Request to the Department of Health and Human Services to halt plans for unethical placebo-controlled study of drug for respiratory distress syndrome in Latin America unless it is redesigned to treat all patients (HRG Publication #1558)
Note: You will need Adobe Acrobat Reader to open and read an attachment noted below. If you do not have Acrobat Reader, you can access this free software by clicking on the graphic. You will only have to complete this process once.
February 22, 2001
U.S. Department of Health and Human Services
200 Independence Avenue, S.W.
Washington, D.C. 20201
Fax: (202) 690-7203
Dear Secretary Thompson:
We have learned that the Food and Drug Administration (FDA) is seriously considering endorsing the design of a drug-company study of a new surfactant drug in four Latin American countries in which a control group of 325 premature newborn infants with potentially fatal Respiratory Distress Syndrome (RDS) would be treated with placebos instead of a lifesaving and already FDA-approved surfactant drug. In a clear demonstration of the exploitive nature of this proposed research, the same manufacturer that is seeking approval for this drug plans a study in Europe in which all infants not receiving the new surfactant drug will receive an already FDA-approved surfactant drug. Internal FDA documents state that “Conduct of a placebo controlled surfactant trial for premature infants with RDS is considered unethical in the USA.”
Surfactants like the drug to be tested here have been shown to reduce neonatal mortality by 34%(1) and the FDA estimates the neonatal (28-day) mortality rate among premature infants in these countries to be at least 30%. If half of the infant deaths were due to RDS (this was the case in the U.S. in the pre-surfactant era)(2) the provision of placebo (instead of another form of surfactant) to the 325 infants in the control group will result in the preventable deaths of 17 infants (325 X 0.3 X 0.5 X 0.34).
Particularly because the FDA approved another surfactant (Infasurf) in 1998 on the basis of studies performed between 1991 and 1993 in which all infants were treated with a presumably effective drug and none were given placebo, we call on you to immediately put a stop to plans for this unethical and exploitive study in its present design. The study is unethical because it violates the principle that placebos not be employed when there exists a standard treatment that may reduce or prevent harm, improve health or prolong life. If the study produces findings that are beneficial to patients in wealthier countries but the drug is not widely available in the countries in which it was tested, an additional dimension of unethical behavior will have been added. For the study to take place ethically, all infants must be provided with a treatment either known or expected to be effective; a comparison of the new surfactant to an already approved one would therefore be acceptable.
The information on the trial we present here is based on documents (see attached) made available to hundreds of FDA employees in conjunction with an internal FDA Scientific Rounds on January 24, 2001. The meeting had the extraordinarily inappropriate, but revealing, title “Use of placebo-controls in life threatening diseases: is the developing world the answer?” As noted, information presented by the FDA at the meeting acknowledges that “Conduct of a placebo controlled surfactant trial for premature infants with RDS is considered unethical in the USA.” Moreover, in the countries where the trial would take place (Bolivia, Ecuador, Peru and Mexico), surfactants are used in some hospitals, but “surfactants are completely unavailable to infants at many other hospitals, secondary to rationing or economic limitations,” the FDA documents say. It is in these latter hospitals that the studies are planned. The cost of a course of surfactant for a 1500g premature infant is estimated to range between $1079 and $2440 (Average Wholesale Prices; personal communication, Tony Mavrantonis, Medical Economics, February 15, 2001). Because the drug is still an Investigational New Drug, neither the name of the product nor that of the manufacturer is available to us.
