Background:  After successful induction treatment with lopinavir/ritonavir (LPV/r) + zidovudine/stavudine (AZT/3TC), LPV/r monotherapy maintained plasma viral load suppression <50 copies/mL in a majority of subjects treated with monotherapy for as long as 72 weeks in Study M03-613. However, subjects receiving LPV/r monotherapy experienced more confirmed plasma viral load rebound above 50 copies/mL than subjects receiving efavirenz (EFV) + zidovudine (AZT)/3TC. We sought to determine predictors of loss of virologic response to LPV/r monotherapy in Study M03-613.

Methods:  Of 104 subjects randomized to LPV/r+AZT/3TC, 92 achieved 3 consecutive plasma viral load <50 copies/mL, deintensified to LPV/r monotherapy, and were followed for a median of 68 weeks on monotherapy. Endpoints included confirmed plasma viral load rebound >50 or 500 copies/mL. Potential predictors of response (subject demographics, baseline plasma viral load and CD4 count, and adherence [4-day subject recall]) were assessed using the Cox proportional hazards model. Adherence was also assessed as a time-dependent covariate.

Results:  Subjects who deintensified to LPV/r monotherapy were primarily white (59%) and male (78%), with mean baseline plasma viral load 4.9 log₁₀ copies/mL and CD4 count 226 cells/mm³. Average adherence by self-reported 4-day recall was 97.4%; 32 subjects had confirmed plasma viral load rebound >50 copies/mL (12 with confirmed rebound >500 copies/mL). By multivariable Cox proportional hazards model, subjects reporting at least 1 missed LPV/r dose had a 2.7-fold higher risk of confirmed plasma viral load rebound >50 copies/mL (p = 0.01). Risk of rebound decreased 17% for each additional 50 CD4 cells/mm³ at baseline (p = 0.01). Median baseline CD4 count in subjects with versus without rebound was 169 vs 235 cells/mm³. Baseline plasma viral load and demographic variables were not significantly associated with plasma viral load rebound. Both adherence and baseline CD4 count were marginally associated with confirmed plasma viral load rebound >500 copies/mL (p = 0.07, p = 0.06, respectively).  When adherence was assessed as a time-dependent covariate, subjects reporting at least 1 missed dose in the preceding 4 days had a 16-fold higher risk of confirmed plasma viral load rebound >500 copies/mL.

Conclusions:  Adherence was associated with confirmed virologic rebound on LPV/r monotherapy. Baseline CD4 count may be a determinant in the success of an induction maintenance strategy.