TRANSCRIPT: Meeting Six, Session 7

Trial Design and International Standards


August 30, 2011


Washington, DC


Ruth Macklin, Ph.D.

Professor of Bioethics, Department of Epidemiology and Population Health, Albert Einstein College of Medicine

Robert Temple, M.D.

Deputy Center Director for Clinical Science, Center for Drug Evaluation and Research; Acting Director, Office of Drug Evaluation I, U.S. Food and Drug Administration (FDA)

Download the Transcript


            DR. GUTMANN:  Welcome back, everybody.  We are

  now in session seven of this two-day commission

  meeting, and I am very pleased to welcome our two

  guests, Ruth Macklin and Robert Temple.  And I

  would -- will briefly introduce both of them.  They are

  widely known and respected, and will present us with

  two different views on the topic of trial design and

  international standards.

            Ruth Macklin is Professor of Bioethics in the

  Department of Epidemiology and Population Health at

  Albert Einstein College of Medicine in New York.  She

  currently co-directs an NIH Fogarty International Center

  training program in research ethics, which takes place

  in Buenos Aires, Argentina.  She is a member of the

  research protocol review panel in the Human

  Reproduction program at the WHO, and she also serves on

  the Vaccine Advisory Committee at WHO.

            She has published more than 200 scholarly

  articles and chapters in bioethics, law, medicine,

  philosophy, and the social sciences.  She is also

  author or editor of 11 books, including "Against

  Relativism," and, most recently, "Double Standards in

  Medical Research in Developing Countries."  Ruth,


            I will also introduce Dr. Robert Temple, who

  is Deputy Center Director for Clinical Science at the

  FDA's Center for Drug Evaluation and Research.  He is

  acting Director of the Office of Drug Evaluation, which

  oversees the regulation of cardiorenal,

  neuropharmacological, and psychopharmacological drug


            Dr. Temple has a long-standing interest in the

  design and conduct of clinical trials, and has written

  extensively on the subject, especially on the choice of

  control groups in clinical trials, evaluation of active

  control trials, trials to evaluate dose response, and

  trials using enrichment designs.  He also has a

  long-standing interest in the hepatotoxicity of drugs—I have to confess   I’m not sure what that is-liver?-Okay, now I know what that means, I should have known that--

  having participated in the first detailed FDA-NIH

  outside discussion of the subject in 1978.

            Look forward to both your presentations.

  Ruth, would you please begin?

            DR. MACKLIN:  Thank you very much.  I am goi to have one very simple message, and the simple message

  is to defend a single ethical global standard, or a

  single global ethical standard.  That is a simple

  message.  But understanding it and interpreting it is

  not so easy.

            So, there are some complications.  The first

  complication is what counts as an ethical standard.

  How would you frame the statement?  How would you frame

  the standard that you claim that one might claim to be

  a single global standard?  So I'm going to give a

  couple of candidates for such standards, and I'm going

  to throw them out -- reject them, I mean.


            DR. MACKLIN:  One candidate for a single

  standard might be phrased as follows:  If it's

  unethical to carry out research with a particular

  design in a developed country, or an industrialized

  country, it is unethical to do that same research in a

  developing country.  That's one candidate for how one

  would compare the design, the research design, in a

  developing and developed country.

            Well, I argue that that's flawed, because the

  particular circumstances can be sufficiently different

  to warrant different designs.  So, for example -- and

  this is the example I am going to use twice -- research

  on a preventive or therapeutic method that could be

  used or designed to be used in remote, rural areas with

  poor infrastructure could require an experimental

  intervention or comparator that would not be used in an

  industrialized country.

            And I want to point out here and emphasize

  that it's the infrastructure that is the problem, the

  inability to conduct the trial, or to bring the

  products of the trial.

            Here is a second example that I am going to

  throw out, another candidate for a single standard in

  research design might be phrased as follows.  All

  participants in research should receive the level of

  care they would receive in a developed country.

            Now, some people around the table may

  recognize that statement.  I'm not sure anybody has

  ever argued for it.  But this is flawed for basically

  the same reason as the previous one, because adherence

  to this requirement would make it impossible to do

  research designed to develop treatments for some

  diseases or some conditions, again, in remote rural

  areas of developing countries.  Or, even if they're not

  rural areas, there are places even in a city where the

  infrastructure simply doesn't allow you -- you might

  need an ICU, for example, an intensive care unit.  So,

  such research, however, is critically important to test

  interventions that can benefit people in countries with

  poor health infrastructures.