Respiratory Distress Syndrome
RDS is a common condition in premature infants. In 1998, 1328 U.S. infants died of RDS, making it the fourth largest cause of infant mortality.(3) Historically, approximately 50% of all neonatal deaths resulted from RDS,2 although with the advent of surfactant this percentage has decreased greatly. However, the 50% figure is probably still a reasonable estimate for areas without access to surfactant or neonatal intensive care. Annual incidence is estimated at 60,000-70,000 cases in the U.S.(4) The condition occurs in 60%-80% of infants under 28 weeks gestation and in 15%-30% of those born between 32 and 36 weeks.2
In a major scientific breakthrough in the 1960s, scientists discovered that one key cause of RDS is a deficiency or dysfunction of a naturally occurring chemical called surfactant. Surfactant reduces surface tension in alveoli (small air sacs) in the lung. In the absence of functional surfactant, markedly increased pressures in the alveoli are required to inflate the lungs, preventing proper aeration. In the 1970s and 1980s, scientists worked to produce a safe and effective surfactant that could be instilled in the lungs of premature infants to reverse the devastating course of RDS. The first surfactant (Exosurf) was approved in the U.S. in 1990, the second (Survanta) in 1991, the third (Infasurf) in 1998 and the fourth (Curosurf) in 1999. Exosurf is a synthetic compound, Survanta and Infasurf are isolated from cow lungs and Curosurf is isolated from pig lungs. All are administered while the infant is receiving advanced life support; the liquid surfactant is instilled down the endotracheal tube, generally with the infant sequentially in four different positions to maximize dispersal of the liquid. Infants often respond within minutes. Surfactant may be administered shortly after birth, before the infant develops respiratory distress; this is known as prophylactic therapy. Alternatively, the clinician may wait to see if respiratory distress develops and, if it does, administer surfactant; this is known as treatment or rescue therapy.
Clear Evidence that Surfactant Saves Newborn Lives
Multiple clinical trials attest to the efficacy of both synthetic and natural surfactants in reducing neonatal mortality. For this reason, surfactant was described in an article in the New England Journal of Medicine as long ago as 1993 as “without doubt the most thoroughly studied new therapy in neonatal care” and as “a major advance in neonatal care.”(5) By then, there had been 35 randomized, controlled trials (most with placebo) which together demonstrated efficacy in reducing neonatal mortality for both natural and synthetic surfactants for both prophylaxis and treatment of RDS.
To assess the current state of knowledge regarding surfactant efficacy, we consulted the Cochrane Library, an ongoing international collaborative effort among scientists to collect and synthesize the results of all clinical trials for a large number of conditions using a statistical technique called meta-analysis. There have been seven Cochrane reviews involving surfactant, each including about seven separate clinical trials. There is some overlap between the reviews. These include studies of prophylaxis and treatment of RDS as well as studies of natural and synthetic surfactants. Because the surfactant to be tested in the proposed study is a synthetic surfactant and because the FDA documents indicate that this is a treatment trial, we have paid most attention to the review comparing synthetic surfactant to placebo for the treatment of RDS.1 (In these studies, “placebo” generally means the administration of “sham air” rather than surfactant down the endotracheal tube.) However, the results of prophylactic trials with synthetic surfactants are generally similar.(6) There is no Cochrane review of natural surfactant for treatment of RDS, although an earlier meta-analysis by the same Cochrane author concludes that “mortality is reduced in infants with RDS who were treated with [natural surfactant].”(7)
Six placebo-controlled studies of synthetic surfactant in the treatment of RDS in premature infants were deemed to be adequate for inclusion in the meta-analysis by the Cochrane author; all but the smallest of these involved Exosurf. The Table summarizes the results of the six studies. The studies evaluated a number of different patient outcome variables, the seven most important of which appear in the Table. Not all studies evaluated all outcomes. For each of these outcomes, at least one (and often more) of the studies reached statistical significance (surfactant was significantly better than placebo), although different outcomes were statistically significant in different studies.
The probability of finding a statistically significant result for a given outcome was strongly related to sample size. The largest study(8),(9),(10) had a total of 1237 patients in two arms and five out of seven outcomes showed statistically significant beneficial effects for surfactant. The smallest study(11) included only 24 patients and, not surprisingly, showed no statistically significant finding for any of the three outcome variables it evaluated. Importantly, all but the largest study are smaller than the study proposed here (and so “false-negative” results are less likely); the five smaller studies account for 22 of the 24 non-statistically significant findings. The failure of small studies to show statistically significant findings that would be apparent if all studies were combined is the very rationale for meta-analysis.
