DEPARTMENT OF HEALTH AND HUMAN SERVICES

FOOD AND DRUG ADMINISTRATION

NATIONAL INSTITUTE FOR CHILD HEALTH
AND HUMAN DEVELOPMENT










FDA/NIH CONFERENCE

CLINICAL PHARMACOLOGY DURING PREGNANCY

ADDRESSING CLINICAL NEEDS THROUGH SCIENCE





Monday, December 4, 2000

9 o'clock a.m.




















Holiday Inn Capitol
Washington D.C.


C O N T E N T S

AM Introductory Remarks:
	Susan Wood, Ph.D.	3

Welcome and Opening Remarks:
	Janet Woodcock, M.D.	4

Keynote Addresses
	NIH: Jane Henney, M.D. (for Ruth Kirschstein)	6
	FDA: Jane Henney, M.D.	11

Current Requirements for Labeling Products for Use in
Pregnancy and FDA Initiatives:
	Sandra Kweder, M.D.	17

Consumer Perspective:
	Diana Zuckerman, Ph.D.	43

Physiological Changes In Pregnancy: Effect on Drug Disposition:
	Marilynn Frederiksen, M.D.	57

PM Introductory Remarks:
	George Giacoia, M.D.

Current Therapeutics: Evidence and Needs:
	Asthma: Michael Schatz, M.D.
	Hypertension: Baha Sibai, M.D.
	Seizure Disorders: Mark Yerby, M.D.
	Psychiatric Disorders: Katherine Wisner, M.D.

Interface of Clinical Pharmacology and Drug Safety:
	Gideon Koren, M.D.