            So these are some of the reasons why those

  descriptions, or those interpretations of a single

  ethical standard, will not work.  So let me turn to

  placebos, because that's the area in which Bob Temple

  and I have grappled on many occasions, and the one that

  seems to raise -- it's not the only research design

  question--but seems to raise the most controversy.

            And here is the double-standard position.

  Placebo controls are acceptable in developing

  countries, where many people lack access to proven

  interventions.  And the ethical defense is that

  subjects are not made worse off than they would be if

  they were not enrolled in the research at all, since

  the placebo control group would get the placebo, which

  would be, let's say, equivalent to nothing, and the

  other half would get an experimental product that may

  or may not work.

            A second point is that a cost -- this is the

  defense of double standards -- a costly proven

  intervention would never be available to the

  population.  So why include it, according to this

  argument, as an arm in the study if it's never going to

  be available?

            And the final point -- and this is boiling

  down the argument to the barest essentials -- is that

  research subjects are treated equitably.  Now, what

  does equitably mean here?  The argument is they receive

  the level of care that patients in the same community

  receive for the same disease, or same condition.  That

  may be nothing if it's a placebo-controlled trial.

            So, this is basically the argument that people

  are not being made worse off.  And the question is

  whether that is a minimal ethical standard, or

  a -- let's call it the minimalist view of ethics--that

  you're not making people worse off, even if you might

  design the trial in such a way that they could be made

  better off.

            Here is the single standard position on

  placebos.  This holds that the same standards -- again,

  standards for use of placebo -- in the sponsoring,

  industrialized country should be applied in the

  resource-poor country.  And here the statement is, "If

  patients in a developing country who are not enrolled

  in a clinical trial would receive no treatment, that

  cannot justify -- that alone cannot justify withholding

  an effective treatment from subjects in the research,"

  if, in fact, one could design the research


            I mean there are some premises here, that is

  that you have adequate research design that can provide

  answers.  So that is the underlying premise, that is

  that the science is sufficient to be able to obtain


            So, the third point, placebos may not be used

  in a control group in the poor country when an

  effective treatment for that condition exists in the

  industrialized country.  So the question of justice

  here is global justice.  On the previous slide I said

  local justice, meaning that the people who are in the

  trial would be treated equitably, according to local

  circumstances.  But now we're looking at something

  broader, and that's global justice.

            So, I'm not sure how much the commissioners

  are completely familiar with the international

  guidelines, since we are talking about standards.  So

  the next few slides are simply going to pull out what

  the international guidance – several--international

  guidance documents say about research design.

            Declaration of Helsinki in 2008.  The

  benefits, risks, burdens, and effectiveness of a new

  intervention must be tested against those of the best

  current proven intervention, except in the following

  circumstances.  And the first circumstance -- these are

  probably better known than some of the other

  guidelines -- the first circumstance is the use of

  placebo or no treatment is acceptable in studies where

  no current proven intervention exists.  This is

  non-controversial.  Nobody rebuts that.

            However, here is the tricky part -- and this

  is my italics that I inserted -- "Wherefore compelling

  and scientifically sound methodological reasons, the

  use of placebo is necessary to determine the efficacy

  or safety of an intervention, and the patients who

  receive placebo or no treatment will not be subject to

  any risk of serious or irreversible harm."

            Now, the reason I italicized those words,

  "compelling and scientifically sound methodological

  reasons," is precisely because experts may disagree on

  what is compelling and what is scientifically sound.

  So there is where some of the debate is going to lie,

  not so much with the first issue -- because the first

  question, if there are no treatments, there is no

  problem with using placebo.  So, it's in the analysis

  and determination of what is -- what are those

  compelling and scientifically sound reasons.

            Here is the Council for International

  Organizations of Medical Sciences, their statement of a

  single standard.  "In externally sponsored research,

  the ethical standards applied should be no less

  stringent than they would be for research carried out

  in the sponsoring country."  So this is now a statement

  about an international standard, a global standard.

  But it is not enough clarity on what constitutes


            The UN AIDS document, which is a companion

  document to the one that Mitchell Warren talked about

  this morning, here is what they say.  This is their

  comment on control groups.  "The use of a placebo

  control arm is ethically acceptable in a biomedical HIV

  prevention trial, only when there is no HIV prevention

  modality of the type being studied that has been shown

  to be effective in comparable populations."

            Here, however -- and they do give a couple of

  examples:  a vaccine that is not known to be effective

  against a virus that is prevalent, or a microbicide

  shown to be effective for vaginal intercourse but not

  for rectal intercourse.  So it can be different

  populations, it can be a different product.  So, those

  are examples that can show when it would be possible or

  acceptable to use placebo.

