Predicting HIV RNA virologic outcome at 52-weeks follow-up in antiretroviral clinical trials

Objective: To determine the ability of intermediate plasma viral load (pVL) measurements to predict virologic outcome at 52 weeks of follow-up in clin ical trials of antiretroviral therapy. Methods: Individual patient data from three clinical trials (INCAS, AVANTI- 2 and AVANTI-3) were combined into a single database. Virologic success was defined to be plasma viral load (pVL) <500 copies/ml at week 52. The sensi tivity and specificity of intermediate pVL measurements below the limit of detection, 100, 500, 1000, and 5000 copies/ml to predict virologic success were calculated. Results: The sensitivity, specificity, and positive and negative predictive values of a pVL measurement <1000 copies/ml at week 16 to predict virologi c outcome at week 52 were 74%, 74%, 48%, and 90%, respectively, for patient s on double therapy. For patients on triple therapy, the sensitivity, speci ficity, and positive and negative predictive values of a pVL measurement <5 00 copies/ml at week 16 to predict virologic outcome were 68%, 68%, 80%, an d 47%, respectively. Conclusions: For patients receiving double therapy, a poor virologic result , at an intermediate week of follow-up is a strong indicator of virologic f ailure at 52 weeks whereas intermediate virologic success is no guarantee o f success at 1 year. For patients on triple therapy, disappointing intermed iate results do not preclude virologic success at 1 year and intermediate s uccesses are more likely to be sustained.