HIV Treatment as Prevention: To be or not to be?

Now 30 years after the first case of AIDS was described we have yet to develop reliable methods-either behavioral or biological-to prevent the spread of HIV. The spectacular and life-saving success of antiretroviral therapy has been more than offset by the challenge of treating millions of people world wide for their entire lives, and the sobering observation that for every person we treat several more become newly infected.1 Given this sad reality, no prevention stone should go unturned. And it seems more than reasonable to believe that antiretroviral drugs can be used for prevention.2 Indeed, these drugs have demonstrated the ability to nearly eliminate vertical (mother to baby) transmission of HIV,3 implementation challenges notwithstanding. Antiretroviral agents could be used to prevent the sexual transmission of HIV as preexposure prophylaxis, postexposure prophylaxis or to reduce transmission from infected patients.2 The latter approach has great promise: because personal health can benefit from antiretroviral therapy (ART) why not exploit the public health benefit as well? Indeed, many drugs concentrate in the genital secretions leading to profound and prolonged (albeit incomplete) suppression of HIV in the genital tract.2 The widespread availability of ever safer ART has led to clinical observations that support treatment as prevention. Several studies of discordant couples demonstrate dramatically reduced risk to the HIV-negative sexual partner when the infected person takes ART.4-7 And some (but not all) population based studies suggest the potential for falling prevalence of HIV in communities where ART is broadly used.8-12 Under idealized conditions, where most HIV-infected people are tested and treated and suppressed for life, it is possible to formulate a mathematical model where we literally “treat our way out of the epidemic.13” But 3 big concerns have clouded the picture: (1) transmission of resistant viral strains14; (2) the contribution of patients with acute and early HIV (subjects who are not likely to be detected routinely but contribute substantially to the spread of HIV15; and (3) the practicality of the idea.16 In this issue of J Acquir Immune Defic Syndr, a Chinese team of investigators have offered us an eye-opening “real world view.17” The Chinese HIV epidemic includes a large subgroup of paid plasma donors (now called former plasma donors in deference to the elimination of the practice) who were inadvertently infected with HIV through flawed procedures.18 In the current report, Wang et al17 retrospectively evaluated 1927 discordant couples for 2 years, 2006-2008. The index case received HIV care (including ART) free of charge. The results related to HIV transmission in these couples are important, surprising and cautionary. Eighty-four seroconversion events were observed in 4.3% of couples with 2 critical twists. First, the transmission events increased with time of follow-up, regardless of counseling, knowledge, and condom availability. This result demonstrates no protective immunity in partners who were, by definition, exposed and uninfected at the time the study started. Second, and perhaps most important, transmission events occurred with equal frequency in couples regardless of whether the partner was provided free ART. Of course, it is possible that HIV was acquired from another sexual partner (as has been observed in studies in Africa5,7). Or the patients may have used their drugs poorly and/or developed resistance to therapy. But the main point is that the Chinese patients were receiving routine health care. Studies reporting the great success of ART in reducing HIV transmission4-7 represent couples managed under special conditions with (for the most part) frequent visits, laboratory monitoring, and some unique health care resources. The “test and treat” idea did not spring to life overnight. A great deal of effort has gone into proving the biological plausibility of this approach,2,19 and National Institutes of Health is supporting a multinational randomized clinical trial (HPTN052) of 1750 discordant couples to prove that ART durably reduces HIV transmission over 5 years.19 But at the end of the day, the success of the “test and treat” idea for a population rests on the ability of ART to produce durable reduction of HIV transmission from treated people to their sexual partners. The results from Wang et al17 in China demand a giant pause. Will ART suppress the transmission of HIV under “real life” conditions? It would seem more than wise to try to answer this question before we fully deploy a “Test and Treat strategy”, expecting to detect a benefit to the general population.