Risk factors analysis and validation of early treatment non-response in children with primary ITP: a multicenter retrospective cohort study

Objective To identify factors associated with early non-response to first-line treatment in children with primary immune thrombocytopenia (ITP), and to develop and externally validate a prediction model. Methods This multicenter retrospective cohort study included 335 children with newly diagnosed primary ITP as the derivation cohort, including 255 responders and 80 non-responders. Clinical data obtained after admission were collected. Candidate variables were screened using correlation analysis and least absolute shrinkage and selection operator (LASSO) regression, followed by multivariable logistic regression to identify factors associated with early treatment non-response. Model discrimination was evaluated using the area under the receiver operating characteristic curve (AUC). An independent external validation cohort of 78 children was additionally included to assess model discrimination. Results Among the 335 children, 255 (76.1%) achieved a treatment response within 14 days after first-line therapy, whereas 80 (23.9%) did not. Multivariable analysis showed that older age (OR = 1.769, 95% CI: 1.629–1.920, P < 0.001), higher platelet-to-lymphocyte ratio (PLR) (OR = 1.622, 95% CI: 1.112–2.367, P = 0.012), and antinuclear antibody (ANA) positivity (OR = 3.142, 95% CI: 1.626–6.071, P < 0.001) were independently associated with early treatment non-response. The combined model incorporating age, PLR, and ANA achieved an AUC of 0.831 (95% CI: 0.778–0.885) in the derivation cohort and 0.718 (95% CI: 0.591–0.845) in the independent external validation cohort. Calibration was acceptable in both cohorts, with Hosmer–Lemeshow P values of 0.609 and 0.220 and Brier scores of 0.129 and 0.137, respectively. Conclusion Older age, higher PLR, and ANA positivity were independently associated with early non-response to first-line treatment within 14 days in children with newly diagnosed primary ITP. These findings may help identify children at high risk of treatment non-response and support individualized monitoring and treatment strategies.