Aims: Although high-dose methotrexate (HDMTX) is an effective means for the treatment of acute lymphoblastic leukemia (ALL), the development of renal dysfunction remains a significant management challenge. This study aimed to identify the key factors in HDMTX-induced acute kidney injury (AKI) in childhood ALL. Methods: We retrospectively analyzed the clinical data in 1,329 courses of HDMTX treatment in 336 Chinese ALL children at the First Affiliated Hospital of Guangxi Medical University from September 2012 to November 2016. The clinical data were compared between the groups of children with development of AKI and those without. Risk factors were identified by multiple logistic regression analysis, and the diagnostic performance of plasma MTX concentration was evaluated by receiver operating characteristic (ROC) curve analysis. Results: AKI was observed in 88 patients (26.2%) and 104 courses (7.8%). Binary logistic regression revealed that age (OR 1.349; p = 0.005), first HDMTX course (OR 1.767; p = 0.013), MTX dose per body surface area (BSA; OR 1.944; p = 0.015), and baseline serum total protein (OR 0.929; p = 0.021) significantly correlated with AKI. The area under the ROC for 48-h plasma MTX concentration was 0.890 (95% CI 0.850–0.930), and sensitivity and specificity values of the cut-off value were 78.8 and 90.4%, respectively. Conclusion: Increasing age, higher MTX dose per BSA, lower baseline serum protein, and first HDMTX course were significant risk factors for developing HDMTX-induced AKI in childhood ALL. The threshold of 48-h MTX plasma concentration is valuable for the prediction of HDMTX-induced AKI.

Objectives This study aimed to establish a vancomycin population pharmacokinetics (PPK) model based on serum cystatin C and to optimize dosing for achieving targeted steady‐state trough concentrations (Css) of 10–15 and 15–20 mg/l. Methods Patients aged ≥18 years were prospectively enrolled. A vancomycin PPK model was built with glomerular filtration rate (GFR) as a renal covariate estimated by cystatin C. A new group of patients were used for external evaluation. PPK analysis and Monte Carlo simulations were performed using nonlinear mixed effect modelling programme. Key findings Two hundreds of patients with 514 samples were included. The final model was CL (L/h) = (5.07 × (GFR/105.5)0.524 × (AGE/48.5)−0.309 × (WT/60)0.491); V (l) = 46.3. Internal and external evaluations demonstrated good stability and predictability. The average probability of target attainment (PTA) of optimal dosing regimens for targeted Css achieving 10–15 and 15–20 mg/l were 51.2% and 40.6%, respectively. An average PTA ≥71% for targeted concentration of 10–20 mg/l was obtained. Conclusions A vancomycin PPK model with cystatin C as the renal marker has good stability and predictability. The new proposed dosing regimens were predicted to achieve a good PTA.