THURSDAY, December 8, 2005

Session 4: Ethical Issues in Pediatric Research

Sara Goldkind, M.D., M.A., Bioethicist, Office of Pediatric Therapeutics, Office of the Commissioner, Food and Drug Administration

DR. GOLDKIND:  I would like to shift the conversation from the clinical arena to the area of pediatric research and particularly to regulated research.

What I would like to do is first start with where we have been, what the evolution is in pediatric clinical trials.  And I'm going to look at societal, legislative, and regulatory history to do so.

Then I would like to talk about where we are now and the importance and findings of pediatric research, what I call a shift in paradigm from the view that research was risk-laden to the notion that research is actually hope-laden and look at medical and societal influences on that view and discuss two specific topics which are very challenging issues, some of which we have heard a little bit about from Norm on risk levels.

And in order to do so, I'm going to look specifically at the FDA regulations on subpart D, which are the additional safeguards for children in clinical investigations.

I would also like to touch upon the issue of inclusion of healthy children that has already been raised this morning and look at some ethical documents and subpart D referrals that we have actually had to the agency to see if we could flesh out a little bit more information on the inclusion of healthy children.

To begin so, I thought that I would begin with actually the birth of the FDA and take us through about a 100-year history of where we were and where we are now.  It's interesting that while the FDA was formed in 1906 in reaction to muckraking journalists to consumer activists to professional societies to documents such as Upton Sinclair's novel The Jungle, which exposed cure-all claims and misbranded products with the idea that we should try and keep products away from consumers who are not wary of what they're purchasing and also that there ought to be even regulations among products that are transmitted via physician prescription.

Prior to the 1906 regulation, there was a 1902 Biologics Act, which was stimulated because of children who died after receiving tetanus-laced diphtheria antitoxin.  In the next couple of slides, I'm going to mention some of the untoward and catastrophic effects that have occurred in children which influenced some of the fundamental FDA regulations that are still in effect today.

So two landmark dates are 1938, when safety requirements for drugs were initiated and legislated, and 1962, pre-market efficacy requirements for drugs were legislated.  Both legislative initiatives were stimulated, in part, by catastrophic events in children.   Yet, even at the time, the ethical approach to children was thought to be exclusion from research.

The 1938 Act, as I said, was passed, in part, in response to the sulfanilamide disaster, where an elixir of sulfanilamide was prepared with a poisonous solvent, which was actually antifreeze.  And 100 out of 107 of the people who were killed were children.  This led to safety requirements that are still in effect today.

And the sponsors specifically must show before marketing that a product is safe for the use under the conditions of prescribed, recommended, or suggested in the proposed labelling.

I would like to point out that the FDA is very specifically focused on what is written in the label.  And you will see this for the efficacy requirements, too.

So research is conducted.  And the information from those research protocols are found in the label.  So the label is a very important document and reflective of some of the clinical findings.

In 1962, the Kefauver-Harrison amendments were passed, in part, in response to the thalidomide tragedy.  That is what we call the efficacy requirements, which, again, are still in effect today.  And both these pieces of legislation are actually quite complex.  And I'm just giving you a few of the highlights of these pieces of legislation.

It basically states that sponsors must show before marketing that, based on substantial evidence, which is adequate and well.controlled clinical trials, the drug will have the effect it is claimed to have under the condition of use, prescribed, recommended, or suggested in the proposed labeling.

It also required maintenance of strict distribution records, which was part of the problem in the thalidomide circumstance.  Once it was already recognized that thalidomide caused phocomelia or limb deformations, they couldn't retrace all the pills that were distributed.  And it required that investigators supervise trials.  It required informed consent from study subjects.  And it required that animal studies be done prior to human exposure.

I put these up because these are certainly significant for the protection of children.  They're significant for the protection of all human subjects.

So now I would like to sort of trace what I think is the beginning, if you will, of some of the history related to the inclusion of demographic subgroups in clinical trials and specifically children.

And I would like you all to recognize that it's not a dichotomous situation where one day we had no children in clinical trials and then the next day we did include them.  In the 1950s, children were already being enrolled in multi.site oncology research, which was the harbinger of what we have today, which is the Children's Oncology Group, which is a national and at times multinational research endeavor on behalf of pediatric oncology.

In 1966, of course, Henry Beecher published his famous New England Journal article that described about 20 or a little bit over 20 research infractions.  And Dr. Fost described Willowbrook, which was a different perspective on Willowbrook, but that was one of the studies that was mentioned in Dr. Beecher's publication.

In 1972, of course, we have the Tuskegee atrocities that were brought to full light, which stimulated the formation of the National Research Act and the national commission.

So at that point, there was this general, although not complete, orientation to restrict children's participation in research and protect them accordingly.

