August 14, 2006
Does treatment during acute or early HIV infection alter the virologic or immunologic set point or delay the time for initiation of therapy? As of yet there are no conclusive, scientific data. However, two multicenter, German cohorts (Prime-DAG and Ac-DAG) that have been recruiting patients with recent HIV infection (documented to have occurred in the past six months) have given us new insights into this issue.1
Of the 200 patients who were identified since 2001, 144 received some form of highly active antiretroviral therapy (HAART), while 56 did not. This was not a randomized study, and those who were treated had much higher median HIV plasma viral load measures (more than 500,000 versus 240,000 copies/mL) and lower CD4+ cell counts (453 versus 621 cells/mm3). Of the patients who received HAART, 98 stopped treatment after a median of nine (range 1.2 to 28.7) months, with a median CD4+ cell count of 797 cells/mm3, and 79/98 had plasma viral load measures below 50 copies/mL.
The analysis presented at the XVI International AIDS Conference compared the 98 individuals who stopped treatment to the 56 who never received HAART in terms of the time to a CD4+ cell count of less than 350 cells/mm3 and/or an HIV plasma viral load of more than 100,000 copies/mL, representing a disease stage at which the initiation of HAART should be considered.
Over the period of observation, similar proportions of both groups (37/98 [38%] versus 20/56 [36%]) reached the pre-established endpoint. However, those who had received HAART reached this endpoint much later (after 14.3 versus 8.3 months, P = .016) -- a difference that may be even more significant in patients who have baseline HIV plasma viral load measures above 50,000 copies/mL. The authors concluded that "early treatment of primary HIV infection delays the time until possible treatment indication in patients presenting with high viral loads during seroconversion."
These data are quite impressive, representing one of the largest cohorts of patients with acute/early HIV infection to have been so carefully evaluated. It is now quite clear that the maintenance of an undetectable viral load after time-limited HAART in acute or early HIV infection will be an unusual phenomenon. If any benefit of therapy exists, it will manifest as a more favorable set point and/or a measurable delay in the need to initiate HAART.
These observational data were, by the very design of the study, clearly biased against showing such a benefit. Patients who received treatment had more advanced disease at baseline, while the time to the need to initiate (or re-initiate) HAART was not counted from the time of initial diagnosis. In my opinion, if we add this period of time, the potential benefit of therapy could be as much as three years rather than six months. Finally, the cohort included patients who were infected well after developing a full antibody response to HIV, and this may represent a group that would benefit less significantly from therapy. Their inclusion in this study would dilute the potential benefit of the intervention that could have been measured. Because of all of these considerations in the context of a non-randomized study, the information presented in this poster cannot be taken as conclusive. However, the fact that it supports the hypothesis -- that a benefit of a time-limited course of HAART may exist -- justifies the conduct of randomized, controlled clinical trials comparing treatment to observation in patients with "recent" HIV infection. The definition of "recent," however, must be rigorous and designed to ensure that the population that may benefit the most from the proposed intervention is recruited. The endpoint of such studies must no longer be the maintenance of an undetectable viral load after treatment is discontinued, but must be based on the achievement of clinically significant virologic and immunologic benefits.
We are indeed fortunate that such studies are currently underway in North America2 and Australia,3 although they are unlikely to yield a definitive answer for a few years. In the meantime, it appears sensible to make HAART available to individuals with early HIV infection, if such individuals (in consultation with their health care providers) feel this is the most appropriate course of action to take.