            I am going to skip this, because the time is

  running, and go to the ICHGCP, which is not an ethics

  document, but this section of their efficacy section

  has this to say.  "In cases where an available

  treatment is known to prevent serious harm, such as

  death or irreversible morbidity in the study

  population, it is generally inappropriate to use a

  placebo control."  Note that it says "generally."

            "In other situations where there is no serious

  harm, it is generally considered ethical to ask

  patients to participate in a placebo-controlled trial,

  even if they may experience discomfort as a result."

  And here are the two ethical points:  "provided the

  setting is non-coercive, and patients are fully

  informed about available therapies and the consequences

  of delaying treatment."

            So, my last slide here, next-to-the-last

  slide, is a statement from this European group.  And

  this is the point at  which I want to emphasize,

  which is that economics alone should not be the

  determinant about -- of whether or not one may use a

  placebo in a poor country.  The European group says,

  "Research activities involving human subjects cannot

  exclusively be assimilated to an economic activity

  subject to market rules."

            On the contrary, in the context of solidarity,

  regarding health as a public good, rather than as a

  commodity, it needs to be regulated according to

  fundamental principles.  The general approach chosen

  within this opinion is that fundamental ethical rules

  applied to clinical trials in industrial countries are

  to be applicable everywhere.  So that is the global one

  single standard.

            And my conclusions -- I have one more slide,

  sorry -- the basic value underlying the defense of a

  single ethical standard is global justice.  And the

  disparity in economic circumstances is not a morally

  relevant factor for determining ethical standards for

  research design, when the external sponsor is a wealthy

  country or a pharmaceutical company.

            And in -- my critique of double standards

  concludes that financial inequalities -- to endorse

  double standards would be to include that financial

  inequalities should be a significant determinant of

  global justice in the design of research.

            DR. GUTMANN:  Thank you, Ruth.  Robert?

            DR. TEMPLE:  Now, let me just say

  preliminarily that the main issue in a lot of this is

   -- turns on what the consequence of not getting the

  standard therapy is, and that will come up repeatedly.

            The major concerns in all this about the use

  of placebo are two main issues.  One is there is a

  general concern about the use of placebos, mostly

  raised by the 2000 Declaration of Helsinki -- I will

  touch on that -- and then the fundamental question of

  what a person in a trial is entitled to, best local

  versus best global therapy, all of which Ruth has

  touched on.

            And a second important issue on the case where

  being deprived of the therapy could have consequences

  to the patient that are serious is where this is being

  done, and whose interest the trial is serving.  And I

  will try to get to those.

            In symptomatic conditions -- this is a long

  story; I will be glad to dilate on it later, if you

  want -- in symptomatic conditions with very few

  exceptions, only a trial showing a difference between

  treatments is going to be interpretable.  If you merely

  see no difference, that is what is sometimes called

  equivalence or non-inferiority in symptomatic

  conditions, it is usually impossible to determine

  whether the drug, in fact, worked.  We have a long

  guidance out on non-inferiority studies.  This was also

  addressed in the ICH document E10.

            The main trouble is, with symptomatic

  conditions, you don't know whether the supposed act of

  control actually had the effect it was supposed to have

  in the trial.  A typical example is depression, where

  drugs we are quite sure work beat placebo about 50

  percent of the time.  So if you do a trial and you see

  no difference from the active control in a depression

  trial, how do you know whether this was a trial that

  could tell, or a trial that couldn't tell the

  difference between active and inactive treatments?

            So, with exceptions, symptomatic conditions

  generally need to show a difference.  You can be

  better, or you can beat a placebo.  That's what you

  have to do.

            So, in 2000, when the World Medical

  Association essentially banned placebos whenever there

  was a known effective therapy -- as Ruth showed you, if

  there is no effective therapy, placebos are fine -- had

  people followed that advice, there would have been no

  more development of symptomatic treatments if there was

  an existing therapy.

            Now, some people would say if the new therapy

  isn't better than the old ones, who cares anyway, but

  we can go into that.  New treatments often have

  advantages that are not in the form of superiority.

  They may be better tolerated, there is lots of reasons.

  They may have additive effects, when you study them

  later.  You do want new symptomatic treatments, in


            So, let me just go quickly through the

  Declaration.  So, from the very beginning, at least as

  early as 1975, the Declaration said in every medical

  study, every patient, including those in control, if

  any, should be assured of the best proven diagnostic

  and therapeutic method.  It has always been unclear to

  me what they meant by that, because even in an active

  control trial people aren't getting the best therapy,

  they are getting something somebody wants to test.

