Objective: To evaluate fosamprenavir/lopinavir (LPV)/ritonavir (RTV), fosamprenavir/RTV, or LPV/RTV in antiretroviral treatment-experienced patients. Lack of drug interaction data prompted a pharmacokinetic substudy to minimize subject risk.
Design: Multi-center, open-label, selectively randomized, steady-state pharmacokinetic study in HIV-infected subjects.
Methods: A planned independent interim review occurred after at least eight subjects were randomized to each arm. Subjects received twice daily LPV/RTV 400/100 mg (arm A; n = 8); fosamprenavir/RTV 700/100 mg (arm B; n = 8) or LPV/RTV/fosamprenavir 400/100/700 mg (arm C; n = 17). Plasma samples were collected over 12 h between study weeks 2 and 4. Pharmacokinetic parameters were compared based on a one-sided t-test on log-transformed data with a Peto stopping boundary (P < 0.001).
Results: Amprenavir mean area under the curve over 12 h (AUC0-12 h) and concentration at 12 h (C12 h) (microg/ml) were, respectively, 42.7 microg x h/ml (range, 33.1-55.1) and 2.4 microg/ml (range, 1.4-3.2) in arm B and 17.4 microg x h/ml (range, 4.6-41.3) and 0.9 microg/ml (range, 0.2-2.7) in arm C: geometric mean ratio (GMR) arm C:B was 0.36 [99.9% upper confidence boundary (UCB), 0.64] and 0.31 (99.9% h UCB, 0.61), respectively (P < or = 0.0001). Lopinavir AUC0-12 h and C12 h were, respectively, 95.3 microg x h/ml (range, 60.3-119.3) and 6.3 microg/ml (range, 2.2-9.2) in arm A and 54.4 microg x h/ml (range, 23.5-112.2) and 3.0 microg/ml (range, 0.4-7.9) in arm C: GMR arm C:A of 0.52 (99.9% UCB, 0.89) and 0.39 (99.9% UCB, 0.98), respectively (P < or = 0.0008). Ritonavir exposure was not significantly different between arms.
Conclusion: APV and LPV exposures are significantly reduced using LPV/RTV/fosamprenavir, possibly increasing the risk of virologic failure. Consequently, A5143 was closed to enrollment.