A clinical prediction model for sepsis in chemotherapy-induced neutropenia among children with acute lymphoblastic leukemia: development and validation

This study aimed to develop and internally validate a nomogram for estimating the risk of neutropenia-associated sepsis in children with acute lymphoblastic leukemia (ALL) and chemotherapy-induced neutropenia. This single-center retrospective cohort study included 264 pediatric patients with ALL treated at the First Affiliated Hospital of Xinjiang Medical University between January 2022 and July 2024. The cohort was randomly divided into a training cohort and an internal validation cohort at a ratio of 7:3. Candidate predictors were evaluated using univariate logistic regression, least absolute shrinkage and selection operator regression, and multivariable logistic regression. Model discrimination, calibration, and clinical utility were assessed using receiver operating characteristic (ROC) curves, calibration analysis, and decision curve analysis (DCA), respectively. Bootstrap internal validation with 1,000 resamples was additionally performed in the training cohort. Among the 264 patients, 39 developed neutropenia-associated sepsis after chemotherapy. Duration of neutropenia, oral infection, glucocorticoid use, and serum interleukin-6 level were independently associated with neutropenia-associated sepsis and were incorporated into the final nomogram (P < 0.05). The nomogram showed good discrimination, with an AUC of 0.884 (95% CI, 0.802–0.966) in the training cohort and 0.893 (95% CI, 0.770–1.000) in the internal validation cohort. The Hosmer–Lemeshow test indicated no significant lack of fit in either cohort (P = 0.11 and 0.36, respectively). DCA demonstrated potential clinical net benefit across threshold probabilities of approximately 10%–70%. This developed nomogram showed promising performance for estimating the risk of neutropenia-associated sepsis in children with ALL and chemotherapy-induced neutropenia. It may support clinical risk stratification and individualized assessment; however, independent external validation is required before routine clinical implementation.