P R O C E E D I N G S
AM Moderator Introductory Remarks
	DR. WOOD:  Good morning.  I am Susan Wood.  I am the new Director of the 
Office of Women's Health at the Food and Drug Administration.  I am delighted to 
be the moderator of this panel which going to open what I think is going to be a 
fascinating conference over the next two days.  So, I want to welcome you and I 
look forward to hearing not only our presentations but some of the discussions 
that will ensue and, hopefully, be able to look at the end of this and think 
that perhaps we can identify some ideas and strategies for moving forward.
	Some housekeeping notes this morning, we may have some rearrangement of 
the calendar.  We have had the unfortunate news that Dr. Kirschstein is ill but 
that speech will be given.  She has a wonderful replacement.  Dr. Henney is 
going to give two talks today -- eminently qualified.  We also may shift the 
schedule so that Dr. Sandra Kweder, who is scheduled for the end of the day -- 
we may move her up, and I am sure it won't break anyone's heart if we have to 
end a little bit early today but it will provide more time for more discussion 
at the end of the day.
	So, to open the conference, we are delighted to have Dr. Janet Woodcock, 
who is the Director of the Center for Drug Evaluation and Research, commonly 
known as CDER, at the Food and Drug Administration.  She has been Director of 
CDER for the last six years and has served at FDA for over fourteen years in a 
number of senior positions.  She received her M.D. from Northwestern Medical 
School, and has had faculty appointments at Penn State and at UCSF.  She is 
going to talk about what we are going to be doing today, and we look forward to 
hearing her remarks.
Welcome and Opening Remarks
	DR. WOODCOCK:  Good morning.  On behalf of the FDA and the NIH, I would 
like to welcome you to this conference.  All of us are here because we have a 
shared interest in the proper use of medicines during pregnancy and lactation.  
At these times of life, just like any time, it is best to avoid medicines, of 
course, unless they are truly necessary but chronic and acute illnesses will 
strike us all and require therapy.  When this happens women deserve better than 
an educated guess at the proper dose and regimen or at the optimal drug choice.
	The challenge for the biomedical community is to find a way through the 
thorny social, legal, ethical and financial issues to ensure that appropriate 
scientific studies are performed.  This will provide pregnant or lactating women 
the benefits of the same scientific evidence-based treatment that we all hope 
for in all of medicine.
	In this conference we seek to identify the key opportunities where 
scientific study will bring tangible results.  We hope this can stimulate the 
kind of research and the generation of scientific information that pregnant and 
lactating women deserve and need and, as a clinician who has had the challenge 
of treating women with diseases, chronic diseases, who were lactating or were 
pregnant and who were desperately in need of therapy and not finding the answers 
I need, I certainly hope we can find a way forward.  Thank you very much.
	[Applause]
	DR. WOOD:  Thank you.  We are now going to have a double presentation.  I 
think it is wonderful that Dr. Henney is going to be able to do that because she 
is going to give Dr. Kirschstein's presentation as well as her own and, given 
that she has spent so many years at the NIH -- nine years at the National Cancer 
Institute and five years as the Deputy Director of the NCI, I think she has 
insight to the NIH as well as to the FDA.
	Dr. Henney, as many of you know -- well, I hope all of you know, is the 
Commissioner of the FDA and she has been here for the last two years.  But that 
was not the first time she came to the FDA either.  She spent two years, several 
years back, as the Deputy Commissioner of FDA after she had left NCI.  In the 
meanwhile, she spent seven you years at the University of Kansas in a number of 
senior positions and, after serving as Deputy Commissioner of FDA, went back to 
New Mexico and spent four years there as the Vice President of the University of 
New Mexico Health Sciences Center.  She is a graduate of the Indiana University 
School of Medicine, and she was a fellow in medical oncology at M.D. Anderson 
Hospital and Tumor Institute and also did a great deal of work at the Cancer 
Therapy Evaluation Program at the NCI.  She has received numerous honors, 
including the recent Honorary Fellowship of the American Colleges of Healthcare 
Executives.  Back in 1979 and '81, she also received Public Health Service 
Commendation Medals and the Commissioner's Special Citation in 1994.  Recently, 
she also received the Jacobs Institute's Excellence in Women's Health Award, and 
I think that is particularly appropriate given today's topic and its focus on 
the health of pregnant women.
	We welcome you, Dr. Henney.
Keynote Address
	DR. HENNEY:  The other affiliation that I have with the NIH is that I live 
there.  When I was coming back to the FDA I called Harold Varmus, who was then 
Director of the NIH, and asked him if he could see his way clear to let an old 
NIH-er, a fellow PHS-er and a person who had one of their centers located on his 
campus be one of the residents of NIH.  So, he said, "sure, as long as you pay 
your rent."  So, every year I watch the NIH budget very carefully.  I have no 
interest at all in how many grants they are funding; it is what is the 
maintenance budget.
	[Laughter]
	Yesterday Bob and I were out, pulling up all of the cannas and planting 
lilies, and working like little dogs out in our yard when up came Alan Rabsin 
who, many of you may know, is the Deputy Director of the NCI but his most 
distinguished appointment was to become Ruth Kirschstein's husband fifty years 
ago.  He had this paper he was waving and he said, "would you give it?  Would 
you give it?"  And, I thought at this point I will do anything.  Ruth has been 
ill I think just with a flu-like illness for a while, and I said, "certainly, I 
would do anything for Ruth Kirschstein who has been a very old and dear both 
friend and colleague for probably twenty years.
	I thought that you should know that we are friends and colleagues because 
our remarks really have a remarkable convergence.  I think that we see this 
issue very clearly and very similarly.  But so that I could break up, not saying 
the same thing twice, I thought I at least ought to try to look a little bit 
more like Ruth Kirschstein as I was giving this talk and so I will become short 
--
	[Laughter]
	-- Ruth is short!  This is what she wanted me to tell you.  I will tell 
any of those from NIH who have participated in her speech-making that all 
changes were made by Ruth, not by me, and there are little edits all throughout 
the paper so I will try to give the correct delivery of her talk.
	The National Institutes of Child Health and Human Development of the NIH 
is pleased to join the Food and Drug Administration in co-sponsoring this 
conference on Clinical Pharmacology During Pregnancy, Addressing Clinical Needs 
Through Science.  The focus of the meeting will be therapeutics during the 
second and third trimester of pregnancy.  The Food and Drug Administration 
already has initiated a major effort concerning the onset of labor during the 
first trimester.
	The goals of this conference are threefold:  To summarize the state of 
knowledge regarding clinical pharmacology and pregnancy; two, to raise awareness 
among clinical researchers about the need for such research and for 
collaboration in these areas; and, three, to garner support for such research 
from health advocacy groups.
	Although most women and their physicians seek to minimize the use of 
medications during pregnancy in order to protect the developing fetus from any 
potential adverse effects, it is not completely unheard of that over-the-
counter-drugs and on occasion even prescription drugs are used by pregnancy 
women.  We know that based on the National Collaborative Perinatal Project, 
conducted by the National Institutes of Neurological Diseases and Stroke years 
ago, slightly more than three pharmaceutical products were used during per 
pregnancy.  Similarly, in Scotland, it has been reported that 82 percent of 
pregnancy women receive prescription drugs.  In addition, pregnancy women often 
use non-prescription drugs for minor symptoms.  In one study it was observed 
that 75 percent of the pregnancy women surveyed used mild analgesics during 
pregnancy, and 41 percent used cough and cold preparations.
	There are many issues that must be considered prior to the administration 
of a drug during pregnancy.  For example, if a drug treatment being considered 
is for a maternal disease, one needs to know if the pregnancy alters the 
therapeutic effectiveness of the drug by altering its pharmacodynamic or 
pharmacokinetic characteristics.  One also needs to be aware as to whether the 
pregnancy will increase or decrease the chance of toxic drug effects on the 
mother or on the fetus.  If the drug is planned to be used to treat fetal 
disease, one must decide on the appropriate route of administration via the 
maternal organism into the amniotic fluid or directly into the fetus.  If one 
were to decide to treat the fetus through the mother, the pharmacokinetic levels 
must be determined to be assured that the drug can reach in utero for a 
significant length of time to influence the course of the fetal disease.
	A fundamental tenet of clinical pharmacology is the relationship that 
exists between the pharmacologic and the toxic response to the drug, and the 
concentration of the drug at the site of the action.  The many physiologic 
variables that will dictate adjustment of drug dosage in pregnant women often do 
so as a result of pharmacokinetic parameters.  The three most important 
parameters are bioavailability, or the fraction of drug absorbed into the 
systemic circulation; the volume of distribution, the measure of the apparent 
space in the body available to contain the drug; clearance, or a measure of the 
body's ability to eliminate the drug.  All of these important parameters are 
changed as a result of physiologic alterations of pregnancy.
	Data regarding pharmacokinetics and pharmacodynamics, as mentioned 
previously, are rarely available.  The study of a drug used in pregnancy is one 
of the most neglected areas in the field of drug research.  We all know the 
reasons for this.  Of interest is the fact that most studies dealing with 
pharmacokinetics have used male subjects.  We need to compare the effects of 
drugs on pregnant and non-pregnant women to ascertain the physiological 
adaptations during pregnancy that could profoundly affect pharmacokinetics.
	As you know, the NIH requires that clinical trials include an appropriate 
number of female subjects.  However, since most drug studies are supported by 
the pharmaceutical industry, these rules do not apply.  Physicians who treat 
pregnant women frequently prescribe medications that have never been approved by 
the FDA for obstetric patients.  This off-label use and the lack of obstetric 
labeling is the direct result of the lack of research and clinical trials of 
drugs in this special population.  Studies are needed in this regard.
	So, it is my hope that this conference will provide the background and set 
the stage for NIH and FDA to work together to develop the necessary studies 
regarding drug usage in pregnant patients, to provide data to practicing 
physicians and to allow informal decisions regarding drugs prescribed during 
pregnancy.  This will lead to rational therapeutics in this special population.
	And now for my talk, on behalf of the Food and Drug Administration, I 
would too like to join Dr. Woodcock in thanking each of you for taking the time 
to join us today to hear our perspectives and to share your views about the 
important subject of clinical pharmacology curing pregnancy.  I would also like 
to thank Dr. Kirschstein, Dr. Dwayne Alexander from NICHD and Dr. Vivian Penn 
from the NIH Office of Women's Health for their support and partnership.  It is 
with your help that we have been able to hold this conference to raise awareness 
of this issue and to facilitate the development of much needed research.
	It is a great privilege to jointly host this conference with NIH, and 
particularly with the NICHD.  The NICHD workshop, held last September, generated 
much enthusiasm and support for expanded research efforts in this area.  Our 
goal for this two-day conference is to broaden the range of ideas and actions 
that can be taken to facilitate the research that is needed in this area, and to 
urge other organizations to participate in a matter of importance to the health 
of women and their unborn children.  Acquiring the data necessary to make good 
treatment decisions will take the cooperation of all interested parties, from 
clinical study patient volunteers to academic investigators, to industry-
supported R&D, to government agencies.
	Interest and concern related to the pharmacology or drugs used by pregnant 
women is not new.  This issue has traditionally been addressed through pregnancy 
labeling of medicines using a letter category system.  The system assigns the 
letters A, B, C, D or X to a drug based on its risk to the developing fetus, 
with drugs in category A presenting the least risk and drugs in category X 
presenting the greatest risk.  The categories are applied to drugs depending on 
the human and animal data available, and for most drugs data are sparse.  Fully 
two-thirds of approved drugs are in category C because data in pregnant women 
are lacking and animal studies have either revealed adverse effects or have not 
been done.  Having this label doesn't give the prescribing physician the kind of 
information that would help to guide the physician in their decision-making.
	Most recently, FDA's pregnancy labeling task force has made great strides 
in developing a new model label that would distinguish clinical advice from risk 
information, providing different levels of information with narrative text 
instead of the letter categories.  Whether the old category or new narrative 
system is used, it is time -- some would say long overdue, to assess our current 
understanding of the pharmacology in pregnant women and start acquiring the 
additional data necessary to allow health practitioners to appropriately 
prescribe drugs for women during pregnancy.  Drug labeling in any form can only 
go so far without appropriate data to support it.
	In spite of the current lack of information, there would also appear to be 
little reluctance to prescribe.  Prescription drugs used by pregnant women are 
more common than you might think.  In 1995 FDA conducted a study that examined 
an HMO database over two years, looking specifically at the number of 
prescriptions taken by women during pregnancy.  This study excluded the commonly 
prescribed prenatal vitamins, iron supplements and tocolytic drugs.  
Surprisingly, the findings were that women under 35 years of age were taking on 
average three prescriptions during the course of their pregnancies, in women 
over 35 this average rose to five prescriptions.  For someone who is getting 
well over 50 and takes only one prescription a day, I think this seems like a 
lot.
	Some of the drugs used most extensively by pregnancy women fall into the 
therapeutic categories of antibiotics, analgesics and narcotics.  Other commonly 
used drugs include anti-epileptics, antihypertensives, anti-nauseants, 
psychotherapeutics and respiratory therapeutics.  For most of these drugs no 
well-controlled studies have been conducted in pregnant women.  These therapies 
are used to treat conditions that affect not only the women's health but also 
impact the health of the developing fetus.  In most cases, stopping these 
treatments during pregnancy is not a reasonable option even though the risk may 
be unknown or poorly characterized.
	Certainly the well intended but misguided practice of prescribing lower 
doses of drugs to pregnant women to minimize exposure to the fetus is not the 
answer.  What often happens when this is done is that the patient receives a 
lower dose but requires a longer period of treatment or even a different drug.  
It has long been appreciated that the changes in women's physiology during 
pregnancy are dramatic, particularly during the second and third trimester, and 
can significantly influence pharmacokinetic profiles of many drugs.  Because the 
clinical pharmacology of many drugs has not been systematically studied, our 
understanding remains limited.  The lack of data impedes the careful selection 
of safe treatments, appropriate dosages and what might be done when treatments 
fail.  Simply put, this lack of knowledge puts women and their unborn children 
at risk.
	It is very troubling that while pregnant women are a population of 
patients where we want to have the most certainty about the effects of drugs on 
both the patient and the fetus, it is an area where we have the least data.  For 
what some may view as ethical and others as paternalistic reasons, pregnant 
women have generally been excluded from clinical trials.  For many 
investigational products this may well be a sound and reasonable course.  
However, once a drug is on the market we know that there may be inadvertent 
exposures during pregnancy, for example, when a woman learns that she is 
pregnant while taking a drug.  In add, pregnant women often require treatment 
for conditions or illnesses that occur during pregnancy or that were 
preexisting.  These drug exposures provide opportunities to collect data on the 
pharmacology of specific drugs in pregnant women and, conceivably, in every 
trimester of pregnancy.
	These opportunities to collect data have often been missed or ignored.  We 
need to do a better job of gathering this information and making it available in 
a way that will help physicians and the women that they care for make the best 
treatment decisions.
