Accurate tumor response determination remains inconclusive after preoperative chemoradiation therapy (CRT) for rectal cancer. This study aimed to investigate whether clinical assessment, such as endoscopy and magnetic resonance imaging (MRI), can accurately predict ypT stage and select candidates for pelvic organ-preserving surgery in rectal cancer after preoperative CRT. A total of 110 patients who underwent preoperative CRT followed by curative resection for rectal cancer were prospectively enrolled. Magnetic resonance tumor regression grade (mrTRG) using T2-MRI, endoscopic evaluation, and combination modality (combination of endoscopy and mrTRG) were used to analyze tumor response after preoperative CRT. Endoscopic findings were categorized as 3 grades and the mrTRG was assessed into 5 grades. Twenty-nine patients (26.4%) had achieved pathologic complete response. When predicting ypT0, endoscopy showed significantly higher area under the curve (AUC 0.818) than did mrTRG (AUC 0.568) and combination modality (AUC 0.768) in differentiating good response from poor response (P < .001). Both endoscopy and combination modality showed significantly higher diagnostic performance in sensitivity (79.31%), positive predictive value (PPV 67.65%), negative predictive value (NPV 92.11%), and accuracy (84.55%) than those of MR tumor response (sensitivity 37.93%, PPV 36.67%, NPV 77.50%, and accuracy 66.36%) for the prediction of ypT0 (P < .001). Combination modality showed significantly higher diagnostic performance in sensitivity (56.92%), NPV (56.92%), and accuracy (67.27%) compared with those of mrTRG. Neither endoscopy, nor mrTRG, nor the combination modality had adequate diagnostic performances to be clinically acceptable in selecting candidates for nonoperative treatment strategies. However, endoscopy may be incorporated in clinical restaging strategy in planning the extent of surgical resection in patients with rectal cancer.
All Science Journal Classification (ASJC) codes