BACKGROUND: High-dose methotrexate (HD-MTX)-based polychemotherapy is widely used for patients with central nervous system (CNS) lymphoma. The pharmacokinetic (PK) variability and unpredictable occurrence of toxicity remain major concerns in HD-MTX treatment. OBJECTIVES: This study aimed to characterize the population PK of HD-MTX in patients with CNS lymphoma and to identify baseline predictors and exposure thresholds that predict a high risk of nephro- and hepatotoxicity. METHODS: Routinely monitored serum MTX concentrations after intravenous infusion of HD-MTX and MTX dosing information were collected retrospectively. Acute event of toxicity (≥ grade 1) was defined according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0 on the basis of serum creatinine and alanine aminotransferase. A population PK model was developed in NONMEM. Toxicity data were analyzed using a logistic regression model, and potential baseline and exposure-related predictors were investigated. RESULTS: In total, 1584 MTX concentrations from 110 patients were available for analysis. A two-compartment population PK model adequately described the data. Estimated glomerular filtration rate (eGFR), treatment regimen, albumin, alkaline phosphatase, and body weight were identified as significant covariates that explain the PK variability of HD-MTX. Baseline eGFR and sex were identified as significant predictors for renal toxicity, and MTX dose (mg/m CONCLUSIONS: A population PK model was developed for HD-MTX in patients with CNS lymphoma. The toxicity analysis showed that lower baseline eGFR and male sex, and higher MTX dose are associated with increased risk of acute nephro- and hepatotoxicity, respectively. The proposed exposure thresholds that predict high risk of renal toxicity and the developed models hold the potential to guide HD-MTX dosage individualization and better prevent acute toxicity.