AIDS TREATMENT NEWS #234, November 3, 1995
John S. James

[Do not rely on us to get this message across. We cannot attend the meeting, and we missed the deadline for advance distribution of this statement. Nor do we know what issues are being addressed in the evaluation process, as that is not public information; sometimes it has not been available even to Working Group members themselves.]

*****

We are hearing widespread concern that OAR may go overboard in the shift to basic research, and seriously limit the NIH role in AIDS trials. Granted, previous AIDS clinical-trial research often lacked focus, and as a result was unproductive. But some AIDS trials clearly belong at NIH.

Here are four kinds of AIDS trials (not a complete list) that NIH must do:

(1) Pathogenesis-intensive trials. For example, one ACTG trial now being designed will suppress viral replication as completely as possible with multiple drugs, in a small number of patients, to allow intensive study of the changes in CD4 and other blood cells as the CD4 counts increase. Clearly this clinical trial is basic science as well -- basic science which is especially relevant, as it is done in people. Industry is unlikely to do such work, and community organizations are seldom equipped for it.

Note that pathogenesis-intensive trials -- the most scientific of all trials -- do not always need to be controlled.

(2) Small, rapid, proof-of-principle trials of treatment leads, including "alternative" treatments. The history of medicine is full of surprises; treatment progress seldom moves in predictable straight lines. Testing the dozens of treatment leads which have known safety information and some credible rationale will almost certainly find no measurable benefit from most of them. But a few probably will prove unexpectedly interesting -- producing incalculable SCIENTIFIC benefit by opening new research avenues, in addition to any immediate value the potential treatments may have to patients and physicians. (Other benefits at no extra cost include the public-health importance of showing that many popular treatments fail to work as intended -- as well as the economic value of any positive discovery in reducing the cost of medical care, since most of the treatment leads in this class are non-proprietary and readily available at negligible cost.)

The primary outcome could be viral load, immunological markers, or other credible measurements. The aim of this research would be to find some measurable, reproducible effect indicating drug activity. Once activity is known, further research could be planned from there.

The National Cancer Institute has traditionally done such early testing in cancer; but in AIDS, no one has done much of it so far. Community organizations could conduct these trials, but usually they have not. Part of the problem may be regulatory cost, as many treatment leads in common human use do not exist in forms that meet FDA requirements for research -- prepared under GMP (Good Manufacturing Practices) rules, and with the active ingredient properly standardized from batch to batch (which is probably impossible if the active ingredient(s) are unknown). Running a trial legally in the U.S. could require the major expense of setting up a new production facility -- in addition to laboratory research to identify, test for, and perhaps synthesize active ingredient(s). Because of these barriers, nothing happens -- when a safe, simple trial could quickly identify the few valuable leads, which then could easily find the support required for further research. NIH could deal with this problem more effectively than community organizations can.

Another problem for community research is that only organizations with a highly "respectable" face can raise the money required, and those groups tend to follow the same scientific fashions as NIH and others, reducing their potential contribution to that of a small drop in the big bucket of research already being done. Technically these trials would be easy to run; but it is hard to find the required money, expertise, good judgment, and also openness to new ideas, all in the same organization.

(3) Development and standardization of "markers" and other tests. NIH is already the leader here. Industry can develop specific tests, but there are major gaps that industry will not fill.

NIH should do more to develop new immunological markers -- especially those which can be measured by flow cytometry or related methods, since this technology is well suited to standardization and quality control.

(4) Confirmatory trials -- or rather, a single master protocol for clinical confirmation of antiretrovirals which have received accelerated approval. When new treatments are likely to become part of the standard of care and be administered to hundreds of thousands of people, many researchers, physicians, and patients want long-term clinical data, to confirm that treatments approved based on short-term trials really are ultimately beneficial, and do not have detrimental surprises later.

There is no way that every new treatment and promising combination can continue to have its own confirmatory trial, with its own control group, etc. We need a master protocol to compare the most important possibilities; and government must be involved in the administration and data management, instead of leaving this to the competing companies. Government should pay for and control the aspects of this research which are not drug-specific.

Confirmatory trials will become much more feasible if we can accept a certain philosophical shift. We need to realize that these trials -- although they are randomized and controlled -- do not need to produce a winner and a loser. Therefore, this master protocol does not need to have a loser arm (traditionally provided by an obsolescent standard of care, such as AZT monotherapy before the ACTG 175 and Delta results were reported), and can avoid all the ethical, recruitment, and compliance problems that requiring a loser entails.

And these trials do not need to prove equivalence either; they do not need to reject any null hypothesis. Instead, their purpose should be to monitor the major therapeutic options likely to become part of the standard of care, looking for any unexpected problems (or unexpected benefits as well). If the master protocol we propose never proves that anything is better than anything else, and also never proves that anything is equivalent to anything else, it will still have been a complete success. It will have safeguarded the public from unexpected toxicities or treatment failures in all of the treatments tested. (All these treatments will already have proven superiority or at least equivalence to standard of care -- in the surrogate-marker trials required to gain accelerated approval -- before being eligible to enter the confirmatory master protocol.)

Such a master-protocol trial will also produce a treasure chest of associated information, such as which patients are likely to do well on which treatments -- information not as readily available from separate confirmatory trials which are not designed to be comparable. And additional substudies can be nested to address specific questions.

To illustrate with specific drugs, such a protocol could randomize volunteers to the Roche, Merck, Abbott, or other protease inhibitor -- or any other drug combination with equally good safety and surrogate-marker results -- after those drugs have received accelerated approval. Each developer would recommend other drugs (for example, AZT plus 3TC) designed to make the combination with their product work as well as possible; and volunteers would not have to pay for these recommended drugs. (Volunteers would be allowed to use other, non-recommended drugs on their own; therefore, they would have no reason to conceal anything from the researchers, but could fully report what they did.)

NIH, as administrator of the master protocol, would decide which combination treatments were good enough to become an arm in the trial -- meaning that participation would be a major marketing advantage for a company (not an unavoidable burden like confirmatory trials today) because it would imply official recognition that, based on all current knowledge, the company's treatment was as good as any treatment known. And participation would of course fulfill the accelerated approval confirmatory trial requirements -- almost regardless of the result of the trial, as there would be no loser arm, and therefore no winner/loser relationship expected. Only the discovery of a serious toxicity or other drug failure would be held against a drug, greatly reducing the developer's risk from the head-to-head comparison with competitors. With these incentives, industry will compete to get into these trials, greatly facilitating inter-company cooperation (since a winning combination will usually require products of more than one company), and helping to finance the project as well.

But this whole structure requires government management of the master protocol, requires an outside party which answers to the public interest. Competing companies probably cannot organize such a program on their own. NIH has already shown that it can fulfill this kind of management role, for example in ACTG 175. No one else could do so.

951103
ATN23404

AEGiS presents published material, reprinted with permission and neither endorses nor opposes any material. All information contained on this website, including information relating to health conditions, products, and treatments, is for informational purposes only. It is often presented in summary or aggregate form. It is not meant to be a substitute for the advice provided by your own physician or other medical professionals. Always discuss treatment options with a doctor who specializes in treating HIV.

Copyright ©1980, 1995. AEGiS. All materials appearing on AEGiS are protected by copyright as a collective work or compilation under U.S. copyright and other laws and are the property of AEGiS, or the party credited as the provider of the content. .