3
March 12 of last year and received a report from
4
FDA this year. In point of fact,
three different
5
agencies, under the leadership of Secretary
6
Thompson, did major initiatives on obesity. These
7
were six-month programs to bring together the
8
expertise in the National Institutes of Health, the
9
Centers for Disease Control and the Food and Drug
10
Administration.
11 During the time that those three
12
committees were working on the problem of obesity,
13
from its very fundamental scientific rationale all
14
the way up to ameliorative steps, the estimated
15
number of deaths from obesity-related diseases in
16
the United States increased from 300,000 per year
17
to 400,000 per year. A major
path-breaking paper
18
by Dr. Julie Gerberding and others at the Centers
19
for Disease Control and Prevention have estimated
20
that obesity is going to overtake smoking as the
21
leading cause of death in the United States by the
22
end of this decade.
14
1 These are chilling figures. It is a
2
horrible disease, a complex of diseases that is
3
lurching out of control. FDA,
through the
4
authorities it has under the Nutrition, Labeling
5
and Education Act, must do something.
Dr. Bob
6
Brackett is going to detail that for you. He came
7
in as Director of the Center for Food Safety and
8
Applied Nutrition in the middle of the obesity
9
initiative but his expertise and leadership are
10
very much part and parcel of our accomplishing of
11
this mission. He will talk about
the other people
12
from his Center and elsewhere who were on the
13
committee, but it is a great accomplishment for FDA
14
and I am very pleased that you are going to hear
15
about it in depth from Dr. Brackett himself.
16 The second major thing is the Critical
17
Path initiative which was developed in the last
18
stages of Dr. McClellan's tenure here.
It has
19 stood
the test of introduction; it has not stood
20
the test of time yet. But with
your help we will
21
get the Critical Path from the research laboratory
22
to the bedside, to the pharmacy, to the hospital
15
1
and all other places where we need to be sure that
2
FDA is not an impediment to the development and
3
introduction of new drugs and other modalities of
4
therapy, and also fine-tuning our procedures so
5
that we actually encourage by efficiency the
6
development of much needed drugs in much needed
7
areas.
8 One of the things that you will hear about
9
today is a publication, on a regular basis, of
10
Critical Path opportunities that we will list to
11
the public, to the industry that we serve, and also
12
to all other stakeholders about what we think are
13
the opportunities for new development that FDA
14
would like to work with sponsors on, work with
15
research institutions on and anybody and everybody
16
else. This is something new for
FDA and I think we
17
are ready for it, but you will also have some
18
targets of opportunity this morning to help shape
19
and develop that critical initiative.
I believe it
20
will be with us for a long time.
We have had some
21
similar kinds of things in the past that we have
22
come up with, but this one has a system to it and
16
1
we believe, if we do it right, it will be enduring
2
and it will mark a new era for the Food and Drug
3
Administration. So, your
critical attendance to
4
that is very much solicited. Dr.
Woodcock will be
5 talking
about that.
6 At this point I would like to turn the
7
floor back over to Ken Shine, with many thanks once
8
again for his leadership on this committee. I am
9
looking forward to the day, as I hope all of you
10 are. Thank you.
11 DR. SHINE:
Thank you very much, Dr.
12
Crawford. One of the
responsibilities of this
13
committee is to be the final judges in the FDA
14
scientific achievement awards program.
Those
15
awards apparently will be given in another month or
16
so, but I think one of the parts of our activities
17
has been the opportunity to look at the quality of
18
the science. I had a great deal
of difficulty in
19
choosing in many categories the more outstanding of
20
the proposal because the science was so good. I
21
think that is an area that has been extremely
22
encouraging to the committee.
17
1 We do have some other business we have to
2
undertake before we begin the program, and Jan
3
Johannessen is going to take care of that.
4 DR. JOHANNESSEN:
Thank you, Dr. Shine.
5
The following announcement addresses the issue of
6
conflict of interest with respect to this meeting,
7
and is made part of the public record to preclude
8
even the appearance of such at the meeting.
9 The Food and Drug Administration has
10
prepared general matters waivers for Drs. Shine,
11
Principe, Pickett, Grima, Riviere, Laurencin,
12
Swanson, Thomas, Roses, Pi-Sunyer, Cassell,
13
Harlander and one of the guest speakers, Dr.
14
Califf. A copy of the waiver
statements may be
15
obtained by submitting a written request to our
16
Freedom of Information office.
The waivers permit
17
them to participate in the committee's discussion
18
of FDA's obesity working group report and Critical
19
Path initiative.
20 Topics of today's meeting are of broad
21
applicability and, unlike issues before a committee
22
in which a particular product is discussed, issues
18
1
of broader applicability involve many industrial
2
sponsors and academic institutions.
The
3
participating committee members have been screened
4
for their financial interests as they may apply to
5
these general topics at hand.
Because general
6
topics impact so many institutions, it is not
7
practical to recite all the potential conflicts of
8
interest as they apply to each participant. The
9
FDA acknowledges that there may be potential
10
conflicts of interest but, because of the general
11
nature of the discussion before the committee,
12
these potential conflicts are mitigated.
13 We have open public comments scheduled for
14
eleven o'clock. I would just
remind everyone to
15
turn your microphones on when you speak so that the
16
transcriber can pick everything up.
Thank you.
17 DR. SHINE:
Thank you very much, Jan. I
18
think we are prepared now to go to our program. I
19
would just make a couple of observations. This is
20 a
very dense program and we are going to have to,
21
on the one hand, hold our speakers to the time that
22
is allotted to them and, at the same time, I would
19
1
urge you to make notes and recognize that if they
2
use their full allotted time we are not going to be
3
able to question them at that time but, rather, we
4
have a specific period at 10:15 for a period of
5
questions. So, if there are
issues where you
6 really need clarification in order to understand
7
what is being proposed, then by all means we ought
8
to do that. On the other hand,
we are going to
9
have to move expeditiously through the
10
presentations if we are going to get today's work
11
done.
12 We are pleased that the overview for the
13
FDA initiative on obesity, as you have heard, will
14
be provided by Dr. Robert Brackett and we invite
15
him to proceed.
16
Overview of the FDA Initiative on Obesity
17 DR. BRACKETT:
Thank you, Dr. Shine.
18 [Slide]
19 Good morning, everybody. I am Bob
20
Brackett and what I am going to do is give a very
21
brief overview of sort of the major points of the
22
obesity initiative, and then following me the next
20
1
three speakers will actually get into some of the
2
more final details that were part of the
3
initiative, as well as some of the summary of the
4
report.
5 [Slide]
6 I think the background for why we did this
7
is pretty much apparent to most people in that
8
people have sort of rediscovered that obesity in
9 this country and
worldwide is, in fact, of epidemic
10
proportions in that in the United States overweight
11
and obese people have increased risks of many other
12
chronic diseases that are relating to the deaths
13
that Dr. Crawford mentioned, including heart
14
disease, diabetes and certain types of cancer.
15
Also, as Dr. Crawford mentioned, it is enough so
16
that there are significant deaths, to the point
17
where that is competing with tobacco as a public
18 health
problem. So, it is something that we do
19
need to address within FDA, something that we have
20
partnered on with other entities within Health and
21
Human Services. In fact, this is
sort of a
22
nationwide program where it takes the participation
21
1
and the partnership of many different groups. That
2
is one of the things we have attempted to do and
3
will attempt to do.
4 [Slide]
5 One thing
that is lost, in addition to the
6
400,000 deaths that are estimated related to
7
obesity, is the economic cost of obesity to this
8
country. That is estimated at
117 billion dollars
9
per year. So, that has a major
impact on the
10
health costs in this country and does drive the
11
cost for everybody. So, it is
something that we
12
all need to participate in trying to eliminate.
13 [Slide]
14 The obesity working group was actually
15
created in 2003, in August, by Dr. McClellan when
16
he was here. At that time Dr.
Crawford was the
17
chair and my predecessor, Joe Levitt, was the
18
deputy chair at that time.
19 [Slide]
20 They were given a very simple charge, and
21
that was to prepare a report that outlines an
22
action plan that would cover the critical
22
1
dimensions of the obesity problem that FDA could
2
participate in with the other agencies as well.
3 [Slide]
4 During that time the group was very
5
active. From 2003 to 2004--that
is not much time
6
to do all of the activities that they had to and
7
they were very, very active. It
was on a fast
8
track. The obesity working group
met eight times
9
during that short period of time.
They received
10
briefings from a number of different invited
11
experts from the Department of Health and Human
12
Services, as well as and including Centers for
13
Disease Control and NIH. We did
hold one public
14
meeting and one workshop; two round table
15
discussions; and also solicited from the public
16
many different comments on our obesity related
17
issues by docket submissions.
So, in fact, we did
18
try to absorb as much information to put in this
19
report as we possibly could, and then tried to
20
synthesize all of what was provided and provide the
21
report that was required.
22 [Slide]
23
1 This is just a copy of the report, and
2
this has been submitted. It is
called "Calories
3
Count" and it is the report of the working group on
4
obesity. What the report
actually does is to
5
provide a range of both short- and long-term
6
recommendations to address this epidemic.
7 An important part is that we did try to
8
base all of these recommendations on known
9
scientific facts. So, it did
take some teasing
10
apart of what was thought to be contributing to
11
obesity versus what is known from a nutritional
12
standpoint. The one thing it
also attempted to do,
13
and did, is address the multiple facets that
14
contribute to the obesity problem, particularly
15
those that are under FDA's purview.
16 [Slide]
17 Some of the recommendation highlights that
18
were given from it are, first of all, developing
19
appropriate and effective consumer messages to aid
20
consumers in making wiser dietary choices. That
21
was one of the most important things, that is, to
22
actually get consumers to understand that they are
24
1
in control of their weight problems, and to
2
establish educational strategies and partnerships
3
to support these appropriate methods to teach
4
people, and particularly children with regard to
5
childhood obesity, how they can lead better lives
6
through better nutrition. So,
this is getting back
7
to some really fundamental issues that needed to be
8
addressed to the American people.
9 [Slide]
10 It also involved pursuing improvements to
11
the labeling of packaged foods with respect to
12
caloric and other nutritional information, that is,
13
giving the consumers the information they need to
14
make the judgments that we were trying to educate
15
them about.
16 Then, encouraging and enlisting
17
restaurants in voluntary, collaborative efforts to
18
combat obesity and then also provide nutritional
19
content information to consumers at the point of
20
sale. Since many of the meals
are eaten away from
21
the home, this was particularly important.
22 [Slide]
25
1 It also recommended facilitating the
2
development of new therapeutics that could be used
3
in the treatment of obesity, and then,
4
coincidentally, designing and collaborating with
5
others--and that "others" is very broad--effective
6
research in the fight against obesity.
So, it was
7
really trying to encompass the whole scope of
8
activities that could be done.
Again, the
9
important part was really involving the
10
stakeholders continuously throughout these various
11
processes. We had to not be
dictating what this
12
was but it was actually a very participatory
13
process.
14 [Slide]
15 The conclusions, in summary were, as many
16
of us already know, that the problem of obesity in
17
America really has no single cause.
There is no
18
one thing that you could point to; it is really the
19
result of a variety of different factors that when
20
they act together over time--and this includes the
21
genetic component for consumers and environmental
22
factors--do contribute to increasing obesity and,
26
1
consequently, there also is no single solution to
2
this problem in that the current trends will only
3
be reversed if you have well-coordinated,
4
complementary efforts from virtually all sectors of
5
society to combat this.
6 [Slide]
7 The other thing, and the part that we
8
needed to particularly emphasize to consumers, is
9
that the obesity epidemic is not going to be solved
10
quickly, neither will their individual weight
11
problems be solved quickly. Any
long-lasting
12
reversal of the phenomenon will itself be a
13
long-term process. We took a
long time to get to
14
this point and it is going to take some time to
15
reverse it.
16 [Slide]
17 Just to give you a little bit of update
18
about the obesity working group, Dr. Rulis, who
19
will be speaking this morning as well, has been
20
engaging in a national policy dialogue of what
21
things can and could be done with regard to
22
obesity, especially FDA's participation. Dr.
27
1
Orloff, who will also be following up, has dealt a
2
lot with the therapeutics aspects of obesity and,
3
of course, the research component, which will be
4
discussed by Dr. Acheson, can provide more details
5
on where the research sort of would be heading in
6
the future.
7 With that, I will end and, again, leave
8
some of the details to the speakers that follow,
9
and with them you will get a lot more of the
10
nitty-gritty of what the obesity working group has
11
accomplished in the last year.
12 DR. SHINE:
Thank you, Dr. Brackett. Why
13
don't we proceed then to hear from David G. Orloff,
14
who is the Director of the Division of Metabolic
15
and Endocrine Drugs in the Center for Drug
16
Evaluation and Research, to talk about therapeutics
17
involved in obesity?
18 Obesity - Therapeutics
19 DR. ORLOFF:
Good morning. Thank you very
20
much.
21 [Slide]
22 What I am going to use my time on to do
28
1
today is to give you a brief overview of what the
2
therapeutics subgroup which largely consisted of
3
members of the Division of Metabolic and Endocrine
4
Drug Products--myself, Dr. Eric Colman and Dr.
5
Patricia Beaston--reported to the overall obesity
6
working group according to our charge from Dr.
7
McClellan and the original charge to the group
8
itself.
9 Specifically, I will give you a sense and
10
an understanding, I hope, of where we are today in
11
our regulatory stance; how we got here; and what
12
our plans are for the immediate future as we
13
participate in this enterprise to address this
14
public health problem.
15 [Slide]
16 Let me start with two quotes which
17
actually are cited by Dr. Bray, who is a leader in
18
this field, in a recent paper that is actually
19
included in a volume edited by Dr. Pi-Sunyer who is
20
with us.
21 The first is that sudden death is more
22
common in those who are naturally fat than lean.
29
1
This sounds like a contemporary summary statement
2
from a modern epidemiologic study.
3 The second, which is a bit older sounding
4
but perhaps more broadly conclusive in a sense,
5
reads that corpulency, when in an extraordinary
6
degree, may be reckoned a disease, as it in some
7
measure obstructs the free exercise of the animal
8
functions, and has a tendency to shorten life by
9
paving the way to dangerous distempers.
10 Well, as the title of this slide suggests,
11
these are actually not new observations. The first
12
is a quote from Hippocrates and the second is a
13
quote from Dr. Fleming of the Edinborough School,
14
writing in 1760.
15 [Slide]
16 Today we call these risks associated with
17
obesity co-morbidities. While
they are certainly
18
better enumerated and their pathophysiology better
19 understood, as I have suggested, this is not a new
20
problem. As you can see and as
everyone is aware
21
around the table, these co-morbidities run the
22
gamut from cardiovascular disease and all of its
30
1
manifestations through the big problem of metabolic
2
syndrome and burgeoning epidemic or existing
3
epidemic of type 2 diabetes, to increased risk for
4 a
variety of malignancies and a whole host of
5
psychological disorders, including depression and
6
eating disorders.
7 One of the reasons I start with this slide
8
is because, frankly, the drug side of FDA has been,
9
for lack of a better term, accused in the past of
10
being somewhat insensitive to the magnitude of this
11
problem as reflected in all of these risks, and I
12
want to assure everyone that we fully understand
13
the overall metabolic and health consequences of
14
obesity and it drives our interest in ongoing
15
active participation in this initiative.
16 [Slide]
17 So, with that, Dr. McClellan gave us
18
really two charges that asked us to address the
19
problem, we believe, from two different sides. The
20
first was to assess the real or perceived barriers
21
to development of new or enhanced therapeutics.
22
Then, as I said, coming at it from the other side,
31
1
to make recommendations on ways to encourage
2
development of new or enhanced therapeutics.
3 We took this as a call, on the one hand,
4
to make sure, as always, that we were well informed
5
and up to date in our understanding of obesity and
6
its risks, but also that we really ought to be
7
revisiting our FDA guidance to industry on
8
development of drugs for obesity.
Needless to say,
9
FDA guidance to industry is supposed to represent
10
current thinking and I think it is worth at least
11
considering that 1996 is retreating into the past
12
and we ought to make sure that that is still our
13
current thinking.
14 So, we have an open comment period on that
15
document that is closing in a few weeks. We have
16
already had face-to-face dialogue with the American
17
Obesity Association of PhRMA, of which Dr. Crawford
18
was president, and we will be convening an advisory
19
committee later this year to discuss potential
20
changes to our guidance.
21 [Slide]
22 The harsh reality of the state of medical
32
1
therapeutics is really where we start.
The fact is
2
that although there are a number of drugs on the
3
market for the treatment of obesity, current
4
therapies are really not the panacea or in any way
5
miracle drugs. They generally
induce only modest
6
degrees of weight loss. Weight
maintenance is
7
really a central problem in addressing this public
8
health issue.
9 It is important to understand, with regard
10
to the state of the evidence, that we have limited
11
data available on the impact of drug-associated
12
weight loss on morbid outcomes.
The Xendos trial
13
with Orlistat in obese patients at risk for type 2
14
diabetes is one recent example of a finding that
15
touches or that addresses the effects of weight
16
loss on that significant outcome.
But we really
17
have no mortality data and we think that that is a
18
big hole. I think everybody
agrees that that is a
19
big hole in our understanding.
20 [Slide]
21 But having said that, opportunities
22
abound. This is a list of the
potential targets
33
1
and mechanistic approaches to the treatment of
2
obesity that are really under development and
3
consideration and research, and the list grows
4
daily with an increasing understanding by those
5
involved in the scientific side of this enterprise,
6
with increasing understanding not only of the
7
physiology of weight maintenance and of energy
8
economy in general, but also the pathophysiology of
9
obesity.
10 [Slide]
11 Let me tell you where we see the path
12
forward to safe and effective obesity drugs from
13
the standpoint of the Center for Drugs.
To do so,
14 I
need to tell you a little bit about the modern
15
history of obesity therapeutics; to talk to you
16
about standards of evidence for approval of drugs
17
prior to our 1996 guidance; and the transformation
18
in medical perception of obesity to the way it is
19
looked at today that led ultimately to the
20
development of our standards and rationale in our
21
1996 guidance. There are
multiple areas for
22
discussion. I will really touch
mostly on just one
34
1
of them.
2 These, if you will, constitute some of the
3
barriers which Dr. McClellan asked us to address,
4
as well as a number of unanswered questions that,
5
in our mind, constitute barriers as well, although
6
perhaps not barriers in the same way as regulatory
7
barriers.
8 With regard to the modern history, as I
9
think most people realize, it is a checkered one.
10
It has been fraught with ill-conceived mechanistic
11
approaches, unscrupulous investigators and
12
practitioners and a lot of bad luck along the way,
13
if nothing else.
14 [Slide]
15 It begins in the 1880s with the use of
16
thyroid extract and drug-induced hyperthyroidism as
17 a
path to weight loss with its attendant risks,
18
into the 1930s with dinitrophenol which,
19
incidentally, continues to rear its head even today
20
and which, as many of you realize, is one of the
21
poster children in the FDA chamber of horrors that
22
led ultimately to the passage of the 1938 Food,
35
1
Drug and Cosmetic Act, along with the elixir of
2
sulfinamide tragedy, to amphetamines which really
3
came to market in the mid-1930s and, as I will say
4
in a few minutes, dominated the scene with regard
5
to obesity therapeutics over the last half century
6
and their problems related to addiction and CNS and
7
cardiac toxicity.
8 The late 1960s marked really the end of
9
the Rainbow pill problem which had begun some 20 or
10
25 years before. These were
medical regimens which
11
included fixed-dose combinations of digitalis and
12
thyroid hormone but, in the collective, included
13
the addition of thiazide and other diuretics,
14
purgatives as well as amphetamines.
It was that
15
lethal combination of digitalis and thiazides with
16
hypokalemia and accentuated digitalis toxicity that
17
led to numerous tragedies and ultimately to
18
regulatory action against this enterprise.
19 The 1970s saw a mini epidemic in primary
20
pulmonary hypertension, mostly in Europe, related
21
to Aminorex and I think people are aware of the
22
unfortunate case of Redux and the unexpected, but
36
1
in some cases, frankly malignant cardiac
2
valvulopathy that resulted from the use of those
3
drugs.
4 I think it is worth pointing out, and
5
everyone should realize, that in terms of a barrier
6
that this history, on the one hand definitely
7
tempers our approach from a regulatory standpoint
8
to the development of therapeutics in this area,
9
but I also think that it has affected in the past,
10
and continues to affect, the overall integrity of
11
the anti-obesity enterprise and it is something, if
12
you will, that from a public relations standpoint
13
needs to be dealt with.
14 [Slide]
15 This is a list of the centrally-acting
16
anorexigens approved after 1938.
I put it up
17
really to show you that the last 50 years is the
18
half century of amphetamine congeners in obesity
19
research and therapeutics.
20 Two drugs on this list, mazindol and
21
sibutramine, are not amphetamines per se but I
22
think all these drugs are stimulant anorexigens and
37
1
they have all been marketed as controlled
2
substances. All, except for
sibutramine, as I will
3
explain in a moment, were approved and labeled for
4
only short-term use.
5 [Slide]
6 So, what were the standards of evidence
7
that we had to support the use of these drugs prior
8
to 1996? As I said, all of the
drugs approved
9
prior to that time, and so labeled for treatment of
10
obesity were for short-term use.
The trials that
11
supported the efficacy and safety of these products
12
were up to 12 weeks in duration and were limited in
13
size with 200-400 patients.
There were obviously
14
concerns about abuse and addiction potential that I
15
mentioned a moment ago with these products and, as
16 I
said, they were labeled for short-term use,
17
really dating back to the late 1960s and early
18
1970s.
19 Their efficacy was modest and, as you will
20
see in a moment, not really all that different from
21
what we have today, with the mean loss of
22
approximately 5 kg for placebo and a range of
38
1
placebo-subtracted means across multiple studies of
2
1-10 kg.
3 [Slide]
4 Although this is an age list epidemic and
5
the observations of its association with chronic
6
morbid and mortal conditions, you know, goes back
7
to the ancients, nevertheless, it took a sort of
8
reexamination to lead to a transformation in the
9
way this disease was perceived and in the way that
10 it
ought to be treated. As I said, today
it is
11
looked at as a chronic condition associated with
12
metabolic derangements and conferring the risk for
13
long-term morbid and mortal sequelae.
14 As I said, there is a big problem with the
15
high rate of weight regain following
16
discontinuation of drugs, and there is a
17
recognition that maintenance of healthy weight is
18
critical to reduction in risk for
19
obesity-associated adverse outcomes as opposed to
20
cycling of weight.
21 [Slide]
22 With this recognition or this casting in a
39
1
metabolic light, in 1992 the obesity drugs were
2
transferred to Endocrine and Metabolic, and in
3
1995, before my time, an advisory committee was
4
held to discuss this evolution and a disease model,
5
the standards for lifelong treatment algorithms
6
with these drugs, and discussion of clinical trial
7
design and evidentiary standards.
8 [Slide]
9 In 1996 a draft guidance was issued. The
10
cardinal features of that draft guidance include
11
the patient populations which are to be targeted,
12
which include those patients who have significant
13
obesity and who are at high risk for sequelae.
14
This is in keeping with the NIH evidence-based
15
treatment guidelines which were issued in that same
16
time frame.
17 The duration of Phase 3 trials which we
18
have adhered to up until now include a first year
19
of placebo-controlled investigation to provide
20
proof or principle of efficacy, followed by an
21
open-label year, in the second year, to provide
22
further information on durable efficacy and safety
40
1
in long-term use. This harkens
back to the
2
historical bad luck to which I referred and,
3
frankly, to the absence of outcomes data with
4
existing drugs and to the fact that we are not
5
requiring any outcomes data for establishing the
6
balance of risk and benefit for these products.
7 [Slide]
8 With regard to the efficacy criteria, it
9
is notable that the FDA's efficacy standard is less
10
rigorous than the European one.
We require either
11 a
demonstration of a mean placebo-subtracted weight
12
loss greater than or equal to five percent of
13
baseline body weight, or that the proportion of
14
subjects who lose greater than or equal to five
15
percent of their baseline body weight is greater in
16
the drug than the placebo group.
The EMEA, as you
17
can see, has ten percent cutoffs for both those
18
criteria.
19 [Slide]
20 Three drugs have been approved for
21
long-term treatment under the current guidelines.
22
The first, dexfenfluramine, was withdrawn less than
41
1
18 months after its approval, as I mentioned
2
earlier, related to unexpected cardiac
3
valvulopathy. Sibutramine, a
centrally-acting
4
anorexigen, and Orlistat, a non-absorbed intestinal
5 lipase inhibitor, remain on the market. These are
6
indicated for long-term use. As
I said before,
7
their efficacy over the loan haul, over the one to
8
two years of exposure in the clinical trials, is
9
about along the lines of the earlier products
10
approved for short-term use.
11 [Slide]
12 With regard to barriers, as I said, I will
13
mention just one, and this relates to the safety
14
exposures that have been required of sponsors
15 working
in this area. The FDA has stated a
minimum
16
of 1,500 patients for one year should be exposed to
17
these products prior to submission of an NDA, and
18
that 200-500 patients complete a second year to get
19
an understanding of durable efficacy and safety.
20 Those who have held that this might be too
21
rigorous have evoked the International Conference
22
on Harmonization E1A document on the development of
42
1
drugs for long-term treatment of
2
non-life-threatening conditions which cite 300-600
3
patients for 6 months and 100 for 1 year. They
4
cite also 1,500 patients total in all of the
5
studies, including Phase I.
6 [Slide]
7 But that document also explains that
8
longer and larger exposures are readily
9
rationalized and perhaps necessary if the benefit
10
of the drug is noted or expected to be small,
11
experienced only by a fraction of treated patients
12
and of uncertain magnitude as in reliance on a
13
surrogate, such as weight loss.
14 I think this characterizes the state of
15
the evidence and the state of the efficacy with
16
these products and it is really our rationale
17
behind requiring larger exposures.
Also, I think
18
we all understand, as has been said many times
19
today, that the magnitude of this epidemic is such
20
that the anticipated target populations in
21
open-market use is absolutely astronomical and it
22
is our responsibility to be as sure as we can be
43
1
that drugs are going to be safe and effective when
2
they go out into these millions and millions of
3
patients.
4 [Slide]
5 With regard to therapeutic gaps and
6
unanswered questions, one of the issues that really
7
we believe bears further investigation is
8
head-to-head comparisons of approved agents. This
9
is an issue that comes up in a lot of different
10
fields of medical therapeutics and is certainly one
11
in which I think patients and physicians are
12
under-served by the pharmaceutical and clinical
13
investigational communities as a means of providing
14
evidence for individualized, rational choices in
15
treatment.
16 [Slide]
17 Long-term safety and efficacy of older
18
drugs also needs better understanding, particularly
19
phentermine.
20 [Slide]
21 I show you this slide of drug use data
22
from 1991-2002 to demonstrate that throughout these
44
1
last 12-plus years phentermine leads the pack with
2
regard to U.S. prescriptions.
These were projected
3
prescriptions for 2002 but I think they were borne
4
out, and also projected for 2003.
Phentermine
5
prescriptions are double those for sibutramine and
6
Orlistat, yet, this is a drug that is approved only
7
for short-term use. It is likely
that in this
8
context it is being used for long-term use and we
9
really are poorly informed as to its long-term
10
safety and efficacy.
11 [Slide]
12 Studies of combination therapy, in order
13
that we can better label these drugs, are also
14
necessary and, parenthetically, studies of drug
15
cycling which I believe are commonplace in
16
therapeutic interventions in obesity--more
17
information is required for purposes of drug
18
labeling. Finally, the
"Holy Grail" that is the
19
question that we all yearn for an answer about, and
20
that is, do the drugs that we are going to use in
21
this area, both today's and tomorrow's, confer
22
long-term individual and population reductions in
45
1
morbid and mortal sequelae of obesity?
2 [Slide]
3 Finally, or actually next to finally, I
4
want to read you another quote, and this is from J.
5
Diamond, writing an interesting piece that he
6
published in Nature last year.
The piece is really
7
about the worldwide epidemic of type 2 diabetes,
8
diabetes and obesity that humans generally,
9
although some more than others, are programmed as a
10
result of energy economy genes in the aggregate to
11
develop when food is overly abundant.
That is
12
obviously why we are here today.
In discussing
13
this problem he states that an epidemic of a
14
genetic disease waxes because of a rise in
15
environmental risk factors, and then wanes when the
16
number of susceptible potential victims falls, but
17
only because of the preferential deaths of those
18
who are genetically more susceptible.
19 This is a sobering concept and that is why
20 I
wanted to bring it to our attention.
However
21
long obesity has been a problem, it is clear that
22
we are still on the waxing side of this story.
46
1
When the obesity and type 2 diabetes epidemics
2
begin to wane we want it to be because of the
3
success of our interventions and not because of a
4
culling of the susceptibles.
5 [Slide]
6 In conclusion, from the standpoint of the
7
therapeutic subgroup the magnitude of the obesity
8
epidemic, its contribution to chronic disease, its
9
costs to individuals and society, and the absence
10
of broadly effective therapeutics constitute a call
11
to action by the collective medical community,
12
including FDA, and we are certainly committed in
13
that regard.
14 Resolution of issues around real or
15
perceived barriers to development is paramount to
16
advancing the field of obesity therapeutics, though
17
we want to emphasize that this must be without
18
sacrificing the overall quality of the obesity
19
armamentarium.
20 With that, diet and exercise remain the
21
mainstays of prevention and treatment of obesity;
22
drugs are adjunctive to hygienic measures. They
47
1
were yesterday, they are today and they will be
2
tomorrow we believe.
3 Precedent with older as well as newer
4
drugs, reliance on weight loss alone as a measure
5
of health effects of these products directs a
6
cautious, measured approach in obesity
7
therapeutics. Questions, among
others, of
8
comparative efficacy and safety, long-term clinical
9
outcomes of treatment and effects of combination
10
therapy regimens must be addressed sooner rather
11
than later. Thank you very much.
12 DR. SHINE:
Thank you very much. I think
13
we will go ahead and have Dr. Acheson make his
14
presentation and then it looks like we may have a
15
few minutes before the break to inquire about both
16
of these presentations.
17 David Acheson is Director of Food Safety
18
and Security Staff in the Center for Food Safety
19 and Applied Nutrition, and he is going to discuss
20
issues related to obesity and research.
21 Obesity - Research
22 DR. ACHESON:
Good morning and thank you,
48
1
Dr. Shine, for that introduction.
2 [Slide]
3 What I am proposing to do over the next
4
few minutes is really to describe the process that
5
the obesity subgroup of the FDA obesity work group
6
underwent to begin to address the issue of obesity
7
research. This clearly is an
enormous area and so
8 I
want to begin with really just a focus on what
9
the overall approach was for this.
One of the
10
mandates for the obesity work group was to identify
11
the applied basic research needs that include the
12
development of healthier foods, as well as a better
13
understanding of consumer behavior and motivation.
14 [Slide]
15 The overall approach that we took here was
16
really to begin with focusing on research topics
17
that were mission-relevant to FDA.
When we began
18
to think about research in the context of obesity,
19
clearly, there is an enormous amount of research
20
going on and we needed to focus it down and target
21
it towards FDA relevant issues.
22 The approach that we took to try to do
49
1
this was to think through and to document current
2
and relevant research that was related to these
3
mission-relevant topics, and I am going to go
4
through some of those, and based on that, then to
5
identify knowledge gaps.
Essentially, my
6
presentation is set up to look at some of the
7
research issues and subsequent knowledge gaps.
8 [Slide]
9 As we discussed this, we really tried to
10
narrow this down and came up with three principal
11
areas. The first was in relation
to labeling and I
12
would say this is probably one of the largest areas
13
that we discussed. The labeling
was broken down
14
into labeling in relation to restaurants and
15
labeling in relation to the Nutrition Facts Panel.
16 We also considered translational research,
17
and where I am going there is in relation to some
18
of the translation between the basic science and
19
the mission-relevant issues for FDA.
We identified
20
three areas there that are of potential relevance:
21
the first, neonatal imprinting; second, "omics",
22
essentially genomics, proteomics and metabolomics;
50
1
and, thirdly, the impact of caloric restriction and
2
what research there was going on in FDA and other
3
areas that was related to that.
4 The third point was just to consider the
5
other areas. You just heard from
Dr. Orloff
6
extensive discussion about drugs in relation to
7
obesity. But we also wanted to
consider drugs and
8
devices, food additives and dietary supplements
9
and, essentially, those areas were largely excluded
10
from our consideration because they weren't mission
11
relevant research for FDA. I am
not saying that
12
they are not relevant to the quest of tackling the
13
obesity problem but, as you will see, my
14
presentation doesn't cover those in any great
15
detail and the explanation is really that the
16
research in those areas is not relevant to the
17
mission of FDA.
18 [Slide]
19 So, to focus firstly on the labeling
20
research, the current research and some research
21
that was undertaken once the obesity work group was
22
established worked through the root of focus
51
1
groups. One of the areas that
was addressed was
2
restaurant labeling and to determine consumer
3
reactions to menus that include caloric information
4
and, essentially, to ask the question whether, if
5
you provide caloric information on a menu, it makes
6 a
difference to consumers' choice.
7 The second area in relation to the food
8
label was to consider consumer reactions to the
9
nutrition facts panel.
Obviously, there is a
10
limited amount of information you can get on a
11
facts panel, but what are the critical elements
12
that one needs to get on? Some
of the areas that
13
were addressed there were the issue of calories of
14
the daily value, what did the consumers understand
15
by that; what the impact would be of eliminating
16
the calories from the fat section of the nutrition
17
facts panel; and then to consider what the impact
18
would be on adding more information on
19
multi-serving packages to show the actual calories
20
and the percent daily value in the package.
21 Where we were going there was essentially
22
that if you look at items that are typically
52
1
consumed in a single serving, such as a muffin, it
2
may be labeled in such a way that it is considered
3
that a muffin is two servings and not many of us
4
going to consume just half a muffin and put it on
5
the side for consuming later.
So, it was more
6
logical to think through the total amount of
7
calories of the serving.
8 Obviously, the third point was to consider
9
messages and to determine what is the most
10
effective message for conveying these nutritional
11
issues.
12 [Slide]
13 As part of this, FDA has been involved in
14
the development of a social sciences model to help
15
determine the factors that influence dietary and
16
weight management. That has
really involved three
17
areas The first is the review of
literature to try
18
to identify factors that affect food behavior and
19
to catalog existing data; the second, to develop a
20
quantitative model really to get at what an
21
individual's decisions are affecting weight. This
22
works into physical activity and choice of food and
53
1
that, obviously, involves attitudes, behaviors and
2
environmental factors. Then, to
get into the issue
3
of cost benefit analysis and the role of food
4
labeling regulation and the development of policy.
5 [Slide]
6 Some of the consumer perceptions and
7
attitudes--and this is just a brief summary--are
8
listed on this slide. Some of
the things that we
9
learned from this brief study of the research
10
environment was the perception of overweight versus
11
obesity and the fact that, in fact, most or many
12
consumers don't consider overweight as a "problem"
13
and indicate that it is of little consequence to
14
consumers. It is different with
obesity, but just
15
simply being overweight is not a significant
16
consequence.
17 Secondly, the perception of a person's
18
weight and typically adults and teenagers
19
mis-perceive their weight. Men
tend to
20
underestimate their weight and healthy, underweight
21
women tend to overestimate their weight. Also,
22
parents misjudge the weight of their children and
54
1
will often determine that their children are at
2
healthy weight where, in fact, they may be a little
3
overweight.
4 The third point is what consumers'
5
perception is of their diet, and there is a
6
tendency to think that generally you are eating a
7
more healthy diet than you actually are. If you
8
analyze the content of the diet, the perception
9
certainly is that you are eating better than, in
10
fact, you are eating.
11 Some of the recent focus groups have
12
really tried to target parents and children.
13
Obviously, part of our thinking here was the
14
research in relation to product obesity and trying
15
to work on that angle of it.
16 The second point there is the perception
17
of obesity, which I have alluded to, in terms of
18
access for information and the perceived barriers
19
and the motivating factors that will allow people
20
to alter these perceptions.
21 The conclusion of all of this was that
22
there is emphasis on incremental change. I think
55
1
as you have already heard, nothing is going to
2
happen here quickly. Also, we
have to be cautious
3
of over-saturation of consumers.
They obviously
4
want to receive health information but we have to
5
do this in a targeted, judged and careful way.
6
Finally, the focus on child education, I think, was
7 a
very clear message as an area of research in
8
relation to this whole consumer aspect that has to
9
be considered as an important focus.
10 [Slide]
11 The knowledge gaps: the first knowledge
12
gap that comes out of this is information used to
13
facilitate consumers' weight management decisions.
14
There is a very clear need for research in this in
15
both a qualitative and quantitative fashion. I
16
have just summarized three points there. Firstly,
17
the consumer reaction to the food label. I have
18
gone over some of the issues there, but what are
19
the advantages of highlighting the calories, or
20
listing the quantitative amounts of nutrients in
21
multi-size packages? Our initial
focus groups
22
indicated that these are important areas. What we
56
1
need now is to take this into quantitative research
2
to really document the impact that we might get
3
from some of these changes.
4 The second point, the consumer reaction to
5
and the effectiveness of restaurant nutrition
6
information. Does it make a
difference if we list
7
the information about the calories, the fat, the
8
sodium next to the items on the menu list when you
9
go to a fast-food environment?
10 Finally, the consumer's dietary behavior
11
and attitudes toward weight management, which
12
obviously encompasses what I have already talked
13
about but gets into the area of physical activity
14
as well.
15 [Slide]
16 The second knowledge gap is the
17
relationship between obesity and food consumption
18
patterns. Where we are going
there is essentially
19
the relationship between obesity and the frequency
20
of foods consumed in different locations. The
21
target there is to examine the impact of consuming
22
food in the home environment versus the fast-food
57
1
environment versus a restaurant environment, and
2
does it make a difference in terms of the frequency
3
and the types of foods that are consumed in those
4
environments?
5 Also, we recognize that there is an
6
important impact of both socioeconomic status in
7
this as well as ethnic background, and there is a
8
very clear need to generate data in relation to
9
both of those issues.
10 Finally in this second knowledge gap are
11
the factors that actually contribute to overeating.
12
Obviously, those are encompassed with many of the
13
things I have already discussed, but what is the
14
role of the super-size portion in all of this, and
15
that really needs to be investigated.
I am just
16
using that as an example of factors that would
17
contribute to overeating.
18 [Slide]
19 The second area is in relation to
20
formulation research, and really what this was
21
addressing was what do we know about the factors
22
that will drive reformulation to develop healthier
58
1
foods? The questions that were
posed there were
2
whether current regulations offer a barrier or an
3
incentive to the production of healthier foods.
4
The research that is currently ongoing there is
5
that there are currently discussions with key
6
industry personnel, and this is being done through
7 a
third-party contractor, so that we can get some
8
insight into the barrier and regulatory hurdles
9
that may be present that could impact product
10
reformulation to develop healthier food.
11 The third bullet there gets into the
12
issues focused on manufacturers--what they can put
13
on the label; what claims they can make; possible
14
changes in regulations that would facilitate
15
healthier foods; and what incentives we may be able
16
to put in place, as a regulatory agency, that would
17
encourage the development of healthier foods.
18 [Slide]
19 So, the third knowledge gap in relation to
20
this product reformulation is where do we need to
21
go? What do we need to do? We really need to
22
further explore these barriers.
