Development and Validation of a Prediction Model for Severe Pre‐Engraftment Syndrome During the Treatment of Unrelated Umbilical Cord Blood Transplantation in Pediatric Patients

ABSTRACT Background Pre‐engraftment syndrome (PES) is a common immune response during the early stages of umbilical cord blood transplantation (UCBT), but the severity classification of PES remains uncertain. Objectives To develop and validate a predictive model for PES severity in pediatric patients receiving UCBT. Study Design This retrospective study (2017–2025) included 123 pediatric patients (training cohort) and 41 external validation cases. Uni‐ and multivariable analysis was conducted to assess the potential parameters affecting PES severity. Based on the multivariable Cox model, the optimal model's performance was illustrated using a nomogram and evaluated through discrimination and calibration. Results Infused CD34 + counts (hazard ratio [HR]: 1.249, 95% confidence interval [CI]: 1.037–1.503, p  < 0.05), donor chimerism on day 7 (HR: 1.025, 95% CI: 1.010–1.040, p  < 0.05), and serum IL‐6 level on symptoms onset (HR: 1.101, 95% CI: 1.000–1.123, p  < 0.05) were identified as independent risk factors. The predictive performance of this nomogram model was evaluated by C‐index and AUC, with C‐index of 0.754 (95% CI: 0.693–0.815) and 9‐day and 14‐day AUC of 0.757 (95% CI: 0.655–0.859) and 0.856 (95% CI: 0.699–0.981), respectively. High‐risk patients (score ≥104.9) had higher 9‐day severe PES incidence (65.4% vs. 11.4%, p  < 0.05) and Grade II‐IV/III‐IV aGVHD rates (84.2%/59.9% vs. 48.2%/28.2%, p  < 0.05). External validation in an independent cohort ( n = 41) displayed a C‐index of 0.821 (95% CI: 0.689–0.953) and 14‐day AUC of 0.933 (95% CI: 0.873–0.993). Conclusion This predictive model could predict the severity of PES in pediatric patients undergoing UCBT, thereby promoting preemptive treatment.