In 1977, the commission issued its report and recommendations on research involving children, which forms the ethical basis of our current pediatric regulations.  It was really a novel and insightful and innovative and very thoughtful report.  The subpart D regulations that I am going to discuss later come directly from that report.

So already in the 1970s, the American Academy of Pediatrics' Committee on Drugs called for new drugs to be studied in children under controlled circumstances.

And at that time, the AAP said that it was unethical to adhere to a system which forces physicians to use therapeutic agents in an uncontrolled experimental situation virtually every time they prescribe for children.  Dr. Fost alluded to that today.

The figures are somewhere between 75 and 80 percent of drugs that are used in children, prescribed for them, had not been rigorously, systematically tested.  And so that children, when they go to the pediatricians' offices, are prescribed medication and basically, in an unsystematic, uncontrolled manner are turned into subjects of one.  So that's what that piece of those guidelines was in reaction to.

In 1979, the FDA added a pediatric section to the labeling.  So if you open the Physician's Desk Reference in the 1970s, it looks a lot different than it does now in that there was no pediatric section at that time and there was no pregnancy section at that time.  So there have been marked changes in the way that we study demographic subgroups.

Now, I think one of the pivotal times in history for research and the notion of this shift from risk-laden to hope-laden was the 1980s AIDS crisis, where clearly there was a devastating and life-threatening illness with no known treatments.  And the AIDS activists lobbied very hard for research access, which is the only way to get any hope at therapeutics.

In 1985, the FDA adopted a regulation which required integrated summaries of safety and efficacy by gender, age, racial subgroups for any new drug application.

And, actually, just to give you some perspective, children were not the only ones who were recommended to be excluded from trials, pregnant women or women of child.bearing potential were also.

In 1977, there was a guideline at the FDA that women of child-bearing potential should not be enrolled in clinical trials.  And that certainly has also shifted dramatically to the point that in 2004, the FDA issued guidance on how you do pharmacokinetic studies in pregnant women because there's an understanding now that certainly many medications are used and needed in pregnancy.

I would say and many would say that beginning in 1994 through to 2003 was the near decade of most significant advancements in pediatric legislation and research endeavors and initiatives.

In 1994, there was a regulation that was published that said that sponsors should review pediatric data to determine whether existing data was adequate to support pediatric labeling.  It basically introduced the concept of extrapolation of efficacy from adults to children.

So, in other words, if you have a disease such as, let's say, Crohn's disease, that could occur in adults as well as children, sponsors were supposed to think about whether you could take the information you know from adults and transfer that down to the pediatric population.  That's basically what part of that regulation says.  Now, it did not require that clinical studies be done.  And that, you'll see, will change later.

In 1995, again, the American Academy of Pediatrics' Committee on Drugs issued a statement that there is a moral imperative to formally study drugs in children so that they can enjoy equal access to existing as well as new therapeutic agents.  I think that their statement was very much in line with what it was in the 1970s, although they wanted to contemporize it.  But they basically agreed with the fact that there was this moral imperative, which was their earlier statement as well.

FDAMA was an extremely important piece of legislation for pediatric research.  And that occurred in 1997 because it authorized what's called "pediatric exclusivity incentives," that were designed to address the need for improved pediatric information.

And what it did was say that if the FDA issued a request for pediatric studies and if those studies were designed and approved by the FDA and they were completed, then the sponsor could get an additional six-month period of market protection, not only in the specific drug that they tested in children but in the entire moiety, in any preparation that used that particular chemical.

So sildenafil, for example, is now tested for pulmonary hypertension in children, but it's the compound that's known as Viagra.  So if that company were to get six months of exclusivity, then it would apply to all of the products and all the uses of that moiety.  So this was a very important incentive to encourage pediatric research.  And that was I guess commonly called "the carrot."

In 1998, there is the final pediatric rule, which is also called "the stick."  It requires the evaluation of pediatric information and supporting pediatric evidence.  However, that was enjoined early after it was legislated by a federal court, stating that the FDA had overstepped its regulatory authority.

In 2000, the Children's Health Act was mandated and that all research on children conducted, supported, or regulated by HHS comply with subpart D.  And their Code of Federal Regulations is 45 CFR 46.

In 2000, the ICH document, which is a document that's a global document meant to harmonize amongst all the nations that participated in it — and that was Japan, the European Union, and the United States.  They issued a document called "Clinical Investigation of Medicinal Products in the Pediatric Population."  That gives nod to the recognition that pediatric research could be done globally and that it would be important to try and harmonize the human subjects protections across those participating nations.  In 2001, the FDA adopted the subpart D regulations, which we'll talk about in much more detail later.

And then, finally, there are two more pieces of legislation that I want to go over before we shift to today.  In 2002, the Best Pharmaceuticals for Children Act basically reauthorized FDAMA's pediatrics section sunset, the pediatric exclusivity.