As I said, there
59
1
is research that is going to give some pointers but
2
the three key areas there are, firstly, do
3
incentives, e.g., label prominence, impact industry
4
on the development of healthier foods.
What is
5
going to drive them to do that?
Do the barriers,
6
e.g., regulatory hurdles, have an impact on the
7
development of healthier foods?
We really need to
8
understand that, understand what those hurdles are
9
and, if possible, see what we can do to get around
10 them. Obviously, the third point there, once we
11
have tried to balance the incentives and the
12
hurdles, is how can these be addressed.
13 [Slide]
14 The next area in relation to drugs,
15
devices, food additives and dietary supplements--as
16 I
mentioned, this was not a primary focus for the
17
reasons that I have already given.
18 [Slide]
19 However, we did identify one knowledge gap
20
here that we thought was important. That was the
21
potential for FDA-regulated products to be
22
unintentionally contributing to the obesity
60
1
problem. The two points there
were to consider
2
whether weight gain may be an unintended and
3
unrecognized complication of certain medications.
4
Obviously, there are examples of that.
A research
5
gap here or a research need is that this has not
6
been consistently measured. It
has not been
7
consistently evaluated, and we need to consider
8
whether this should be an adverse effect that has
9
to be taken into account in the approval of a drug.
10 So the sub-bullets there are that, first
11
of all, we have to determine if this is a problem.
12
We identified it as a knowledge gap, and I
13
emphasize knowledge gap. We have
to figure out
14
whether it is a problem that needs to be addressed.
15 Then, obviously, the corollary of that is
16
to develop animal models to study the long-term
17
effects on weight management and medications.
18
Obviously, animal models can go in many other
19
directions too but the issue of unintended
20
consequences of a drug that is taken for a
21
completely unrelated reason could be tested through
22
appropriate animal models.
61
1 [Slide]
2 Finally, there is the basic science
3
component of this. I mentioned
the three areas:
4
Neonatal imprinting, and where we were going there
5
was to try to get a better understanding of the
6
impact of developmental programming, particularly
7
during early development so, obviously, fetal
8
exposure, neonatal exposure, infant exposure when
9
these metabolic pathways are being
10
established--does what mommy eats when she is
11
pregnant, does what the neonate is fed or the young
12
infant fed, what is the impact of that on
13
subsequent development of obesity?
Part of our
14
thinking on neonatal imprinting was the impact of
15
infant formula often on childhood obesity and
16
subsequent adult obesity.
Certainly, an area there
17
that is of interest is the push towards different
18
types of formula, without really I think major
19
consideration on the impact of the obesity issue in
20
relation to new formulas.
21 The second area is obviously the "omics"
22
issue to identify susceptibilities.
This is really
62
1
getting at linking genetic susceptibilities.
2
Obviously, this goes beyond just the susceptibility
3
to obesity but the susceptibility of obese people
4
to develop other problems such as type 2 diabetes.
5 A
clear area there is through genomics and
6
subsequently proteomics and then metabolomics to
7
really get a better handle on how that all fits
8 together.
9 Finally, the effects of caloric
10
restriction, and over the last ten years NCTR has
11
placed some emphasis on this.
They have been
12
working on that and have been looking at the effect
13
of caloric restriction, and this is in relation to
14
free radical formation, malignant tumor rates and
15
"less than ad lib" feeding on rapid weight
16
reduction. so, there is also a
need there in terms
17
of further research.
18 [Slide]
19 The fifth knowledge gap relates to this
20
area of translational research, and we felt that it
21
was essential that FDA use basic research to drive
22
and develop regulatory policies.
This is obviously
63
1
going to be especially with NIH but also many or
2
the other stakeholders.
3 Just as a small aside there, I was
4
recently at a meeting at NIH where I presented some
5
of what we have been doing and, interestingly,
6
while there was a lot of discussion about the basic
7
metabolic pathways and many of the potential drug
8
therapies that Dr. Orloff discussed were raised,
9
the group really was beginning to focus on that
10
obviously a lot of this is about calories, and to
11
get a better understanding of caloric management
12
and some of the "social" sciences around food
13
intake.
14 But the knowledge gaps here include
15
obviously the area of getting a better
16
understanding of developmental imprinting, the use
17
of "omics" and the development of animal models for
18
the effects of diet, drug therapy and long-term
19
weight maintenance. So, there
are some common
20
themes there.
21 [Slide]
22 To conclude, obviously, calories are a
64
1
critical element and the research that we were
2
focused on was really how can we, having taken that
3
basic premise that calories are what this is all
4
about--how can we focus our research to get at
5
that. Probably the most critical
element of all of
6
this is understanding consumers.
We need to
7
understand food labeling and how the consumers
8
react to it; their eating habits and their weight
9
management. When we have a
better understanding of
10
that, we can then use food labels, education
11
programs or whatever to try to move in the right
12
direction.
13 The third point there is the development
14
of healthier foods. I have
already discussed that
15
obviously in the context of what can we do as a
16
regulatory agency to encourage healthier food
17
development, and then, finally, the input from
18
basic research in the development of regulatory
19
policy which is always an underpinning, the good
20
science has to be there first to move forward.
21
Thank you very much.
22 DR. SHINE:
Thank you very much. We have
65
1
an opportunity now, before our break, to engage our
2
three speakers on any issues that the committee
3
would like to raise. What kinds
of questions or
4
issues would you like to inquire about?
Xavier?
5 DR. PI-SUNYER:
I would just like to ask
6
Dr. Orloff--I agree with him completely about the
7
need to do studies of combination therapy to look
8
at either the safety or the efficacy of combination
9
drugs, but as he knows and I know, the drug
10
companies only want to test their own drug and not
11 a
combination because of the expense of doing that.
12
So, how would he propose that this get done?
13 DR. ORLOFF:
Well, the first thing is that
14
this is a problem obviously, as you point out,
15
during the patent life of drugs.
I suppose, to
16
some extent, the community would have to give
17
thought to really how large and long the
18
combination therapy trials might actually have to
19
be. Again, I am just thinking
off the top of my
20
head here but it occurs to me that if the long-term
21
safety and efficacy of the individual products has
22
been established as a basis on which they have been
66
1
approved by the FDA, then that principle stands
2
and, obviously, on a case-by-case basis this would
3
have to be considered but it seems reasonable to
4
think that we might be able to prove principle of
5
efficacy and acceptable safety of combinations,
6
perhaps not to the standards of directing labeling
7
but certainly to a standards that would provide
8
some guidance to physicians in this field, thus,
9
standards for publication in peer-reviewed journals
10
which, for better or for worse, is not always the
11
same as standards for labeling.
So, that is one
12
possibility.
13 FDA cannot force drug companies to do
14
trials with drugs other than their own unless, I
15
suppose, there was some clear-cut rationale, for
16
example, that the combination of drug with another
17 was such that it spared toxicity of the
proprietary
18
drug and FDA stated that they would be really
19
unwilling to approve--and this would be precedent
20
setting--the proprietary drug at its optimal
21
effective dose but would consider approving it in
22
combination with another product which was still
67
1
effective, the two together, but where the toxicity
2
was spared. Those are just a few
thoughts.
3 DR. SHINE: Dr.
Orloff, just to follow-up
4
on another element here, you quite appropriately
5
showed that obesity is a chronic illness and that
6
it is likely that one might have to think about it
7
much like hypertension in that it may require
8
lifetime treatment. What do we
know about the
9
capacity to select individuals who, in fact, might
10
be most effectively treated with drugs long term in
11
terms of either their behavioral characteristics or
12
their genetic characteristics?
That is, could one
13
stratify the obese population into categories where
14
it would make more sense to make major efforts with
15
drug therapy as opposed to other kinds of
16
approaches?
17
DR. ORLOFF: I don't think we have that
18
information available to us. I
am certainly not
19
the expert to give the definitive answer but what
20
you have touched on is clearly a problem that is
21
active in all areas of medical therapeutics, and it
22
is the great hope for obesity, as well as you
68
1
suggested for hypertension and other sort of risk
2
factor modification approaches to disease
3
prevention, that we would be better up-front at
4
identifying patients, on the one hand, who are at
5
greatest risk for the sequelae--and we are better
6
and better at doing that for cardiovascular disease
7
risk based on biomarkers and such--but, on the
8
other hand, are likely to benefit most from the
9
drug from the standpoint of, for example,
10
pharmacogenomic directed responsiveness. I don't
11
know if anybody else around the table--Dr.
12
Pi-Sunyer, if you have any thoughts on where the
13
state of the science is there.
14 DR. PI-SUNYER:
No, I think that is a very
15
good point but I think we are not there in terms of
16
knowledge that we can select people who we know
17
will get a better effect or are less likely to have
18
safety problems with a drug than others.
19 DR. SHINE: I
am fascinated by the
20
implications of the fat gene argument made by Jerry
21
Diamond and a number of other people, including the
22
Rockefeller people, and the question again of
69
1
whether over some period of time we could identify
2
targets that are more likely to be responsive than
3
others.
4 Both of you touched on this, but in terms
5
of the science one of the biggest gaps we have is
6
the research on behavior and how and in what way we
7
can influence behavior. We know
that for a whole
8
variety of other addictions the only way that one
9
is usually successful in dealing with tobacco or
10
alcohol, or whatever, is total cessation.
11
Unfortunately, that doesn't work for obesity.
12 DR. ORLOFF:
Yes, I think it is an
13
interesting point that overall poor or less than
14
satisfactory outcomes in obesity treatment, even in
15
patients who remain on drugs, is a combination
16
likely of at least two factors.
One is perhaps
17
that in truth the efficacy of the drug wanes but
18
probably more than likely because the behavioral
19
components that determine the ultimate success of
20
the intervention are, you know, that the patient
21
falls off the wagon, if you will.
I talked about
22
the concept of drug cycling, which I gather is
70
1
utilized by a number of obesity docs.
The degree
2
to which it is successful after the first couple of
3
cycles I think is in question.
4 DR. SHINE: Dr.
Swanson?
5 DR. SWANSON:
This is a question for Dr.
6
Acheson. Building on the
behavioral issue, we all
7
know that part of the issue is that calories count
8
but it is an intake-output balance that is
9
necessary. Was the consideration
of how to
10
motivate people to expend more energy or some kind
11
of a link between X number of calories--to eat this
12
thing that contains X number of calories you would
13
need to expend so many steps?
Or, was that out of
14
scope because it is kind of out of FDA's purview?
15 DR. ACHESON:
Well, it certainly was
16
discussed, Dr. Swanson. The
issue of caloric
17
balance is obviously input-output.
Unfortunately
18
FDA can't regulate exercise. If
we could, maybe we
19
would solve the problem. But
what we did there, we
20
had a number of presentations from people who are
21
involved in that.
22 For example, the VA has a program that
71
1
they are just initiating called "Move" in which
2
they are essentially identifying their obese at
3
various points through their healthcare system and
4
enrolling them, as much as possible, in some kind
5
of exercise and weight management program. So,
6
they are trying to tackle it from both sides. Our
7
approach was, well, what can we do on a label, etc.
8
that is going to have an impact, and that is half
9
the story. You are right, it is
not just
10
understanding caloric intake; it is also output.
11
So, we didn't ignore it. We need
to partner with
12
people to try to get that piece across.
13 DR. SWANSON: I
just think that,
14
considered from a prevention aspect, rather than
15
targeting obese or overweight people and this is
16
what you need to get the pounds off, another
17
approach that I really haven't seen is how do you
18
prevent the pounds from going on, and you can have
19
the super-size meal if you walk to a certain
20
restaurant versus driving.
21 DR. ACHESON: I
totally agree and that is
22
part of understanding consumer behavior. There is
72
1
no question that that is important.
Also, I think
2
let's try and get in there and prevent the problem
3
before it ever starts.
4 DR. SHINE: Let
me follow-up on that. The
5
experience with tobacco is that you can do
6
substantial amounts of public education. You can
7
probably change the overall attitudes of a subject
8
but, in fact, much of the progress made in tobacco
9
was related to taxing tobacco; to changing the
10
environment in which people could smoke, and so
11
forth. In the case of obesity,
clearly there is no
12
population that is at greater risk than children
13
where, on the one hand as Dr. Swanson points out,
14
physical exercise--gym is often gone from the
15
curriculum but, on the other hand, they are being
16
exposed to a substantial amount of fast foods.
17 I was impressed by the way in which in
18
your focus groups you identified that consumers
19
like to get the information about a group of foods
20
that they would likely eat together, the burger
21
example of the combination of fries, Cokes, and so
22
on and so forth. My question is
whether an
73
1
appropriate target would be schools and school
2
administrators around the combination of what kids
3
are eating in school to try to educate them and
4
their parents as to what the risks are in having
5
access to what in many institutions is just awful.
6
Now, some school districts are clearly making an
7
attempt to modify that but changing the environment
8
by virtue of changing what is available to kids
9
might be a very important strategy here, and there
10
the FDA might be able to come up with some labeling
11
activities, similar to the way in which you have
12
grouped the information about burgers, in a way
13
that would be useful to administrators and PTAs.
14 DR. ACHESON: I
agree with you. It does
15
focus on the product population which is possibly
16
the primary target here. I think
that is one
17
avenue to pursue. Obviously,
schools have many
18
things that they have to deal with and think about,
19
and this would just be one component of managing a
20
child's life, so to speak, to try to control their
21
diet.
22 That speaks also to developing educational
74
1
programs to help parents understand the importance
2
of caloric balance for their children.
Maybe that
3
could be done through the schools as well,
4
potentially working with the school lunch program,
5
for example, to try to think about healthier foods
6
in that context. So, I agree
with you. It is
7
certainly an avenue that one should consider.
8 DR. SHINE:
What I am arguing about is not
9
just about the education; it is actually changing
10
the environment in a substantial way.
Dr. Roses?
11 DR. ROSES: I
was also going to make the
12
point about the only reason that we eat is so we
13
can burn the calories, and if you are only
14
measuring one side of the equation as part of the
15
studies that are done and as part of the advice we
16
give we may be just dealing with, one might say, a
17
minor part of the problem.
18 I was pleased to hear that there is now an
19
acceptability that we are genetically different and
20
we have different susceptibilities, and some of us
21
will gain weight on different calories and others
22
won't. But I think, rather than
be in the second
75
1
point about "omics" as if there were some magic in
2
applying techniques to biological problems, there
3
is some evidence already that we can stratify
4
populations by their genetic polymorphisms. For
5
instance, this room could be stratified by your
6
blood types. However, just in
the beginning state,
7
the science called pharmacogenetics is applying
8
that to responders versus non-responders the
9
extremes of a drug trial.
10 With our experience with a recent
11
candidate, in a very small study which is in the
12
public domain and I can show it if you like in a
13
single slide, we were able to put stratifying
14 markers in a very simple way on people who
were the
15
big responders and people who could not respond and
16
people who, by the way, had a difference in weight
17
gain if they gained weight on the drug.
18 The third part has to do with head-to-head
19
trials. Head-to-head trials take
the average group
20
of patients and they test them against a drug. But
21
in the world of stratification comparing one third
22
that might respond that might respond in this group
76
1
with the one third that might respond in this group
2
is not really head-to-head, and it is now possible
3
to stratify people by their metabolic enzymes, by
4
their markers, by a variety of other things so that
5
we can truly do, if you are interested in making
6
that a gold standard, head-to-head trials that at
7
least have some common denominators.
8 DR. SHINE: Any
comments? If not, I am
9
going to take the chair's prerogative and go to a
10
break. We are going to have an
opportunity after
11
the next speaker to have a continued discussion,
12
and I presume you will be available for further
13
conversation. Let's go ahead and
take a break. We
14
will reconvene at 9:45. Thank
you.
15 [Brief recess]
16 DR. SHINE: We
will come back to our
17
discussion after the next presentation.
We are
18
pleased that Alan Rulis, Senior Advisor for Special
19
Projects in the Center for Food Safety and Applied
20
Nutrition, will tell us about the highlights of the
21
obesity working group report.
Alan?
22
Highlights of the Obesity Working Group Report
77
1 DR. RULIS:
Thank you, Chairman.
2 [Slide]
3 I am Alan Rulis.
I am currently Senior
4
Advisor for Special Projects in the Center for Food
5
Safety and Applied Nutrition. I
have been involved
6
with the obesity working group writing team and
7
throughout the process of putting that report
8
together.
9 To summarize very briefly, the report
10
recommends a number of thrusts in several different
11
areas that you are all aware of now, in particular
12
the food label and in particular there calories,
13
serving size, claims and so forth, even the issue
14
of carbohydrates. We have
recommendations in the
15
report about enforcement against products with
16
misleading weight loss claims and, as you have
17
heard this morning, there are sections that deal
18
with therapeutics and research.
19 [Slide]
20 What I would like to focus on in my
21
presentation is a couple of other areas that are
22
extremely important to the success of FDA's efforts
78
1
to stem the tide of obesity.
They relate to
2
education and restaurants, under the rubric of
3
trying to find means by which we can get around the
4
table the people involved in these issues, not just
5
the agency but stakeholders, educators, industry,
6
consumer groups, and so forth, to discuss next
7
steps and to move forward, to make some sort of
8
forward progress.
9 In that regard, the report does recommend
10
in fact that the FDA use a third-party facilitator
11
as a convener to essentially create a national
12
dialogue on obesity. So, what I
would like to
13
cover in my remarks this morning are some slides
14
that relate to development of that national
15
dialogue by means of a facilitator to get the
16
discussion going.
17 I looked over my slides and realized that
18
25 of my 30 slides are really questions. Of the
19
last 12, 9 of them are questions to the Board. So,
20
my slides will be primarily in that vein, looking
21
forward to your insights, comments, recommendations
22
and feedback to us.
79
1 As I said, the report does in fact
2
recommend that FDA work through a facilitator to
3
provide a forum for stakeholders to seek a
4
consensus-based approach that addresses two
5
specific aspects of obesity in the United States,
6
in particular, developing options for providing
7
voluntary nutrition information for foods consumed
8
away from home, for example in the restaurant
9
setting and, secondly, education to combat product
10
obesity.
11 The reasons are quite obvious. We know
12
that Americans now spend about 46 percent of their
13
food budget on food eaten outside the home, as our
14
report states. We also know that
body weight and
15
trends towards obesity that begin at puberty or in
16
adolescence are often propagated through adulthood
17
and are very difficult to reverse.
So, we think
18
that those are two prime areas for establishing a
19
dialogue.
20 [Slide]
21 With respect to foods eaten away from
22
home, we imagined putting together a contract for a
80
1
facilitator to start the dialogue and we imagined
2
having the facilitator address the following
3
questions: What nutrition
information would be the
4
most helpful for consumers to have before ordering
5
food in a restaurant?
6 [Slide]
7 What are the best options for providing
8
nutrition information in a restaurant setting? We
9
have had many discussions with the restaurant
10 industry throughout the course of the obesity
11
working group's efforts focused on this question.
12
It is a complicated question because the restaurant
13
industry is very diverse. There
are quick service
14
restaurants where people walk in and get their meal
15
right away and there are others where there are
16
white tablecloths and meals are prepared to order,
17
and there are a lot of specific changes that are
18
made to menu items by the cook.
It is very
19
difficult to say that there is one set of best
20
options, of course.
21 [Slide]
22 Should nutrition information be listed for
81
1
all menu items or just some items?
What are the
2
practical considerations there?
If we do have
3
restaurants put information on some menu items,
4
then what criteria should determine which items
5
should have nutrition information listed? We have,
6
as a matter of fact, a very interesting menu right
7
now in the agency from a major chain restaurant
8
that has caloric information next to all of the
9
menu items and it is very interesting, very
10
remarkable to look at because you are not used to
11
seeing it but it can be done and it is there.
12 [Slide]
13 When providing nutrition information in
14
the restaurant setting, what consideration should
15
be given to the differences between chain and
16
non-chain restaurants? Of
course, that relates to
17
the whole question of the diversity of that segment
18
of the industry.
19 [Slide]
20 How should restaurants tailor the kinds of
21
nutrition information presented based on expected
22
clientele? You can imagine the
need there to focus
82
1
the kinds of information that is presented to the
2
type of restaurant and the clientele that frequent
3
that restaurant.
4 [Slide]
5 Number six, should nutrition information
6
be presented in context, for example, as a percent
7
of daily value, comparing for example to a 2,000
8
calorie per day diet or, for example, comparing to
9
other menu items? This is
something that I think
10
many consumers would find very helpful, the notion
11
that when you walk into a restaurant and order an
12
item off the menu, you have some sense of whether
13
consuming that item is going to comprise 20 percent
14
or 80 percent or 150 percent of your normal daily
15
caloric intake. That sort of
feel for what that
16
meal represents would be extremely helpful we think
17
to restaurant patrons.
18 [Slide]
19 Number seven, how should FDA proceed to
20
encourage restaurants to participate in a voluntary
21
pilot program to test various options?
In the
22
obesity working group report we say as part of the
83
1
recommendations that we would like to see a pilot
2
study conducted that would get into the question,
3
in specific terms, about how such nutrition
4 information,
caloric information for example, could
5
be placed at the point of sale for consumers to
6
use, and we envision the restaurant industry coming
7
with ideas about this. We don't
envision the FDA
8
funding and creating such a pilot study out of thin
9
air. This is the kind of thing
that would need to
10
be done around the table, in communication with the
11
industry and, thus, again the idea of a third-party
12
facilitator.
13 [Slide]
14
Number eight, how can
industry and Food
15
and Drug Administration measure the effectiveness
16
of providing nutrition information to consumers in
17
restaurants? One of the things
we learned in our
18
deliberations in the obesity working group is that
19
information itself doesn't really necessarily
20
answer the concern; that imparting information to
21
people can be done but it is not necessarily
22
absorbed and, once absorbed, it does not
84
1
necessarily change behavior. So,
I think a
2
pertinent question that we would pose to our
3
third-party facilitator would be to get at the
4
answer to this question.
5 [Slide]
6 On the subject of pediatric
7
nutrition/obesity education, we have similar
8
questions that we would like a facilitator to help
9
us address. On what age groups
is it most
10
appropriate for Food and Drug Administration to
11
focus education efforts? This is
a sociological
12
question as much as a pharmacodynamic one.
13 [Slide]
14 What are the most appropriate settings,
15
for example school or home, health settings, social
16
organizations or clubs, for such educational
17
efforts? Again, those of us who
were on the
18
obesity working group are well aware that there are
19
many groups which have spent years working on these
20
kinds of issues and know a lot about imparting
21
information to children, to families and the
22
pitfalls of trying to create an awareness about
85
1
such subjects as obesity and then to change
2
behavior. It is very
complicated. It is almost an
3
art form. It is something that I
think many of us
4
at the agency are feeling like we really do need to
5
have a participatory discussion about and, again,
6
is another reason for having a third-party
7
facilitator bring in people with different areas of
8
expertise to get around this question.
9 [Slide]
10 Another question on the pediatric
11
nutrition/obesity education issue is how important
12
is it for education efforts to be conducted using
13
mass media, for example television, radio and
14
print? When I think of
adolescents today, they are
15
wired, they are electronic, they are not reading
16
newspapers; they are plugged in and they can be
17
reached very effectively in many ways through the
18
Internet, through the media, through the various
19
things that they are tuned into constantly. So, we
20
need to be contemporaneous here.
We can't be
21 thinking
in terms of 20th century communication
22
tools.
86
1 [Slide]
2 Over what time period should an education
3
effort be extended to achieve optimal impact and
4
lasting effects? Another thing
we learned in our
5
deliberations in the group was that education is
6
really not a short-term phenomenon.
It is possible
7
to generate messages, to send them out and, you
8
know, clean the dust off your hands and walk away
9
and affect absolutely nothing.
What has to happen
10
is that there needs to be a long-term effort that
11
is continuous, that is reinforced and that has
12
mid-course corrections to make it more effective.
13 [Slide]
14 What types of messages are the most
15
effective, and in which age groups, for educating
16
children about nutrition and health?
Again, this
17
ties back into the question of prevention, the
18
notion that if we can prevent adolescents or young
19
people from starting on a path of being overweight
20
we have done a whale of a lot more than we can do
21
by worrying about how to get people who are
22
currently overweight as adults to lose some weight.
87
1
That is also important but it is extremely
2
important to realize that once that pattern is set
3
it is very hard to reverse.
4
[Slide]
5 To what extent should education and/or
6
other types of messages be tailored to different
7
ethnic and/or socioeconomic groups?
This is
8
another issue we heard strong signals on in our
9
facilitated discussions during the course of our
10
obesity working group. We are a
heterogeneous
11
nation, extremely so and becoming more so. We have
12
large numbers of immigrants coming to our country
13
from different cultures where eating patterns and
14
signals about what is right to eat and what is
15
appropriate to eat, what is appropriate to feed
16
your family are very, very different.
Even those
17
of us who have been here a number of generations
18
across our country have very different views about
19
how we eat and what is appropriate.
So, we need to
20
be able to tailor our efforts and I think our
21
facilitator needs to be able to help us to do that.
22 [Slide]
88
1 Then, what are the effective means that we
2
might be able to find for partnering in the public
3
and private sectors to develop and deliver obesity
4
education? The Department of
Health and Human
5
Services has published a proposal in the Federal
6
Register, last summer, that invited comments and
7
suggestions on how to create public/private
8
partnerships on a number of health issues that face
9
the country, and this is one of them.
We are I
10
think open to the question. We
would put the
11
question to our third-party facilitator. Let's get
12
to the essential factors that help us partner with
13
those stakeholder segments out there, the industry
14
segment and the consumer segment and the citizen
15
segment of our country, to get everybody around the
16
table to chew on this issue.
17 It is not simple because the FDA, as a
18
regulatory agency, needs to be careful about how it
19
does those things so it does not present either the
20
appearance or the actuality of favoritism to any
21
one company over other companies, and so forth.
22
But we also do see the value of partnering. A lot
89
1
can be done that isn't regulatory in the
2
traditional sense. Talking to
and working with
3
effective partners on the outside can be extremely
4
effective, and we would like our facilitator to
5
help us uncover effective examples that are
6
currently operating, and there are some. We, at
7
the FDA, are very familiar with the "Fight Back"
8
campaign that was addressing the issue of microbial
9
contamination of food and that is a very effective
10
program that has educated a lot of people and
11
prevented a lot of illness in this country. It is
12 a
nice example of how that can work.
There are
13
undoubtedly many others and we want to find out
14
what they are, learn about what makes them
15
successful and then adopt the aspects of them that
16
make them successful to the programs that we would
17
generate.
18 [Slide]
19 That brings me then to my last dozen
20
slides or so on which there are still nine more
21
questions, and these are really questions that go
22
to the Board, and you have all seen them in your
90
1 materials
and I hope you have given them some
2
thought. They are focused on the
two prongs of the
3
issue as I have presented them here, namely, the
4
scope of work that we would present to our
5
contractor or third-party facilitator relating to
6
food eaten away from home, and work on pediatric
7
obesity education.
8 Chairman, I would I guess defer to you in
9
terms of how you want to proceed.
I can go through
10
these questions quickly if you would like me to
11
just to remind folks of what they are, and then we
12
can kind of revisit them.
13 DR. SHINE: Why
don't you go through them?
14 DR. RULIS:
Okay.
15 [Slide]
16 Number one, are FDA's proposed questions
17
and issues likely to provide appropriate
18
information to proceed with a pilot program with
19
restaurants? In other words, are
we asking the
20
right questions here? Are we
missing something
21
obvious? We really need your
thoughtful input and
22
your recommendations about that.
If we haven't hit
91
1
the target right on the head, then can you think of
2
some other questions or issues that should be
3
addressed and that we should have our facilitator
4
focus on?
5 [Slide]
6 Secondly, what kind of evaluation is
7
appropriate to assess the effectiveness of a pilot
8
program? As I mentioned, we
wanted to get together
9
and have the restaurant industry create some sort
10
of program by which this can occur.
I should say
11
very quickly that these programs are already going
12
on out there. You know,
McDonalds and other
13
companies have been in the news recently about
14
programs that they are putting together to move the
15
country in the right direction in terms of
16
increased physical activity and wise food choices.
17
So, we are well aware of that.
We want to engage
18
with them on something specific in this context and
19
so how do you evaluate the effectiveness of such a
20
program? I think we need your
help on that.
21 [Slide]
22 What advice would you have for a
92
1
facilitator? If we were to go
forward and contract
2
out this work, what advice would you have for a
3
facilitator concerning the basis for evaluating
4
recommendations on providing nutrition information
5
in a restaurant setting? Again,
are we targeting
6
this right? Have we asked the
right questions or
7
are we missing something obvious?
8 [Slide]
9 Fourth, what research would be helpful for
10 a
facilitator to know about to help them provide
11
the best guidance to the agency on this subject?
12
Again, you all, on the Board, have many different
13
areas of research that you are familiar with.
14
Perhaps you are aware of some that could be of
15
value for us to hear about on the record today that
16
could help us focus this in the right direction.
17 [Slide]
18 Fifth, in view of the materials that you
19
have been provided, is there any other advice or
20
information you believe is important to give the
21
FDA on this issue?
22 [Slide]
93
1 That also goes for the pediatric obesity
2
issues as well, and I will go through those
3
questions briefly. Are FDA's
proposed questions
4
and issues likely to provide appropriate
5
information to guide the development of useful and
6
understandable nutrition/obesity education efforts?
7 I will tell you why I put
8
nutrition/obesity on this slide.
In our
9
discussions with people who have been doing this
10
for years, they say one of the things you don't do
11
when you talk to adolescents about obesity is use
12
the word "obesity." As
I say, it is an art form
13
and children are extremely interested in knowing
14
how to live healthy lives, and they are very
15
conscious of their bodies, their body weights and
16
their physical appearance, and they very definitely
17
want to be healthy. But if you
come at them with
18
information that talks about obesity, it is a
19
turnoff. These are things that
are sort of the
20 subtleties of this issue that we have to get
around
21
and understand.
22 [Slide]
94
1 What research would be helpful for a
2
facilitator to know about?
Again, this is the same
3
as in the other aspect, to provide the best
4
guidance to the agency on this subject.
5 [Slide]
6 What other questions should FDA be asking
7 a
facilitator to explore in order to help the
8
agency develop effective educational strategies?
9 [Slide]
10 In view of the materials you have been
11
provided, is there any other advice or information
12
you believe is important to give FDA on this issue?
13 [Slide]
14 Let me finish by saying just two things
15
here. I have a couple of
quotes. One is from Dr.
16
Crawford: "We're going back to basics, designing a
17
comprehensive effort to attack obesity through an
18
aggressive, science-based, consumer-friendly
19
program with the simple message that 'Calories
20
Count.'" That really in a
sentence summarizes what
21
it is all about on that subject.
22 [Slide]
95
1 Our Secretary, Tommy Thompson: "Counting
2
calories is critical for people trying to achieve
3
and maintain a healthy weight.
This new report
4
highlights FDA's overall strategy for getting
5
consumers accurate, helpful information that allows
6
them to make wise food choices at home, at
7
supermarkets and in restaurants.
Taking small
8
steps to eat a more balanced diet and to stay
9
physically active can go a long way to reversing
10
the epidemic of obesity that harms far too many
11
Americans."
12 I think we are at a point where we have
13
written the report and we have delivered the
14
report. The report has
recommendations and we now
15
want to move forward with those recommendations to
16
take the next steps. What I have
described by way
17
of third-party facilitated national dialogue on
18
these two main issues about restaurants and
19
pediatric obesity are concrete first steps that I
20
think we can take beyond just delivering a report.
21
We look forward to your feedback on these questions
22
and your good guidance to the agency.
That is my
96
1
presentation. Thank you.
2
Questions and Discussion with the Board/Presenters
3 DR. SHINE:
Thank you very much, Dr.
4
Rulis. I know there are a couple
of questions that
5
we were going to come back to with our first two
6
speakers, and we will do that at the end of this
7
session, but I would like to focus over the next
8
few minutes on the challenge from Dr. Rulis with
9
regard to commenting on the charge to the
10
facilitator. I gather he is not
asking us the
11
answer to those questions. He is
asking us are
12
those the right questions and are there other
13
questions or other approaches that would be
14
important for FDA to undertake.
We also have a
15
written copy of these questions.
Jim, why don't
16
you go ahead?
17 DR. RIVIERE: I
just have one comment.
18
Focus groups seems to be the primary mechanism that
19
you are going to ask this facilitator to address
20
these questions, and just that there be a lot of
21
concern about the makeup of those focus groups. It
22
seems trivial, but even from personal experience
97
1
with focus groups, people who will go to focus
2
groups, there is a selection process in that area,
3
especially for some of the questions on teens--and
4 I
still have two teenagers. One has a
problem in
5
this area and the other one doesn't.
You know,
6
same environment; obviously different genetics--but
7
the point really is in looking at who would
8
participate in this and how you would get them to
9
participate because otherwise your results are
10
going to be strongly biased to people who may not
11
even have a problem to begin with.
And I agree, if
12
you say it is an obesity focus group you are not
13
going to get the right target population you want,
14
but just to try to look at some clinical trial
15
sampling protocols to make sure you have a
16
representative group.
17 DR. RULIS:
Thank you. Let me just add in
18
response, if I might, that in the course of our
19
deliberations we did have the opportunity to do
20
some focus groups and we did them, and they
21
contributed to the report. But
we were all very
22
much aware that, yes, indeed, they are not really
98
1
research; they are focus groups, and that what
2
needs to happen is research that has a much broader
3
and deeper sort of foundation and a long-term
4
lasting effect. So, yes, thank
you. I think it is
5 a
very important point and we need to keep that
6
very clearly in front of us.
7 DR. SHINE:
Susan?
8 DR. HARLANDER:
When you mentioned that
9
you were going to get a third-party facilitator
10
involved, I am assuming that that is going to go
11
way beyond consumer-based focus groups.
My
12
understanding is that that facilitator would also
13
be engaging the other stakeholders, like the
14
restaurant industry or the food industry--
15 DR. RULIS:
Very definitely.
16
DR. HARLANDER: --and I think it is
17
critically important to involve stakeholders
18
because food companies and restaurants understand
19
their consumer base and they do a tremendous amount
20
of proprietary research internally that they might
21
be quite willing to share with you under certain
22
circumstances. So, I think
industry involvement is
99
1
critical.
2 Personally, I tend to believe that the
3
market is going to sort itself out on this issue.
4
You already see that certain restaurants, if they
5
perceive that their consumers are interested in
6
more healthy alternatives, are including them on
7
the menu already. So, you know,
I think that they
8
can bring a tremendous amount of knowledge to what
9
you are trying to do. I guess my
understanding is
10
we are going way beyond consumers and will involve
11
the affected industry as well.
That is true?
12 DR. RULIS:
Yes, thank you.
13 DR. SHINE:
Susan, given your background
14
in food science, what is your general reaction to
15
the set of questions that have been raised? Are
16
these the right questions, do you think?
17 DR. HARLANDER:
Well, there is one
18
critical question that I would add.
I would agree
19
with all the questions that you have but I think
20
one key area is that all of this information that
21
will be provided is not free, and nowhere in the
22
research that I have reviewed prior to this do we
100
1
ask consumers if they are willing to pay for
2
additional information. If we are
going to expect
3
the food industry to change their labels, we have
4
to understand that that is not free.
You know, it
5
costs money to make all those changes, particularly
6
if analytical work is going to be required. The
7 same will be true for restaurants. They are not
8
typically giving that kind of information to
9
consumers so either they are going to have to use
10
some sort of software package that is going to
11
predict this, which they haven't typically used in
12
the past, or they are going to have to do
13
analytical--they are going to have concerns about
14
liability.
15 So, I think one question I would add is if
16
there are costs associated with that, are consumers
17
willing to pay. I just spoke
with one gentleman
18
and we talked about, you know, maybe consumers
19
would be willing to pay more for a smaller bag of
20
French fries than for a larger bag of French fries.
21
If those costs were there, would that influence
22
what consumers would buy?
Because I think that
101
1
kind of information would be very helpful as you
2
approach restaurants to try to get them involved.
3
So, I think if there is a cost associated with
4
this, we need to understand that.
5 DR. SHINE:
Other comments? Yes, John?
6 DR. THOMAS: I
would like to make some
7
comments with respect to focusing on our youngsters
8
and the various inequities in school systems across
9
the country. Certainly, it seems
to me, and I am
10
generalizing now because I am sure some schools are
11
much more advanced than others and maybe this is
12 more detail than you are really searching for at
13
this point in time, but establishing under the
14
rubric of PTAs, for example, some sort of
15
nutritional council where, in fact, they bring the
16
PE teacher, the health educator, other dieticians,
17
assuming that particular school board has access to
18
that because I realize there are going to be some
19
small schools that may not have those--I know, for
20
example, a lot of schools have begun to put
21
restrictions on school vending machines which
22
heretofore was a major source of discretionary
102
1
monies for the school principal where he or she
2
could go out and buy books, or whatever, which
3
otherwise was not funded by their particular
4
budgets.
5 Also, building into that overall concept
6
some sort of notion of nutritional incentive.
7
Weigh these kids at the beginning of the year and
8 see
what happens at the end of the year.
You know,
9
give them a goal to shoot for.
Don't just talk
10
about it on the first day and then forget about it.
11 We can't change the busing system with
12
respect to physical activities.
We all live in the
13
suburbs now so that one is taken away from the
14
pediatric cohort.
15 Finally, I think there needs to be a
16
reemphasis on individual sports in our school
17
systems. Team sports have
obviously taken
18
precedence but they need to have lifelong physical
19
activities and we need to encourage them through
20
the physical education departments and whatever
21
structure they have to go forth with lifelong,
22
single competition sports.
Nowadays you can't even
103
1
get kids to run out to the ball field because it
2
has to be organized. There is no
such thing as
3
getting a group of youngsters just to play ball for
4
the fun of it. It has to be a
lesson or it has to
5
be a league. Somehow we need to
encourage tennis,
6
golf and other things that they are going to do for
7
the rest of their life. Again, I
am sure there are
8 schools
that are already doing this but it has to
9
be a coordinated effort for these various skill
10
sets.
11 DR. SHINE:
Thank you. Other comments?
12
Yes, Cato?
13 DR. LAURENCIN:
Well, I would like to echo
14
some of the thoughts that were said earlier. I
15
think that a number of groups are already trying to
16
do this; a number of large chains are already doing
17
this. My one comment is that I
think the FDA can
18
be very helpful in terms of young people by trying
19
to keep it simple. I went to one
restaurant where
20
the place mats actually had all the nutritional
21
information for all the different foods that they
22
had. I have two doctorate
degrees and I was just
104
1
about able to get through it, so I think that a
2
young person might have a difficult time getting
3
through it.
4 We are trying a pilot program at our
5 university where we are
labeling the foods in the
6
vending machines with green, yellow and red. The
7
foods that are labeled red have actually five
8
percent surcharge that goes to research. So, these
9
sorts of methodologies are where perhaps the FDA
10
can be helpful in terms of being able to decide
11
what is green, what is yellow and what is red and
12
also, of course, the possibility of recommendations
13
in terms of how we can disseminate this sort of
14 information in terms of what the different
levels
15
are, and things like that.
16 One, I think it is important to try and
17
keep it simple in terms of actually trying to help
18
young people. Number two, we can
even place some
19
monetary incentives and make that a part of it, and
20 I
think it would be good.
21 DR. SHINE:
Thank you. Josephine?
22 DR. GRIMA: In
recent years the public has
105
1
been bombarded with information about low
2
carbohydrate diets, and although this is not a
3
trend that has been endorsed by the AMA or
4
government associations, the fact of the matter is
5
that that information is out there in the public.
