Discordance in TME Specimen Grading in a Prospective Phase II Multicenter Rectal Cancer Trial: Are We Overestimating the Quality of our Resections ?

Overview

abstract

OBJECTIVES: To report the results of a rigorous quality control (QC) process in the grading of total mesorectal excision (TME) specimens during a multicenter prospective phase II trial of transanal TME. SUMMARY BACKGROUND DATA: Grading of TME specimens is based on macroscopic assessment of the mesorectum and standardized through synoptic pathology reporting. TME grade is a strong predictor of outcomes with incomplete (IC) TME associated with increased rates of local recurrence relative to complete or near complete (C/NC) TME. Although TME grade serves as an endpoint in most rectal cancer trials, in protocols incorporating centralized review of TME specimens for quality assurance, discordance in grading and the management thereof has not been previously described. METHODS: A phase II prospective taTME trial was conducted from 2017-2022 across 11 North American centers with TME quality as primary study endpoint. QC measures included training of site pathologists in TME protocols, (2) blinded grading of de-identified TME specimen photographs by central pathologists, and (3) reconciliation of major discordance prior to trial reporting. Cohen's Kappa statistic was used to assess agreement in grading. RESULTS: Overall agreement in grading of 100 TME specimens between site and central reviewer was rated as fair, (κ=0.35 (95% CI, 0.10-0.61, P<0.0001). Concordance was noted in 54%, with minor and major discordance in 32% and 14% of cases respectively. Upon reconciliation, 13/14 (93%) major discordances were resolved. Pre- versus post-reconciliation rates of C/NC and IC TME are 77%/16% and 7% versus 69%/21% and 10%. Reconciliation resulted in a major upgrade (IC to NC, N=1) or major downgrade (NC/C to IC, N=4) in 5 cases overall (5%). CONCLUSIONS: A 14% rate of major discordance was observed in TME grading between site and central reviewers. Resolution resulted in a major change in final TME grade in 5% of cases, which suggests that reported rates or TME completeness are likely overestimated in trials. QC through central review of TME photographs and reconciliation of major discordances is strongly recommended.