RESUMO
OBJECTIVE: To explore the effect of perioperative and postoperative chemotherapy on the prognosis of patients with advanced gastric cancer after resection. METHODS: From January 2004 to December 2016, a retrospective cohort of 277 stage IV gastric cancer patients after surgical treatment of gastric resection was enrolled in Department of General Surgery,Nanfang Hospital, Southern Medical University. Among them, 228 underwent surgical resection combined with postoperative adjuvant chemotherapy(postoperative chemotherapy group, group A), and preoperative chemotherapy plus surgery resection and postoperative chemotherapy in 49 cases(perioperative chemotherapy group, group B). After generating propensity scores with eight covariates, including gender, age, biological classifications, completion of chemotherapy, depth of tumor infiltration, lymph nodemetastasis, the extent of lymph nodes dissection and type of gastrectomy, 49 patients in group A were one-to-one matched with 49 patients in group B. Kaplan-merier method was used for survival analysis, and Cox proportional risk regression model was used to analyze independent survival risk factors of patients with advanced gastric cancer undergoing surgical resection. RESULTS: Before propensity scores matching(PSM), biological classifications(P<0.001),the completion of chemotherapy(P<0.001), depth of tumor infiltration(P<0.001), lymph node metastasis(P=0.049), the extent of lymphnode dissection(P=0.001) and the type of gastrectomy(P=0.001) significantly differed between two groups. While after PSM, only the completion of chemotherapy were vital different between two groups. After PSM,median survival time of group A and B were 16(95% CI 10.36-21.64) vs 29(95% CI 17.24-40.76) months, which showed not significant difference(P=0.191). The univariate analysis showed that biological classifications, the completion of chemotherapy, lymph node metastasis and the extent of lymphnode dissection were survival prognosis factors. And the multivariate analysis showed that the chemotherapy cycles≤ 2 cycles, lymph node metastasis and the extent of lymphadenectomyless than D2 were independent poor prognostic factors for advanced gastric cancer patients performed with gastrectomy. Notably, the order of chemotherapy and surgery was not independent variate of prognosis(perioperative chemotherapy vs. postoperative chemotherapy: HR 0.986,95%CI 0.539-1.806,P=0.964). CONCLUSION: Perioperative chemotherapy is not the independent variable to improve prognosis of resected advanced gastric cancer patients.Nonetheless, perioperative chemotherapy could benefit patient's chemotherapy tolerance and compliance, which could be attributed to the superiority in survival compared to postoperative chemotherapy. The finding could offer reference and guidance for further design of perspective studies for advanced gastric cancer patients.