Failure of antiretroviral therapy (ART) has been a particular problem among HIV-1 infected illicit drug users. This failure has primarily been attributed to poor compliance with antiretroviral drug regimens. Studies from this laboratory have indicated that a higher viral mutation frequency that we have documented in injection drug users (IDU) and the resulting higher frequency of primary resistance mutations in pol might contribute to the poor response in this group. Genotypic resistance analysis using population sequencing has proved useful in guiding selection of antiretroviral therapy in individuals who have developed viral relapse after initial treatment. Our preliminary data analyzing up to 10 viral clones from protease inhibitor (PI) and non-nucleoside reverse transcriptase inhibitor (NNRTI) naive IDU have indicated that a strikingly and significantly high proportion of IDU subject-visits (28%) carry PI resistance mutations compared to the much lower proportion (8%) found in non-IDU. Almost none of these resistance mutations were detected using standard population genotyping techniques. Similar high levels of primary resistance have been found in a cohort of non-injection illicit drug users followed by investigators at Vanderbilt Medical School. This proposal hypothesizes that use of sensitive sequencing techniques prior to initiation of HAART will be predictive of rapid development of resistance, therefore enabling the development of personalized and more effective initial HAART regimens. Specifically in cohorts from the Johns Hopkins and Vanderbilt Schools of Medicine of 150 HAART-naive drug users who rapidly failed HAART and 150 comparable subjects for whom therapy was successful we will 1) Determine whether identification of resistance mutations by standard clonal analysis using the Sanger sequencing method to study the pol region from 20 HIV-1 clones from a single visit predicts risk of subsequent therapy failure better than population genotyping from that same visit 2) evaluate whether identification of resistance mutations by high throughput clonal analysis of relatively short sequences from the RT and protease regions (454 sequencing) predicts risk of subsequent therapy failure better than: i) standard population genotyping and ii) analysis of 20 clones (sequenced using the Sanger method) in which both NRTI and NNRTI or PI resistance mutations can be identified on the same viral clone and 3) Evaluate, using 454 sequencing technology, the frequency of clonal resistance needed for clones with PI or NNRTI resistance to predict increased risk of rapid development of therapy failure. The results of this study could provide a new standard of care for initiation of HAART and could greatly reduce the financial and social impact of HAART failure. This study is designed to evaluate new technologies that will render anti-HIV-1 drug therapy more effective. This new technology will permit better characterization of the viral strains that are infecting an individual so that the therapy can be specifically targeted to those viruses.
