DRUG DEVELOPMENT: WHAT'S NEEDED NOW?

For those familiar with clinical trials and drug development, the last few months have been particularly difficult. Aside from ddI, where the research and distribution programs have essentially worked and much of the news at this time is good, research on new treatments appears to have greatly slowed. We have been unable to confirm that the situation is really as bad as it appears, however, because of the widespread confusion; no one seemed to know what was happening. We could not find a vision or direction for AIDS treatment research, a plan with a reasonable chance of success.

Now the outlines of how to overcome the current obstacles and to make research productive again are becoming visible. Scientifically, what needs to be done is already fairly clear. In the past, however, the dismal political landscape ruled out any possibility that what should be done would be done, and led to the current paralysis. Today elements are in place for new consensus, new social contracts among the parties involved. This article describes key elements of consensus which we believe could and should develop.

What's Wrong Now?

The basic problem has been the expectation that all new antivirals must go through huge trials, usually scheduled for two years and requiring hundreds of carefully selected volunteers who are willing to have their treatment selected at random, before any drug can be recognized as effective. In practice, enormous administrative, financial, proprietary, scientific, ethical, and personnel-shortage problems make these trials take much longer than two years, and guarantee that the great majority of drugs worthy of human testing will not be in trials at all.

Early-access arrangements, such as the current test with ddI or the proposed "parallel track" to provide treatment during the trials for persons with no other medical options who do not qualify for trials, can save many lives; these vital efforts must be supported. But early access only works when there is a public-spirited and well-financed drug company, willing to spend money to save lives when sales of the drug may be years away. It does not address the shortage of researchers and facilities caused by the resource demands of the "dinosaur" (large and slow) trials -- demands which in practice prevent drugs from being tested, approved, or considered for early access. In short, while we continue to work for early access, we must also develop more fundamental reforms to improve the productivity of the drug testing and approval system.

What's Needed, And Why It Hasn't Happened Already

The reason for the "dinosaur" trials in the first place is that disease progression in AIDS is highly variable, and the laboratory tests to measure progression (such as T-cell count and p24 antigen) have drawbacks; many researchers have not been willing to trust these tests to tell whether drugs are working. The two-year trials with hundreds of patients, comparing deaths and serious infections in those who do and do not get the new drug, are used to obtain definitive statistical proof that the drugs are helping people. The problem, of course, is not only that these trials take a long time to get results, but also that they are so unwieldy that in practice they seldom happen at all.

The most important new technology for potentially changing this situation is the development of better tests of disease progression, especially plasma viremia (see "New Viral Measurement for Rapid Drug Testing," AIDS TREATMENT NEWS # 94, January 5, 1990). A small experiment using AZT showed that it was possible to get definitive statistical proof in a few weeks, with only a small number of volunteers, that an antiviral is working in humans to reduce the level of infectious HIV in the blood. If this kind of test could be used instead of the huge two-year trials required until now, the logjam of AIDS clinical trials could be broken, and all the important drugs could be tested quickly.

At this time the plasma viremia test is gaining widespread professional acceptance. The only objections we have heard are that some researchers question whether the test is quite ready, and some want to wait for future tests which will be more economical. And some leading experts believe that plasma viremia isn't really necessary, because rapid drug testing could have been done all along, using p24 antigen, despite its drawbacks. The big problem now with plasma viremia is that it is very expensive, costing over a thousand dollars per test. Few labs have experience doing this test, which is not available commercially.

The professional consensus now developing sees plasma viremia as something useful to do. But should it be used only to select interesting candidate drugs for the two-year "dinosaur" trials, or can it be accepted as the proof of efficacy required for marketing approval? Could a drug be approved with official labeling stating that it reduces HIV levels in blood, although clinical benefit has not yet been proved? On this question, consensus has not yet developed either way.

Here is why we believe it is vitally important that the answer to the latter question be "Yes":

* Avoiding the need for two-year, several-hundred-patient multicenter efficacy trials could quickly end the current drugjam, and allow all important antivirals to be tested.

* Additional safety information could be obtained from trials (or supervised access programs) designed to answer safety questions -- not dredged as a by-product out of trials structured primarily to get statistical proof of efficacy.

* A major study by the FDA found that all five vitally important anti-infective drugs recently approved for marketing were approved based on pivotal clinical trials none of which used survival as an endpoint. At least some of these trials did use reduction in microbial levels, instead. AIDS drugs should not be bottlenecked by unworkable requirements which are not applied to other drugs.

