DDI AND PARALLEL TRACK: OVERVIEW AND EDITORIAL

Both the medicine and the public policy around DDI are developing rapidly. But the real progress should not obscure the fact that we still have a serious problem.

The bottom line is that unless major toxicity is found with DDI (which could still happen), the current plans, intentions, and directions will lead to an unprecedented situation of a clear medical consensus in favor of a treatment which is blocked by red tape. There will be adequate supplies of the drug, and it will not be especially expensive to produce. But treatment will be denied because of scientists' fears that they will not get subjects for trials of other drugs such as AZT or CD4 if patients have viable options, because of regulators' fears that precedents set with AIDS will damage their control over routine drugs in the future, and because even if the drug is inexpensive there is no way to charge for it, no one to pay for it, and reluctance to provide a free alternative to high-priced AZT. In other words, thousands of people will be denied lifesaving treatment because the lives of persons with AIDS are still not valuable enough to arouse the political will to treat this epidemic as an emergency.

Formal efficacy trials of DDI should start soon -- hopefully within the next few weeks or months. There will probably be three different double-blind trials: DDI vs AZT for those who have never taken an antiviral before; low-dose DDI vs high-dose DDI for those intolerant to AZT; and AZT vs DDI for those who have already been on AZT for over a year. No one will get a placebo.

These trials appear to be ethical for those who can get into them. The problem is that they will take two years or more to get results -- the time needed to wait for statistically significant numbers of deaths or serious infections to accumulate in the control groups. Meanwhile, for each person who can get into these trials, others who need DDI will be excluded, for various reasons.

It now seems clear that some form of early access -- whether called "parallel track", "treatment IND", or "compassionate use" -- will be provided for certain people who cannot get into the trials, during the two or more years it takes the trials to run, and the additional time for paperwork before DDI is approved for marketing. The issue then becomes what criteria are used to select those eligible for early access. There is much concern that the criteria will be narrow, in order to define away the parallel track as much as possible. Patients and their physicians benefit by early access, but other interests involved are likely to see it as an expense and bother, if not a threat.

An advisory committee including persons with AIDS or AIDS advocates is now being formed to advise Dr. James Mason, Assistant Secretary of Health, about the parallel track on or before August 21. You can write to Dr. Mason about the parallel track and access to DDI; for more information, see below. Telling Dr. Mason and the committee about your own situation, or that of your patients, could be especially helpful, as committees often produce inappropriate rules or criteria because they did not think of enough examples during their deliberations.

What will be done on DDI is haggling about the specific rules and boundaries of access -- who will and who will not be allowed the option of using this treatment, with decisions being made primarily for institutional interests, not patients'
interests. Consequently there will also be increasing activity around "underground" DDI. At this time it is hard to find, and costs about $600 per month or more; but several early samples have been tested and all were good quality. Further chemical testing will be essential, however, in order to prevent fraudulent and/or dangerously contaminated products.

What should be done is to move toward more control of treatment decisions by patients and their physicians. But clinical research will have to reform itself to work well in a more open environment (or in any environment, for that matter). Today it is argued that we must restrict patients' access to treatment options, in order to balance the interests of the "few" (those now ill with AIDS or HIV) against the interests of the "many" (those who will become ill in the future). The fear is that if patients have attractive treatment options, they will not volunteer for scientific trials, and therefore we will not get the answers needed. Even the more humane wing of the current professional debate, Fauci's "parallel track" proposal, is based on the idea that access must be denied those who could participate in trials, in order to force them to do so.

The cheap and easy way to get research subjects would seem to be to deny patients other treatment options. This approach not only is unethical, however, but also it has failed to produce subjects for trials even when there is no access to alternatives. The right way to get volunteers would be to reform what is currently wrong in clinical trial design. Too often trials are not designed for the real world. For example:

(1) If a trial asks a real, practical question, then if follows that all the treatment arms to which patients are randomized must be viable, medically sound therapeutic choices. Why ask patients to undergo a course of therapy which is known in advance to be unattractive and not a viable option? What useful knowledge is gained?

The answer is that commercial considerations often override scientific and medical ones. The drugs which get studied tend to be those with the most commercial and professional momentum behind them, whether they are the best scientific prospects or not. Or patients are forced to take single drugs only (when in practice those drugs will be used in combination with others), because each company wants its product to be approved by itself, not in combination with a rival's product.

When patients are asked to undergo bad medicine for the sake of "science", then naturally they will seek other options. But if trials were redesigned to compare good therapies with other good therapies, then few patients would object to the randomization and blinding which the researchers want, and the conflict would go away. An added bonus is that trials so designed would produce more useful information, because they would focus directly on obtaining the answers which physicians need.

It is finally being acknowledged that the information sought by the FDA for the drug-approval process, and the information sought by physicians for making treatment decisions, are not the same. How odd it is that they should be different! Imagine how much more efficient the system could be if clinical trials were designed to answer the same questions that good physicians will ask when they decide whether or not to use the drug. (One fundamental problem, however, is that the FDA is set up to approve products, not to approve treatments, and clinical trials are therefore designed accordingly.)

Some progress is being made. Trial design now is better than it was two or three years ago. But progress is inherently slow, because it means that academic researchers, who often look down on clinicians, need to learn new ways of thinking, and learn them well enough to be successful in using them.

(2) Another major problem in clinical trials is the discourtesy with which potential volunteers are often treated, the lack of concern for their interests and practical needs. For example, patients often must stop using other treatments which are important for their health in order to qualify for trials. Too often when they do so they are strung along, told that the trial is not ready for them at the expected date, and left to wait indefinitely without their standard medication, with no clear answers about when they can start the trial. (Notice that this problem is not considered in the ethical review of the trial, or the informed consent, because it occurs before the patient begins an official relationship with the study.)

Better administration is part of the answer, but not all of it. In addition, research on human beings must be recognized as a special calling. It differs profoundly from testing drugs on laboratory animals -- and not only because stricter ethical guidelines are required. For human research to be successful, the whole approach to the problem must be different.

The practical effect that such reforms could have on recruitment is illustrated by community-based research organizations which have persons with AIDS or HIV on all their review boards and decision-making bodies. Organizations such as New York's Community Research Initiative have done far better than academic medical centers in recruiting volunteers for trials. The reason is that persons with AIDS and their physicians are well represented on the scientific and ethical review boards which approve these trials. Before studies begin, they are modified to take into account their medical and ethical impacts as felt by those most affected.

In contrast, the academic centers have done so poorly that a number of studies have gone through all the required steps of approval, at considerable trouble and expense, and then failed to recruit a single qualified volunteer. It is preposterous to blame the patients for this situation, or to blame access to "underground" treatment options, or sanctioned options such as the proposed parallel track -- instead of analyzing the problem case by case to find out why patients do not volunteer, and how future trials could be improved so that they would.