Rohan Benecke 1, Veshni Pillay-Fuentes Lorente 1, Roland van Rensburg 1
1 Stellenbosch University (Cape Town, South Africa)
Introduction: The antiretroviral drug dolutegravir (DTG) inhibits the renal organic cation transporter 2, the primary elimination pathway for the glucose-lowering drug metformin. In a healthy volunteer study by Song et al [1], metformin area under the concentration-time curve (AUC) increased by 79% when administered concomitantly with DTG 50 mg/day, informing subsequent guideline recommendations to limit metformin to 1,000 mg/day. The largest target population who will require metformin and DTG co-administration are women with HIV (WWH) with dysglycaemia who are often obese. Our previous non-compartmental analysis replicating the Song study [2] showed that among 15 obese WWH, metformin and DTG AUC0–24 was 40.9% and 54.4% lower, respectively, compared to Song. Metformin and DTG clearance were 1.7- and 2.2-fold higher, respectively. Accordingly, the metformin dose restriction may not be appropriate in obese WWH.
Objectives:
We developed a mechanistic population pharmacokinetic (popPK) model for metformin to:
• Asses DTG AUC on metformin clearance, and
• Simulate dosing scenarios up to metformin 2,000 mg/day to explore exposures beyond the current dose restriction
Methods: South African, obese, adult WWH receiving once-daily extended-release (XR) metformin 1,000 mg and DTG 50 mg contributed rich plasma concentration-time profiles (pre-dose and 1, 3, 4, 6, 8, 10 and 12 hours post-dose). Size descriptors for body weight (total body mass, fat mass, and lean body mass [LBM]) were obtained from direct imaging. DTG exposure and renal estimated glomerular filtration rate (eGFR) were evaluated as covariates on apparent clearance (CL/F). PopPK modelling was performed in Monolix (version 2023R2) using SAEM [3]. Steady-state simulations (1,000 individuals/scenario) explored metformin 1,000, 1,500 and 2,000 mg/day across a wide body mass index (BMI) range (15–60 kg/m2); metformin AUC0–24 and Cmax were summarised, and Cmax >5 mg/L was used as the reported conservative concentration-based safety threshold [4]. To contextualise our findings, we performed comparative simulations using the popPK model developed by Bardin et al [5] for immediate-release, twice-daily metformin in a population without DTG. The Bardin model included participants with BMI 20.5−51 kg/m2, and used a one-compartment model with first-order absorption and dose-dependent bioavailability.
Results: Thirteen participants contributed evaluable metformin PK profiles. Mean BMI was 45.6±6.5 kg/m2, and eGFR 86.4±11.6 mL/min/1.73m2. Metformin PK was described by a one-compartment model with first-order elimination; the XR input was described by zero-order release into an absorption compartment followed by first-order absorption. LBM was selected as the optimal size descriptor and used for allometric scaling of CL/F (fixed exponent 0.75) and volume of distribution (V/F; fixed exponent 1). LBM explained a substantial proportion of interindividual variability in clearance, reducing it from approximately 40% in the base model to 24% in the final model. The final parameter estimates with RSE were ka 0.124 h⁻¹ (6%), CL/F 70.2 L/h (6.4%), V/F 35.7 L (29.6%), and a zero-order XR tablet release duration of 5.15 h (4.2%). Notably, DTG exposure and eGFR did not significantly improve model fit. Simulations across the BMI range showed minimal impact of BMI on metformin exposure, likely due to low LBM variance throughout the broad BMI strata. At metformin 1,000 mg/day, median AUC0–24 was approximately 11.7–12.4 mg.h/L and Cmax 1.25–1.31 mg/L across the BMI range (0% above the 5 mg/L safety threshold). At 1,500 mg, median AUC0–24 increased to 17.8–18.6 mg.h/L and Cmax to 1.89–1.98 mg/L (0% >5 mg/L). At 2,000 mg, median AUC0–24 was 23.3–25.0 mg.h/L with Cmax 2.49–2.64 mg/L. At this dose, only 0.3–1.1% exceeded the 5 mg/L threshold. Comparative simulations with the Bardin model at matched daily doses showed substantially overlapping dose-exposure relationships in the 1,000–2,000 mg/day dosing range.
Conclusions: In obese WWH, DTG was not a significant covariate on metformin exposure. Our simulations suggest that metformin doses above the 1,000 mg/day restriction increase exposure in a near dose-proportional manner while rarely exceeding the conservative Cmax safety threshold at 2,000 mg/day. The similar exposures found in our comparative simulations in a population not on DTG increase the credibility of our model and results. These findings may indicate that obesity-related increases in metformin clearance offsets potential DTG inhibition. Future studies should evaluate a broader range of LBM, renal function, and DTG exposures, and optimise metformin exposure to glycaemic response to support population-specific clinical dosing recommendations.
References:
[1] Song IH, Zong J, Borland J, et al. The effect of dolutegravir on the pharmacokinetics of metformin in healthy subjects. J Acquir Immune Defic Syndr. 2016 Aug 1;72(4):400-407.
[2] Van Rensburg R, Kellermann T, Pillay-Fuentes Lorente V, et al. Reduced Metformin Concentrations in Obese Women with Human Immunodeficiency Virus Treated with Dolutegravir. J Infect Dis. 2025;232(6):e1012-1021.
[3] MonolixSuite, 2025, Available from: https://www.simulations-plus.com/software/monolix/ (Accessed 23 February 2026)
[4] Kuan IH, Wilson LC, Leishman JC, et al. Metformin doses to ensure efficacy and safety in patients with reduced kidney function. PLoS One. 2021;16(2):e0246247.
[5] Bardin C, Nobecourt E, Larger E, et al. Population pharmacokinetics of metformin in obese and non-obese patients with type 2 diabetes mellitus. Eur J Clin Pharmacol. 2012;68:961–968.
Reference: PAGE 34 (2026) Abstr 11859 [www.page-meeting.org/?abstract=11859]
Poster: Drug/Disease Modelling - Infection