|Number of Studies Evaluating this Outcome||Number of Studies with Statistically Significant Protective Effect||Range of Odds Ratios||Combined Odds Ratio||95% Confidence Interval for Combined Odds Ratio|
|Neonatal mortality||6||1||0.60-0.87 ****||0.66||0.52-0.84|
|BPD or death||4||2||0.50-0.69||0.57||0.46-0.71|
|Mortality prior to discharge||6||1||0.47-0.97****||0.71||0.57-0.89|
|Mortality at 1 year||4||1||0.49-0.97||0.73||0.58-0.92|
* Lung collapse
** Pulmonary interstitial emphysema; loss of lung tissue
*** Bronchopulmonary dysplasia; newborn lung disease due to administration of oxygen
**** Excludes one trial with only 12 patients in each study group
We next turn to the fourth column in the Table. Even studies with sample sizes that are too small to detect an effect should estimate the magnitude of the effect approximately correctly, even if the finding does not reach statistical significance. Odds ratios are estimates of the size of an effect; in these studies, an odds ratio less than 1 (but greater than zero) means that surfactant protects the patient from a given outcome better than a placebo; odds ratios greater than one mean that surfactant causes the outcome to happen more often than in the placebo group. Excluding the extremely small trial, every study’s estimate of the odds ratio for every outcome gives an odds ratio less than one, generally in a fairly narrow range. This is important because one justification for conducting a placebo-controlled trial, even when a known effective therapy exists, is that there is variability in the outcomes of the studies. This is not the case here to any significant extent.
The final two columns in the table show the results of the meta-analysis. For each of the seven outcomes evaluated, there is evidence of substantial efficacy and each finding reaches statistical significance. For example, neonatal mortality is reduced 34% by surfactant (relative to a placebo), mortality at one year by 27%, and pneumothorax (collapsed lung) by 43%. These are the benefits being denied the patients in the placebo group. As noted, for the 325 patients in the placebo group, this is equivalent to 17 excess deaths by 28 days.
Placebo Controls no Longer Justified
This Cochrane review concludes that the use of synthetic surfactant for the treatment RDS “has been demonstrated to improve clinical outcome.”1 In a section of the review entitled “Implications for research,” the reviewer states unequivocally: “Further placebo controlled trials of synthetic surfactant are no longer warranted.” (The Cochrane reviewer reaches the same conclusion for prophylactic synthetic surfactant treatment.(6) The American Academy of Pediatrics has also strongly endorsed the use of surfactant for RDS: “Surfactant therapy substantially reduces mortality and respiratory morbidity for this population.”(12)
Not only have other synthetic (and natural7) surfactants proved clearly superior to placebo in trials of RDS, but synthetic and natural treatments have also often been compared to one another in non-placebo-controlled trials for the treatment of RDS.(13) The Cochrane review on this topic identifies seven studies in which two surfactants were compared to one another in premature infants, dating back to 1993. It identifies eight additional active-controlled (no placebo) studies dating back to 1992 which were excluded from the review because they were not randomized and/or did not have clinical outcomes.
As noted above, there are a total of seven Cochrane reviews of surfactant efficacy. If we exclude the two reviews that address only prophylactic surfactant, remove all duplicate studies and restrict ourselves to those clinical trials in premature infants deemed adequate for inclusion in the reviews, there have been a total of nine placebo-controlled trials. The first of these appeared in 1985,11 and all studies until 1991, when the first active-controlled trial was published,(14) were placebo-controlled trials. Between 1991 and 1995 both placebo- and active-controlled trials were conducted. From 1996 onward, there are a total five active-controlled trials in the Cochrane database (there have been 17 since 1991) and not a single placebo-controlled trial in premature infants. At least three additional active-controlled treatment trials (and no placebo-controlled ones) have been conducted in premature infants since the Cochrane databases were last updated.(15),(16),(17) Clearly, the trend in surfactant clinical trials has been toward active-controlled trials. This study, if it ever occurs, will therefore be a landmark of unethical behavior — a turning to the developing world to conduct studies that the FDA admits could never occur here. Again, the title of the FDA session says it all: “Use of placebo-controls in life threatening diseases: is the developing world the answer?”
Moreover, the only two clinical trials supporting the 1998 approval of Infasurf in the label were active-controlled: both the treatment and prophylaxis indications were supported by trials comparing Infasurf to Exosurf. From the product label and the Cochrane review, there is no evidence that a placebo-controlled trial of Infasurf has been done. ***** Clearly, the FDA has accepted active-controlled trials in the past to support approval of a surfactant. Why can it not do so now?
