When should HAART be started?

17 December 2000. Related: Conference reports, HIV 5th Glasgow 2000.

Brian A. Boyle, MD, for HIVandHepatitis.com

Some studies presented support the concept of delaying therapy until the CD4 T cell count is significantly less than 300 cells/mm³, perhaps as low as 200 cells/mm³.

One of the questions most often heard at this conference was “When should HAART be started?” This question encompasses many of the issues of HAART therapy including tolerance, toxicities, adherence, and preservation of immune function. Currently, the guidelines generally recommend starting HAART while the CD4+ cell count remains above 350 cells/mm³.

Recently, some retrospective studies have been presented which challenge this recommendation and indicate that patient health and virologic success are not affected by waiting until the CD4+ cell count is significantly less than 350 cells/mm³, and perhaps even as low as 200 cells/mm³. [1]

Some interesting retrospective studies were presented at this conference, which appear to support this concept of delaying therapy; however, these studies do not provide hard evidence that delaying therapy has no adverse effects and a clear answer to the question of when to start HAART will likely require a massive clinical trial which may take years to complete. So for now, we are left to our own best clinical judgment based on the information at hand.

Dr. Andrew Phillips from London presented retrospective data derived from chart reviews of 2742 patients who were part of either the Swiss HIV Study, EuroSIDA Study or Frankfurt HIV Clinic cohort [2]. The screening criteria for entry into the study required that the patients be antiretroviral naive when they started HAART, consisting of a regimen of at least 3 antiretrovirals (ARVs).

The cohort was divided into 3 groups based upon baseline CD4+ cell count at the time therapy was started: less than 200 cells/mm³ (baseline median CD4+ cell count 78 cells/mm³ and viral load 5.2 log₁₀ copies/mL), 200 to 349 cells/mm³ (baseline median CD4+ cell count 274 cells/mm³ and viral load 4.8 log₁₀ copies/mL) and more than 349 cells/mm³ (baseline median CD4+ cell count 465 cells/mm³ and viral load 4.6 log₁₀ copies/mL). The endpoints for the study were (1) virologic suppression less than 500 copies/mL within 32 weeks, and (2) in those who achieved less than 500 copies/mL, the time to viral rebound.

2346 (86%) of the patients achieved a viral load less than 500 copies/mL by 32 weeks. Compared with patients with more than 349 cells/mm³, the relative hazards, using a Cox multivariate analysis, of achieving this degree of viral suppression were 1.07 and 0.88 for baseline CD4+ cell counts of 200-349 cells/mm³ and less than 200 cells/mm³, respectively. Dr. Phillips interpreted this data as meaning that the less than 200 cells/mm³ group did “slightly” worse than the other two groups, but admitted that the multivariate analysis virtually eliminated any significant difference between the groups. An analysis based upon viral load showed that patients with a viral load at baseline higher than 100,000 copies/mL did worse as well, with relative hazards of never achieving virologic success of 1.03 and 0.70, for baseline viral loads of 10,000 to 99,999 copies/mL and greater than 100,000 copies/mL, respectively, relative to those patients with a baseline viral load less than 10,000 copies/mL. Of the 2346 patients that successfully achieved a viral load less than 500 copies/mL, 461 rebounded to greater than 500 copies/mL. All groups of CD4+ cell counts and viral load had comparable levels of rebound.

Based upon this data, Dr. Phillips concluded that so long as the CD4+ cell count remains greater than 200 cells/mm3 and the viral load less than 100,000 copies/mL, virologic response rates are roughly equivalent. But, as discussed later, this may not be the entire story.

Dr. Alessandro Cozzi-Lepri from London presented another retrospective, multicentre study evaluating virologic failure in ARV-naive patients starting HAART (with at least 3 ARVs) from the Italian Cohort of Patients Naive to Antiretroviral (I.C.O.N.A.) database [3]. This study included 1484 patients and its primary endpoints were increase in CD4+ cell count, time to virologic failure and rate of AIDS/death after HAART. For the purposes of this study, virologic failure was defined as a failure to reach less than 500 copies/mL by week 32 of HAART or, if initially successful, 2 consecutive viral loads greater than 500 copies/mL.

