Objective To investigate the clinical value of multi-modal magnetic resonance imaging using magnetic resonance tumor regression grading (mrTRG) combined with MRI morphological parameters in evaluating the prediction efficacy after neoadjuvant concurrent chemoradiotherapy for locally advanced rectal cancer preoperatively.Methods The clinical data were collected of 48 patients with locally advanced rectal cancer (stage T3~T4) patients undergoing surgical treatment after neoadjuvant concurrent chemoradiotherapy (NCRT) at Hefei Cancer Hospital of Chinese Academy of Sciences and the Western Branch of the First Affiliated Hospital of University of Science and Technology of China (Anhui Cancer Hospital) from May 2019 to September 2023. According to the postoperative pathological tumor regression grade (pTRG), the patients were divided into the good curative effect group (19 cases) and the poor curative effect group (29 cases). All patients underwent high-resolution rectal multi-modal MRI before and after neoadjuvant chemoradiotherapy. The diagnostic efficacy of mrTRG was analyzed using the Kappa test with pTRG as the gold standard, and baseline MRI morphological parameters were compared between the two groups. Receiver operating characteristic (ROC) curves were used to determine the efficacy of multi-modal magnetic resonance imaging using mrTRG combined with MRI morphological parameters in evaluating the therapeutic effect after neoadjuvant concurrent chemoradiotherapy. Results The modified dichotomous tumor regression grading scheme was adopted. mrTRG showed medium-high concordance with pTRG, with a kappa value of 0.698 (95% CI: 0.492~0.904,P<0.05) . Single factor analysis showed that the differences in T3a stage, circumferential resection margin (CRM)and mrTRGbiomarks between the good and poor curative effect groups were statistically significant (P<0.05). The area under the curve (AUC) of NCRT response assessed by mrTRG, baseline T3a stage, and CRM was 0.842, 0.693, and 0.721, respectively, and the combined model had a sensitivity of 89.5%, a specificity of 82.8%, and an AUC of 0.907, which was significantly superior to baseline morphologic parameters (P<0.05), and there was no statistically significant differences between the combined model and mrTRG mark(P=0.07).Conclusion Preoperative mrTRG combined with baseline morphological parameters has high diagnostic efficacy in assessing NCRT response in rectal cancer, and it provides an imaging evidence for the clinical setting of individualized treatment plan. |