Correlation of Pre- and Post-radio-chemotherapy MRI Texture Features With Tumor Response in Rectal Cancer

Area of Science:

• Radiology
• Oncology
• Medical Imaging Analysis

Background:

• Locally advanced rectal cancer (LARC) requires effective neoadjuvant treatment.
• Magnetic resonance imaging (MRI) plays a crucial role in assessing treatment response.
• Radiomics offers quantitative analysis of imaging features.

Purpose of the Study:

• To investigate radiomics features from MRI for predicting treatment response in LARC patients.
• To evaluate the association of radiomics features with tumor regression grade (TRG) after neoadjuvant chemoradiotherapy (CRT).

Main Methods:

• Retrospective analysis of 53 LARC patients (T3/T4 or N+) who underwent MRI before and after CRT.
• Extraction of 27 texture radiomics features from T2-weighted MRI images.
• Statistical analysis to correlate radiomics features with TRG.

Main Results:

• All 27 radiomics features extracted pre-CRT significantly associated with TRG (p<0.05).
• Post-CRT, only Cluster Prominence remained a significant predictor of TRG (p=0.037, r=0.291).

Conclusions:

• Pre-treatment radiomics features effectively predict CRT response in LARC.
• Cluster Prominence shows sustained predictive value for TRG after CRT.
• Further investigation of MRI-derived radiomics in LARC is warranted.