	The need for research is apparent, and must be undertaken in a way that is 
based on the deliberative process of science but encouraged and expedited by the 
support of both health professionals and women's health advocates.  As such an 
effort is undertaken, the studies need to be carefully designed and the highest 
ethical standards met as all matters related to human subject protection are 
considered.
	There is another concern that relates to unknowns.  It is that many 
pregnant women are turning to herbal remedies and dietary supplements, believing 
that they might be safer than synthetic drugs.  But since there is often even 
less safety information about the effects of these remedies in general, and 
especially in pregnant women, this represents a risk of unknown magnitude.  
Combining drugs with these products complicates the picture even further.
	Let me leave you with this message, it is time to commit and undertake 
this work.  It is encouraging to see that we are on the right track but we need 
action, and it is going to take action on the part of each of us.  Will Rogers 
once said even if you are on the right track you will get run over if you just 
sit there.
	[Laughter]
	We have been sitting long enough.  I thank you for being here and for your 
interest, your participation and your willingness to engage in this important 
effort.  Thanks.
	[Applause]
	DR. WOOD:  Thank you very much, Dr. Henney.  I think both of your 
presentations laid out the importance of this work and the importance of both 
the FDA and the NIH to engage very actively on it, and engage very actively on 
it not only among ourselves but with the research community and with the 
industry to, again, come up with ways to really get at this problem because it 
is not a trivial one, as your presentation demonstrated.
	As I mentioned, we are going to shift our order.  So, I will now introduce 
Dr. Sandy Kweder.  She is, as you will note, at the end of your day but we are 
going to move her up now.  I think it is really very relevant as we hear from 
our government leaders to also hear some details about FDA's current practices.
	Dr. Kweder is the deputy director of one of FDA's offices of drug 
evaluation within CDER.  Her office oversees the regulation of drugs to treat 
infectious diseases, including antibiotics and antivirals, as well as managing 
several large cross-center initiatives.
	Dr. Kweder graduated from the Uniform Services University School of 
Medicine and began her career in what was then the division of antiviral drugs, 
in 1988, and she has been involved throughout the agency for a number of years 
in senior positions.  Her interest in the medical issues related to pregnancy is 
long-standing.  In 1995 she took a sabbatical and too, a two-year fellowship in 
Obstetrics and Consultative Medicine at Brown University's Women and Infants 
Hospital.  Since she has returned, she has led the FDA's efforts to revise how 
drugs are labeled for use in pregnancy and lactation, and serves as the Chair of 
the Pregnancy Labeling Task Force.  Dr. Kweder?
Current Requirements for Labeling Products for Use in
Pregnancy and FDA Initiatives
	DR. KWEDER:  Good morning, everyone.  I appreciate all of you braving the 
cold to come and join us here this morning.
	[Slide]
	Some of the themes that you have heard from Dr. Woodcock, Dr. Henney and 
Dr. Kirschstein in absentia I am going to try and put into a little bit more of 
a historical context and ordered framework so that you can understand how we, at 
the agency and the NIH, came to feel that it was time to begin to bring people 
together to talk about the science in this field.
	[Slide]
	So, my objectives for today are to talk a little bit more specifically 
about our experience with labeling drugs for use in pregnancy.  I will also talk 
about the development of a new label model that we have been working on at the 
agency and with others.  I will talk, third, about some related initiatives to 
this project that this conference comes out of and, specifically with regard to 
this conference, the critical role for clinical pharmacology data in 
appropriately labeling drugs for special populations.
	[Slide]
	First, I have to ask you to endure a couple of slides about what I call 
labeling 101.  I think I have two or three.  This is the real sleeper stuff, but 
I think it is important that we all begin at the same point in thinking about 
this.
	Product labels, as we have heard the term a couple of times already today, 
are legal documents, those things that in the tiny print that go into the 
Physician's Desk Reference.  When I say label, that is what I am talking about.  
There are other definitions; that is the one we are talking about, when it is 
folded up into little teeny pieces at the back of your new coffee mug.  That 
label is a legal document.  It is a vetted source of information vetted 
information that is widely available and the vetting is that the label is 
produced by a pharmaceutical company but is vetted by the FDA for accuracy.  One 
of the reasons it is extremely important is because it provides information 
which forms the basis for any particular product approval for marketing or 
licensing in the case of biologics.
Because of that, it provides a summary of the basis of information upon which a 
company can promote the product or advertise it.  It is also a basic source of 
information from which many other sources of information come.
	The FDA doesn't conduct the research that goes into those product labels.  
I think it is important to be clear about that.  We are not talking here today 
about research that the FDA does.  Companies conduct the research.
	[Slide]
	What the label is supposed to do is that it is supposed to provide 
clinicians with information about the indications for which the product has been 
shown to be effective.  For example, to treat high blood pressure, or to treat 
migraine headaches -- this product has been shown to be effective to treat those 
conditions.
	It also, importantly, provides a description of the safety information 
available about that product.  That may include information about special 
populations such as women, the elderly, children, ethnic minorities and, in some 
cases, pregnancy.
	The product label also guides the clinician regarding appropriate dosing 
and administration of the product -- what is the right dose in altered 
physiologic states?  For example, lots of times you can go to a product label 
and see things about what to do if your patient has altered renal physiology, or 
altered liver function.
	[Slide]
	The pregnancy section of the label was first addressed in regulations in 
the late 1970's, and the goal of those regulations was to try to guide the 
clinicians who were struggling with making decisions about prescribing medicines 
for women who were pregnant.  The goal of those regulations was not to help 
physicians who were working with patients struggling with decisions about 
inadvertent exposures, which are perhaps even more common than specific intended 
use in pregnancy.
	As Dr. Henney mentioned, the system post risk and benefit into a letter 
graded category system which I think most of you are familiar with.  
Interestingly, although it appears to be risk graded, A the best, X the worst, 
it is not quite so simple and X is not an X, is not an X, is not an X 
necessarily because it combines risk and benefit.
	As a final point, and I will say a little more about that in a minute, I 
want us to be clear that historically the agency has not considered pregnancy 
its own indication.  Pregnancy is an indication for use in regulatory terms when 
a drug or biologic product is developed to treat a condition that is unique to 
pregnancy.  We historically always considered pregnancy a special population.  
For example, if a drug is approved and has a labeled indication to treat 
migraine headache in adults, it is not that it doesn't have an indication in 
pregnancy, the pregnancy information part of the label is designed to provide 
additional safety data to guide the clinician who may be dealing with a woman 
who has been exposed to that drug in pregnancy or has headaches and needs a 
prescription in pregnancy.  So, it is a special population, not necessarily its 
own indication.
	[Slide]
	I want to say a few more things about our experience in applying the 
pregnancy categories, the A through X.  Most products, at the time they come to 
the market and forever really, in labeling only have information available from 
animal studies.  The nature of those animal studies is such that most drugs end 
up in pregnancy category C, which says that we saw something in the animal 
studies and it didn't look quite right.  Okay?  The nature of the way animal 
studies are done is such that they are designed to stress the system and elicit 
toxicity.  That is not wrong, it is just that that is what we then have to deal 
with.
	Unfortunately, we aren't always certain what the predictive value is, 
positive or negative, of those animal reproductive toxicology studies.  We don't 
know if a positive in rats is likely to lead to a positive in humans all the 
time, or if a negative in dogs is likely to lead to a safe drug in people all 
the time.  We don't really understand that predictive science in this field very 
well.  Consequently, when you have positive findings in animal studies they are 
very difficult to make go away because the studies are the studies.  Therefore, 
the information in labels is rarely changed or updated.  Even with additional 
human data, you can never get rid -- we have a very difficult time ever saying, 
you know what, those animal studies where we saw something or we didn't see 
something were helpful.  They always remain.  Consequently, labeling and the 
pregnancy categories that we have are almost never changed.  They are rarely 
updated.  Historically, our experience at the agency is that companies and many 
others, as well as ourselves included, have for a long time perceived warning 
language in labeling in this section as optimal.
	[Slide]
	Yet, the real conundrum for us is that what we want is more and better 
data, particularly in the clinical setting of use of these products.  When a 
clinician is dealing with a woman who is pregnant and her unborn child, what she 
and the clinician need is the best and most data available and, unfortunately, 
it is the area of science where we have the least data and we have less than 
good knowledge about the predictive value of some of the data.
	[Slide]
	To try to address that, in the mid 1990's the FDA established a Pregnancy 
Labeling Task force which is a group mostly of people within the agency but also 
has included some people from outside the agency to begin to look at this 
conundrum.  We had three major tasks before us:  First, to examine that A, B, C, 
D, X letter category system; second, to try and recommend changes if we thought 
they are needed; and, third, to try to consider the bigger picture of related 
needs, particularly in the area of science.
	[Slide]
	We began the process with task A, looking at our current regulations.  
When the government has regulations and they want to know what other people 
think of them, they do this thing called "hold a hearing" and there is even a 
name for it, they are called Part 15 hearings.  That means that we sit at the 
table and let people come, tell us what they think.  They are always 
interesting.  We don't have to do a lot of work for them; we just have to 
listen.  We held one of those in 1997 and we asked the questions on this slide:  
Is the pregnancy section of the labeling relied on by practicing healthcare 
providers?  Is it useful?  If it is useful, how is it useful to you?  We wanted 
to know if it is good, what is good about it.  If it is not good, what is bad 
about it.  Tell us specifically.  If, overall, you think that the way we do 
this, this letter category system, is not informative or it is excessively 
problematic, what do you suggest we do to improve it for you?  How can we make 
it better?
	We had a full day of feedback, and there were people who gave us positive 
feedback about the system.  We were told that the information in labels, in this 
section of the label in particular, is relied upon by healthcare providers, and 
the things that people cited as being good about it were its simplicity.  There 
were people who came to the hearing who told us that what they liked was that 
you could condense the information in the label, the body of data, down to 
single ordered letters.  They liked the idea that A was good and X was bad.  
They also liked the idea that it helped because you could make decisions quickly 
as a clinician by not having to interpret very complex data.  It was all done 
for you in a simple letter-graded system.  They also liked it because it was 
something that was familiar.  We all like things that are familiar to us and 
that we know.
	[Slide]
	On this slide I have four bullets that are a sampling of the criticisms we 
had of the system.  The first time I tried to give an overview of the Part 15 
hearing to a group -- it was actually a group of internal people, I was able, 
with great pain, to cut it down to ten slides.  So, I think I have come a long 
way in getting this down to one, with four bullets.
	But overwhelmingly people were critical of the system.  The kinds of 
people that came to the meeting were professional societies, consumer groups, 
practicing healthcare providers and the industry.  These sample criticisms I 
think sum up most of the comments of the day.  What people liked about the 
system was also its greatest criticism, and that was its simplicity.  We had 
many speakers who told us that it is overly simplistic and were able to provide 
numerous examples.  The concern about over-simplicity was not because it is not 
good to simplify but the concern was because of what happens when one tries to 
apply a system that is overly simplistic.  The concern was that what often 
happens to women and their clinicians is that they end up taking a very passive 
approach to an extremely complex scientific and clinical situation.  Because of 
that, many of the presenters told us the system does not facilitate clinical 
decision making.  What it does, it allows people to just knee-jerk -- "oh, I'm 
going to prescribe this drug because it is a B, not a C or a D" when, in fact, 
there is not very much difference between them.
	The other criticisms were the fact that the descriptions of the data that 
did exist, when they existed were very difficult to understand and, most 
importantly, that updates of the labels were extremely rare.  The point was made 
to us that you have no teeth in your regulations.  There is nothing that is 
making anyone update these things when we know there is literature out there 
that tells us that there is more information than what you have in the labels.
	[Slide]
	So, the take home message for us was that the current system is 
uninformative and, most importantly, that we needed to move away from an overly 
simplistic system of letter categories and move to something that is more 
descriptive.
	[Slide]
	That was in late 1997.  Since that time, we have been in the process of 
developing a new model for how to do this.  Our goals in doing that are to 
provide a practical system that is useful to clinicians and patients, a system 
that clearly provides a risk assessment statement and what the basis is for that 
statement -- if it is animal data or, in particular, human data.  We also 
recognize that there are different levels of needs.  We know that patients read 
this section of the label.  We know that primary care providers read this 
section of this label and we know that specialists and consultants read this 
section of the label.  They all have very different needs, needs for very 
different kinds of information and we need to at least find a middle ground and 
meet most of them.  The process that we embarked on ourselves was to try and 
seek multi-disciplinary consensus within the agency, but not do so in a vacuum 
and seek public comment along the way, which we have been doing.
	[Slide]
	I will give you a preview of the new format that we have been developing.  
It will have three components for the pregnancy section.  There will be a 
section that we are right now calling clinical considerations which tries to 
bring in information that clinicians need to consider to link risk assessments 
to practical applications of the data.  For example, is there anything special 
for an antihypertensive?  Is there anything special about hypertension in 
pregnancy that the clinician, who may not be familiar with this, needs to 
remember before they think about the risk assessment provided?
	The second section is the risk assessment itself.  We need to do a better 
job of incorporating both the human and animal data into that risk assessment 
statement than we currently do, and state the relevant factors that we have 
considered in doing so.
	Of course, there is always the discussion of data, and we need to do a 
better job at providing descriptions of the kind of data that went into the risk 
assessment statement in the first place.
	[Slide]
	So, we are here in December, 2000.  We are in this thing where we actually 
have a draft version that we are circulating within the agency right now.  We 
have taken this model to a public advisory committee meeting in the past year.  
We have taken sections of it to outside groups for focus testing among 
healthcare providers, particularly in the area of standard risk assessment 
language.  That has been a very useful exercise for us because we want to make 
sure that the messages that might be conveyed are the ones that we intend -- 
that people hear what we intend them to hear.  So, using the focus testing 
method has been a very a good tool to help us do that and we will probably 
continue to use that as we try to develop this further.
	We have also begun the process ourselves of what I call beginning the 
internal cultural shift.  You know, one of the struggles with this is that this 
