Objective: To evaluate the pharmacokinetics and pharmacogenomics of efavirenz (EFV) and lopinavir/ritonavir (LPV/RTV) in HIV-infected persons requiring hemodialysis.
Design: Prospective, observational study of HIV-infected hemodialysis patients receiving one 600 mg tablet daily of EFV (N = 13) or three 133.3/33.3 mg capsules twice daily of LPV/RTV (N = 13).
Methods: Twenty-four-hour EFV and 12-h LPV/RTV pharmacokinetics were assessed. Geometric mean ratios were calculated using historical controls with normal renal function. The effects of several candidate gene polymorphisms were also explored.
Results: The geometric mean [95% confidence interval (CI); percentage of coefficient of variation (% CV)] Cmin, Cmax, and area under the curve (AUC) for the EFV group were 1.81 microg/ml (0.93, 3.53; 103%), 5.04 microg/ml (3.48, 7.29; 72%), and 71.5 microg h/ml (43.2, 118.3; 93%), respectively. These parameters were 2.76 microg/ml (1.86, 4.11; 53%), 8.45 microg/ml (6.41, 11.15; 52%), and 69.6 microg h/ml (55.6, 87.2; 37%) for LPV and 0.08 microg/ml (0.05, 0.14; 63%), 0.58 microg/ml (0.44, 0.76; 41%), and 3.74 microg h/ml (2.91, 4.80; 37%) for RTV. The AUC geometric mean ratios (90% CI) for EFV, LPV, and RTV were 132% (89, 197), 81% (67, 97), and 92% (76, 111), respectively. LPV Cmin was lower than expected in the hemodialysis group. Higher EFV concentrations were associated with the CYP2B6 516G>T polymorphism.
Conclusion: The pharmacokinetics of EFV and LPV/RTV in hemodialysis suggests that no dosing adjustments are necessary in treatment-naive patients. As HIV-infected hemodialysis patients are disproportionately black, the increased frequency of the CYP2B6 516G>T polymorphism may lead to higher EFV levels. The potentially lower LPV trough levels in this population suggest that LPV/RTV should be used with caution in protease-inhibitor-experienced patients.