It also established not only the process of studying drugs that were on patent but a process of studying them, the drugs, that were off patent.  And it mandated that FDA and NIH establish a collaboration in the study of drugs.

It also mandated that there be public dissemination of summary findings for those studies conducted under the provisions of BPCA.  So this is a significant change because it mandated that there be a public dissemination of findings, whether they were positive or negative, because there was this recognition that there was a dearth of information about the pediatric population and that any information would be very helpful information.  So it did not necessarily have to be positive findings.

BPCA also stated that the FDA written requests would ask sponsors to provide information on the representation of children of ethnic and racial minorities as appropriate, that neonates should be considered as appropriate.  And it established the Office of Pediatric Therapeutics at the FDA, which I'm in.  And it also required that there be an ethicist in that office and a pediatrician.

It mandated the public review of safety reports so that the Pediatric Advisory Committee would be a body that would hear the public review of adverse event reports that were accumulated as drugs were used.  And it mandated the dissemination of pediatric information, as I had said before, obtained under BPCA.

The pediatric rule, which I had mentioned had been previously enjoined, was reinstated in 2003 through the Pediatric Research Equity Act.  And this required that studies be done of certain drugs and biological products if certain conditions were met, such as there was a new indication, dosage form, route of administration, et cetera, or if there were meaningful therapeutic benefits to this proposed new medication over existing therapies.  And that could be in the realm of treatment, diagnosis, or prevention.

And it amended BPCA to broaden the functions of the Pediatric Advisory Committee.  And I'll talk a little bit more about that committee as we go through the slides.

So what is the result of all of these regulations and legislations?  Well, currently we have had 100 products reflect new labeling changes based on these pediatric studies.  And this labeling includes new indications, safety warnings, dosing information, and sometimes extension to lower age groups.

A hundred and twenty-five products have been studied.  And those results have been submitted to the FDA.  Some are still under review, and some did not lead to any labeling changes.

Here are some of the lessons learned from pediatric trials.  Not all drugs that work in adults actually work in children.  And I have included some examples.  There is rosiglitazone, which is for type II diabetes; and zolmitriptan for migraine headaches.

Some drugs may need different trial designs for children than adults to demonstrate efficacy.  So antidepressants are difficult to study, even in adults, but in pediatrics, we have not been able to demonstrate efficacy.  And it's unclear whether it's the trial design or the drug itself.

Weight-based extrapolation of dose from adult data may be incorrect.  So in the past, the way pediatricians would prescribe medications is simply to alter the dosage based on body weight adjustments.  In many of the drugs that we have gotten data back on, we find that this assumption is incorrect.  And this can lead to overdosing, as in Luvox for girls between the ages of 8 to 11 may require lower doses, and Lodine for juvenile rheumatoid arthritis requires about a two times higher dose per kilogram than is required in younger children than adults.

Additional lessons that we have learned are that there are undefined unique pediatric adverse events.  Lotrisone for inflammatory dermatoses, for example, demonstrated that a third of 3- to 12-year olds had hypopituitary adrenal access suppression and decreased the amount of corticosteroids that they produced.

Camptosar, which is for chemotherapy, had marked dehydration and hypokalemia in children when compared to adults.  And then a number of drugs — I just picked two.  A number of drugs have effects on growth and behavior, many of which are reversible when the children come off the drugs, Ribavirin for hepatitis C and budesonide for asthma.

And then, finally, additional lessons that we have learned are that even within the pediatric population, which is quite a heterogeneous population, — we have heard from Skip.  He has talked about premature infants and children who need tremendous care in the NICU.  That's a very different population than your healthy 17-year-old football player.

There's tremendous heterogeneity within the pediatric population itself.  And the way that pharmacokinetics, the way that drugs are metabolized is more variable within the pediatric population than originally anticipated.

So, for example, Pepcid clearance values change with age.  And Sotalol or a beta blocker, in that particular trial, they found that you couldn't even make weight-adjusted changes based on pediatric information.  Body surface area adjustments were needed to avoid cardiac arrhythmias.

And then, additionally, I wanted to point out that new technology now allows modifications that we can make to the risk-benefit ratio when we do pediatric trials, such as seen in pharmacokinetic studies, where it used to be that you had to do population pharmacokinetics and draw a lot of blood samples from a small number of children to generate a dose-response curve.

Now there has been the understanding that you can do sample pharmacokinetic studies, where you draw less blood samples from each individual child but you draw them from a greater number of children so you can get an integrated curve.  So technology is also informing what we understand about pediatric trials.

I want to look further at the shift from risk-laden to hope-laden.  This is a simple schematic that discusses the ethics in the altered paradigm; whereas, before there was paternalism and protectionism, with the focus on non.maleficence.  The thought was that children should be excluded from unnecessary research risks.  With research inclusion, there is more an honoring of the freedom to participate, accessibility, and beneficence.