6 I
think one of the important things to address is
7
how much information is really the consumer taking
8
in, and I think the facilitator has to use that in
9
their plans in order to educate the public on this.
10 DR. RULIS:
Yes, I will just respond
11
briefly on the carbs issue. The
carbs issue is
12
front and center right now. It
is on a lot of
13
packages we see in the stores and we are well aware
14
that this is an issue that is really crying out for
15
some clarification. As we have
stated in our
16
report, we are currently in possession of some
17
petitions that are intending to try to clarify some
18
of these statements on labels and we are working
19
very hard. Those of us who have
spent many years
20
at FDA, we relish the opportunity to beat our head
21
against hard problems--
22 [Laughter]
106
1
--and this is a difficult one
because, you
2
know, what is a carb, first of all, and then how do
3
you impart information in a way that is not
4
misleading and, yet, doesn't impinge on people's
5
right to say things under the First Amendment?
6
These are difficult questions but we look forward
7
to trying to solve them.
8 DR. SHINE:
Susan?
9 DR. HARLANDER:
I just have one reaction
10
to what Cato brought up. I think
that there is a
11 lot of concern about labeling foods as good
foods
12
or bad foods, and that you will get a lot more
13
cooperation from the food industry and the
14
restaurant industry if any food, whether it is a
15
green dot, yellow dot or a red dot, can fit into a
16
balanced diet, and it has a lot to do with how much
17
you are going to consume of that.
But I think
18
there is a lot of concern about is something going
19
to be categorized as a bad food as a result of some
20
of these efforts, and it might create some lack of
21
willingness to cooperate.
22 DR. SHINE:
Yes, Xavier?
107
1 DR. PI-SUNYER:
I think you get around
2
that by talking about total calories, and you get
3
around the carbohydrate issue also by talking about
4
total calories, which I think is what the FDA is
5
trying to do. I think the really
important thing
6
is how many calories people are eating in relation
7
to how much exercise they are doing.
8 I would just like to say a couple of
9
things. Number one is that I
think it would be
10
nice if the FDA would collaborate with other HHS
11
agencies on these campaigns. For
instance, CDC is
12
very much involved with the physical activity
13
aspect of trying to educate the public, and it
14
would seem to me that eventually, if and when the
15
FDA goes out with an educational program for
16 raising
the consciousness of the American people or
17
to raise the consciousness of the restaurants, or
18
whatever, that there be kind of a concerted effort
19
with agencies like the CDC which are working on the
20
other side, which is the expenditure side. That is
21
number one.
22 Number two, I think that the main emphasis
108
1
really ought to be on prevention.
I agree with Dr.
2
Swanson completely that weight loss is extremely
3
difficult, and it is expensive, and it is hard to
4
do and it is hard to maintain.
What I think the
5
message needs to be for children and for adults is
6
to control their weight. The
average American is
7
gaining an enormous amount of weight every year and
8
over a decade they are gaining between 10 and 20
9
pounds. So, what we want to do
is to prevent that
10
weight gain from age 21 or age 18 on, and I think
11
that is where the focus really ought to be of the
12
campaign.
13 DR. SHINE: Dr.
Crawford is going to
14
comment, but again, looking at the questions that
15
you were asked, are these the right questions for a
16
facilitator? Are you comfortable
with the
17
direction that this is taking?
18 DR. PI-SUNYER:
Well, I am comfortable
19
with that direction. I think we
need to know more
20
about behavior but, you know, the trouble is that
21
we do know quite a bit about behavior already,
22
particularly from other campaigns with tobacco and
109
1
other substances of abuse, and I think changing
2
behavior is very hard. I am not
sure that these
3
focus groups are going to give you the answers
4
totally that you are looking for.
5 I do think talking to the focus groups,
6
particularly about the labeling, is very important
7
because clearly, as Cato says, you don't want to
8
give them a label that they are not going to be
9
able to read or that they are going to throw up
10
their hands and say I don't understand this. So, I
11
think the whole label reading aspect of this focus
12
group initiative is very good.
13 DR. CRAWFORD:
Your question about
14
coordination within HHS was a good one.
The way
15
Secretary Thompson dealt with this as he moved this
16
towards the top of his priority list was that, as I
17
mentioned, about a year ago he caused NIH, CDC and
18
FDA to put together these in-depth studies, headed
19
up by working groups chaired by the deputies at
20
each one of those organizations.
I was the one who
21
chaired the FDA one.
22 Then, one week about three months ago each
110
1
of the agencies presented their findings to the
2
public. It was obesity week for
Secretary Thompson
3
and he had three separate and then one unified
4
press conference. At that point,
the Department of
5
Health and Human Services took over the public
6
education initiative. FDA's role
will be to feed
7
into that, as will the CDC's.
That initiative of
8
education has begun with an ad council campaign.
9
Then it will be followed up with some more in-depth
10
approach as we attempt to reach all strata of
11
American society. The message
will be refined. It
12
will be more sophisticated as it moves forward from
13
the base, which is like a cartoon type presentation
14
which is called sometimes the "body parts"
15
presentation because in the most famous one of
16
those ads someone is walking through a supermarket
17
aisle and there is some protoplasm on the floor,
18
and a little girl asks her parent what is that on
19
the floor, and the parent responds it looks like
20
someone lost their double chin from eating well.
21 [Laughter]
22 It is a very catchy campaign. I am
111
1
compelled to say so and that is how we get
2
coordinated.
3 DR. SHINE: Let
me just follow-up on this
4
for a minute. I think there is a
real question
5
here to understand the niche of the FDA in this
6
area. I think one needs to
explore very carefully,
7
not only in terms of the federal agencies but state
8
health departments, medical associations and others
9
who are confronting this problem at local
10
communities. Changing this with
national publicity
11
creates a certain kind of environment, if you will.
12
But where the rubber hits the road is going to be
13
concerted efforts in communities that will address
14
all of the elements. We have
talked about
15
restaurants and schools but what about company
16
cafeterias and the whole question of whether, in
17
fact, it is economically desirable to a company
18
which has X amount of turnover to be paying Y
19
amount of dollars in order to minimize the obesity
20
that goes on in that company because it might help
21
them save healthcare costs with diabetics and
22
hypertensives, but that depends on turnover.
112
1 So, the kinds of programs that we are
2
talking about are going to be partnerships. They
3
are going to be efforts to get the media, to get
4
the schools, to get industry to relate to this
5
issue of everything from bike lanes to exercise or
6
fitness programs in companies.
7 My point is that I think the extent to
8
which the facilitator can really hone in--the
9
agency is famous for its capacity to analyze
10
science and to find science-based information which
11
could form the basis of all of these programs.
12
Articulating that, modifying that, working on that
13
so that it is part of these overall programs is
14
going to be key. Therefore, I
would urge them to
15
look fairly broadly as to what we are talking about
16
and I think, again, I would not exclude state
17
health departments. I wouldn't
exclude both urban
18
and rural environments as places to understand what
19
the niche is.
20 I would also suggest to you that
21
increasingly in 8th grade biology there is an
22
interest of young people about their bodies, and so
113
1
forth. Anorexia-bulimia has been
a major target
2
for those educational programs.
The question is
3
whether those educational programs could provide a
4
balanced view of diet, exercise, and so forth, and
5
is the FDA an ideal source of information packages
6
and materials that could be used in public schools?
7
With all due respect to USDA, I worry about the
8
USDA's way of presenting that.
Perhaps between the
9
two agencies one might be able to come up with
10
material that would be part of the curriculum in
11
terms of that.
12 Finally, I want to emphasize, as I
13
mentioned before about tobacco, that everything we
14
have learned about changing behavior has to do with
15
not just the information but changing the
16
environment, whether it is because you pay more or
17
whether because something is accessible or not.
18
Therefore, when you talk about children, I happen
19
to think that is where an enormous amount of the
20
focus should be. I think the
public cares about
21
their kids. I think they will
often do one thing
22
and do something else for their kids.
So, as far
114
1
as they are concerned, starting in kindergarten,
2
the environment in which they are living will make
3
an enormous difference as to what their lifetime
4
habits are. So, I don't think
you can start too
5
early. Although I have focused
on the 8th grade
6
biology course, the fact is that if a second grader
7
sees vending machines all around with high cal food
8
or high fat food, they are going to assume that it
9
is part of the environment in which they live. So,
10 I
think the emphasis on pediatrics is extremely
11
important but that one ought to think about it in
12
terms of how and in what way you can use the data
13
that you have to change the environment in terms of
14
what is available to people.
Katherine?
15 DR. SWANSON:
Kind of building on the
16
notion that behavior is important, I personally
17
think that your question related to how do you
18
measure effectiveness is the most important one.
19
If you go down the path of trying all of these
20
different concepts and you don't have a way to
21
measure behavior change--not just what people say
22
they will do but what they will do--you really
115
1
won't be able to target what is going to have the
2
biggest impact. So, I would make
sure that that
3
question might be thought of first with every
4
strategy that you are looking at, how would we
5
impact the magnitude of the change.
6 Another thing to consider with regard to
7
the restaurants, listing the calories I think is
8
very interesting. The
segregation out of healthy
9
choices is another. Another
option that would be
10
very interesting to explore that would be
11
applicable to the white tablecloth restaurants as
12
well would be smaller portion sizes.
I mean, many
13
of us who are on the road all the time and you go
14
to the restaurant and you simply cannot eat all of
15
the food that is put on your plate, and would you
16
be able to pay not half as much because there are
17
labor costs there, but offering multiple service
18
sizes might be a very interesting thing to study
19
because if you reduce it by half you have cut your
20
calory intake in half.
21 One more question that you might want to
22
add is, is there a way that you can link the
116
1
caloric intake piece to the caloric output piece in
2
some kind of a message?
3 DR. SHINE:
Thank you, Katherine. I would
4
suggest that I would like a facilitator to create a
5
model for continuous quality improvement in this
6
area. What I mean by that is we
don't know what
7
works. We don't know what will
not work. We need
8 a
concept that there is a lot of experimentation
9 going on to learn from that what impact it
has, and
10
then constantly reassess how and in what way we are
11
using our activity. I think the
notion is that it
12
is for the long haul. I think
that looking at that
13
in terms of socioeconomic status, ethnic groups,
14
and so forth, is going to be pretty central
15
because, in fact, they may behave very differently
16
and it is not at all clear to me, once you get the
17
data and put something in place, that it will work
18
in the same way in a variety of settings. So, I
19
think that notion of what is the long-term model
20
for constantly reassessing where one is and where
21
do you get the data, how do you get the data, what
22
are you trying to measure in order to make those
117
1
decisions would be an important part of how to
2
conceptualize this.
3 Incidentally, we were talking at the break
4
about differences in terms of environment and
5
culture. You might be interested
to know that the
6
Japanese, about a year or two ago, redid the
7
height/weight tables for boys because the boys are
8
getting so much bigger and putting on more weight.
9
They are also getting higher cholesterol levels,
10
incidentally. But girls are
not. The culture in
11
Japan is so focused on young women being thin that
12
they have shown no change in the height/weight
13
environment. So, these cultural
elements make an
14
enormous difference in terms of how people behave.
15
Other comments?
16 DR. THOMAS:
Just a couple of follow-up
17
points, and these are certainly not revelations but
18 I
would strongly urge as you look at behavioral
19
profiles that you make every effort to get a
20
quantifiable endpoint. That is
almost a
21
contradiction when you are dealing with behavior,
22
but as you go into this I think you really need to
118
1
look at those endpoints.
2 The other thing, I was struck by the
3
simplicity of the color coding that Cato mentioned.
4
While it has a downside and you can get into the
5
good and bad foods, I think the general public
6
would be more inclined to relate to that that as
7
opposed to getting hung up on numbers.
So, a plea
8
would be that if you try to quantify these sorts of
9
things and use a color code, keep it simple. Even
10
though there are some downsides to that, I would
11
encourage your work groups to look at that.
12 DR. RULIS:
Great, thank you.
13 DR. SHINE:
What I heard, particularly
14
from Xavier's comments, is that the color code
15
might have more to do with caloric density, for
16
example, than value judgments about bad and good,
17
and I think that is an important observation.
18 DR. PI-SUNYER:
There is actually quite a
19
lot of research on color coding, mostly out of Les
20
Epstein's lab at University of Syracuse. He has
21
done a lot of work on treating children for weight
22
loss and weight maintenance using color codes for
119
1
children, which for them is a lot simpler, and he
2
has gotten very nice results doing that. That is
3
an observation.
4 DR. SHINE:
This is a little bit of a
5
digression because this is a question Allen Roses
6
wanted to raise before. Allen
wanted to show just
7
one slide which was relevant to our earlier
8
discussion about interventions.
Why don't we just
9
take a look at that? Cecil, why
don't you ask your
10
question while they are setting up?
11 DR. PICKETT:
It was really a comment in
12
response to Dr. Orloff's presentation.
I really
13
just wanted to compliment the FDA for being really
14
proactive in revisiting the guidelines for the
15
treatment of obesity. I strongly
encourage the FDA
16
to really have the advisory committee in late 2004.
17 I
think that would be very, very useful, very
18
beneficial.
19 In terms of unanswered questions that Dr.
20
Orloff discussed, particularly with head-to-head
21
comparisons of approved agents for the treatment of
22
obesity, I think that will happen.
I think that
120
1
will happen in Phase IV studies that will occur,
2
and market dynamics will basically drive those
3
studies in terms of companies trying to position
4
obesity therapy out in the marketplace, as we are
5
seeing in other areas in terms of hyperlipidemia
6
for example where these comparative studies have
7
been done.
8 Combination therapy is a little more
9
difficult because of intellectual property issues
10
with approved agents but, again, if in fact
11
combination therapy provides benefit to patients,
12
then I think companies, because of commercial need,
13
will find a way to develop combination therapy for
14
the treatment of obesity. So,
yes, those are
15
unanswered but I see those areas as less of an
16
issue.
17 DR. SHINE: To
what extent, Cecil, do you
18
think, given that the NIH is also going to make a
19
major move on obesity, they ought to be into, for
20
example, combination trials and things of this sort
21
as opposed to the industry? What
do you think the
22
balance is in terms of what the NIH ought to be
121
1
doing and what industry can do in this area?
2 DR. PICKETT:
It certainly could provide a
3
benefit here, but generally the size of the trial
4
designs that would have to be done to show efficacy
5
and safety, particularly of combination therapy, I
6
think is really something the industry should
7
undertake.
8 DR. SHINE:
And, in terms of the advisory
9
committee that is being constituted for later this
10
year, what do you think is the principal kind of
11
issue that it should be taking on in terms of
12
revisiting the situation?
13 DR. PICKETT:
Well, this is obesity
14
therapy and drug development has had a real jaded
15
past, but I think it is important for the advisory
16
committee to take a look at the size of the trials
17
and the efficacy endpoints that would be required,
18
other than just simply weight loss, and also the
19
size of the studies to show long-term safety. I
20
think this is something that is still a bit of an
21 issue,
at least in my own company, as we think
22
about development of drugs to treat obesity. So, I
122
1
think that would be extremely useful.
2 DR. SHINE:
Thank you very much. Allen,
3
you wanted to show us a slide?
4 DR. ROSES: As
a scientist, when I
5
mentioned "omics" I saw a lot of blank faces and I
6
think a picture is worth 10,000 words and I just
7
wanted to show you some the ways that are currently
8
being used.
9 [Slide]
10 It doesn't really matter what this
11
molecule is but it is a molecule that is being
12
tested for weight loss in obese people.
If you
13
look at this, the people on the capsules are in red
14
over the course of the trial and the people on the
15
placebo are in blue. If you look
to your far right
16
where the weight loss is between 4 and 10 percent
17
of weight during this particular interval, there is
18 a
spike of people who are hyper-responders.
19 Interestingly, if you look to the far
20
left, people who took this drug during the course
21
of the trial gained less weight than people who
22
were on placebo. So, we also
have that kind of
123
1
information.
2 But in this case it was fairly simple to
3
find several polymorphisms, ABO blood group. In
4
this case it is not the ABO blood group, but it is
5 a
1, 1 or 1,2 or 2,2 polymorphisms from 3 different
6
genes that have something to do with weight loss
7
that segregate or enrich these folks.
So, the 2,2
8
patients are the hyper-responders; the 1,1 would be
9
those that wouldn't seem to have that much of a
10
response; and the 1,2 are those in the middle.
11 From the point of view of taking something
12
with a partial response, if you want to go into
13
Phase III trials, partial response requires larger
14
trials. But, in fact, if you can
segment the
15
population, and I think we are going to hear about
16
that later this afternoon from Dr. Temple, but if
17
you can segment the population and withdraw from
18
your selection the non-responding population or the
19
1,1, you are more likely to get an effect with
20
fewer people, a shorter period of time and a better
21
proof of principle that the molecule has true
22
efficacy. This is
"omics." It is called
124
1
pharmacogenomics and it is being applied now.
2 DR. SHINE:
Thank you, Allen. I didn't
3
know that Allen had that slide but it is relevant
4
to this notion that, given the necessity for folks
5
to be on these agents I suspect for a long time
6
given the way this chronic problem persists,
7
finding ways to stratify the population, on
8
whatever basis, would be important in terms of
9
minimizing the risks. Dan?
10 DR. CASCIANO:
I am just curious, is there
11 a
phenotype associated with that hyper-responder?
12 DR. ROSES: Not
that we could see
13
prospectively that there was a genotype.
14 DR. SHINE:
Xavier?
15 DR. PI-SUNYER:
I would like to comment on
16
that further and also on something Dr. Orloff said,
17
and that has to do with weight gain with drugs. I
18
really do believe that that is another problem that
19
the FDA really needs to look at very closely.
20
These very active, very good antipsychotic drugs
21
are having an enormous impact on weight in a number
22
of people. We see people who
have gained 60, 70
125
1
pounds on the psychotropic drugs.
They don't want
2
to go off them because their mental state is so
3
much improved by these drugs.
Maybe the kind of
4
research that Dr. Roses just commented on would be
5
important in that, that is to say, not the
6
responders for the efficacy of the drug but the
7
responders for the side effects of the drug which
8
are really pretty bad for some people.
I think
9
certainly in the label of these drugs a lot more
10
should be said about the side effect of weight
11
gain. I think every psychiatrist
today should have
12 a
scale in his office and weigh his or her patients
13
every time they come. You know,
if you talk to
14
psychiatrists about that, there isn't a scale in a
15
psychiatrist's office, I don't think anywhere.
16 DR. SHINE:
Your comment is very well
17
taken and I think very important.
Dr. Orloff, do
18
you want to make any response to that?
No? But
19
you are, in fact, going to be looking closely at
20
this issue in terms of the identification and
21
potential labeling in terms of drugs which are
22
associated with weight gain. Any
other comments on
126
1
any of the material that we have heard this morning
2
on obesity? John?
3 DR. THOMAS:
Just a quick note, certainly
4
the troika of obesity is exercise, diet and drugs
5
and my concern, as we go forward, would be in the
6
drug development area and that we don't lead people
7
into a false expectation. I
mean, finding a good
8
anti-obesity drug would be like finding the "Holy
9
Grail." In fact, until that
is found we want to
10
make sure that there is continued emphasis on
11
exercise and diet. Everyone
would love to have
12
that magic bullet or pill so they can still be a
13
couch potato but that is not going to happen for a
14
while.
15 DR. SHINE: Let
me just summarize. I
16
think we have seen very good evidence that obesity
17
is both a national and an international problem.
18
You might be interested to know that if you go to
19
Beijing today, close to 20 percent of the residents
20
in Beijing are obese.
Twenty-five years ago you
21
never saw obese Chinese. But
gravitating to the
22
city, losing the capacity to bicycle or work in
127
1
agriculture, the availability of fast foods, the
2
movement from a diet that was rich in rice and in
3
some places fish to one that has shown
4
extraordinary increases in pork consumption and
5
beef consumption has produced a significant number
6
of obese Chinese.
7 We, in this country, are suffering from
8
all of the effects of being able to afford to eat
9
as much as we want and to travel in vehicles
10
everywhere we want to go, and we have made some
11
very poor investments. On the
one hand, as John
12
pointed out, we have invested in vending machines
13
to produce more income for schools at the same time
14
that we did away with gym and a number of other
15
physical exercise programs.
16 What we have heard I think is that the FDA
17
has a critically important role in dealing with a
18
multifactorial problem which the FDA in itself is
19
not going to solve. But given
the capacity to
20
understand the data, to refine that and to provide
21
good information, the challenge, it seems to me, is
22
going to be providing that information to the right
128
1
people at the right time so that they can make the
2
right kinds of decisions. That
includes not just
3
the individual consumer but includes school boards,
4
PTAs, employers, health departments and others who
5
can use that information as part of a broader
6
activity, whether it is involved in color coding
7
calory density or something else.
8 So, I think the work that I have seen and
9
the report on "Calories Count" is very useful. I
10
do like the way in which the focus groups responded
11
to the various ways to present the information.
12
And I think that doing further trials and creating
13 a
sense that this is a long-term, continuous
14
quality improvement process will be the way in
15
which we can contribute the most to this activity.
16 I
want to thank the presenters this morning, and I
17
would encourage the members of the committee, if
18
you have additional thoughts as you digest--and
19
that is not meant to be a pun--digest this
20
information, please communicate it because I think
21
the staff would like to get as much input and
22
perspective as they can. If you
might come into
129
1
some restaurant in your travels that does an
2
exemplary approach to this, you might even let us
3
know. Thank you again.
4 We will go on to our presentation prior to
5
the public hearing. John
Marzilli, who is Deputy
6
Associate Commissioner for Regulatory Affairs, is
7
going to give us an update on the Office of
8
Regulatory Affairs peer review process.
John?
9 Update on Office of Regulatory Affairs
10 Peer Review Process
11 DR. MARZILLI:
Thank you, Dr. Shine and
12
members of the Board.
13 [Slide]
14 I am John Marzilli.
I am from the Office
15
of Regulatory Affairs, FDA's field organization,
16
and I am here this morning to give you an update of
17
our peer review of our pesticide and industrial
18
chemicals program.
19 [Slide]
20 With me today is Dr. John Specchio. Dr.
21
Specchio is ORA's senior science advisor and
22
professor of food science at Montclair State
130
1
University in Montclair, New Jersey.
2 [Slide]
3 Briefly, the objectives of our peer review
4
of the pesticides and industrial chemical program
5
in the Office of Regulatory Affairs are to assess
6
the quality of science across our organization,
7
from program planning with the Center for Food
8
Safety, our inspectional program, investigations,
9
laboratory analysis, regulatory actions taken by
10
the agency and across the entire system our quality
11
management system. Since we are
a science-based
12
regulatory agency, the quality management system is
13
an underpinning for all of the activities that we
14
undertake.
15 We are also taking the opportunity to look
16
at the adequacy of resources for this program in
17
particular, the skills and expertise of our staff
18
across the country, the technologies employed in
19
terms of laboratories, equipment information
20
systems utilized, as well as our organizational
21
infrastructure, with a keen view toward the mission
22
relevance of this program for the Office of
131
1
Regulatory Affairs in particular and the Food and
2
Drug Administration in its mission.
3 [Slide]
4 I would like you to focus your attention
5
on the last sentence here in FDA's mission
6
statement which we feel is an important aspect for
7
us in looking at this program.
It is helping the
8
public to get the accurate science-based
9 information that they need to use medicines
and
10
foods to improve their health.
One of the
11
underpinnings of this study is to make sure that
12
our information is accurate, cutting-edge science
13
and timely that we get to our constituents, to the
14
regulated industry and to the public as well.
15 [Slide]
16 The goal of the ORA peer review of the
17
pesticides program in particular has been to
18
establish a roadmap for future improvement of this
19
program and coordination of the pesticide program
20
across the ORA field organization.
Also, we want
21
to use this as a model for the future review of
22
additional programs, scientific programs within the
132
1
Office of Regulatory Affairs.
2 [Slide]
3 As I mentioned, our team is a
4
cross-section of the Office of Regulatory Affairs.
5
As the Deputy Associate Commissioner for Regulatory
6
Affairs, I serve as the executive sponsor for this
7
study and our future studies.
Our senior advisor
8
to the team is Susan Setterberg, our regional
9
director from the central region.
Co-chairing
10
along with John Specchio, our science advisor, is
11
Pat Schafer, Director of Compliance from New
12
Orleans. On the team we also
have Tom Gardine,
13
District Director from Philadelphia, Sandra
14
Whetstone, from our headquarters Office of
15
Enforcement, Michael Ellison, a national food
16
expert with our Division of Field Investigations.
17
We also have a representative of the import program
18
at FDA from the New England District, Frank
19
Mazzoni, and for administrative support, the New
20
Orleans District has been gracious enough to have
21
Barbara George working with us, and we thank the
22
District for her assistance.
133
1 [Slide]
2 Our team members from the lab side, as I
3
said and I will introduce him shortly, Dr.
4
Specchio, our senior science advisor for the
5
Division of Field Science is co-chair of the peer
6
review team. From our Division
of Field Science
7
here, in Rockville, we have Tom Savage, the Deputy
8
Director. We have two senior
science advisors,
9
Larry D'Hoostelaere and Charlie Parfitt
10
participating on the team, and representing our
11
Pacific Regional Office, is Tom Sidebottom,
12
assistant to the regional director, and formerly
13
from our Southwest Regional Office, who has since
14
retired, Meredith Grahn participated in some of the
15
work that we began a year ago in this endeavor. I
16
would like to acknowledge the work done by our
17
senior science advisor for the Office of Regulatory
18
Affairs who retired at the end of the calendar
19
year, Richard Baldwin, who was an active member of
20
this team and the original chair for this endeavor
21
before it was handed over to John Specchio and
22
myself at the beginning of the calendar year.
134
1 [Slide]
2 I would just like to share with you the
3
breadth of the Office of Regulatory Affairs. We
4
are 190 officers are located across the country.
5
We are present in 49 states, the Commonwealth of
6
Puerto Rico and the United States Virgin Islands.
7
The only state where the FDA doesn't happen to have
8
an office is the State of Wyoming, to which we can
9
easily commute from our other states.
But we are
10
present in 190 cities with 4,000 employees
11
nationwide.
12
[Slide]
13 The laboratory support for the work that
14
we do in the Office of Regulatory Affairs is
15
coordinated across 13 field laboratories for the
16
Office of Regulatory Affairs. We
have five
17
multi-purpose laboratories that are responsible for
18
the analysis of all FDA-regulated products, and
19
those laboratories are located in New York,
20
Atlanta, Jefferson, Arkansas, Los Angeles and
21
Seattle. These laboratories are
the underpinning
22
of our pesticide program as are our pesticide
135
1
servicing laboratories. With
have eight specialty
2
laboratories for which Kansas City, which is the
3
sixth laboratory that we included in the lab study
4
portion of the peer review, is our total diet
5
laboratory for the FDA, and that is co-located with
6
our Kansas City district office.
7 Now I would like to turn the podium over
8
to Dr. Specchio, our science advisor, to give you
9
an overview of the program.
10 DR. SPECCHIO:
Thank you, John, and thank
11
you to the FDA Science Board for inviting me.
12 [Slide]
13 One thing I have heard a lot today was a
14
lively discussion of obesity and, no matter which
15
way the FDA goes, more food, less food, high fat
16
food, low fat food, we are going to make sure that
17
it is free of pesticides and industrial chemicals.
18
So, hopefully, we will be on our path to doing
19
that.
20 Just a few other comments, as we all know,
21
ORA's program responsibilities are extremely
22
diverse and very inter-disciplinary.
In order to
136
1
approach the peer review process systematically,
2
our assessment really focused on the pesticide and
3
industrial chemicals program area.
4 [Slide]
5 The components of the peer review process
6
were designed and based upon design and
7
implementation of a data review process. I will
8
talk about that briefly in a short while. We also
9
looked at a comprehensive work product review
10
pesticide and industrial chemical data collected
11
for one entire fiscal year. We
also had structured
12
discussions with managers and technical experts.
13 We actually developed a database research
14
model and a list of questions of in-depth and
15
focused interviews based on our developed research
16
model. We conducted a thorough
examination of
17
major programs in food areas in the pesticide and
18
industrial chemicals total diet study and dioxant
19
programs.
20 [Slide]
21 To give you an overview of the complexity
22
of the research model that was developed by our
137
1
committee, there were eight district offices
2
visited by our site review teams.
There were six
3
ORA laboratories visited. John
mentioned the
4
location of most of those, and there were
5
additional 12 district offices included in the
6
study that were not in the site visits.
We wanted
7
to make sure that we included all the districts in
8
the ORA system. The district
offices and labs
9
visited were picked because of their past and
10
present involvement with the ORA pesticide and
11
industrial chemical program.
12 [Slide]
13 Our on-site visits included review of
14
many, many different things that are important to
15
the program. For instance, local
SOPs for domestic
16
and imported work. We looked at
sample collection
17
reports. We looked very, very
in-depth at the lab
18
worksheets. We looked at the
establishment
19
inspection reports; the compliance actions; the
20
data systems, including FACTS and OASIS, and for
21
those members who are not familiar with those
22
acronyms, the acronym of FACTS, which I had to
138
1
learn myself, was Field Accomplishment Compliance
2
Tracking System which captures all field
3
investigations, sample analytical results and
4
compliance records of FDA regulatory work. OASIS
5
stands for the Operational and Administrative
6
System for Import Support. All
import entries of
7
FDA-regulated products, nearly eight million this
8
year, are tracked through this system.
We looked
9
at all the current guidance for the pesticide and
10
industrial chemical program. We
looked at
11
historical documents and earlier studies, and we
12
interviewed managers and technical experts and
13
calculated their responses.
14 [Slide]
15 The database research model which was
16
developed by the team was utilized to validate the
17
on-site reviews. The team
reviewed over 357 work
18
products, as I outlined before.
We focused on all
19
violative work products and reviewed a portion of
20
non-violative as well. We
interviewed over 150
21
individuals, both in terms of management and lab
22
and science experts. We had
additional comments
139
1
received from all interested individuals. After
2
consolidation, the committee received nearly 100
3
pages of comments which were reviewed by the
4
committee as well.
5 The comments that we received had a
6
working knowledge of the pesticide and industrial
7
chemicals program and, again, the committee had
8
over 500 comments to review, to consider
9
established trends and to develop ideas for
10
improvement of the program.
11 [Slide]
12 As we progressed with the analysis of the
13
data five key areas emerged and we are still
14
working on this, and hopefully our recommendations
15
will center around these five key areas. One is
16
program management. We are
already seeing several
17
types of major recommendations that may be on the
18
horizon. In terms of program
management, some
19 possibilities
would be redesign of the work
20
distribution for the domestic and imported programs
21
around our laboratory system.
22 We spent a lot of time on laboratory
140
1
science issues. Some questions
we are looking at
2
are, are we using the right methods?
Are the
3
methods current? Is our
instrumentation adequate
4
for the job at hand? How does
FDA stand versus
5
other organizations in both our compliance and also
6
our lab science?
7 In terms of the productivity and
8
timeliness, are we checking the right foods? Are
9
we looking at imports differently than we are
10
looking at domestic foods? Are we
getting the
11
results back in a timely fashion so that we can
12
adequately protect the consumer?
Again, are we
13
looking at the type of foods versus the consumption
14
of the population, the proportion of foods that are
15
out there versus consumption patterns?
Are we
16
targeting the right types of foods?
17 Resources--we are looking at lab
18
capabilities. We are looking at
the lab
19
assignments and the personnel.
Again, we are
20
looking at do we have the right equipment? Do we
21
need to improve our training? Do
we need to look
22
at our budgets for this particular program too or
141
1
is the budget adequate? Of course,
compliance and
2
regulatory policies are going to play a major role
3
in looking at recommendations, whether or not we
4
need to address this and are the compliance and
5
regulatory policies effectively meeting the
6
original task our original task that it was
7
intended to do?
8 [Slide]
9 Finally, what are our next steps? Well,
10
we are in the process right now of finalizing the
11
review. The committee met
several seeks ago to
12
start writing the preliminary report and analyzing
13
most of the data. We are going
to meet again in a
14
few weeks to continue this process.
In addition to
15
that, we will design a timeline for execution of
16
proposed improvements. As John
mentioned, the
17
purpose of this is to propose improvements to the
18
pesticide and industrial chemical program. We will
19
look at a timeline and we will, hopefully, make
20
recommendations to the people who will move forward
21
with our recommendations. Our
final report will be
22
finished in the summer of 2004 and we intend to
142
1
make a full presentation to all of you, the Science
2
Board at the fall, 2004 meeting.
3 [Slide]
4 I am very, very pleased to tell you that
5
the committee was a very, very good committee. We
6
all worked very well together. I
think there was
7
great cooperation and I think we are on the right
8
track for looking at our proposed objective, which
9
was to look at the state of science in ORA of the
10
pesticide and chemical contaminants program. With
11
that, I will close and I thank you for your time.
12 DR. SHINE:
Thank you. Questions?
13
Observations? Yes, Jim?
14 DR. RIVIERE:
One observation, this
15
committee seems to have been made up primarily of
16
FDA people. Why weren't there
more external
17
people, especially for an assessment of the state
18
of the science? I am just
curious because most
19
other center reviews I have seen involve outside
20
people.
21 DR. ALDERSON:
Let me address this. This
22
part of this review, what they are reporting on, is
143
1
the first step of the process.
This is their
2
internal assessment. What they
are doing, what
3
will be presented to the Board is what is used to
4
go to the next steps for the external, and Dr.
5
Shine will appoint an outside committee this fall
6
and do the external review, just like we have done
7
with the other three Centers.
8 DR. SHINE:
Other observations? John?
9 DR. THOMAS: I
am just curious as to
10
whether you have added more chemicals or more
11
pesticides to your list of monitoring, and have you
12
changed any of the levels of residues?
13 DR. MARZILLI:
Only God can add to the
14
list of residues--God and the EPA--
15 [Laughter]
16 I think one thing that we have done, not
17
in terms of adding to the list of industrial
18
chemicals or pesticides that we are reviewing, is
19
we have brought more improved communications across
20
the field organization here, and I think life in
21
general has brought new challenges to our program
22
in terms of looking at residues and contaminants
144
1
that we weren't looking at a few years ago, such as
2
chloramphenicol residues and the dioxant work that
3
has been ramped up as well. You
have seen recent
4
coverage of that in the news.
Our organization has
5
ramped up to address these new challenges and
6
brought forward the equipment and the personnel and
7
gotten them trained, working in collaboration with
8
the Center for Food Safety and the Center for
9
Veterinary Medicine to get these programs added to
10
the existing programs in the field laboratories.
11
So, every day new challenges do face us.
12 DR. THOMAS: I
was just going to add has
13
there been a shift insofar as the analysis of, say,
14
border versus non-border?
15 DR. MARZILLI:
Actually, border is a funny
16
term to discuss when you are talking about the
17
import program because, for instance, for us right
18
now the major border location could be a city like
19
Louisville, Kentucky--and I just use that as an
20
example--where you would have plane loads of
21
produce from anywhere in the world where a plane
22
would touch down in Miami or Newark, New Jersey,
145
1
and the final destination would be Louisville or
2
any other city in the middle of the country, and
3
that becomes a border location for us and we have
4
to deploy investigators to get out there, collect
5
samples, have the samples shipped to the
6
laboratory, and we have a commitment to get timely
7
results out so that, especially in the case of
8
fresh produce, that product can be distributed into
9
the marketplace.
10 So, we have incredible new challenges as
11
an agency. Years ago our border
locations were our
12
major port cities along the east and west coast and
13
the border between our partners here in the North
14
American continent. But those
borders have moved
15
all over the United States now.
It has been a
16
tremendous challenge for us, but that has been the
17
underpinning of our program probably for the past
18
ten years to get timely results to the customers
19
that are waiting to get the produce to market.
20 DR. SHINE:
Clearly, food safety has been
21
one of the issues in dealing with terrorism. The
22
agency has ramped up its activities in this regard.
146
1
Is your analysis taking into consideration the
2
potential for materials to be introduced into the
3
food supply?
4 DR. MARZILLI:
Yes, sir, absolutely. As a
5
matter of fact, from the foundation of our
6
pesticides and industrial chemicals program we have
7
ramped up across eight of our field laboratories,
8
in the chemistry area, a food emergency response
9
network. The focus for those
efforts in terms of
10
methods and instrumentation has been collaborative
11
efforts with other government agencies, and we have
12
ramped from a food safety to a food security
13
program in our chemistry labs so that we can
14
provide this.
15 At some point in the near future I would
16
like to share with the Science Board the investment
17
that we are bringing on-line now with our mobile
18
laboratories that can be deployed anywhere in the
19
country. We have two that we
have designed now,
20
that have been designed under contract with the
21
Department of Defense and will have a biological
22
platform for handling bio agents, and the other
147
1
will have a chem platform for handling chem agents,
2
and they can be deployed anywhere in the country.
3 So, FDA has a hundred years of experience
4
that we are using as a foundation for this work
5 but, as you know, the challenges grow every day.
6 DR. SHINE:
Susan, did you want to make a
7
comment?
8 DR. HARLANDER:
Is this review process for
9
all of the Centers? Is this
something we will look
10
forward to at each meeting?
11 DR. ALDERSON:
It is an ongoing process
12
across the Centers and CBER, CDRH and CFSAN have
13
already undergone their reviews.
So, ORA was next
14
in line and that will be followed CDER.
15
DR. SHINE: Any other comments or
16
questions? Observations? We look forward to a
17
report in the fall and also for working out next
18
steps in terms of the external portion of this
19
review. Thank you very much.
20 Open
Public Hearing
21 We will now move to an open public
22
hearing. I have been instructed
to read the
148
1
following statement for the record:
Both the Food
2
and Drug Administration and the public believe in a
3
transparent process for information gathering and
4
decision-making. To ensure such
transparency at
5
the open public hearing session of the advisory
6
committee meeting, FDA believes that it is
7
important to understand the context of an
8
individual's presentation. For
this reason, FDA
9
encourages you, the open public hearing speaker, at
10
the beginning of your written or oral statement to
11 advise
the committee of any financial relationship
12
that you may have with any company or any group
13
that is likely to be impacted by the topic of this
14
meeting. For example, the
financial information
15
may include a company's or a group's payment of
16
your travel, lodging or other expenses in
17
connection with your attendance at the meeting.
18
Likewise, FDA encourages you at the beginning of
19
your statement to advise the committee if you do
20
not have any such financial relationships. If you
21
choose not to address this issue of financial
22
relationships at the beginning of your statement,
149
1
it will not preclude you from speaking.
2 We have three individuals who will be
3
making particulars. We also have
received a
4
written document from Robert Rheinhart on behalf of
5
the AIDS Vaccine Advocacy Coalition and the San
6
Francisco AIDS Foundation, speaking to the Critical
7
Path issue that we will be discussing this
8
afternoon. That statement will
be posted on the
9
web site and will be available for public scrutiny.
10
We have two statements that we know about with
11 regard to the obesity working group report and one
12
with regard to the Critical Path.
13 In regard to the obesity working group
14
report, we will ask Richard Atkinson, M.D.,
15
President of the American Obesity Association and
16
Past President of the North American Association
17
for the Study of Obesity and the American Society
18
for Clinical Nutrition, to make a statement. We
19
want to limit the statements to five minutes. The
20
committee may have questions and we would
21
appreciate it if the speakers would use the
22
microphone so that we can get their comments on
150
1
record. Is Dr. Atkinson
here? No? Well, we have
2 a
letter from Dr. Atkinson indicating that he was
3
going to be here. If there is no
objection, I
4
think that letter could be placed in the record as
5
well.
