We at The Center for AIDS acknowledge ourselves to be supporters of the Community Programs for Clinical Research on AIDS (CPCRA) and its SMART study. However, the running of such a study poses formidable challenges to the CPCRA; the study design itself, while intelligent, is not without limitations. In fact, the very nature of long-term effectiveness trials such as SMART is subject to reasonable criticism, which is why we have included an opposing essay from Jeff Schouten, a physician and activist (see The Case Against a When-To-Start Antiretroviral Therapy Trial).

Accruing, retaining, and following 6000 patients for as long as 8 years is no small task. Indeed, the CPCRA itself does not have access to 6000 eligible volunteers, which is why the consortium has sought international collaborators. Moreover, once accrued, SMART will ask volunteers in its virologic suppression arm to do what patients in clinical practice often fail to do: adhere to toxic treatment regimens for years on end. Conversely, SMART will ask volunteers in its drug conservation arm to sit tight as their CD4 T cell counts fall to uncomfortable lows before intervening with treatment. For these reasons, it is an open question whether SMART is feasible. Against this backdrop, many will ask why even try? The answer is a halting one.

At its core, SMART makes the sobering -- and we fear, accurate -- assumption that the development of anti-HIV therapeutics is static. This is SMART's essential premise. Why else would SMART inquire about the better approach to the long-term use of today's therapy unless today's therapy is what we expect to have tomorrow?

Yes, the trade association for the pharmaceutical industry lists 23 agents under investigation for HIV infection. But close inspection of the list reveals, among other things, a gel for prophylaxis against sexual transmission and a sustained-release formulation of an existing nucleoside analog. In other words, the list reveals nothing that would fundamentally alter the current approach to the treatment of HIV disease.

In the coming years, we may see marginal improvements in treatment (e.g. fusion inhibitors and "resistant-repellant" protease inhibitors) but we are unlikely, at the current pace of development, to see the implementation of new concepts in treatment. Some observers, motivated by hope and optimism, may argue otherwise, but their position is inconvenienced by the facts.

The reasons for this paucity of novel therapeutics are beyond the scope of this essay, but if the designers of SMART are correct in the assumption that the current approach to treatment will hold sway for years to come, then the justification for the study is compelling. Moreover, the very findings of SMART could provide the impetus for a new round of private industry investment in HIV drug discovery. Specifically, if SMART demonstrates that antiretroviral therapy can be prudently delayed until the CD4 T cell count has fallen below 250 cells/mm³, the pharmaceutical industry will lose hundreds of millions of dollars. On the other hand, if SMART demonstrates that earlier, tightly suppressive therapy is necessary for long-term clinical benefit, but patients continue to effectively refuse treatment through non-adherence, the industry still has an incentive to find new approaches. Either way, the study results could ignite a fresh search for novel, safer, easier, and more profitable therapeutics.

Since the beginning of the HIV epidemic, patients and clinicians have trusted expert opinion to guide decisions about treatment. That trust has not been well placed. As David Barr persuasively demonstrates (see Too Smart for Our Own Good), expert opinion has been unreliable and often wrong. Only 5 years ago, experts thought that patients in early HIV disease -- and hence at low-risk for clinical progression -- should undertake aggressive therapy. Today, however, experts suggest that the toxicity of treatment outweighs the benefits of its early use. Undoubtedly, expert opinion will shift yet again in the years to come.

By contrast, SMART offers HIV-infected patients the opportunity to have their treatment decisions informed not by opinion, but by evidence. It is undeniable that the practice of medicine is part science and part art, but the art of medicine, properly practiced, refers to the uses of evidence-based judgment and creative decision-making. The art does not include clinical practice guidelines drafted in a vacuum of knowledge. Of course, even SMART will not teach clinicians how to apply their art to every patient, but no study ever teaches that. The best clinical trials reveal only general insights into the treatment of a patient population, not the treatment of a particular individual. Health care providers will always have the task of making decisions case by case.

What of the HIV-infected patients today who sacrifice flexibility in their own decision-making to volunteer for SMART to help answer the question of how best to use antiretroviral therapy? If one of SMART's two strategies proves superior to the other, then half of the study's volunteers will not benefit from its findings, a fact informed consent documents should make clear to them. This is, however, nothing new; it is the nature of clinical trials. Indeed, the only ethical study is one where, as is the case with SMART, the outcome is truly unknown. The history of the epidemic is replete with study volunteers who risked their own welfare to advance the science and treatment of HIV infection. Important research would grind to a halt without such altruism.

Research often involves risks not associated with garden-variety clinical care. Patients averse to the risks of research are in no way obligated to participate. Perhaps, as some have asserted, HIV treatment activists with certain beliefs would not themselves sign on to SMART. However, we activists are only another type of expert, and our intellectual propeller heads are not imbued with clairvoyance by virtue of our HIV infection.

If it should prove impossible for SMART to accrue or follow 6000 volunteers for several years, then HIV-infected patients will have voted with their feet not to answer important questions about long-term management of the disease. So be it. Today, the only thing any of us can do is speculate about the prospects of running the trial to completion. If the study flounders, the government will have made, at worst, a good-faith inquiry into the public health implications of contrasting strategies to the long-term use of antiretroviral therapy. If the study succeeds, the government will have made a vital contribution to understanding and improving the treatment of people with HIV. The CPCRA's investigators, staff, and community advisors are manifestly committed to the latter.

Patients, properly educated, decide as individuals when and if to enroll in a study. But the National Institutes of Health (NIH), acting through subsidiaries such as the CPCRA, is responsible for designing and executing trials that answer questions of public health. Today, the medical establishment, led by the NIH, lacks authoritative answers to an array of questions important to the management of HIV infection. This is why SMART is vital to the interests of HIV-infected people as a whole. The study offers an opportunity to gain evidentiary insight into the long-term clinical consequences of immediate and deferred treatment, uninterrupted and episodic treatment, treatment guided by viral load and treatment guided by CD4 T cell count. Could any other inquiries be more relevant to exploiting the long-term benefits of therapy?

With antiretroviral chemotherapy the treatment option for HIV disease today, and the most likely treatment option for tomorrow, it is time to get SMART.

Back to the RITA! Spring 2001 contents page.