"Lessons learned" is a popular phrase in international circles. Typically used to sum up the results of a review of experiences, it has a reassuring pedagogic tone. The "students" have done their home- (or field-) work, and have acquired the relevant knowledge. Presumably any mistakes they might have made in the course of their studies will not be repeated, because the "lessons" have been "learned." The real world of international clinical trials differs significantly from the tidy schoolroom metaphor. Nevertheless, it is still useful to ask: Are there lessons to be learned about future trials from the controversy over placebo-controlled trials of zidovudine (AZT) to prevent mother-to-child HIV transmission in developing countries? The answer is that there are some procedural cautions, but few unchallenged "lessons."
The trials involved over 12,000 women in seven countries and were funded by the U.S. Centers for Disease Control and Prevention and the National Institutes of Health. UNAIDS and French agencies were also involved in similar trials. Based on preliminary results of the trials in Thailand, which showed a reduction of 51 percent in transmission in mothers given a short course of AZT compared to those who received a placebo, similar trials under way in Africa and the Dominican Republic were suspended. Both enthusiastic proponents of the trials and their vociferous opponents asserted that their views had been vindicated. The proponents declared that such significant results could only have been achieved with placebo arms; the opponents saw the results as evidence that placebos were never needed in the first place. First Lesson Learned: Even unambiguous scientific results of controversial clinical trials do not necessarily result in ethical consensus.
In retrospect, it is not surprising that these trials became so controversial. They brought together two issues that have been the subject of ethical debate for years: the use of placebo controls in clinical trials and the ethical standards for conducting trials in developing countries. In addition, this trial added the highly charged elements of the rapid spread of HIV in the developing world, the administration of drugs to pregnant women, and the transmission of a fatal disease to newborns. Although the issues had been discussed among the investigators and sponsoring agencies and their immediate circles of advisers, there was no broad discussion in ethics forums or in other arenas where some of the complexities could have been explored without the distraction of media-aimed charges and countercharges. Second Lesson Learned: Any trials with a combination of these elements are likely to be controversial; broad and open discussion should take place before they are implemented.
The study on which all the developing trials studies were based--AIDS Clinical Trial Group (ACTG) 076--was itself fraught with controversy. Conducted in the United States and France, it demonstrated that AZT was effective in reducing the rate of maternal-infant HIV transmission of HIV. The rate of transmission in the treated group was 8 percent compared to 25 percent in the placebo group. The difference between the two groups was so substantial that the data safety and monitoring board felt it was unethical to continue and stopped the trial early. Subsequent data have confirmed the effectiveness of AZT in preventing perinatal transmission in women who would not have been eligible for 076 because of their lower CD4 count and more serious disease progression. From 1984 through 1992, the estimated number of children with perinatally acquired HIV increased every year, then declined 42 percent during the period 1992 to 1996. Nevertheless, the 076 regimen is not 100 percent effective.
The 076 trial--perhaps one of the few trials that is known around the world by its numerical designation--went through several significant revisions before it was implemented. …