Similar to the prior study, the patients in this study were divided into groups based upon CD4+ cell count (200 or less cells/mm3, 201 to 350 cells/mm³ and greater than 350 cells/mm³) and viral load. At 96 weeks, the patients with a baseline CD4+ cell count less than 200 cells/mm3 had a higher rate of virologic failure when compared to the group that started with greater than 350 cells/mm³ (relative hazard 1.33). Further, patients with a baseline CD4+ cell count less than 200 cells/mm³ or a baseline viral load greater than 100,000 copies/mL had a higher rate of virologic failure than the other groups, approximately 15% and 30%, respectively, but these rates were not statistically significant.

The mean rise in CD4+ cell count was 296, 312 and 151 cells/mm³ in the groups that had a baseline CD4+ cell count of less than 200, 210-350 and greater than 350 cells/mm³, respectively. There was a highly significant risk of an opportunistic infection with a CD4+ cell count less than 50 cells/mm³ which decreased rapidly as cell counts rose, and there was no difference in this respect between the groups that started with 201 to 350 or greater than 350 cells/mm³.

Dr. Cozzi-Lepri concluded that this study and the study presented by Dr. Phillips reached similar conclusions, and indicate that starting therapy at a CD4+ cell count greater than 200 cells/mm³ and a viral load less than 100,000 copies/mL is relatively safe.

Dr. Julio Montaner raised a significant cautionary note during a question to Dr. Cozzi-Lepri. He raised several confounding factors that could have significantly affected these results, including issues of adherence and medication regimens selected, and stated that he was concerned regarding the long-term effects of allowing significant immunocompromise to occur.

The latter is an extremely important point we do not have an answer for, and several studies presented at this conference found, as prior studies have, that once the immunologic repertoire is lost it may never be fully regained and both cellular and antibody defects persist even after the CD4+ cell count returns to “normal” [4, 5].

While we know that the most critical parts of the immune system may be preserved despite significant immunocompromise and may significantly recover on HAART, preventing the occurrence of most opportunistic infections once certain threshold CD4+ cell counts are achieved, we do not know what other problems may occur due to lost immune function which may be less critical and not recovered.

Are patients who appear “normal” immunologically still at higher risk for certain infections or cancers? Only a long-term, carefully enrolled and controlled trial will be able to definitively answer that question, and until one does, it is my opinion that in most patients we must err on the side of caution and continue to follow the current guidelines.

References:

Montaner J, Hogg R, Yip B, et al. To start or not to start? Diminished effectiveness of anti-retroviral therapy among patients initiating therapy with CD4+ cell counts below 200/mm3. In: Program and Abstracts of the 13th International AIDS Conference, July, 2000, Durban, South Africa. Abstract LbPeB7050.
Phillips A, Staszewski S, Weber R, et al. Viral load changes in response to antiretroviral therapy according to the baseline CD4 lymphocyte count and viral load. In: Program and Abstracts of the Fifth International Congress on Drug Therapy in HIV Infection, October, 2000, Glasgow, Scotland. Abstract PL3.4.
Cozzi-Lepri A, Phillips A, d’Arminio Monforte A, et al. When to start HAART in chronically HIV-infected patients? A collection of pieces of evidence from the I.C.O.N.A. study. In: Program and Abstracts of the Fifth International Congress on Drug Therapy in HIV Infection, October, 2000, Glasgow, Scotland. Abstract PL3.5.
Lane C. What are the differences in immune restoration in response to HAART at different levels of immune suppression/immunodeficiency at initiation of treatment and do they matter? In: Program and Abstracts of the Fifth International Congress on Drug Therapy in HIV Infection, October, 2000, Glasgow, Scotland. Abstract PL7.1.
Lederman M, Valdez H, Medvik K, et al. Functional significance of CD4 cell numbers on the way down versus on the way up. In: Program and Abstracts of the Fifth International Congress on Drug Therapy in HIV Infection, October, 2000, Glasgow, Scotland. Abstract PL7.4.

http://www.hivandhepatitis.com