Prognostic value of tumor regression grade (TRG) in patients with gastric/gastroesophageal junction cancer treated by peri-operative chemotherapy and surgery: A single-center experience (Institut Mutualiste Montsouris, Paris, France).

Abstract

486 Background: Management of locally advanced gastric and gastroesophageal junction (GOJ) cancer relies on peri-operative chemotherapy and radical surgery. Response to neoadjuvant chemotherapy can be evaluated by Mandard Tumor Regression Grade (TRG) performed on the removed specimens. The aim of this study was to investigate the prognostic value of TRG, in terms of post-surgical disease-free survival (DFS) and overall survival (OS). Methods: Retrospective analysis of all consecutive patients (Jan 2010 to Dec 2021) who underwent oncological gastrectomy for gastric or GOJ adenocarcinoma after neoadjuvant chemotherapy was performed. TRG was evaluated according to Mandard classification, ranging from 1 (complete pathological response) to 5 (absence of response), TRG 1-2 being considered as “good responders” and TRG 3-5 as “bad responders”. Univariate and multivariate analysis were performed for DFS and OS. Results: One hundred and ninety-nine patients were included, 157 treated with FOLFOX and 42 with FOLFOX + Taxanes (either FLOT or FOLFOX + Abraxane). TRG 1-2 was observed in 30% in patients (pts) treated with FOLFOX and 24% in pts treated with FOLFOX + Taxanes. With a median follow-up over 5 years (yrs), median DFS and OS were not reached (DFS rates at 3, 5 and 10yrs of 66.5%, 64.1% and 55.7%, respectively; OS rates at 3, 5 and 10yrs of 75.6%, 66.8% and 60.4%, respectively). In univariate analysis, TRG 1-2 was significantly associated with favorable outcome compared to TRG 3-5 in terms of DFS (HR 2.4, p-value 0.0056) and OS (HR 2.6, p-value 0.0061), as well as ECOG at diagnosis (0 vs. 1), ypT stage (yp T0-2 vs. yp T3-4), ypN stage (ypN- vs. ypN+), tumor localization (antrum + body vs. cardia + esophagus), and lymphatic, vascular and peri nervous embolisms (negative vs. positive) for both DFS and OS. However, in multivariate analysis (selection backward Cox-model), only ypT stage and vascular embolisms for DFS, and ypT and ypN stages for OS were found to be independent prognostic factors. Conclusions: Although TRG is a prognostic factor in patients with gastric or GOJ cancer, it does not appear as an independent factor for DFS and OS in this series of 199 pts. These results raise the issue of “good responder” definition, especially for TRG 3. Prognostic impact of TRG needs to be investigated for patients treated with combination of immune checkpoint inhibitors and chemotherapy in neoadjuvant regimens.