Quantifying covariate effects on mycophenolate pharmacokinetics in renal transplantation : a review using pharmacokinetic modeling

Background: Mycophenolic acid (MPA) is a cornerstone immunosuppressant in renal transplantation, yet large interindividual variability in its pharmacokinetics (PK) complicates dosing and patient management. While numerous modeling studies have identified covariates affecting MPA exposure, findings have remained fragmented. There remains a lack of updated review-level work that summarizes these findings and capable of translating reported patient-specific factors into a common quantitative output metric for direct comparison across different study types. The objective of this work is to provide a literature review of patient specific factors and their impact on MPA pharmacokinetic with the aim that this information can be incorporated into future integrated models that better describe MPA pharmacokinetics and can guide personalized dosing. Methods: This study represents a review to incorporate population pharmacokinetic modeling principles into evidence synthesis. A systematic search of MEDLINE (January 1, 2009–May 20, 2025) identified studies including ≥20 kidney transplant recipients that investigated MPA PK, or PG using a standardized model informed data conversion approach, heterogeneous model outputs were standardized into interpretable metrics, such as percentage change in MPA exposure (AUC0-12h) per factor. Results: After converting reported results to a common exposure measure, it revealed that serum albumin, and co-administered calcineurin inhibitors were the most impactful and consistent factors. Hypoalbuminemia (32 g/L) reduced MPA AUC0-12h by 17–44%, while hyperalbuminemia (52 g/L) increased it by 16–44%, sufficient to shift patients outside the therapeutic window (30–60 mg·h/L). Co-administration of cyclosporine reduced MPA AUC by 20–60% compared to tacrolimus through inhibition of MPAG biliary excretion. Pharmacogenetic analysis identified significant but variant-specific effects, such as increased AUC in carriers of UGT1A93 and SLCO1B3 T334G variants. Conclusion: This review reconstructs and standardizes published evidence on patient-specific determinants of MPA pharmacokinetics into a common exposure-based format. By assembling fragmented model-based and exposure-based findings into comparable output metrics, this work provides base evidence that can be incorporated into future model-informed dosing tools. Such standardized outputs and reconstructed pharmacokinetics model will support future development of model–based tools in clinical decision making, through which published evidence can be queried, simulated, and translated into patient-specific exposure estimates.