179 Scientific Rationale for Antiretroviral Therapy in 2005 Robert F Siliciano Johns Hopkins Univ Sch of Med and Howard Hughes Med Inst, Baltimore, MD, USA

This talk will discuss key scientific principles underlying antiretroviral therapy, focusing on the unique way in which HIV evolves in vivo.  In viremic patients, there are two important populations of virus: a quantitatively dominant, rapidly evolving population that is replicating in activated CD4+ T cells and macrophages, and small stable populations of viruses that persist in cellular reservoirs.  New data confirm that the reservoir in resting memory CD4+ T cells allows virus to persist in a latent form for decades even when all active replication is stopped.  Active replication in the presence of suboptimal drug regimens leads to the evolution of resistant variants that can enter this latent reservoir, persist there indefinitely, and reemerge at any time in the future when the selective conditions are favorable.  In patients on suppressive HAART regimens, there is always a low level of viremia that represents the sum of residual ongoing replication and virus release from stable reservoirs including the reservoir in resting memory CD4+ T cells and a newly identified “second reservoir”.  The best that HAART can do is to suppress viremia to a new steady state called the “release point” which represents the level of viruses released from stable reservoirs when new cycles of replication are completely stopped by HAART.  Thus the goal of therapy should be to suppress viremia to this release point.  For most patients, the release point is somewhat below 50 copies/ml.  In most patients on HAART, there are episodes of intermittently detectable viremia (“blips”).  A detailed analysis of blips with sampling every 2-3 days indicates that blips are common, short in duration, low in magnitude (<200 copies), and represent random biological and statistical variation around a mean level of viremia below 50 copies/ml.  Ultrasensitive genotyping indicates that no new drug resistance mutations appear before, during or immediately after blips.  Future directions for improving HAART include assays that will determine whether suppression to the release point has been achieved in individual patients and  assays that will detect hidden drug resistance mutations selected by prior non-suppressive therapy that are present in the latent reservoir.