Phenotypic analysis of HIV-1 drug resistance during highly active antiretroviral therapy (HAART)

Highly active antiretroviral therapy (HAART) can not eliminate HIV-1 production in infected patients. Persistent residual viremia under HAART has raised concerns about the evolution of HIV-1 drug resistance and the eventual failure of antiviral treatment. To study HIV-1 drug resistance in patients treated with HAART, we developed a novel phenotypic assay that can measure HIV-1 replication in the absence and presence of drug combinations in vitro over a broad dynamic range of up to 4 logs.; Applying this method to analyze HIV-1 replication in the setting of virological failure under HAART, we demonstrated, in agreement with clinical studies, that continued non-suppressive HAART can confer residual treatment benefit. This treatment benefit can be divided into two additive effects: residual suppression of replication of resistant viruses and selection of resistant viruses that has reduced replication capacity.; In the setting where HAART persistently supresses plasma viral load below the limit of detection (50 HIV-1 RNA copies per ml), we tested whether this level of viral suppression can arrest the evolution of HIV-1 drug resistance by cross-sectionally measuring the heterogeneity of drug-resistance phenotypes of plasma viruses in HAART treated patients. When HAART has failed to suppress high level viral replication, we observed homogeneous resistant phenotypes of plasma viruses, confirming that continuous viral replication under drug selection will lead to deterministic viral evolution.; In the course of this study, we observed that large phenotypic differences between wild type and resistant viruses in response to individual drugs can be obscured by HAART. We analyzed the pharmacokinetic basis of this masking effect of HAART. In the case of combinations of nucleoside reverse transcriptase inhibitors (NRTI) and non-nucleoside reverse transcriptase inhibitor (NNRTI), we showed that such a masking effect was mainly determined by the potency of the new drug added to combination. If the new drug was at equal potency as the original drug, the phenotypic differences between the wild type and resistant viruses was maintained. If the new drug was much more potent than the original drug, the phenotypic difference was obscured. The drug combination effect between NRTI and NNRTI can be predicted by the median-effect model, which generally predicted that drug combination effect was determined the relative potency of individual drugs in combination regimen and should be dominated by the exceptionally potent drug within the regimen. (Abstract shortened by UMI.)...