LAUSR

BACKGROUND Lung metastasis (LM) status is critical for making treatment decisions in soft-tissue sarcoma (STS) patients, yet magnetic resonance imaging (MRI)-based prediction of LM in STSs has not been thoroughly investigated. PURPOSE This study aimed to develop MRI-based radiomics models for identifying LM in STSs. METHODS We enrolled 122 STS patients from our hospital to form a primary cohort. Thirty-two patients from another hospital were included as an external validation cohort. All patients underwent T1-weighted contrast-enhanced (T1-CE) MRI scans before treatment. Radiomics features were extracted from T1-CE MRI sequence and selected by least absolute shrinkage and selection operator (LASSO) to build the radiomics signature. Clinical factors were evaluated using the univariate and multivariate analyses. Multivariable logistic regression analysis was used to construct a clinical-radiomics nomogram incorporating the radiomics signature with margin. Receiver operating characteristic (ROC), calibration and decision curve analysis (DCA) curves were plotted and area under the ROC curves (AUCs) were calculated to assess the predictive performance of nomogram, radiomics signature and margin. RESULTS A total of 5 features was finally identified highly related to the LM status to develop the radiomics signature. The nomogram integrating the radiomics signature and margin achieved the best prediction performance in the training (AUCs, nomogram vs. radiomics signature vs. margin, 0.918 vs. 0.894 vs. 0.609), internal validation (AUCs, nomogram vs. radiomics signature vs. margin, 0.864 vs. 0.841 vs. 0.666) and external validation (AUCs, nomogram vs. radiomics signature vs. margin, 0.843 vs. 0.800 vs. 0.643) sets. CONCLUSIONS The developed nomogram was a promising tool to help make preoperative treatment strategies for STSs. This article is protected by copyright. All rights reserved.