So some of the societal and medical factors that have contributed to this shift are — as I had said before, there's been a lot of experience with children in oncology research, and it's been an extremely positive experience overall— the AIDS epidemic, in which research was the only setting where potentially lifesaving treatments were available, and advocacy groups and professional societies have been significant, Elizabeth Glaser Foundation and the American Academy of Pediatrics.

It's not just societies limited to children's interests, but even groups that have advocated on the parts of women, minorities, and elderly have informed us about demographic subgroups.

There's been a growing sophistication, as I mentioned, in the understanding of the heterogeneity of the pediatric population.  And then, once again, there's been a recognition that because of significant off-label use, children have been exposed to treatment risks in the interest of avoiding research risks.

So where are we today?  Today I would say that there is a general consensus about scientific and ethical necessity of including children in research.  But there still is a debate about how best to protect children in the research environment, and I'd like to talk a little bit more about two challenging areas:  the risk levels found in 21 CFR 50, subpart D; and the involvement of healthy children.

So what I would like to do now is to go over in very schematic, simple fashion what subpart D is.  Subpart D, our additional safeguards for children in clinical investigations, basically says that an institutional review board, an IRB, can authorize a pediatric protocol outright in their own committee's deliberations if they determine that that protocol falls into one of three categories.  It can either be a protocol that is deemed to be only minimal risk exposure for kids, or it can be a protocol that has greater than minimal risk but it presents the prospect of direct benefit to the individual subjects enrolled in the trials, or it can involve a minor increase over minimal risk that presents no prospect of direct benefit to the individual subjects.  However, it's likely to yield generalizable knowledge about the subject's disorder or condition.

Now, this, as I said, came directly from the national commission's report on children.  At that time, they wanted to sort of leave, if you will, an out.  They wanted to protect children, but they also wanted to allow an opportunity for important research to go forward, even if it didn't fall under one of these first three categories.

And so there is an additional category to subpart D.  That basically says that if the IRB feels that the research cannot fall in one of the first three categories of subpart D but the RIB has determined that the research presents an opportunity to understand, prevent, or alleviate a serious problem affecting the health or welfare of children, then it can send that protocol to the federal agency for an expert panel review.

The way that works — and I'm going to show you towards the end of the presentation a few protocols that have come to the FDA and to OHRP under this provision.  The way that works is the IRB will contact the FDA if it's FDA.regulated protocol or OHRP if it's research that's federally supported or conducted for federal review.

We convene an expert panel that draws off of ethicists, physicians, lawyers, education specialists, patient advocates, statisticians.  We comprise a group that is varied and that we think each person's expertise will be significant to helping the deliberation of that protocol.

It's a public process, and it affords the opportunity, not only for written public comment but also for personal appearances on behalf of the public, any advocacy group that wishes to come.  And we'll discuss a little bit more about what the comments have been like in relation to those referrals.

Once the expert panel has completed its deliberations, then the Pediatric Ethics Subcommittee chair presents those deliberations to the Pediatric Advisory Committee, which is the parent committee to the subcommittee.

The Pediatric Advisory Committee is chartered to make a recommendation to the FDA commissioner and, for purposes of these referrals, to the Secretary of HHS as well.  So that we now have because of the liberation of the Pediatric Advisory Committee charter under the Pediatric Research Equity Act of 2003, which I mentioned, the ability to have a single federal panel.  So that a referral that came to the FDA would not be reduplicated in its review by OHRP.  That definitely streamlines the process and makes it more efficient and makes it more ethically and scientifically coherent.

So what I want to look at is in the first three categories, I'm going to talk about the referral process a little later, but now I want to talk about what does subpart D have to say about risk levels that we can expose children to as part of research?

The federal regulations define minimal risk.  And they define minimal risk as the probability and magnitude of the harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests.

So right off the bat, you can see that "daily life" in this particular definition is not tethered to the daily life of home.  And the national commission had said in the daily life of healthy children, over the years it's been discussed and rediscussed and as recently as March of 2004, the Institute of Medicine issued a report on pediatric research and interpreted the "daily life" to mean in the normal average healthy child living in a safe environment.

The other risk category that is found in subpart D is greater than minimal risk.  That has no regulatory definition.  In 50.53 category, it talks about a minor increase over minimal risk, but certainly there can be understood to be more than a minor increase over minimal risk, particularly in the referrals sent under that subpart D referral process.

So if we look at the regulations a little bit further, I've picked out and you can see highlighted in green some terms that might help us put risk into a little bit better focus.

50.52 refers to the notion of justifying the anticipated benefit to the risks based upon the anticipated benefits to the subjects.  So there's a risk.benefit calculus that occurs when you're looking at the direct benefit category of subpart D.

It also says under 50.52 that the relation of the anticipated benefit to the risk should be at least as favorable as that presented by available alternative approaches.

