Impact of different HIV resistance interpretation by distinct systems on clinical utility of resistance testing

Genotypic assays are widely used tools for determining HIV-1 drug resistance and for guiding treatment. Several systems have been developed to interpret the complex influence of amino acid substitutions in HIV reverse transcriptase or protease on the phenotypic susceptibility or clinical response to the 18 available antiretroviral agents. In this review we analyse both studies comparing interpretations by different systems and studies showing correlation of interpretations with clinical outcome, in order to identify discordance and how this may affect prediction of subsequent therapy outcomes.

During the last year, several studies analysing interpretation systems, individually or comparatively, have shown substantial variability of the predicted drug activities and therapeutic outcomes. Discrepant interpretation was detected mostly for nucleoside reverse transcriptase inhibitors and rarely for non-nucleoside reverse transcriptase inhibitors. Better correlation with treatment outcome was found with most recently updated systems, while a weaker prediction was found with systems interpreting activity of nucleoside reverse transcriptase inhibitors solely on the basis of phenotypic susceptibility. Virological, patient-related and treatment-related factors can all affect the results of systems' clinical validations. Refinement of resistance interpretation is possible by introducing rules derived from genotype-outcomes correlation or, at least for protease inhibitors, genotype-phenotype correlation.

Papers showing clinical validation of the available interpretation systems are presented with a critical view to help the readers' evaluation of their possible use. There is a need for developing a consensus towards common interpretations. Large clinical and virological databases with quality data will be useful for future improvements.