6 The next speaker was Arthur Frank, M.D.,
7
from the George Washington University Weight
8
Management program. Dr. Frank,
why don't you
9
proceed?
10 DR. FRANK: My
name is Arthur Frank. I am
11 a
physician and the Medical Director of the George
12
Washington University Weight Management Program in
13
Washington, D.C.
14 I would like to offer some additional
15
perspective or clinical perspective because I work
16
with patients who are trying to manage the disease
17
of obesity. It is a very challenging
world of
18
individual patients who are struggling with this
19
disease. I don't want to sound
skeptical; I don't
20
want to sound petulant, but perhaps I can offer a
21
perspective from the viewpoint of the individual
22
physician who is dealing with the individual
151
1
patient, two people sitting in a room, talking
2
about obesity. I want to make
some observations; I
3
want to raise some questions. I
don't think I have
4
very many answers and I don't think any of us have
5
very many answers.
6 What is noteworthy about the disease of
7
obesity is obvious to us all, that it is a very
8
difficult disease; it is very difficult to control.
9
Our efforts, for the most part, have been very
10
unsuccessful but I would also like to emphasize the
11
enormous amount of effort that is devoted to this
12
task by patients. These are
individual patients;
13
these are good people. They work
very hard to
14
manage this disease, and they want desperately to
15
succeed; they care very much.
They know much of
16
what they need to do. They know
they have to eat
17
less and they know they have to cut down their
18
calories. They are not
uninformed. They may be
19
inconsistently informed; they may be informed with
20
bad information but they are not uninformed. They
21
are not careless; they are not indifferent; and
22
they are not defiant people.
152
1 The therapeutic mandate seems very simple,
2
eat less and exercise more. This
simple and really
3
very valid instruction has been, for the most part,
4 a
monumental failure--eat less and exercise more.
5
Patients do know they need to do this.
This
6
information they already have.
They understand it,
7
but it seems extraordinarily difficult for them to
8
implement this simple set of rules, particularly as
9 a
lifelong effort. I think it is fair to
ask why
10
is it so difficult to do what seems so simple--all
11
you have to do is just eat less.
Perhaps it is
12
because it assumes that eating is merely a matter
13
of choice when much of what we know about eating
14
suggests that the control of eating behavior is
15
much more involved than just willful behavior.
16
Much of eating is a highly regulated phenomenon
17
with an elaborate set of complex neurochemical and
18
peripheral signaling systems which tenaciously
19
control a survival-dependent biological function.
20 I
have to emphasize this is a survival-dependent
21
biological function and it is very hard to control,
22
and the idea that someone is going willfully to
153
1
override and superimpose on that mechanism may, in
2
fact, be a relatively naive kind of assumption.
3 When you talk to patients it does not
take
4
long to appreciate the intensity and the sincerity
5
of their effort, but it soon appears that telling
6
an obese patient to eat less is like telling a
7
depressed person to just pull yourself together, or
8
telling an asthmatic just breathe easy.
If I could
9
pull myself together I wouldn't be depressed. If I
10
could breathe easy I wouldn't have asthma. And,
11
the obese patient says if I could eat less I
12 wouldn't be obese.
13 It appears to be as difficult to control
14
eating in a sustained way, or may be as difficult
15
to control eating in a sustained way as it is to
16
control sleeping or breathing. I
have had the
17
draft of the obesity working group report, and I
18
looked at the report and many of the
19
recommendations of the working group deal with
20
efforts to change behavior, and I am very
21
respectful of that. Yes, we must
change behavior.
22 We must do this in a sophisticated way. We must
154
1
simplify that task as much as possible, and we must
2
understand how to do that as effectively as
3
possible. But from the
perspective of a physician
4
who is trying to help the patients, these
5
recommendations create a sense of merely shifting
6
the deck chairs on the deck of the Titanic,
7
particularly with severely overweight patients.
8
Severely overweight patients, they do know they
9
need to eat less.
10 Do we want to spend our resources in a
11
fight over which micronutrients to consume or how
12
to reword the label, or might we be better devoting
13
our resources in trying to understand how we can
14
modify neuropeptide Y? Some of
the recommendations
15
tend to reinforce the sense of blaming the victim.
16
If you could simply control yourself you would be
17
better, and we will try to help you control
18
yourself but you must control yourself, and you and
19 I
both know that the problem is more complicated,
20
that eating is a regulated phenomenon.
Yes, it is
21
important that we understand behavior; it is
22
important that we teach our children: it is
155
1
important that we know how to count calories; it is
2
necessary that we change labels and we change
3
diets. But these will not have
any effect on
4
changing the basic dysfunctional regulatory system,
5
which is probably a neurochemical regulatory
6
system.
7 Perhaps obesity is mostly a neurochemical
8
disease and, like diabetes of hyperlipemia, is not
9
primarily a nutritional disease.
Even if we could
10
change eating behavior, we would still have to
11
change neurochemistry. My
suggestion is not that
12
we avoid or neglect changing behavior but that
13
somewhere greater emphasis be placed on the fifth
14
item in the set of recommendations, what is labeled
15
therapeutics, and what I and my patients need is a
16
better understanding of the regulatory system and
17
appreciation of the complexity and the redundancy
18 of the regulatory system, a better understanding of
19
whether it is possible that changes in behavior can
20
affect the regulatory system, perhaps more
21
pharmacotherapy, more sophisticated ways of
22
modifying the regulatory system.
156
1 Perhaps obesity is like diabetes. Its
2
management does depend upon controlling diet and
3
exercise but its best control derives from
4
understanding and manipulating the hormonal
5
regulation of the biochemistry of blood sugar. We
6
will not solve the problem merely by telling people
7
that they must change their behavior when they may
8
be at the limit of what they can willfully change.
9
There is just so much that they can willfully
10
change, particularly in a sustainable way. We
11
might do better if we can understand the regulatory
12
system a little more effectively and if we can
13
develop ways of affecting that system, perhaps
14
through pharmacotherapy, and it might make more
15
successful the mandate that you all have to do is
16
to eat less and exercise more.
17 DR. SHINE:
Thank you, sir. Anyone want
18
to comment with regard to this statement?
19 DR. CRAWFORD:
Just an aside, I thought
20
when he was talking about the basically
21
dysfunctional regulatory system he was talking
22
about something other than what he was, and I am
157
1
relieved to know it wasn't the FDA!
2 DR. SHINE: I
guess my response is that by
3
no means do I disagree with the notion that there
4
are a group of individuals who have enormous
5
clinical problems and clearly the agency has
6
identified the notion of therapeutics as an
7
important goal, and neurochemistry is an essential
8
portion of that, and that problem is not going to
9
be solved solely by the FDA.
That is going to
10
require a concerted effort by both basic and
11
applied scientists to deal with it.
12 I think the question that we have as an
13
agency concerned with the health of our public is
14
what are the most cost-effective ways in order to
15
create a healthier society and there, sir, I would
16
argue that creating the proper environment, just as
17
we have done to reduce smoking of about half of
18
adults when I was growing up to about a quarter of
19
adults right now so that we don't end up with lung
20
cancer, is at least as important as the efforts to
21
develop neuropharmacologic agents.
The emphasis I
22
think on information, behavior and environment is
158
1
predicated on the concept that we would like to see
2
whether we can't minimize the number of patients
3
that come to you in your practice, but we know
4
damned well you are going to still have plenty of
5
patients in the foreseeable future.
6 We will move next to a statement by
7
Elizabeth Jacobson. Elizabeth
Jacobson, Ph.D. is
8
Executive Vice President for Technology and
9
Regulatory Affairs at AdvaMed in Washington, and
10
she is going to make certain comments on the
11
Critical Path. We have not
discussed that but the
12
committee has read the material and, hopefully, is
13
knowledgeable about this so we look forward to your
14 comments.
15 DR. JACOBSON:
Good morning. I am hear
16
representing the Advanced Medical Technology
17
Association, or AdvaMed. AdvaMed
is the world's
18
largest association of manufacturers of medical
19
devices, diagnostic products and medical
20
information systems. We have
more than 1,100
21
members who manufacture nearly 90 percent of the
22
devices and diagnostics that are purchased annually
159
1
in the United States, and about 50 percent of the
2
products that are purchased annually around the
3
world.
4 Although we are the largest device trade
5
association, our members range from the largest to
6 the
smallest companies. Nearly 70 percent
of our
7
members are small businesses, with fewer than 30
8
million dollars annually in sales.
9 We would like to thank you for the
10
opportunity to comment on FDA's white paper on the
11
Critical Path. Our bottom line
is that we support
12
FDA's efforts to identify and address issues that
13
prevent innovative products from reaching patients
14
and we have four general points that we would like
15
to make about the paper that you are all going to
16
be talking about this afternoon.
17 First, the paper's underlying assumption
18
about device innovation is not really supported by
19
the data on device submissions.
The paper states
20
in several places that its findings of stagnation
21
apply to drugs, biological products and medical
22
devices. But the illustration
that there is
160
1
stagnation in the industry is really based on
2
metrics for devices and biologics over the past ten
3
years and, in fact, there is a graph that shows
4
that there has been a steady ten-year decline in
5
submissions for these products.
But FDA's own
6
statistics don't demonstrate a similar decline in
7
the number of medical device submissions over the
8
same time period and, in fact, there has been a
9
statistically significant increase in both PMA and
10
IDE submissions over the past ten years, although
11
for devices the numbers do bounce around more than
12
they do for drugs and biologics.
13 We think this is important because in any
14
effort to identify opportunities to move products
15
along in their development it is really critical to
16
understand the key differences between the
17
industries. These differences
may be important in
18
identifying solution strategies.
Moreover, it may
19
be that some device-specific regulatory approaches
20
could be useful in addressing concerns about the
21
developmental process for pharmaceuticals, and the
22
agency and the pharmaceutical industry may, in
161
1
fact, want to look at that.
2 Most important to the device industry are
3
the areas where improvements could further
4
facilitate the movement, obviously, of innovative
5
devices to the patient's bedside, and we are
6
interested in identifying and addressing these
7
areas with the agency and other players, and we are
8
looking forward to the discussion this afternoon,
9
especially to hear more about the proposed Critical
10
Path opportunities list.
11
Our second point is that the
unique
12
characteristics of device innovation are really key
13
to understanding Critical Path opportunities for
14
devices, and I would like to go into those a little
15
bit. Device development has been
characterized as
16 a
continuous iterative process. It is
17
characterized by constant product changes made in
18
response to user needs and preferences, and it
19
really distinguishes device innovation from
20
pharmaceutical development.
21 There is also a learning curve associated
22
with practitioner use of device technologies that
162
1
may be longer than that with drugs.
Importantly,
2
innovations in devices are not restricted to the
3
premarket phase of their development.
Instead,
4
actual use of devices by practitioners in the
5
clinical setting typically spurs additional
6
refinements and improvements.
So, the clinical use
7
of new devices stimulates feedback from medical
8
practitioners, device redesign use, more feedback
9
and, in fact, also leads to ides for breakthrough
10
technologies as well.
11 In addition, medical device refinements
12
often result from advances in other industries, in
13
material sciences and bioengineering, molecular
14
biology and information system, just to mention
15
some. A lot of technologies are
adaptations from
16
other fields and many were developed through the
17
interdisciplinary work of clinicians, physicists,
18
biologists, engineers, computer scientists and
19
others. So, this
interdisciplinary character also
20
contributes to the evolutionary nature of device
21
development and it lends a degree of
22
unpredictability to the process.
163
1 Our third point is that the principles of
2
device regulation to date have been based on these
3
unique characteristics of devices and device
4
innovation. Congress recognized
the broad
5
diversity of devices and realized that the
6
one-size-fits-all of drug regulation probably
7
wouldn't work for devices. Accordingly,
in 1976,
8
Congress based the device regulatory scheme on the
9
degree of risk, with the lowest risk devices
10
subject to only basic requirements and the highest
11
risk devices subject to more extensive premarket
12
review.
13 More recently, in '97, the Food and Drug
14
Modernization Act codified certain medical device
15
policies that enabled more streamlined review of
16
devices, things like exemption for the many of the
17
lowest products, early collaboration meetings,
18
collaborative review processes and other
19
provisions.
20 Perhaps most relevant to a discussion of
21
Critical Path FDMA also contained a least
22
burdensome provision that required FDA to identify
164
1
and utilize the most expedient means to market for
2
novel technologies that would still assure device
3
safety and effectiveness and FDA has found a lot of
4 opportunities to apply least burdensome methods to
5
device review, and I am sure you will hear some of
6
them from the device speakers this afternoon,
7
including the use of Bayesian statistics to reduce
8
the number of subjects required in clinical
9
studies; reliance on valid non-U.S. data in lieu of
10
clinical data from U.S. populations; use of
11
previously submitted information; data from
12
published literature; and the use of objective
13
performance criteria.
14 Among the reasons for this congressional
15
focus on risk-based regulation was recognition that
16
the device industry differs in significant ways
17
from that of the pharmaceutical industry. The vast
18
majority of device companies is small.
The life
19
cycle of the average device is about 18 months,
20
after which the device is replaced by newer
21
technology. Devices typically do
not benefit from
22
extended patent protection, and medical devices
165
1
encompass a huge range of products, from bandages
2
and wheelchairs to in vitro diagnostic products, to
3
implanted defibrillators and artificial joints and
4
drug-coated stents.
5 For these reasons, there was a need for a
6
regulatory system that could be flexible as well as
7
efficient, and since '76 FDA has looked for ways to
8
improve its pre- and postmarket handling of medical
9 devices.
Most recently, it converted its GMP
10
regulation, good manufacturing practices, to a
11
quality system approach that was closely harmonized
12
to the internationally accepted ISO quality systems
13
standard, and it required the industry to apply
14
risk management principles to medical device
15
design.
16 Our final point is that there is always
17
room for improvement in any endeavor, including the
18
regulation of medical devices.
FDA's white paper
19
proposes that FDA, industry and academia work
20
together to ensure that products move along the
21
Critical Path of development and it proposes that
22
the development of a Critical Path opportunities
166
1
list.
2 We certainly support the desired endpoint
3
here and we see the potential for improvement in
4
processes issues and in science issues.
For
5
example, we think the science used in device review
6
could benefit from a critical look at what are the
7
scientific expectations for new and novel
8
technologies, and how can we feed those into an
9
appropriate clinical trial design.
So, we have had
10
some early thoughts on potential candidates for
11
this critical opportunities list but before
12
proceeding any further to refine our list we really
13
want to hear some more specifics in today's
14
discussion about how is the agency thinking about
15
this and how does it intend to proceed with this
16
effort. So, we are very much
looking forward to
17
this afternoon. Thank you very
much.
18 DR. SHINE:
Thank you very much for your
19
comments. I know, Janet, you are
going to be
20
talking about the Critical Path this afternoon. Do
21
you want to make any observations about this
22
question whether devices are following the same
167
1 track as biologics and drugs?
2 DR. WOODCOCK:
I put a slide in the
3
presentation on the PMA submissions so that will be
4
available and we can discuss that.
I think we did
5
acknowledge in the report, although because we had
6
so many things to talk about it was brief, that
7
devices, as Liz just said, have a more iterative
8
development process and you are going back and
9
forth from the clinic with many improvements and
10
engineering technology allows that to happen versus
11
drug development technology.
12 So, it is different and we believe that we
13
need to develop somewhat different approaches, and
14 I
think we all agree with that.
15 DR. SHINE:
Thank you. Any other public
16
testimony? I have a special
admiration for the
17
device industry. As a young
cardiologist I took
18
care of a patient who had aortic insufficiency and
19
had a bird-cage valve put in, a so-called Hoofnagel
20 valve.
Probably no one in the room here has ever
21
seen one. It was a valve that
was placed in the
22
aorta, just behind the trachea, and it made a
168
1
clicking sound every time it opened and closed. If
2
the patient opened his or her mouth--they were
3
mostly males--you could hear it clicking across the
4
room. As long as the patient was
in sinus rhythm,
5 a
regular rhythm, the patients could tolerate it.
6
This young man developed atrial fibrillation and
7
the randomly irregular sound drove him to suicide.
8
He was not the only case. But
the incremental
9
improvement in heart valves and technique which
10
eventually allowed one to put in heart valves which
11
are essentially completely silent so even the
12
patient can't hear them is one of the examples of
13
devices which came a long way.
So, I have a
14
special respect for incremental changes in devices,
15
and we will talk more about that this afternoon.
16 Are there any other items? Janet, could
17
we start at one o'clock? Would
that work for you?
18
We do have some speakers who aren't here yet and I
19
am afraid if we start much before that they may not
20
be here. But we will reconvene
at 1:00. Do any
21
members of the committee have any other comments
22
they want to make before we adjourn for lunch? If
169
1
not, why don't we recess and we will reconvene at
2
one o'clock?
3 [Whereupon, at 11:30 a.m., the proceedings
4
were recessed for lunch, to reconvene at 1:00 p.m.]
170
1 A F T E R N O O N
P R O C E E D I N G S
2 DR. SHINE: In
open session one of our
3
presenters lost so much weight that he became
4
invisible! He has now reappeared
in proper form
5
and we want to give him an opportunity to make a
6
statement. So, if we can
reconvene, at 12:55 we
7
will go ahead and we will be in open session. I
8
have given Dr. Atkinson a copy of the instructions
9
which I read so that for the record he has heard
10
that. Richard Atkinson is the
Director of the
11
Obesity Institute at MedStar Research Institute in
12
Washington, and he is going to take about five
13
minutes to make a presentation as part of our
14
discussion on obesity.
15 DR. ATKINSON:
Thank you very much, Dr.
16
Shine. Sorry, I had another
appointment and was
17
confused about the time I was supposed to be here.
18 First the financial disclosure, I have
19
consulted at least for pretty much every company
20
that makes an obesity drug and a whole bunch of
21
them that don't and would like to, so I am
22
completely conflicted and take everything I say
171
1
with a grain of salt. So, that
is my conflict of
2
interest--
3 DR. SHINE: No,
no, not with a grain of
4
salt--
5 [Laughter]
6 DR. ATKINSON:
I am currently the
7
President of the American Obesity Association, and
8
Past President of the North American Association
9
for the Study of Obesity and the American Society
10
of Clinical Nutrition, but I am not really here
11
today to represent them; I am here as a physician
12
who has treated obese people for more than 30
13
years. I started my first weight
reduction program
14
in 1973.
15 I am a little bit concerned about the
16
direction that I see the federal government, or at
17
least the emphasis that I see the federal
18
government is taking, and I want to just point out
19
some things, and I will try to be brief because I
20
understand Dr. Frank also said the same thing this
21
morning. But just to talk about
obesity as a
22
chronic disease, multiple causes and obviously all
172
1
of them have excess adipose tissue, but the
2
critical point I think is that the biochemistry of
3
the body of obese people is different from that of
4
lean people. If an obese person
loses weight down
5
to that of a lean person their biochemistry is
6
still not that of a lean person.
They are poised
7
to regain weight and they have terrible hunger. If
8
you have not tried going on about a third of your
9
calories for about two weeks and see if you can do
10
that for the rest of your life, and that is,
11
indeed, what obese people do, they have those
12
protective mechanisms.
13 That is why I think the standard treatment
14
of diet, exercise and behavior modification has not
15
worked. We have decades of
research to show that
16
it really does not work well.
That is not to say
17 that
all the things that are coming out are not
18
really important. We do need to
look at what are
19
kids are eating and get at least milk into the
20
vending machines. We need to
increase the
21
activity, and so forth. So, all
those things are
22
very important but terribly important for skinny
173
1
people too; it is not just for fat people and it is
2
not enough.
3 I will point out that the treatment of
4
choice for new onset type 2 diabetes and new onset
5
of essential hypertension in the absence of any
6
obvious renal disease is diet and exercise and both
7
of those work very well. In 80
percent-plus people
8
the blood pressure comes down; their blood sugar
9
comes down. Why aren't the first
words out of the
10
mouth of the physician, okay, I am going to put you
11
on a diet for a year and then we will see if we can
12
use drugs?
13 So, I realize that the drugs that we have
14
for obesity now are not terribly effective but we
15
have not been very creative in using those drugs in
16
different ways, and we have also not, I think,
17
spent the kind of money that needs to be spent from
18
the period from 1973 to 1994 so there were no
19
obesity drugs released on the market.
20 So, I come here to point that out. I
21
think it is great to do these behavior things.
22
These preventive measures are very important, but I
174
1
would hate to see such huge amounts of money go to
2
those and not to developing drugs to treat the
3
disease of obesity. I think I
will stop there.
4
Thank you very much.
5 DR. SHINE:
Thank you very much for your
6
comments. Does anyone want to
make any comment
7
with regard to the presentation?
I think, as with
8
Dr. Frank, none of us denies the critical
9 importance of dealing
with those people who are
10
particularly morbidly obese and drug development
11
and we hope that is going to happen as a
12
consequence of concerted efforts, not only by FDA
13
in terms of what its role is, but the NIH which has
14
increasingly identified obesity as a public health
15
target. I think we agree that a
balanced approach
16
to this is likely to be required.
Thank you, Dr.
17
Atkinson.
18 Let's move back to our agenda. We are
19
going to be taken down the Critical Path by our
20
next speaker. Janet Woodcock I
think is known to
21
many of you. She is the Acting
Deputy Commissioner
22
for Operations here, at the FDA, and we are looking
175
1
forward to hearing from her about the concept of
2
the Critical Path. While they
are tuning up, we
3
are going to welcome Gail Cassell who is going to
4
speak to us but who is also a member of this
5
committee, and we acknowledge that she is now
6
present.
7 Introduction to Critical Path
8 DR. WOODCOCK:
Thank you. Good afternoon.
9 [Slide]
10 I am going to set up this afternoon's
11
presentations on the FDA's Critical Path initiative
12
and I would like to say up front that we are just
13
starting this initiative so one of the
14
purposes--you have probably heard the questions we
15
have for the Science Board--is to get some input.
16
This is a very complicated subject and this is
17
really the first time it has been framed in this
18
manner, to our knowledge, by any agency or
19
organization.
20 As we have been rolling this out and
21
discussing with various groups, it has become clear
22
that people have different parts of the elephant.
176
1
This is a very complex issue and there are many
2
different disciplines involved, many different
3
perspectives. So, what we have
set up this
4
afternoon is to hear from a variety of different
5
perspectives, the basic science perspective, the
6
clinical perspective, the whole range of
7
development perspectives. So, we
don't expect a
8
definitive answer today from the Board.
We would
9
like some provocative questions and help in moving
10
this initiative along, but we are just in the
11
beginning stages. So, let me
just get started, and
12 I
am going to leave some time after my presentation
13
for questions from the Board.
14 [Slide]
15 The real challenge, if you get down to the
16
basics, and you have been talking about this, this
17
morning to some extent, is that there are multiple
18
serious diseases that afflict our population and
19
really need better treatments.
Everything from
20
autism to obesity that has been discussed today,
21 multiple sclerosis--just a huge number of diseases
22
that people are suffering from, schizophrenia,
177
1
stroke, and so on--increasing incidence of chronic
2
disease.
3 [Slide]
4 Prevention is seems to be even more of a
5
challenge. Obviously, there is a
greater
6
opportunity in prevention but primary prevention
7
still eludes us in many areas and early detection
8
and intervention, thus prevention of progression,
9
is also something that raises huge challenges for
10
us.
11 [Slide]
12 From the perspective of society--and FDA
13
doesn't have the total role but for society there
14 is
really an urgency of need for timely development
15
of new technologies driven by the aging of our
16
population, mounting burden of illnesses being
17
experienced and the benefits of prevention versus
18
secondary intervention that we are all going to
19
have to be coping with, say, with the obesity
20
epidemic that was discussed this morning.
21 But, at the same time, society requires a
22
high degree of certainty related to performance of
178
1
these medical technologies. In
other words, people
2
want to have things that work.
They want
3
interventions that do prevent obesity or treat
4
obesity. They want certainty
about safety. Maybe
5
these interventions, these technologies are not
6
going to be 100 percent safe but we want to know
7
what their safety profile is like before we turn
8
them loose on the population.
Also, society is
9
facing tremendous challenges, as you all know, with
10
providing access to technologies that are
11
available, and then affordability of those
12
technologies that are going to be developed. This
13
is kind of the framework under which we are looking
14
at the product development process.
15 [Slide]
16 There is tremendous optimism though that
17
we will surmount these challenges.
We have
18
sequenced human genome and other animal genomes.
19
We have new genomic and proteomic technologies to
20
really probe body systems at the cellular and
21
subcellular level. With have
systems biology to
22
look at the system, the organ system or cellular
179
1
system. There are advances in
medical imaging that
2
are just stunning. And, we have
some presentations
3
recently internally at the FDA about how this
4
imaging technology can be used in development.
5
Nanotechnology in many of our areas has tremendous
6
promise for delivery of devices, drugs, and so
7
forth; tissue engineering; and then the advances in
8
drug discovery, way on the left side of the
9
pipeline, that have really increased the number of
10
drug candidates that are available and out there
11
for evaluation.
12 [Slide]
13 So, there has been a tremendous amount of
14
optimism. Fueling this have been
ten-year
15
investment trends and biomedical research. The NIH
16
funding has been doubled over the past five years.
17
Pharmaceutical R&D investment is following the same
18
trajectory as far as rate of rise, and there have
19
been major investments in biotechnology of all
20 kinds
which are not as easily trackable but they
21
have been significant.
22 [Slide]
180
1 This shows ten-year trends in biomedical
2
research spending, reflected by just U.S.
3
pharmaceutical R&D spending and total NIH budget
4
spending increases. You can see
these basically
5
parallel one another.
6 [Slide]
7 As a result of all this new knowledge, new
8 basic science breakthroughs, as well as the
R&D
9
investment there has been expected a matching
10
acceleration of product development and really a
11
flood of new products coming through the pipeline
12
that we hope to see.
13
[Slide]
14 These are the trends for major drug and
15
biologic product submissions to FDA over the
16
matching ten years. The numbers
do bounce around,
17
and so forth, but we see the reverse of the trend
18
that we are seeing in the investment side.
19 [Slide]
20 To get to Liz Jacobson's point that was
21
made this morning, here is the trend for device
22
premarket applications. The PMAs
are the most
181
1
complicated devices that require special
2
submissions with special data, and so forth. This,
3
was alluded to, is an upward trend but I wouldn't
4
say this is an explosion of new technology either.
5 As we said, devices are different. Even
6
these PMAs may represent incremental changes to
7
existing products, and so forth.
But we don't have
8
as good a handle, I think, on the device world and
9
it is clear that devices are more innovative right
10
now. There is more innovation
actually coming out
11
of the pipeline than there is for pharmaceuticals.
12
And we agree with that.
13 [Slide.]
14 Now, in addition to this picture that I
15
have just showed you, medical-product development
16
costs are escalating very rapidly.
The cost of
17
bringing a successful drug to market is estimated
18
between $0.8 billion and $1.7 billion per single
19
successful drug marketed. This
loads on the costs
20
of all the failures that we discuss in the report.
21
There are a lot of candidates that are taken into
22
clinical development and, perhaps, $500 million may
at 182
1
be spent and then they fail late in development.
2
This total cost to get a single successful product
3
loads in those failure costs.
4 These cost estimates are controversial.
5
Nevertheless, I think there is uniform agreement
6
that it is very expensive. It is
very expensive
7
and it is getting much more expensive.
This gets
8
back to the issue of availability, access and
9
affordability which is one of the societal
10
challenges as we develop new technologies for all
11
the diseases that our population is suffering.
12 In addition to this escalating costs, we
13
do see a higher failure rate in the pharmaceutical
14
world than traditionally, say, than a decade and a
15
half ago when more compounds that got into early
16
clinical trials made it successfully through the
17
development process.
18 It is hypothesized, at least, that this is
19
related to the fact that more difficult diseases,
20
in general, are being treated.
Easier diseases
21
such as certain infectious disease, certain acute
22
diseases, and so forth, are actually easier to
at 183
1
address than some of the chronic diseases, or
2
diseases such as schizophrenia, or whatever, where
3
we don't really understand the underlying etiology.
4 [Slide.]
5 So what is the diagnosis of this problem?
6
If we agree there seems to be a problem, what is
7
the diagnosis? Basically, the
central thesis of
8
our report, the stagnation report that we have put
9
out, is that we think investment and progress in
10 the
basic biomedical science has surpassed the
11
investment and progress in the medical-product
12
development process.
13 So, at the left side of the pipeline,
14
there is this tremendous explosion of basic
15
scientific knowledge. But I am
going to walk you
16
through, now, how we feel it hasn't been applied to
17
actually the process of getting these products
18
developed.
19 This process, which is called the
20
development process, is on the critical path to
21
getting these products to the patients.
We think
22
it is becoming a serious bottleneck to delivery of
at
184
1
new products as the basic science enables more
2
discovery and more candidates to be identified. We
3
are using both the evaluation tools and the
4
infrastructure of the last century to develop this
5
century's advances. I think,
from the FDA, we
6
think there is no doubt about this although it is
7
spotty and we certainly have advances in certain
8
areas. Let me go through this a
little bit more.
9 [Slide.]
10 Our prescription for this diagnosis, which
11
is the central part of this report, is that we
12
believe the society, we collectively, should
13
utilize the new scientific knowledge not only to
14
discover products but to improve the development
15
process and make it more effective.
16 We need to develop robust applied research
17
programs into these critical-path science areas to
18
lead to generalized knowledge that all developers
19
can pick up and use as they are trying to bring a
20
product through the path. We
need to strengthen
21
the academic bases for these critical-path
22
disciplines--and I will discuss this more in a
at
185
1
minute--and we need to intensify FDA involvement in
2
critical-path research and standards development to
3
make this happen because we feel the kind of
4
knowledge is central to helping us move along.
5 [Slide.]
6 Now, what do I mean by the critical path
7
for medical-product development.
This schematic
8
shows a very generalized product-development path
9
that kind of includes devices and drugs and so
10
forth, so it is overly generalized.
But,
11
basically, you start with basic research
12
information and move into, for drugs, what is
13
called the discovery phase, finding candidates and
14
targets.
15 For devices, you are designing prototypes.
16
You move into a phase called preclinical
17
development where you take your selected candidate
18
or candidates and you work them up and you try to
19
select the most promising ones.
If you have a
20
promising candidate, at some point, you can move
21
into the clinic and there you are going to really
22
be spending big bucks and you are going to start
at
186
1
doing progressive clinical testing, evaluating the
2
product and seeing whether or not it has those
3
characteristics of performance, the effectiveness,
4
the safety that you are looking for.
5 We put the critical path starting
6
somewhere in the middle of that prototype, design
7
or discovery phase and moving all the way through
8
to getting the product onto the market.
9 [Slide.]
10 There are really three dimensions that we
11
have identified from the basis of our knowledge in
12
overseeing tens of thousands of these development
13
programs. First of all, you have
to assess safety
14
progressively, as you move down the path. When it
15
is a twinkle in the eye, an idea in the laboratory,
16
it is not that critical. But,
very quickly, you
17
have to start thinking about, is this intervention
18
going to be safe and how do we evaluate the safety
19
of this, first predictively and then
20
confirmatively.
21 Second of all, you want to show, as we
22
said earlier, the product works.
The whole reason
at
187
1
people are going to pay for this, give you money,
2
is that it would be effective and would treat their
3
condition. From the very earliest
phases,
4
developers are evaluating the candidates and trying
5
to predict are they going to work.
This is very
6
difficult to select those candidates or prototypes
7
that actually will be effective.
8 The third dimension, which is really
9
ignored by a lot of folks in the science community
10
because they don't have to deal with it, is the
11
industrialization aspect. You
may have the
12
greatest idea in the world and, if you can't
13
mass-produce it, you are not going to have a
14
product at the end of the day that you can market.
15 So, from the very earliest stages, you
16
have to be tailoring that product and scaling it up
17
and designing it in a way that it can be reliably
18
manufactured. That is an area
where FDA puts a
19
great deal of emphasis because we see, at the end
20
of the day, a lot of failures of products that
21
cannot be reliably manufactured.
22 [Slide.]
at 188
1 Now, the complexity of this and the
2
problem of explaining this and the reason you are
3
going to hear a lot of different perspectives this
4
afternoon is this critical path is extremely
5
complicated for any given product line, shall we
6
say. These three dimensions each
have whole
7
departments, huge departments, in companies devoted
8
to them and there are actually whole industries
9 that work in little
areas here such as
10
contract-toxicology or clinical-research
11
organizations, or all sorts of things,
12
contract-manufacturing organizations and so on,
13
that function throughout these dimensions of the
14 critical
path.
15 But, basically, you are working very early
16
during the prototype design or discovery. You are
17
trying to predict safety and you have a variety of
18
ways that you try to do that.
You try to select
19
materials that have already been known to be safe.
20
You try to look at structure-activity relationship,
21
say, on a computer or other ways that you can
22
predict safety. Medical utility;
back in the
at 189
1
prototype design or discovery area, you are really
2
looking at test-tube models, in vitro models, to
3
see if the compound, if it is a drug, intervenes on
4 a
pathway of interest and what the in vitro
5
findings are.
6 You might do consumption modeling for a
7
device. At the same phase, you
are working with
8
that physical design of the product.
Then you move
9
later into animal testing. You
are trying to do
10
the same thing. Finally, you
move into human
11
testing. So there will be a
great deal of
12
discussion from different perspectives this
13
afternoon from all our presenters.
Many of them
14
will just focus on one area on this critical path
15
in one of these three dimensions.
16 [Slide.]
17 Why do I say that the science base hasn't
18
been robust enough and that is our diagnosis.
19
Well, the science necessary to evaluate and predict
20 safety and effectiveness and to enable
21
manufacturing is actually from the science that
22
generates the new ideas. It is a
different kind.
at
190
1
It is more applied science.
2 In general, NIH and academia do not focus
3
their research in this area. It
has been primarily
4
done in the private sector.
Critical-path research
5
is complementary to basic and translational
6
research but it results in the creation of new
7
tools for product development, and usually
8
generalized tools, not very specific tools. Let me
9
talk about this a little bit more.
10 The basic research feeds into all these
11 types
of research, obviously, because basic science
12
research is foundational.
Translational research,
13
which NIH is now focusing more on its road-map
14
initiative is typically involved in a specific
15
product or type of products and moving those along
16
from discovery into the early clinical development.
17
If you read the literature, there are lots of
18
definitions of translational research but that is
19
sort of the most consensus definition.
20 Critical-path research, we feel, underlies
21
many of these later phases but is a fundamentally
22
different kind of enterprise. So
I am going to
at
191
1
talk a little bit about what kind of tools might be
2
developed by using critical-path research to give
3
you a better idea.
4 [Slide.]
5 We call these, in general, evaluative
6
tools because what the developer is trying to do is
7
figure out whether or not this product is going to
8
work and whether it will be safe.
In the early
9
stages, the developers used scientific tools to
10
select candidates that they predict will have a
11
high probability of safety and effectiveness. This
12
is because the failure rate is such a big problem.
13 If you develop your product all the way
14
into the clinic and then find out it doesn't work,
15
you are burdening the system with a tremendous
16
number of failed products and you are driving up
17
the costs. Typically, developers
use laboratory
18
tests, consumption models and animal studies to
19
make these predictions of safety and effectiveness.
20 [Slide.]
21 In the later stages of development, human
22
testing is used to confirm these predictions. But
at
192
1
the very early human safety-and-efficacy testing
2 isn't extensive enough to do that, Phase I,
Phase
3
II. So their preliminary
estimates are made, "We
4
hope this product is going to work," based on
5
biomarkers, various markers that can be used to
6
look at human response.
7
Then, if you get into a
Phase III trial or
8 a
confirmatory trial, then you would finally be
9
sure that your predictions were correct. So these
10
are the general types of tools we are talking
11
about.
12 [Slide.]
13 To get into specific examples, what would
14
be a predictive safety tool.
This is something we
15
don't have right now. What if
you could use a
16
genomic-expression system of maybe liver cells to
17
evaluate a compound's impact on liver-cell
18
function. This would help a lot,
say, in looking
19
at hepatotoxicity.
20 Right now, over the last decade, we have
21
developed liver-cell assays that actually can tell
22
how the liver metabolizes drug compounds and
at
193
1
predicts hepatic metabolism.
This has been
2
extremely helpful in screening out candidates that
3
have unfavorable metabolic profiles, drug-drug
4
interactions, and so forth. But
right now we don't
5
have a way to screen, say, on a genomic basis, for
6
potential liver toxicology.
7 This is an example of an evaluative tool
8
that everyone could use. You could
use it for
9
device constituents, biomaterials.
You could use
10
it for drugs and biologics and help weed out
11
compounds that, much later, would be found to be an
12
hepatotoxin.
13 [Slide.]
14 What about another predictive safety tool.
15
CBER has been trying to develop some of this,
16
reference standards and test systems for
17
gene-therapy vector potency. It
is obviously
18
extremely important for gene-therapy vectors as a
19
safety issue that you have a very good handle on
20
the potency of those vectors.
That is a
21
standardized safety tool that is needed, so you can
22
compare your lots of your vector in a standard
at 194
1
assay to make sure you are hitting the right
2
target.
3 [Slide.]
4 What about predictive efficacy tools. One
5
of the problems with devices is they are
6
continuously being modified and those modifications
7
can stack up behind human testing because the
8
engineers can work a whole lot faster than the
9
doctors. If we had better
computer models that
10
could model and predict the outcomes of these
11 device modifications, that tool could be used
very
12
extensively to accept certain levels of changes in
13
devices without going to human testing.
Actually,
14
our device reviewers feel that these tools could be
15
developed. They think this is
possible.
16 [Slide.]
17 Another predictive efficacy tool would be
18 a
quantitative biomarker that could be used both in
19
animal and early human trials to indicate the
20
effect of the drug, give us that good predictive
21
test early and guide dose and regimen decisions, so
22
you are not floundering around in the dark until
at
195
1
you get up into the efficacy trials.
2 We have had very few new biomarkers being
3
emerging in the last decade. It
is one of the
4
problems with the science base.
5 [Slide.]
6 Another kind of safety tool, and you might
7
not think of this as a safety tool, but we need
8
standardized, acceptable trial designs to assess
9
specific safety risks, a
consensus view on, okay,
10
if you do this trial design and you run your
11
compound through it, you do this trial design and
12
you run your compound through it, you have
13
addressed this question, this uncertainty, to the
14
degree that you need to; for example,
15
hepatotoxicity, electrocardiographic changes that
16
may cause cardiac arrhythmias, and so forth.
17 We don't have enough of these standardized
18
trial designs out there. So each
developer is
19
forced to work with the FDA and try and figure out
20
how this question will be addressed each time and
21
how much residual uncertainty would be acceptable.
22
So this is a different kind of tool.
at
196
1 [Slide.]
2 I think Rob Califf will be talking about
3 this this afternoon, but another confirmatory tool
4
would be an available clinical-trial network that
5
you could run your compound through to rapidly and
6
efficiently answer specific safety queries using a
7
large simple-trial format; in other words, a
8
network in which compounds could be plugged into to
9
answer these questions.
10 Right now, again, the clinical-trial
11
infrastructure in the United States, the developers
12
are having to--each time, they have to create the
13
infrastructure in the trial network to answer these
14
questions. It is very
laborious. It is very
15
time-consuming, adds years, often, to a development
16
program.
17 [Slide.]
18 A confirmatory efficacy tool that everyone
19
wants to have but I am going to get into later how
20
much actual design work and scientific work has to
21
go into this, would be an FDA guidance document of
22
how to use a specific technology as an accepted
at
197
1
surrogate marker for a condition with a specific
2
clinical-trial framework.
