Objective To explore the effectiveness of synthetic MRI sequences in predicting the treatment response of patients with locally advanced rectal cancer (LARC) undergoing neoadjuvant chemoradiotherapy treatment (nCRT). Methods A total of 51 patients with biopsy-confirmed rectal adenocarcinoma were enrolled at Ruijin Hospital from August 2023 to June 2024. All patients were assessed as having LARC by baseline MRI and received nCRT followed by radical surgery. All subjects completed synthetic MRI and high-resolution T2-weighted imaging (T2WI) scans within two weeks before receiving nCRT treatment. Based on high-resolution T2WI images, radiologists assessed extramural vascular invasion (mrEMVI) at baseline in the subjects. After the synthetic MRI sequence scanning was completed, synthetic images of T1 mapping, T2 mapping, and proton density (PD) mapping were generated using Synthetic MR post-processing software. Histogram-based quantitative parameters at baseline were extracted using python software, including quantitative parame-ters of the primary tumor and peritumoral fat region: T1 relaxation time (T1RT), T2 relaxation time (T2RT), and proton density (PD). Using postoperative pathological results as the gold standard, patients were grouped according to: (1) primary tumor response: pathological complete response (pCR) vs. non-pCR; (2) tumor regression grade: (TRG) 0-1 vs. TRG 2-3; and (3) mesorectal lymph node metastasis status: positive (ypN+) vs. negative (ypN-). Differences in baseline mrEMVI status and quantitative parameters of the primary tumor and peritumoral fat among different groups were compared using Student's t-test, Mann-Whitney U test, and Chi-square test. Binary logistic regression was used to identify independent risk factors for predicting TRG grade, pCR status, and mesorectal lymph node status. Based on the selected risk factors, logistic regression models were established. The predictive performance of the quantitative parameters, mrEMVI status, and the regression models for TRG grade, pCR status, and mesorectal lymph node status was evaluated using receiver operating characteristic (ROC) curves. Results Baseline mrEMVI positivity (P=0.03) and quantitative parameters of peritumoral fat tissue-including the maximum T2RT_Fat (139.53 ms vs. 129.60 ms, P=0.03), 90th percentile (189.18 ms vs. 174.00 ms, P=0.03), root mean square (120.09 ms vs. 115.48 ms, P=0.04), and lower T2RT_Fat uniformity (0.54 vs. 0.61, P=0.04)—were indicative of positive mesorectal lymph node status after nCRT. None of the observed indicators were correlated with the primary tumor response. Logistic regression analysis showed that mrEMVI and elevated T2RT_Fat_P₉₀ were independent risk factors for predicting mesorectal lymph node metastasis. The logistic regression model combining both mrEMVI (AUC=0.667) and T2RT_Fat_P₉₀ (AUC=0.692) demonstrated good predictive performance (AUC=0.747), although the improvement was not statistically significant. Conclusions T2RT_Fat_P₉₀, extracted from baseline MAGiC synthetic MRI serves as a non-invasive imaging biomarker for predicting mesorectal lymph node metastasis after nCRT. The combination of T2RT_Fat_P₉₀ and baseline mrEMVI can be used as an auxiliary tool for predicting the mesorectal lymph node metastasis status in LARC patients following nCRT.