O-017 - The prognostic relevance of MRI Tumor Regression Grade versus histopathological complete response in rectal cancer

Date 04 July 2015
Event WorldGI 2015
Session Oral and LBA abstracts
Topics Rectal Cancer
Imaging, Diagnosis and Staging
Pathology/Molecular Biology
Presenter S. Yu
Citation Annals of Oncology (2015) 26 (suppl_4): 108-116. 10.1093/annonc/mdv235
Authors S. Yu, D. Tait, G. Brown
  • The Royal Marsden, Sutton/UK



In rectal cancer, pathological complete response (pCR) is a favoured endpoint for assessing treatment effectiveness. However, a robust evaluation of treatment efficacy prior to surgery enables further options to improve tumour response and improve overall outcomes. The purpose of this analysis was to test the hypothesis that preoperative MRI assessment of tumor regression grade is as effective in outcome prediction as pCR assessment.


215 patients with rectal cancer treated with CRT were identified between 2003-2009. High resolution MRI following CRT was assessed for mrTRG. Multivariate binary logistic regression was used to identify potential independent predictor(s) for pCR. OS and DFS were calculated using the Kaplan Meier product limit method and differences between survival for mrTRG1, mrTRG2, mrTRG3 and mrTRG4-5 were tested for significance using the Mantel-Cox log rank test.


The 3-year DFS for mrTRG 1 was 84% (HR 2.8, 95%CI: 0.86-9.0) and 82% for mrTRG2 (HR 2.3, 95% CI 1.0 - 5.2) and therefore could be considered as equivalent to pCR (DFS 82%, HR 2.8 95%CI: 1.3-6.0). mrTRG3 patients showed intermediate prognosis with a 3 year DFS of 72%. mrTRG 4-5 patients had significantly poorer outcomes, DFS was 61% and similar to non-pCR DFS of 56% (p = 0.005) (Fig 1).


MRI TRG 1-2 patients can be considered prognostically equivalent to pCR and thus supports its use as a prognostic indicator that is equivalent to histopathology compete response.

Figure: O-017. Kaplan-Meier Analysis and survival. (a) pCR and DFS (b) pCR and OS (c) mrTRG and DFS (d) mrTRG and OS