section of labeling and this kind of information about use of drugs in pregnancy 
has really not been something that clinical reviewers or physicians at the FDA 
have paid a lot of attention to historically over the years.  That gets back to 
the point I made earlier about warning language being perceived as optimal.  No 
one likes to have to confront this.  This is really hard stuff, but we need to 
confront it; it is time, as Dr. Henney said.
	[Slide]
	I think it is a fair question for you sitting in the audience, listening 
to me wax about this, to ask why does this take years?  I mean, you guys started 
this -- what? -- you said, "this Pregnancy Labeling Task Force was in the mid-
90's; hey, it's five years later.  Why does this take so long?"  Well, one is 
because the science itself is sparse and very challenging.  Second, new 
regulations are intentionally slow.  We want to be sure that we don't have to do 
this again in ten years or five years, God forbid.  Third, as Dr. Henney so 
nicely pointed out, more is needed than just a different way of saying the same 
old stuff.  We need more than a new way, you know, editing on your word 
processing document to get information out.  We must raise the expectations for 
data quality and data quantity.  And, that is why we are here today.  There are 
many obstacles to developing the science that is so sorely needed.
	[Slide]
	You remember when I started my talk I said that one of the tasks that we 
were given as a task force within the agency was to look at what are the broader 
needs of labeling; what else do you need to get this done well?  And, we were 
able to identify four areas.
	[Slide]
	The first one of which is agency expertise within the FDA itself, which we 
have been working on since that time.  We have a couple of internal documents 
that we are trying to work with our reviewers on, on how to assess human 
pregnancy outcome data, to look at the safety equation; how to think about 
reproductive toxicology data when we are looking at animal data.  Do we do this 
well?  Do we do this consistently?  Are our standards up to the standards out in 
the broader community of reproductive toxicology?  And, we have spent a lot of 
time and energy working to raise the level of scientific expertise of our 
reviewers.  We have put together an advisory committee of outside experts to 
help us along the way with many of these steps.
	[Slide]
	In the area of improving data, we have several initiatives ongoing and I 
am going to touch first on the point of safety data.  Before I even begin, I 
want to say safety data comes in two areas.  One is general safety, what are the 
safety outcomes that we are concerned about for the mother and baby of taking 
any medicine, regardless of its intended therapeutic effect?  The second safety 
part of the equation is dosing, getting the dose right and I am going to get to 
the dose right part at the end.  So, I want to touch on general safety.
	We do have a new safety reporting regulation that is in the Office of 
Management and Budget now that will specifically say that pregnant women are a 
special population, and where there are new data that tell us something about 
how to use a product safely or about the safety of a product in this special 
population, that information needs to be carefully thought about by the company 
who markets the drug, and incorporate it into the product label.  That issue has 
to be revisited every year.  So, we have a new tool that, hopefully, will be put 
to work in helping us find all the data that is disparately out there and have 
it come into our hands so that we can work with companies to get it into the 
label.
	In the area of providing more and better data, one of our focuses has been 
on how can we do this with prospective data collection.  The approach that we 
have taken to safety data in the past has been solely, for the most part, to 
hunt for toxicities; to wait for the spontaneous reporting system to tell us 
about bad outcomes and then we put those in the label.  That is important.  That 
system should remain as it is and we have to use it well.  But what we hear from 
clinicians is we need more than that.  Please help us move from this hunt for 
toxicity to a better understand of the margins of risk and the margins of 
safety.
	We have tried to do that a little bit by working with companies in 
thinking about how to put together prospective safety studies and capture the 
experience of the medical community where we know that products are being used 
by pregnant women all the time.  One of the things the FDA does is they develop 
these things called guidances.  They are not regulations but they reflect 
current thinking on how to establish a term that has been coined that is 
pregnancy registries, or prospective or active surveillance studies of product 
use in pregnancy so that we can capture some of those margins of risk and 
safety.
	[Slide]
	I think in the future that model is going to change.  Even the system that 
we have now where some companies or some groups do studies for one drug, one 
registry; one drug, one prospective safety study -- I think we are going to see 
an expansion of that.  A conference was  held just last week by the CDC that we 
helped put together to begin to look at what are all the efforts ongoing now in 
this country and internationally where folks are interested in at looking at the 
big safety picture of drugs in pregnancy.  And, boy, there are a lot of them.
	Well, how can we capitalize on that?  How can we bring together all those 
disparate efforts so that they can by synergistic and we can understand the body 
of data that they are telling us?  So, I think what we are going to see in the 
future is a more coordinated approach to collecting these kinds of prospective 
safety data and trying to make them make sense with our passive postmarketing 
surveillance system.  I think the bottom line from that CDC meeting last week 
was that we can do this better.  We have the tools to do it and we must.
	[Slide]
	Area number three of data that really crosses into the safety equation is 
lactation, and this meeting isn't specifically about lactation but it always 
comes up in this context so I wanted to make sure that I said a few words about 
it.  You know, labeling should facilitate clinical decision-making.  People who 
are reading the lactation section of a label ought to be able to understand or 
get a picture of the risks and benefits of a drug to the infant who may be 
exposed to it by breastfeeding.
	Clinical pharmacology has a lot to teach us about what to expect in that 
area, but right now our product labels are not especially informative in this 
area.  For those of you who have read them, you know that that is a gross 
understatement.  In fact, they are probably worse than the pregnancy section of 
labels.  They say almost nothing at all.  We think that there are data available 
or that could be readily developed to go into this section of the label so that 
this section could be informative to clinicians and women who are trying to make 
decisions about how to manage breastfeeding and taking medicines.  One of the 
things that has been well pointed out to us is the sharp contrast between our 
label and what experts say and what some of the goals are regarding 
breastfeeding mothers and healthy people in 2010.
	[Slide]
	We have done a couple of things in this area.  We are really only just 
beginning.  We held a joint meeting of our pregnancy labeling and pediatrics 
advisory committee meetings this fall to begin to consider some of the 
scientific principles that would have to go into any data collection, and we 
actually have a guidance document that is under development now to help 
investigators who are interested in pursuing this area of research to assess the 
presence and extent of drugs in breast milk.  The new labeling rule, the model 
that outlined for you earlier, will apply the same model to lactation as it does 
to pregnancy -- clinical considerations, risk assessment and data.
	[Slide]
	The fourth area of data is clinical pharmacology data, and this principle 
holds for any drug in any population, whether we are talking about pregnant 
women or non-pregnant women, men, children, the elderly.  Once treatment is 
decided to be needed, every patient should get the right dose for them and every 
patient should not be an N of one experiment.  Few labels, however, include 
dosing information or even basic information about the pharmacokinetic profile 
of drugs in pregnancy.  The consequence of that is, as Dr. Henney so nicely 
mentioned, is that many clinicians think -- and this is the right way to think -
- I want this woman to be exposed to the least amount of drug possible.  I want 
to do the least amount of harm I can.  So, I will go with the lowest dose I 
think I can get away with.  As I think you will hear in talks later today, that 
may not always be the right decision.  It is certainly well intended but we have 
the scientific tools to help us help the clinician get the dose right.
	[Slide]
	You will hear later today about some of the physiologic changes in 
pregnancy that are part of the scientific discussion -- an expansion of plasma 
volume, an increase in cardiac output, changes in renal blood flow, and other 
regional blood flows, changes in enzymatic activity and metabolism, altered GI 
motility and possibly absorption, altered plasma albumin and other protein 
levels.
	[Slide]
	All of those things are physiologic changes that have the potential to 
change the pharmacokinetic profile of drugs when they are taken by pregnant 
women.  Now, interestingly, the FDA has published several guidances to help 
researchers and companies assess changes in pharmacokinetics and dosing in 
situations where physiology is expected to be altered.
	In 1998 we published a guidance on how to do studies, what to think about 
in the area of renal impairment.  In 1999 we published a guidance on what to 
think about when drugs might be altered by hepatic impairment.  In 1998 we also 
published one on children where physiology may be altered.  We have not 
published any such document or anything at all that would guide anyone who was 
considering the altered physiology of pregnancy and what might be done to study 
the pharmacokinetics of drugs in that altered physiologic state.
	[Slide]
	Again, our new safety regulation will help us bring some of the data that 
are out there in the scientific literature to attention and for thought about 
whether or not it should go in the labeling.  NICHD held a workshop on this 
topic several months ago that you will hear more about in the course of this 
meeting.  The American Society of Clinical Pharmacology and Therapeutics is 
going to hold a symposium as part of their annual meeting, on this topic of 
clinical pharmacology and pregnancy, in March of next year.  I think what all of 
these efforts will come together to tell us is that there is science to be 
developed, and that is why we are here today.
	[Slide]
	Where we need to go in this area is to collaborate and find ways to 
facilitate this type of research.  The clinical needs of healthcare providers 
and women who have health problems and are struggling with decisions about 
therapeutics in pregnancy are great.  But, what we need to do is we need to try 
and resolve those problems through sound scientific development.
	Clinical pharmacology is one tool, and an important tool, as a part of 
evidence-based medicine which is the gold standard in this country and in the 
world increasingly.  And, we are here to try to be a part of developing 
effective programs that bridge clinical disciplines and help build partnership.  
Over the next day and a half you are going to hear about how we can confront 
real and perceived barriers to conducting ethical and sound scientific research 
in this area.  Some of it comes from perceptions about this part of science 
being reproductive health and not women's health.  It is both.  Some of the 
discussions you will hear are about how can we conduct these studies ethically; 
how can we conduct these studies in a way that makes us not too nervous about 
liability and legal barriers.  Where could funding come from for this kind of 
research.  Hopefully, we will have some consensus on how to move forward in 
those areas.
	[Slide]
	So, in summary, within the FDA and through collaboration with other 
agencies we are making steady progress in many areas that will bring information 
to clinicians through labeling of products for use in pregnancy.  Our focus has 
increasingly been on trying to emphasize and enhance information that clinicians 
and patients need to have at their fingertips on a daily basis.  To do this, our 
job is to think about what the models are for data collection and think beyond 
our standard way of doing things for many years.
	Where that brings us today is to thinking about clinical pharmacology.  To 
do that, collaboration is essential and I want to say that, from our standpoint, 
we feel like the public deserves our best collective effort, and I want to thank 
you for coming today.
	[Applause]
	DR. WOOD:  Thank you.  I think that provided us with a lot of information 
on what the FDA is thinking and where the issues are that we need to develop and 
work on today, backed up by somehow getting the data that is going to be 
required to make all of this functional.
	We are going to move from inside of government to outside of government.  
Dr. Diana Zuckerman is here today to give us the consumer perspective.  But she 
also brings with her a long history as a researcher and author, and as a policy 
person in Washington, representing the interests of consumers and women.
	Dr. Zuckerman is the President of the National Center for Policy Research 
for Women and Families, where she has been for, I guess, a little bit over a 
year.  She has a Ph.D. in psychology from Ohio State University, and did her 
post-doctoral work at Yale, and then served as a faculty member at Vassar and at 
Yale, as well as doing research at Harvard.
	From 1983 to 1993 she served as a congressional staffer, working for the 
House subcommittee that had jurisdiction over the Department of Health and Human 
Services, including FDA, and was the instigator, if I can say such a word, of a 
number of congressional investigations and hearings on health issues, many of 
them relevant to FDA.  She then went on to work at the Center for Mental Health 
Services but then came back to Congress to work on the Senate Committee for 
Veterans Affairs during the healthcare reform days, working for Senator 
Rockefeller, and then moved on to work at the White House, working for the first 
lady and the Office of Science and Technology Policy.  She then moved on to work 
in a number of leadership roles at some non-profit organizations, and then moved 
on to the National Center for Policy Research for Women and Families, which I 
believe she founded in 1999.
	She has been highly cited in the press and has been on all sorts of 
national media, and has authored four books and many articles in both scientific 
journals and national newspapers.  I have known Diana Zuckerman since 1990 and 
have worked with her on a number of initiatives, and I know she tries to tie 
together the scientific basis of questions that lead to health outcomes, 
positive or negative, for women but always trying to bring it back to the data 
that we have and the data that we need.  Diana?
Consumer Perspective
	DR. ZUCKERMAN:  Thank you, Susan.  I always like hearing about my 
checkered past.  Because I am the president of an organization called the 
National Center for Policy Research for Women and Families, when I am talking 
about the consumer perspective I am going to be talking about each pregnant 
woman as representing at least two consumers, the woman and the fetus.  And, 
since we are talking about women who presumably most of the time are aware that 
they are pregnant and want the baby, those are going to be the assumptions that 
will underlie the remarks that I am going to make.  Of course, many women don't 
know that they are pregnant when they are taking prescribed drugs but then, 
again, the issues that we are talking about are mainly going to be focused on 
those who do, I think.
	When I was thinking about what I wanted to say today, I have always 
assumed that one of the reasons why we have so little research on pregnant women 
and drugs effects is because of ethical concerns of doing research on pregnant 
women and concerns about liability.  But, as I heard more about the issues that 
FDA has been dealing with, it actually brought home to me that there might be 
something going on other than that.  Many of us are old enough to remember what 
it was like to go to schools where teachers who were showing their pregnancy 
were not allowed to teach.  We grew up in a time when there was a sense that, 
you know, pregnant women were a different group of people, a special population, 
and there is nothing wrong with being a special population but I think there is 
a little bit of this mentality that, you know, pregnant women are this different 
group.  They aren't the women that take the same prescribed drugs as everybody 
else, or who have the same illnesses sometimes as other people and, of course, 
there are illnesses that tend to get worse for women who are pregnant.
	So, as I was listening, it occurred to me that, on the one hand, I am very 
pleased with so many of the initiatives and the ideas that FDA has about how 
things need to change, and how much more information we need, and how we have to 
deal very differently with these issues but, on the other hand, I was also 
shocked as a consumer rep that somehow we are still thinking of women in this 
other way and we are so far behind in terms of the information that we have that 
we can give to pregnant women, or our recognition of how quickly we should be 
getting information to them when it becomes available.  So that really, really 
struck me.  We are really far behind.  I don't know if we are sitting on the 
