Abstract
BACKGROUND: The role of external beam radiation therapy (RT) in non-surgical gastric cancer (GC) remains controversial due to conflicting trial results and a lack of large-scale real-world evidence. This study utilizes the population-based Surveillance, Epidemiology, and End Results (SEER) database to examine the association between RT and overall survival (OS) in non-surgical GC patients, while acknowledging inherent confounding factors. METHODS: We identified 29,923 non-surgical GC patients [2004-2021] from the SEER database, categorizing them into RT (n=6,629) and non-RT (n=23,294) groups. Baseline demographic and clinicopathological characteristics were collected. Survival was followed until death or censoring. Kaplan-Meier analysis, multivariable Cox regression (adjusting for age, sex, year of diagnosis, marital status, race, tumor site, grade, stage, and chemotherapy), propensity score matching (PSM), and subgroup analyses were employed. RESULTS: This investigation included 29,923 patients with non-surgical GC. Of these, 6,629 (22.2%) received radiotherapy (RT), while 23,294 (77.8%) did not. Baseline characteristics differed significantly between groups (P<0.001), with the RT group having a higher proportion of patients with regional stage disease (30.9% vs. 10.5%) and receiving chemotherapy (82.4% vs. 57.3%). The results showed that the median OS was 10.5 months in both groups (P<0.001), with 3-year survival rates of 12.4% in the RT group versus 17.8% in the non-RT group. Multivariate analysis demonstrated an association between RT and OS [hazard ratio (HR), 1.11, 95% confidence interval (CI): 1.07-1.15, P<0.001], which remained significant after PSM analysis (HR, 1.10, 95% CI: 1.06-1.15). In the subgroup of metastatic patients, the association between RT and OS was attenuated (HR, 1.05, 95% CI: 0.98-1.12). CONCLUSIONS: In this SEER-based analysis, RT was associated with reduced OS in patients with non-surgical gastric cancer. The SEER database's lack of treatment intent and symptom data precludes causal interpretation. RT remains clinically relevant for symptom control, and individualized decision-making is paramount. Prospective studies distinguishing curative versus palliative RT are warranted.