Kidney Week

Abstract: TH-PO0830

Kidney Transporter-Informed Physiologically Based Pharmacokinetic Modeling: A Novel Approach to Characterize Metformin Disposition in Critically Ill Patients with Sepsis

Session Information

• Pharmacology (PharmacoKinetics, -Dynamics, -Genomics)
  October 22, 2026 | Location: Exhibit Hall A, Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Pharmacology (PharmacoKinetics, -Dynamics, -Genomics)

• 1900 Pharmacology (PharmacoKinetics, -Dynamics, -Genomics)

Authors

• Zhao, Yuqian, University of Pittsburgh, Pittsburgh, Pennsylvania, United States

Yuqian Zhao, MS, MSc

• Gómez Danies, Hernando, University of Pittsburgh, Pittsburgh, Pennsylvania, United States

Hernando Gómez Danies, MD, MPH

• Nolin, Thomas D., University of Pittsburgh, Pittsburgh, Pennsylvania, United States

Thomas D. Nolin, PharmD, PhD, FASN

Background

Metformin shows promise for preventing sepsis-associated acute kidney injury (SA-AKI). The LiMiT-AKI study (NCT05900284) is assessing metformin safety in critically ill patients with sepsis; but optimal dosing is unknown. Metformin is eliminated unchanged through the kidney via OCT2, MATE1, and MATE2K transporters, which may be altered in sepsis. We aimed to develop a physiologically based pharmacokinetic (PBPK) model integrating preclinical kidney transporter expression to characterize metformin PK and support future dose selection in this population at risk for SA-AKI.

Methods

Kidney mRNA expression of OCT2, MATE1, and MATE2K was quantified by RT-qPCR in mice with cecal ligation and puncture (CLP) induced sepsis and controls then compared by unpaired t-test. A metformin PBPK model was developed and validated using Simcyp V25 (Certara), then extrapolated to septic patients by incorporating eGFR stratification, sepsis-related physiological changes (cardiac index, hematocrit, albumin, α1-acid glycoprotein), and CLP-induced transporter expression changes. Simulated steady-state C_max and AUC_0-12h were compared to observed values from the LiMiT-AKI study following 5 days of twice-daily enteral dosing of 500mg or 1000mg with acceptance defined as predicted/observed ratio within 2-fold.

Results

CLP mice exhibited significant downregulation of kidney OCT2 (0.54-fold, p=0.02) and MATE1 (0.47-fold, p=0.002), while MATE2K was unchanged. The final sepsis PBPK model predicted mean C_max and AUC_0-12h within 2-fold of observed values across all subgroups at both 500 and 1000mg dose levels (Table 1).

Conclusion

This work establishes a novel PBPK model integrating preclinical kidney transporter expression with sepsis-related physiological changes to characterize metformin PK in critically ill patients with sepsis. The model serves as a clinically translatable tool to inform future metformin dosing strategies for prevention of AKI in this high risk population.

Funding

• NIDDK Support