Visualized clinical-radiomics model for predicting the efficacy of surufatinib in hepatic metastases of neuroendocrine neoplasms.

Background: Hepatic metastatic neuroendocrine neoplasms (HM-NENs) have few treatment biomarkers and low survival rates. We created a clinical-radiomics fusion model to predict surufatinib efficacy in HM-NENs and presented it as a nomogram, meeting unmet requirements in precision hepatology. Methods: This study included 76 HM-NEN patients (131 hepatic metastases) treated with surufatinib. SlicerRadiomics was used to extract radiomics features from arterial phase computed tomography (APCT). The least absolute shrinkage and selection operator (LASSO) was used to select radiomics features and calculate a radiomics score (Radscore). Multivariable logistic regression analysis was utilized to create the clinical-radiomics fusion model, which included clinical characteristics and Radscore and was displayed as a nomogram. The area under the receiver operating characteristic curve (ROC) was used to assess model performance, and internal validation was done using the bootstrap resampling approach. Results: After multivariate logistic regression analysis, the Radscore, Ki67 antigen (Ki67), number of hepatic metastases, and extrahepatic metastasis were included as predictors in the final model. The area under the curve (AUC) of the clinical-radiomics fusion model to predict the response of surufatinib of HM-NENs was 0.928 (95% CI: 0.885 - 0.971). The AUC verified by bootstrap is 0.928 (95% CI: 0.881-0.965), indicating a good performance of the fusion model. Conclusion: The clinical-radiomics fusion model can effectively identify patients with HM-NENs sensitive to surufatinib therapy. The nomogram provided clinicians with a convenient and dependable tool for decision-making.