In an AIDS Study, the Devil Is in the
New York Times ( 04/02/00) By JEROME GROOPMAN
Imagine a medical study in which researchers are tracking the progress of H.I.V. in 415 heterosexual American couples. One of the partners is already infected. The doctors do not tell the uninfected partner that the other has the virus. After receiving intensive instruction on how to prevent transmission, and access to free condoms, the couples are followed for more than two years. Antiviral drugs are withheld. When the study ends, 90 more people are H.I.V. infected.
This is no idle scenario. While the study described above was conducted mainly by Americans, the setting was rural Uganda. The results of the study, which began in 1994 and ended in 1998, were published last week by The New England Journal of Medicine. Even before last week, many doctors have asked how such studies in the developing world could be justified on moral grounds.
It could be argued that the Uganda research, part of a larger study of sexually transmitted diseases among 15,127 people, was an opportunity for scientists to observe the progress and spread of a deadly disease so they could better know how to prevent it in the future. Withholding treatment, too, might offer the chance to study when best to intervene. Still, how could American researchers stand by silently while their subjects contracted the virus? How could they simply watch as a potentially fatal infection took its natural course?
The editor of the New England Journal of Medicine, Dr. Marcia Angell, raised these questions in an editorial accompanying the article. Saying that such a study could not have been ethically conducted in the United States, she believes that it could not be justified based upon its major research finding: that the more virus a person carries the more likely he or she is to infect a sexual partner. The conclusion is of little use to the Africans, who have no access to therapy. It is more valuable to the developed world.
Even this finding was hardly surprising. Studies with similar results involving monkey viruses related to H.I.V. will soon be published. While imperfect, such animal models can be substituted when human studies raise moral concerns.
In defense of the study, its leader, Dr. Thomas Quinn of Johns Hopkins University, said that the research met a "high ethical standard." Ugandan authorities, he said, insisted that researchers not inform study subjects that their partners had H.I.V.; only the infected person could do that. The study was ethical, Dr. Quinn said, "because it provided the best medical care possible in remote regions of rural Africa." The subjects with syphilis and gonorrhea, which are easily treated, were given antibiotics. Johns Hopkins approved the research, which was not designed for therapeutic intervention against the AIDS virus, Dr. Quinn said. In rural Uganda, he contended, there is no way to monitor the antiviral drugs, which can be toxic.
Knowledge about the natural history of a disease is the underpinning of all medical decision-making, and serves as the yardstick with which to measure the risks and benefits of intervention. Sometimes a refined picture of a disease's history has invalidated clinical practices based on apparently reasonable assumptions. Questions still persist about AIDS treatment, particularly therapy given early in the disease when the immune system is intact. The assumption that AIDS researchers like myself make is that it is best to treat early and hard, but this is not yet proven. Drugs used too early might make patients resistant later when they need treatment more.
An example of this is found in cancer medicine, my specialty. I examined a 62-year-old man who had enlarged lymph nodes in his neck, underarms and groin. A biopsy showed a nonaggressive lymphoma. He was shocked because he had no symptoms. After studying laboratory tests, I advised him that treatment was not clearly required at this stage of the illness.
"I have cancer and you are not going to treat me?"" he asked in disbelief. "This is not some kind of experiment, is it?"
For many years the conventional wisdom had been to treat lymphoma early. But in the 1970's, studies found that immediate treatment for this kind of lymphoma was not always in the patient's best interests because it neither extended survival nor improved quality of life.
Other disorders have now been shown to benefit from early treatment. Initially, physicians had scant information on hepatitis C, and many concluded that the hepatitis C virus was relatively innocuous, because it seemed to damage the liver minimally. Few advised their patients about the risk of sexual transmission; therapy was not recommended. Recent studies show that early, aggressive antiviral therapy can protect the liver and sometimes eradicate the virus. The quest for medical data has included extremes in which humans were treated as subhuman. Some of the most accurate information about tissue damage and death due to exposure to cold comes from Nazi experiments on concentration camp inmates. The myriad manifestations of syphilis were meticulously charted in the infamous Tuskegee study, in which black men were not treated with penicillin so that researchers could continue collecting results.
The difference between the abominable Tuskegee trials and the study in Uganda is that anti-H.I.V. drugs were unavailable to rural Ugandans. Treatment would also not be routinely provided after the research ended. The absence of such drugs was consistent with the status quo. Some would argue that although Uganda lacked the resources to monitor drug safety in remote areas, all the subjects in a clinical research study deserve to benefit fully from their participation in it.
If information were available showing that the Ugandans were of early-stage infection, and not in clear need of drugs, a moral argument could be better made to justify the failure to offer antiviral treatment. But these data are absent. "I wish such T-cell data could be gotten in rural Africa," Dr. Quinn said.
The issues here go well beyond this particular study and H.I.V. Scientists are rapidly identifying disease-causing genes and new microbes. The thirst for data on the natural history of an illness need not divert researchers from moral standards. Medical scientists can design studies that keep the developing world from turning into a laboratory for experiments that they would not perform in their own clinics.