Background:  Antiretroviral regimen complexity may contribute to increased toxicity and virologic failure. A5116 assessed safety and efficacy of simplified class-sparing regimens (protease inhibitor [PI]- and nucleoside [NRTI]-sparing regimens) in subjects with long-term virologic suppression on a prior antiretroviral regimen.

Methods:  A5116 was a randomized, open-label study of lopinavir/ritonavir (LPV/r) (533 mg/133 mg twice daily) + efavirenz (EFV) (600 mg once daily) vs EFV+2 NRTI in subjects who received at least 18 months of a 3- or 4-drug PI- or NNRTI-based regimen as a first regimen, had plasma HIV-1 RNA levels £ 200 copies/mL and no documented phenotypic resistance. Due to low virologic failure rates, the original primary study endpoint of time to confirmed virologic failure (plasma HIV-1 RNA > 200 copies/mL) was revised during the study to a combined study endpoint of virologic failure or toxicity-related discontinuation of any component of randomized treatment. Time-to-event distributions were estimated using the method of Kaplan-Meier and compared with log-rank tests stratified by prior ACTG 388 regimen (study of 3- or 4-drug indinavir-based-regimens in advanced HIV disease) and non-388 treatment. Analyses were intent-to-treat.

Results:  A total of 236 subjects (118 per arm) were enrolled and followed a median of 110 weeks. For the EFV+2 NRTI arm, 92 (78%) received zidovudine+lamivudine (AZT+3TC) and 22 (19%) stavudine (d4T) +3TC. Overall, 87% were men; 57% were white, 24% black, and 18% Hispanic. Median age was 43 years. Median baseline CD4 cell count was 475 cells/mm³. A higher rate of treatment discontinuation occurred in the LPV/r+EFV arm (p < 0.001). Overall, 16 (7%) of the endpoints were virologic failure and 25(11%) were toxicity-related treatment discontinuations. Time to the combined study endpoint was longer in the EFV+2 NRTI arm (p = 0.001). There was a trend toward a higher VF rate in the LPV/r+EFV arm (p=0.088 intent-to-treat, p=0.064 censoring subjects who modified treatment). Toxicity-related endpoints occurred more often in the LPV/r+EFV arm than the EFV+NRTI arm (17% vs 5%, respectively, p = 0.002) due mainly to differences in triglyceride levels. Higher triglyceride and cholesterol levels were seen in the LPV/r+EFV arm with no effect seen on glucose or insulin levels. During long-term follow-up, higher CD4 cell counts occurred in the LPV/r+EFV arm, but differences did not reach significance.

Conclusions:  Treatment with EFV+2 NRTI led to lower rates of regimen failure than treatment with LPV/r+EFV, due to decreased toxicity and a trend toward reduced virologic failure.

Keywords: class sparing combination regimens; long term suppressive simplified regimens ; Lopinavir/ritonavir / efavirenz