Alternatives to Placebo-Controlled Trials
Active-controlled trials may be divided into two categories: superiority trials (in which the object is to demonstrate that the new therapy is superior to existing therapy) and non-inferiority trials (in which the objective is to prove that the new therapy is not inferior to existing therapies by a pre-specified amount). In this case, the FDA has raised questions about a non-inferiority trial because previous studies are said to have given inconsistent results. However, we have shown above that the results of previous placebo-controlled studies of synthetic surfactant in the treatment of RDS are remarkably consistent. Even if a study does not reach statistical significance for one endpoint, it may well reach it for another, helping to reach the correct conclusion. In any event, the sponsor is planning a non-inferiority trial of the new drug against Curosurf in Europe.
An alternative would be a superiority study, the basis for approval for Exosurf. Apparently such a study in a developed country is not considered feasible by the sponsor due to enrollment difficulties and “ethics,” according to the FDA documents. Presumably this means that developed country patients are not eager to take the chance of being randomly assigned to the new therapy and the developed country ethics committees might object to patients receiving unproven therapies. According to the FDA documents, “The sponsor has not yet provided justification for why they haven’t planned a superiority trial versus Exosurf in underdeveloped Latin American countries.” One reason may be the claim that a superiority study against Exosurf may take longer to conduct for statistical reasons. Another is that the new drug may be no better than Exosurf. With four surfactants already on the U.S. market and the patients at the participating centers in the four Latin American countries currently receiving none of them, why is speed a factor for either U.S. or developing country patients? Clearly, this issue is primarily of concern to drug companies who stand to make more money the sooner the drug is marketed. We take no position on whether the active-controlled trial should be a superiority or a non-inferiority trial; we simply insist that all patients receive active treatment, not a placebo. (We would have had no objection to the use of a placebo arm had there been no known effective treatment for RDS, but this is obviously not the case.)
A Violation of Ethical Principles
The two leading international ethics documents are the Declaration of Helsinki, maintained by the World Medical Association, and the Council for International Organizations of Medical Sciences (CIOMS) International Ethical Guidelines for Biomedical Research Involving Human Subjects. Both seek to provide guidelines for the ethical conduct of research and provide yardsticks against which research conduct can be measured.
Both the CIOMS document and the Declaration of Helsinki assert that the researchers’ responsibility extends beyond the few hospitals where the research is conducted. ****** According to the CIOMS document, “As a general rule, the sponsoring agency should ensure that, at the completion of successful testing, any product developed will be made reasonably available to inhabitants of the underdeveloped community in which the research was carried out.”(18) The Declaration of Helsinki, in a newly added section, states that “Medical research is only justified if there is a reasonable likelihood that the populations in which the research is carried out stand to benefit from the results of the research.”(19)
In the proposed research, however, it is less than clear that many residents of these developing countries will benefit to any significant degree from the research. According to the FDA documents, the sponsor claims it will make the drug available at “reduced cost” to the participating centers in the Latin American countries. (This introduces an element of coercion into the whole enterprise, as a hospital in a poor country will have every incentive to accept a trial in which some of its patients receive a treatment they would not otherwise receive.) What “reduced cost” means is, of course, unclear and there is no indication of the length of this commitment.
The FDA documents also note that the infants in the placebo group “may receive … benefits this population in Latin America would not otherwise have received.” In addition, “neonatology training provided by the sponsor to local health care professionals may improve local standard of care for all babies.” These statements may be factually correct, but they evade the essential ethical responsibility of researchers toward patients. It is clear that the researchers could do more for the participants in this proposed research project than they plan to do. To compensate for the sub-standard care they plan to provide to the placebo patients, they posit other benefits to local residents of countries poorly placed to turn down the possibility of millions of dollars in research funds. In the words of the Declaration of Helsinki, “In medical research on human subjects, considerations related to the well-being of the human subject should take precedence over the interests of science and society.”19 In this case, the well-being of stockholders appears to be taking precedence over the interests of human subjects.
