Neoadjuvant Chemoradiotherapy on Postoperative Complications of Rectal Cancer: A Retrospective Study Integrating MRI...

Area of Science:

• Oncology
• Medical Imaging
• Artificial Intelligence

Background:

• Neoadjuvant chemoradiotherapy (nCRT) is standard for locally advanced rectal cancer (LARC).
• nCRT increases postoperative complication (POC) risks due to tissue fibrosis and immunosuppression.
• Current predictive models for POC do not account for nCRT-induced tissue changes.

Purpose of the Study:

• To investigate the impact of nCRT on POC in LARC patients.
• To develop a deep learning-based predictive model integrating MRI radiomics and clinical features for early POC risk stratification.

Main Methods:

• Retrospective, dual-center study of 695 LARC patients (272 nCRT, 423 non-nCRT).
• Propensity score matching (PSM) to balance covariates.
• Radiomics and deep learning features extracted from preoperative MRI in the nCRT group; six machine learning algorithms evaluated.

Main Results:

• nCRT was a significant risk factor for POC (OR 3.78, P < 0.001).
• The combined radiomics-deep learning model achieved an AUC of 0.82 in external validation.
• The multimodal model outperformed unimodal approaches in predicting POC.

Conclusions:

• nCRT is a significant risk factor for POC in LARC patients.
• A multimodal deep learning model integrating MRI radiomics and clinical data offers superior predictive accuracy for POC.
• This model provides a clinically actionable tool for personalized perioperative management in LARC patients.