Pre-Treatment Computed Tomography Radiomics for Predicting the Response to Neoadjuvant Chemoradiation in Locally...

Area of Science:

• Oncology
• Radiology
• Medical Imaging

Background:

• Computerized tomography (CT) is crucial for diagnosing and treating locally advanced rectal cancer (LARC).
• Predicting treatment response in LARC is essential for optimizing patient management.

Purpose of the Study:

• To evaluate the efficacy of pretreatment CT-based radiomics in predicting pathological complete response (pCR) to neoadjuvant chemoradiotherapy (nCRT) in LARC patients.
• To develop and validate a predictive model for pCR.

Main Methods:

• Retrospective analysis of 340 radiomic features from pretreatment contrast-enhanced CT scans of 216 LARC patients.
• Feature selection using LASSO, generation of a radiomic signature, and construction of predictive models with clinical variables.
• Model performance assessed using decision curve analysis and validated in an independent cohort.

Main Results:

• Pathological complete response (pCR) was achieved in 20.4% of patients.
• A combined model incorporating radiomic signatures, distance to anal verge, lymphocyte-to-monocyte ratio, and carcinoembryonic antigen demonstrated superior predictive performance (AUC 0.926 training, 0.872 validation).
• The combined model outperformed models using only radiomic or clinical variables.

Conclusions:

• A robust combined predictive model effectively differentiates LARC patients who will achieve pCR after nCRT.
• CT-based radiomics, integrated with clinical data, offers a promising tool for personalized treatment strategies in LARC.