50.53, the category that looks at minor increase over minimal risk, where you have the prospect of generalizable knowledge but not to the direct subjects involved, talks about the interventions or procedures present experiences to the subjects that are reasonably commensurate with those inherent in their actual or expected medical, dental, psychological, social, or educational situations.

And then, finally, 50.53, again, referring to the category of minor increase over minimal risk, states that the anticipated knowledge should be of vital importance to the understanding of amelioration of the subject's disorder or condition.

So the preamble to the FDA regulations gave a series of examples of what they thought might be minimal risk when the regulation was published.  You can see clean catch urinalysis, stool samples, and electroencephalogram, minimal diet or daily routine changes, or standard psychological tests.  This puts the cap at a very low level.

And also one of the points that I wanted to make about the subpart D regulations is if you look at them, 50.52 and 50.53 all refer to subjects with disorders or conditions.  So if you're trying to categorize healthy children in a research protocol or healthy controls, then they either have to fall into the category of 50.51, which is minimal risk, which is described here in this slide, or they have to fall into the category of 50.54, which requires a federal referral for approval.

Now, what do we know about minor increase over minimal risk?  Well, the national commission and the IOM report both suggested that minor increase over minimal risk be understood as slightly more than minimal risk, which, again, still implicates a very low level of risk that is acceptable for children as part of research.

This is a slide that sort of brings together a series of different suggested factors that might help us understand risk.  And what you want to look at is not only the types of interventions but the times that they're repeated and the cumulative nature of the interventions in the protocol, whether there are special considerations for special populations.

So if you have children who are anemic, blood drawing may be more of a risk than it might be for a population of children who are not anemic.  You have to look at the magnitude probability and duration of the risks and also the sense, as I had said before, that risk may be age.graded based on the heterogeneity of the pediatric population.

Look at the equivalence to daily life in terms of experiences and routine physical or psychological examinations or tests and look at commensurability or comparability, which I think are quantitative and qualitative comparisons to experiences already familiar to children being studied.

David Wendler in a very recent article stated that they were going to — this particular article was an attempt to try to quantify what are the risks that children are exposed to as part of daily life.  And so they looked at what are the risks of the average child dying from a car trip or having some kind of an injury from sports.

The figures are up on the board.  Basically, what Wendler and Emanuel concluded after they looked at those figures is that more empirical data is needed to quantify risk levels.  They also suggested that maybe based on the fact that those risk levels seem to be higher than what some IRBs might interpret, that maybe IRBs are overestimating.  Well, wait a minute.  I'll go back up and say this again.

Based on the fact that these seem to be higher risk levels than one might imagine for children as part of daily life, they thought that maybe some IRBs are actually overestimating minimal risk.

An example that they use is that allergy skin testing by 60 percent of IRBs that were polled thought it was more than minimal risk.  Some thought it was a minor increase over minimal risk and some thought it was more than minimal risk, but only about a third of the polled IRBs thought that allergy skin testing was actually minimal risk.

They also suggested that if the current federal regulations index risk to daily life and daily life seems to be more risky than we think it is based on empirical data, maybe the current federal standards are allowing excessive risk in some cases.

Basically they call for more empiric data on risk level.  And when they were asked to look at a definition for minor increase over minimal risk, Wendler and Emanuel proposed that it be understood as the level of risk in the lives of those children who face a greater, yet socially acceptable, risk.

Friedman, Ross, and Nelson suggest that minimal risk should be understood as the probability and magnitude of physical or psychological harm is no more than that to which is appropriate for a scrupulous parent to intentionally expose a child for educational purposes in family life situations.  And they suggest that there ought to be a single standard of research for children with or without the disorder or condition under study.  What you see here is an attempt to try and place pediatric research and our understanding of risk within a social context, I think.

Now shifting over to healthy children, this is the second topic that I would like to cover.  Going back to the Pediatric Advisory Committee, it is a committee of multiple specialists, including patient advocates and consumer representatives, et cetera, that meets to deliberate issues of clinical trials in pediatrics, ethical issues.

And it issued a consensus statement in 1999 on the topic of when should healthy children be included in pediatric research.  And they stated that, in general, pediatric studies should be conducted in subjects who may benefit from participation in the trial.  Usually this implies that the subject has or is susceptible to the disease under study.

Looking at the ICH document, E11, that I referred to earlier, it states that pharmacokinetic studies should generally be conducted in patients with the disease and if the disease process is similar, consider extrapolating efficacy from older to younger children.  And information that can be obtained in a less vulnerable consenting population should not be obtained in a more vulnerable population.

So what you see suggested is that there be perhaps a staged approach to some pediatric research if it's scientifically valid to do so.

ICH E6, which is the guidance on good clinical practice, states that unless an exception is justified, research should be conducted in patients having a disease or condition for which the investigational product is intended.

