What does this research mean for the field?

Machine learning models can accurately predict mycophenolic acid exposure in pediatric lupus nephritis, enhancing personalized dosing strategies. Novelty: ClaimNovelty.NOVEL_FINDING. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

This research aims to develop machine learning models to predict mycophenolic acid exposure in pediatric patients with lupus nephritis.

February 19, 2026

Predicting mycophenolic acid exposure in pediatric lupus nephritis: a real-world study across three clinical scenarios

Key Points

This research aims to develop machine learning models to predict mycophenolic acid exposure in pediatric patients with lupus nephritis.
Data collected from two cohorts of pediatric lupus nephritis patients: routine treatment and dose adjustment.
Models constructed using machine learning algorithms, including XGBoost, LightGBM, and random forest.
Model performance evaluated using various metrics, with SHAP method for interpretation.
XGBoost and GBM models showed superior performance in predicting therapeutic window attainment with an AUC of 0.67.
Random forest model achieved an R2 of 0.84 in predicting MPA area under the curve using clinical indicators.
The framework allows for personalized MPA dosing, improving therapeutic drug monitoring.

Abstract

Abstract Objectives Childhood-onset systemic lupus erythematosus (cSLE) with lupus nephritis (LN) presents therapeutic challenges due to complex mycophenolic acid (MPA) pharmacokinetics. This study developed machine learning (ML)-based models to predict MPA area under the concentration-time curve (MPA-AUC) across three scenarios: determining therapeutic window attainment, precise AUC estimation, and forecasting post-dose-adjustment exposure. Methods Data were collected from cLN on MPA with two cohorts: routine treatment (group 1) and dose adjustment (group 2). Comprehensive patient data, including demographics, medication details, and laboratory test results, were collected. Models were constructed using algorithms such as XGBoost, LightGBM, and RF. Model performance was assessed using multiple metrics and the Shapley Additive Explanations (SHAP) method was employed to interpret the models. Results Data from 154 patients (1,376 follow-ups) were included. In Scenario 1, XGBoost and GBM models demonstrated superior performance, with the LightGBM model excelling in predicting therapeutic window attainment (AUC = 0.67, Precision = 0.64, Recall = 0.66, F1 = 0.65). In Scenario 2, the RF model incorporating clinical indicators and C0.5, C1.5, C4 concentrations achieved the most accurate predictions (R2 =0.84, MAE =3.97, and RMSE =5.97). In Scenario 3, which included data from 246 follow-up records of 99 patients, the RF model showed the best performance (R2 =0.51 and MAE =20.57). SHAP analysis highlighted dose, weight, renal function, and concentration timepoints as key predictors. A web-based tool was developed to support clinical decisions. Conclusion This study establishes the first ML-driven framework for personalized MPA dosing in pediatric LN, addressing TDM limitations and enhancing precision medicine.

Perguntar à IA

Bookmark