Surgery is associated with improved survival in elderly patients with gastric cancer based on a competing risk analysis of the SEER database

Abstract

Abstract Background Gastric cancer (GC) is one of the most common malignancies worldwide and occurs with a particularly high incidence in the elderly population. Given the substantial burden of comorbidities and the elevated risk of non–cancer-related mortality in older patients, the true survival benefit of surgical treatment remains controversial. This study aimed to evaluate the impact of surgery on survival outcomes in elderly patients with gastric cancer using competing risk analysis. Methods Patients aged ≥ 75 years with gastric cancer were identified from the Surveillance, Epidemiology, and End Results (SEER) database and were classified into surgical and non-surgical groups according to whether tumor resection was performed. Propensity score matching (PSM) was applied to balance baseline characteristics between groups. Kaplan-Meier survival analysis and Fine-Gray competing risk models were used to assess the association between surgery and survival outcomes. A nomogram was subsequently developed to predict individualized survival probabilities. Results A total of 9201 elderly patients with gastric cancer were included, including 5,730 who underwent surgery and 3,471 who did not. After PSM, the surgical group showed significantly better 5-year overall survival (OS) than the non-surgery group ( p < 0.001), with a markedly lower 5-year cumulative incidence of cancer-specific death (CSD) (30.60% vs. 62.10%, p < 0.01). The competing risk–based nomogram demonstrated good discrimination, with 1-, 3-, and 5-year AUCs of 0.803, 0.813, and 0.800 in the training cohort and 0.787, 0.803, and 0.792 in the validation cohort, and showed good calibration between predicted and observed outcomes. Conclusions Among patients aged ≥ 75 years with gastric cancer, surgical treatment was associated with improved survival, particularly through a reduction in cancer-specific mortality. Competing risk analysis may provides a more accurate assessment of the association between surgical treatment and survival outcomes in this high-risk population and may facilitate individualized decision-making in elderly patients.