            Anyway, there were some people -- notably

  Rothman and Michaels in a pretty famous New England

  Journal of Medicine article in 1995 -- said that means

  if there is an existing therapy, you can't use a

  placebo.  And you know, in arguments on stages, Ken

  Rothman said you can't test a new drug for baldness

  because there was Rogaine.  You can't have a placebo

  control trial of baldness or seasonal allergic

  rhinitis.  You just can't ever do it.  It's unethical

  and unacceptable.

            Of course, as I said, read literally, the

  active control trial also doesn't give people a known

  effective therapy.  So, as a result, nobody took this

  particularly seriously until 2000, when, as Ruth showed

  you, it now said the benefits, risks, burdens, et

  cetera, have to be compared.  You can use a placebo

  when there is no known therapy.  But, otherwise, it

  clearly says you can't.  So that means if there is an

  existing therapy, an allergic rhinitis, baldness,

  whatever it is, you cannot do a placebo control trial.

  Everybody noticed.  IRBs were very concerned, because

  they thought you wouldn't be able to do trials any


            After a lot of concern by FDA, HHS, and a lot

  of people, the World Medical Association eventually

  changed it.  They made a mistake initially.  They said,

  "Where for compelling" -- this was in 2001 -- they

  said, "Where for compelling methodological reasons it

  was necessary to use a placebo, or where a prophylactic

  diagnostic was being investigated for a minor


            Well, that's not right.  That first one is

  grossly unethical.  That says if you can't get

  information, except with a placebo control trial and

  you have to kill a lot of people to get that

  information, it's okay.  That's wrong.  They fixed it

  in 2000.  It now lumps those two together, and it

  basically says if no harm will come to people -- you've

  seen this already, so I won't repeat it -- if no

  serious harm will come to people, and if it is

  methodologically necessary, placebo control trials are

  okay.  That's what it says as of 2008, and I think

  there is pretty general agreement that that's okay.

            ICH E-10 in 2000, in my view, got it quite

  right, and it said the principal issue in use of

  placebos is the ethical one.  There is no issue when

  there is no effective therapy.  The question is when is

  it acceptable not to give existing therapy to people in

  randomized -- with a drug or placebo.  And I am -- I

  underlined "available," because that's what it said.

  In cases where available therapy is known to prevent

  severe harm, you cannot use a placebo.

            The "generally," by the way, was a hedge

  word -- I remember it, because I threw it

  in -- referred to cases where the treatment is so toxic

  that people won't take it.  So low-dose AZT was tested

  against placebo, because nobody would take the high

  dose.  So that was the only exception.  It wasn't meant

  to hedge too much.

            In other situations where there is no serious

  harm -- and, you know, Zeke Emanuel says, "Well,

  serious harm, irreversible harm, those are all

  ambiguous" -- I don't think it's that ambiguous.  If

  it's really bad for people not to get an available

  drug, you mustn't not give them a placebo.  And there

  could be arguments about how much vomiting is

  unacceptable.  We all agree.  ICH E-10 says that.

            Where there is no serious harm it is generally

  considered ethical to ask patients to participate in a

  placebo-controlled trial, even if they may be

  uncomfortable.  You have to give them adequate

  information, they can't be coerced.

            One could raise debates about how much money

  is coercive -- perfectly good question.  And it points

  out whether a particular placebo-controlled trial will

  be acceptable to subjects, investigators -- could be

  debated.  And one country might conclude one thing

  about highly emetogenic chemotherapy, and another place

  might think something else.  And, of course, the

  patient is supposed to be informed enough to know.

            This is now more or less what the Declaration

  says.  ICH E-10 refers to available therapy.  It is no

  accident they didn't address the question of whether

  available meant in that country, or all over the world,

  because that was just too hard.  They didn't want to get

  in -- we didn't want to get in to the best local versus

  best global.

            So, under ICH E-10,however, this whole issue

  is only of interest when you're talking about a

  treatment that prevents serious harm, because those are

  the treatments that would ordinarily have to be given,

  if they were available.  Symptomatic treatments are

  probably not an issue, and I don't think there is much

  controversy on that.

            So, suppose the treatment really is

  important -- and we actually discussed this at a WMA

  meeting, and I would say there was a fair amount of

  agreement on what I am about to tell you, but we'll see

  what you think.  Suppose a clinically-important

  treatment is not available in a developing country.

  That actually can happen in a developed country, where

  a country chooses not to approve a treatment.  That has

  happened from time to time.

            And -- but let's also suppose -- you have to

  stipulate this -- that a comparison study comparing

  some new treatment with the established treatment would

  not be informed, that this is a case where you have to

  have a placebo to have it be interpretable.  And that

  could be debated, too, but the issue doesn't arise in

  that case.  If a non-inferiority study is available,

  you just do the non-inferiority study.

