Publication Abstract Display
Type: Published Abstract
Title: Divergent evolution of phenotypic antiretroviral susceptibility of HIV isolated from cerebrospinal fluid and plasma.
Authors: Letendre SL, Paxinos E, Wong J, Petropoulos C, Whitcomb J, Galovich J, Hellmann N, McCutchan JA, Ellis RJ
Year: 2002
Publication: Antiviral Therapy
Volume: 7 Issue: Suppl 1 Pages: 66
Abstract:BACKGROUND: HIV strains isolated from immuneprivileged sites, such as the central nervous system (CNS), can differ in env, gag, and pol compared with those from other tissues. Differences in pol may have important clinical consequences since many antiretrovirals (ARVs) poorly penetrate the blood-brain barrier. For example, subtherapeutic ARV concentrations in the CNS might accelerate development of resistant strains in the CNS. Conversely, HIV evolution in the CNS may be restricted in the relative absence of ARV and immune pressures, maintaining wild-type strains that are more susceptible, rather than more resistant, than those in plasma. This alternative would explain observations that therapy can suppress HIV replication in cerebrospinal fluid (CSF) even when it fails to do so in plasma. METHODS: To define inter-compartmental HIV differences, we serially sample plasma and CSF in individuals initiating or interrupting ARV therapy. After a change in therapy, matched blood and CSF samples are obtained within an hour interval up to nine times over 12 weeks. Each subject’s primary medical provider selects their ARV regimen. This report describes the course of an illustrative participant following interruption (TI) and re-initiation of zidovudine-lamivudine-nelfinavir. HIV RNA was measured by the RT-PCR (Roche Amplicor). Susceptibility to ARVs was measured with phenotype resistance testing (PRT, ViroLogic PhenoSense). Population genotyping was performed using GeneSeq HIV. Phylogenetic relationships were analysed using maximum parsimony and maximum likelihood methods. RESULTS: The subject interrupted therapy with HIV RNA levels of 6973 in plasma and <50 in CSF. Six matched pairs obtained over 82 days demonstrated that HIV replication peaked at 72 457 in plasma after 16 days and at 41 081 in CSF after 47 days. Baseline PRT indicated reduced susceptibility to abacavir, lamivudine and nelfinavir in plasma. Strains isolated from plasma and CSF had concordant phenotypes 16 days after TI. Susceptibility patterns diverged 47 days post-TI: the original pattern persisted in plasma but a new pattern was observed in CSF, demonstrating susceptibility to all drugs. Phylogenetic analysis indicated that the strains from CSF were monophyletic. Reinitiation of therapy resulted in suppression of replication in CSF, but not in plasma. CONCLUSIONS: Antiretroviral therapy susceptibility of HIV strains isolated from plasma and CSF can diverge during TI and may predict subsequent therapeutic response. In CSF, HIV replication returned more slowly but, paradoxically, reversion to strains with wildtype susceptibility occurred more quickly. Emergence of a monophyletic, fully susceptible strain in CSF, but not in plasma, suggests derivation from an archived CNS source. CNS-specific selection pressures may favour replication of such wild-type strains, which may explain why HIV replication is more readily suppressed there. By suppressing HIV replication in the CNS even when it fails in plasma, ARV therapy may have neuroprotective benefits in patients with few treatment options.

return to publications listing