Correlation of Magnetic Resonance Imaging With the Histopathological Staging in Mid and Low Rectal Cancers

Background Magnetic resonance imaging (MRI) is the preferred modality for local staging of rectal cancer and plays a crucial role in guiding neoadjuvant therapy and surgical planning. However, its accuracy in post-neoadjuvant chemoradiotherapy (NACRT) restaging-particularly for tumor (T) stage, nodal status, and tumor regression-remains variable, largely due to post-treatment fibrosis, edema, and inflammatory changes that complicate differentiation between residual tumor and treatment-related effects. Objectives To evaluate the correlation between post-NACRT MRI staging and final histopathological staging in patients with carcinoma of the mid and low rectum, and to assess the accuracy of MRI in predicting tumor stage, nodal status, circumferential resection margin (CRM), and tumor regression grade. Methods This prospective observational study included 61 consecutive eligible patients with biopsy-proven, locally advanced, non-metastatic adenocarcinoma of the mid and low rectum treated between June 2020 and October 2021. All patients underwent baseline pelvic MRI for assessment of pre-treatment nodal status (cN0 and cN2), followed by long-course neoadjuvant chemoradiotherapy (NACRT) and repeat MRI six to eight weeks after treatment to document post-therapy nodal status. Post-NACRT MRI findings were compared with final histopathological staging, which served as the reference standard. Agreement was assessed using Cohen's kappa statistic, and diagnostic accuracy was calculated. Results The median age of the study population was 48 years, with an almost equal gender distribution (males: 31/61, 50.8%; females: 30/61, 49.2%). Prior to neoadjuvant chemoradiotherapy (NACRT), most patients had locally advanced disease, with cT3-cT4 tumors in 57 patients (57/61, 93.4%). Pre-treatment nodal staging on MRI demonstrated cN2 disease in 35 patients (35/61, 57.4%) and cN0 status in the remaining patients. Post-treatment MRI demonstrated significant downstaging, with cT3-cT4 disease observed in 34 patients (34/61, 55.7%). Following NACRT, nodal downstaging was evident, with cN0 status in 29 patients (29/61, 47.5%) and persistent cN2 disease in the remaining patients. The overall accuracy of MRI for post-NACRT T staging was 54.1%, showing fair agreement with histopathology (κ = 0.37; p = 0.001), while nodal staging accuracy was higher at 73.8%, with moderate agreement (κ = 0.55; p = 0.001). Pathological complete response was observed in 14 patients (14/61, 23.0%), of whom MRI correctly identified three patients (3/14, 21.4%), yielding a sensitivity of 21.4% and a positive predictive value of 75.0%. MRI-based tumor regression grading demonstrated moderate agreement with pathological tumor regression grade (κ = 0.42), with an overall accuracy of 59.0%. MRI-guided post-treatment surgical planning resulted in negative circumferential resection margins in all patients. Conclusion High-resolution MRI is a valuable tool for post-NACRT assessment and surgical planning in mid and low rectal cancer. While MRI demonstrates moderate accuracy for nodal staging and tumor regression assessment, its performance in T-stage restaging remains limited due to post-treatment changes. MRI plays a critical role in achieving negative resection margins through individualized surgical planning. Advances in MRI technology and larger multicenter studies may further improve staging accuracy and clinical decision-making.