            So, can you study a new drug with same

  therapeutic goal as existing therapy in a developing

  country, where the standard of care is not available?

            And I think there are two distinct cases.  One

  is the one Ruth referred to, where the trial clearly

  serves the interest of the country, because they can't

  deliver the standard therapy, they don't have the

  clinics, they don't have the techniques.  The other is

  where the trial is being done solely because of a

  commercial interest.  Someone wants to make use of the

  fact that the drug is not available in that country to

  do a placebo controlled trial that would plainly not be

  acceptable in his own country.  And you can get into

  whether he plans to market the drug in the other

  country, and all that stuff.

            So, let's consider those two issues.  The

  general view at the WMA -- and I think Ruth referred to

  this as the things that she worried about being

  excluded -- was if you could say that the country

  needed the information, that it was important to them,

  and they needed to know the results -- and remember, it

  couldn't be obtained with an active control trial -- it

  was okay to do the trial.

            The famous HIV transmission study which got a

  lot of discussion is a good example of that.  If it's

  true that a non-inferiority study would not have been

  informative and, as Ruth and I were just discussing,

  that a historical controlled experience wouldn't have

  been informative, the only way to find out whether

  short-course HIV was effective was to do the placebo

  controlled study.  And that's why Barnes and Thatcher

  wrote that it was okay.

            The disagreement was largely from Wolfe &

  Lurie, because they thought an active control trial

  would have been acceptable, and I don't agree with

  that.  But that's not the philosophical argument.

            Another example, we've been talking about the

  use of rectal artesunate, where it really would have

  been better to just give, you know, IV quinine or

  something.  And, in general, this refers to any therapy

  really needed by the county that they have good reason

  to get.

            So, the thought there -- I think the general

  agreement was that it would be acceptable not to give

  the therapy that was available in the advanced country,

  because you had no choice; it couldn't be given.

            Briefly take the other --

            DR. GUTMANN:  Robert, I'm going to just ask

  you to try to wind up.  I need to give --

            DR. TEMPLE:  I can do that.

            DR. GUTMANN:  You will have a chance to answer


            DR. TEMPLE:  Okay.  The other case is the

  famous Surfactin case, which I won't dwell on, where

  the purpose of the trial, plainly, was to get it

  marketed in the U.S.  There was good reason to believe

  that nobody in the trial would be better off -- would

  be worse off than they would have been without the

  trial, but they clearly would not have gotten standard

  therapy.  Most people did not feel comfortable with

  that approach.  It was not for the purpose of a

  country, it was so that somebody could get the data,

  and to go market the drug in the United States.

            I just want to pose the question that, having

  abandoned that trial, which the company did under

  publicity, it's very clear that more babies died in

  Latin America than would have died if the trial had

  happened.  Now, nobody seems to be bothered by that,

  but it seems worth thinking about.  Because they

  weren't getting the drug.

            And I think that may be my end.  Okay.  We

  sort of addressed this.  Okay.

            DR. GUTMANN:  Thank you very much.  This is an

  active issue for us to consider.  And I want to open it

  up to commission members to make comments or ask


            (No response.)

            DR. GUTMANN:  I will begin, if nobody -- let

  me just ask both of you, because you have, no doubt,

  read the sounding board response, "The Ethics of

  Placebo-Controlled Trials:  A Middle Ground," that was

  in the New England Journal, co-authored by Ezekiel

  Emanuel and Franklin Miller.  It would be helpful for

  us if we could hear your response.

            Robert, to you, the authors say psychological

  and social harm caused by depression, for example, are

  either not considered or dismissed.  In other words,

  when -- you began to address this, but what

  actually -- how broad are you -- do you want us to

  interpret harm?

            And let me just -- I will just pose the

  complementary  -- there are many questions here, but

  this suggests a middle ground, just for us to

  understand how far apart you really are.

            Ruth, the co-authors say that the dichotomy

  you've proposed between rigorous science and ethical

  protections is false; scientific validity constitutes a

  fundamental ethical protection.  I'm sure you would

  agree with that.  But here is the -- if placebo

  controls are necessary or desirable for scientific

  reasons, that constitutes an ethical reason to use

  them.  Although it may not be a sufficient reason, it

  is a reason.

            So, if you -- if I start, either one of you,

  if you could, just say where you see the common ground

  lacking here, because it does seem like once you take

  into account the need for scientific validity, which is

  an ethical consideration, we believe, and once you take

  into account a range of harms that would be serious

  harms, there is not -- it is at least hard for me to

  see why we can't agree on this middle ground from both

  your perspectives, unless you really are orthodox, as

  orthodox as this suggests you are.