3 So the question is, you do this trial and
4
you use this marker in this fashion, and then that
5
is acceptable as a marker for efficacy in this
6
condition for this claim.
Everyone probably who is
7 a
product developer would like to have access to a
8
tool like that, but we have very few of them and
9
there are not that many on the horizon.
10 [Slide.]
11 Devices has actually done this one in one
12
case, and we have it in the paper, to have a
13
consortium-based trial framework to allow
14
manufacturers to pool resources in a single trial
15
to answer a specific efficacy question pertaining
16
to a class of devices or drugs, or a specific
17
safety question.
18 In the example in the report, they pooled
19
studies for digital mammography to look at was
20
there an advantage over plain-film mammography.
21
Several manufacturers were able to pool their data
22
to attempt to answer that question.
But this is,
at 198
1
again, something that isn't at hand for people
2
right now.
3 [Slide.]
4 Another tool, and I hope Rob will talk
5
about this a little, would be to standardize the
6 clinical infrastructure more so that we have
7
standardized case-report forms and data-collection
8
and data-format standards for clinical trials so
9
that developers are not constantly inventing new
10
case-report forms, new data standards, the
11
investigators are not having to deal with this each
12
time different--this would be a tremendous benefit
13
to the clinical-research enterprise.
14 [Slide.]
15 So those are examples. You could see it
16
is a very wide range of examples, all the way from
17
new laboratory tests of different kinds to
18
clinical-trial networks and data standards. I
19
think the presenters this afternoon will go over
20
all the different aspects of this in different
21
ways.
22 I just have a few more. I want to finish
at
199
1
up so we have some time for questions, but I want
2
to talk about our initiative and how this fits in.
3
The first part of my talk was really to frame this
4
problem and we seek input on whether this is a
5
correct definition, diagnosis, framing of the
6
problem.
7 This is what we are doing now. Our
8
initiative is a serious attempt to bring attention
9
and focus to the need for targeted scientific
10
efforts to modernize these techniques and methods
11
that are used to evaluate the safety, efficacy and
12
quality of the medical products as they move from
13
selection and design all the way to the bedside.
14
So it is this evaluation pathway.
15 [Slide.]
16 We are trying basically to address the
17
problem I presented at the beginning of my talk.
18
We need to get more innovative products to
19
patients. We want the fruits of
the biomedical
20
revolution to actually translate into better
21
health. We want to achieve
robust
22
product-development pathways that are efficient and
at
200
1
predictable and I can tell you we simply do not
2
have that now. That is what is
driving the cost
3
and taking the time.
4 We think we need to develop new tool kits
5
that bring the scientific advances into the
6
development process, not just the discovery
7
process. We, and others, we
feel, should perform
8
research on tools that remove specific identified
9 obstacles in product
development. That is how we
10
are planning to address this.
11 The question is why--and we have gotten
12
this when we issued the report--like, why did FDA
13
issue this report? What role do
you have?
14 [Slide.]
15 I would like to point out FDA scientists
16
are involved in the review during the product
17
development. We are there soon,
actually, often,
18
right before, these products get into people. The
19
scientists are involved in evaluation all during
20
the development process and then the review
21
process. Then, once the products
are out on the
22
market, sometimes, they experience safety failures
at 201
1
even at that time.
2 But, during development, our scientists
3
see the successes, failures and the missed
4
opportunities that are occurring.
FDA guidance
5
documents, we have noticed, are known to foster
6
innovation and improve chances of success, the
7
existence of these guidance documents, and they
8
seem to. Where there is clarity
in a development
9
pathway, in a critical path, it fosters attention
10
and innovation and investment in that product area.
11
So, where developers see a clear path forward, they
12
will focus their innovation in that area.
13 We have, in the past, had a convening and
14
coordinating role for new biomarker and
15
clinical-methods development and, when we have done
16
that, it has been extremely successful; for
17
example, in the HIV area in stimulating and helping
18
that development move forward efficiently.
19 [Slide.]
20 This is in the Critical Path Report. It
21
is a little too complicated to explain totally in a
22
slide, but what we see happening is, on the
at
202
1
left-hand corner up there, during the application
2
review, either the investigational product
3
oversight or the marketing application, the
4
problems arise. That is what we
are always dealing
5
with. If there are no problems
in product
6
development, then things just go right through.
7 But we are seeing all these problems that
8
arise; difficulty in prediction, unexpected safety
9
problems that arise, difficulty in determining the
10
correct dose or design. As we
identify these
11
problems and if they become generalizable--in other
12
words, we see them across more than one product--we
13
often try to refer them, through scientific
14
research, to academia. We take
them to our
15
advisory committees, try to get a discussion going
16
about what the problem is and how it can be
17
surmounted.
18 We go back and forth between public input
19
and research and, hopefully, we achieve some
20
solutions. You will hear about
that from some of
21
our presenters, the FDA presenters, this afternoon.
22
We achieve solutions in many areas, but it is
at
203
1
limited by the amount of resources we can put
2
against it.
3 Eventually, if we have recommended a
4
scientific approach, as we take these to our
5
advisory committees or we publish guidance, we have
6 a
public process for discussion and then that
7
becomes a public standard that developers can use.
8
So they can pick up that standard, if they have a
9
product, and they can just apply it.
They don't
10
have to develop it. They don't
have to agonize
11
over it. That is the standard
they can use. We
12
find that this really facilitates development.
13 [Slide.]
14 Let me just use an example, and this slide
15
is out of order. So skip to the
next one. For
16
biomarker or surrogate-marker development, what
17
usually occurs now. And you are
in the scientific
18
community. This is what I think
occurs; right? A
19
marker is developed by an academic or industrial
20
scientist as part of a research project.
21 They publish the method and it get picked
22
up in additional laboratories and people
at
204
1
start--they are interested in this marker, whatever
2
it is, some blood test or metabolic test or
3
whatever. Then it begins to be
adopted by academic
4
clinicians and it might be in a kind of home-brew
5
format; in other words, it is not an FDA-approved
6
test. It is something that is
available and people
7
use.
8 Then this biomarker gets into widespread
9
clinical use and people start trying it on their
10
patients in the clinic, not usually in trials,
11
maybe series; we tested this in people and they
12 progressed
and here it is, a ten-patient series,
13
100-patient series. Then
gradually it begins to be
14
used in trials. People think,
maybe this is a
15
promising biomarker. We will put
it our trial and
16
we will use it.
17
The sponsors will present it
to FDA to
18
say, we have this biomarker in our trial. We don't
19
know what it means, but here are the results of
20
this biomarker. Then publicly,
and in scientific
21
journals, there are calls for us, as a biomarker or
22 a
surrogate marker, to use this biomarker, okay,
at
205
1
because everybody has gotten used to it and think
2
it should be used. It might be
promising.
3 But then we get stuck as a society. What
4
happens next? There are calls to
use it. We have
5
had biomarkers that have been in this status for
6
twenty years.
7 [Slide.]
8 Now I will get back to that other slide.
9
The question is how do you accumulate this existing
10
data into knowledge. We have all
these
11
publications about this biomarker.
People have
12
kind of used it and everything, and like it, but
13
how do you turn it into something that is actually
14
knowledge or a generalized principle that you can
15
actually use and rely upon.
16 That is really one of the issues here.
17
Who is responsible for doing this?
That is
18
something that we are raising in the critical-path
19
arena. I am using biomarkers as
an example because
20
they are so important to this endeavor.
21 Who is going to pool all the data and
22
analyze it, do this cross-series and cross-study
at
206
1
analysis? Who is going to
evaluate the primary
2
data, not just published summary data, but who is
3
going to look at the primary data and say, how does
4
this really stack up? Who is
going to identify the
5
gaps?
6 Usually, what you find--everybody is
7
talking about the biomarker.
They say it is great.
8
When you really get to look at the data and you
9
look at the outcomes, you find gaps.
There are
10
gaps in our knowledge. We don't
really have
11
generalizable knowledge. We
don't have principles
12
yet.
13 Who is going to conceive and conduct
14
research that would actually close these gaps and
15
make this into a biomarker that developers can rely
16
upon or a true surrogate marker that could be used
17
for approval?
18 I am using biomarker as an example here
19
because I think this really summarizes the problem.
20
This is the development of an evaluative tool to a
21
level where it can be used in product development.
22 I
think our thesis is, right now, nobody is in
at 207
1
charge of doing this. And this
is one of the
2
problems because then it doesn't happen.
3 So we will skip on.
4 [Slide.]
5 Now, why could FDA play a role here? We
6
believe, and we have been told by Congress, that
7
access to medical technology, providing that is
8
part of our mission. The FDA is
responsible to
9
advance the public health by helping to speed
10
innovation. This is from one of
our publications.
11
We are in a health-promotion role as well as a
12
health-protection role at the FDA and so we feel
13
not only are we integral in the process because we
14
see all these problems, but improving this process
15
is part of our mission.
16 [Slide.]
17 So, what we have done, we have published
18
this initial report called The Innovation
19
Stagnation Report that was announced by
20
Commissioner McClellan and Duputy Commissioner
21
Crawford on March 16. Today's
presentation to the
22
Science Board is really our science kickoff of this
at
208
1
in the sense of having a substantive discussion.
2 We have been started extensive discussions
3
with a lot of stakeholder groups, industry,
4
academia, government and the patient groups are
5
very important. We feel that
patient groups really
6
need to have a voice in this. We
have one letter
7
written, I think, by one of the HIV activist
8
groups. We agree. We agree that, when you are
9
talking about biomarkers and outcome measures and
10
clinical-trial design and with claim, the patient
11
groups must also have a real voice in this.
12 We have had really positive feedback so
13
far from all these groups about the fact that we
14
have identified a problem that needs to be
15
addressed, but we really seek your input today.
16
[Slide.]
17 What we plan to do, in concrete terms, is
18
we can't take on this whole problem.
We think
19
identifying it is extremely helpful.
We plan to
20
identify and prioritize the most severe barriers,
21
development barriers and problems, in the areas
22
that provide the greatest opportunity by soliciting
at
209
1
input from a wide variety of sources.
2 Then we are going to construct a national
3
critical-path opportunities list because we think
4
identifying these barriers and opportunities will
5
give us something that will be concrete enough for
6
people to make this real. We
will publicize this
7 list.
8 We are also really making sure we focus
9
our internal activities as far as our process, make
10
sure we have efficient ways of surfacing and
11
identifying these problems. We
can't solve them
12
all but we need to surface them up and make them
13
public and identify them in a way that the
14
community can deal with them.
And then we seek
15
community concurrence on additional steps once we
16
get our final list together which we hope we should
17
do by early fall.
18 [Slide.]
19 Conclusions; I think FDA's Critical Path
20
Report raises serious concerns about the ability of
21
the current development process to get innovations
22
to patients rapidly and efficiently.
We believe
at
210
1
this is becoming more serious with the pressure on
2
the basic-science side.
Preliminary discussions
3
with key experts in industry and academia are
4
validating this analysis of the development
5
problem.
6 [Slide.]
7 We are continuing our dialogue with a wide
8
range of stakeholders to try and reach agreement on
9
the problem scope and the problem definition. So
10
we are still early. We are also
working on
11
defining specific opportunities to overcome these
12
hurdles with concrete deliverable steps that could
13
be accomplished.
14 [Slide.]
15 We have just opened a web address and
a
16
docket. I have these here and
they are in the
17
handouts for people. So we
welcome comments both
18
to our docket and people can refer to our web
19
address, and we will be updating that with events
20
and other information.
21 Thank you.
22 DR. SHINE:
Janet, thank you very much.
at
211
1 I think it is worth spending a few minutes
2
for questions for Janet that you might have on the
3
way she has laid out these issues.
4 Janet, a couple of things I would put on
5
the table. One is you have shown
us data for ten
6
years. It would be very helpful
to have a
7
long-term view of how much of this is cyclical,
8
have we seen examples of ups and downs over time,
9
what is the periodicity, and so forth.
10 The second question I would have is, one
11
of the major changes in the '90's was the move to
12
genetics, genomics, and so forth.
It was clear in
13
talking to many of my colleagues who are in the
14
pharmaceutical business that they really thought
15
that it was going to be pretty easy to identify
16
targets as the genome went forward and that it was
17
going to be a slam-dunk in terms of getting
18
products out.
19 It turned out that was very hard. I am
20
wondering, and I think I would like to hear from
21
some of our other speakers about this issue as
22
well, how much of the downturn that you have
at
212
1
described in the '90's is, in fact, betting on the
2
wrong paradigm and that there is a fundamental
3
problem with that paradigm that will require
4
evolution of the paradigm if it is going to produce
5
new products.
6 Finally, you have pointed out that a great
7
deal of the difficulty arises out of having the
8
methodologies that allow development and
9
association of efficacy, safety and so forth, to
10
develop. In a number of other
industries--I am
11
thinking now about the roadmap in the
12
semi-conductor industry.
13 SemiTech in the computer industry, in
14
fact, put together resources around common
15
technologies. They still
maintain their
16
proprietary interest in the drug that they were
17
developing, but when there was a need for having
18
some methodology which was the infrastructure, have
19
you had any discussions about the concept of an
20
industrial public-private kind of enterprise which
21
would, if you will, recapitulate, and I am using
22
SemiTech as an example. There
are things that are
at
213
1
similar. There are things that
are different. I
2
understand. But I would be
interested in your
3
comments.
4 DR. WOODCOCK:
I actually violently agree
5
with that. I think that this is
a matter of almost
6
industrial policy in the United States, this
7
industry, the medical-product industry and its
8
research base which is applied research as well as
9
basic-science research.
10 We have looked, in the context of our
11
manufacturing initiative, our GMP initiative--we
12
looked at the collaborations in the semi-conductor
13
industry. In fact, this is a
very good model in
14
how that improved quality and the speed, agility,
15
of that industry and so forth.
We have had
16
preliminary discussions with the manufacturing
17
sector of medical products around this very
18
concept. I think this is a very
important concept
19
for us to explore.
20 As far as your second question, we think
21
that the downturn is not the whole story here. We
22
recognize that is multifactorial and might have
at
214
1
resulted from corporate mergers.
It might have
2
resulted from the fact that, in our estimation, I
3
think our technical estimation, and I will ask some
4
of the CDER scientists in the room, that genomics
5
or any of these new technologies take about fifteen
6
to twenty years before they really start paying
7
off.
8 But I think the problem we see is here are
9
these tremendous societal needs and expectations
10
out there and this is what we are getting out of
11
the pipeline right now. And the
causes of that you
12
can debate. But the question is
what can we do
13
about it, right now?
14 DR. HARLANDER:
I really applaud FDA for
15
creating this Critical Path Document because I
16
think it identifies precisely the issues that they
17
are facing. But I think it also
opens up--puts
18
tremendous pressure on the agency, itself,
19
because--and this is my first meeting, so I haven't
20
been able to tour all of the centers that would be
21
involved in providing the kind of scientific
22
support that you would need to really provide the
at 215
1
foundation for policy development, standards and
2
guidance development.
3 Do you have the scientific expertise and
4
the financial resources that you would really need
5
to have in order to support the development of
6
these kinds of technologies?
7 DR. WOODCOCK:
We do what we can with what
8
we have. No. I mean, we are doing what we can now
9
with what we have.
10 DR. HARLANDER:
But you don't have a lot.
11 DR. WOODCOCK:
We do not have a lot. Say,
12
in certain areas like the drugs
area, there is
13
virtually almost no laboratory support available.
14
But we don't conceive of this initiative as all
15 being
done internal to FDA. We believe one of
our
16
roles is really being able to see and generate and
17
identify all these problems and that a wide variety
18
of stakeholders needs to work together, as was just
19
alluded to, to kind of try to address some of
20
these.
21 But, even to do that role effectively, in
22 a
broad way, would require more scientific
at
216
1
resources than we have available to us now.
2 DR. SHINE:
Cecil?
3 DR. PICKETT:
Again, Janet, I would like
4
to echo Susan's remarks, particularly about
5
complementing, again, the FDA on taking up this
6
initiative. I think it is going
to be very
7
valuable. I think it will be a
framework that both
8
the FDA and industry and other stakeholders can
9
have a rationale conversation about how to improve
10
the development of new drugs.
11 Just a couple of comments. One of the
12
things that I have seen missing in the
13
documentation has to do with regulatory guidelines
14
and harmonization of regulatory guidelines. As you
15
know, we oftentimes have slightly different
16
guidelines depending on whether or not we are
17
filing in the European Union or Japan or here in
18
the U.S. I think that is one
area that can
19
certainly help speed up overall drug development.
20
So I would hope that could somehow be part of the
21
discussion.
22 DR. WOODCOCK:
What we did here was to--we
at
217
1
wanted, in the critical-path part, to have a
2
discussion of the science because I believe that if
3
we--and we agree that regulatory streamlining, or
4
whatever, is another effort that is a continuous
5
challenge and we need to keep working on and we are
6
working on.
7 But, if you default your discussion or
8
start your discussion, frankly, around regulatory
9
standards, then people tend to forget that there is
10 a
basic science infrastructure that must be in
11
place. They tend to forget the
point of this
12
report and they default to the issue of, well, it
13
is just regulatory requirements and everything.
14 So I think we all have to be careful
15
because, if it is just regulatory requirements,
16
then there is no science problem.
If we agree
17
there is a science problem, then we really to
18
address the science problem.
19 DR. PICKETT:
But, again, regulatory
20
requirements are based upon sound science.
21 DR. WOODCOCK:
Agreed. The more robust
22
applied scientific infrastructure we have available
at
218
1
to us, the more rationale and targeted and actually
2
less conservative our requirements can be because
3
it drives more certainty.
4 DR. PICKETT: I
just want to follow up on
5
Ken's comment about whether or not it is the wrong
6
paradigm in terms of drug discovery and
7
development. I don't think the
paradigm is wrong.
8 I
think what has been incorrect is the expectation
9
part.
10 DR. SHINE:
Much like gene therapy, the
11
expectation is that it would come very quickly.
12 Let's go to John and then I think, Gail,
13
we are going to let you talk and you can engage in
14
the conversation at that point.
John?
15 DR. THOMAS:
Establishing priorities is a
16
noble adventure and certainly some companies are
17
going to stake out their area based on their
18
expertise. To some extent, we
have already done
19
that. If you just want to be so
crass as to go
20
down the list of diseases and discover how much we
21
spend per patient, that could translate back to
22
what a society can afford to pay for.
at
219
1 But then you get into the predicament of
2
diabetes melitis, for example, probably one of the
3
most expensive diseases to treat because of all the
4
other ramifications that you have.
So there you
5
are not really treating one disease.
You are
6
treating several, a constellation, of diseases.
7 The other thing, in establishing
8
priorities in terms of therapeutics, it seems to me
9
you have go to accommodate short-term versus
10
long-term therapy and then something, which we
11
already have in place, is to deal with orphan
12
drugs. That has got to be given
some consideration
13
as you go forth in establishing priorities. I
14
don't have the answer to that, but just
15
observations that throw money at one disease versus
16
another disease. I guess the
empty tin can rattles
17
the loudest and that, in some cases, historically,
18
has determined what would constitute resources
19
going towards a given pathology or disease entity,
20
which you are not going to be able to exclude.
21 DR. WOODCOCK:
That is an intriguing
22
suggestion, that we look at the burden of disease,
at
220
1
economic burden or other burden of disease, in
2
thinking about this. Many of
these technologies,
3
these evaluative technologies, will cut across many
4
products and diseases but, in the later stages,
5
obviously for the effectiveness prediction, then
6
you are narrowing it down to a specific disorder.
7 DR. SHINE:
Janet, I know you are going to
8 stay around because one of the questions--the
9
report emphasizes that the strategies used for
10
orphan diseases could be used more generally. I
11
want to come back to that in terms of how you think
12
that is happening. But thank you
very much.
13 We have got six presentations we want to
14
hear from and I want to be sure that they each get
15
their chance. Each of them is
twenty minutes. The
16
first is a perspective on anti-infectives and
17
vaccines that Gail is doing. As
you know, she is
18
Vice President for Scientific Affairs and the
19
Distinguished Lilly Research Scholar for Infectious
20
Diseases.
21 You should be aware that our Acting
22
Commissioner made reference to the fact that you
at
221
1
also have an Alabama heritage where Gail was an
2
extraordinarily distinguished Chair of the
3
Department of Microbiology as well as a national
4
leader in this area. So we are
looking forward to
5
her perspective.
6 Gail?
7
Perspective on Anti-Infectives and Vaccines
8 DR. CASSELL:
Thank you, Ken, for those
9
nice comments, and Lester. I
really am
10
appreciative of the opportunity to share with you
11
some thoughts this afternoon as it relates to
12
anti-infectives and vaccines.
Obviously, in twenty
13
minutes, I can't do justice to both.
So what I
14
will try to focus on are antibiotics and the
15
critical need for new antibiotics.
16 First of all, I would just like to say
17
that I would begin by talking about the current
18
need, the current status of development of
19
anti-infectives and then what relationship I think
20
the Critical Path Initiative has to these needs,
21
opportunities, priorities and next steps.
22 First of all, I would like to say that I
at 222
1
am wildly enthusiastic about this initiative and I
2
shared the document very early on with Josh
3
Lederberg, a friend who won the Nobel Prize when he
4
was in his early 40's for microbial genetics and
5
has had a long-term interest in FDA and development
6
of new drugs. I don't think he
would mind my
7
sharing with you this afternoon his comment, and
8
that was he found this document to be one of the
9
most refreshing that he has read in a long time.
10
He was very excited to see this initiative and FDA
11
really taking leadership. I
think we could all say
12
that.
13 With regards to the current need for
14
antibiotics but also antivirals and vaccines, I
15
would draw your attention to an IOM report that was
16
released last March, March of 2003, in which
17
antibiotic drug discovery and development as well
18
as vaccine development was declared a crisis
19
worldwide.
20 The reason for the declaration of
the
21
crisis is because of the current thinness of the
22
pipeline, both as it relates to vaccines as well as
at
223
1
antibiotics and antivirals, other than HIV and,
2
perhaps, hepatitis as well as maybe herpes, but,
3
other than that, certainly lacking in that regard.
4 With respect to antibiotics, the reason
5
for the declaration of a crisis is because of the
6
current experience that we have with antibiotic
7
resistance and the fact that we have several
8
pathogens for which now strains have been
9
identified for which there are no remaining
10
antibiotics left. In fact, last
week, you may have
11
read about the new strain of Salmonella isolated in
12
Taiwan which was resistant to every antibiotic
13
tested. This is a food-borne
pathogen which does
14
give cause for concern. And I
could go down the
15
list.
16 What you may not appreciate, however, is
17
that tuberculosis is one of the oldest infectious
18
diseases. One-third of the
world's population
19
remain infected with latent t.b. and now there is a
20
tremendous increase in multi-drug-resistant strains
21
of t.b. If you heard, on March
16, the release of
22
the new WHO report. These
strains are increasing
at
224
1
with a great deal of frequency worldwide and the
2
five antibiotics, four at most, left to treat MDR
3
t.b., all with the exception of a quinolone, are
4
over 40 years old.
5 So this, if nothing else, should
6
illustrate the need for new therapies with respect
7
to antibiotics. We also have new
and emerging
8
infections, and I don't need to remind you of SARS
9
or the ongoing problem that we have right now with
10
bird flu which would require an antiviral and
11 vaccine.
12 Lastly, in terms of need and the reason
13
for the statement of the crisis by the IOM report
14
is the fact that we have the threat of
15
intentionally released microorganisms in the way of
16
bioterrorism threats and, if you were to take a
17
look at the threats on the Select A, B, C list and
18
try to match that up with our armamentarium
19
currently with respect to antibiotics, vaccines and
20
antivirals, you would see just how bare the
21 cupboard
really is.
22 Enough said about the need. What does the
at
225
1
pipeline look like? I have just
completed, along
2
with some of my colleagues, a review of the current
3
antibiotics in the pipeline, and antivirals, using
4
the four most commonly used private databases in
5
terms of what is in clinical development. I can
6
tell you this afternoon that we should not rest
7
assured that the pipeline is full.
8 In fact, it is currently bare with respect
9
to any new antibiotics in clinical development with
10
most of those, if not all of them, being slight
11
versions of the old classes with particular
12
emphasis on new quinolones listed.
There is an
13
obvious difference between what we see in the
14
pipeline compared to what we would have seen ten
15
years ago, and that is that most of the antibiotics
16
are being developed by biotech companies and
17
companies outside the U.S. as opposed to in the
18
U.S. as opposed to typical large pharma.
19 You have been seeing a lot in the public
20
press about large companies having gotten out of
21
development of antibiotics and, often, it is said,
22
that the only reason for that is lack of market or
at
226
1
market insufficient to recoup investment in
2
development of new antibiotics.
3 What I want to impress upon you today is
4
that is just one of many reasons that most
5
companies have chosen to invest resources in other
6
areas. There are equally
compelling unmet medical
7
needs that do have larger markets and the
8
scientific opportunities are as great, or greater,
9
in those areas.
10 What I do want you to hear me say is that
11
there are still significant technical challenges in
12
antibiotic drug discovery. From
1980 to 1990,
13
antibiotic drug discovery was primarily
14
serendipity, as was most drug discovery. However,
15
in the early '90's, mainly because of public
16
pressure and the view for need of new antibiotics,
17 a
great effort was put in place by most
18
pharmaceutical companies using the new
19
technologies.
20 The first microbial genome was completed
21
in 1993, and many followed shortly thereafter. So
22
it is not due to lack of targets.
We have a
at
227
1
plethora of targets. Where there
is a need is an
2
increase in chemical diversity and a better
3
understanding of the interactions of the different
4
targets that we currently are trying to develop
5
inhibitors for.
6 Now, what I would like to also say,
7
however, is that, with respect to antibiotic
8
development, it is unlike other drugs that one
9
might develop and that is you generally don't
10
develop a single antibiotic for a single
11
indication. So the large numbers
of patients
12
required for the different indications that one
13
would want to develop are very important
14
considerations when you look at cost.
15 The other thing that I think is a problem
16
is that many in the infectious-disease community,
17
leading authorities, in the late '90's and even as
18
late as 2000 were saying there was no need for new
19
antibiotics, that we had enough antibiotics
20
especially those that were practicing in primary
21
care, pediatrics, OB-GYN. When
one would conduct
22
focus groups to gather data to support continued
at 228
1
development, the response was that we had enough.
2 So this raises the problem that we have
3
when we talk about treatment of infectious diseases
4
particularly those that are bacterially induced,
5
and also virally induced, and that is we need new
6
and improved diagnostics. So I
just wanted to say
7
that I think that some of the critical-path issues
8
are particularly relevant to the ability to develop
9
new diagnostics and the use of them to promote more
10
judicious use of the agents that we do have but
11
also to identify new opportunities with respect to
12
new agents that are out there.
13 Now, I just wanted to talk a little bit
14
about the potential changes that have occurred just
15
in the last two or three years that I think will
16
have a significant impact on antibiotic and
17
antiviral drug development.
18 One is a major advance which, actually,
19
the Critical Path Document does refer to and I want
20
to elaborate just a little bit, and that is, when
21
it became obvious that no one was focused on
22
development of new drugs for t.b. and malaria, the
at
229
1
public pressure grew and public-private
2
partnerships were established, the Gates Foundation
3
and the Rockefeller Foundation being the primary
4
funders for these public-private partnerships.
5 The two primary ones are the Malarias
6
Ventures for Medicine and the Global Alliance for
7
T.B. I spent last week at the
Wellcome Trust
8
reviewing these PPPs, as we call them, and I think
9 it is a very exciting new approach to drug
10
discovery and development. In
fact, the thing I
11
like best about the Critical Path Initiative is the
12
realization that there is a lot of synergy between
13
research in the public and private sector that is
14
relevant as it relates to drug discovery
15
development. We should admit
that, all of us, as
16
well as the public and Congress and figure out how
17
we can better have these two sectors working
18
together in the area of drug discovery and develop.
19 The PPPs actually have the potential to
20
bring together the best of the best from both the
21
public and the private sectors.
So I think that,
22
if anything, I would say, Janet, I hope that, in
at
230
1
the next steps that you have outlined and the
2
groups that you are bringing together, that you
3
won't talk to the public sector in isolation from
4 industry, et cetera.
5 I think it is going to be the combination
6
of the group that is important, and to not ignore
7
the fact that now the lines are very blurred in
8
terms of how much drug discovery and development is
9
being done in the public sector compared to even
10
five years ago. As most of you
probably know, many
11
universities now have GMP facilities not just for
12
small molecules but also for biologics.
13 So I think the key to implementing the
14
Critical Path Initiative is going to be to try to
15
figure out how the public and private sector can
16
actually work better together.
17 If one had to prioritize tools that would
18
be most helpful for development of new antibiotics
19
or new antivirals and new vaccines, I have given
20
this a lot of thought and have had a lot of
21
conversations with people. I
think most of us
22
would agree that we would probably put at the top
at
231
1
of the list more emphasis on systems biology to use
2
to determine and predict the interaction of the
3
different targets.
4 Ken, my response to you in terms of the
5
genomics approach and why we feel like we are empty
6
handed is because I think we really didn't
7
appreciate how interactive these new targets were
8
going to be. Now we have the
ability to develop
9
inhibitors not just for a single enzyme in a given
10
pathway but actually to develop inhibitors, and in
11
some cases multiple inhibitors, for an entire
12
network.
13 Some of the most exciting work being done
14
in the public sector in terms of new antibiotic
15
discovery, as a result of the
$1.7 billion
16
increase in funding for NIH to focus on development
17
on new antibiotics and new antivirals, is exactly
18
this approach and that is a systems-biology
19
approach.
20 The other thing that I would say I would
21
put at the top of the list is the systems-biology
22
approach, particular that used by Lee Hood and
at
232
1
members of the Systems Biology Institute in Seattle
2
to try to better predict toxicology and efficacy.
3 I
think this is where your public-private
4
interactions and the public-private partnerships
5
that are beginning to be established will be very
6
useful.
7 There is an extremely important piece of
8
legislation that was passed in December of this
9
year in the DoD appropriations.
That was to allow
10
the establishment of what we call another
11
transaction authority which actually facilitates
12
establishment of consortia of different companies
13
and also public institutions getting around
14
antitrust issues to do exactly the kind of things
15
that, Ken, you were referring to.
16 Last week in London, I learned, in fact,
17
there is a Brain Initiative there where, in fact,
18
multiple companies are getting together to try to
19
address some of the nonproprietary issues that are
20
bottlenecks as far as drug discovery and
21
development. The MRC is very
interested, as the
22
Wellcome Trust, in finding translation research.
at
233
1
But, again, it is the public-private partnership
2
that has people most excited.
3 I think other priorities would have to
4
rest with manufacturing and improved manufacturing,
5
particularly as it might relate to use of natural
6
products for development of new antibiotics and
7
also management of clinical data.
The one think, I
8
think, I would like to have seen more of in the
9
Critical Path Initiative is more on the need for IT
10
infrastructure and innovation in IT as it relates
11
to collection and management and analysis of
12
clinical data, and knowledge-management data-mining
13
tools that could be developed.
14 Then the last thing I would just emphasize
15
is that I had a chance to review your slides, Dr.
16
Goodman, that you are going to speaking about this
17
afternoon and would be very supportive of the
18
efforts--all efforts--that we can put forth with
19
regards to development of tools for screening for
20
adventitious agents and biologics as well as other
21
products because I think this is a huge looming
22
potential crisis that we need to be addressing.
at
234
1 I would like to lastly say, and I think I
2
am going to end five minutes early, Ken, that the
3
American Society for Microbiology that has over
4
42,000 members who are very concerned about the
5
crisis that we have in vaccines and antibiotics. I
6
chair the Public and Scientific Affairs Board for
7
ASM. We are in the process of
establishing a task
8
force to establish priorities and opportunities.
9 The one question that I would have for FDA
10
would be, in thinking about the next steps, what,
11
then, do we do with those priorities and could we
12
talk more specifically about who will be
13
responsible for what.
14 I think you know I have the same questions
15
that were posed earlier by Sue, and it is one of
16
resources. How can we help
empower FDA to play a
17
leadership role because I think it is FDA's role to
18
do that.
19 The other thing I would encourage is that,
20
with the new centers, Research Centers of
21
Excellence, that have been established by NIH with
22
this increased funding, whose primary emphasis is
at
235
1
on development of new antivirals, vaccines and
2
antimicrobials, is that there be very close
3
interactions between those RCEs and FDA from the
4
very outset and, also, hopefully very close
5
interaction with the new product-development PPPs
6
to facilitate that process as much as possible
7
because I think there are a lot of opportunities.
8 Lastly, with regards to development of
9
countermeasures for bioterrorism, having in place
10
the Animal Rule which, if you are not familiar
11 with,
maybe there will be time in discussion to
12
bring that up. But I think it
will force all of us
13
to take a new approach as we think about
14
development of new antibiotics, new antivirals and
15
vaccines and what the limitations are or are not
16
when, in fact, you don't have the ability to test
17
efficacy in humans.
18 So thank you.
19 DR. SHINE:
Gail, thank you very much. I
20
think most of us, if not all of us, recognize that
21 a
number of the issues which are laid out in the
22
critical path are critical in antibiotics areas as
at
236
1
well as others. But, as I hear
you talking, the
2
rate-limiting step still seems to be a fundamental
3
science issue; that is, it sounds as if it
4
is--whether you call it systems biology or looking
5
at interactions, and so forth.
Is that a
6
misinterpretation?
7 DR. CASSELL: No; I don't think so, Ken.
8 I
just would say that, again, going back to that
9
public pressure that was placed on the industry in
10
the early 90's when we first had the
11
vancomycin-resistant enterococci and the historic
12
meeting at the Rockefeller, the OTA report, the ASM
13
Task Force on Antimicrobial Resistance, the
14
establishment of the Interagency Committee and Task
15
Force on Antimicrobial Resistance, companies did
16
begin to invest tremendously in the development of
17
new antibiotics using the best of the latest tools,
18
genomics--I mean, in infectious diseases, we have
19
long been using proteomics and have the ability to
20
do that.
21 I can tell you that at least one company
22
that I am aware of, over a hundred high-throughput
at
237
1
screens using intact organisms, pathway screens, as
2
well as enzyme screens--you know, people have come
3
up empty-handed. So I think that
it is not
4
necessarily the systems biology or the need for
5
targets. I think that would
be--certainly, it is
6
going to help but I think, in honesty, I believe it
7
has to do with chemical diversity and the fact that
8
maybe, because most companies moved out of the
9
natural-product area, might be one factor.
10 But, there again, it goes back to the
11
tools. Natural products are much
more expensive in
12 terms of cost of production but also just in
terms
13
of development and also side effects in many cases.
14
There are a lot of people looking at the reasons
15
why, in fact, we have not had any more success in
16
developing new classes of antibiotics, who, I am
17
sure, know that in forty years of intensive effort,
18
there have only been two new classes.
19 So why is that?
Is it because all the
20
low-hanging fruit has been picked?
Or maybe there
21
are no more. I don't know. But, at any rate, I
22
can tell you that one other really exciting thing
at
238
1
is going on right now and that is the response to
2
the Gates Global Health Challenges.
A lot of those
3
applications deal with drug development discovery.
4
Gates is very committed to helping, in terms of
5
drug development.
6 Again, I just close by saying I am
7
optimistic that people are taking a completely
8
fresh look at how we are doing business; in fact,
9
in the area of anti-infectives, maybe where we do
10
come up with a new paradigm for drug development.
11 DR. SHINE: We
need to load Rob's slides,
12
so we have got a couple of minutes.
Do other
13
people want to raise questions for Gail?
14 DR. THOMAS: I
have got a question. I
15
would certainly like to echo the importance of
16
systems biology. We have gotten
to a point in our
17
training programs now that our young people are
18
coming out of the very advanced universities and
19
they have such a niche in research, you can't find
20 a
pharmacologist or toxicologist that has ever
21
dealt with a whole animal. So
you are talking
22
about systems research. We need
to go back and
at
239
1
train some of these people.
2 DR. CASSELL: I
chair the Board of
3
Directors for the Burroughs Wellcome and I am
4
pleased to see that we have one of our
5
distinguished investigators here on the Science
6
Board. We actually have taken a
very close look at
7 a
lot of issues related to translational research.
8
You know, Burroughs Wellcome Fund has really played
9 a
lead role, I think, in funding investigators in
10
this field.
11 We had Marv Cassman come to our last Board
12
meeting, the former Director of NIGMS at the NIH.
13 I
heard something I thought I would never hear Marv
14
say. He said, "The two
things that are the biggest
15
bottlenecks are the lack of people trained in
16
integrative biology, systems biology, and microbial
17
physiology," which we have been trying to convince
18
him of for twenty years. But I
think you are
19
absolutely right about that.
20 DR. THOMAS:
The other comment is when you
21
look at--again, going back to the critical path and
22
list of priorities, heretofore, we have never had
at
240
1
this globalization concept to deal with and yet it
2
has been really reinforced with the various viruses
3
that we are now dealing with.
4 Two or three decades ago, the U.S.
5
Military of Operations was interested in selected
6
infected disease around the world.
Now, that is
7
all out the window. We have got
to think globally
8
in our whole list of priorities and not just focus
9
on what is a major disease in the United States.
10 DR. CASSELL: I
am glad you reminded me
11
because one of the other things that I hope maybe
12
Dr. Goodman will address is the need to rapidly
13
speed up our ability to develop new vaccines and
14
antivirals and, perhaps, broad spectrum antivirals
15
which, I think, the systems-biology approach would
16
allow us to better do using both the human genome
17
and the infectious-agent genomes.
18 But when you have got something like SARS,
19
or the bird flu, or potentially new strains of
20
influenza, we need to fast-forward anything we can
21
in terms of development and approval.
I don't
22
know, Jesse, if you plan to talk about this, but
at
241
1
John LaMontaigne has been filling my ear full about
2
the need, as far as speeding up the process.
3 DR. SHINE:
Gail. Thank you very, very
4
much. Hopefully, we will have a
chance, at the end
5
of these presentations, to revisit some of these
6
questions.
7 I want to thank Rob Califf, who just flew
8
in. Those of you who know Rob
know that he is
9
probably the nation's leader in terms of academic
10
involvement in clinical trials and evaluation of
11
drugs and has a particularly important perspective.
12 Rob, we are particularly interested in
13
your comments in the area of chronic diseases where
14
the elements in the path that Janet has emphasized
15
are critical. How do you see not
just the
16
fundamental science but getting it through that
17 path
and how we should be thinking about FDA
18
initiatives in that area?
19 Perspective on Chronic Disease Therapies
20 DR. CALIFF:
Thanks, Ken. It is great to
21
be here. Actually, I wish we
would hurry up with
22
the antibiotics. I am running a
CCU today and now
at
242
1
that we are putting devices in everybody, they are
2
all infected, and the old antibiotics don't work
3 very well when they get infected with devices
in.
4 [Slide.]
5 But this is actually from, Gail, your
6
talk. This is a perfect
handoff. I am going to
7
stress, if we don't cut the transaction costs of
8
the public-private partnership, we are not going to
9
solve these problems. It is also
not just
10
fundamental science. I am going
to talk about what
11 I
see as the blocks in the other areas.
12 [Slide.]
13 So the key points are that chronic disease
14
is the dominant health issue. I
think predicting
15
whether chronic-disease therapy causes a net
16
benefit or harm is extremely difficult and the
17
difficulty of this is typically underestimated by
18
almost everybody involved.