And then the Council for International Organizations of Medical Sciences in collaboration with the World Health Organization in 2002 issued 21 or 22 guidelines with a company in commentary.  This is also understood as an international document.  In it, they have a guideline specifically addressing research in children.

They state that before undertaking research involving children, the investigator must ensure that the research might not equally well be carried out with adults.

Now I would like to shift back to what our experience has been with the subpart D referrals and talk about three referrals that we have had since December of 2003, when the Pediatric Research Equity Act reauthorized the Pediatric Advisory Committee, and go through a little bit more before I do that of what a referral looks like.

So once an IRB goes through its own internal deliberations on a protocol and they determine that they think that there is an opportunity to understand, prevent, or alleviate a serious problem affecting the health and welfare of children but they just cannot approve this under one of the first three categories of subpart D, then they make the referral to the pertinent or joint federal agencies.

We convene an expert panel, as I mentioned to you before, and usually take about a day's worth of deliberations.  And those deliberations include any scientific expertise that we think needs to be presented to the expert panel.  It includes a presentation by the principal investigator.  And it includes a representative of the institutional review board of record to flesh out what their applicable concerns have been with this particular protocol.

Then the referral is presented to the Pediatric Advisory Committee once a recommendation has been determined by the Pediatric Ethics Subcommittee.  And those recommendations have to include what approval category the subcommittee recommends and whether or not there are any required modifications to the protocol or parental permission or assent documents or whether there are any recommended modifications.

That gets presented to the advisory committee, which deliberates.  And in two out of three cases, the advisory committee added they didn't disagree with the actual recommendations of the subcommittee, but they added some stipulations to the list of recommendations.

Then our office writes a memorandum and transmits those recommendations along with a memorandum to the commissioner and some additional supporting documents.  And the commissioner of the FDA makes a determination on whether or not the protocol can go forward.  The commissioner could also decide the protocol can't go forward or could make modifications in the recommendations.

If it involves federally supported or conducted research, then the commissioner's findings are then transmitted to the Office for Human Research Protections.  And they submit a memorandum to the Assistant Secretary for Health, who has been authorized to make these decisions on behalf of the Secretary.

And once the Secretary or the ASH, I should say, makes the final determination about whether or not that project can be funded or supported, then the project can go forward.

So it has actually been a process that sounds rather long and complex, and it is.  But it has actually been a process that has been greatly streamlined since prior subpart D referrals before the initiation of the advisory committee.

So here are three examples of subpart D referrals.  All three of them were initiated because the IRBs felt that the healthy children or the children who did not have the disorder or condition under study (the control group) did not fall into the first category of subpart D of minimal risk.  And so, however, in each case, the IRBs felt that the research was significant research.  And so these were referred.

In the first case, a single dose of dextroamphetamine — it was a ten.milligram dose — was to be given to children with and without attention deficit hyperactivity disorder.  And then functional magnetic resonance studies were going to be done on all the children to see if there were differences in the neural anatomy of children with ADHD and children without ADHD.

In the second case, precursors, phospholipid precursors, were to be administered to neonates to try and determine preferential use of these phospholipids in the synthesis of surfactant.

In that case, these were premature infants who had respiratory distress syndrome, previously known as hyaline membrane disease.  And they wanted to look at the control group.  In order to minimize any risk that they exposed the control group to, they selected full-term infants who were also intubated and had arterial access in the intensive care unit, but they were there for purposes other than lung disease, such as awaiting cardiac surgery, et cetera.

So, in actuality, in that case, the main research intervention was the administration of the phospholipids, which can be found in hyperalimentation.  There were no additional blood draws, and any of the tracheal aspirates that they were going to get were going to be gotten when routine care was delivered.  So that was the comparison group there.

And then in a third case, gonadotropin-releasing hormone agonists tests in disorders of puberty, there has been a commercial lack, for various reasons, at this point in an adequate diagnostic test to try and differentiate some disorders of puberty.

So in this case, the principal investigator wanted to administer Lupron, which is a gonadotropin-releasing hormone agonist, to children who both have disorders of puberty and to children who do not have disorders of puberty, to try and establish some sense of whether this challenge test could be used as an adequate means to make this differential in diagnosis.

So it involved the administration of a single dose at the levels that are currently approved for the treatment of precocious puberty in children.

And I can expand on any of these descriptions, but the point that I was trying to make in this slide is that these were still fairly low levels of risk for the healthy children, but because of the way our subpart D regulations are written, they required a subpart D referral.

In the first case, the commissioner approved that the study went forward, but the study was eventually withdrawn by the sponsoring agency.  In the second case, both the commissioner and the ASH approved that the study go forward.  And in both of those cases, those were with stipulated and recommended modifications.  And in the third case, that was just heard in November of '05.  So we don't have determinations on that one yet.

