Objectivev To investigate the clinical characteristics of patients with anti-MDA5 antibody-positive dermatomyositis with interstitial lung disease (MDA5+ DM-ILD), explore prognostic risk factors, and evaluate the predictive value of relevant indicators for mortality, thereby providing a basis for the early identification of high-risk patients and precise intervention. MethodsThe clinical data and relevant laboratory test results within 24 hours of admission of 33 patients with MDA5+ DM-ILD diagnosed in Fuzhou Pulmonary Hospital of Fujian Province from January 2021 to December 2024 were retrospectively analyzed. According to medical records and telephone follow-up results, the patients were divided into a survival group (n=19) and a death group (n=14). Univariate analysis and multivariate COX regression analysis were conducted to identify the risk factors in the death group. The statistically significant risk factors from the multivariate analysis were used to draw receiver operating characteristic (ROC) curves. ResultsThe median follow-up time of the 33 patients was 18.5 months (interquartile range: 10.2-26.8 months). Univariate analysis showed that the neutrophil count, serum ferritin (SF), erythrocyte sedimentation rate (ESR), carcinoembryonic antigen (CEA), pro-gastrin-releasing peptide (ProGRP), cytokeratin fragment antigen 21-1 (Cyfra21-1), neuron-specific enolase (NSE), lactate dehydrogenase (LDH), cancer antigen 153 (CA153), chest CT lesion extent score, and ground-glass opacity (GGO) score for each lung lobe were significantly higher in the death group than those in the survival group. Conversely, the oxygenation index (PaO2/FiO2) was significantly lower in the death group. These differences were statistically significant (P<0.05). Multivariate Cox regression analysis identified SF (OR=1.001, 95%CI 1.000-1.002, P<0.05) and Cyfra21-1 (OR=1.100, 95%CI 1.037-1.166, P<0.05) as independent risk factors for death. ROC curve analysis determined the optimal cutoff values as 1688 ng/mL for SF (AUC=0.778, sensitivity=57.1%, specificity=94.7%) and 5.63 ng/mL for Cyfra21-1 (AUC=0.853, sensitivity=100%, specificity=73.7%). ConclusionSF and CYFRA21-1 serve as independent predictors of mortality in patients with MDA5+ DM-ILD. These biomarkers enable early identification of high-risk subgroups, and guide targeted therapeutic interventions to improve prognoses when integrated with clinical and computed tomography characteristics.