Radiomics Combined with Dosiomics and Clinical Omics for Predicting Response to Neoadjuvant Therapy in Rectal Cancer

Motivation: Approximately 15-27% of patients with LARC achieve pCR. However, determining whether these patients should undergo surgery or follow "watch-and-wait" approach requires accurately predicting preoperatively if they have achieved pCR. Goal(s): To developed a comprehensive multi-omics model to predict pCR. Approach: We used logistic regression to generate a non-imaging model, three radiomics-based models, and a dosiomics-based model and were then combined into a comprehensive model, its performance was tested on multi-center datasets. Results: The radiomic model exhibits relatively comparable performance, and the final comprehensive multi-omics model exhibits good performance after combining clinical characteristics and dositomics. Impact: Predicting patients who achieve pCR after nCRT can assist doctors in formulating personalized treatment strategies. This helps determine whether patients require surgery or if the "watch-and-wait" approach can be adopted.