When to Start Antiretroviral Therapy?
Faye A, Le Chenadec J, Dollfus C, et al. Early versus deferred antiretroviral multidrug therapy in infants infected with HIV type 1. Clin Infect Dis. December 1, 2004;39(11):1692-1698.
Holmberg SD, Palella FJ Jr, Lichtenstein KA, Havlir DV. The case for earlier treatment of HIV infection. Clin Infect Dis. December 1, 2004;39(11):1699-1704.
Cohen CJ, Boyle BA. Editorial commentary: antiretroviral therapy: the "when to start" debates. Clin Infect Dis. December 1, 2004;39(11):1705-1708.
Few topics in HIV medicine evoke as much heated debate as the subject of when to start antiretroviral therapy. The December 1, 2004 issue of Clinical Infectious Diseases provides a relevant research paper and excellent viewpoint articles.
In the first research article, Albert Faye and his team examined the effects of early versus delayed antiretroviral therapy in 83 HIV-infected infants in the French Perinatal cohort. Given the potential for life-long use of combination HIV treatment for HIV-infected infants and children, this group of patients amplifies the concerns about long-term drug toxicities and the interest in strategies that might decrease the total time of drug exposure. Additionally, about one fifth of infected children go on to develop a more severe form of HIV disease (including HIV encephalopathy) and an understanding of prognostic factors for children at risk are lacking.
In this study, early drug therapy was defined as initiation of treatment before the infants turned 6 months. Baseline characteristics of the 2 study groups were similar. Forty of the infants received early multi-drug therapy: 74% of the infants received a regimen of 2 nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs) and a protease inhibitor (PI), 10% of the infants received a regimen of 3 NRTIs and 3% received 2 NRTIs with an non-nucleoside reverse transcriptase inhibitor (NNRTI). Forty-three infants received delayed therapy -- 18 initiated treatment between ages 6 and 24 months and 14 started dual-drug treatment before age 24 months and 11 patients did not receive treatment before 24 months of age.
Among the early treatment group, there were no AIDS-defining events during the study. By contrast, in the delayed group, there were 7 AIDS-defining events, including 3 cases of encephalopathy. All AIDS-defining events occurred before the initiation of highly active antiretroviral therapy (HAART). Indeed, no other demographic or laboratory characteristic (such as CD4+ cell count, CD8+ cell count or HIV viral load) was predictive of having an AIDS-defining event. The authors conclude that only receipt of multi-drug therapy before age 6 months was protective against having the severe form of childhood HIV disease.
In my view, because of the high proportion of patients that develop early AIDS manifestations, this population of infant patients provides a temporally compressed view of adult HIV disease. The suggestion that treatment of the entire population of infants protects 20% from developing severe disease, therefore, is not surprising, but rather reinforces the concept that HAART results in immune reconstitution. Lacking from the analysis of this small cohort (and potentially interesting long-term follow-up) is a description of the negative consequences of the treatment, if any. It is this intersection of benefit and risk that forms the core of the debate concerning when is the optimal time for the initiation of antiretroviral therapy.
There are two accompanying articles also in this journal on this topic that are well worth the read. The first, authored by Scott Holmberg and other prominent clinical investigators (including some who have collaborated with me on research), summarized recent information about the timing of starting HAART. While there was initially some controversy regarding the authors' incomplete disclosure of potential conflict (which once again raised awareness of how important it is for writers to disclose pharmaceutical honoraria since it clearly has the potential to bias), I believe that the points raised by this viewpoint article nevertheless deserve attention.
Indeed, many of the endpoints used to evaluate the merits of early or late initiation of treatment are measures of advanced AIDS; potential benefits of HAART could result in the prevention of non-AIDS, HIV complications. Revelations by Lichtenstein1-2 reveal the relationship between nadir CD4+ cell count and the development of lipodystrophy and peripheral neuropathy -- two complications that were once believed only to be the result of drug exposure.
HAART, once burdened by high pill burden and side effects, is clearly easier to adhere to now than in the mid 1990s; drug resistance risk also appears to be improved with the newer treatment combinations, particularly those regimens that include a boosted PI. Lastly, the notion that HAART, once started, can never be discontinued is also being challenged by strategic clinical trials. Hence, a contemporary view of the risks and benefits of HIV treatment also must include a revisit of the risks and benefits of HIV-related adverse events.
In the second editorial, Boyle and Cohen (both contributors to TheBodyPro.com) provide a concise review of arguments, both pro and con for early and late initiation of treatments. The article also reviews advances in treatment and the understanding of HIV complications. Boyle and Cohen point out that while newer regimens are both easier to take and less toxic, there are remaining disadvantages, including toxicity and drug resistance. In an era of increasingly evidence-based decision-making for HIV, we must hope and wait that the optimal study that addresses these issues will be conducted. Until that time, the debates will continue.