Incredibly, consideration of this unethical study occurs in the aftermath of a lengthy debate over revisions to the Declaration of Helsinki, which resulted in a clear rejection of the ethical position implicitly advanced by the sponsor and potentially endorsed by the FDA. After the controversy over the provision of placebos instead of AZT to HIV-positive pregnant women in trials in Africa and Asia (see below), researchers and government scientists, primarily from the U.S., embarked on a campaign to revise the Declaration of Helsinki. At the time, the Declaration required that “In any medical study, every patient – including those of a control group, if any – should be assured of the best proven diagnostic and therapeutic method.”(20) The researchers argued that this apparently straightforward language was never meant to extend to developing country residents (or, implicitly, poor people in developed countries); researchers were only required to provide what patients would receive in the absence of the trial (the so-called “standard of care argument”). The researchers were faced with a clear choice: having violated the Declaration, they could either bring their behavior into conformity with the Declaration or bring the Declaration into conformity with their unethical behavior. They chose the latter. In an early draft of the revised Declaration, researchers were to be required to provide “the best proven diagnostic, prophylactic or therapeutic method that would otherwise be available to him or her.” (emphasis added) After an international protest against this language, which would have institutionalized the provision of second rate care (or no care at all) to residents of poor countries, the draft was rejected. Indeed, Delon Human, secretary-general of the World Medical Association, said of the draft: “It was very useful … because it defined exactly what we didn’t want.”(21) After more international debate, the members of the World Medical Assembly finally settled on the following language: “The benefits, risks, burdens and effectiveness of a new method should be tested against those of the best current prophylactic, diagnostic, and therapeutic methods.”19 This brief history makes it clear that, while the previous and current Declaration language are quite similar, the fact that the so-called “standard of care argument” was considered and squarely rejected means that the sort of study contemplated here is expressly forbidden by the internationally recognized Declaration.
This study could, however, meet the lower-than-Helsinki standards in the current version of the U.S. National Bioethics Advisory Commission (NBAC) report on international research ethics. Although that report requires that research participants in developing countries be provided with “effective” (as opposed to the preferable “best”) interventions, the NBAC’s draft also permits an exception whereby even the “effective” therapy need not be provided if this was not the “most appropriate study design.”(22) The researchers would have to justify this exception to an ethics committee. The proposed trial would exploit just this exception, illustrating precisely why the NBAC report is so grossly deficient.
A Violation of U.S. Regulations
The proposed research is not only blatantly unethical, but also may also violate U.S. federal regulations which require that ethics review committees called Institutional Review Boards (IRBs) ensure that: “Risks to subjects are minimized … by using procedures which are consistent with sound research design and which do not unnecessarily expose subjects to risk.”(23) (The risks to the infants in the placebo group are not minimized, because the administration of an approved surfactant could reduce their probability of dying by 34%.) The regulation is also clear that it also applies to “research conducted, supported or otherwise subject to regulation by the Federal Government outside the United States.”(24) Even if the sponsor is a foreign drug company, once the study is submitted to the FDA as part of a New Drug Application, it becomes subject to FDA regulation.
FDA Support for Placebo Controls
Nonetheless, FDA officials, particularly Dr. Robert Temple, Director of the FDA’s Office of Medical Policy, have already voiced their hostility to the new Declaration. Dr. Temple has written a number of largely repetitive articles over the years in which he has questioned the usefulness of active-controlled trials.(25),(26),(27),(28),(29),(30) The sponsors of the proposed trial are therefore likely to have been fully aware that they would receive a receptive ear from at least some FDA officials.
It is ironic, therefore, that neither laws nor FDA regulations actually require placebo-controlled trials for drug approval. Rather, the regulations require “adequate and well-controlled studies,” and list five types of acceptable studies: 1. randomized, placebo-controlled trials; 2. dose-response studies; 3. active-controlled studies; 4. no treatment concurrent controlled studies; and 5. historical controls.(31) So new drugs can be approved in the absence of placebo-controlled studies. Indeed, in some divisions of the FDA, active-controlled trials are commonly used as the basis for drug approval. The field of oncology has for years eschewed placebo controls in trials of treatments of cancers for which effective therapy exists.
Comparison to Perinatal HIV Prevention Trials in Africa and Asia
In 1997, we criticized a series of 15 clinical trials in Africa and Thailand, including nine conducted or funded by the U.S. government, in which researchers gave HIV-positive pregnant women in the control arms placebos or drugs not proven to be effective, rather than the proven-effective drug AZT.(32) The trials attempted to identify affordable drug regimens to prevent the transmission of HIV from mother to infant in developing countries.
