Baseline HIV drug resistance profile predicts response to ritonavir-saquinavir protease inhibitor therapy in a community setting

Objective: To determine whether baseline drug resistance assays could help to predict treatment failure with the protease inhibitor combination ritona vir-saquinavir. Methods: Baseline HIV-1 drug resistance was determined for 76 consecutive p atients who started treatment with the dual protease inhibitor combination ritonavir-saquinavir between September 1996 and June 1997 either alone or i n combination with other antiviral agents. Resistance to 10 different antiv iral agents was assessed by both phenotype (Virco Antivirogram) and genotyp e (Vircogen). Results: Resistance inferred from viral genotype was similar to measured ph enotypic resistance for both ritonavir and saquinavir (P < 0.01). Baseline drug resistance phenotype was predictive of poor virological response to th is dual protease inhibitor combination, despite the confounding effects of other antivirals. Patients were at least four times less likely to achieve a 0.5 log(10) decrease in plasma HIV RNA viral load if their viral isolates were resistant to ritonavir or saquinavir. Patients classified as resistan t to either drug using either method had median decreases in plasma viral l oad of 0.05 log(10) HIV RNA copies/ml or less, compared to > 0.8 log(10) fo r those with sensitive virus. Patients resistant to both drugs never achiev ed plasma viral loads < 100 000 copies/ml. As little as fourfold increases in baseline resistance appeared to be sufficient to compromise even dual pr otease inhibitor therapy. Conclusion: Baseline resistance to ritonavir or saquinavir or both was asso ciated with a poor antiviral response. Our data suggest that the measuremen t of drug resistance may assist in optimizing antiretroviral therapy in the clinic. (C) 1999 Lippincott Williams & Wilkins.