Serum YKL-40, periostin, and eosinophil-derived neurotoxin as complementary biomarkers for assessing disease severity in adults with bronchial asthma: A retrospective observational study

Background: Bronchial asthma is a heterogeneous inflammatory airway disease, and conventional clinical and inflammatory indicators may not fully reflect its biological complexity or severity. YKL-40, periostin, and eosinophil-derived neurotoxin (EDN) represent distinct aspects of airway remodeling, type 2 inflammation, and eosinophil activation, respectively. This study investigated their individual and combined clinical value for assessing disease severity in adults with bronchial asthma. Methods: This single-center retrospective observational study included 82 adults with bronchial asthma, comprising 23 patients with mild, 39 with moderate, and 20 with severe asthma. Serum YKL-40, periostin, and EDN concentrations were measured using enzyme-linked immunosorbent assays. Clinical characteristics, Asthma Control Test (ACT) scores, pulmonary function parameters, fractional exhaled nitric oxide (FeNO), peripheral blood eosinophils, and total immunoglobulin E were collected. Correlation and logistic regression analyses were performed to evaluate associations between serum biomarkers and asthma severity. Receiver operating characteristic analysis was used to assess the ability of individual biomarkers and their combined model to distinguish severe from non-severe asthma. Results: Serum YKL-40, periostin, and EDN levels creased progressively from mild to moderate and severe asthma (all P for trend <0.001). YKL-40 showed the strongest inverse correlation with percentage predicted forced expiratory volume in one second (FEV1) (r=−0.60, P<0.001), whereas periostin was closely correlated with FeNO (r=0.57, P<0.001) and peripheral eosinophil count (r=0.53, P<0.001). EDN showed the strongest correlations with eosinophil count (r=0.66, P<0.001), eosinophil percentage (r=0.63, P<0.001), and FeNO (r=0.60, P<0.001). In multivariable logistic regression, lower percentage predicted FEV1 and higher YKL-40, periostin, and EDN concentrations remained independently associated with severe asthma. The areas under the receiver operating characteristic curve for YKL-40, periostin, and EDN were 0.801, 0.776, and 0.818, respectively. Their combined model achieved an area under the curve of 0.899 (95% CI, 0.811–0.951), with 85.0% sensitivity and 83.9% specificity. Bootstrap internal validation yielded an optimism-corrected area under the curve of 0.897. Conclusions: Serum YKL-40, periostin, and EDN are as-sociated with distinct but complementary clinical and inflammatory features of adult asthma. Their concentrations increase with disease severity, and combined assessment provides greater discrimination of severe asthma than individual biomarkers. These circulating biomarkers may serve as complementary tools for biological assessment and risk stratification in adults with bronchial asthma.