The proposed trial differs from the controversial perinatal HIV trials in at least three important respects. First, for the perinatal trials, we believed that the single well-conducted, placebo-controlled trial(33) and its subanalysis,(34) combined with evidence that perinatal HIV transmission occurred mostly late in pregnancy(35) was sufficient to make it very likely that short courses of AZT late in pregnancy would prevent HIV transmission (as proved to be the case). In this case, there are literally dozens of clinical trials, including many with placebo, that demonstrate effectiveness. This makes the use of a placebo control all-the-more unacceptable.
Second, whereas in the perinatal HIV trials the researchers were government and academic scientists testing a regimen specifically for use in developing countries, this time the sponsoring institution is a pharmaceutical company motivated by profit. The company’s primary concern appears to be to obtain approval in developed countries, where they are likely to reap by far their greatest sales. Pharmaceutical sales in Latin America represented a mere 7% of international pharmaceutical sales in 1999, compared with 40% in North America and 27% in Europe.(36) These infants are being used by the company for reasons having much more to do with corporate bottom lines than neonatal health in their countries.
Third, the perinatal HIV trial researchers claimed (incorrectly) that it was impossible to deliver the AZT regimen already proved effective due to the lack of intravenous facilities, the late presentation for prenatal care of many pregnant women in developing countries, etc. More clearly than in the perinatal trials, this excuse does not apply here. By definition, this trial will be conducted in fairly sophisticated medical settings, because both the placebo and surfactant patients will have endotracheal tubes and will probably be on ventilators. Since the group randomized to the new surfactant will be administered active treatment by this route, it is obviously quite straightforward to administer surfactant under these conditions. These studies must have budgets in the hundreds of thousands, if not millions, of dollars. The physician-investigators who conduct these studies would clearly be making the choice not to provide these lifesaving therapies.
Moreover, the investigators are claiming that they will be providing benefits to the hospital by training neonatologists and providing the drug to those hospitals after the trial. In other words, they are willing to aid these countries in several different ways — just not the simplest way (treating the control group) which, although sparing many infants’ lives, would preclude the company from doing the study the way it wishes.
In sum, although the lives of fewer infants are at stake here than in the perinatal HIV trials, the proposed study may be guilty of even worse ethical infractions.
Common Justifications for Unethical Research
Researchers commonly seek to justify research of this type by claiming that participants are protected by informed consent, IRBs and Data Safety Monitoring Boards (DSMBs). We agree that all three of these elements should be in place in such trials, but their presence is no guarantee of ethical research. An unethical research study is an unethical research study even with informed consent, IRB approval and a DSMB. Several studies have shown the lack of informed consent in research conducted in developing countries.(37) An adequate informed consent process would have to make reference, at a minimum, to the almost two dozen well-conducted, placebo-controlled studies showing the known effectiveness of other surfactants and to explain why these are not being provided to the infants in the study. Even so, the desperate parents of these infants are likely to assent to the trial, for at least they have a 50% chance of receiving treatment. No doubt they would be at least as likely to sign up if they knew they had a 100% chance of receiving an active treatment.
Approval by an IRB is critical, but even U.S. IRBs have been severely criticized.(38) Recently, the staff of the NBAC floated a proposal to require only local IRB approval for U.S.-funded foreign studies, but a storm of protest resulted in the removal of the proposal because local IRBs are often not experienced enough to adequately review research proposals. More recently, the Washington Post reported that the apparent local IRB approval of a Pfizer-sponsored meningitis study in Nigeria was based on a document back-dated by one year.(39)
It is also useful to have a DSMB in studies of this kind. These independent committees review the data periodically as the trial progresses and may halt a study if there is evidence that a treatment is harming or benefitting its recipients more than those in the comparison arm. Given the strong benefits demonstrated by other surfactants in previous trials, it is quite possible that the DSMB will have to terminate this trial early when the new drug proves better than nothing. By then several infants will have died unnecessarily.