track or sitting on the train that is not moving, which is what happened to me 
on the Metro this morning --
	[Laughter]
	-- but whatever it is, we have a long way to go and I am really proud and 
pleased to be here, and hope that we can start moving in the direction that we 
need to go.
	So, it seems to me we can't talk about safety issues and efficacy issues 
for the woman who has some medical needs and is pregnant without talking about 
these two things -- what is safe and effective for her as a woman and what is 
safe for the fetus, the baby that she wants to have.  We could even get into the 
whole issue of the consumer being even broader, the family as a whole, but let's 
just stick with these two to start with.  I think you have already heard this 
morning that the FDA has found that the average woman has three prescribed drugs 
during pregnancy, and the older she gets the more she has, and women over 35 
average five prescribed drugs, not counting the vitamins and other things 
related to labor and delivery.
	When you think about these older women, some of them who very much want to 
have a child and this may be their last chance -- their first chance and their 
last chance or, at the very least, their last chance -- and how concerned they 
are to be able to have a healthy baby, that is a particular concern that they 
are taking so many prescribed drugs.  And, one wonders exactly what is going on 
and how we can help these women to have the information they need to make the 
choices that are best for them.  Clearly, the strategy that has been the focus 
so far, which is waiting for bad outcomes to be reported on drugs, is clearly 
not the way to go.
	I am skipping over some stuff, you will be happy to know, because some of 
it has really already been discussed.  The basic problem here that I think we 
have to address, as we talk about what the consumers need to know, the basic 
problem, of course, is the lack of research.  We can't go anywhere without 
acknowledging that up front, that women aren't told what they need to know 
because at least in some cases nobody knows what it is that these consumers need 
to know.  We don't have the research that we need.  And, we can't talk about 
improving the warnings in a really meaningful way until we have better research.
	So, because of that, it might feel like talking about labeling and talking 
about how we should change the labeling system, and should we have consumer 
brochures, and how should we get this information to consumers -- it might seem 
that we are talking about, you know, shuffling the deck chairs on the Titanic if 
we don't have the basic research information that we need.  But I actually don't 
think that is true.  I think we do need to talk about what information we can 
get to women in the meantime, women and their doctors, even though we need to go 
full steam ahead on research.  But until we have that research and for women who 
might be participating in such research, what is it that we can tell them and 
what do we need to tell them?
	I know that some people think that the goal of these warnings should be to 
warn women about what the risks are without scaring them.  I actually disagree 
with that.  I don't think it is possible to adequately warn pregnant women about 
potential risks without scaring them.  I think that pregnancy for many women is 
scary, and I think that the lack of information about most drugs is going to be 
scary to many women.  In a way, I think that the whole purpose of warnings is to 
scare women or to scare consumers, to give people a sense that something bad can 
happen.  I mean, you don't want to make them hysterical; you don't want to have 
them panic, but if you haven't adequately explained that there are risks and 
that, guess what, even if those risks are rare it can happen to you or your 
baby, then we haven't really adequately done the job.  So, I think that this 
concern about scaring patients is a little bit paternalistic -- a term I don't 
use that much or lightly.  I think we really have to assume that consumers need 
information and that we should be providing as much information as we can to 
them, as clearly as possible, so that they can be part of the decision-making 
process.  It should not be up to the doctor only to tell consumers what the 
doctor thinks the consumer needs to know.  The consumer really needs to be a 
part of that decision-making process and they can't be part of that process 
unless they have all the information they can in a form that they can understand 
and that makes sense to somebody who may not be trained in epidemiology or 
pharmacology or whatever.
	So if there is no research on a particular product that a woman might want 
to take, a medical product that she might want to take or her doctor thinks she 
needs to take, or if she might be in some kind of registry or some kind of 
prospective clinical trial for somebody who really needs to take a drug but 
happens to be pregnant, then she really needs to be told very, very clearly that 
there is no research.  I mean, if that is all there is to tell her, she needs to 
be told that clearly.  And, it is fine to talk about, well, we don't think that 
this drug is dangerous because we haven't had bad reports of adverse reactions 
to pregnant women or to the fetus, or whatever, that is fine but I really think 
the consumers deserve to know when there is no research and that should be said 
just as clearly as the results of research would be told. 
	Similarly, if there are potentially relevant data on animals, whether it 
is good news or bad news, I think consumers need to know that.  Of course, they 
need to be told that we don't know exactly what the animal data implications are 
for humans but, still, I believe that consumers deserve to know that.  For that 
reason, I think that the current labeling system does have certain strengths.
	As you have heard, there are basically what seems like grades of A, B, C, 
D and X, and I know that there are some concerns about that system.  It is 
simple, but is it simplistic?  But really the main problem, I believe, with the 
labeling as it is, is not the set up of the labeling per se, I think the main 
problem is that there are so few drugs that get an A because there are so few 
drugs where there is clear evidence that they are safe, and there are not that  
many drugs getting a B, which has some evidence that suggests that they are 
safe, and so many -- hundreds and hundreds -- of drugs are getting Cs and Ds 
which means that basically we don't know very much about what the impact is 
going to be on the fetus.
	Personally, I think that the X, which indicates that there are studies 
demonstrating clear evidence of fetal abnormalities -- I think that X is a 
pretty clear warning and there is something to be said for having something that 
is a clear warning that you hope will at least get people asking the questions 
that they need to ask, whether that is the physician or the consumer.  It is 
good to have something that really is a warning flag, a red flag that gets 
people's attention.
	Similarly, a D, which indicates and conveys that there are concerns about 
the risks but also a belief that the benefits might outweigh the risks, I think 
again that as long as you have something that really has a kind of a red flag 
that brings concern so that people are focusing attention on what the risks 
might be, and then the physician can clearly state what they think the benefits 
are and why the benefits might outweigh the risk.  I think it is fine.  I am not 
saying that the physician doesn't have an important role here.  I am only saying 
that the consumer really needs to hear as much information as possible so that 
the consumer can be part of this decision-making process.
	Another part of the issue is how to get this information to consumers.  
Some consumer advocates complain that the current system isn't always that 
effective because women who are pregnant are not always told as much information 
as they should have even about these rating systems of drugs.  So, we need to 
figure out a way to get this information or, if it is revised in some way, 
whatever the information is, we need to figure out a way to get it more widely 
distributed and clearly presented to consumers, and it shouldn't depend on the 
individual doctor.  There would be other mechanisms and ways that women can get 
this information.
	I don't want to digress too far but, in a way, the situation we face is 
actually quite similar to the movie rating system.  Most of you probably know 
what a PG13 movie means, more or less, and what an R-rated movie means.  Yet, 
those rating systems have also been criticized lately and, in a way, for similar 
reasons.  One of the problems, if you have a 12-year old child and you want that 
child to be able to go to a movie that they would like to go to, is that the 
child is probably going to want to go to a PG13 movie or an R-rated movie and 
that means that those movies have content that a parent may not approve of.  But 
it is hard to make a choice of other kinds of movies because, just like the 
labeling system here, there are so many movies that are in those categories.  
You know, it is really very similar.  You have too many kinds of things in the 
same categories and not enough in the other categories.  So, parents have a very 
hard time finding a movie that they might approve of, especially for their pre-
teen children, because the kids don't want to go to the PG movies and the 
parents don't want them to go to these other movies.  In the same way, if you 
have too many drugs in the "we don't know" category of your labeling system you 
don't really have very many choices and consumers are stuck and doctors feel 
stuck because they want to be able to provide medication for their patients, and 
patients feel stuck because they want medications that are going to help them 
with the various medical problems they have when they are pregnant and, yet, 
there is not enough in this case research to put enough drugs in other 
categories so that consumers have some kind of choice of being able to choose a 
medication where there is research evidence that shows that it is probably safe 
or looks like it would be safe.  If we have more research, we will have a better 
rating system for drugs.  I wish it were so easy for movies but I actually think 
it is going to be easier for drugs.
	Now, we all understand that they are basic ethical dilemmas of doing 
studies on pregnant women, and I know that Ruth Faden will discuss those 
tomorrow.  Ellen Clayton will be talking about the legal implications.  So, I am 
not going to get into that.  But we already know that women are taking a lot of 
prescription drugs when they are pregnant --more than I expected, I must admit.  
So, if on average women are taking three prescription drugs while pregnant and 
older women are taking five, then it seems to me this does present clear 
opportunities to do research on these women as long as we are just as careful -- 
just as careful that women participating in such research know what the risks 
are, even if that means they are just being told we don't know what the risks 
are -- the risk is that we don't know what the risks are because there is no 
research.
	So, certainly a Phase I kind of study, for example, comparing blood levels 
of pregnant and non-pregnant women who are taking the same dose of the same 
drug, you know, it makes perfect sense to do that kind of research in the 
context of women who would be taking this drug anyway and are told ahead of time 
what is known and not known about the drug they are taking for pregnant women.
	It seems to me that also that at the same time that we are doing those 
kinds of clinical trials, we also need to collect data on the health outcome of 
the baby for that woman.  I mean, I just can't see doing one and not the other.  
It seems to me that both have to be part of the same research.
	We know that drug companies haven't shown much interest in doing this 
research, and you can't really blame them in a way, and it is really, really 
essentially, as Dr. Henney pointed out, that we have total confidence -- total 
confidence in the research.  For that reason, I think it is terribly important 
for the federal government to be doing that research and to be funding the 
research.  As we all know, the research that the FDA bases its decisions on is 
research conducted by the manufacturers.  I think in this particular case we 
really need research conducted by NIH, perhaps AHQR, the Agency for Healthcare 
Quality Research.  Whatever the appropriate HHS agencies are, I think that is 
where we should focus our attention, and I think that means that research should 
be done within the agencies, intramural research when possible, and when 
extramural research is done, if the agencies are funding research elsewhere they 
should use extreme care, looking for conflicts of interest for who is conducting 
the research.  I think it is very, very important that if independent 
researchers at medical centers or universities are doing the research that they 
not be getting funding also from the companies that make those products, even 
for other research that is totally unrelated.  I think that these issues are so 
important to consumers, and are such sensitive issues that we have to use even 
more caution than we normally do.  I, of course, think we should be using more 
caution all the time.
	So, there really needs to be some kind of fire wall or maybe a lock box, I 
am not sure, but there should be something that really protects this research so 
that people feel that the folks conducting the research and reporting on it have 
no conflict of interest at all and are just able to focus on exactly what the 
results are.
	I think everybody in this room knows that some of the greatest disasters 
in women's health have revolved around the impacts of drugs on pregnant women, 
thalidomide and DES being just two examples that come to mind.  So, it really is 
amazing that so many years after those disasters we are still in this situation 
of having so little research and so little focus on protecting consumers so that 
they can make reasonable decisions and have the information they need to make 
those decisions.
	I guess I can't help but say that thalidomide and DES are particularly 
good examples of how the impact, when things go wrong, doesn't just affect the 
baby but the entire family.  So, we really have to think about how can we make 
sure that consumers have all the information they need to keep themselves 
healthy, to keep their babies healthy and to keep their families functioning.
	I really want to thank the FDA and NICHD for holding this meeting.  I 
really am very proud to be a part of it, and I hope and I think that we can all 
can start out by agreeing that we are long past time -- long past time when we 
should have been taking these issues seriously, and that it is really essential 
that we move forward as quickly but as carefully as we can to get this research 
done and to make sure that consumers have information.  I guess I should say 
often that this information absolutely has to be clearly written down, in plain 
English, so that we are absolutely certain that the consumers are getting the 
information and aren't entirely dependent on who their physician or healthcare 
provider happens to be.  Thank you.
	[Applause]
	DR. WOOD:  Thank  you, Diana.  I think all of our presentations this 
morning have really helped us identify the issues, be it the health of the 
woman, the health of the offspring, the scientific and regulatory needs and 
struggles that we face.  So, I thank you, Diana, for bringing us the consumer 
perspective because I think it is a voice that we need to keep in mind 
throughout the next two days.
	We are going to move forward for the rest of the day to looking at some of 
the very specific science and clinical issues that face pregnant women, and the 
drugs that they are taking, so that will be sort of a shift, then we will move 
back tomorrow to looking at some of the policy and legal and ethical issues.
	Some housekeeping for this morning, we are going to take a 15-minute break 
but I do want to note that tomorrow morning we do have a change in schedule 
which will add yet another perspective, and that is we do have a member of 
Congress coming tomorrow morning, Representative Joe Ann Emerson, who is a 
Republican representative from Missouri, the 8th District.  She will be coming 
here tomorrow morning to address the conference from the perspective of 
Congress.  So that will provide us a perspective from the other branch of 
government that clearly influences the type of work that both FDA and NIH do and 
the interest of the women members of Congress.
	I think we are going to take a break for 15 minutes.  So, if we can 
reconvene at about 10:50, we will go on to our final morning presentation.
	[Brief recess]
	DR. WOOD:  After our next presentation we will have a question and answer 
period for all the three presenters.  That is our plan.  We will go ahead and 
get started with this next presentation but then we will open to questions and 
answers and discussion with the audience.
	Our next speaker is Dr. Marilyn Frederiksen.  Dr. Frederiksen is Associate 
Professor of Obstetrics and Gynecology at Northwestern University Medical 
School.  She is board certified in OBGYN, maternal fetal medicine and clinical 
pharmacology.  Her research has been centered on the use of drugs and drug 
therapeutics during pregnancy, and her work on drugs used for pregnancy 
asthmatics led to her being asked to serve on the NHLBI Task Force on Asthma and 
Pregnancy, which led to guidelines for the care and treatment of the pregnant 
patient with asthma.
	She is a fellow of the American College of OBGYN, a member of the Society 
for Maternal Fetal Medicine, and the American Society for Clinical Pharmacology 
and Therapeutics.  She was elected in April of this year to the USP Council of 
Experts serving as the chair of OBGYN.  She will be speaking on physiological 
