De novo metastatic versus recurrent gastroesophageal cancer: A single-centre retrospective analysis.

Abstract

282 Background: Outcomes remain poor for patients with metastatic gastric, gastroesophageal and esophageal adenocarcinoma (GEAC) with median survival of 14 months with combination chemotherapy and immunotherapy as per the Checkmate 649 trial. There is limited data to clarify if any differences exist between denovo metastatic and recurrent GEAC in terms of survival and response to systemic therapy. We compare the baseline characteristics and outcomes for patients with denovo metastatic and recurrent GEAC in our cohort. Methods: A retrospective observational analysis was conducted to include patients with metastatic GEAC reviewed at Princess Margaret Cancer Centre between 2007 and 2021. Baseline characteristics including age, sex, race, performance status and Charlson Comorbidity index (CCI) were reviewed. Outcomes of interest included overall survival (OS), progression free survival on first line (PFS1) and second line (PFS2) systemic therapy and disease-free interval (DFI) in the recurrent group (defined as time between initial diagnosis and development of recurrent disease). OS was calculated from diagnosis of stage IV disease. Cox proportional hazards and Kaplan Meier curves were used to compare OS, PFS1 and PFS2. Outcomes were adjusted for baseline characteristics. The effect of DFI on outcomes was evaluated in the recurrent group. Results: Our cohort consisted of 619 cases, of which 456 (73.6%) were denovo metastatic and 163(26.4%) were recurrent. The majority (71%) were male, non-Asian (85%) and median age for the cohort was 60 years. There were no significant differences in baseline characteristics and biomarker status (HER2, MSI) between denovo and recurrent groups. Number of metastatic sites was significantly higher in denovo group (Mean:2.3 vs 1.8, p<0.001). A significantly higher percentage of patients received immunotherapy in the first line setting in the recurrent group (8% vs 4%, p:0.04). Using multivariate cox regression and after adjusting for age, performance status and CCI, there was a significant difference in OS favoring recurrent group (aHR:0.70, 95% CI 0.57-0.86, p=<0.001). Using Kaplan Meier curves, median OS was 18.08 months in the recurrent group vs 14.62 months in the denovo metastatic group. There was a significant difference in PFS1 using a multivariate model in favor of the recurrent group (aHR 0.81,95% CI 0.67-0.99, p=0.04), as well as similar improvement in PFS2 for the recurrent group (aHR:0.69,95% CI 0.52-0.92, p=0.01). In the recurrent group, OS did not differ significantly between DFI <6 months, 6-12 months or >12 months. Conclusions: Our results show a significantly better OS, PFS1 and PFS 2 for recurrent as compared to denovo metastatic GEAC for patients treated in the 1st line metastatic setting in a large single-centre retrospective analysis. More in depth analysis of molecular differences in patient tumors is underway and may help to explain better outcomes.