            You want to start, Ruth, and then we will do


            DR. MACKLIN:  I should say that where I stand

  with regard to that claim and the article, I am not a

  defender of active control orthodoxy.

            DR. GUTMANN:  Okay.

            DR. MACKLIN:  The middle ground is between

  placebo control orthodoxy and active control orthodoxy.

            DR. GUTMANN:  Correct, correct.

            DR. MACKLIN:  I accept the middle ground.  I

  accept the arguments.  And I also accept the examples

  that Bob gives of the circumstances in which -- which

  he refers to as the symptomatic cases, in which you

  really can't tell whether it's the disease, the

  remission, the -- I accept all of that.  So -- and the

  middle ground actually tries to work between those two,

  and --

            DR. GUTMANN:  Right.

            DR. MACKLIN:  -- that's the middle ground.

  So --

            DR. GUTMANN:  Correct.

            DR. MACKLIN:  -- that's where I stand on that


            DR. GUTMANN:  Okay, okay.

            DR. MACKLIN:  Where I -- let me say the other

  thing is I think there may not be agreement on what are

  the scientifically and methodologically compelling

  reasons to use placebo.  And so the challenge there is

  to see whether --

            DR. GUTMANN:  Okay.

            DR. MACKLIN:  -- there are other

  methodologists who might challenge Bob's view on when a

  trial might be uninformative in the non-symptomatic


            DR. GUTMANN:  Right, fair enough.  Very good.

  That's very helpful to us.  Robert?

            DR. TEMPLE:  There might be somebody who would

  do that, but we haven't been exposed to them.  I think

  there is broad agreement that if, under the

  circumstances, we say an active control is not

  informative -- I mean we just wrote a long guidance on

  this, 40 pages of stuff to people to disagree with if

  they want to.  We have not had disagreements with the

  fundamental principle; we have had disagreements about

  some of the edges.

            I didn't feel that Zeke described a middle

  ground at all.  I think he described what exactly our

  position was.  If you really believe that -- yeah, we

   -- Susan and I talked about that considerably.  We

  didn't think -- it reads like E-10 to me.  If you

  really believe that not getting an anti-depressant is

  going to be dangerous, if you have evidence to that,

  then you can't study new antidepressants.  Put it away.

  We totally agree.  The question is whether that is


            And if you look at, you know, thousands of

  depression trials, you won't see an increased suicide

  rate in the people who didn't get treated; you actually

  see increased suicidality in the people who are

  treated -- he didn't know that at the time he wrote

  that; that's more recent news -- but if there is a

  risk, even a low risk of something bad happening, then

  you can't do it.  We agree with that.

            His example of severely emetogenic

  chemotherapy, we have always been troubled by that and

  thought that was at the edges.  If it keeps you from

  getting your chemotherapy, you have to either not do

  it, design around it, or something like that.

            And there are ways to make the period during

  which a person is miserable and suffering shorter.  You

  can have your end point be the first sign of something,

  instead of, you know, having to vomit for hours and

  hours.  You don't have to do that.

            In the case of severely emetogenic

  chemotherapy, actually, I'm quite sure you could do an

  active control trial.  But don't mistake that for

  thinking that all studies work -- there has been a

  review of studies of Ondansetron, which is the way you

  prevent emesis.  And in dozens and dozens and dozens of studies,

  it failed to be placebo when it was being used

  post-surgically.  And the reason was simple.  The

  people didn't vomit, so you couldn't show an advantage.

            So, in emetogenic chemotherapy, you probably

  could use a non-inferiority study.  There is a

  legitimate question about whether people should be

  exposed to this kind of misery.  That's a perfectly

  good question.  There is -- we are -- I don't know what

  the orthodoxy is.  Everybody acknowledges that, so --

            DR. GUTMANN:  We are very pleased to --

            DR. TEMPLE:  I am comfortable with his


            DR. GUTMANN:  We are very pleased to hear

  that.  We are not going to settle every scientific

  dispute about particular studies, but if there is a

  ground that the two of you agree on, that is an

  important step forward.

            So, let me call on Christine Grady.

            DR. TEMPLE:  I'll say one thing.  I don't

  think Ruth and I differ on very much.  Probably the one

  thing is the Surfactin case, where I'm not so sure that

  is really wrong, and that deserves discussion.  But I

  realize most people don't agree with me.  So I know


            DR. GUTMANN:  Yeah.  But there will be -- and

  we could all agree on all the standards at a level of

  specificity, and still disagree on particular cases

  because of specifying the facts and the


            So that is extremely important, what the two

  of you have just said in response to my question.  Let

  me go on.  Christine?