19 We all agree, I think, biomarkers and
20
imaging are the way to go. But,
really, mostly is
21 a
screen which I think the document emphasizes and
22 I
really do not believe that this can substitute
at
243
1
for clinical outcomes in most circumstances. Then,
2
finally, there is tremendous inefficiency in
3
clinical trials.
4 I am just going to make these points with
5 a
series of slides very quickly. I will
go through
6
the slides quickly.
7 [Slide.]
8 The average life expectancy for women in
9
our society on the left, 90, almost.
Over 80 for
10 men. So these problems--I am not lamenting these
11
problems to have but, if you go to any hospital in
12
the U.S. today, you are going to find it full of
13
people over age 70 many of whom are quite
14
functional with a variety of chronic and acute
15
illnesses.
16 [Slide.]
17 The dominant forms of illness from the
18
World Health Organization in developed countries
19
and, increasingly in developing countries, are the
20
same old plugged up blood vessels and sequelae of
21
smoking, et cetera, chronic illness.
22 [Slide.]
at
244
1 Disability is dropping dramatically,
2
probably due to better medical care.
Actually,
3
disability is dropping faster age-adjusted than
4
longevity is increasing, and longevity is
5
increasing very nicely.
6 [Slide.]
7 The consequences of this, Dr. Zahooney[?]
8 shows over and over. A
lot of people forget that
9
we are at this phase right now where we have had a
10
relatively flat portion of this curve but we are
11
just hitting the upstroke.
People in specialty
12
practice in the U.S. are feeling this incredibly
13
dramatically right now. There is
an unbelievable
14
shortage of cardiologists, oncologists,
15
nephrologists, all the people dealing--and it is
16
not that we are finding patients that we didn't
17
have before. It is just that we
are in this
18
beginning phase of the elderly group.
19 [Slide.]
20 This is going to impact things in ways we
21
haven't imagined. I don't think
most people have
22
come to grips with this slide yet.
at
245
1 [Slide.]
2 Mr. Greenspan made the point about a month
3
ago. I felt like he was finally
coming clean about
4
this, but this is probably about halfway clean as
5
far as I can tell in terms of the situation that we
6
are in.
7 [Slide.]
8 It is going to be driven even harder by
9
the people that we are dealing--my average
10
75-year-old clinic patient is a lot different today
11
than twenty years ago in a lot of ways.
12 [Slide.]
13 We also have this phenomenon that people
14
tend to forget on the product-development side, and
15
that is we are increasingly sophisticated in
16
measuring whether what is being given to people
17
actually does anything worthwhile.
It is happening
18
well in the postmarketing phase and I think will
19
increasingly happen in the postmarketing phase. It
20
still is interesting that they would see on medical
21
students, as I did today, who have been through
22
several years of basic science and then show up
at
246
1
imagining that what they have heard about in basic
2
science is going to have these dramatic results in
3
clinical outcomes. We are
talking about 10 percent
4
reductions in events with most of the things that
5
we do.
6
[Slide.]
7 This is leading to a view well beyond
8
product development by those paying for medical
9
care that we can actually begin to measure this.
10
This year is the big year. I
don't know whether to
11
be glad or sad that cardiology is leading the way
12
here in that we now feel like, for our big
13
diseases, heart failure and acute coronary
14
syndromes in the hospital, we have got about ten
15
things for each of those that we all agree on that
16
ought to be done routinely and as a standard of
17
care.
18 This year, Medicare is paying us
19
differently based on whether we do those ten things
20
and rewarding hospitals if they post that
21
information on the internet for everyone to look
22
at. This is a big change and, if
you think about
at
247
1
product development and what may drive it in the
2
future, if you don't get into this category of
3
being a performance indicator, it is going to be
4
hard to get your product paid for.
That is the way
5
it is going to be driven.
6 [Slide.]
7 We have looked at what it takes to find
8
out if things work. My
conclusion, unfortunately,
9
is that there are no shortcuts.
This is a whole
10
list of reasons. I am not going
to dwell on them.
11 I
left references. The slides will be
posted. I
12
did send them by e-mail yesterday, but I don't
13
think you got them. There are a
whole variety of
14
reasons for this.
15 [Slide.]
16 I am going to just show you examples of a
17
couple. Most of what we do has a
very small
18
effect. This is a slide that
changed my career
19
when it was published. I thought
I could develop
20
great new treatments for heart attack, the leading
21
cause of death and disability in our society. The
22
bottom line is the best thing we have would take
at
248
1
6,000 to 10,000 patients to tell the difference.
2
In fact, we are just starting a trial of what we
3
think may be a great new therapy.
8,000 patients
4
is what it is going to take. It
takes a global
5
study. If you do anything short
of that, you don't
6
really know.
7 [Slide.]
8 Things that we think ought to work based
9
on surrogates sometimes not only don't work, they
10
kill people. The preeminent
example, of course, is
11
antiarrhythmic drugs and the CAST study.
12 [Slide.]
13 But we can take the entire field of heart
14
failure. On average, if you are
enrolled in a
15
Phase III trial in heart failure, you would be
16
better off getting placebo, not only for being
17
alive but also for quality of life.
None of these
18
were developed in the absence of biomarkers and
19
arrays of biomarkers and lately even transgenic
20
models showing you can reverse the condition in a
21
transgenic-mouse model. That is
the case of the
22
TNF-alpha inhibitors.
at
249
1 [Slide.]
2 So the simple-minded notion that you can
3
measure a few biomarkers and assure the public that
4
you know that you are giving them something that is
5
good for them, I think, is largely discredited.
6 [Slide.]
7 The reason is remarkably simple and
8
published in the Annals of Internal Medicine over a
9
decade ago by DeMets and Fleming who have been in
10
this position over the years to watch the hosts of
11
people coming through with their great treatments
12
and to do the statistics on them, clinical trials
13
and watch the disappointment or the elation that
14
occurs at the end of it.
15 In a simple world, all diseases would work
16
through a linear mechanism and we would be able to
17
intervene on the surrogate and be assured that we
18
would have an effect on the true clinical outcome.
19 [Slide.]
20 But, unfortunately, and I think Dale did a
21
good job of describing it, not only is it not
22
linear. I mean, in the simple
case I gave, there
at
250
1
are just two things going on.
One is the surrogate
2
pathway that you know about. The
other is some
3
other pathway of the disease that you didn't know
4
about. So you have an incomplete
effect or you may
5
not only affect the surrogate but actually have a
6
direct effect on the true outcome that could be
7
good or bad that you didn't think about measuring.
8
Or you could affect all three at the same time.
9 [Slide.]
10 All this, frankly, is unpredictable and
11
leads to this situation with every treatment we
12
give. Even the things we are
absolutely sure
13
work--we have had a rash of left-main disease on
14
our CCU. It is like the 1980s
again. I am not
15
sure why. We know surgery is
beneficial but you
16
are telling people, you have got a 3 percent chance
17
of being dead tomorrow if we do the surgery. It is
18 a
mixture of good and bad.
19 I haven't seen much come out of giving
20
drugs to entire people, every cell in their body
21
that is any more predictable than that.
22 [Slide.]
at
251
1 The king of biomarkers in cardiology today
2
is troponin which, in the course of the decade, has
3
gone from being unknown to being a standard of care
4
measured in 99.8 percent of American hospitals.
5 [Slide.]
6 It stratifies prognosis. I am going to
7
come back to it in just a minute.
The problem with
8
the biomarkers is that it is not enough to have
9
something that predicts out come.
You have also
10
got to show that it predicts the effect of the
11
treatment. Specifically, if we
say it segregates a
12
population where the treatment effect is greater,
13
you now have got a subgroup problem.
So I am going
14
to make a little fun of the FDA but it is really
15
making fun of all of us.
16 We had a great example a couple of years
17
ago, the Val-HeFT trial. It is
sort of what I call
18 a
typical beneficial chronic-disease effect, about
19 a
13 percent reduction in the setting of heart
20
failure in the composite of death and
21
rehospitalization.
22 [Slide.]
at
252
1 As we all do, we looked at all the
2
subgroups and, lo and behold, when we looked at
3
them, patients who were not taking an ACE
4
inhibitor, as you would expect, got a tremendous
5
benefit from the angiotensin-receptor blocker. But
6
if the patients were on a beta blocker already, it
7
wasn't quite so pretty and, in fact, when we looked
8
at patients who were on ACE inhibitor and beta
9
blocker together, there was actually a 42 percent
10
increase in the risk of death.
11 This looked very convincing. The p-values
12
are very small by old criteria.
It was post hoc.
13
But how could you ignore it?
14 [Slide.]
15 So, in fact, the labeling was written. It
16
was actually fairly careful labeling--I did verify
17
just for you all this morning that still there are
18
no doctors who read any labels so don't really
19
worry about it very much--but not recommend a
20
concomitant use of ACE inhibitor and beta blocker.
21
Actually, that by itself is not necessarily bad
22
advice but the implication here, taken by most of
at 253
1
cardiology, was it made sense because if you block
2
both the renin angiotension system three ways, beta
3
blockers, ACE inhibitors and ARVs, it has got to be
4
too much and it has got to be bad.
5 We have not done three clinical trials
6
that have totally refuted this as being a bad
7
thing. It is just not the
case. It was play of
8
chance, a subgroup finding.
9 [Slide.]
10
So this brings us back to the
Richard Peto
11
quote. You could put false
positives here just as
12
well, but the fact is, as you may remember, back in
13
ISIS 2, if you were Gemini or Libra, you got no
14
benefit from aspirin. If you had
the other
15
astrological signs, you got a tremendous benefit.
16 The point I want to make with regard to
17
biomarkers, as we get into the era of protemoics
18
and metabolomics, is that, if we do enough studies,
19
we will be able to replicate the findings. We have
20
actually replicated this finding now in SYMPHONY 2
21
where exactly the same astrological signs have the
22
same effect. It is now submitted
for publication
at 254
1
just as an example.
2 [Slide.]
3 But not only that, if we have too few
4
outcomes, we can get results that used to be
5
considered definitive. You would
never ethically
6
repeat the experiment. In heart
failure, we repeat
7
the experiment and we get results that are
8
sometimes negative and sometimes worse and several
9
detrimental therapies have been kept from the
10
market or taken off early because of the courage of
11
people to go ahead and do the experiment again.
12 [Slide.]
13 So all this is magnified. The subgroup
14
problem is magnified now that we can measure
15
multiple biomarkers. So here is
BNP and troponin T
16
together.
17 [Slide.]
18 Here is BNP and CRP.
You see in these
19
plots that if you got badness in both of them, it
20
is really bad, goodness in both of them, it is
21
really good.
22 [Slide.]
at
255
1 BNP and creatinine clearance.
2 [Slide.]
3 Troponin T and CRP.
So you begin to get
4
ten factorial biomarkers all of which add a little
5
bit and you divide that into the ten factorial
6
subgroups, you have got an enormous mathematical
7
problem.
8 [Slide.]
9 My dean and chancellor are making it more
10
difficult by predicting that we are going to put
11
this together with images and genomic data and
12
somehow you will get a printout that will tell you
13
what to do with your patient.
Well, I think this
14
is just an enormous mathematical problem.
15 [Slide.]
16 So Point No. 1 is biomarkers and imaging
17
are the way to go. It is well
laid out in the
18
document. I think it is
critical. But it only
19
should be used for screening.
And not in my slide
20
is that I think this is only going to work through
21 a
public-private partnership. There is
not a
22
company on the face of the earth as big, Allen, as
at
256
1
you may be that can do enough of the negative
2
experiments to find out what the array of
3
biomarkers is that predicts a beneficial treatment
4
over time, or even toxicity.
5 It is especially bad if the experiments
6
are secret. The companies are
not releasing their
7
negative data. This is creating
an enormous
8
problem. Many of us that make
good money
9
consulting with companies do it because the
10
companies call and say, we have got this problem
11
with a biomarker. What have you
seen with the
12
other companies? You can't tell
us exactly what
13
you saw but you might have some experience because
14
it is not being shared across the industry.
15 I think unleashing this on the public
16
without further study of true outcomes is an
17
enormous game of Russian roulette in chronic
18
therapies.
19 [Slide.]
20 Last point; we waste a lot of money on
21
clinical trials completely unnecessarily. There
22
seems to be a myth, and I don't know where it came
at
257
1
from, that the industry has figured out how to do
2
clinical trials efficiently.
This is not the case.
3 A
huge amount of money is being spent and mostly
4
because of the tremendous risk aversion. I think
5
there are very simple things the FDA could do if
6
there was a carrot added to the simple things that
7
would cause industry to change its behavior.
8 [Slide.]
9 We did a little study--this is in press in
10
American Heart Journal--where we took the
11
information, we created hypothetical clinical
12
trials and asked them how they spend their money
13
and what might be done to change it using some
14
pricing models.
15 Interestingly, there aren't any medical
16
journals interested in this. I
am the Editor of
17
American Heart Journal. That is
why it is
18
published there.
19 [Slide.]
20 But, if you take an average heart-failure
21
trial, now, you are talking about $142 million; an
22
average acute-coronary-syndromes trial, $83 million
at
258
1
being spent. You can see the
breakdown of how the
2
money is spent. Most of the
money is going to
3
collect data that is never used for anything and to
4
fly people all over the world checking the data
5
that is not going to be used for anything to make
6
sure that it matches a medical record which, in and
7
of itself, is inaccurate. It is
really quite
8
remarkable.
9 What is also interesting about the
10
dynamics of the cost here is that when you begin to
11
cut down on some of this nonsense, it has an effect
12
on all segments. So, if you
collect less data, you
13
need less monitoring. If you
require less
14
monitoring, you spend less money paying the sites
15
because they don't have to spend a lot of money to
16
deal with the monitors. You can
read this on your
17
own.
18 [Slide.]
19 What is really fascinating about this to
20
me is that there is nobody studying how to do this
21
and, in fact, trying to get the industry to talk
22
about it is almost impossible because they are
at 259
1
scared to death that, if the question comes up that
2
it might be imperfect, that somehow it will hold up
3
an application.
4 We have a case right now, something that
5
is being reviewed by the FDA. We
had some very
6
creative people who wanted to study learning-curve
7
phenomenon which is a huge issue in medicine now.
8
It is no surprise, when you do a big clinical
9
trial, the first hundred patients, there are a lot
10
of mistakes made. But the
company has blocked us
11
from sending this in in a timely fashion until the
12
FDA review is over because they are afraid that the
13
FDA might look at this and raise questions about
14
the study.
15 [Slide.]
16 So we need people studying
this. If you
17
don't believe this is a problem, I would just
18
remind you, my defense always now is HRT.
19 [Slide.]
20 And remarkable consistency in the clinical
21
trials we are seeing between WHI and HERS.
22 [Slide.]
at
260
1 But I think the lesson here is that we
2
focused on the biomarkers that look good and we
3
forgot that there are at least 20 biomarkers we
4
know of now. In our latest
finding, which I am
5
really excited about, is that HRT improves
6
availability of vascular progenitor cells. So the
7
hottest thing in cardiology, HRT is good for.
8
Unfortunately, when you give it to people
9
chronically, it doesn't do good because there are
10
other offsetting effects.
11 So the bottom line here is I think
12
biomarkers are good but we better work out
13
public-private partnership arrangements that
14
facilitate the sharing of the information and we
15
should really take clinical trials seriously as an
16
enterprise that requires study on how to do it
17
better and more efficiently.
18 Dennis Gillings and I probably only agree
19
on one thing and that is that we could reduce the
20
cost of clinical trials by 50 percent and actually
21
have better data at the end than we currently have.
22 So I will stop there.
I think I was on
at
261
1
time.
2 DR. SHINE:
Terrific. Thank you very
3
much. Lots of provocative
content. Questions from
4
the committee? We have got a
couple of minutes
5
here before we go to our next speaker.
6 You have identified, in terms of getting
7
drug to where it will do the most good.
I have
8
identified concerns about biomarkers.
You have
9
talked about the design and expense of clinical
10
trials. From your perspective,
in terms of the
11
products that you have been involved in, trying to
12
see them move through the pipeline, and given
13
Janet's notion about trying to develop science that
14
would accelerate safety which would accelerate
15
utility which would look at industrialization,
16
could you give us an idea as to where you think
17
that science ought to go?
18 Obviously, the drug trial is a big part of
19
the utility story and so are biomarkers, but I am
20
just curious as to whether you have thoughts about
21
areas where you would like to see additional
22
science underpinning that process.
at 262
1 DR. CALIFF: It
depends on your
2
definition. Let me just talk
about two aspects of
3
science. I think the issues of
the underlying
4
science of cellular networks, proteomics, that is
5
very well described by a lot of people and I think
6
it is going to move forward. We
have got to figure
7
out how to get the NIH better integrated into the
8
rest of the world. There is a
lot of work going on
9 that I think you all know about.
10 But there is also the industrial science
11
of studying what you do. It is a
funny business to
12
me that we do this quantitative stuff and no one is
13
studying how to do it better. I
am sure every big
14
pharmaceutical company has people that are
15
studying, most consultants, by the way.
But there
16
is no national enterprise to study how to do
17
clinical research better and a lot of it can be
18
empirically studied, I believe.
19 We make rules and regulations based on
20
somebody's opinion. We never
study whether they
21
make things better or worse.
22 We took a poll of our investigators last
at 263
1
week and 80 percent say that their lives has been
2
made more difficult. The
majority say they are
3
considering stopping doing clinical research in
4
this country because the transactions have become
5
so complicated and difficult, the cost is
6
skyrocketing and they are not being reimbursed.
7 Every other industry studies itself to
8
make itself more efficient, but not this one.
9 DR. SHINE: We
talked a little bit before
10
you came about the semi-conductor industry,
11
SemiTech and a variety of those models.
We want to
12
come back to them.
13 Rob, are you going to be able to stay for
14 a
little bit?
15
DR. CALIFF: I have got to go back. I am
16
on call today.
17 DR. SHINE: Any
other questions for Rob
18
before he leaves, then? Xavier?
19 DR. PI-SUNYER:
Dr. Califf, I don't
20
know--were you here during Dr. Woodcock's
21
presentation?
22 DR. CALIFF:
No, but I have talked with
at
264
1
Janet quite a bit and she showed me her slides.
2 DR. PI-SUNYER:
What you have said about
3
surrogate markers really goes quite against, I
4
think, what she said earlier about trying to
5
develop these further. You are
very skeptical.
6 DR. CALIFF:
Let's see if we disagree. I
7
think surrogate markers need to be used for
8
screening. Out of the,
hopefully, hundreds of
9
things to come from proteomics, for example, and
10
with proteomic arrays, to pick the things to really
11
try to go the distance with. I
don't know of any
12
other way of doing it. I think
it is the
13
right--and imaging, which she may not have talked
14
about as much, I think is equally as important.
15 It is a political issue, largely, as to
16
when you let things on the market and then how you
17
follow up and make sure you really know at what
18
level you do. For a chronic
disease, you may not
19
be able to wait until you really know.
An example
20 I
used that Larry Kessler and I had several
21
meetings about would be cardiac valves.
You can't
22
wait the eighteen years to know if a cardiac valve
at
265
1
is really safe to put it on the market.
2
That is not right, but you
sure better
3
study it. And we have examples
of that.
4 DR. SHINE: I
want Janet to comment. But
5
what I heard was something a little different.
6
What I heard was a skepticism that final decisions
7
about products that rest only on surrogate markers
8
is dangerous, that surrogate markers for screening
9
makes sense if they are appropriately set up and
10
that maybe surrogate markers, CD4s, viral load in
11
HIV or whatever, which have a high degree of
12
prediction of the outcome, but, until you have
13
predicted the outcome with the surrogate marker,
14
you want to be very cautious about how far that is
15
taken down the--
16 DR. CALIFF:
Let me say one more sentence
17
based on what--if you say "final," yes; I am
18
adamant and even CD4. Talk to
Tom Fleming about
19
CD4 now. Yes; it is a good way
to know that a drug
20
will have an effect. But if you
make rounds with a
21 doctor now, you are talking about dozens of drugs
22
and it doesn't distinguish which--a CD4 count, in
at
266
1
the short term, doesn't tell you, for HIV, which is
2 now a chronic disease, which one is going to
give
3
you better outcomes in the long term.
4 DR. SHINE:
Janet?
5 DR. WOODCOCK:
I think there is a
6
spectrum. I made a distinction
between the
7
predictive, the need for better prediction, which
8
is driving a lot of the costs.
We need to be able
9
to identify, as Gail said, viable candidates and
10
move them along versus confirmation, which I think
11
is what Rob is talking about.
12
Both of us talked, I think,
about society
13
has a need for certainty. Rob
said it better, I
14
think. It isn't just the
FDA. The payers,
15
everybody, people who take the drugs, they want to
16
know they are going to work. So
I don't totally
17
agree with Rob. I think there
are other areas--I
18
think he is influenced by cardiology.
I think
19
there are other areas, my field, for example, where
20
you have very good certainty with a surrogate and
21
they have been very successful.
22 DR. CALIFF:
What field is that?
at
267
1 DR. WOODCOCK:
I am a rheumatologist.
2 DR. CALIFF:
Let's talk about COX2
3
inhibitors, just another paper in Circulation about
4 a
higher MI rate.
5 DR. WOODCOCK:
Yes; but that is a side
6
effect. We agree there are side
effects.
7 DR. CALIFF:
Bob Temple is probably going
8
to talk about this. We will
probably agree more
9
than--
10 DR. WOODCOCK:
Can I just say one more
11
thing, not on the biomarker issue, but I think the
12
issue Rob has raised about the academic base of
13
clinical trials and how we do clinical trials is an
14
extremely important point. I
tried to make that
15
point as well, probably not as well, again, as he
16
did, but it is an extremely important point because
17
that is the enterprise that you are involved in
18
once you get into people.
19 We don't study it.
We wouldn't build
20
bridges if we didn't have engineering schools or
21
they wouldn't be very successful bridges, at least.
22
And that, I think, is the situation we are in.
at
268
1
That is also science. It is not
regulatory. If
2
you are just making a regulatory standard without a
3
science base, again, you are not making the most
4
informed recommendation on what should be done.
5
That is the problem.
6 DR. SHINE:
Thank you very much, Rob.
7
Thank you for whipping up here and have a safe trip
8
back.
9 DR. CALIFF:
Thank you. I would love to
10
argue about this more.
11 DR. SHINE:
Good. I would, too.
12 Our next speaker is Robert S. Langer, Bob
13
Langer, from MIT. Bob is an
extraordinarily
14
distinguished scientist, one of, I think, a
15
relatively small number of people who are members
16
of the National Academy of Sciences, the National
17
Academy of Engineering and the Institute of
18
Medicine which reflects, I think, the extraordinary
19
breadth of his contributions. He
is going to talk
20
about the critical path as it relates to drug
21
formulation and development and tissue-engineering
22
issues.
at
269
1
Bob, thank you very much for
coming down
2
for this purpose. We are
particularly interested
3
in this interface from your perspective.
4 Drug Formulation and Development
5 and Tissue Engineering Issues.
6 DR. LANGER:
Thank you very much.
7
Actually, I was going to say, Ken, that probably
8
all three of those, I was probably that rare
9
instance where all three of those organizations
10
made a big mistake.
11
DR. SHINE: Don't you believe it.
12 DR. LANGER: I
wasn't sure exactly what
13
the best thing would be to do. I
had the pleasure
14
of talking with both Janet and David and others in
15
preparation for this. So I kind
of actually broke
16
this talk up into two parts.
17 [Slide.]
18 Basically, the first half of the talk, I
19
wanted to talk about areas where new development
20
tools, which was kind of the theme of the report,
21
might accelerate progress. I
deliberately tried to
22
pick things that were not in the report because I
at
270
1
thought that that might be more helpful. Also, I
2
have tried to pick things that I think may aid in
3
some of the things that are going on, if you look
4
at it from a more generic perspective.
Hopefully,
5
that will be clear.
6 The two I picked were formulation and
7
glycosylation.
8 [Slide.]
9 The second area that I will end the talk
10
with are sort of new areas that are coming down the
11
pike. These would be areas that
are new
12
technologies that people have heard about it, but
13
there really have been very few, if any, products
14
that have been approved yet. So
these are going to
15
be areas where I think we will all need to think
16
about it even further. Really,
my talk here is, at
17
best, an introduction to what we might think about.
18 I picked nanotechnology and tissue
19
engineering based on my conversations.
20 Let me just start out with formulation.
21
Part of the reason I picked this is because I
22
wanted to pick something I knew a little bit about.
at
271
1
The area that I have worked on for many years has
2
been drug-delivery systems. But
one of the things
3
that I noticed as I have gotten into that field--I
4
probably also had the pleasure of consulting with
5
just about every pharmaceutical company in the
6
world. One of the things that
you
7
notice when you do this is that formulation is sort
8
of the lowest rung on the totem pole.
It comes
9
last. That has actually created
huge problems.
10
When you go to some of the really top companies in
11
the world, what you see is that a drug might have
12
fantastic effects and cells and things like that,
13
but they can't formulate it.
They can't make it
14
soluble. It happens, actually,
40 or 50 percent of
15
the time. You can't get it in
the right crystal
16
structure.
17 What you see is companies literally
18
spending years, many years, sometimes decades
19
trying to figure out whether to kill the project or
20
to solve it. Really, what you
see largely is
21
people working by hand in the laboratory trying to
22
solve it.
at
272
1 A couple of years ago, and part of this is
2
companies that I have been involved with--a couple
3
of years I was talking to some people about this
4
and we decided to try to create a company to help
5
this.
6 [Slide.]
7 In fact, the person who is leading the
8
effort there is--the company is TransForm, but the
9
man who is leading the effort, actually,
10
scientifically is a many named Colin Gardner who
11
used to be worldwide Vice President of
12
Pharmaceutical R&D an Merck.
He would see this
13
problem over and over again and wanted to see if we
14
could make an impact.
15
I will sort of try to
highlight a little
16
bit about what they have tried to do.
This is kind
17
of just a summary and then I am going to show you a
18
video to give you the idea of high throughput.
19 Basically, the idea is generally,
20
traditionally, what would happen is they call it
21
form and formulation. Basically,
that is making
22
drugs in new crystal structures or making them more
at 273
1
soluble, things like that.
Normally, you might do
2
ten or twenty in one of two months.
What they are
3
doing, because of what I will show you, high
4
throughput, now they are able to do up to 20,000 in
5
two to four weeks.
6 They also are using tools that have been
7
talked about in the report, the FDA report, like
8
informatics, so that, once you do this 20,000, you
9
can go back into the next 20,000 and do a lot
10
better. These kinds of tools are
being used in
11
genomics but they really haven't been used in
12
certain other areas.
13 So that is basically the idea. Basically,
14
what I thought I would try to do now, and this is
15
going to go very fast, so it is going to give you a
16
video of what they are doing.
Part of why I wanted
17
to show it to you is maybe people could also use
18
their imaginations.
19 See, here, we are applying to formulation
20
but one of the things you might ask is could you
21
apply this to other things that might also
22
accelerate the kinds of things we are talking
at
274
1
about.
2 So let me try to show you the video.
3 Basically, what you will be seeing, just
4
to give you an introduction, you kind of have all
5
these vials and they actually call them hotels with
6
different rooms. Each of them,
what you will be
7
doing is trying to create a new crystal structure.
8
It is all done by robotics. The
robots are going
9
to put these powders in. Then
they have these
10
on-line systems that measure, like, X-ray
11
defraction for crystal structure or Raman
12
spectroscopy for crystal structure.
You are going
13
to see real-time measurements.
14 Then they have informatics packages that
15
take all this stuff and then organize it. Then
16
they will look at it and say, how many crystals do
17
you have, what are their crystal structures, and so
18
forth.
19 So let me go to that video.
20 [Video.]
21 Here it is.
This is illuminated. It is
22
all robotics. Like, you can see
the vials going
at
275
1
in. You shine the light through
it and you
2
get--this is an X-ray defraction done on line.
3
They are putting it just in again by robotics. All
4
this is real time. They are
looking at it. These
5
are all the Raman spectrographs, real-time. You
6
organize it. You get the idea of
high throughput.
7 They then organize it. They bend the
8
data. This is the
informatics. You see three
9
crystal structures here. Then
you can take that
10
data and now you can organize it.
This is a
11
slightly different way, but, again, you see three
12
crystal structures.
13 This last thing is just a way of having a
14
target. You could have a target
of a certain
15
crystal structure or drug solubility.
This just
16
shows, again, real data how they have improved each
17
time by using informatics package to go up and up.
18 Just to give you a couple of examples of
19
that. What they have done,
actually, for Jim's
20
benefit, they have transformed--this is all
21
public--they have had a very close relationship
22
with Alza so they have been able to do, in a month,
at
276
1
the same number of transdermal experiments that
2
Alza had done in thirty years just because of using
3
these kinds of approaches.
4 [Slide.]
5 I thought a real practical example which
6
has been published and probably very relevant to
7
this audience is actually ritonavir.
This is an
8
AIDS drug that Abbott made.
People may know that
9
they made it 1.5 years after they launched
10
it--well, it was originally in crystal Form 1. But
11
1.5 years after they launched it, it formed into a
12
new polymorph. So it was 50
percent less soluble
13
and they had to basically recall it and try to
14
reformulate it. Basically, they
could never get it
15
back. They could never get it
back to Form 1.
16
Somehow, it had been contaminated and, no matter
17
what they did, it would never get back.
18 But, again, using a high throughput
19
approach, and this was published in Proceedings of
20
the National Academy, in two weeks, basically, what
21
they were able to do is not only make Form 2 but
22
make Form 1 again, and make three new forms that
at
277
1
had never been discovered before.
2 So, basically, this is just one of quite a
3
number of ways of using high throughput to try to
4 aid
discovery and, in my opinion, a very
5
underdeveloped area, formulation.
6 [Slide.]
7 The second area that I was going to go
8
onto was, when I read through the report, and,
9
again, my goal was to try to pick things that were
10
not so much mentioned, but some of the, I think,
11
exciting new tools that are coming along are
12
imaging, informatics, genomics and proteomics.
13 I thought I would pick a fifth one
14
because, I think, again, it is potentially very
15
important and very underdeveloped which is
16
glycomics, sugars. It is
interesting. Whenever
17
somebody gives a protein, you know the sequence.
18
It is a pure molecule. But, if
you think about it,
19
whenever you are giving a polysaccharide or a
20
glycoprotein, that part of it you are not.
21 In other words, if you took heparin, and I
22
will come back to this is a second, heparins
at 278
1
poly-disperse. All glycoproteins
from a glyco part
2
are poly-dispersed. So, as
people start to look
3
forward to maybe making bio-generics someday or
4
things like that, how do you ever know what you
5
have really got.
6 [Slide.]
7 So, anyhow, just to go over a little bit,
8
cracking the code of sugars, you could view in some
9
ways as being analogous to sequencing DNA.
10
Sequencing of DNA, obviously, led to a major
11
revolution but, like DNA and proteins, sugars also
12
can play very important roles.
They exist as
13
sequences and building blocks similar to DNA but
14
here the sequencing has been a lot more complex, a
15
lot harder.
16 As I mention, it is important, I mean,
17
just to pick two drugs. One, it
might be Lovenox,
18
which is a low-molecular heparin.
Let's say one
19
wanted to make betters ones or generic ones.
20
Another example could be glycosylated proteins,
21
like epogen. So these are
important areas.
22 [Slide.]
at
279
1 What's been done, then, or maybe just to
2
go over the problems, so there is an inherent
3
complexity of sugars. There has
been structural
4
complexity and they are information-dense. There
5
haven't been ways to amplify them.
There hasn't
6
been polymerized chain reaction, so to speak. And,
7
as I mentioned, a particular issue is that they are
8
heterogeneous. You don't give
one heparin.
9 So the problem is lack of technology and
10
tools to sequence sugars so it has made it
11
difficult to characterize and engineer them and to
12
decipher their role in biology.
This work was
13
actually, again, published in Science and being
14
developed by another company I have been involved
15
with called Momenta. Basically,
it has been led by
16
Ram Sasisekharan and others.
17 But, in our lab, a number of years ago, we
18
were able to clone heparinase, heparinases, all
19
kinds of heparin-cleaving enzymes.
What they have
20
done, and published in Science, is use these to
21
cleave specific portions of these molecules and
22
then use a vanity of techniques like mass spec,
at
280
1
capillary electrophoresis, NRM and basically have
2
coupled with an informatics package and basically
3
sequenced pretty much any kind of sugar.
4 So this kind of approach may some day, as
5
an example, may lead to being able to make better
6
polysaccharides, better glycoproteins and so forth
7
because now you might have the opportunity to know
8
what you are dealing with.
9 [Slide.]
10 So that is kind of what I wanted to go
11
over in the first part of the talk.
The second
12
part of the talk is more raising questions rather
13
than even giving any suggestions for answers,
14
although I will try to do that just a little. But
15
two future areas I wanted to focus on in medical
16
development are nanotechnology and tissue
17
engineering.
18 Nanotechnology is a term that gets widely
19
used and has many different meanings.
I thought I
20
would pick two examples, though, where I thought it
21
might be helpful to go over in medicine. One is
22
actually things, again, which came out of our lab
at
281
1
which are what I will call microelectromechanical
2
devices. These are what are
called MEMs devices.
3
In particular, these are little microchips, but
4
what you are doing in these microchips is, if you
5
have potent medicines, you can make, and I will
6
show you some pictures in a second, little wells.
7 Unlike regular chips which are all
8
electrical, these chips you can put drugs in or
9
chemicals in. You can store them
there
10
indefinitely. These wells are
nano-sized wells.
11
They are very small but you can pack a good deal of
12 drug
in them. The idea is that they could
actually
13
in the body or in a patch or in a pill, but what
14
you can do, if you want, is you can--and, again,
15
this is still in the animal-testing stage.
16 But, basically, all these little wells are
17
individually addressable. What
you can do is the
18
drug will basically stay there, but if you just
19
apply one volt selectively, and you might do this
20
by telemetry, like the way you open up a garage
21
door. There are all kinds of
electrical things you
22
can do. You can basically cause
the gold to
at
282
1
dissolve.
2 We picked gold here.
We picked gold here.
3
You can also use other materials.
But basically
4
the idea is that you have all these nano-wells.
5
They could have different doses of the same drug in
6
or they could have many doses of a drug. Let's say
7
somebody wanted to do combination chemotherapy.
8
You could literally put a pharmacy on a chip some
9
day.
10 [Slide.]
11 These, as I mentioned, can be made quite
12
small. These are pencils and
here is the chip.
13
They can be made bigger, too, but, basically, the
14
idea is that they are battery-powered, they are
15
going to be controlled by telemetry, ideally. This
16
is, like I say, the way you might open up a garage
17
door.
18 You could base the design based on things
19
like pacemakers, things like that.
Again, all this
20
is kind of futuristic but let me again show you a
21
video to illustrate what might happen.
What I am
22
going to do in this video is show you a single
at
283
1
well, and you have to look, because it happens
2
quickly and then we are just going to apply one
3
volt and that gold will dissolve.
When it does,
4
sort of you will get a change in the background.
5 So we will take a look at this, see if the
6
video works.
7 [Video.]
8 Here it is. We
are applying the voltage.
9
Just like that, it opens up.
And, as soon as it
10
does, whatever is underneath it could come out. It
11
could come out right away if you put no polymer or
12
gel underneath it or it can come out more slowly if
13
you put a polymer or gel underneath it.
14 [Slide.]
15 Here is just an example. These are done
16
in rats. It is actually quite
reproducible where
17
you get multiple pulses at different times. So
18
this is a way of getting a pulse delivery from
19
these little chips. So that is
one example of a
20
nanotechnology kind of thing.
21 [Slide.]
22 The other, which some of these are already
at
284
1
in clinical trials, is to make what I will call
2
nano-sized medicines. The idea
is that--and this
3
is being done by a number of people--but the idea
4
is that you could take drugs and couple them to
5
polymers. These are just five
different designs
6 that people are looking
at, like polymer drug
7
conjugates.
8 The reason people are doing this is
9
because you could lengthen the drug's lifetime in
10
the body. You might, somebody,
and people are
11
trying to do this, target the drug to a particular
12
type of cell, like, say, a cancer cell.
So all
13
these are things that are under active research
14
because they can impart new properties.
15 Also, gene therapy would be another
16
example. People are trying to
couple, make
17
polymers that could deliver a DNA or RNAI to
18
patients somebody. But, again,
these might need to
19
be nano in size so they will be taken up by
20
receptor-mediated endocytosis.
21
[Slide.]
22 So here I just tried to pick some issues
at
285
1
with some of these nanotechnology approaches. How
2
will you assess safety? How do
you characterize
3
them? These are whole new issues
that will come
4
up. What are the right
biological
5
characterizations and what are the right
6
physical-chemical characterizations.
What animal
7
models are appropriate and, of course, finally,
8
which I didn't even get into because we are not
9
there, what human tests will obviously need to be
10
done.
11 [Slide.]
12 The last area that I wanted to go over
13
today, and, again, it is a future area, is tissue
14
engineering. Of course, so much
of what we always
15
here about are drugs or medical devices, but tissue
16
engineering really addresses things that are very
17
hard to fix with either drugs or devices.
18
Just to give you some
statistics from an
19
article I wrote in Science with Jay Vacanti a few
20
years ago, this is just hugh.
Annual tissue loss
21
and end-stage organ failure can be up to $500
22
billion in healthcare costs.
Part of the reason
at
286
1
for that is these are serious diseases where people
2
are dying of liver failure or heart failure, so you
3
have to do surgery. They are in
the hospital for
4
long times. There are 8 million
surgical
5
procedures per year.
6 [Slide.]
7 On this slide, I will just go over it
8
fairly quickly. These are just
some of the
9
incidences, like, again--or you might need to do
10
reconstructive surgery so you see problems with
11
bone, cartilage, tendon, skin, blood vessel, liver,
12
pancreas, heart. And this is a
very incomplete
13
list, but it just gives you the idea that the
14
numbers are big and there are not good ways to
15
solve it.
16 [Slide.]
17 So tissue engineering, the paradigm that
18
we started thinking about, and I will just sort of
19
go through it quickly, is that, what you might do
20
some day is take isolated dissociated cells--these
21
are bone, cartilage, liver, intestine, urothelial.
22 I
will mention in a minute, and obviously, this
at
287
1 will be another big
issue for you some day is stem
2
cells. What you might do is take
stem cells and
3
convert them to one of these.
4 What we have done is we have put them on a
5
polymer to give them the right kind of structure.
6
You can make them into literally any shape or form
7
and use the polymer as a template for the cells to
8
reorganize depending on the polymer chemistry. I
9
will, again, give an example of two.
10 Grow them outside the body and then put
11
them inside the body where they might form a new
12
tissue.
13 [Slide.]
14 Just to show you a few pictures. This is
15 a
scanning electron micrograph of the polymer
16
fibers with cells on them. This
is liver cells.
17 [Slide.]
18 But then, to sort of illustrate the shape
19
idea, I thought I would pick an example that
20
Prashad Shastri who is now at Penn, he is going to
21
Vanderbilt, did is let's say we moved--I will make
22
this up--twenty or thirty years into the future and
at
288
1
somebody comes in and they want plastic surgery.
2
They say, "We would like a new nose."
3 Maybe they want an upturned nose, or maybe
4
they will want a hooked nose.
Probably not, but
5
the point is what I believe will happen someday is,
6
by using polymer fabrication techniques, coupled
7 with cad-cam techniques like computerated design,
8
you will be able to make a nose in any shape you
9
want or anything else.
