Abstract
Background
Intravenous cyclophosphamide (IV-CYC) is a first-line treatment for idiopathic inflammatory myopathy (IIMs) complicated with interstitial lung disease (ILD) but has been hampered by a lack of targeted clinical trials and heterogeneous patient responses. Therefore, this study established an evaluation model for IV-CYC treatment response in IIMs-ILD patients by combining clinical characteristics and radiomics.Methods
Retrospectively included total of 62 IIMs-ILD patients were enrolled and categorized into two groups on the basis of treatment response: the Effective group (n = 18) and the Non-effective group (n = 44). Baseline clinical and imaging characteristics were analyzed, with no significant differences observed in demographic or clinical factors between the groups. Least absolute shrinkage and selection operator (LASSO) regression identified five key predictors of treatment response: ROl_Lobe2 infected rates, ROl_Segment1 infected rates, anti-MDA5 antibody, anti-Jo1 antibody, and IgA levels.Results
A nomogram incorporating these factors was developed, with an area under the curve (AUC) of 0.876 (95% CI: 0.788-0.964), a sensitivity of 0.889 and a specificity of 0.790. The model's Youden's index was 0.639. The calibration curve indicated strong agreement between the predicted and actual outcomes. Decision curve analysis (DCA) and a clinical impact curve (CIC) confirmed the model's clinical utility, showing significant net benefits in predicting the treatment response to IV-CYC in IIMs-ILD patients.Conclusion
The proposed nomogram provides a reliable tool for predicting treatment response in IIMs-ILD patients receiving IV-CYC therapy, potentially guiding personalized treatment strategies.