changes in pregnancy: effect on drug disposition.  Dr. Frederiksen?
Physiological Changes in Pregnancy:
Effect on Drug Disposition
	DR. FREDERIKSEN:  Good morning.
	[Slide]
	We have spent a good deal of time over the last ten years figuring out 
that drug trials or drug studies done in man don't equal or are not necessarily 
pertinent to the study of women and to their use in women.  We now realize 
something that probably some of us have thought about over the last twenty 
years, that is, studies in men certainly don't pertain to drug use during 
pregnancy.
	[Slide]
	As Sandy put up a slide about physiologic changes, I am going to review 
them for you.  We have cardiovascular system changes, including plasma volume 
expansion; increase in cardiac output and changes in regional blood flow during 
pregnancy.  There is a decrease in albumin concentration which will affect 
protein binding of drugs bound to albumin.  There are enzymatic activity 
changes; increase in glomerular filtration rate which could potentially affect 
renal clearance of drugs, as well as gastrointestinal changes during pregnancy, 
which we need to explore and explore their effect on drug use during pregnancy.
	[Slide]
	Cardiovascular system changes -- plasma volume, increase in cardiac output 
and regional blood flows -- just to review this for you, plasma volume expansion 
begins at 6-8 weeks of gestation.  So, what we are going to say is that the 
changes are pertinent to the first trimester.  They begin in the first trimester 
and they usually increase and peak somewhere in the third trimester.  The volume 
increase is about 1400-155 cc and peaks at 32 weeks of gestation, which 
represents about an increase in 1200-1600 cc above the non-pregnant patient.  
With regard to the multiple gestation which has been increasing in our 
population due to artificial reproductive technologies, that plasma volume 
increase is even higher.
	[Slide]
	Cardiac output increases 30-50 percent and can be as high as 50 percent by 
8 weeks of gestation.  Again, the theme is that in the first trimester you see 
the beginning of the changes, and usually those increase during the second 
trimester, peaking in the third trimester.  That cardiac output is accomplished 
through an increase in stroke volume as well as heart rate.  The stroke volume 
happens early in the pregnancy.  Patients complain about perceptions of heart 
beat in the first and early second trimester, and then the heart rate increases 
later in the pregnancy as well.
	[Slide]
	The regional blood flow changes -- there is increased blood flow to the 
uterus which can be about 20-25 percent of cardiac output at term.  There is 
increased renal blood flow which you will see as an increase in glomerular 
filtration rate.  There is increase in skin blood flow because the fetus makes 
heat and the heat needs to be dissipated by the pregnant patient.  So, there is 
increase in blood flow to the skin.  There is also increase in blood flow to the 
mammary glands during pregnancy.  All those increases have to be compensated by 
a decrease in some other blood flow, and that is assumed to be a decrease in 
skeletal muscle blood flow which probably represents why we see muscle cramping 
during pregnancy.  The cardiac output can only sum to 100 percent.
	[Slide]
	One issue for the clinical pharmacologist is whether or not drugs that are 
cleared by blood flow changes through the liver change during pregnancy.  We are 
thankful to our clinical pharmacologists in Great Britain who did a very nice 
study of hepatic blood flow during pregnancy, and let me take you through this 
slide.
	The blue represents the absolute hepatic blood flow during pregnancy as 
measured by indocyanine green, and this did not seem to change and probably is 
the reason why we do not see a clearance change for drugs such as propranolol or 
atenolol during pregnancy because the absolute blood flow does not change.  
However, when you think about blood flow you can also express it as a percentage 
of cardiac output and that, during pregnancy, is actually decreased compared to 
the postpartum state and really represents the change in blood flow, the 
increases that we see to the uterus and the kidneys, and as a percent of 100 
that will then decrease.
	[Slide]
	There are also respiratory changes which occur during pregnancy.  There is 
a compensated respiratory alkalosis, a lowered PaCo2 which ends up to raise our 
pH a little bit, pH 7.44 being normal during pregnancy.  This will change our 
binding parameters for certain drugs which are sensitive to pH changes.
	[Slide]
	We also have a decrease in albumin concentration.  If you measure albumin 
during the second, third and the postpartum state, albumin does fall in 
comparison to the postpartum, however, this probably is not a physiologic 
hypoalbuminemia or dilutional effect because if you look at the total protein 
seen during the second, third and the postpartum state they are equivalent.  So, 
this may be a change in the albumin production or catabolism during pregnancy 
which will affect this lower albumin compared to the postpartum state.
	[Slide]
	We also see enzymatic activity changes during pregnancy, and those are 
largely thought to be related to the pregnancy hormone changes and there has 
been a perception that the hormone that really changes or is responsible for the 
most changes is estrogen because pregnancy is the most estrogenic state that we 
know of.  However, the N-demethylation inhibition has been shown to be done by 
progesterone.  You have to remember that progesterone rises during pregnancy as 
well and can equally affect the activity of the hepatic enzymes.
	[Slide]
	We know that CYP3A4 which causes hydroxylation actually has increased 
activity during pregnancy.  So, not every enzyme in the hepatic armamentarium 
goes down during pregnancy.  Hydroxylation has been shown to be increased.
	[Slide]
	This is a study from our Swedish colleagues, looking at CYP2D6 activity.  
This is a portion of the cychrome p450 system which has been felt to be 
genetically determined so that once you have received your genetic endowment it 
won't change.  What they showed was that there was an increased clearance of 
metoprolol which is metabolized by CYP2D6 during pregnancy.  So, they typed 
those patients with dextromethorphan to see whether or not the actual activity 
of the enzyme did change and showed that for a lot of patients who are either 
homozygous or heterozygous extensive metabolizers, they had a decreased 
dextromethorphan/dextrophan ratio, while our poor metabolizers actually had an 
increase in the dextromethorphan/dextrophan ratio, which means that we have a 
change in an enzymatic activity which we felt previously had been determined by 
your genetic makeup.
	[Slide]
	We also have an increase in GFR or glomerular filtration rate.  Compared 
to the non-pregnant patient, glomerular filtration rises by 15-18 weeks of 
pregnancy, peaks somewhere in the late second trimester, early third trimester, 
and then sort of falls toward term.  That may be study related because of 
changes in either cardiac output or blood flow back to the heart caused by the 
pregnant uterus sitting on the vena cava and the aorta.  A lot of these studies 
actually were done before that change in physiology was accomplished, and they 
were done in a supine position during pregnancy which will affect your cardiac 
output and your blood flow.
	[Slide]
	Gastrointestinal changes are also known to occur.  Progesterone does 
decrease activity of smooth muscle, but there are also decreases in gastric 
acidity, delay in gastric emptying and there is increase in transit time which 
has the potential of changing absorption if there is a window for that 
absorption in the gut.
	[Slide]
	Those are the theoretical or the known physiologic changes that we have, 
and have the potential to affect the pharmacokinetics of drug use during 
pregnancy.  What I would like to do now is to show you some of our best studies 
that exist, looking at these elements.
	One of them is the volume expansion that is caused by the plasma volume 
expansion during pregnancy.  That can affect your volume of distribution.
	[Slide]
	There is a very nice study by Aldrich looking at caffeine pharmacokinetics 
in pregnancy, and this was an oral study and they used saliva as their 
measurement of the caffeine.  They gave a known dose of caffeine to pregnant 
patients and calculated the apparent volume of distribution, which didn't seem 
to change during pregnancy compared to the postpartum period.
	[Slide]
	In the 80's we looked at theophylline which is also a methyl xanthine, 
similar to caffeine.  We did an intravenous study to avoid the problems of oral 
absorption and assumptions that are necessary to calculate volume of 
distribution in that population.  We did pregnant asthmatics and we did it in 
the second trimester, third trimester and the initial postpartum period, six 
weeks, and then we did it in the remote period at about 12 weeks.  What we 
showed was that the volume of distribution on a purely liters basis was elevated 
in the third trimester.  But when you normalize for weight on a liters/kilogram 
basis we showed that the volume didn't seem to change during pregnancy.
	[Slide]
	Protein binding is also something which can be affected both by the pH and 
the concentration of albumin, and that will affect the free fraction of the drug 
and that will be available to interact with receptors.  From our work, we looked 
at theophylline free fraction.  This is the percent of free fraction over our 
four time periods.  In the second and third trimester there was a marked 
elevation in the free fraction of the drug.  So, we had more drug available to 
either be cleared or interact with the receptors.
	[Slide]
	When you do more careful analysis of the changes in protein binding with 
theophylline you can actually determine the binding capacity.  This is a cross-
sectional study, looking at non-pregnant and pregnant patients and the binding 
capacity in a pregnant patient is actually increased.  In other words, there are 
more binding sites for the albumin.  Well, does that make sense?  If you think 
that there is actually a higher free fraction, what is actually going on?
	What is actually going on is that the affinity constant is decreased so 
even though there are more binding sites in pregnancy, they don't seem to hold 
the theophylline as well and the binding affinity constant is less in pregnancy 
so it may interact but it may become free more often.
	[Slide]
	Then there are clearance changes.  To go back to our caffeine 
pharmacokinetics, the caffeine clearance really fell, especially in the third 
trimester, and was a factor of three.  That really accounts for why a pregnant 
patient can drink a cup of coffee and have it last all day.  She doesn't need to 
continue to drink multiple cups of coffee because the clearance of her caffeine 
has dropped by a factor of three during pregnancy.
	[Slide]
	What we did for the theophylline is that we looked at renal clearance as 
well as hepatic clearance.  So, when initially I had hypothesized that perhaps 
it might be cleared through the kidneys I got very, very negative looks from 
colleagues around the room, saying that theophylline is known to be hepatically 
metabolized and there wouldn't be any metabolism through the kidney.
	What we showed was that theophylline renal clearance is markedly increased 
during the second and third trimester or pregnancy compared to remote 
postpartum, and that is an increase in unbound theophylline, increased blood 
flow and, therefore, clearing it through the kidney.
	[Slide]
	Because we looked at renal clearance of this hepatically metabolized drug, 
we could also look at the hepatic clearance, and total hepatic clearance and 
intrinsic hepatic clearance, which really accounts for the changes in protein 
binding, were decreased in pregnancy compared to the remote postpartum period.
	[Slide]
	We couldn't show that overall clearance changed but we did show that the 
half-life was prolonged in the third trimester of pregnancy.
	[Slide]
	This is a study which Mary Jean Creek published.  The first author is 
Susan Pond.  It is looking at methadone clearance.  Really, the reason the study 
was done was because patients required methadone and we really didn't know how 
to use methadone in pregnancy.  They showed that the clearance in the second and 
third trimester was markedly elevated, increased compared to the postpartum 
period and, therefore, patients, if they had increased clearance, would actually 
require more methadone to keep their fetus and themselves from not withdrawing 
during pregnancy.
	[Slide]
	This is a study looking at total -- this isn't clearance but actually 
plasma concentrations.  This is carbamazepine used in our epileptic population, 
and we see the fall in plasma clearance.  But if you look at these little blue 
bars, down here, that is the free fraction and that is the drug which will then 
interact with the receptors and that doesn't change during pregnancy, and can 
account for the fact that you can get no change in seizure control despite the 
fact that the absolute plasma concentration or the total plasma concentration 
has fallen during pregnancy.
	[Slide]
	Similarly, we know that the clearance of phenytoin changes, and this is 
shown by the fact that on a total plasma concentration phenytoin falls during 
pregnancy.  Again, the free fraction is the same.  These are first studies to 
look at real changes in clearance, but also related to the changes in the plasma 
protein binding of the particular drug during pregnancy.
	[Slide]
	This study was looking at comparison of metabolites, looking at the 
changes in the way the hepatic enzymes or the cytochrome p450 enzymes change 
during pregnancy.  This represents CYP1A2 which is decreased during pregnancy.  
The yellow bars are during pregnancy and the blue are comparisons to the 
postpartum state.  Xanthine oxidase is decreased N-acetyl transferase is 
decreased but hydroxylation is actually increased, and the relative nature of 
all these is masked because it is on one slide with one You axis.
	[Slide]
	We are aware that we have clearance changes and changes in the peak 
concentrations of drugs.  One area which is very important to the practicing 
obstetrician is the use of antibiotics.  Tobramycin is a drug which we don't use 
very often -- we usually use gentamicin -- but it has been shown to have a 
higher clearance in the mid-trimester and a shorter half-life.  The clearance is 
lower in the third trimester with a longer half-life.  The volume on a per 
kilogram basis shows really no changes.  And, this supports really single dose 
tobramycin use during pregnancy but that specific use for pregnancy hasn't 
really been looked at.
	[Slide]
	Another drug that we use, enoxaparin, is a low molecular weight heparin, 
which may have less osteoporosis changes during pregnancy.  It has been shown 
that there is no change in the Tmax.  The Cmax is lower during pregnancy.  In 
other words, the concentration at the maximal point is lower during pregnancy.  
Clearance decreases in late pregnancy, and there is lower anti-factor Xa 
activity and area under the curve has been shown to be lower during pregnancy -- 
things that we need to know to treat patients with this drug effectively which 
are not in labeling.
	[Slide]
	Lastly, the gastrointestinal changes have actually been looked at.  This 
is really the only study looking at oral bioavailability during pregnancy.  It 
is a 1977 article, and if you look at area under the curve, we have a lower area 
under the curve during pregnancy as compared to the non-pregnant patient, a 
lower peak level as compared to the non-pregnant patient, but that 
bioavailability or the effect of the gut during pregnancy has been shown to be 
no different.  So, this is the only drug where we know that the amount of drug 
that is absorbed during pregnancy is about equivalent to the non-pregnant state, 
and there aren't any other studies in the literature.
	[Slide]
	This is a more recent study looking at the conversion by the gut from an 
inactive drug to an active drug.  Valacyclovir is the pro-drug and acyclovir is 
the drug.  The pro-drug is converted by first-pass metabolism, and this group 
looked at the drug levels of acyclovir and valacyclovir in pregnant and non-
pregnant patients.  In the non-pregnant patient valacyclovir levels are about 3-
5 times higher.  That is the normal state.  They give a higher level than the 
rather poorly absorbed acyclovir.  Valacyclovir in pregnancy is only about 3 
times higher than acyclovir.  The gut works but probably the lower level 
reflects the change in plasma concentration of the drug or the plasma volume.
	[Slide]
	Those are examples of the pregnancy physiology that are in the literature, 
and I didn't go into an exhaustive survey but I tried to pick and choose studies 
that really showed the examples and the best studies that are in the literature.
	[Slide]
	You might want to ask yourself can't we do this intrapartum or peripartum 
to decrease exposure to the patient and to the fetus.  Well, in the peripartum 
period we have increased cardiac output.  We have blood flow changes beyond what 
we have seen in during pregnancy.  We have the onset of uterine contractions 
which will change blood distribution and drug distribution to the fetus, and 
there can be a question of whether or not there is a pharmacodynamic change, or 
the effects of the drug may be slightly different or altered during this period 
of time.
	[Slide]
	This is a study looking at cephalosporin during pregnancy, and what these 
investigators did was to study those patients in the mid-trimester of pregnancy, 
at delivery, and then the remote postpartum period.  What they showed was that 
with this steady state pregnant patient, not under stress for the portion of 
delivery, clearance was actually higher than what it was determined to be at the 
time of delivery.  The half-life was different correspondingly.  This shows us 
