P-171 Does metastasectomy really improve survival in gastric cancer?

Abstract

Previous observational studies suggest that metastasectomy improves overall survival (OS) for patients with metastatic gastric/gastroesophageal junction cancer (mG/GEJC). However, most conclusions were based on comparisons of distinct populations, with surgery groups having favorable characteristics that could have influenced the results. All consecutive patients with mG/GEJC who underwent metastasectomy in our institution were retrospectively recruited. After, a non-metastasectomy control group, with a single site of metastasis and ECOG 0-1, was paired in a 1:1 ratio by the nearest neighbor propensity score matching method, using the following pairing categories: age at diagnosis < 60 years vs ≥ 60 years; intestinal vs diffuse subtype; synchronous vs metachronous metastasis; peritoneal vs other site of metastasis. The primary objective was to compare the overall survival (from the metastasis diagnosis to death by any cause) between the groups. Secondarily, prognostic factors associated with OS were evaluated. Time-to-event variables were analyzed by Kaplan-Meier curves and compared by Log-rank test. Cox regression was used for multivariable analysis. Between September 2007 to January 2020, 138 mG/GEJC patients were included (69 in each group). The median follow-up was 37 months. The median age at diagnosis was 54 years (31% were ≥ 60 years old); most were men (57.2%), with comorbidities (55.1%) and without malnutrition (only 8.7%). Gastric cancer (88.4%), diffuse subtype (65.2%), synchronous (77.5%) and peritoneal metastasis (65.2%) were predominant. The characteristics between the metastasectomy and the control group were well balanced, with a non-significant increased proportion of GEJC in the control group (17.4% vs 5.8%, p=0.06). Patients who underwent metastasectomy were more exposed to FLOT and less exposed to FOLFOX regimens (16.4% vs 3.0% and 37.3% vs 52.2%, respectively; p=0.048); and tended to receive triplet regimens more often (41.8% vs 29.9%; p=0.207). The median OS was superior in the metastasectomy group (26.0 vs 14.0 months; HR 0.52, 95%CI 0.35-0.78; p=0.001), as it was the median progression-free survival in first-line (12.0 vs 6.0 months; HR 0.56, 95%CI 0.39-0.80; p=0.001). The median time between the metastasectomy to death was 17.0 months (95%CI 11.4-22.6) and this was not influenced by triplet vs doublet schemes or by the metastasectomy site (peritoneal vs others, linfonodal vs others, or visceral vs non-visceral). Data were immature to evaluate the influence of type of perioperative chemotherapy on these results. Metastasectomy, FLOT regimen and ECOG 0 were independent prognostic factors for improved overall survival in multivariate analysis. Metastasectomy in gastric/gastroesophageal junction cancer was associated with improved overall survival, even when compared to a matched-paired population of metastatic patients with a favorable prognostic profile. This study reinforces the importance of considering this approach, when reasonable, for mG/GEJC patients.