* Doctors and patients considering using antivirals which had been proved to reduce HIV (but not proved to increase survival or reduce opportunistic infections, because those studies had not been done) would also have other options available, including AZT, about which much more is known. They can and should be trusted to decide among these options.

* Further clinical research could be required after marketing approval, when necessary.

* Companies with a promising drug would have a realistic chance to get it to market in the foreseeable future, and make an attractive but fair profit. (Today, only companies with great resources and the right connections can seriously hope to do so.) Incentive would stimulate AIDS drug-development projects throughout the clinical-trials pipeline, all the way back to the laboratory, and could get many more companies involved in AIDS research (only a few are today).

* Competition between drugs would help to control runaway prices, potentially saving billions of dollars for governments, other third-party payors, and individuals.

* Faster development of improved treatments could save additional billions of dollars in hospitalization and other expensive care -- as well as saving lives.

* The alternative -- continuing to require huge, two-year trials before drug approval -- is exceptionally grim. U. S. AIDS drug development is now close to a standstill, and there are no prospects for finding enough resources to get it going again under current procedures.

But what if the AIDS researcher establishment will not support plasma viremia or any other alternative to the years-long, "dinosaur" trials for meeting the efficacy requirement for marketing approval? If that happens, then the AIDS community should use the consensus it can get -- that plasma viremia is worth doing to help prioritize drugs -- and push for parallel track access to the drugs which this test shows works best. And also, we would have to examine how much of the objection to a workable approval system stems from scientific concerns, and how much stems from the empires which have been built under the present system. How much results from the influence which Burroughs-Wellcome (which stands to lose most if alternatives to AZT are approved for marketing) has purchased throughout the AIDS research community?

In the past, AIDS researchers were inadvertently given incentives to build empires, not to get practical results. No one spoke out when existing procedures, policies, regulatory standards, and scientific conventional wisdom virtually
guaranteed that no new drugs would be developed. Today there is much more pressure for results. All the parties involved have a common interest in improving the productivity and efficiency of drug development and regulation.

Price Vs. Incentive: Toward A New Social Contract?

Recently we have heard concerns that pressures from the AIDS community to lower drug prices will deter companies from engaging in research. We do not know how serious this problem is, but it is hard to see how price will cease being an issue. Usually those who want to let companies charge anything, to give them maximum incentive to develop new drugs, are those who will always get the treatments they need at any price. But for the great majority, the price issue will not go away.

Another approach to improving incentives is to make potential profits more likely, and not as far away in time -- instead of insuring the right to make exorbitant profits in some hazy, distant future when the drug in question will probably be obsolete anyway. Consider the following proposed three-way understanding between government, industry, and the AIDS community:

* In an emergency, the FDA would develop standards designed to be feasible for pharmaceutical companies to meet, to promote two public purposes: getting drugs approved sooner, and giving pharmaceutical companies more incentive to be involved in research. The FDA is already moving in this direction. It should explicitly recognize that part of its job is to develop regulations which stimulate commercial development when required to meet a national emergency.

* Companies which respond to an emergency must be able to recover their research costs and make a profit. Some would say the profit should be above industry average, because of the extra risks and expenses for an accelerated or even crash program. But also there must be some restraint. In return for government assistance, through financing of research, streamlining of regulations, or otherwise, companies with a unique product for a life-and-death emergency must not be allowed to profiteer and charge whatever the market will bear. Companies can no longer take advantage of the bigotry toward AIDS to pioneer unconscionable policies, which they then apply to other diseases later.

* The AIDS community should accept reasonable profits (but not profiteering) for drug companies as a requirement for making the system work. We cannot wait for a new economy -- either for socialism or for libertarian free enterprise -- but must make the existing system work now. We should continue to support companies which are willing to consider the public interest and work together in mutually beneficial public-private partnerships. Also, the AIDS community will need to communicate and negotiate with the consumer-protection movement, and with advocates and organizations for other diseases, more than it has in the past.

Until last summer, government-funded researchers and FDA officials had little contact with "front line" private or clinic physicians with large AIDS practices; now there is much better communication. This improvement should make the government-funded research more responsive to practical needs.

Patients and researchers often have different interests; until recently there was little communication, and often hostility, between them. Now both groups are learning that they need to work together.

Today all parties involved in AIDS treatment development realize that there are serious problems in the current system, and that consensus and collaboration are necessary to overcome them. The elements of a new consensus are now taking shape.