Integrating human immunodeficiency virus and tuberculosis drug treatment.

Abstract

The human immunodeficiency virus (HIV) and tuberculosis (TB) epidemics are major global public health challenges. Worldwide, approximately 42% of TB patients are also co-infected with HIV, and sub-Saharan Africa (SSA) is home to the majority of the world’s infections of both HIV and TB. Dual infection has been shown to be associated with a higher risk of death. Integrating drug treatment for both diseases is therefore essential to improve survival. However, drug interactions between antiretroviral therapy (ART) and anti-TB medication remain a challenge to effective treatment integration. Although several drug interactions have been identified, only some are clinically relevant. The impact of significant interactions on public health outcomes is expected to be greatest when large numbers of patients are prescribed interacting drugs. Efavirenz (EFV) is the most commonly prescribed nucleoside reverse transcriptase inhibitor (NNRTI) component of first line ART in sub-Saharan Africa, particularly when rifampicin (RIF) based TB treatment is co-administered. RIF is known to up-regulate cytochrome P450 (CYP450) drug metabolizing enzymes resulting in decreased exposure to concomitantly administered drugs that utilize similar metabolic pathways. Therefore, the concomitant use of EFV with RIF would be expected to increase EFV clearance while absorption of TB drugs may also be compromised by advanced HIV disease. The efficacy of both TB and HIV treatment may thus be compromised by pharmacokinetic interactions, while more recent evidence also implicates genetic variation in drug metabolism as a predictor of drug exposure. To understand the significance of the EFV-RIF interaction better in a South African population, the pharmacokinetics of EFV during and after RIF-based TB treatment were investigated as an ancillary study of the ‘Starting Tuberculosis and Antiretroviral Therapy’ (START) trial (CAPRISA 001: NCT00091936). Participants were randomized to receive both ART and TB treatment simultaneously (integrated arm) or to initiate ART only on completion of TB treatment (sequential arm). In both arms, the ART regimen included once daily enteric-coated didanosine (400 mg for participants >60 kg; 250 mg for participants <60 kg), lamivudine 300mg and efavirenz. Based on the expected drug interactions, when EFV was administered in the presence of TB treatment, participants weighing less than 50kg received 600mg and those weighing 50kg or more received 800mg daily. After TB treatment was successfully completed, all patients received EFV 600mg. Blood samples for trough EFV plasma concentrations were obtained at the end of months 1, 2 and 3 during TB treatment and at the same time points after TB treatment was successfully completed. Additionally, approximated peak RIF concentrations were measured 2.5 hours post-dose at the end of months 1, 2 and 3 of TB treatment. The influence of single nucleotide polymorphisms, in CYP2B6, CYP2A6, and UGT2B7 on EFV concentrations, and in drug transporter genes (SLCO1B1) on RIF concentrations, was assessed post-trial from stored peripheral blood mononuclear cell (PBMC) samples. EFV concentration-time data were analyzed using a population pharmacokinetic nonlinear mixed effects model (NONMEM) to quantify the impact of RIF-based TB treatment on EFV clearance. Unexpectedly, there was an overall 29.5% reduction in EFV clearance during TB treatment. A bimodal distribution of EFV apparent clearance (CL/F) was evident and indicated that slow EFV metabolisers accounted for 21.9% of the population. EFV clearance after oral administration in fast metabolisers was 11.5 L/h/70kg off TB treatment and 7.6 L/h/70kg when on TB treatment. In slow metabolisers, however, the clearance estimates were 2.9 and 4.3 L/h/70kg in the presence and absence of TB treatment respectively. Building on the findings of the NONMEM analysis and in response to the US FDA prescribing change in 2012, that approved an EFV dose increase from 600mg to 800mg in patients weighing 50kg and more when on concomitant RIF, the presence and influence of pharmacogenetic polymorphisms of the CYP450 enzyme system on NNRTI plasma exposure during and after TB co-treatment and the effect of increasing the EFV dose was investigated. During TB treatment, median (IQR) EFV Cmin was 3.2 (2.6-6.3) μg/mL and 3.3 (2.4-9.5) μg/mL in the EFV 800mg and 600mg groups respectively, while off TB treatment Cmin was 2.0 (1.4 - 3.5) μg/mL. The frequency of the CYP2B6 *1, *6 and *18 haplotypes was 18.5%, 38.9% and 25.9% respectively. Polymorphisms in all three CYP2B6 genes studied (516T-785G-983C) were present in 11.1% of patients. Median (IQR) EFV concentrations in patients with the three mutations were 19.2 (9.5-20) μg/mL and 4.7 (3.5-5.6) μg/mL when on and off TB treatment. TB treatment, composite genotypes CYP2B6 516 GT/TT, CYP2B6 983 TC/CC or being a CYP2A6*9B carrier predicted median EFV Cmin > 4 μg/mL. Therefore, increasing the EFV dose to 800mg during TB treatment is unnecessary in African patients with these polymorphisms. As a critical component of first line TB treatment concerns about sub-optimal TB drug bioavailability were examined for RIF. The influence of drug transporter gene polymorphisms on RIF concentrations was also assessed. Median RIF (IQR) C2.5hr was found to be 3.6 (2.8-5.0) μg/mL while polymorphism frequency of the SLCO1B1 (rs4149032) drug transporter gene was high (0.76) and was associated with low RIF concentrations as was male gender and having a low haemoglobin. Increased RIF dosage warrants urgent consideration in African TB-HIV co-infected patients. In conclusion, concomitant RIF-containing TB treatment unexpectedly reduced EFV CL/F with a corresponding increase in EFV exposure. Polymorphisms of EFV metabolizing enzymes were frequent in this population and contribute to this outcome. While in South Africa where TB-HIV co-treatment is associated with elevated EFV concentrations, peak RIF concentrations were alarmingly low and well below the recommended target range of 8 to 24 μg/mL. Increased RIF dosage may be warranted in African TB-HIV co-infected patients whilst the need for EFV dose increase is not supported by these data. Recommendations for public health benefit, in this generalized epidemic in South Africa, include the consideration of an EFV dose reduction as a cost saving to improve life-long treatment sustainability, and a RIF dose increase to curb TB treatment failure and future development of multiple-drug resistant (MDR) TB.