            DR. GRADY:  Thank you both.  I think my

  question follows a little bit on Amy's, in a certain

  way, although I was thinking of it in a different way

  before she started.

            You both have been through the war, so to

  speak, in terms of this controversy.  And it sounded

  like, from each of your presentations, that

  with -- now, many years after the debates began, and

  the changes to Helsinki, and the E-10, and the

  literature that has been available to sort of carve

  middle grounds, that actually, the disagreement between

  you is hardly there at all.

            And so, I guess the question that I would have

  is, at this juncture in history, what do you think

  needs to be done to sort of put this question to be, so

  to speak?  I mean what's still out there that is

  troubling people, in terms of this question about

  placebo control, or standard of care?

            (No response.)

            DR. GRADY:  Or do you think we should put our

  energy someone else?  I guess that's part of the


            DR. MACKLIN:  Well, I do think we should put

  our energy someone else.  But I think it doesn't help

  to revisit the 076, that controversy, because people

  are entrenched, and don't like to give up their

  positions, whatever position they took.  So that's not

  helpful.  New examples --

            DR. GUTMANN:  It's not helpful to go back to

  those people who are entrenched, but that, if you ask a

  lot of educated people who are following this debate,

  that's where a lot of people think this is in this

  debate, not where the two of you are now.

            DR. MACKLIN:  Well, I think what we have

  to -- let me -- in order to answer the question, what

  is the remaining disagreement between Bob and me?

            Now, if it comes down to the Surfactin trial,

  there was a -- later, an active control Surfactin trial

  that was done, not a placebo-controlled trial.  This is

  not one of those cases of the symptomatic trial, of the

  symptomatic situation.

            So, the defense that Bob gave was fewer

  babies -- yeah, some babies died that wouldn't

  otherwise have died if the trial had been done.  But

  the question that arises is, "Why wasn't an active

  control trial done?  Why could it not have been done?

  Would it not have yielded scientific information?"  And

  that's what I didn't hear.  That, it seems to me --

            DR. GUTMANN:  So that's -- we're not going to

  spend all our time on Surfactin --

            DR. TEMPLE:  No, but I will tell you we

  concluded -- and whether this is right, in some sense,

  doesn't matter -- but we concluded that showing

  equivalence to the existing animal -- bovine-derived

  surfactant would not be persuasive, because some of the

  trials of the initial surfactant had not been

  successful, even though we all know it's a wonderful

  therapy, and it saves babies' lives.

            What they eventually did was beat a synthetic

  surfactant, which probably isn't as good as the actual

  surfactant.  So they were able to do a superiority

  trial.  A superiority trial would always have been

  acceptable.  No one would have ever doubted that.  But

  a non-inferiority trial to the bovine surfactant would

  not have been interpretable.

            Again, I am not all that knowledgeable about

  the trial, so I don't know if that's true.  All I would

  say, though, is that if that is true, then you needed

  to do a trial, show a difference.

            By the way, they would never have done a

  superiority trial in the Latin American countries, they

  would have done it in the U.S.  So that trial would

  never have gone there, because, among other things, we

  would have been less certain about the applicability to

  the U.S.  We worry about that.

            So, it really is a case where the trial they

  wanted to do could not have been done in the U.S.

  Nobody disagrees about that.  And it really focuses on

  whether -- it really -- sort of depends on whether you

  focus on the people in the trial, and how they are, or

  how you feel about what it tells you about the world,

  that you can do that trial there.

            And I -- you know, I am into social justice,

  but you've got to think about the people in the trial,

  too.  I think that's the tension.

            DR. GUTMANN:  John?

            DR. ARRAS:  Okay.  So, Bob, I am detecting a

  bit of a tension within your presentation, okay?  So,

  earlier in your presentation, you did make a

  distinction that I think you thought was important

  between studies that were done for the benefit of the

  local population versus studies that are done simply

  to, you know, get approval back here and market the

  drug here.  Right?

            So -- but then, near the end -- and just now

   -- you seemed to float a very different sort of

  standard, which is what you might want to call a kind

  of Pareto standard, you know.  If people are made

  better off, and nobody is made worse off, then it's

  okay to do the study.

            So, there is, I think, a tension between those

  two sorts of standards, because you could say, well,

  the fact that some people will benefit, you know, if

  the trial is done, that's not a decisive reason if that

  first principle is really paramount:  namely, needed

  within the local community, needed within that

  country's public health structure.

            So, could you discuss that a little bit?  I

  mean how do you reconcile those two?  Because it could

  be that you could just go with the second, right?

            DR. TEMPLE:  No.

            DR. ARRAS:  Okay.

