PROBLEM WITH AZT?

Our last column summarized the only published report so far of clinical tests with the experimental drug azidothymidine. This early trial brought clear benefits and few side effects to the 19 patients in the study group.

Later we heard anecdotal reports that several patients in later trials in two cities (not including San Francisco) developed pancytopenia, a serious bone-marrow disease. We called Burroughs-Wellcome Corporation, which is developing AZT; they told us that there have been some changes in hematological measurements in both phase 1 and phase 2 tests, but that sometimes these changes occur in persons with AIDS without any
therapy, so more study is necessary before we can draw conclusions.

We will have to wait for more information. Meanwhile, here are some points to consider about all clinical tests on experimental drugs, not only AZT.

First, this writer and many others have been urging faster action on the development of new treatments. Often there is a trade-off between speed and risk; for example, researchers could reduce the risks of side effects by scheduling months or years of animal tests before any human trials begin. But persons with AIDS should have the right to make their own decisions about
accepting the risks of trying new treatments, instead of being given a flat "No". And everyone benefits when treatment research moves quickly ahead. For these reasons, we should not criticize but commend researchers and physicians who make decisions which are right at the time, even if wrong in hindsight -- which balance the risks of doing something against the risks of doing
nothing. They are taking a personal risk when they could so easily cover themselves by refusing to do clinical trials until exhaustive preliminaries have been completed.

Also, we should make sure that no treatment gets lost unnecessarily, even if side effects occur. For example, AZT worked very well in at least one clinical trial (published in Lancet, March 15, 1986). Problems discovered in other trials may be manageable. For example, side effects could be dose-related; the published results of AZT suggest this possibility, as the
higher doses tested sometimes produced a B-cell deficiency, though never a serious one. Perhaps it will prove possible to predict who will react adversely to a drug, or to monitor blood tests in order to catch problems in time. Side effects can even result from impurities in experimental drugs, which are not regularly manufactured but often made up in a new batch for each
study, by scientists who have probably never prepared that substance in pharmaceutical purity before. And even if it's not possible to avoid a danger, the benefit of the treatment may outweigh it.

Unfortunately, research institutions which conduct a test which fails due to side effects may never find it in their interest to publish or announce the results, due to fear of legal liability or other reasons. If they don't publish, other clinicians who have heard of the problems but don't have the details won't dare to try the drug again, and a useful treatment
which might be safe if used differently will be lost. Sometimes we may have to insist that when persons have risked their lives in drug trials, the public has at least a moral right to have the results published so that they can help other scientists and doctors to save lives in the future.