Multivariate prognostic study on large gastric cancer

Background: Although many authors investigate the prognostic factors of gastric cancer, there are few comprehensive. studies on the prognosis of patients with large gastric cancer. The aim of this study was to clarify the prognostic factors of large gastric cancer using multivariate analysis. Methods: The study included 95 patients who underwent gastrectomy for gastric cancer measuring 10 cm or more in diameter. We examined 24 clinicopathologic factors based on patient, operation, and tumor findings. Survival rates were analyzed by the Kaplan-Meier and Mantel-Cox method, and multivariate analysis was done using the Cox proportional hazards model. Results: Overall 5-year survival rate was 22%, and median survival period was 15 months: The 5-year survival-rate was influenced by the tumor size, gross type, serosal invasion, extragastric lymph node metastasis, liver metastasis, peritoneal dissemination, stage of disease (I, II vs. III, IV), resection margin, and operative curability (R0 vs. R1, R2). Of these, independent prognostic factors were three tumor findings: serosal invasion (absent vs. present, odds ratio 3.06, P < 0.01), extragastric lymph node metastasis (absent vs. present, odds ratio 2.13, P < 0.05), and liver metastasis (absent vs. present, odds ratio 3.77, P < 0.05). The survival was not significantly associated with any of the patient factors or operation factors including the extent of lymph node dissection. Conclusion: In patients with large gastric cancer, independent prognostic factors were serosal invasion, extragastric lymph node metastasis, and liver metastasis. Prognosis after gastectomy was determined by these tumor factors and was not associated with the patient or operation factors.