            DR. TEMPLE:  The distinction is between what

  the consequence of not getting the standard therapy is.

  If it's a symptomatic condition, where the only

  consequence of not getting the standard therapy is

  you're symptomatic, I think that's okay anywhere you

  want to do it.

            And, for what it's worth -- I didn't mention

  this -- the sorts of trials that people do of

  depression in Eastern Europe and elsewhere are the very

  same trials they do at home.  So, everybody is treated

  the same, and nobody comes to harm.  I have no trouble

  with any of those, wherever you do them.

            There are people who are uncomfortable about

  how many trials are being farmed out to poorer parts of

  the world.  Separate question.  I have not been

  addressing that.

            This arises where there is clearly harm from

  not getting the standard of care.  And there I want to

  make the distinction between -- this isn't necessarily

  my position, but everyone is quite comfortable with the

  idea that where doing the trial and not giving the

  standard therapy is very much in the interest of the

  country, because they will get a treatment for -- to

  prevent HIV transmission that ain't the best, but it's

  pretty good, or they'll treat malaria better than they

  otherwise could have, even if it's not the best,

  everybody, I thought, at WMA was very comfortable with


            What they're not comfortable with is my

  thought that maybe it's okay to do a trial if everybody

  in the trial is at least as well off as they otherwise

  would be.  People, in general, are not comfortable with

  doing that in, say, Latin America for -- in order to

  market a drug in the U.S.  Okay?

            And I think it's the social injustice.  They

  don't like the fact that you would have to do a trial

  like -- that you could even get to do a trial like that

  in those countries.  They resent the disparity of

  wealth, and things like that, all of which are

  perfectly legitimate.  I just think it's also important

  to think about the people in the trial, because they

  would have been better off.

            DR. ARRAS:  Well, yeah.  And if you were a

  parent of a child with that kind of lung condition, you

  probably would want your child in that study.

            DR. TEMPLE:  I think you would.  And they did.

  That's why they wanted to do the trial.

            DR. GUTMANN:  Christine?  Quickly, because we

  have a hard stop.

            DR. GRADY:  I just wanted to follow up --

            DR. GUTMANN:  But I know Christine wanted to

  follow up.

            DR. GRADY:  -- because it seemed like the

  disagreement about Surfactin has to do more with what

  some people call responsiveness or, you know, local

  needs, than it does with the design of the trial, the


            So, is that right?  I mean is the placebo

  question settled, and we have to worry about these

  other things, like whether or not responsiveness

  matters?  Or do we still have work to do on

  what -- agreeing on when placebo is okay and not okay?

            DR. TEMPLE:  See, I think to approach the

  ethical issue you should assume that you, in fact, did

  need to do a placebo-controlled trial to get

  information.  I mean if you could do an equivalence

  trial, then the issue doesn't arise.  If you could find

  a loser surfactant to beat, then the issue doesn't

  arise, although you wouldn't do it in Latin America.

            The issue arises where, for one reason or

  another -- let's assume it's true -- you really can't

  do an equivalence trial, and no one will let you do a

  placebo-controlled trial where the drug is available,

  because harm would come to people.  You know, there is

  a million drugs --

            PARTICIPANT:  Is there an example of that?

            DR. TEMPLE:  Oh, yeah.  Suppose I want to know

  whether a new -- okay, ACE inhibitors prevent death in

  heart failure.  Okay.  How do I get a new ACE

  inhibitor?  How do I get that claim, if I'm a new ACE


            Well, I can't do an equivalence trial for a

  variety of reasons, and the main reason is therapy has

  marched on and now everybody gets a beta blocker and

  Spironolactone, in addition to the ACE inhibitor.  So I

  don't know what the effect of the ACE inhibitor is.  I

  can't do the non-inferiority margin.  So I can't do

  that.  But I could go to a country where there are no

  ACE inhibitors, and show that the ACE inhibitor works,

  compared to placebo.

            Is that a good idea?  Well, maybe if everybody

  is better, and everybody gets a Spironolactone and a

  beta blocker and a diuretic, and they're better off

  than they currently are, and you compare added ACE

  inhibitor with placebo, I'm not sure I'd object to that


            But that's where the issue -- there is a

  million things you can't study any more, because an

  equivalence trial is uninformative, and you can't

  deprive people of a known therapy that saves your life.

            DR. GUTMANN:  This has been enormously helpful

  to us, and I think is an example of how bringing two

  people who have sparred, but are very thoughtful and

  responsive, can make a difference.

            So, on behalf of the whole commission and

  everybody present, we thank you very much.



This is a work of the U.S. Government and is not subject to copyright protection in the United States. Foreign copyrights may apply.