Development of a Clinical Risk Prediction Model for Risk Stratification of Capecitabine-Induced Grade 2–3 Hand–Foot Syndrome

Background: Hand-foot syndrome (HFS) is a dose-dependent toxicity of capecitabine that affects quality of life and may lead to treatment interruption. Currently, no clinical prediction tool exists to identify patients at high risk of developing moderate to severe HFS (grades 2 to 3). This study aimed to develop a clinical prediction model to estimate the risk of grade 2 to 3 HFS in patients receiving capecitabine. Methods: A retrospective cohort study was conducted in capecitabine-treated cancer patients at a tertiary cancer center between January 2018 and December 2024. Nine prespecified predictors including capecitabine dosage, sex, age, body mass index, body surface area, combination chemotherapy, estimated glomerular filtration rate, topical diclofenac use, and serum albumin were analyzed using multinomial logistic regression to develop the prediction model. Results: Of 596 patients, 132 (22.2%) developed grade 2–3 HFS. The final model included five predictors: sex, age, body mass index, combination chemotherapy, and capecitabine dosage. The model demonstrated good discrimination, with an optimism-adjusted C-statistic of 0.72 (95% CI, 0.67–0.77), and good calibration, with a calibration slope of 1.012, calibration-in-the-large of 0.000, and an observed-to-expected ratio of 0.996. Conclusions: The model was implemented as a web-based application to estimate individual patient risk and facilitate risk stratification. This internally validated prediction model may serve as a pilot clinical tool for identifying patients at increased risk of capecitabine-induced grade 2–3 HFS and warrants external validation before broader clinical implementation.