A modern trend in biomedical research is to conduct research in developing countries, rather than in industrialized ones. The number of new foreign investigators in the FDA’s database grew from 988 in the 1990-1992 period to 5,380 in the 1996-1998 period.(40) If the result is therapies for or knowledge relevant to developing country scourges such as malaria or onchocerciasis, this would clearly be a step in the right direction, particularly for pharmaceutical companies that have essentially turned a blind eye on the needs of developing country residents whose purchasing power is limited. But if the consequence is conducting research in developing countries that has little to do with the major public health concerns locally (as is the case here), is conducted in a fashion that would be unacceptable in the sponsoring country or does not result in effective interventions being made available after the trial, we face a future in which studies of the kind described here will become increasingly common. U.S. government endorsement of the exploitive research proposed here would be a major setback for the image of U.S.- sponsored international health research and for relations with these developing countries in general.
This trial is one of the best examples of the race to the ethical bottom that the “standard of care argument” ensures. Unable to conduct a placebo-controlled study in a developed country, or even in the wealthier parts of these developing countries, the researchers have hit upon the idea of experimenting on the poorest of the poor, even as they propose an active-controlled trial in Europe. Such behavior might be expected from a profit-driven drug company. The FDA’s role is to prevent such unethical behavior, not to give it the agency’s stamp of approval.
Peter Lurie, MD, MPH
Sidney M. Wolfe, MD
Public Citizen’s Health Research Group
Marshall Klaus, MD
Adjunct Professor of Pediatrics
University of California, San Francisco
Co-editor, Care of the High-Risk Infant (Saunders)
cc: Greg Koski, PhD, MD, Director, Office for Human Research Protections
***** Some placebo-controlled trials in the Cochrane reviews are of unnamed calf surfactants.
1. Soll RF. Synthetic surfactant for respiratory distress syndrome in preterm infants (Cochrane Review). In The Cochrane Library, Issue 4, 2000. Oxford: Update Software.
2. Behrman RE, Kliegman RM, Nelson WE, Vaughan VC, eds. Nelson Textbook of Pediatrics, 14th edition. W.B. Saunders Company: Philadelphia, 1992, p. 463.
3. Guyer B, Hoyert DL, Martin JA, Ventura SJ, MacDorman MF, Strobino DM. Annual summary of vital statistics — 1998. Pediatrics 1999;104:1229-46.
4. Verma RP. Respiratory distress syndrome of the newborn infant. Obstetrical and Gynecological Survey 1995;50:542-55.
5. Jobe AH. Pulmonary surfactant therapy. New England Journal of Medicine 1993;328:861-8.
6. Soll RF. Prophylactic synthetic surfactant for preventing morbidity and mortality in preterm infants (Cochrane Review). In The Cochrane Library, Issue 4, 2000. Oxford: Update Software.
7. Soll RF, McQueen MC. Respiratory distress syndrome. In: Sinclair JC, Bracken MB, eds. Effective Care of the Newborn Infant. Oxford University Press, Oxford, 1992, p. 342.
8. Courtney SE, Long W, McMillan A, et al. Double-blind 1 year follow-up of 1540 infants with respiratory distress syndrome randomized to rescue treatment with two doses of synthetic surfactant or air in four clinical trials. Journal of Pediatrics 1995;126:543-52.
9. Long W, Corbet A, Cotton R, et al. A controlled trial of synthetic surfactant in infants weighing 1250g or more with respiratory distress syndrome. New England Journal of Medicine 1991;325:1696-703.
10. Sauve R, Long W, Vincer M, et al. Outcome at 1-year adjusted age of 957 infants weighing more than 1250 grams with respiratory distress syndrome randomized to receive synthetic surfactant or air placebo. Journal of Pediatrics 1995;126:575-80.
11. Wilkinson A, Jenkins PA, Jeffrey JA. Two controlled trials of dry artificial surfactant: early effects and later outcome in babies with surfactant deficiency. Lancet 1985;2:287-91.
12. American Academy of Pediatrics. Surfactant replacement therapy for Respiratory Distress Syndrome. Pediatrics 1999;103:684-5.
13. Soll RF. Natural surfactant extract versus synthetic surfactant for neonatal respiratory distress syndrome (Cochrane Review). In: The Cochrane Library, Issue 3, 2000. Oxford: Update Software.
14. Kendig JW, Notter RH, Cox C, et al. A comparison of surfactant as immediate prophylaxis and as rescue therapy in newborns of less than 30 weeks’ gestation. New England Journal of Medicine 1991;324:865-71.