that studying drugs at the time of delivery may give us data which is not 
applicable to the period of time which is the greater period of time, during 
gestation, during those nine months prior to the onset of labor and delivery.
	[Slide]
	This is a study of morphine pharmacokinetics during labor, showing much 
elevated increase in clearance compared to the postpartum state.  This is higher 
than what we would expect even for the methadone studies.  There seems to be a 
change, again, in pharmacokinetics during the peripartum state.
	[Slide]
	In the postpartum state, let's say could we study patients during that 
period of time, after delivery has actually occurred because we do have an 
elevated cardiac output which is maintained for many weeks afterwards.  We have 
glomerular filtration rate which is sustained.  We do encounter a diuresis but 
maybe 3-5 days after delivery we may be able to capture patients during this 
period of time.  We do have the onset of breastfeeding, and what I am going to 
show you is some great variability in this postpartum period, which is sort of 
disappointing to anyone who wants to study patients during this period of time.  
But it also may be an area of opportunity for us to sort of look at the 
peripartum period or the postpartum period with a different perspective which 
may show us why there is such great danger to the woman during this period of 
time.
	[Slide]
	Clindamycin is the drug which we commonly use in the postpartum state.  It 
is used for postpartum endometritis.  This is a study which was done looking at 
the oral use of clindamycin.  What they showed was that the peak level varies.  
This is a six-hour period of time, and the peaks are all different so that there 
is great variability in when that peak concentration can be effectively seen.
	[Slide]
	This is a study from our group, looking at postpartum gentamicin and the 
distribution volumes that we saw.  One caveat about this study is that we looked 
at postpartum endometritis patients and endometritis may have happened in the 
first seven days and it may have been a patient who was readmitted to the 
hospital, and we combined that data and that may be a problem.
	We showed over 0.1 to 0.5 difference in plasma volume and distribution 
frequency data.  In other words, the average plasma distribution volume actually 
increased.  If you look at the non-pregnant or the commonly used patient that 
uses gentamicin, the variability is between 0.2 and 0.3.  So, we had greater 
variability in what we showed here.
	[Slide]
	When you are thinking about doing drug work in pregnancy, you also have 
ethical and IRB concerns.  The pregnant patient is really not a normal 
volunteer.  We don't have pregnant patients presenting themselves for drug 
trials outside our clinical research organization.  They are not available in a 
clinical setting.  So, oftentimes this work has to be done in multiple centers 
to get data that is applicable.  The IRB wants you to have used this drug or 
make sure that you are using the drug for a valid clinical indication.  So, you 
are going to be studying patients who have hypertension, epilepsy, endometritis, 
urinary track infections.  You do need to have serial studies.  They  need to 
span pregnancy.  As I have shown you, what is applicable at the time of delivery 
is not applicable during pregnancy.  If you are going to use a drug for labor 
and delivery, then that is probably the only place that you need to study that 
drug.  And, you do need to use a valid control population.  It is nice to have 
the patients act as their own controls and not to use historic controls or male 
controls.  You are going to get a lot of erroneous answers or erroneous 
interpretation of that data if you use those control populations.
	[Slide]
	Your study design should use population pharmacokinetic analysis to 
ascertain whether there is a problem or whether you should be studying that drug 
in pregnancy.  You should be using tracer substances as  reference markers so 
that you can differentiate drug clearance changes or changes due to drug effect 
from the underlying physiological changes of pregnancy.  You should incorporate 
in vitro protein binding studies, and use of stable isotopes for bioavailability 
will decrease the number of studies.
	These patients are connected to family.  They are the caregivers for young 
children.  You have to sometimes have a GCRC where you can provide child care or 
have a bassinet for a postpartum patient to bring her child in.  So, those 
allowances are very important.
	[Slide]
	What drugs should we study?  Well, if we are using a drug specific to 
pregnancy, such as a tocolytic agent or a drug for that mythical eclampsia 
agent, that is going to be studied during pregnancy.  But probably more relevant 
and more relevant to what we are talking about today are drugs commonly used by 
women of childbearing potential.  Antidepressants and asthma drugs and 
antihypertensive drugs are drugs that need to be studied during pregnancy so 
that we do have dosing regimens as well as safety elements.
	And, that is the end of my talk.
	[Applause]
	DR. WOOD:  Thank you so much.  I think, again, we are having a wonderful 
presentation to lay out some of the physiologic issues, and then translating 
that into what that means in the research questions that need to be considered 
during design and carrying out of studies that would give us answers that would 
be both valid and useful.
	I think we want to open it up now to some discussion with the audience.  I 
am noticing that there is no microphone out there.  Going back to both Sandra 
Kweder's presentation as well as Diana Zuckerman's, I know there are a lot of 
issues around FDA's policy as well as some of the consumer perspectives.  Yes?
	DR. SIBAI:  Baha Sibai, from Cincinnati.  What I have noticed this morning 
is that there has been a lot of emphasis on the risks of the drugs used in 
pregnancy.  I think this has been my biggest concern, that the risks that have 
been overemphasized have been an over-kill.  There is a wealth of information to 
inform the physician as well as the patient on what the risks from not taking 
the medications.
	To give you an example, in one year we had four women die, four women and 
four infants died together because the physician was reluctant to use the 
medication because of the overemphasis, which I call over-kill, by the FDA and 
by the labeling process.  We really need to put an addition the scare tactic of 
what is the absolute risk which, again, has been very, very lacking in much of 
this.  A risk of 0.6 percent is really nothing if a physician is going to 
withhold a drug where a woman could die.  To give you an example, I had two 
women die from thrombosis of their prostatic valve.  They were referred to us 
because the physicians were reluctant to use coumadin, for example, at 16 or 18 
weeks because you have overemphasized for them the effect of coumadin.  Coumadin 
effects are restricted up to 9 weeks.  Beyond that the risk is very minimal, 
which has been de-emphasized with this labeling.  The same thing with anti-
epileptic drugs.  Two women died because of status epilepticus because their 
physicians stopped their medications.  One was stopped at 9 weeks, one at 13 
weeks.  This I think is what we really need to emphasize more and more.
	DR. KWEDER:  I want to thank Dr. Sibai for raising those points.  From my 
standpoint in representing FDA, I wholeheartedly agree with you.  I think one of 
the comments that I made during my talk was that historically, particularly 
since thalidomide, we have perceived the warning language as most optimal and in 
many cases it is highly appropriate.  But we need to take a step back and look 
at margins of risk and what are the components of risk.  In many situations the 
risks to the mother, the known risks to the mother and baby of not taking a 
medicine far outweigh any potential or theoretical risks that may be posed by a 
product.
	In many situations we don't know how to balance those, but we need to be 
able to provide the information in a way that helps clinicians think about both 
sides of the equation because both are important.
	DR. FREDERIKSEN:  Can I say something?  When we did the asthma and 
pregnancy guidelines, and you are going to hear more about that from Dr. Schatz, 
in the back of the room, this afternoon, but one of the things that we had to 
balance was uncontrolled disease against drug risk.  I think that is a concept 
which the medical community has not been communicative about at all.  We haven't 
used it.  We have not talked to patients about uncontrolled disease versus the 
drug risk.  I think we have to reeducate and rethink how we use drugs in 
pregnancy, and that is the risk/benefit ratio that we have to weigh.  It is not 
always the teratogenic risk that we are talking about.  That can be very small 
against the huge risk of uncontrolled disease.  So, when we talk about 
risk/benefit analysis, that is the one that we have to remember to talk about.
	DR. ZUCKERMAN:  I just wanted to say that I agree, of course, that 
consumers need to know all the information.  My assumption is that at least the 
research has already been done on what the risks are of not taking medication 
when a woman is sick.  I suppose it might be a little different when she is 
pregnant, what the risks are to her and it would certainly be very important to 
get that information too.
	So, I really agree with almost everything you said, and the only thing I 
would just caution about is I think we shouldn't be talking about informed 
consent as scare tactics.  I think informed consent is providing information and 
we should provide all information.  Sometimes the information is scary on both 
sides.  There are risks if you don't take it; there are risks if you do.  But we 
provide that as best we can, as clearly as we can in writing, in lettering that 
is big enough for people to read and in a format that is user friendly, and we 
provide it to the physicians and to the patients, and then they can discuss what 
the balance is of the risks and benefits.
	DR. KWEDER:  Yes, I want to pick up on that point because I think that one 
of the things that we have to acknowledge is that sometimes patients make 
decisions that healthcare providers don't necessarily agree with.  They are not 
the decisions that we would make, and they may be more worried about a potential 
risk of a medicine than a known risk to themselves and to the baby.  But that is 
their decision but our job is to provide the information so that they and their 
healthcare providers can make those decisions together and in an informed way.
	DR. WOOD:  I would like to add on to that one aspect, which is it is that 
unknown part.  If you feel like you know the risks of something you are less 
afraid of it to some degree.  This unknown stuff, which is the lack of data on 
either how effective a drug is going to be, what dose we should give you and 
maybe we need to give it to you higher but we would really prefer to give it to 
you lower, and we are going to try this out on you and we are really not sure -- 
to many women, I think that is a scarier thing than just taking nothing at all.  
Or, if you had information that said this is a risk of 1/1000 or 1/10,000 and, 
therefore, we can compare it to your risk of going off this medication, but if 
the risk is unknown, well, that could mean it is 100 percent, and that is where 
the lack of information is a disservice, I think, to many women.
	MS. HARE:  Hello.  Doris Hare, American Foundation for Maternal and Child 
Health.  I was one of those women forty years ago who was asked whether I should 
take a drug.  I was a severe asthmatic.  I was told by my woman obstetrician 
that they didn't know the long-term effects of the drug on human development, 
nor did they know the long-term effects of my asthma.  I chose to take the drug.  
My child died.  I never felt bitterness.  You can live with heartache but to 
live with bitterness and deception is a terrible thing.  I wholly and heartily 
recommend the concept of informed consent.
	MS. TRAPNELL:  I am Carol Trapnell.  I just have a question, mostly food 
for thought.  You showed a lot of pharmacokinetic data from studies that are 
published, and I always am left, as a clinical pharmacologist, wondering, well, 
what do I do with that information?  I am wondering if you have any comments 
about how people can take literature studies where the pharmacokinetics are 
evaluated but the pharmacodynamics are really not because you are left 
wondering, well, if the clearance is changed by 30 percent, what does that 
really mean and how do we translate those kinds of data into clinically 
meaningful comments and changes for the patients?
	DR. FREDERIKSEN:  Well, your comment is right on.  It is not only PK but 
it is also PD, and there is even less data with respect to pharmacodynamics.  
But let me give you an example.  The ampicillin data was published in 1977, 
which means we really need to use more ampicillin, not less ampicillin, during 
pregnancy.  If anyone follows the emergence of resistant organisms that has 
occurred, they know that the bugs that we are trying to kill during pregnancy -- 
the resistance has gone up.  In fact, we have had an increasing number of 
pregnant patients having urinary track infections in the mid-trimester where the 
only drug that we can use is gentamicin, which is a systemically available 
antibiotic, because all of the other medications will not cure the urinary track 
infection.
	You have to think in terms that if we are going to treat urinary track 
infections and the resistant organisms and not see increases in resistance, we 
are going to have to use better dosing.  That is one example.  I am seeing more 
patients with these problems.
	DR. KWEDER:  Can you clarify that?  Is your point that perhaps by treating 
these women with recurrent urinary track infections with sub-optimal --
	DR. FREDERIKSEN:  Sub-optimal dosing.
	DR. KWEDER:  -- what you are doing is you are increasing the prevalence of 
resistance, which is a known phenomenon in infections.
	DR. YAFFEE:  Sumner Yaffee, NIHCH.  A comment and question for Dr. Kweder 
and then a comment for Dr. Henney, but she is not here to answer that.  Talking 
about the predictability of lack of predictability with animal studies for risk 
assessment, which Sandy mentioned -- that there is no predictability, and my 
comment is that we will never have predictability until we understand the 
molecular mechanisms underlying birth defects, which is what you are talking 
about, structural birth defects.  So, at the NIH we have embarked on a large 
initiative dealing with the molecular mechanisms of birth defects and, 
hopefully, Congress will appropriate the money.  They have just come to town 
this morning, and if they can get their act together we will have a budget to 
deal with the problem.
	So, the question, Sandy, is why does the FDA continue to require two 
species of animal studies with respect to a new drug?
	DR. KWEDER:  I will do my best.  I am not an expert in animal reproductive 
toxicology by any stretch, but historically it has been acknowledged that most 
products, when they reach the market, will be used, either intentionally or 
there will be inadvertent exposure in pregnancy.  People who are experts in the 
field tell us that there is some predictive value.  You get some general sense 
of what you might see in humans.  What my colleagues tell me is that 
increasingly we understand more and better about how to use these studies.
	The area where I think we have a lot of work to do is trying to make that 
kind of information more readily accessible to the clinical community.  There 
are not a lot of people in the scientific community, at least in my experience, 
who are familiar with and well versed in both the animal data and analysis side 
and the clinical side, and there is a huge divide there.  It is important for us 
to try and bridge that.
	This isn't only an issue in reproductive toxicology.  It is the case in 
any area of science where we use non-human data to try and anticipate risks to 
people.  Another good example is with carcinogenicity.  We have some of the same 
challenges.  But I think that there is increasing dialogue between those 
communities in both the area of carcinogenicity and reproductive toxicology, and 
we need to find ways to better use the information that we have.
	DR. YAFFEE:  I agree with what you said, but I also agree that there is a 
lack of predictability and the proponents of the use of animal data are the 
people investigating these chemicals in animals.  So, they have a certain self-
interest.  I won't go on; you can see where I am coming from.
	DR. KWEDER:  Yes, I can.
	DR. YAFFEE:  But you still require for an NDA, for a new chemical entity 
you have to show -- the regulations require it --
	DR. KWEDER:  That is right.
	DR. YAFFEE:  Why don't you change it?
	DR. KWEDER:  I will say, Sumner -- let me see if I can summarize this, I 
think that they are probably very useful for us in looking for big signals, for 
big signals, for problems that are potentially enormous.  They may have less 
value in cases that are more subtle.  So, we can find things, for example like 
Accutane --
	DR. YAFFEE:  That is a big one.
	DR. KWEDER:  Right.  How well they help us with lesser findings I think is 
where the scientific struggle and discourse has to continue.
	DR. YAFFEE:  So, basic science will help you in the end.
	DR. KWEDER:  Absolutely.
	DR. YAFFEE:  Then just adding to Dr. Henney's comments, in terms of the 
epidemiology of drug intake during pregnancy, two weeks ago the French published 
a study in Lancet, which Dr. Henney knows about, which demonstrated that the 
problem is still in existence, at least in France, where perhaps drugs, 
chemicals, wine is taken in abundance, and it showed that in 1000 women the mean 
average number of drugs taken during pregnancy is 13.6.  So, they characterized 
not just the first trimester but all of pregnancy.  Secondly, 10 women out of 
