The University Record, March 11, 1998

Placebo-controlled trials on human subjects in developing countries draw criticism

Editor's Note: Peter Lurie, assistant professor in the departments of family and community medicine, epidemiology and biostatistics, and the Institute for Health Policy Studies at University of California, San Francisco, is currently a visiting assistant professor in the Residential College and visiting assistant research scientist, Institute for Social Research. On Feb. 26, he gave a talk on his research as a special topic in the Research Responsibility Program sponsored by the Office of the Vice President for Research.

By Paula Saha

While at a professional conference in Cote d'Ivoire, Peter Lurie came across a group of placebo-controlled drug trials that struck him as alarmingly unethical.

Researchers were looking for a shorter AZT regimen for HIV-infected pregnant women that would be the most effective in reducing the chance of transferring the infection to their baby.

Sound harmless enough? Lurie's problem with the studies was that most of them compared a short regimen of AZT to a placebo group--essentially, no treatment at all--despite the proven efficacy of a longer regimen of treatment.

As Lurie saw it, women were being denied available, effective treatment, and scientists were contributing to the needless spread of HIV infections.

The question, says Lurie, is not whether the shorter regimen is better than nothing, but whether there is a cheaper, shorter regimen that would retain most or all of the efficacy of the 076 regimen.

"If you're not sure if the short version is as good as the long," says he, "it seems to me that you would go ahead and compare those two."

In November 1997, Lurie and Sidney Wolfe published an article in the New England Journal of Medicine criticizing a number of Centers for Disease Control and Prevention (CDC) and National Institutes of Health (NIH)-funded drug trials to reduce the risk of mother-to-infant HIV transmission in the developing world. Their research, done with the Ralph Nader-funded Public Citizens Health Research Group, spawned a nationwide debate on the ethics of placebo-controlled drug trials on human subjects in the developing world.

The Problem

"Transmission of HIV from mother to infant is widely acknowledged as a major public health problem," Lurie says. "It is estimated that by the year 2000, there will be six million HIV-positive pregnant women in the developing world."

Approximately one-quarter of these women, says Lurie, will unwittingly pass the infection on to their infants, resulting in millions of new HIV infections.

Though in smaller numbers, the same problem of mother-to-infant HIV transmission exists in the developed world. The number, already far less than that in developing countries, is rapidly falling, thanks to ACTG (AIDS Clinical Trial Group) 076. The regimen consists of the administration of AZT, a well-known antiviral drug effective against HIV, to the pregnant mother for 28 weeks during her pregnancy, intravenous administration during delivery, and six weeks of AZT to the baby after birth.

ACTG 076 costs $800-$1,000 per person--too costly in the developing countries that have the most cases of HIV.

This difficulty prompted researchers to launch additional studies to find a shorter, yet effective AZT regimen in various countries including, Uganda, Thailand and Cote d'Ivoire. Of the 16 studies launched, 15 of them were placebo-controlled.

The Violations

The placebo-controlled studies, Lurie charges, violate "a number of core principles in medical ethics," including the 1947 Nuremberg code that was developed after Nazi experiments during WWII, and the World Health Organization's Declaration of Helsinki, widely regarded as the fundamental guiding principles of research involving human subjects. The Declaration states that in any medical experiment, every patient including those of the control group should be assured of the best proven diagnostical therapy. "It's clear that this is for physicians all over the world, not just Western physicians," Lurie says.

The Defense

Lurie's opponents and supporters of placebo-controlled trials in developing countries claim their research is ethical for a number of reasons.

AZT may not work in developing countries

This defense implies that the study should be repeated in a variety of developing countries with the assumption that results will be different. "I think that that is sometimes true," says Lurie. "But there is no reason to believe AZT would be more or less effective in developing countries--it's not as if when ibuprofen comes out in this country, people say let's go repeat the study in Cote d'Ivoire."

The one difference between developing countries and the United States which may make a difference is that while the United States recommends HIV positive mothers not breastfeed, while developing countries recommend that they do, believing the chance of risk of HIV transmission is less than the risk of other infections. Lurie contends, however, the placebo-controlled studies as they exist now are not addressing this difference. In addition, there is no reason to believe that the 076 regimen would still be a more effective treatment on breastfeeding mothers than no treatment at all.

Placebo is standard of care in developing countries

Researchers say that if the study was not being conducted, the HIV positive pregnant women would get no treatment at all. This way, they say, those in the placebo group are no worse off than they otherwise would be, and those in treatment group are better off.

"Those two statements are true," concedes Lurie. However, "whether they justify conducting placebo-controlled trials is another matter altogether."

Lurie believes that using the standard of care argument is taking advantage of a developing country's social predicament.

Lurie argues that "if it is true that socially deprived people in developing countries can receive nothing because that's the standard of care, then it must be true that if there are groups of people in our country that don't get adequate medical or social services, that you could just use them, you could make the same argument and conduct otherwise unacceptable experiments on them.

"If we make the standard of care argument," warns Lurie, "it will come home to roost."

Placebo-controlled trials are more rapid

Proponents of placebo-controlled trials argue that more subjects are needed for equivalency trials. Lurie and several others have said, if that's the problem, researchers should get more subjects.

However, Lurie does not necessarily concede that researchers need more subjects for an equivalency trial. "An equivalency study asks a different question," he says. "You want to see if one regimen is more effective than another, so researchers must establish a tolerance."

Informed consent was obtained

"Certainly if there's no informed consent, the study is unethical," agrees Lurie. "But if there is informed consent it doesn't mean that the study is ethical. If the study is designed in a way that is inherently unethical, you can get all the informed consent in the world and it won't make it better."

Lurie also has some doubt about the validity of the informed consent that was obtained. He points to an early October article in the New York Times that quoted women who were part of the CDC trial in Cote d'Ivoire:

"They gave me a bunch of pills to take, told me to take them. I figured, if one of them didn't work against AIDS, then one of the other ones would," said one woman. In another case, informed consent was obtained from a woman within five minutes of being told she was HIV positive.

"There are real questions about whether the informed consent was any good," Lurie maintains.

Institutional Review Boards (IRB) approved the studies

Just as with informed consent, Lurie believes "it is important studies be approved by the IRB, but that does not make a study ethical. Lots of unethical studies have been approved by the IRB."

As it happens, Lurie believes there are serious questions about the actual approval by the IRBs in developing countries. Lurie and Wolfe found that CDC's records on the Cote d'Ivoire IRB review shows on the day the study was approved, of the nine committee members, only two were present to vote. In a similar study conducted by the NIH in Ethiopia, a new IRB, one that had never before formally voted on any study, approved the study by only a 6-5 margin. In Bukina Faso, there is no ethics commission--the Ministry of Health decides which studies are ethical.