Predicting antiretroviral drug resistance from the latest or the cumulative genotype.

BACKGROUND This study evaluates the added benefit when estimating antiretroviral drug resistance of combining all available resistance test results in a cumulative genotype relative to using the latest genotype alone. METHODS The prevalence of resistance and genotypic sensitivity scores (GSS) predicted by the latest and the cumulative genotype, together with virological outcomes after the latest genotype, were measured in treatment-experienced patients who underwent ≥2 resistance tests in 1999-2008. RESULTS Comparing the latest with the cumulative genotype in 227 patients, 4 (1.7%) versus 0 (0.0%) showed no major resistance mutations, whereas 74 (32.6%) versus 46 (20.3%), 88 (38.8%) versus 76 (33.5%) and 61 (26.9%) versus 105 (46.3%) showed single-class, dual-class and triple-class resistance mutations, respectively. The median (IQR) number of fully or partially active drugs was 6 (5-6) versus 5 (4-6) for the nucleoside/nucleotide reverse transcriptase inhibitors, 3 (1-3) versus 1 (1-3) for the non-nucleoside reverse transcriptase inhibitors and 7 (7-7) versus 7 (7-7) for the protease inhibitors, respectively. Among 163 patients who started a new regimen after the latest genotype, both the latest and the cumulative GSS were predictive of early (≤24 weeks) virological responses. The GSS decreased by median 1 unit (IQR 0.5-1.0) in the cumulative genotype and larger differences relative to the latest genotype corresponded to smaller decreases in viral load. CONCLUSIONS The cumulative genotype offers a more comprehensive evaluation of the burden of resistance. This approach can guide small but appreciable improvements in the selection of antiretroviral regimens for treatment-experienced patients.