Several sirolimus (SRL) population pharmacokinetics (PopPK) were conducted to explain its pharmacokinetic variability, and the results varied across studies. Thus, we conducted a systematic review to summarize significant predictors influencing SRL pharmacokinetic variability. Moreover, discrepancies in model methodologies across studies were also reviewed and discussed. Four databases (PubMed, CINAHL Complete, Science Direct, and Scopus) were systematically searched. The PICO framework was used to identify eligible studies conducted in humans and employ a nonlinear-mixed effects strategy. Based on the inclusion and exclusion criteria, 20 studies were included. SRL pharmacokinetics were explained using 1- or 2-compartment models. Only one study assessed the model using an external approach, while the rest employed basic or advanced internal approaches. Significant covariates influencing SRL pharmacokinetics were bodyweight, age, CYP3A5 polymorphism, gender, BSA, height, cyclosporine dose or trough concentration, triglyceride, total cholesterol, hematocrit, albumin, aspartate aminotransferase, alanine aminotransferase, and total bilirubin. Of these, bodyweight, age, and CYP3A5 polymorphism were the three most identified significant predictors for SRL clearance. This review summarizes significant predictors to predict SRL clearance, which can subsequently be used to individualize SRL maintenance dose. However, the PopPK model selected for such prediction should be based on the resemblance of population characteristics between the target population and those used to conduct the model. Moreover, the predictability of the models in the target population should be assessed before implementation in clinical practice.
Keywords: anticancer; immunosuppressant; nonlinear mixed-effects; population pharmacokinetics; sirolimus.