10 So here is one of the noses that was made.
11
Then you could take cartilage cells on.
In fact,
12
even though the nose is not being done clinically,
13
there are examples out in clinical trials where
14
people are taking cartilage cells from minimally
15
invasive procedures like ear biopsies,
16
arthroscopies, and making tissues for different
17
cosmetic or structural things, usually under
18
physician-sponsored INDs.
19 [Slide.]
20 Anyhow, I thought I would just give you
21
two examples to end the talk of tissue engineering
22
to give an idea of some of the considerations. The
at
289
1
first example is blood vessels.
This is very
2
multidisciplinary, too, so it has been very hard to
3
make blood vessels less than 6 millimeters in
4
diameter. In fact, nobody has
succeeded. That is
5
why they do a transplant.
6 [Slide.]
7 But here is a polymer that is 3
8
millimeters in diameter, 97 percent porous. This
9
work was done by Laura Nicholson at Duke when she
10
was a post-doc in my lab. But,
basically, the
11
polymer is chemically modified.
A blood vessel, of
12
course, has smooth-muscle cells--that is SMC--on
13
the outside and endothelial cells on the inside.
14 But one of the things that was tricky to
15
make blood vessels are, see, normally, if any of
16
you grow cells in culture and you go to any lab,
17
what you will see is they will sit there in 96 well
18
plates or tissue-culture flasks.
They will just
19
sit there. But tissue
engineering is more complex.
20 What we found is that none of those would
21
work to make blood vessels. What
Laura did was
22
recognize that, if she was going to really get this
at
290
1
to work, she had to create an environment much like
2
the body. So, in the body, blood
vessels are not
3
sitting there in an incubator.
They are actually
4
hooked up to a pulsative pump, your heart.
5 So she actually created a series of
6
bioreactors which would mimic that and create
7
pulsatile radial stress.
8 [Slide.]
9
Here is a picture of
that. She figured
10
out the right media. This was
years of work.
11
Pulsatile pump beating at 165 beats per minute,
12
pulsing it over eight weeks.
13 [Slide.]
14 At the end of that time, you could
15
actually make a little blood vessel.
16 [Slide.]
17 And, if you characterized it--this we also
18
published in Science. It is 50
percent collagen.
19
It is quite strong. Its rupture
strength is
20 greater than 2000 millimeters of
mercury. You can
21
suture them into animals. It has
got similar
22
pharmacology. We just picked
three markers, two
at
291
1
regular blood vessels.
2 [Slide.]
3 These were then put into pigs as the
4
animal model and, using angiography, you could see
5
the blood vessels are open months later.
6 One of the other issues that will come up
7
with a technology like this is how do you get the
8
cells. Well, one possibility is
the patient, but
9
if you try to get them from the patient and
10
somebody had a heart attack and you wanted to do
11
this, then how do you grow them up?
It is going to
12
take a long time so you might need a cell source.
13 That is where stem cells come in.
14 [Slide.]
15 Here is another study, this is just a
16
summary, that Levenberg in our lab did.
This was
17
published in PNAS. She was able
to take human
18
embryonic stem cells and actually clone out by
19
characterizing them appropriately the--figuring out
20
the right day, basically, and the right antibodies,
21
she was actually to clone out human embryonic
22
endothelial cells and then put those in the
at
292
1
polymers.
2 These are, just, again, some markers.
3 This
is histology, platelet endothelial-cell
4
adhesion Molecule 1, CD34. But,
again, if you look
5
closely, you can see the lumen in the animals and
6
the blood vessels in there. So
that is one example
7
that I wanted to give you and I will come back to
8
this in just a second in closing.
9 The other example, very quickly, is tissue
10
engineering, I think, actually provides a way
11
where, someday, you might even be able to aid in
12
spinal-cord repair.
13 So I thought I would pick another example
14
where Erin Lavick did this study.
She, basically,
15
working with Evan Snyder, took neuronal stem cells,
16
put them on a special polymer she designed and did
17
about 50 rats. Most of these
rats are basically
18
made--well, I should say all the rats were made
19
paraplegic.
20 Let me just show you what happens
21
normally.
22 This is 100 days after, and you focus on
at 293
1
the rat's hind legs. This is
another video. When
2
you make the rats paraplegic, here is what happens.
3 [Video]
4 He is just sort of dragging his feet.
5
This will go on for about 15 seconds.
Also, the
6
paws are splayed in a kind of funny fashion.
7
Again, it is very important to realize that there
8
is a huge distance, I am sure everybody knows, from
9
animal models to humans, and this experiment is
10
still very early stages. But it
will illustrate
11
what you can achieve some day.
12 The experiments were basically--the
13
controls were nothing, the neuronal stem cells by
14
themselves, the polymer by itself.
15 [Video]
16 Now I am going to show you 100 days out
17
when these stem cells and the implant were put in
18
in the experimental animal within 24 hours of the
19
injury. Here you see a marked
improvement. This
20
is actually not our best animal, either. These are
21
the means from the paper. Again,
you can see this
22
animal, as pretty much all of the treated ones do
at 294
1
100 days out, are able to support their own weight
2
and the paws are splayed in a much more normal
3
fashion.
4 Again, this will be many years before this
5
is ready--again, look at the paws--before this is
6
ready probably for FDA review, but it gives you an
7
idea of what you might see someday.
8 [Slide.]
9 So, to end the talk, as I look at this
10
area, tissue engineering, here are just four points
11
that I wanted to raise; again, how should safety be
12
assessed? What are the
appropriate markers? How
13
do you judge success? Also, how
do you judge it
14
from a functional standpoint?
Say, you took the
15
blood vessels of the spinal-cord--by the way, these
16
are just examples. You can take
this across twenty
17
or thirty different tissues or organs.
And, again,
18
what are the appropriate animal models and then,
19
obviously, as you go further, how does one design a
20 clinical trial.
21 Basically, that is what I wanted to go
22
over today to both give you an idea of certain
at
295
1
areas which might be able to help accelerate
2
progress and others that I am sure we will need
3
plenty of help.
4 Thank you.
5 DR. SHINE:
Just terrific. Again, we see
6
the opportunity that you have taken to move
7
biology, physics, engineering, technology,
8
materials together in a remarkable way with a
9
variety of these kinds of approaches.
10 I guess the principle question that I have
11
for you is, given what we have been talking about,
12
namely what has to happen, how do we develop the
13
science that underlies the process for moving drugs
14
from one end of the spectrum to the other in terms
15
of applicability, your work suggests that there is
16 a
real potential for individual investigators or
17
groups of investigators doing the research which,
18
in fact, could underlie that process.
19 We have been struggling with the question
20
of who would fund that research and whether one has
21
to use, for example, public-private approaches,
22
industry-private approaches, and so forth in order
at
296
1
to see whether one can create enough of an activity
2
so that it could satisfy the needs of the process.
3 So I am just curious as to your
4
perspective on where the kind of unique
5
capabilities you have fit into this overall area of
6
science in support of the drug development and
7
application process.
8 DR. LANGER: I
think it has to come from
9
both. I think the important
thing that I read in
10
the FDA report is really the emphasis on not just
11
basic research but on these kinds of research. I
12
think, as the government--if the government, say,
13
NIH is the obvious place, but there are obviously
14
other government arms--looks at that, I think that
15
is an important step.
16 I mean, we have seen some of that from the
17
NIH in recent years where they are not just funding
18
hypothesis-driven research but also
19
technology-driven research. So I
think certainly
20
the government and I think reports like the one
21
that was written help on that because they create
22
awareness.
at
297
1 I also think, if the opportunity is there
2
for companies, both small companies and large, to
3
make money, there also is an opportunity there.
4
The companies that I have mentioned, I think, the
5
small companies like TransForm or Momenta, they
6
basically saw unmet needs that they felt were huge
7
and, therefore, could create opportunities where
8
they might get funding from larger companies or
9
might be able to create brand-new products
10
themselves.
11 I think that is sort of the good thing
12
about America. You have this
whole
13
venture-capital, entrepreneurial, area and people
14
see those opportunities. They
say, boy, here is an
15
area that people haven't focused on at all. Maybe
16 I
can make an impact.
17 But my feeling is what has been good
18
about, what I enjoyed reading in the report, is I
19
think it creates the awareness, to the extent that
20
various groups read it, on everybody's part. This
21
is an important area that could accelerate
22
progress.
at 298
1 DR. SHINE:
Thank you. Does anybody have
2 a
quick comment before we take a break?
Also,
3
hopefully, Bob will be around in the next few
4
minutes.
5 DR. LANGER:
Yes; I will be around until
6
4:00 and then I have to give a lecture at another
7
part of FDA.
8 DR. SHINE: If
not, why don't we go ahead
9
and take a break. We will
reconvene at 3:20
10
promptly so that we can get to our last three
11
presentations
12 Bob, thank you very, very much.
13 In the next few minutes, we are going to
14
get a perspective with regard to the role of the
15
critical path in drugs, devices and biologics from
16
folks within the agency who have responsibilities
17
in this area. The first will be
Bob Temple who is
18
Director of the Office of Medical Policy, Center
19
for Drug Evaluation and Research who is going to
20
give an overview of opportunities related to drugs.
21 I believe that is what you are going to
22
do, Bob.
at
299
1 DR. TEMPLE:
Right. If my slides are
2
loaded--
3 Overview of Opportunities
4 Drugs
5 DR. TEMPLE: I
will talk very fast and
6
will say nothing about biomarkers, by the way.
7 [Slide.]
8 I want to talk generally about efficiency
9
in drug development. As I go, I
will tell you why
10
that seems relevant. In
particular, I want to talk
11
about how to make studies that give you the answer
12
more unequivocally than we now do.
That is the
13
general message.
14 [Slide.]
15 It is worth asking what would represent
16
efficiency in drug development.
The first most
17
obvious thing you think of is smaller studies,
18
simpler studies, collecting less data, as Rob said,
19
doing trials in other countries, having a network
20
that you just drop trials into, using central IRBs
21
so you don't have to go to a million different
22
ones, all that kind of stuff.
at 300
1 There are real possibilities in all those
2
things but I am not mostly going to talk about
3
that. Some other day with
another twenty minutes.
4 [Slide.]
5 What I want to talk about is improving the
6
quality of development so that you get valid
7
answers earlier, in particular that you learn
8
whether your drug is likely to work in Phase II
9
which is when you are supposed to learn that, or we
10 always
thought so. If you knew that, you could
11
terminate development of things that aren't going
12
to work or that are unsafe earlier.
You could
13
proceed into development only with drugs that are
14
likely to succeed.
15
Also, you wouldn't lose
effective and safe
16
drugs because you got the wrong dose or something
17
like that. So that is what I
want to talk about.
18 [Slide.]
19 I have been interested in efficiencies in
20
design for a long time, but we were recently told
21
of something that, to me, is unbelievable and
22
astonishing. PhRMA says, and
there is no reason
at
301
1
not to believe them, they should know, that of
2
drugs completing Phase II, about half of them fail
3
in Phase III and often because they can't be shown
4
to be effective. Now that, to
me, is amazing. It
5
is not what we always thought the way things were.
6 You are supposed to know in Phase II, for
7
most kinds of drugs, whether they are going to
8
work. That is what Phase II is
for. So what is
9
the matter? Well, we don't know
what is the matter
10
and I would say to PhRMA that their number-one task
11
is to find out. They ought to
review all these
12
failures and figure it out.
Doing anything else
13
doesn't make any sense. But they
don't work for
14
me.
15 We have some thoughts, though, about what
16
could improve this and that is what I want to talk
17
about.
18 [Slide.]
19 We are doing a number of things--I think I
20
won't dwell on this--to try to meet earlier with
21
companies, to try to give what information we can
22
and what we know about because we see a lot of
at
302
1
mistakes. Those are going on
and, because I have
2
got to finish on time, I won't do that.
And we are
3
writing guidance, too.
4 But the failures have to be understood.
5
Some of them are going to be unavoidable. Some
6
rare adverse effect comes along.
It is an
7
hepatotoxin. Okay; those drugs
are going down.
8
Sometimes, there is no valid biomarker.
9
Platelet-active drugs don't have a plausible
10
biomarker so you have to do a 3,000-patient study
11
to see if they work and sometimes it doesn't work
12
out. You can't fix that. Sometimes adverse
13
effects show up after a long time.
14 [Slide.]
15 But sometimes there are avoidable
16
problems. For example, people
don't use biomarkers
17
that do exist to find the right dose and they carry
18
out all the studies at the wrong dose, too high,
19
toxicity, too low, doesn't work.
They don't study
20 a
wide enough dose range so that things don't work
21
out right and there is nothing they can do about
22
it.
at
303
1 Sometimes, the Phase II study isn't really
2
positive but there is a subset in which it might
3
work, and it was trending favorably.
So they
4 launch
Phase III and they haven't really got a
5
basis for it. That is, as Lou
Sheiner would say,
6
moving to confirm before you have learned enough.
7 I think failure to consider dose-finding
8
in Phase III is a bad mistake for reasons that I
9
will get into. Sometimes, they
don't do
10
appropriate workup. And then,
sometimes, there is
11
subset chasing, a very dangerous thing because
12
subset findings are unreliable as Rob showed you
13
even when we put them in our recommendations.
14 He didn't tell you that the rest of the
15
labeling was very clear that that was not
16
necessarily true.
17 [Slide.]
18 What I want to talk about is ways that
19
Phase II controlled trials could be redesigned,
20
designed differently, to give more unequivocal
21
answers. I am going to talk
about three
22
possibilities. One is the
general concept of
at 304
1
enrichment, putting people into trials who are
2
likely to have larger effects because then you get
3 a
surer answer.
4 Reversing the usual sequence, we usually
5
randomize people to drug or placebo.
You can take
6
people who are on drug and do a randomized
7
withdrawal and, in some cases, find a much more
8
efficient design. If I get to it
and have time,
9
which I don't think I will, I will talk to you
10
about a titration design that Dr. Sheiner
11
suggested. I can't mention his
name without noting
12
that he is in terrible distress and we are all very
13
saddened by that. Anyway, I will
see if we get to
14
that.
15 [Slide.]
16
Enrichment, as a general
concept, is any
17
selection maneuver that makes the population more
18
likely to be able to participate properly in the
19
study, have the endpoint of interest or respond to
20
treatment because all of those things increase the
21
study power for showing an effect.
That is sort of
22
obvious. Enrichment goes on all
the time, even if
at
305
1
it is not identified that way. We
always do that.
2 Some of it, as I said, is routine and
3
nobody has any trouble with it.
You try to find
4
likely compliers. You choose
people who won't drop
5
out because of other diseases.
You have a placebo
6
baseline to eliminate the people whose condition
7
goes away. You eliminate people
who give
8
inconsistent treadmill results in an angina trial.
9
People do these all the time.
Actually, it is not
10
absolutely true that nobody doesn't like this.
11
Sometimes, they say, oh, then it won't be
12
generalizable. But, this is
fairly common
13
practice and everybody does it.
14 People who like to model trials can build
15
these things into their models and see how helpful
16
it is likely to be. That is
pretty standard, not a
17
big deal.
18 [Slide.]
19 Where you understand that one disease is
20
different from another disease, it is perfectly
21
acceptable to do that. Just to take some obvious
22
examples, we know now, because we are smarter, that
at
306
1
edema can be hepatic, renal or cardiac.
You
2
wouldn't study an inotrope in renal edema. That
3
would be silly. Nobody has any
troubles with
4
those.
5 Now that we are even smarter, we know that
6
heart failure can be systolic or diastolic
7
dysfunction. So, if you want to
study an inotrope,
8 you don't study it in
diastolic dysfunction. It
9
will make it worse. So those
kinds of distinctions
10
are fine and we distinguish causes of pain and
11
study particular kinds. We do
that all the time.
12 [Slide.]
13 You can do the same thing in places
where
14
we don't do it. For drugs that
have a
15
renin-angiotensin mechanism, you could study them
16
in high-renin hypertension. I am
sure you could
17
get a smaller trial, get better dose response if
18
you did it, but working up renin-angiotension
19
systems is sort of hard so nobody bothers. But you
20
could.
21 We are going to see a time when well
22
established genetic differences, sometimes
at 307
1
descriptive, sometimes really pathophysiological,
2
are going to be the basis for identifying people
3
who do or don't have a disease and who are not
4
likely to respond. I guess what
impressed me
5
recently is a study in the New England Journal that
6
showed that a high placental growth factor predicts
7
the likelihood of an MI, is four times greater than
8
in people who have a low placental growth factor.
9 Well, if you are going to do a trial of
10
preventing heart attacks, who are you going to put
11
in your trial? High placental
growth factor.
12 [Slide.]
13 Something that is not done because it
14
raises a question about generalizability is, in
15
difficult cases, to examine the patient response
16
before entry to identify likely responders. And
17
you then put them in a rigorous trial.
That makes
18
people nervous, and I am really talking here about
19
Phase II. There might be another
trial to pin down
20
the more general population, but this might give
21
you a better answer.
22 People do these sorts of things. CAST,
at 308
1
ill-fated as it was, was conducted in people who
2
had a 70 percent reduction in their ventricular
3
premature beats based on the idea that, if it
4
didn't do that, how could it possibly work. Very
5
sensible.
6 We have studied topical nitrates in people
7
who have a known blood-pressure response to
8
sublingual nitroglycerine. If
they can't have
9
that, then they are not going to respond to the
10
nitrate.
11 [Slide.]
12 You could also screen for the true drug
13
response and then randomize only the responders to
14
get an idea of who is going to respond and then do
15 a
rigorous study to really prove it. In
fact,
16
Oates, Woosely and Roden did all their
17
antiarrhythmic studies that way.
They gave the
18
drug open and then they found the responders and
19
put them into a trial. Thy
didn't emphasize that,
20
but that is what they did.
21 [Slide.]
22 One might also look at the history of
at
309
1
response to a class. You
probably can't see this
2
very well, but the third study down,
3
Sanchez-Torres, was in known responders to
4
diuretics. In this study of
indapamide, they had a
5
response at 5 milligrams of 37 over 15.
That ain't
6
the normal response to a diuretic.
So the
7
possibility that you could shorten the development
8
process by picking people known to respond is of
9
interest.
10 [Slide.]
11 There are many other possibilities where
12
this might be an efficient mode of development.
13 Where,
as a general matter, you do a clinical
14
screen, you give the drug to the group, find the
15
responders, take the drug away and than randomize
16
to the drug. It has been
notoriously difficult in
17
GI diseases--not things that reduce acid. Those
18
have been pretty easy--but GI diseases where you
19
are trying to modify motility and do something
20
useful, it has been notoriously difficult to show
21
that these work. Cisapride,
domperidone, things
22
like that.
at
310
1 One possibility is to do that sort of
2
screen, look at the apparent responders and then
3
randomize and see if you can confirm what appears
4
to be a response. All of things
make the
5
assumption and depend on the idea that, within a
6
group of apparently similar people, there are, in
7
fact, subsets--genetic, we don't know.
You don't
8
have to know--who respond differently.
The reason
9
it is so hard to show anything in GI disease is
10
that it is really multiple diseases, different
11
mechanisms and so only 10, 20 percent of the
12
population can respond to any given thing. So that
13
is a possibility.
14 The various pulmonary, anti-asthmatic
15
drugs, other than beta-agonists, have
16
disappointingly small average effects.
Well, one
17
possible reason is that there are really multiple
18
diseases with different mechanisms and we don't
19
know it yet. Well, there is a
possibility that you
20
could find out. I won't give my
cromolyn anecdote.
21
It involves my son being cured of his asthma by a
22
drug that hardly works on average, suggesting to me
at
311
1
that there is a subset.
2 I just want to point out, any randomized
3
withdrawal study, which I will describe, has some
4
elements of this.
5 [Slide.]
6 A third opportunity, really important, is
7
picking people likely to have the event. Now, this
8
is automatic in a symptomatic disease.
Everybody
9
has the symptoms so 100 percent of the people are
10
candidates to get whatever you want.
11 If you look at outcome trials, whether it
12
is in cholesterol lowering or heart failure, we
13
have systematically studied the sickest people
14
first, shown an effect in them and then moved down.
15
Well, the sickest people have more events. In
16
consensus, a study of maybe 400
people, the
17
mortality was something like 40 percent at six
18
months; a lot of events, easy to do the study. And
19
you saw the same things in cholesterol where you
20
started with 4S where everybody had cholesterol of
21
260 and recently had a heart attack, and then we
22
have been moving down.
at
312
1 We already know that certain genetic risk
2
factors are going to be able to do the same way.
3
If you want to do prophylaxis against ovarian
4
cancer, you pick people with the characteristic
5
genetic markup that says they are going to get
6
ovarian cancer. So we are going
to see more and
7
more possibilities like that.
8 There is another more mundane example.
9
One of the most successful sets of treatments for
10
motility-modifying drugs has been, and also for
11
acid resisters, in preventing heartburn in response
12
to meals. The way you do that is
you find people
13
who respond to a classic meal.
The classic meal is
14
pizza with everything and unlimited chianti wine.
15
You put them in your trial and very modest effects
16
have been successfully demonstrated in those trials
17
because the population is enriched with people who
18
will have the disease instead of waiting for them
19
to have it.
20 [Slide.]
21 I guess I want to say this again. An
22
enriched population has to be recognized for what
at
313
1
it is. It is not a slice of general
population,
2
but it tells you--it provides proof of concept. It
3
tells you this is a therapy that, at least in
4
somebody, is going to work.
Then, I would argue,
5
in Phase III, you go find out who it works in and
6
pin that down.
7 How is my time?
8 DR. SHINE: You
have got eight minutes.
9 DR. TEMPLE:
Ach. I can talk slow. In
10
that case, maybe I will go back to that anecdote.
11 [Slide.]
12 In 1975, Amery, a Belgian cardiologist,
13
proposed what he thought was a more ethical design
14
for doing angina trials. That is
because, in 1975,
15
our criteria, our requirements for angina trials
16
involved more than a six-month placebo-controlled
17
period. It is really hard to
imagine.
18 He thought there were better ways to do
19
it. What he said was, instead of
randomizing
20
people with angina to drug or treatment, he gave
21
people the treatment, which they thought was going
22
to work, and then randomized people, in fact,
at
314
1
people who appeared to respond to it, to either the
2
test drug or to placebo. You can
randomize to
3
multiple doses and get dose-response information
4
out of this.
5 The endpoint could either be
6
time-to-failure. Your angina
comes back, you are
7
out of the study, and a lot of people find that
8 attractive because you
don't get treatment that
9
doesn't work for a long time, or some conventional
10
measure like angina attacks per week.
11 This is actually the way antidepressant
12
maintenance studies are carried out now, in some
13
sense, how else could you carry them out. But it
14
is useful whenever you want to assess long-term
15
effectiveness but couldn't use a long-term placebo.
16
Nobody is going to have a six-month
17
placebo-controlled hypertension trial anymore. It
18
wouldn't be ethical, but you can test for
19
persistent effectiveness by taking the drug away
20
randomly at the end of six months and seeing if the
21
pressure comes back up.
22 You probably don't want to withdraw it
at
315
1
abruptly because there are withdrawal effects,
2
things like that.
3 [Slide.]
4 That could also be used for blood sugar,
5
conceivably for heart failure, things like that.
6
It turns out that design is very attractive for
7
pediatric studies because our pediatric committees
8
really don't like the idea of having kids having
9
symptoms for a long time. In
this design, as soon
10
as they have a symptom, they are out of the study.
11
They have failed.
12 It is also worth noting that in
13
depression, which we have been reading about all
14
the time, where standard placebo-controlled trials
15
fail about half the time, these trials almost never
16
fail. We are aware of just one
trial that failed.
17
That is because the population is enriched because
18
all the support that goes one in a
19
placebo-controlled trial isn't there because you
20
wait until the event happens. It
doesn't have to
21
happen at any particular time.
Anyway, it is a
22
very attractive kind of design.
at 316
1 [Slide.]
2 I want to show you two examples. This
3
goes back a long time ago. This
is nifedipine in
4
vasospastic angina. This is the
original data we
5
were given out of what was described as a study but
6
really wasn't. It was just a
hundred people who
7
were treated.
8 Our advisory committee said, we don't know
9
what the natural history of vasospastic angina is.
10
There had never been a drug for it.
So maybe it
11
was just going away. So we got
Pfizer to do a
12
trial, a very simple trial. They
took a bunch of
13
people who were on nifedipine of which there were
14
many. They watched them for two
weeks to make sure
15
they were paying attention and then did a
16
randomized withdrawal study of four weeks duration.
17 [Slide.]
18 You can see this was not a very large
19
trial, 28 people. I also should
note that the
20
patients were all available. You
didn't have to
21
wait for people with a rare disease to come to your
22
clinic. There they were being
treated because of
at
317
1
the open follow up.
2 Early withdrawal was the primary endpoint.
3
They had five of those and one person who should
4
have withdrawn but didn't know she was supposed to
5
and had a heart attack, so it is either five or six
6
versus zero. The other measures
of success all
7
were quite good and we were able to approve the
8
drug on the basis of this extremely easy and
9
extremely rapid study.
Obviously, only the people
10
who were doing well on the drug got into this trial
11
so it is enriched.
12 [Slide.]
13 Recently, we approved sodium oxybate for
14
cataplexy with narcolepsy. We
had one conventional
15
trial where people were randomized but a second
16
confirmatory trial was done in people who were on
17
the drug because they were doing well on it. And
18
they were all, obviously, quite available. You
19
just had to call them on the phone because you were
20
providing the drug for them.
21 We looked at median attacks per two weeks
22
in people who were randomized to placebo or to
at
318
1
sodium oxybate after being on it.
So, at baseline,
2
the placebo group had had four attacks per two
3
weeks. They went up to 21. The sodium oxybate had
4
had 1.9 attacks per week and they stayed at 1.9.
5
So, in no time at all, you have a nice confirmatory
6
study, very solid.
7 [Slide.]
8 Other things you could do with randomized
9
withdrawal is make this subset observation, which
10
is highly suspect, into something real.
It might
11
take more than one trial group to do it, but, the
12
overall study is negative, maybe leaning favorably
13
and it is very tempting to find the obviously
14
pharmacologically based subset--all
15
pharmacologically based subsets seem obvious when
16
the drug looks better in them--and actually test
17
it; that is, take those people and do the
18
randomized withdrawal study, thus turning an
19
observation that is highly suspect and likely to be
20
misleading into something that could be pretty
21
solid.
22 [Slide.]
at
319
1 There are some other interesting things.
2
Sometimes, a drug has some side effects. You are
3
not sure whether you really think it works well
4
enough. One way to find out is
to take people who
5
seem to have had a dramatic response and, of
6
course, in a controlled trial, you don't really
7
know whether they did. You just
see a range of
8
relationships. You look at
medians. Take that
9
population and do a randomized withdrawal study and
10
see if they fall apart.
11 We suggested that be done, actually, for
12
lotrenex and our suggestion was not taken. There
13
was probably other evidence that some people
14
responded very well, but that would have been an
15
unbelievably quick way to get an answer to that
16
question because all of the people were out there.
17
They were on the drug. It wasn't
done.
18 [Slide.]
19
The randomized withdrawal, as I
said, has
20 a
lot of advantages. It is enriched; that
is, you
21
are likely to be able to see an effect if there is
22
one because you are picking people who seem to have
at 320
1
had a response. And if there is
a mixture of
2
potential responders and non-responders, this is
3
the way to find them, and do the study in the
4
people who do it.
5 It is efficient.
They already exist. It
6
has an ethical attractiveness because, as soon as
7
people go bad, they are out of the trial. It could
8
verify a subset finding and it is easy to turn
9
these into dose-response studies.
10
[Slide.]
11 So I have got a couple of minutes. Let me
12
just touch on dose finding. We
won't really know
13
this until PhRMA or somebody goes back and sees why
14
all these trials, all these drugs, fail in Phase
15
III. But one of the things that
we believe is the
16
dose-finding isn't very well done.
17 [Slide.]
18 Some people think that dose finding is
19
finished in Phase II. That is
just totally wrong.
20
You have very little safety information. Phase III
21
should be filled with dose-response studies. That
22
is true in hypertension and it is not reliably true
at
321
1
on some other places.
2 [Slide.]
3 Obviously, everybody should use their
4
pharmaco--this is for Larry Lesko.
They should all
5
use their pharmacodynamic information to design
6
their dose-response trial.
7 [Slide.]
8 Sometimes, you can redesign the trial. I
9
have given this example so many times I am sure Bob
10
Meyer would wish I would go away.
But it is a
11
striking observation that, if you do field studies
12
of antihistamines, you put hundreds of people into
13
each group and you fail a large percentage of the
14
time. That is because you can't
control the
15
pollen--did I say for allergic rhinitis?
16
Antihistamines in allergic rhinitis.
17 It is also clear that if you blow the
18
antigen into a chamber in a controlled way, much
19
smaller studies give much clearer evidence of
20
effectiveness and you can learn things very well
21
about dose response, time-of-onset,
22
duration-of-action, and stuff like that. So this
at
322
1
is what people do now. They do
their workup in the
2
chamber and then they confirm it with a field study
3
and everybody is reasonably happy.
4 What I am hoping we will find over the
5
months and years that we are working on this is
6
other examples of where you can enormously increase
7
the efficiency of trial design by doing similar
8
things. We will ask people. I don't have other
9
examples yet, I have to tell you.
10 [Slide.]
11 One thing that is never done but could be
12
is to carry out dose-response studies related to
13
effectiveness in people who are known to respond to
14
the drug. What is the point of
putting people who
15
can't respond into a trial to find out what the
16
dose response is when they can't possibly respond.
17 [Slide.]
18 You have already seen that.
19 [Slide.]
20 Other places to do it. Study a full range
21
of doses.
22 [Slide.]
at
323
1 There is a small pitch for examining the
2
maintenance dose through a randomized withdrawal
3
study, and the last thing, which I am not going to
4
be able to get to but I will want to mention it is
5 a
design that Lou Sheiner proposed in 1982 to use
6
titration designs in an informative way.
7 One of the things about many of the
8
studies we do is a lot of people never get the best
9
dose they could, the highest dose they could take,
10
partly because if you don't interpret those studies
11
properly, you get misled. You
get an
12
umbrella-shaped dose-response curve because the
13
high dose only goes to people who can't respond.
14 What he did was look at the individual
15
dose responses and teased dose-response information
16
out of a titration design, a very efficient thing.
17
If you want to study eight doses, you don't need
18
eight groups. You need one group
on drug and one
19
group on placebo. It is worth
thinking about. You
20
will see it in the slides. It is
a picture from
21
his thing.
22 [Slide.]
at
324
1 All of these designs--I want to make one
2
last emphasis and then I will go away--all of these
3
things have been used. They are
all possible.
4
They make people nervous about the question of
5
generalizability but these should not be thought of
6
as the only trials that are ever going to be done
7
on the drug. They are trials
during Phase II to
8
convince you that you actually have a drug which,
9
it turns out, from PhRMA's data, is a big problem.
10 So we are going to be thinking about these
11
things more as time goes by.
12 Thanks.
13 DR. SHINE:
Thank you, Robert. So we have
14
learned at least four approaches to designing
15
clinical trials that may help the process.
16 We are going to go on with the
17
presentations and then, if you would be available
18
for questions after Jesse's presentation, that
19
would be helpful.
20 DR. TEMPLE:
Yes.
21 DR. SHINE: We
are now going to look at
22
issues with regard to devices.
Larry Kessler,
at
325
1
Director of the Office of Science and Engineering
2
Laboratories in the Center for Devices and
3
Radiologic Health
4 Take it away, Larry.
5 Dr. KESSLER:
Thank you.
6 Overview of Opportunity
7 Devices
8 DR. KESSLER:
Thank you. You may have
9
noticed Dr. Feigal, who is leaving the Center, has
10
let me do this. I thought it was
because he was
11
just leaving but I think it is because he decided,
12
once he saw the agenda and asked me to follow Bob
13
Langer and Bob Temple, he thought, let me follow
14
Bob Langer and Bob Temple and not him.
So I want
15
to thank David, in his absence, for that favor. It
16
is very kind.
17 [Slide.]
18 I am actually going to do something a
19
little different than Dr. Temple.
I think I will
20
be relatively brief.
21 [Slide.]
22 I am going to first mention--I always
at
326
1
mention to an audience I don't know that devices
2
are different than a lot of other medical products.
3
The are varied, and I will talk about that a little
4
bit. I will talk a little bit
about the vision in
5
the Center for Devices and Radiological Health and
6
our version of the critical path.
7 Then I am going to mention some
8
opportunities in the device world in very concrete
9
terms and I will be very brief about each of them,
10
and then mention some of the things that we are
11
trying to do to take this into the future.
12 [Slide.]
13 The device world is broad and vast;
14
blood-pressure cuffs, contact lenses, heart valves,
15
hip implants, infusion pumps, pacemakers. So the
16
critical path for a contact lens is not the same as
17
the critical path for a test strip or for a hip
18
implant. They are different
animals. They are
19
very different animals; how they are regulated, how
20
they are studied, how they are approved. So there
21
isn't a good "one size fits all."
22 I know it is not true, that just one size
at
327
1
fits all, in the other product areas, but, for
2
devices, we are much weirder than other product
3
areas. So thinking through how
we solve that
4
problem, the problem of critical path is going to
5
different.
6 I did not hear Dr. Jacobson's comments
7
this morning but she and I have talked, and the
8
pipeline issue for the device world, as Janet has
9
mentioned and I know Liz mentioned, is different.
10
It is a different problem.
Trying to accelerate
11
product is no less important in the device area.
12
But pipeline product is a little different problem
13
for us although I will point to at least one area
14
that we share with the other medical-product
15
centers, and I will talk about that at the end of
16
my talk, where it is not dissimilar for the device
17
area from drugs and biologics.
18 [Slide.]
19 We have been using this wheel to talk
20
about the designed obsolescence circle that the
21
device world lives in. So, in
our critical path,
22
trying to get things through the preclinical world,
at
328
1
the clinical world into manufacturing represents
2
just as big a hurdle for the device world as it
3
does for pharmaceuticals and biologics.
4 [Slide.]
5 I am going to skip that.
6 [Slide.]
7 I am going to talk about five examples.
8
The first one I am going to talk about actually is
9 a
success story. This is where the
agency, in
10
collaboration with the National Cancer Institute,
11
industry and academia actually helped solve a
12
critical-path problem about four or five years ago.
13 It started in the mid-1990s when digital
14
mammography was trying to come on line.
The way to
15
put digital mammography on the market was to try
16
and do adequate studies to demonstrate its
17
effectiveness and its safety.
Well, the initial
18
studies done by the imaging companies were far too
19
small to actually give us adequate evidence because
20
they did not take into account reader variability.
21
So the early studies brought to the agency were,
22
sadly, difficult and more or less failed designs.
at
329
1 We went back with the industry and sat
2
down and said, what can we do.
And we, and they,
3
and academia and the NCI developed a variety of
4
multiple-reader, multiple-case receiver
5
operating-characteristic curves.
This is
6
methodology. It allowed us to
bring these products
7
to market with enriched studies.
They are not
8
unlike some of the things that Bob Temple was just
9
talking about. These
multiple-reader,
10
multiple-case studies were done with enriched
11
populations and allowed us to get digital onto the
12
market faster.
13 Note that we followed up with the NCI with
14 a
large study now being conducted by the NCI's and
15
the American College of Radiology's Imaging Network
16
Consortium to look at the performance of this
17
product in the postmarket period because we still
18
have questions about this product now that it is on
19
the market.
20 For us, that is part of our critical path.
21
We don't end with putting something on the market
22
and saying we are all done.
There are serious and
at
330
1
significant postmarket issues and digital is one of
2
the areas where we have remained concerns and those
3
results won't be in for a few years while this
4
product is on the market imaging women for breast
5
cancer.
6 [Slide.]
7 I am now going to talk about a couple of
8
examples and be very concrete about the kind of
9
problems we face in the device world and the things
10
that can be done jointly where the government has a
11
role and we hope industry, academia, NIH and others
12
can play with us to help solve critical-path
13
problems.
14 One of them, we find, is an imaging of
15
peripheral vasculature. As we
all know,
16
drug-eluting stents have been proven successful for
17
the coronary world but the question we have is,
18
will they work in the periphery and will standard
19
imaging of lower extremities provide adequate data
20
for us to figure out how to go to market. The
21
answer is, we don't know, but we are acutely aware
22
that the way to do this kind of study is going to
at
331
1
require the size of patient populations usually not
2
available to most of the device industry in this
3
ballpark.
4 So, figuring out how we can do cooperative
5
and collaborative studies across sponsors where
6
peripheral imaging is done well and done accurately
7
is going to be a challenge.
Right now, the imaging
8
techniques for things like ultrasound, the
9
endpoints are not well known and it is not clear we
10
have confidence in the imaging results from this
11
part of the anatomy.
12 [Slide.]
13 Ablative therapies; atrial fibrillation is
14
one of the most common forms of fibrillation in
15
chest problems. It is commonly
diagnosed and there
16
are problems in demonstrating true benefit in some
17
of the studies associated with ablative therapies
18
for atrial fib. Like the last
study problem, this
19
will require a large patient series.
20 But, before we even get there, there is
21
not universal agreement between the agency and the
22
academic community on what are relevant endpoints
at
332
1
to study for the effectiveness of ablative
2 therapies and how do we
do the analysis.
3 So, again, here is a critical problem. It
4
is solvable but it requires a joint collaborative
5
approach between the industry and between the
6
agency as well as academic input.
7 [Slide.]
8 Intraocular lenses; an enormously
9
interesting area because we have had a lot of
10
success in treating cataracts in older patients.
11
But the current IOLs have limits because they have
12 a
single focal length and the human eye has
13
multiple focal lengths. The
industry has been
14
trying to solve this problem for a long time and
15
what it would like to do is, when it takes a
16
crystalline lens out of an aged person, an aged
17
crystalline lens, it would like to keep the
18
capsular membrane intact and put in a polymer, and
19
the polymer would form a shell and form a new lens.
20 The problems are we don't have really good
21
control of optical quality, material properties and
22
biocompatibility endpoints. So
this is an area
at
333
1
where we know what the industry wants to do, we
2
know the problem they are trying to solve, but
3
trying to get from the possible solutions of this
4
problem--we have been trying to do this for about a
5
decade to solve the IOL problem by flexible polymer
6
lenses in the capsular membrane has not been solved
7
and we haven't yet solved these problems.
8 So, trying to work here, this is really
9
very much lab-based work and correlating the
10
lab-based work with clinical-outcome studies is
11
where we are trying to go with this.
12 [Slide.]
13 The last example I am going to give is one
14
that all of the product areas share, and it is
15
genomic and proteomic devices.
The reason I am
16
raising this is last week I attended a jointly
17 sponsored PhRMA-FDA meeting on the science of
18
genomics and proteomics, where are we going.
19 At the end of one of the talks from
20
industry--it was a terrific talk, but the last
21
slide, in the last entry to the slide, given to a
22
wide number of FDA statisticians and regulators
at
334
1
was, "Our algorithms are learning algorithms,"
2
which is great in the discovery phase.
But now he
3
wants to bring that to us to approve as a product.
4
The product we would approve would be an
5
inconsistent algorithm that would update itself.
6 For discovery, that is terrific. For
7
science, that is wonderful. For
product
8
regulation, that is going to give us heartburn. It
9
is going to give us heartburn.
It is going to give
10
heartburn to the Center for Drugs and for
11
Biologics. How we are going to
allow algorithms
12
like this to be used outside of the investigational
13
arena is not yet clear.
