AIDS Drug Points Up Fairness Issue in Medical Tests

The expected announcement today to stop the clinical trials of AZT, an experimental AIDS drug, is part of the continuing medical debate over how to evaluate new therapies for life-threatening diseases.

Most researchers agree that clinical trials, in which a promising drug is compared to another drug or to a placebo, are among the most reliable means of obtaining data on a drug’s efficacy and safety.

But such studies, known as randomized controlled trials, often seem unfair, particularly for tests of drugs against fatal diseases in which some of the patients may die because they were deprived of an effective drug.

In such tests, neither the patients nor their physicians know what drug, if any, the patients are getting. New surgeries and diagnostic tests can also be evaluated in a similar manner. A control group is necessary to measure whether a therapy benefits or harms patients.

Board Makes Decisions

Decisions to stop a major clinical trial are usually made by an independent board, whose members often include statisticians, ethicists, lawyers and lay people as well as physicians. Such boards meet periodically to review data, as was the case in the Burroughs Wellcome-sponsored trial of azidothymidine, or AZT.

Rules for stopping trials are usually set before the first patient receives the experimental treatment.

Trials usually are not stopped prematurely. However, if the review board members are at least 95% confident that the medicine has been proved safe and effective, they often stop the trial, according to Dr. Richard J. Whitley of the University of Alabama at Birmingham. Trials are also stopped when unexpected complications develop.

“A clinical trial is only justified when it is reasonable to assume that either of two alternative treatments might turn out to be the best,” said Robert M. Veatch, a professor of medical ethics at Georgetown University. As soon as one therapy seems better than the other, the trial should be stopped, he said.

Since World War II, “medicine is littered with failed therapies that have been proved not to work after controlled trials,” said Arthur Caplan, an expert in medical ethics at the Hastings Center, in Hastings-on-Hudson, N.Y.

Among such examples, he said, are the now-forgotten practice of freezing the stomach with liquid nitrogen to treat ulcers and the use of bypass surgery on arteries supplying blood to the brain to prevent strokes.

More recently, for AIDS patients, suramin, a once promising drug, was shown in controlled trials to shorten the lives of some who received it.

Similarities Cited

The tests of AZT and other AIDS drugs have many similarities to current tests by the National Institute of Allergy and Infectious Diseases of new drugs against life-threatening herpes virus infections, such as a rare brain infection called herpes simplex encephalitis.

Nearly a decade ago, promising test results in one of these studies raised similar issues about when to stop a drug trial.

Study Halted Abruptly

In that study, some patients were given the test drug while the others received a placebo. The 28-patient study was stopped abruptly because 72% of patients who received the test drug were still alive, while 70% of the placebo recipients had died.

And even though researchers quickly recommended that the drug be used for all patients with the infection, there was nevertheless criticism that they allowed some patients to die while the data was being collected.