15. Ainsworth SB, Beresford MW, Milligan DW, Shaw NJ, Matthews JN, Fenton AC. Pumactant and poractant alfa for treatment of respiratory distress syndrome in neonates born at 25-29 weeks’ gestation: a randomised trial. Lancet. 2000;355:1387-92.
16. da Costa DE, Pai MG, Al Khusaiby SM. Comparative trial of artificial and natural surfactants in the treatment of respiratory distress syndrome of prematurity: experiences in a developing country. Pediatric Pulmonology 1999;27:312-7.
17. Kattwinkel J, Bloom BT, Delmore P, et al. High-versus low-threshold surfactant retreatment for neonatal respiratory distress syndrome. Pediatrics 2000;106:282-8.
18. Council for International Organizations of Medical Sciences, World Health Organization. International ethical guidelines for biomedical research involving human subjects. Geneva, 1993.
19. World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. Adopted by the 18th World Medical Assembly, Helsinki, 1964 and revised by the 29th World Medical Assembly, Tokyo, 1975, the 35th World Medical Assembly, Venice, 1983, the 41st World Medical Assembly, Hong Kong, 1989, the 48th World Medical Assembly, Somerset West, 1996 and the 52nd World Medical Assembly, Edinburgh, 2000.
20. World Medical Association Declaration of Helsinki: recommendations guiding physicians in biomedical research involving human subjects. Adopted by the 18th World Medical Assembly, Helsinki, 1964 and revised by the 29th World Medical Assembly, Tokyo, 1975, the 35th World Medical Assembly, Venice, 1983 and the 41st World Medical Assembly, Hong Kong, 1989 and the 48th World Medical Assembly, Somerset West, 1996.
21. Enserink M. Helsinki’s new clinical rules: fewer placebos, more disclosure. Science 2000;290:418-9.
22. Meslin EM. International report recommendations. National Bioethics Advisory Commission, December 6, 2000.
23. 45 CFR 46.111(a)(1).
24. 45 CFR 46.101(9).
25. Temple RJ. When are clinical trials of a given agent vs. placebo no longer appropriate or feasible? Controlled Clinical Trials 1997;18:613-20.
26. Temple RJ. Special study designs: early escape, enrichment, studies in non-responders. Communications in Statistics 1994;23:499-531.
27. Temple RJ. Problems in interpreting active control equivalence trials. Accountability in Research 1996;4:267-75.
28. Temple R. Difficulties in evaluating positive control trials. Proceedings of the American Statistical Association, Biopharmaceutical Section. 1983:1-7.
29. Temple R, Ellenberg SS. Placebo-controlled trials and active-control trials in the evaluation of new treatments. Part 1: ethical and scientific issues. Annals of Internal Medicine 2000;133:455-63.
30. Ellenberg SS, Temple R. Placebo-controlled trials and active-control trials in the evaluation of new treatments. Part 2: practical issues and specific cases. Annals of Internal Medicine 2000;133:464-70.
31. 21 CFR 314.126(b)(2) (1991).
32. Lurie P, Wolfe SM. Unethical trials of interventions to reduce perinatal transmission of the human immunodeficiency virus in developing countries. New England Journal of Medicine 1997;337:853-6.
33. Connor EM, Sperling RS, Gelber R, et al. Reduction of maternal-infant transmission of human immunodeficiency virus type I with ziduvudine treatment. New England Journal of Medicine 1994;331:1173-80.
35. Rouzioux C, Costagliola D, Burgard M, et al. Timing of mother-to-child HIV-1 transmission depends on maternal status. AIDS 1993;7(suppl 2):S49-S52.
36. IMS. World-wide Pharmaceutical Market 1999. Available at http://www.ims-global.com/insight/world_in_brief/review99/year.htm.
37. Karim QA, Karim SSA, Coovadia HM, Susser M. Informed consent for HIV testing in a South African hospital: is it truly informed and truly voluntary? American Journal of Public Health 1998;88:637-40.
38. Office of the Inspector General. Institutional Review Boards: a time for reform. Department of Health and Human Services, June 1998.
39. Stephens J. Doctors say drug trial’s approval was backdated. Washington Post, January 16, 2001, pp. A1, A17.
40. Office of the Inspector General. Recruiting human subjects: pressures in industry-sponsored clinical research. Department of Health and Human Services, June 2000.