Background
Tuberculosis (TB) remains the leading infectious cause of death, especially in people with HIV. Metformin is a promising immunometabolic host-directed therapy that might improve TB treatment. However, drug-drug interactions should be considered when including metformin in the current TB and HIV treatment regimens. Metformin has been reported to lower rifampicin and isoniazid exposure. In addition, dolutegravir increases metformin exposures while rifampicin reduces dolutegravir exposures, requiring twice-daily dosing. Here, we report the pharmacokinetics (PK) of rifampicin, isoniazid and pyrazinamide co-administered with or without metformin from METHOD, a Phase II trial of metformin adjunctive to standard 1st-line TB treatment in adults with HIV/TB.
Methods
At enrolment, all participants initiated standard 1st-line TB treatment. Half of the participants were randomised to also receive adjunctive metformin, 500 mg OD for 1 week, then 500 mg BD until week 12. ART-naïve participants were started on dolutegravir-based ART within 8 weeks. PK sampling (intensive and sparse) was conducted at week 5. Concentration–time data were analysed using nonlinear mixed-effects modelling.
Results
Eighty adults (43 on metformin) provided 316 samples, median weight 60.8 kg (IQR 53.9–67.8 kg), 29 (37.2%) were female. At the time of the PK visit, 63 (79.0%) were receiving dolutegravir-based ART. Rifampicin and pyrazinamide were described by one-compartment models with linear elimination and their typical values of CL were 15.8 L/h and 3.82 L/h, respectively. Isoniazid was described by a two-compartment model with elimination via hepatic extraction, with a mixture model estimating the clearance for slow (10.6 L/h) and fast (27.9 L/h) acetylators. Isoniazid bioavailability was 15.9% (95% CI: 4.45– 26.4; p<0.009) lower in the metformin arm (observed AUC of 8.60 with vs. 10.8 mg.h/L without metformin). For rifampicin, the bioavailability was 24.0% (95% CI: 7.27–37.8%, p<0.007) lower in the metformin arm (observed AUC of 29.0 with vs. 37.5 mg.h/L without metformin).
Conclusions
Participants on metformin had lower isoniazid and rifampicin bioavailability, resulting in reduced overall exposure compared to participants not on metformin. This effect may be especially relevant for rifampicin, given its critical role in TB therapy.