LAUSR

INTRODUCTION In patients with locally advanced rectal cancer, restaging pelvis magnetic resonance imaging (MRI) after neoadjuvant concurrent chemoradiotherapy is recommended despite its limited accuracy in predicting pathologic T (ypT) and N (ypN) stage. Neoadjuvant rectal (NAR) score is a novel short-term surrogate endpoint for disease-free survival (DFS) and overall survival (OS). We tested the agreement between restaging MRI T (yT) and N (yN) with ypT and ypN stages, respectively, and explored the prognostic significance of restaging MRI NAR (mNAR) score. PATIENTS AND METHODS Between 2014 and 2018, 43 patients with locally advanced rectal cancer completed neoadjuvant concurrent chemoradiotherapy, had a restaging MRI, and underwent surgery. Weighted kappa was used to test the agreement between yT and yN with ypT and ypN, respectively. A kappa value of less than 0.5 was deemed unacceptable. Paired t test was used to compare NAR and mNAR mean scores. Survival was estimated by Kaplan-Meier curves. RESULTS Restaging MRI could not predict ypT stage (slight agreement, κ = 0.111) or ypN stage (fair agreement, κ = 0.278). The mean mNAR score was higher than the mean NAR score (20 vs. 16, P = .0079). The median DFS for patients with low-intermediate NAR and high NAR was not reached vs. 30 months (P = .0063). The median OS for patients with low-intermediate NAR and high NAR was not reached vs. 40 months (P = .0056). There was a trend for longer DFS and OS in patients with low-intermediate mNAR scores (not reached in both groups, P = .058) compared to patients with high mNAR scores (not reached in both groups, P = .15). CONCLUSION Restaging MRI could not predict ypT and ypN stage. The mean mNAR score was higher than the mean NAR score. There was a trend for longer DFS and OS in patients with low-intermediate mNAR scores compared to patients with high mNAR scores.