Predicting chronic immune thrombocytopenia in children using IFNA17 rs9298814 polymorphism and clinical features: a study from a low- and middle-income country.

Immune thrombocytopenia (ITP) is a common acquired hematologic disorder in childhood, with approximately 20-25% of children progressing to chronic disease. Early identification of those at risk remains a major clinical challenge. In this prospective cohort study, 205 children with newly diagnosed ITP at a tertiary center in Vietnam were followed for 12 months. Clinical and laboratory variables, along with IFNA17 rs9298814 genotype, were assessed. Multivariable analysis identified independent predictors of chronic ITP, defined as platelet counts <100 × 109/L persisting beyond 12 months. Of the 205 children, 48 (23.4%) developed chronic ITP. Older age at onset, a longer bleeding-onset duration, and a lower platelet count at 4 weeks were associated with chronic ITP, as was the IFNA17 rs9298814 TT genotype (homozygosity for the reference allele; i.e. absence of the G variant); conversely, carriage of the minor G allele was protective (G-allele odds ratio [OR] 0.10, 95% confidence interval [CI] 0.05-0.20; TT versus G-carrier OR 26.4, 95% CI 11.1-63.1). Two prediction models were developed and expressed as a risk score: a clinical model (age at onset, bleeding-onset duration, platelet count at 4 weeks, and antinuclear antibody status) with an area under the receiver operating characteristic curve (AUC) of 0.92 (95% CI 0.88-0.96), and an IFNA17-integrated model with a significantly improved AUC of 0.95 (95% CI 0.92-0.98; DeLong test p = 0.005). The integrated model demonstrated good calibration. A nomogram and bedside score were constructed for clinical application. Both clinical factors and host genetic variation contribute to the risk of chronic ITP in children. Integrating IFNA17 rs9298814 genotype into clinical prediction models improves early risk stratification. External validation in independent cohorts is warranted.