I look upon these particular subpart D referrals as almost case precedents because they help us advance our understanding of what we think is acceptable or not acceptable in pediatric research.

So, in conclusion, there has been significant progress in pediatric therapeutic research from both the ethical and scientific perspectives.  Federal legislative initiatives have proven to be important and useful tools in obtaining pediatric information.

And, as a consequence of study in children, there is an improved understanding of pharmacokinetics criteria for extrapolation, unique safety concerns, and trial designs for children.  And a public dissemination of summaries of pediatric trial results is important because of the limited numbers of these trials.

Thank you.

CHAIRMAN PELLEGRINO:  Thank you very much.

Rebecca?

PROF. DRESSER:  At one point you were having a shifting from risk-laden to hope-laden in research.  And so it sounds as though you think it's a happy ending and everything is great, but I wonder if you still have some ethical concerns about what has happened along with this shift.

DR. GOLDKIND:  Well, as my slides have demonstrated, I do think that pediatric research is extremely important.  I don't think that the willy-nilly inclusion of children in pediatric research is appropriate.

I think that the graded introduction of research into the pediatric population is important.  And I think that there are a lot of considerations that have to go into whether or not children should be in the research.

I think, first and foremost, there needs to be attention to the scientific question and is this scientifically valid research, do children have to be included to answer the question, is this necessary research.  Even if it's perhaps scientifically valid, it might not be important research.

I think that once you get to the point — and then you, of course, have to ask whether information can be extrapolated from adults.  Sometimes there are disease states where you can extrapolate efficacy, but you still may have to do some supporting studies, such as the pharmacokinetic studies or the safety studies, in children.

So you have to ask what studies do I absolutely have to do in children, what studies can I use from adults.  I think you always have to ask what does the animal data show, the preclinical work, and is this a risk that we can take with children.

I think that the subpart D regulations are an important part of protecting children in research.  I think that we do still have to answer the second part of your question, which is, what are some of the ethical issues in pediatric research?

One of them would be a better understanding of those risk levels.  Another would be understanding the ability to globalize those risk levels as we look more and more towards globalization of research.

PROF. HURLBUT:  Just a little clarification.  You mentioned at one point that some studies were acceptable if they serve to prevent or alleviate serious problems in the health and welfare of children.  I wanted to understand what that would include, because some conditions that you might want to study you could only study in children but don't manifest themselves until later.

Why would you separate off children except for the fact that if you can do it in a later group; therefore, more voluntary subjects or with informed consent?   The fact that it's benefitting children, is that to the point of the regulation or is it that —

DR. GOLDKIND:  Well, these regulations were specifically adopted to protect children.  They require the additional safeguards for children in clinical investigations.  So they do specifically address pediatric research.

PROF. HURLBUT:  Let's use a hypothetical example that you wanted to study later effects of obesity and some circulating element or something, diabetes or something like that, and you could see a benefit to adults by studying children.  Would that be acceptable?

DR. GOLDKIND:  Disorder or condition is one of the aspects that I didn't discuss in this particular regulation.  And disorder or condition is, again, another one of those issues that requires some further discussion, but it has elements of current illness or potentially an at-risk component to it or a future element to it.  That's certainly at least the way the Institute of Medicine has understood it and other bodies as well.

So if you can think that there is a condition in children that may put them at risk for something in the future when they become adults, then there's the understanding that you can go ahead and look at the pediatric population and include them in research on that particular issue, given all the caveats that I've already addressed.

The reverse, where you have a disease entity in adults, you have to make the case for why it's scientifically valid and necessary to study the pediatric population, rather than a consenting adult population.  And when you have to make that case, I mean, the onus is upon the investigators, the researchers, the IRBs to understand what it is about studying that pediatric population that warrants using a population that cannot consent.

PROF. HURLBUT:  Just to take it one more step, you're saying, though, if you could justify that, you could do it.  I'm thinking kind of hypothetically here, but there are quite a few conditions in adult life, like osteoporosis, for example, that you might gain some interesting information by invasive studies, slightly invasive studies and maybe very invasive studies.

I mean, this is hypothetical, but suppose by coring out a little piece of bone marrow you could predict something that would happen at age 50.  I mean, could you theoretically justify that or age processes in studies like this eventually?

DR. GOLDKIND:  Well, you know, as they say, the devil is in the details, but if you could in some way justify why you have to study that pediatric population and why they may eventually become at risk as adults and there were some meaningful invention or preventive mechanism that you could learn from that research, I would imagine that you could study children.

But there are a lot of caveats.

CHAIRMAN PELLEGRINO:  Leon?

DR. KASS:  Thank you very much.  I learned a lot, starting from a very low level of understanding.  So I appreciate this.

My general impression is that people are being very, very cautious, notwithstanding the shift of the paradigm if these are the cases, if these are the sort of illustrative kinds of cases that come for referral under the 50.54.