1000 had no drugs.  That is extraordinary.  Lastly, a number of the drugs taken 
were classified as X.  They use the FDA system.  And, there were a large number 
of herbal medications.
	DR. KWEDER:  Right.  There was a large number of herbal medications and 
over-the-counter products, and that was with no direct-to-consumer advertising.  
They don't have that in France.
	MR. RAAFAT:  Bishai Raafat, Philadelphia.  I have a question regarding 
labeling.  If we are changing now the actual classification from A, B, C to a 
more descriptive kind representing the studies, my concern is that it could be 
more confusing for the physician and the perception of patients.  In a sense, it 
will differ from one person to another if we are just describing a retrospective 
study of 35 patients.  Somebody will consider it as not sufficient; somebody 
will consider it as, yes, it is a good study and I will consider this drug as a 
safe one.
	So, my concern is that we are just putting the liability and the 
responsibility on the shoulders of the physician, and he will take the decision 
more than the regulatory people.  So, if we can have like a formula combining 
both, the actual specification as A, B, C, and the description of the study 
available.
	DR. FREDERIKSEN:  But the legal situation from the standpoint of the 
physician stands right on the shoulders because most of the drugs carry the 
phrase that you have to weigh the risks against the benefits --
	DR. RAAFAT:  Which is more confusing.
	DR. FREDERIKSEN:  Well, that is where we are right now.  So, nothing is 
changing.  You are going to have to make that decision when you use a drug in 
pregnancy.  You have to put into the equation the fact that we are not 
necessarily training physicians to look at this data appropriately.  The amount 
of time that the average person spends looking at toxicology data or any 
background for toxicology data from a medical school perspective, from a 
residency perspective or fellowship turns out to be one or two hours of formal 
education in the whole curriculum of medicine, and we are asking them to make a 
decision on a split time basis, in other words, minute per minute, is that drug 
safe or is that drug not safe?  We are not even telling them now to use that 
drug in pregnancy.  So, it is a step forward.  The toxicology data may not even 
apply in the third trimester to the first trimester, and you are using one set 
of data for the whole span of pregnancy.  And, the difference between one drug 
and another drug, C versus D, may be inconsequential.
	DR. KWEDER:  I don't think our purpose today is to go into detail about 
any new labeling proposal.  I just wanted to share with you some highlights of 
it.  I would say that you can be assured that there will be a way to have some 
standardized risk assessment statements but still allow for more and better 
information to be brought to the discussion so that the risk assessment is the 
risk assessment but there are many other factors that have to be considered in 
clinical decision-making and our goal is to try and highlight what some of those 
others may be.
	DR. SIBAI:  Can I have another question?  Regarding the availability of 
all kinds of information to the patients, sometimes they are over-reacting by 
going to termination of pregnancy.  So, sometimes it should be under the 
supervision of a physician, and I think this is very important.  I have been in 
a teratogen center and I face patients and their perceptions differ from one to 
the other.  So, giving the information under the supervision of a physician I 
think is very important.
	DR. ZUCKERMAN:  I certainly agree, and this is a hard one.  I mean, what 
concerns me is that I think too many physicians don't give information because 
they think the patient will overreact and this is a very tricky issue because, 
of course, if a patient doesn't understand research or doesn't understand all 
the information that the physician is trying to give it will be very hard for 
her or him to make a reasonable decision.  On the other hand, I do think it has 
to be the consumer's decision and the patient has to do the best that, in this 
case, she can to understand and the physician and, I hope, the FDA or NICHD or 
whoever has to help provide that information in a way that is understandable and 
meaningful.  Ultimately, I don't think we protect patients by not giving them 
information even though there is that risk that they will overreact.  I think 
that has to be the patient's decision.
	DR. SIBAI:  I guess what I just heard from you is what nobody here really 
wants to hear, neither the patient nor the physician.  You are saying the 
patient is responsible.  Responsible for what?  You only told her what the risk 
is from this medication.  There is nowhere she can go, on the Internet or 
anywhere, where it tells her what the risk is for not taking this medication.  
Physicians don't have this information.  They cannot give her meaningful 
information.  So, we have to be very, very careful.  The FDA should make every 
effort to put this information for the consumer as well as for the physician in 
the labeling.
	DR. ZUCKERMAN:  Well, I don't disagree with you, and I guess my assumption 
is that a medical product would not have been approved if there wasn't evidence 
that it worked and that, I would have thought, would have been the data 
available to show that this is a medication -- silly me!  At any rate, I guess 
the point is the information needs to be on both sides.  We need the data on the 
effect of not taking the drug, as well as whatever the risks might be.  Both of 
those things are crucially important.  Patients should not be getting this -- of 
course, they will be getting it from the Internet but what I am saying is we 
need to make sure there are other ways that they can get this information, 
preferably from a government agency, and that is part of the job of informing 
consumers not just about what the risks are but also of the benefits.  Both 
things are absolutely crucial.
	DR. SIBAI:  This is why we are here.
	MS. ANDERSON:  I am Ann Anderson, from NIDA.  I am a project officer for a 
study of bupernorphine in pregnancy.  I want to change the focus a little bit.  
My colleague, who has a longer history of this conferences, says that this is 
the same as the conference he went to in '97 -- the same discussion, the same 
problems, no new data.  So, tomorrow afternoon we are going to talk about 
funding and we are going to try to figure out ways to fund the studies that we 
need that will additional the new information, the new data, that will help us 
make decisions.  Funding is a creative enterprise that involves taking a little 
bit from this pot, a little bit from that pot, putting it together and mixing it 
up.  That is what we need to do.  That is my plea.
	DR. CLARK:  I am Richard Clark with Glaxo Wellcome, in Research Triangle.  
I am in the clinical pharmacology division, after being an endocrinologist in 
academic medicine.  I have a question for the panel and a comment.
	One is, in looking at the spectrum of pregnancy, as Dr. Frederiksen 
showed, each trimester is characterized by different changes and then the 
postpartum period.  So, as a clinical pharmacologist looking at areas to study, 
can you give us some guidelines as to where an ideal profiling of changes and 
drug metabolism and drug effect during pregnancy would be?
	Then my second is a comment to the panel that I am not sure that an 
earlier comment made that drug companies should not be involved in doing these 
studies is reasonable.  I think drug companies can do these studies.  We are 
very interested in treating patients carefully and well, and we are very 
interested in doing ethical studies.
	DR. FREDERIKSEN:  With respect to how to do studies in pregnancy, I would 
advocate a mid-trimester study, a third trimester study, initial postpartum as 
well as remote postpartum.  That would give you the maximal amount of data to 
dose a patient and effectively treat her throughout pregnancy and the postpartum 
period.
	DR. CLARK:  Thank you.
	DR. ZUCKERMAN:  I can't resist.  I guess I think the point is that for the 
most part the drug companies have not done much of this research so far and, 
therefore, since we have been waiting for this research and we have not gotten 
this research, I think it is time for the federal government to step in.  
Certainly, if the drug companies want to also do research, that is great.
	DR. WEISS:  Robert Weiss.  I am from Centicore, and I am gratified to hear 
that there are still a lot of issues with the development of small molecules in 
pregnancy.  We have a different set of problems in developing biologics, and one 
of the things that I would like to especially get out of this particular 
conference is we have an interesting set of problems in that many biologics are 
species specific.  So, for me to do preliminary animal studies and hope to come 
out of those animal studies with findings that may be necessarily relevant or 
faithful to what happens in human beings is a real issue.  So, I would like to 
ask both this panel and anybody else who may be speaking if they can make any 
remarks that are pertinent to the development of biologics.  In particular, I 
have an interest in that area, as I am sure do many people.
	DR. KWEDER:  I can comment a little bit -- not much but a little bit.  As 
we have tackled some of these issues at the agency -- I am from the Center for 
Drugs, but we work closely with our colleague from the Center for Biologics 
every step of the way -- we have acknowledged that many of the potential tools 
that we have for drugs we don't have for biologics because they just don't 
apply.  The data are much more difficult to interpret yet the clinical 
conundrums are the same.  So, the collection of human data, whether it is on 
pregnancy outcomes for the broader safety picture or, you know, inducing immune 
responses in vaccines, become all the more important.  We are trying to work 
together to figure out just how to deal with those issues.  But the bottom line 
is that the clinical issues and the decision-making are the same.  So, we just 
have to find different tools and ways to get the information that is needed.
	DR. MUELBROCK:  Eileen Muelbrock, Eli Lilly.  I have a question, as we are 
also desperately trying to get more and more information on pregnancy.  What is 
your feedback been from the healthcare profession, from OBGYN, and what they are 
willing to do to provide us information?  Because, obviously, a big battle we 
have is giving out information out of controlled clinical trials.
	DR. KWEDER:  You know, you are talking about the broader safety picture of 
outcomes.  That is not really the specific topic of this meeting, which is why I 
mentioned a meeting that was held last week that the CDC really took a lead on.  
There are different kinds of data collection than, say, clinical pharmacology 
studies which tend to be smaller, but safety data in those is important because 
you have the opportunity in a clinical pharmacology study to collect it in a 
more focused and intensive way.
	In my talk I mentioned that, you know, one of the different models we have 
had to date for collecting safety information in the broad picture have been 
postmarketing surveillance, the spontaneous reporting system which we will 
always have, and there have been companies that have tried to make intensive 
efforts to collect data prospectively where there have been exposures in 
pregnant women to see what happens to the babies.  There are good examples of 
successful efforts to do that and unsuccessful efforts to do that.  There are 
examples of quality data that have come from those sorts of efforts and poor 
quality data.  But historically they have relied very heavily on the practicing 
clinician to provide the information, and there has been a lot of discussion 
about whether or not that is the most efficient and effective way to collect 
that kind of data, and what are the other models that might do it differently.  
So, it is a much bigger topic but I just want to be clear that FDA, NIH and the 
CDC recognize that there are challenges with the models we have now and we need 
to bring some of those independent efforts together to try and resolve some of 
these tough issues.
	DR. SIBAI:  I want to respond to this for a minute, particularly for the 
Glaxo representative.  You know, being a clinician and somebody who did 
randomized trials, it is very disappointing and frustrating to me to contact a 
pharmaceutical company and get any response from them when it comes to doing 
studies in pregnancy.  The Glaxo company is an example actually.  I am going to 
show you data.  I am glad this was mentioned.  For the labetalol study I 
actually called them.  I wrote to them, telling them we really need to do 
studies with this drug in pregnancy.  It is very important.  All I got from them 
was nasty letters from their lawyers.  They wouldn't even let you talk to them.  
I had this example again when I tried to do this for nephedapine.  The 
interesting thing, after I finished my trial, was I supported myself from my own 
money.  I went ahead and did the trial, irrespective of that.  Then about two 
years later I received a letter from them.  They wanted all of this information 
as post-surveillance about the use of labetalol in pregnancy.  And, I told them 
go to a very, very hot place.
	[Laughter]
	They waited for me to spend all my money and resources to come to get the 
information.  This is why I really still don't believe pharmaceutical companies, 
at least for drugs early in pregnancy, have the interest in them because of the 
medical-legal implications.
	DR. YAFFEE:  I just wanted to additional to what Dr. Sibai said about 
industry's willingness to support, and we have had our own experiences first 
with Prozac which, as you know, is taken by let me say 25 percent, 28 percent -- 
it is given to 28 percent; it is not taken.  It has to be prescribed.  After the 
drug was proposed by us for study a protocol was developed, supported by Lilly, 
by the medical staff of Lilly, and we didn't ask for money -- unusual!  But the 
NIH was going to pay for this study to look at the safety and efficacy.  Then, 
when it hit the corporate level of Lilly, particularly the legal department, 
they said absolutely no.
	I will tell you another story.  I won't go into detail.  The same thing 
happened with Pfizer with an antihypertension drug.  So, we had the cooperation 
of the medical staff of the company.  As was mentioned, they were willing to do 
it.  I can tell you, you won't get any place.
	DR. FREDERIKSEN:  One thing that you brought up from the initial question 
was what is the role of the practicing obstetrician.  I think the practicing 
obstetrician is over-burdened with clinical responsibilities, and I think that 
if you think about reporting a particular case you are much, much more apt to 
report a patient who has taken the drug and had an adverse outcome than you are 
to report a patient who has taken the drug and had a good outcome.  So, it takes 
a clinical epidemiologist to look through this data to actually see whether or 
not what you are reporting is not the background risk that we all have of having 
an abnormal child, and that is 2 to 3 per 100, rather than a real effect of the 
drug.
	Our literature also has problems.  There are case reports of theophylline 
use in the first trimester causing spontaneous miscarriages.  I think the 
majority of us are going to say that the majority of miscarriages are caused by 
genetic abnormalities and not having anything to do with drugs.  But there is a 
case report in the recent literature which is disturbing because somebody from a 
peer review organization didn't do their job there.  But the information is 
difficult to get and it is difficult sometimes to interpret.  But the practicing 
obstetrician cannot always be looked at as someone who is not cooperative; we 
are just essentially overburdened.
	DR. WOOD:  Just to comment on that, the area that Dr. Kweder has brought 
up a couple of times of understanding when something is safe, and if you had 
data done in a rigorous fashion that showed that there was no increased risk, 
that A category, if you will, for which we know that the majority of drugs are 
not going to cause a problem if we only had the data to show it -- that 
information would be so useful, not dependent upon independent physician 
reporting "oh, I had no problem," but to be able to make that comparison between 
the actual background risk of a problem with something that is additive due to a 
particular drug or other ingredient.
	MS. HARE:  Dr. Frederiksen, I greatly enjoyed your presentation.  You 
mentioned that research done during pregnancy is not necessarily applicable to 
research findings found in the interpartum period.  Does the effect of the 
fetus' condition, such as hypoxia, affect the transfer of drug to the fetus?  
And, who is studying that as a part of the equation?
	DR. FREDERIKSEN:  Well, number one, the majority of fetuses are not 
hypoxic.  Okay?  The majority of fetuses have totally normal oxygenation, and 
they do during labor and delivery as well.  That is really a different topic 
entirely and it is not really pertinent to the use of drugs with respect to 
pregnancy.
	When we were talking about labor and delivery and contractions, what I was 
trying to get across was the physiologic changes which will make that period 
unique to pregnancy, and if you are marketing a drug for the use during labor 
and delivery that needs to be studied during that period of time and not prior 
or after.
	DR. JOHNSON:  Rolley Johnson, from Johns Hopkins University School of 
Medicine.  I would like to commend you for organizing this meeting here today.  
I wanted to give a little history.  I am the PI for the trial that Dr. Anderson 
was talking about.  So, we are actually doing a randomized, controlled clinical 
trial in opioid-dependent pregnant women.
	I have to start out by telling everyone a joke that a colle