Limited Sampling Strategies to Predict Mycophenolic Acid and Tacrolimus Area Under the Concentration-Time Curve in Steroid-Free Kidney Transplant Patients.

This study aimed to develop limited sampling strategies to predict oral mycophenolic acid (MPA) exposure from one to three blood samples in stable steroid-free kidney-transplanted patients and assess if the same scheme could predict tacrolimus exposure. Additionally, we aimed to validate existing strategies, and to describe the pharmacokinetics of MPA and its inactive metabolite, MPA-glucuronide (MPAG), in our cohort. We analysed data from dense pharmacokinetic sampling from 15 steroid-free kidney-transplanted patients, which were prospectively enrolled as part of larger cohort. Drug concentration was analysed in plasma (MPA, MPAG) or in whole blood (tacrolimus) using LC-MS. Exposure (AUC0-12h) was based on non-compartmental analysis (MPA, MPAG) or model-derived (tacrolimus). Limited sampling strategies were developed using multiple stepwise linear regression analysis and evaluated for bias and imprecision and using Bland-Altman analysis. Median AUC0-12h was 31.2 μg/mL·h, 346.6 μg/mL·h and 81.9 ng/mL·h for MPA, MPAG and tacrolimus, respectively. Limited sampling strategies incorporating measurements at C0 and C1.5, or at C0, C0.5 and C1.5 could predict MPA and tacrolimus AUC0-12h with low (< 15%) bias and imprecision. None of the previous strategies could adequately predict MPA AUC0-12h. Limited sampling strategies for MPA and tacrolimus can potentially replace full pharmacokinetic profiling in steroid-free kidney transplant patients. External validation is needed before implementation. SUMMARY: Kidney-transplanted patients are treated with immunosuppressive drugs throughout the lifespan of the transplanted organ. In this study, we aimed to derive mathematical equations that can simultaneously predict the total oral drug exposure of two of these drugs (tacrolimus and mycophenolate mofetil) based on few blood samples. We analysed drug concentration in blood samples from 15 patients, calculated their exposure and assessed how accurately the mathematical equations could predict the exposure. We could predict the total drug exposure from two or three samples, and these equations can be used in future research and in the clinic to ensure proper immunosuppressive levels.