I'm trying to think ahead if we're told that there's a huge mushrooming of pharmacological agents, especially directed at children, and if the general rule is children are different and you need to do special kinds of treatment.

I guess the question is, what is the magnitude, anticipated magnitude, of the need for involving healthy children as research subjects, not for, say, a single shot to find the rate of renal clearance of a drug but for questions of toxicity or things where your population of patients is small; to do meaningful studies, you need larger groups.  I'm thinking out loud.

I could see that once you begin to open the door to recruitment of healthy children as control groups here, for one reason or another, and we're talking about a greatly increased pharmacopeia, that this could be a very large business.

Am I right in thinking about that?  Are people sort of concerned about the question of recruitment?  Just to say, I mean, when the question of imposing risks on children is raised and then Norman Fost says, you know, venipuncture, well, I could persuade an eight or ten-year-old I think to be a good sport for that.  I would be much more reluctant to persuade them to injections of medications, the very hazard of which we're trying to study here.

So can you help me sort of think about that with respect to the healthy children question?

DR. GOLDKIND:  Well, I think there are a number of components to your question.  One is that I can hearken back to some of the data that we have.  We have had about 45,000 children enrolled in clinical trials since the 1997 exclusivity regulations, legislation, so 45,000 children in about 8 years of trials that the FDA has requested.  That's not all pediatric trials for sure.  Those are trials that the FDA requested as part of their exclusivity arrangements.

Since FDA adopted its subpart D regulations in 1983, they have had roughly 20 subpart D referrals.  So that's about one per year at the most, probably a little less than that.

And there are all kinds of factors that influence taking this in to the practical settings to be able to answer your question.  There are all kinds of factors that influence pediatric research.

One of the reasons that the pediatric exclusivity incentives were so important to being able to accumulate the information that I've shown you today is because there were financial incentives associated with them.

Children tend to be, generally speaking, thankfully, healthy.  The market for medications is smaller than it is for adults.  And certainly for devices as well.  And that has actually been one of the hindrances to pediatric device development.  I've talked a lot today about drugs and biologics, but devices is another whole spectrum of products.

And sponsors are not always interested in pursuing that market.  So that is one component, one real component, one practical component, to keeping a check, if you will, on how many kids are in pediatric trials.

Now, I think that the subpart D regulations are protective to children because of precisely the scenario that you described.  And that is that when you have a cap on the amount of risk that children can be exposed to, the nice thing is there is this safety valve in subpart D, which is the referral system.

So, even though there is a cap, there is a low level of risk that they can be exposed to, there is a mechanism by which perhaps that research can still be advanced if needed.

I think that that cap mitigates just sort of the free flow of healthy children into research.

CHAIRMAN PELLEGRINO:  Gil?  Dr. Meilaender?  And let me say we have exceeded our time.  So this may be the last question.

PROF. MEILAENDER:  This is a related question in some ways.  It has to do with healthy children again.  The shift from the risk-laden to the hope-laden paradigm, I mean, I think I understand what you mean by that and understand how it might be appealing in certain circumstances, but just conceptually or theoretically, it seems like the wrong paradigm to apply when thinking about healthy children.

I mean, in the hope-laden paradigm, what are they hoping for in doing it, just theoretically?  Can you explain why, if we're thinking about inclusion of healthy children in research, we shouldn't think in terms of the risk-laden, rather than the hope-laden?

I mean, I think the hope-laden paradigm emerged precisely in terms of people with serious illnesses wanting to get early and better access to care and so forth.  And I can't make it fit the healthy children model, but maybe I'm missing something.

DR. GOLDKIND:  Well, you know, I think that perhaps we have to think of that in a somewhat flexible sort of way.  I think that if you look at the figures that I presented and that Dr. Fost presented, that up until very recently, 80 percent of children who went to the pediatrician got medicines that were not tested in children.

So that meant that there were a lot of risks that children were being exposed to in just the average course of going to their pediatrician.  This was not recognized by a lot of parents.  This wasn't recognized by children who were there getting their medications.

And there were — I presented maybe ten or so changes in the labeling just as illustrative changes, but there are over 100 changes.  There are changes in over 100 labels.

So, I mean, I really think of the shift in paradigm really, as you say, to more children who have disorders or conditions who are trying to get some benefit out of the research, but I think that you can, if you're flexible, understand this notion of hope in a more general way if you're reducing the risks that kids are exposed to in their average course of encounter with the medical system.

CHAIRMAN PELLEGRINO:  Well, thank you very much, Dr. Goldkind.

DR. GOLDKIND:  Thank you.

CHAIRMAN PELLEGRINO:  A very, very complete review.  Thank you.

(Applause.)

(Whereupon, the foregoing matter was concluded at 5:11 p.m.)