Nonlinear mixed-effects modelling of drug-drug interactions between antiretroviral therapy and tuberculosis treatment

Human immunodeficiency virus (HIV) remains a significant global health challenge that affected approximately 39 million individuals in 2022, with majority residing in Africa. Among people with HIV (PWH), co-infection with tuberculosis (TB) is a leading cause of death. However, the concurrent treatment of HIV and TB often results in drug-drug interactions (DDIs), mediated especially by rifampicin, a key component of the TB regimen and potent enzyme and transporter inducer. These DDIs may compromise treatment safety and efficacy, potentially leading to therapeutic failure and increased risk of drug resistance. In this thesis, we utilized non-linear mixed effects modelling and data from studies in PWH and healthy volunteers to characterize DDIs between first- and second-line antiretroviral (ARV) and anti-TB drugs. Additionally, we performed simulations to assess treatment target attainment following current dosing recommendations in PWH. Our pharmacokinetic model of standard- and high-dose rifampicin in PWH identified lower bioavailability of the top-up capsule formulation as the cause of lower-than-expected drug exposures in participants on high-dose rifampicin. Furthermore, the reduced dolutegravir exposures in participants on concurrent high-dose rifampicin, compared to those on the standard-dose, were attributed to reduced bioavailability rather than enhanced clearance. Notably, our simulations demonstrated that doubling the dosing frequency of dolutegravir effectively counteracted the DDI with both standard- and high-dose rifampicin. Secondly, we characterized the DDI between ritonavir-boosted atazanavir (ATV/r) and rifampicin, both in plasma and within peripheral blood mononuclear cells (PBMCs). Rifampicin increased the clearance of ATV/r by threefold, and doubling the dosing frequency of ATV/r was sufficient to counteract this interaction and restore treatment target attainment. Notably, rifampicin did not affect atazanavir equilibration or accumulation in PBMCs, suggesting that plasma studies can reliably reflect intracellular processes. We also applied our model to an external dataset, estimating a twofold decrease in atazanavir clearance, likely due to ritonavir co-administration. Lastly, we found clofazimine, a second-line drug resistant TB (DR-TB) drug, to increase the clearance of levofloxacin by 15% but not affect the pharmacokinetics of cycloserine, linezolid, or isoniazid. This confirmed that clofazimine can be safely co-administered with other DR-TB drugs, as it poses minimal risk of significant DDIs. In conclusion, non-linear mixed effects modelling can be used to evaluate DDIs, and we recommend its incorporation in routine dose optimization and therapeutic drug monitoring programs to enhance treatment outcomes.

Grantor dc:publisher.institution

Department of Medicine

Year dc:date.issued

2025

Author dc:creator

• Kengo, Allan

Advisors dc:contributor.advisor

• Denti, Paolo
• Resendiz Galvan, Juan Eduardo

• Drug
• Therapy
• HIV
• TB

Language dc:language.iso

en

Handle dc:identifier.uri

http://hdl.handle.net/11427/42352

OAI identifier oai:identifier

oai:open.uct.ac.za:11427/42352