P-152 Real-world data in patients with localized esophageal and gastro-esophageal junction cancer undergoing surgery in France: Results from the FREGAT database

Abstract

Esophageal cancer (EC) is the third most common digestive cancer in France with an incidence of 5450 new cases in 2018. All stages combined the prognosis remains poor with 5-year survival rates around 14%. In non-metastatic patients, the benefits of adding perioperative chemotherapy (CT) or neoadjuvant chemoradiotherapy (NACRT) to surgery have been demonstrated and are currently recommended. Immune checkpoint inhibitors (ICI) have showed promising results. The aim of the study was to provide real-world treatment outcomes before arrival of ICI in the therapeutic arsenal. We conducted a retrospective cohort study of patients with EC including gastro-esophageal junction cancer (GEJC) who underwent surgery. Patients who underwent surgery between 2014-2019 were identified from the FREGAT database which collected clinical characteristics and treatment for approximately 4000 patients from 35 French centres. Overall survival (OS) and Disease-Free Survival (DFS) were assessed using the Kaplan-Meier method. The FREGAT database included 2262 patients with EC/GEJC, 1366 (60.3%) had stage II or III disease. From them, 833 (60.9%) underwent surgery. Median age was 64 years, 82% were male and 97% had an ECOG 0-1. 228 (27.4%) patients presented with esophageal adenocarcinoma (EAC), 218 (26.2%) with esophageal squamous cell carcinoma (ESCC) and 387 (46.4%) with GEJC. NACRT was administered to 382 patients (45.8%), neoadjuvant chemotherapy (NACT) to 350 (42.0%) and 101 patients (12.1%) had primary surgery. NACRT was almost equally used in both histologies (43.7% - 56.3%) whereas NACT was prominently used in adenocarcinoma (96.6%). 258 patients (30.9%) received an adjuvant CT of whom 23 had no neoadjuvant therapy. Among patients receiving NACRT, only 20 (5.2%) underwent adjuvant CT in accordance with local guidelines and 259 (67.8%) had residual disease on pathological analysis. Considering neoadjuvant treatment, the most frequently used CT regimen was FOLFOX (48.5%) followed by carboplatin+paclitaxel (18.2%). Neoadjuvant FOLFOX use was consistent across histologies (47.0 % for EAC and GEJC and 53.1% for ESCC). Median OS and DFS after surgery were 37.0 months (95% CI: 32.8-43.0) and 19.7 months (95% CI: 16.9-24.1) respectively with no significant differences by cancer localization and histology. Also, by initial treatment (NACRT/NACT/primary surgery), there were no significant differences in median OS: after NACRT 35.2 months (95% CI: 28.2-43.0), after NACT: 37.5 months (95% CI: 31.4-md) and after primary surgery 38.4 months (95% CI: 26.0-md). There were no significant differences in median DFS (after NACRT 16.9 months (95% CI: 13.3-23.3) vs after NACT 21.6 months (95% CI: 16.0-md) vs after primary surgery 23.5 months (95% CI: 18.8-36.1), respectively, p=0.166). In resected population receiving NACRT, 232 patients (60.7%) met the inclusion criteria of the adjuvant nivolumab phase III trial (complete resection, and residual pathological disease) and could benefit from this new therapeutic option. This study offers insights into treatments options and outcomes in French patients with resectable EC/GEJC before arrival of ICI.