Potential Biomarkers for Disease Stratification and Prognosis Prediction in Pediatric Asthma: <scp>LCN2</scp> , <scp>sST2</scp>  and <scp>FGF21</scp>

Pediatric asthma is a common chronic airway inflammatory disease, where accurate assessment of disease severity and prognosis prediction is crucial for treatment decisions. Currently, there is a lack of precise and effective biomarkers. This study aimed to investigate the association between serum levels of lipocalin-2 (LCN2), soluble suppression of tumorigenicity 2 (sST2), and fibroblast growth factor 21 (FGF21) with asthma severity and to evaluate their predictive value for prognosis. A total of 110 asthmatic children (asthma group) were enrolled and stratified by severity into mild (n = 50), moderate (n = 34), and severe (n = 26) subgroups. After 4 weeks of treatment, the asthma group was further divided into good prognosis (n = 48) and poor prognosis (n = 64) subgroups based on Childhood Asthma Control Test (C-ACT) scores. Additionally, 110 healthy children were randomly selected as controls. Serum LCN2, sST2, and FGF21 levels were measured using enzyme-linked immunosorbent assay. Clinical data and pulmonary function parameters were collected. Spearman correlation analysis assessed the relationship between biomarker levels and C-ACT scores. Multivariate logistic regression identified risk factors for poor prognosis. Receiver operating characteristic (ROC) curve analysis evaluated the predictive performance of LCN2, sST2, and FGF21 for poor prognosis. Serum LCN2, sST2, and FGF21 levels were significantly higher in asthmatic children than in controls (p < 0.05), with levels progressively elevating as disease severity increased (p < 0.05). The good prognosis group exhibited lower levels of all three biomarkers compared to the poor prognosis group (p < 0.05). Spearman analysis revealed negative correlations between biomarker levels and C-ACT scores (p < 0.05). Multivariate logistic regression confirmed LCN2, sST2, and FGF21 as independent risk factors for poor prognosis (p < 0.05). ROC analysis demonstrated moderate predictive efficacy for individual biomarkers, while their combination achieved an area under the curve of 0.938, with 93.33% sensitivity and 90.00% specificity, suggesting potential clinical utility for severity assessment and prognosis prediction. Serum LCN2, sST2, and FGF21 levels increase with asthma severity and demonstrate high predictive value for poor prognosis when combined. These biomarkers may serve as early prognostic indicators in pediatric asthma management.