A pharmacokinetic study of rifabutin and its interaction with antiretrovirals in African patients with TB-HIV co-infection.

Abstract

The management of HIV-associated tuberculosis (TB) is complicated by the pharmacokinetic interactions between rifampicin (RMP) and co-administered protease inhibitors (PIs) and non-nucleoside reverse transcriptase inhibitors. Rifabutin (RBT) is an alternative rifamycin, preferred in patients requiring PIs. Recent studies suggest the current recommended dose of RBT in combination with boosted lopinavir (LPV/r) is suboptimal and there are insufficient pharmacokinetic data evaluating the interaction between RBT coadministered with efavirenz (EFV) and nevirapine (NVP). Pharmacogenomic studies have shown that RMP concentrations are lower in patients from sub-Saharan Africa with polymorphisms of the SLCO1B1gene but there is currently no data on the pharmacogenetic determinants of RBT exposure. The pharmacokinetics of RBT were evaluated at two different doses in HIV co-infected patients before and after the introduction of LPV/r, EFV and NVPbased antiretroviral therapy (ART). After six weeks of standard TB therapy, RBT 300 mg daily was started for four weeks. Thereafter patients were randomized to receive either RBT 150 mg daily or RBT 150 mg three times a week (TPW) with LPV/r, RBT 300mg or 450mg with NVP or RBT- 450mg or 600mg with efavirenz. After four weeks on the first RBT dose, patients switched to the alternate dose and continued until the end of TB treatment. Serial RBT and 25-O-desacetylrifabutin (dRBT) concentrations were measured during a dose interval before patients switched RBT doses. The median AUC0-24 and Cmax, of RBT in patients taking 150mg RBT TPW was significantly reduced when compared to the other treatment arms. 86% of patients whilst on this intermittent RBT arm had an AUC0-24 < 4.5 μg.h/mL, level that has been associated with acquired rifamycin resistance. Rifabutin exposure was maintained within the range of AUCs that have been shown to prevent acquired rifamycin resistance (ARR) with 150mg daily dosing in combination with LPV/r. In addition, the combination of RBT with NVP 300mg resulted in significantly increased exposure of RBT, with significantly higher exposure observed with 600mg RBT. However, the combination of RBT 450mg with EFV resulted in RBT exposure lower than 300mg RBT given alone in the same patients, whereas RBT 600mg plus NVP results in bioavailability of RBT equivalent to 300mg given alone. Rifabutin was well tolerated at all doses. Only three grade 4 laboratory toxicities, elevated transaminases, neutropenia, and uveitis, possibly related to RBT were reported in patients taking NVP. SLCO1B1 rs4149032 C>T polymorphism occurs frequently in African patients in Durban and may be associated with low RBT bioavailability. These findings support recommendations for the higher dose of RBT in combination with LPV and EFV but not with NVP.