14 There are some very exciting algorithms
15
out there. Some developments,
particular in
16
screening for cancer with proteomics, look
17
incredibly promising. But, if
they are going to be
18
algorithms that change over time, the regulatory
19
hurdles that we have to face to get through the
20
critical path are going to be tough.
21 Dr. Roses is smiling.
Do you know
22
something I don't know, because if you know how to
at
335
1
solve this problem, then I am going to go visit him
2
at SmithKlineGlaxo and we are going to help solve
3
that problem together.
4 [Slide.]
5 I am going to end with some of our future
6
directions. One of the things we
are trying to do
7
is identify some of the key problems in the device
8
world and we hope to work closely with industry in
9
going that. We have begun to
prioritize our
10
scientific efforts and some of those are directed
11
to some of the things I talked about with you
12
today.
13 Finally, we have been pursuing
14
collaborations with the agency within FDA, with the
15
NIH, particularly with the new National Institute
16
of Biomedical Imaging and Bioengineering and with
17
the National Cancer Institute and, more recently,
18
with industry. We are hoping to
have an active
19
partnership with them in solving some of our
20
critical-path problems.
21 I have run short.
I hope you don't mind,
22
Dr. Shine.
at
336
1 DR. SHINE: I am
delighted. That gives us
2
an opportunity to have a conversation.
I would ask
3
if any members of the committee have any questions
4
or observations for Larry.
5 One of my questions is to what extent, for
6
example, working out the digital-mammography
7
problem, it sounds like that was a problem and you
8
did an ad hoc solution in terms of dealing with
9
that problem in collaboration with multiple
10
companies getting a large enough population and so
11
forth. To what extent are you in
a position to
12
prospectively propose a variety of those kinds of
13
approaches to device manufacturers so that they can
14
think about a variety of ways that they would deal
15
with these kinds of issues as they are developing
16
new products?
17 DR. KESSLER:
It is going to vary across
18
the device spectrum. There are
certain areas, and
19 I
think the genomics and proteomics ones are
20
exactly the kind of area where we have identified
21
the regulatory problem and the scientific problem.
22
We don't have solutions yet but I think the
at
337
1
workshop that I talked about that was sponsored by
2
CDER is exactly along the road to say, we know
3
there are regulatory problems.
We can see them
4
down the road. What can we do
together, from the
5
methods standpoint, that can anticipate those
6
problems.
7 We are doing it in a couple of other
8
areas. We are doing it in the
pathology of the
9
vasculature. The Coronary
Division is doing a lot
10
of outreach. We have been coming
to the
11
transcatheter meetings every year to talk about
12
regulatory problems that we see down the road and
13
inviting both the academic world and the industry
14
who are well represented at TCT every year to
15
anticipate the science and regulatory problems and
16
working on them prospectively.
17 Other areas, I don't think we have been
18
nearly as foreshadowing as that.
I would just
19
challenge you a little bit. The
solution in the
20
digital-mammography area, I don't think it was
21
quite as ad hoc as it appeared.
It was actually
22
part of a systematic series of research that has
at
338
1
been done in the imaging area where we have been, I
2
think, on the cutting edge of imaging science in
3
the FDA. It has been an unusual
collection of
4
scientists that have been there.
We are starting
5
to use those methods that we have designed for the
6
diagnostic imaging world outside that.
7 Again, we are trying to stretch, because
8
what you can do with diagnostic imaging is not that
9
much different than in diagnostic in vitros, for
10
example. So we are starting to
translate some of
11 that
ROC work elsewhere.
12 DR. SHINE: By
ad hoc, I was referring to
13
the imaging activity. What I am
interested in, it
14
sounds like you are moving so that, in other areas,
15
you can articulate what these future algorithms
16
might look like, whether they are computer
17
algorithms or other kinds of--other observations,
18
comments about this interesting presentation?
19 DR. KESSLER: I
know he has got something
20
to say.
21 DR. ROSES: I
could talk about genetics
22
all day. Genomics, I limit
myself. I think, in
at
339
1
terms of discovery, that genomics has been used and
2
is highly overrated. I think, in
terms of its role
3
in the early development and the problem, seeing
4
whether problems that you might run into are part
5
of the genetic makeup of the animal versus the drug
6
being used, there are some very useful uses for
7
this tool.
8 On the other hand, algorithms for using
9
that with respect to changes that might occur as a
10
result of the drug, I fully agree with you. On the
11
other hand, I think one of the things that the--I
12
haven't said anything about this today, but I think
13
one of the things that needs to be clearly
14
differentiated in the critical-path documents and
15
in the discussions about it is the virtual total
16
difference between efficacy pharmacogenetics,
17
adverse-event or safety pharmacogenetics and
18
genomics, "omics," things like that. I think we
19
get very sloppy if we start thinking about those
20
things as the same thing.
21 DR. SHINE:
Good. Any other comments?
22 Larry, thank you very, very much.
at
340
1 DR. KESSLER:
Thank you.
2 DR. SHINE: Our
last presentation is from
3 Jesse
Goodman, Director of the Center for Biologics
4
Evaluation and Research. He is
going to look at
5
opportunities with regard to biologics.
6 Overview of Opportunities
7 Biologics
8
DR. GOODMAN: Good afternoon. I guess I
9
am the last person and have the most slides, so I
10
will try to correct for that by going quickly.
11 [Slide.]
12 But, basically, what I am going to do is
13
sort of strongly endorse the concept of this
14
research initiative, talk about efforts we have
15
made already within the center to be sure that our
16
science base is applied and some of the things I
17
think can be models for how we can successfully
18
work with others and identify product problems and
19
help solve them.
20 Then, as Larry and Bob did, I would
21
suggest a few what I have kind of called things
22
that have, both in talking to the industry and
at
341
1
innovators we work with and then, from our own
2
experience, and I think Janet may have mentioned
3
this, but we see an array of products across the
4
field and have the advantage of being able to
5
recognize things which multiple people are
6
potentially confronting and sort of have a list at
7
the end of some of what we see as rather large or
8
grand opportunities for science having an impact on
9
product development.
10 [Slide.]
11 Some of this I will go through kind of
12
fast, but this is the spectrum of current products.
13
Like Larry, I want to remind people that there are
14
many critical paths. There are
many different
15
types of products that, even within the world of
16
small molecules, there are many critical paths,
17
certainly, when you get to a diversity of biologics
18
ranging from cells to vaccines, et cetera. And
19
there are also many different acceptable clinical
20
outcomes, like, certainly, a 10 percent efficacy in
21 a
cancer drug might be acceptable. That
is not
22
going to be acceptable in a vaccine.
We don't even
at 342
1
have an acceptability of a 10 percent failure rate
2
in a vaccine.
3 So that diversity is something we have to
4
keep in mind and there is also a diversity of tools
5
that are needed to achieve that.
6 [Slide.]
7 Janet has pretty much set the groundwork
8
for this but I would like to go over a couple of
9
concepts. What we are really
trying to do here is
10
facilitate product development through better tools
11
and the latest technologies and also to focus our
12
intramural and extramural science as the resources
13
permit on things that I would characterize as good
14
investments, things that are problems out there
15
that we can help to solve with you.
16 Some examples and things that, within the
17
last year at the center and previously, we have
18
tried to focus resources on is identifying areas,
19
especially in new technologies where--and
20
especially in these diverse biologics or
21
public-health technologies where there may not be
22
large research infrastructures or the products may
at 343
1
be so diverse and unsettled that one person's
2
problem is not another person's problem as far as
3
they know.
4 This includes things like standards,
5
methods, assays and also just to emphasize the
6
importance of guidance, especially with new
7
technologies.
8 Examples in our shop are things you have
9
heard about and others have mentioned, gene
10
therapy, tissue engineering, which we work together
11
with CDRH on, stem cells, new vaccine technologies,
12
blood substitutes, pathogen inactivation. These
13
are areas that have generally had much less focused
14
resources on the science of actualization of
15
products rather than on the basic science of what
16
is the gene, what is the antigen, what is the
17
T-cell.
18 I think another important thing,
19
particularly in cutting-edge technologies is
20
assuring adequate internal expertise, be sure we
21
have the right partnerships with industry, et
22
cetera.
at
344
1 Just a couple of quick examples. I am
2
going to through a number more also quickly, but
3
identify roadblocks and develop appropriate
4
solutions. One example we had
about a year ago was
5
that essentially all these new smallpox vaccines
6
were being developed, there was a need for Vaccinia
7
immunoglobulin. There was only a
rather archaic
8
1950's six-well, hand-done, culture-plate
9
neutralization assay for Vaccinia virus.
10 So we set some of our people who had
11
developed similar assays for other antibodies loose
12
to develop a Vaccinia gene-reporter assay to titer
13
antibody levels. We just, then,
gave that out to
14
various companies that wanted to use it, and this
15
allowed VIG to be developed and the clinical trials
16
to be on a fast track.
17 That is not necessarily my endorsement,
18
then, of the vaccines because obviously they have
19
been challenging.
20 [Slide.]
21 Again, the emphasize it is not just about
22
science but it is about science-based guidance and
at 345
1
standards and outreach and working with others.
2
Again, I would like to emphasize a couple of
3
characteristics about this. I
think internal
4
expertise is very important. I
think product
5
developers, whether they are academic or
6
industrial, would rather talk to people who
7
understand what it is they are doing and they can
8
be helpful.
9 This could benefit multiple sponsors. It
10
can maintain staff cutting-edge expertise. I also
11
think there is a real connection to the quality of
12
the review process that FDA provides, that knowing
13
what you are talking about, having been there on
14
some level, scientific, clinical, looking at that,
15
having been somewhere broadly, is very important
16
and it reduces the risk that you are just going to
17
say, oh, my god; I don't understand this. This
18
scares me, or something that is not good for
19
sponsors either, not identifying problems because
20
you didn't figure it out from the last time you saw
21
it and you could have just saved this person, what,
22
$500 million. I don't know. All I want is a small
at 346
1
percent.
2 [Slide.]
3 Just to provide some examples. You know,
4
we were thinking about, well, what are some of the
5
things that we feel good about that the center has
6
done in the last year or so and then,
7
retrospectively, how does this concept of critical
8
path and scientific contribution fit into this and
9
what have we done.
10 Well, we are really excited about what we
11
were able to accomplish with the blood industry,
12
the diagnostics industry and CDC on West Nile virus
13
screening. Part of that was
recognizing the
14
problem quickly, bringing together the right
15
people. But part of that was the
science and the
16
critical-path science.
17 It was recognizing that we were going to
18
need standards. It was having
worked with the
19
broad diagnostics industry over the years to
20
develop a critical path in a way for
21
nucleic-acid-based testing and then being able to
22
show them that and how to use it very quickly and
at
347
1
provide samples, et cetera. This
worked out well.
2 But we don't have mechanisms or
3
infrastructure in place to do this on a routine
4
basis. We are, more and more,
cobbling this
5
together, especially dealing with emerging
6
infectious-disease threats, et cetera, but it is
7
something that we need to support prospectively
8
both infrastructurewise and then sciencewise. We
9
need to be ready with the science when all these
10
things come along.
11 There are some other things, but this has
12
been part of what we have been able to do, for
13
example, in terms of the response to SARS where,
14
within a couple of--I was told this morning--I was
15
at a meeting where we were discussing vaccine
16
issues with industry and others and I was told that
17
there are now at least five viable vaccine
18
candidates in development.
19 Well, this is in part because of how
20
industry and us could work together in this
21
paradigm and work with our colleagues like CDC and
22
WHO, provide strains suitable for vaccine
at
348
1
manufacturing quickly, get together about assays
2
and pathways. There is a long way
to go on that
3
particular issue but, again, it is an example.
4
Given the time, I will just skip through these, I
5
think.
6 [Slide.]
7 But some examples; new technologies, as I
8
mentioned, are particularly good areas for
9
investment here. I would say new
technologies and
10
unmet public-health needs. But
we have worked with
11
NIH on some of these like trying to understand the
12
gene-therapy adverse events in terms of
13
leukemogenesis and then trying to have an
14
intelligent science-based regulatory response to
15
those events so that we get the products safer and
16
improve them but don't kill a promising field to
17
get that right.
18 It requires knowing what are the questions
19
you need to ask to move the field forward. We have
20
increasingly used our advisory committees almost
21
prophylactically to help define a pathway forward
22
in product development. So we
have islet cells as
at
349
1
incredibly promising treatment for severe
2
refractory diabetes.
Promising. Disparate results
3
all over the place. Issues with
cellular
4
characterization. So we have
tried to bring
5
together people to define how to move the field
6
forward. Same thing with cardiac
cellular therapy.
7 One person over there, I don't remember
8
who, mentioned this issue of harmonization. I
9
think the global harmonization is an important
10
efficiency, as Janet replied, in product
11
development. It is an important
cost saver and we
12
all support. It intersects with
the science, for
13
example, in this later area; for example, providing
14
standards. We can't have global
harmonization
15
unless we know that when somebody in Europe does an
16
adenovirus gene-therapy experiment, they are
17
talking about the same amount of adenovirus that we
18
are, when somebody in Europe makes a Factor VIII
19
concentrate, we know that it is the Factor VIII
20
standard we have all agreed on, and it is the same.
21 Those are the areas where we have really
22
focused some of our scientists' efforts.
at
350
1 [Slide.]
2 So existing research programs within the
3
center have been focused increasingly towards this.
4 I
think, as Janet said, we are all going to work to
5
be sure that we are doing as much as we can
6
efficiently with whatever resources we do have.
7
Many of these are directly connected to the--we
8
have started a good database of this.
We have
9
explored their connections to the applications and
10
products they are supporting.
11 One thing I was going to say is that we
12
are planning to use our advisory committees, for
13
example, to obtain broad scientific input about
14
opportunities in these fields.
Are we looking at
15
the right things? So it is an
extension of the
16
kind of input being got here for this.
17 I think, based on time, I am just going to
18
skip through these.
19 [Slide.]
20 I will just give a couple of examples of
21
things we already look at that really are critical
22
path. When I called up some of
our industry about
at 351
1
their input into this initiative, I was very happy
2
to hear them say things like, well, that is what
3
you are doing. You know, we are
having problems
4
manufacturing this vaccine and you are working with
5
us to figure out why this particular component of
6
this vaccine isn't working and you provide helpful
7
input.
8 So here are some examples; research and
9
toxicity. Why do have
adenovirus. Lung toxicity;
10
why do we have toxicity from artificial blood
11
products. Assay validation for
retroviruses, et
12
cetera.
13 [Slide.]
14 This is the model of lung toxicity due to
15
adenovirus. This helped focus on
what, in future
16
adenovirus vectors in the animal studies should you
17
be looking for to predict toxicity.
18 [Slide.]
19 In the blood products, donor-derived blood
20
products carry considerable risk, have short
21
half-lives. Donors are often in
limited supply.
22
People have been talking about, and working on,
at
352
1
blood substitutes for many, many years.
There is
2
tremendous promise there but there have been a lot
3
of problems.
4 [Slide.]
5 This is just some examples of the physical
6
chemistry of the blood products.
Some of these
7
have been associated with substantial toxicities,
8
particularly vasospasm, tissue damage.
Again, we
9
have had a small group that has worked with the
10
developers of these products focusing on what is
11
the molecular basis of toxicity, identifying the
12
oxidative chemistry of the toxicity and developing
13
an in vitro model to screen products.
14 [Slide.]
15 This is an example where there is not a
16
lot of other investment through other sources in
17
this, whether industry or government.
It has been
18 a
small investigation but, again, we have heard
19
from the manufacturers. This has
taught us where
20
to go with our next-generation products.
21 [Slide.]
22 Adventitious agents; Gail asked me to
at
353
1
comment on that and she saw it in the talk. But
2
this is increasingly a problem in biologics. We
3
have worked with people, developed a number of
4
assays. I will talk more about
that.
5 [Slide.]
6 Product characterization is another area.
7
Again, I mentioned adenovirus standards, product
8
identity.
9 [Slide.]
10 I will give an example because of your
11
comment about--I will linger on this for a second
12
because of your comment about genomics, et cetera.
13
It may not have immediate yield in terms of target,
14
but some of the same technologies may help us, for
15
example, characterize complex products.
Cellular
16
therapies are an example. This
is just a very
17
quick and superficial mention, because of time, of
18
work that Raj Puri, who is in the back, I think Dr.
19
Mahendra Rao from the Institute
of Aging and a
20
number of other people may have contributed to.
21 [Slide.]
22 But, looking at stem cells and saying,
at
354
1
what is it that makes a stem cell at a
2
sophisticated genetic level and reasons for doing
3
this; safety concerns about unregulated growth,
4
being able to characterize them so that they are
5
reproducible for use in medical therapy.
6
When new stem-cell lines are
developed,
7
how do we make sure that they really share the
8
genetic characteristics of stem cells, of having a
9
molecular vocabulary of what is a stem cell.
10 [Slide.]
11 This is just an example from their
12
microarray studies which identified a large number
13
of new genes associated with what we say, a
14
"stemness." Now this
is pretty early kind of
15
understanding of these things but this is an
16
example and I know one or two other laboratories
17
have looked and some of the genes they have found
18
are overlapping, too. But this
helps define what
19
these populations are.
20 [Slide.]
21 Assays for product purity; we have done
22
potency, other methods development.
at
355
1 [Slide.]
2 I think efficacy is very important. Bob
3
Califf mentioned, and he and Janet had a little
4
thing about, surrogate markers.
But certainly, in
5
many of our biologic products, particularly
6
vaccines, or clotting factors, there can be
7
excellent markers of efficacy but they are still
8
missing in lots and lots of very important
9
products. So, knowing what a
protective antibody
10
resource is--for instance, it is not my area so
11
Center for Drugs can get made at me, but it is
12
certainly my area clinically, in terms of
13
antibiotics. Knowing what kind
of pharmacodynamic
14
models and compartment models might predict
15
effectiveness of an antibiotic could be very
16
important. Again, I am not going
to go through all
17
these studies because of time.
18 Identifying what are the protective
19
antibodies, not just what are the levels but
20
actually what antibodies are protective; what kind
21
of T-cells are protective against different
22
diseases. If we know that, then
you are not
at
356
1
talking about 10,000-person efficacy studies for
2
vaccines.
3 [Slide.]
4 And then statistical and epidemiological
5
analysis. Again, both Bobs
touched on that.
6 [Slide.]
7 So, moving from that background where we
8
do a significant amount of this, and it is driven
9
by what people observe from the products and as
10
regulatory issues, what are some of the big
11
opportunities out there that ourselves or others
12
have looked at.
13 [Slide.]
14 Here are some examples. New vaccine
15
delivery systems. Rapid-use
vectors so that, in
16
response to an emerging infectious disease or a
17
bioterrorist event, one could create a new vaccine
18
quickly and have some confidence about its efficacy
19
or the predictability of its efficacy or safety.
20 Making available well-characterized cell
21
banks for production of vaccines and other
22
biologics, characterized in terms of their safety
at
357
1
such as they don't cause cancer and their freedom
2
from adventitious agents. Again,
these are
3
resources which are scattered in very limited
4
amounts throughout the field.
There isn't a
5
library of resources or tools that one could say, I
6
have a new virus, SARS. I want
to produce a
7
vaccine and cell line. I go to a
partner like ATTC
8
or something and just take those off the shelf and
9 I
know they are good with the FDA. They
are safe,
10
and we don't have do two years of studies to figure
11
that out, or maybe ten years.
12 Characterize of cell therapies and links
13
to standardized outcomes. I have
more about these.
14
How is my time?
15 DR. SHINE: You
have about four or five
16
minutes.
17
DR. GOODMAN: That's tons of time.
18
Methods and validation for pathogen inactivation.
19
Multi-pathogen and rapid detection methodologies,
20
longevity and storage of blood and tissues.
21 [Slide.]
22 So I will just get down into the most
at
358
1
superficial details in a couple of these areas.
2
New delivery methods for vaccines and gene
3
therapies. As I said, we have
got all these things
4
out there but there are all kinds of safety issues
5
that are holding up the field.
How much residual
6
DNA is okay in a vaccine? Is DNA
in a vaccine even
7
unsafe on any level? We don't
know that. It is an
8
important area for study.
9 I mentioned new vaccine platforms.
10
Transgenic platforms. Adjuvants
and immune
11
stimulants. Again, specific
companies may have
12
these things and, the minute they encounter a
13
problem, unless it is absolutely essential for
14
their product development, they are going aside to
15
it. But there are probably
things out there that
16
would be helpful to everyone if they were known to
17
be safe and effective.
18
[Slide.]
19 This is one I think is very important,
20
characterization of cellular products and links to
21
outcomes. You know, we have got
multiple labs
22
making just amazing stuff, like the stuff Dr.
at 359
1
Langer said or like people are doing with stem-cell
2
experiments at my former university and elsewhere,
3
where they are taking things, they are hitting them
4
with growth factors, they are expanding them in
5
culture, but what are they?
6 When we have seven different labs doing
7
cardiac-cell therapy from the same source and they
8
exhibit different cell markers or--what is working?
9
What doesn't work? We need to
work with the
10
sponsors to collect the data from this in some kind
11
of standardized way with reasonable assays and
12
connect it to some kind of standardized clinical
13
outcomes so that we can predict and know that, when
14
you make a cellular therapy, this is a product that
15
is, indeed, what it is.
16 [Slide.]
17 I mentioned the cell substrates. I think
18 I
have said enough about that, but I think there is
19 a
real need for a more diverse bank of cell
20
substrates that have been well screened and
21
characterized. This is the kind
of thing where--I
22
feel like if a few really strong groups spent not
at 360
1
huge amounts of money over five years, we could
2
probably get this. But there is
not enough there
3
for any one company, probably, to want to do these
4
or make that investment.
5 Again, it might be a consortium kind of
6
issue, or it might be something government can help
7
with or at least partner in or coordinate an effort
8
and everybody knows it is a level playing field.
9 [Slide.]
10 Very important things in emerging
11
infectious diseases. We spend so
much of our time
12
trying to keep blood, and then our biologic
13
products substrates, tissues, which are going to be
14
increasingly--we are moving forward with our Tissue
15
Rules which has been a long gestation process,
16
probably longer than the average elephant. But we
17
are moving forward with those things.
We need
18
better technologies for dealing with this.
19 Like I said, with West Nile, it was great
20
but if, each year or two, we are having to add new
21
pathogens or take old ones out, we need even more
22
robust and better technologies.
There are things
at 361
1
out there in terms of multi-pathogen testing, new
2
platforms, nanotechnology, or, for major remaining
3
problems with the blood supply, like bacterial
4
contamination, TSE.
5 We need blood and plasma products that,
6
ideally, as part of their process of manufacture,
7
are robustly inactivated for anything that might be
8
there that you don't even know it is there. So we
9
would like to say, when TSE came along, we know,
10 from
the steps we are putting this through, that
11
prion agents are inactivated.
12 In fact, we have done some of those
13
studies. That is another perfect
example of
14
studies that require specialized laboratories,
15
special equipment, benefit people across industry
16
and are usually very difficult, when we ask an
17
individual company to do them.
Or how about to
18
develop better methods for inactivating prions.
19 Tissues; that is an industry that is just
20
in its infancy. But I wonder if
people here
21
realize--some of those numbers were on Dr. Langer's
22
screen, but our current belief is that up to a
at 362
1
million medical tissue transplants are done in a
2
year. These are tendons,
cartilage, et cetera,
3
skin and we don't know how to safely decontaminate
4
tissue. It is basically not
sterile. This is very
5
important stuff.
6 [Slide.]
7 I put this together under one rubric;
8
better, longer-lasting blood, cellular and tissue
9
products. How do we improve the
shelf life of
10
blood. Right now platelets is a
huge issue. We
11
have shortages of platelets. We
have issues where
12
we can't test them for bacterial contamination
13
because, by the time the tests come back, they are
14
no good to use anymore. That is
a little
15
overstatement, but it is there.
16 So, as these cells are increasingly
used,
17
as we have some of the threats, in terms of
18
terrorism, war, et cetera, it becomes more
19
important to have a less fragile blood system. We
20
have a fragile blood system that is held together
21
by one or two people, I think, sometimes. But they
22
are great people.
at
363
1 Then data quality and analysis. This
2
isn't particularly my area of expertise but just to
3
add onto what both Bobs said. I
think not only
4
could we do a better job about how we do studies,
5
how we collect data, but probably also how we
6
analyze it.
7 We have really tried to push, in the
8
vaccine area, for more use of large, simple trials.
9
We have to reexamine FDA's own practices here and
10
what messages we give about clinical trials and how
11
to do it more efficiently.
12 I will just leave it at that. Just to say
13
we thing FDA can be helpful in working with people
14
to identify opportunities and better tools and we
15
think the engagement, no matter--we should be doing
16
this all the time. Again, as
many of you have
17
said, we are very resource-constrained, but we view
18
this as part of what will improve the outcome we
19
are doing anyhow, which is to approve safety and
20
effective products.
21 Hopefully, with that, we can enhance
22
development of good products that people need. I
at
364
1
do think this is particularly important where
2
products present complex new technologies or maybe
3 don't
always have huge economic incentives driving
4
research and development in those fields.
5 Thanks.
6 DR. SHINE:
Jesse, thank you.
7 Questions and Discussion with the Board
8 Why don't you stay there because I have
9
got some questions for you and for Janet. I think
10
this is an opportunity to pull some things together
11
with regard to some of the material that we have
12
been talking about.
13 Let me start and I hope my colleagues will
14
jump in here. Both of you have
demonstrated a
15
complex set of issues in terms of the science and
16
we are also faced with a complex set of players in
17
terms of dealing with it. So we
are talking about
18
new science, new technology, standards, agreements
19
about procedures, and so forth.
We are talking
20
about what the role should be of FDA in its own
21
laboratories, what FDA does in collaboration with
22
the NIH or other agencies, what FDA does in
at
365
1
relation to industry.
2 We have talked about a potential SemiTech
3
in terms of a public-private activity that might
4 involve, in this case, the pharmaceutical industry
5
doing some of these kinds of activities.
6 The question is, from a process point of
7
view, how would you propose priority setting in
8
this area, both in terms of the nature of the
9
problem and the question of who should do it,
10
because it seems to me that the issue here is not
11
that many people are sitting around here who
12
disagree about the need for this kind of knowledge
13
generation, but how do you go from the day-to-day
14
problems? You described the West
Nile story.
15
Again, using the term ad hoc is probably not
16
totally accurate because of the past experience,
17
you solved a lot of problems in relation to an
18
unanticipated set of issues.
19 Yet, if we want to improve the path, that
20
means looking ahead and trying to do some strategic
21
things. So I wonder whether you
have thoughts
22
about this, how you think about, if you had some
at
366
1
additional resources, what you would do as opposed
2
to what other people would be doing.
3 DR. GOODMAN:
Right. And I would really
4
welcome Janet's comments, too. I
would first start
5
by saying I think part of what we want to do is
6
open this dialogue and get input into where we
7
should be playing a role and what that role should
8
be.
9 I think all the partners are important. I
10
guess I would say that my--and, again, I am
11
somebody who came to FDA from a different
12
environment, from academia and from having done
13
some industry-sponsored studies, so I think I have
14 a
view not just from the FDA. I have a
fairly
15
broad spectrum of the product-development endeavor,
16
although I haven't been directly in industry.
17 But what I was going to say is I
18
think--and the document points this out. I think
19
that we do have a unique vantage point.
We see
20
multiple products. For example,
my gene and
21
cell-therapy reviewers see the twenty-something
22
cardiac-cell INDs and we sort of start to recognize
at 367
1
things that may not be in the literature, may not
2
be published. So sometimes, we
can play a unique
3
role in identifying questions.
4 Sometimes, we can sort of help. It is
5
almost like an arranged marriage, the solving of
6
some of those questions. On the
other hand, I
7
think that there is a spectrum--this is so huge.
8
Just look at the few things highlighted in the few
9
talks here today both from inside and outside FDA.
10 Our ambitions to improve product
11
development are so huge that there is a large
12
spectrum of activity that should be brought to bear
13
on that. So, for example--I
mean, the West Nile is
14
an interesting example. So there
NIH, at the one
15
spectrum, we had them work with us and they did
16
some developmental funding of one of the companies,
17
for example.
18 We provided standard and actual samples
19
and sort of library and resources and technologies
20
that would be able to be validated for the in vitro
21
diagnostics. CDC partnered with
us in providing
22
samples in some of their methodology that they had
at 368
1
used in state labs, in their own labs.
2 So I think--I am not trying to dodge the
3
thing. I think the answer to
that would be I think
4
there is a lot more that could be done across the
5
spectrum. I think FDA could do a
lot more and
6
could benefit from doing a lot more.
I don't think
7
FDA should do that in a vacuum, in some cases--for
8
example, in some of these ideas I discussed with
9
vaccine companies, they would say, we would want
10
you to do that, because actually they would want us
11
to do it because they would want this assay that,
12
then, they knew we would accept and make it
13
available to everyone. And it
would be free of
14
politics and free of favoritism, hopefully, if we
15
did our job right.
16 But there might be other things, like in
17
cellular therapies, that should be--FDA can't--even
18
that little piece I laid out, we should be working
19
with all those investigators, the academic
20
organizations, the NIH, in the consortium model and
21
each person contributing in their area.
22 DR. SHINE: I
would suggest to you that
at 369
1
the original concept of a roadmap was created by
2
the then President of Motorola.
It was a roadmap
3
for accomplishing things in the semi-conductor
4
industry.
5 One of the issues is whether the agency,
6
in collaboration with industry and the academic
7
community, should consider a series of roadmaps in
8
product development in a number of areas and then,
9
having agreed on what that roadmap might look like
10
in terms of the elements in getting from here to
11
there, then agree on who takes responsibility for
12
the lead in a particular portion of that roadmap.
13 All I am saying is that I would like to
14
see us think about ways of implementing this that
15
are--because of the very complexity you are talking
16
about, that we can move to the next step which is
17
how do you articulate that in a clear enough way so
18
that people can understand what they are trying to
19
buy into.
20 Let me throw things open to the rest of my
21
panel here and see if there are other comments, and
22
then I want to give Janet an opportunity to give a
at 370
1
benediction.
2 Susan?
3 DR. HARLANDER:
I guess I would just
4
encourage you, and the problem is daunting, I
5
realize, but I think it is so critical to involve
6
all the stakeholders. I don't
know what mechanisms
7
you do for that, because it is not just trade
8
associations. They represent a
segment of the
9
industry and they don't represent academia at all,
10
probably. So I don't know how to
involve that but
11
to recognize that there are a lot of stakeholders
12
and to get their input early.
13 And if there is an opportunity to identify
14 a
cross-cutting fundamental area where you could do
15 a
trial of getting academia and the private sector
16
and the public sector together to work on these
17
things--can't take all of it on but there may be
18
some fundamental thing where you can assure
19
industry that they are not going to be taken down
20
because of--I wrote it down and now I can't
21
remember--that it is okay for them to collaborate
22
on these things from a government perspective.
at
371
1 DR. SHINE: I
think doing some special
2
things in terms of dealing with the anti-trust--
3 DR. HARLANDER:
Anti-trust. I just
4
couldn't think of that; sorry.
5 DR. SHINE:
--are relevant. There are
6
ways to do that.
7 DR. HARLANDER: Yes.
8 DR. GOODMAN:
That is way that FDA, I
9
think, can be helpful by sort of helping
10
orchestrate. Now, a caveat, and
you all raised it,
11
it is the effort and resources that we can put into
12 this, and we frequently are doing this anyhow, for
13
example, with our colleagues and we would like to
14
do a lot more. But we are doing
a lot of this in
15
terms of helping, for example, NIAID in its
16
development of counterterrorism products, but we
17
could do even more.
18 DR. CASSELL:
Ken, I like your idea of a
19
roadmap. I actually have
suggested that taking
20
maybe even just two therapeutic areas initially and
21
dissecting them in terms of trying to set
22
priorities, lay out the roadmap, and then you could
at
372
1
see what kind of overlap there might be between
2
even those two therapeutic areas, might be a
3
rational way to go.
4 Of course, I am biased, but I argued that
5
infectious diseases would be a natural because we
6
are faced with such urgency in terms of need. I
7
think you don't run into a lot of the anti-trust
8
issues that you might in such highly competitive
9
other arenas, say, in neurosciences or
10
cardiovascular because so many companies are not in
11
the area of anti-infectives today.
12 So it might be a natural. There is
13
already this new investment of $1.7 billion at
14
NIAID for development of anti-infectives, so you
15
have got a lot going on in the public sector
16
already focused on drug development, how can we do
17
it better.
18 DR. SHINE: I
would agree with Susan. We
19
are not talking just about government and industry,
20
for example, or government and industry and
21
academia. Consumers have a
major stake in this
22
and they can keep a lot of this process honest in
at
373
1
terms of the various players.
2 DR. CASSELL: I
would just add that the
3
Institute of Medicine, the Board on Health Science
4
Policy has just approved a series of fora on drug
5
development and drug discovery and issues. I think
6
that type of forum is also going to be very useful
7
for discussing some of these things and developing
8 a
roadmap, maybe.
9
DR. SHINE: Good.
Other comments?
10 Janet, I think the sense of the committee
11
is that this is a very important effort, that you
12
and your colleagues have articulated a
13
fundamentally important issue.
It has a lot of
14
analogies to healthcare in the sense that we spent
15
decades delivering all kinds of goods but we never
16
wanted to study the process by which care is
17
provided. So, in this case, it
is to study the
18
process by which products are developed, moved
19
forward, industrialized, as you put it, and so
20
forth.
21 So I think it is a very important area. I
22
think that the overall sense of the committee is
at 374
1
that this is an area that they would be very
2
strongly supportive of. I think
the trick is going
3
to be, obviously, in the articulation of this into
4 a
form--you consult like crazy, but that is not
5
going to ultimately get you a workable structure.
6 All I am raising is, in one way or
7
another, and I think Gail is right; you pick one or
8
two or three areas and then you really look closely
9
at what are the steps that would be helpful over an
10
extended period of time. For
example, we have
11
heard about four aspects of effective clinical
12
trials. Those could form part of
the way in which
13
the clinical trials are done in that area.
14
We have heard about some
innovative
15
approaches to thinking about devices.
That could
16
be part of it, depending on what area you pick.
17
But I think the devil is going to be in structure
18
here so that people know where you want to go and
19
what the products are and who is responsible.
20
Again, I applaud you for at least considering some
21
of the models in other industries because I think
22
we have a lot to learn from those models in terms
at
375
1
of public-private activities.
2 I also think that one can make a strong
3
argument that investments in this area, and I think
4
we ought to make that argument--that investments in
5
this area, through the Congress, through the
6
pharmaceutical industry, through other interested
7
players, could go a long way to improving
8
productivity.
9 I still want to see the overall--I mean,
10
the last thirty years of drugs rather than the last
11
ten years just because I want to be sure that we
12
are not looking at just one part of a cycling
13
phenomenon. But I do think that
one can make a
14
strong argument that solving some of these problems
15
would have enormous benefit to the industry and to
16
the public in a way that people may not have
17
thought about.
18 So just as it has been a tough sell to
19
sell health-services research, this is, if you
20
will, science services in the drug and device
21
industry.
22 Would you like to make any final comments?
at
376
1 DR. WOODCOCK: I want to thank the Board
2
for your attention. This is a
starting-out area
3
and so it is a bit chaotic, I agree, because there
4
are so many factors. It is a
very complex area.
5
It goes everywhere from characterization of viral
6 contamination of cell
lines all the way to advanced
7
clinical-trial methodology.
8 I would reiterate, I think each one of
9
these is interconnected. I think
Gail actually
10
said it the best; each barrier along the path,
11
wherever it might be in the development process, is
12
brought to bear in decisions that are made on what
13
is developed, or what is feasible, or how fields
14
are developed. So this ends up
being really
15
impacting what healthcare products are delivered to
16
the public at the end of the day.
17 We will need your advice over a prolonged
18
period as we begin to shape this initiative. I
19
think you have given us some very important
20
guidance on making this real in the sense that we
21
need to lay out some paths. We
need to reduce this
22
to, okay, what if we wanted to get this healthcare
at
377
1
product or this class of healthcare products more
2
efficiently in the hands of the doctors and
3
patients. Where are the
barriers, an example or
4
two, is what you are talking about and then how
5
would people play on that and where would we see
6 different sectors
contributing and that could help
7
make this much more real, is what I understand you
8
are saying, to everyone because it is so large and
9
complex and hard to bring together.
10 We will do that and, with your indulgence,
11
we will probably bring that back to you probably at
12
your next meeting.
13 DR. SHINE: I
guess if you don't have
14
infectious disease on the list somewhere, Gail
15
will--
16 DR. WOODCOCK:
We're toast.
17 DR. SHINE: Any
other final words from the
18
members of the committee. Then I
think Dr.
19
Crawford wants to make a comment.
20 DR. CASSELL: I
was just going to suggest
21
that, since we really don't meet again, as I
22
understand it, until December, or November. But
at
378
1
still it is a long way away. I
think you have
2
gained a lot of momentum since this was announced
3
on March 16. I would hate to see
you lose that so
4 I
hope that you will feel free to get this group
5
involved. I don't know how we
can do that or what
6
you are at liberty to do, but I think it would be a
7
good idea not to wait until November.
8 DR. SHINE: I
would encourage you to think
9
about using e-mail on various aspects of this. A
10
number of us are very comfortable in working on
11
projects with other agencies and so forth over time
12
using e-mail. So, please don't
hesitate to do that
13
in a way if you think we can be helpful to you in
14
any of these areas.
15 DR. WOODCOCK:
Thank you. We do have an
16
extensive series of activities planned over the
17
spring and summer and we can definitely keep you
18
updated and even, perhaps, plan for substantive
19
input if we get to some critical points.
20 DR. SHINE: Dr.
Crawford?
21 DR. CRAWFORD:
Let me thank Ken and all
22
the committee for all your wise comments and
at
379
1
indulgence and attention today.
Also, I would be
2
remiss if I didn't thank Janet personally for what
3
she has done to pull all this together.
Ken
4
mentioned earlier that she is a widely known
5
innovator and that is certainly the case. This is
6
going to take a lot of hard work and, as Susan
7
said, a lot of interaction with the stakeholders,
8
but also with individual members of this committee.
9 You know, I can promise you that we will
10
stay in touch over the spring, fall and winter, as
11
we go forward.
12 The other thing I would want to say is
13
that the idea of the critical path is getting, in
14
fact, a lot of traction on Capital Hill. We have
15
begun making some courtesy visits with important
16
key members of the U.S. Congress.
They are all
17
key, some are more key than others, perhaps. But
18
also these have been included in our speeches and
19
some other sorts of public interactions that we are
20
making.
21 I think it, by now, is an easily
22
understood concept on the surface.
As we move
at
380
1
forward, I think we are going to be able to get
2
resources here. I also think it
is going to give
3
us a lot better interaction along functional lines
4
with NIH and other sister organizations and brother
5
organizations that we deal with.
6 This is a big thing for FDA. As I
7
mentioned in my opening remarks, something like
8
this has been mentioned usually along the lines of
9
science-based future for FDA and so forth. But
10
this puts legs and gives a structure to it and we
11
are all very enthused about it.
You can count on
12
us to wage a good fight with your help.
13
Thanks very much.
14 DR. SHINE:
Thank you. If there is no
15
other business, we are adjourned.
16 [Whereupon, at 4:40 p.m., the meeting was
17
adjourned.]
18 - - -