Outcome of a second-line protease inhibitor-containing regimen in patientsfailing or intolerant of a first highly active antiretroviral therapy

The outcome of second-line protease inhibitor (PI)-containing highly active antiretroviral therapy (HAART) was investigated in 263 patients who were f ailed by (n = 148) or intolerant of (n = 115) a first HAART regimen. The en dpoints were virologic failure (decline in HIV RNA <1 log(10) copies/ml aft er greater than or equal to 2 months) and discontinuation due to intoleranc e/toxicity. During a median follow-up of 483 days (33-1087 days), 154 patie nts (59%) discontinued the second regimen, 86 (33%) because of intolerance/ toxicity; another 135 patients (51.3%) showed virologic failure. Independen t factors associated with virologic failure (Cox's model) were 7 to 12 mont hs of first HAART (hazard ratio [HR] 1.70 versus less than or equal to 6 mo nths: 95% confidence interval [CI], 1.08-2.70) and gender (HR 1.58 males ve rsus females: 95% CI, 1.04-2.30); the negatively associated factors were ad vanced age (HR 0.61 >34 years versus less than or equal to 34 years: 95% CI , 0.42-0.88), a saquinavir-containing first HAART (HR 0.57 versus indinavir : 95% CI, 0.34-0.93) and change due to intolerance/toxicity (HR 0.58 versus failure: 95% CI, 0.35-0.98). The independent variables predictive of disco ntinuation due to intolerance/toxicity were the reason for switching (HR 1. 79 intolerance versus failure: 95% CI, 1.02-3.16) and the first protease in hibitor (PI) regimen (HR 0.42 ritonavir versus indinavir: 95% CI, 0.22-0.80 ). Given that patients who are failed by a first regimen are at high risk o f having rescue therapy fail as well, second-line regimens including therap ies directed by testing of drug resistance patterns of clinical viral isola tes are warranted. Patients experiencing toxicity due to a first PI-contain ing regimen are at risk of toxicity to other PIs and should be addressed to PI-sparing HAART.