Application of Cox model for analyzing prognosis factors of stage Ⅳ gastric cancer / 重庆医学

ABSTRACT

Objective To investigate the related prognostic factors of stage Ⅳ gastric cancer.Methods The clinical data of 248 patients with stage Ⅳ gastric cancer and intact follow up data in the Tumor Prevention and Treatment Center of Sun Yat-Sen University from 2000 to 2010 were retrospectively summarized.The twelve clinicopathological parameters served as the observation indicators,including age,sex,body mass reduction,H b,CEA,CA19-9,Borrmann type,tumor location,tumor size,pathological pattern,operative mode,metastatic sites and therapeutic model.The survival curve was drawn by using the Kaplan-Meier method.The median survival time was calculated.The univariate analysis was conducted with Log-rank test.The prognosis multivariate analysis was conducted by the Cox's proportional hazards regression analysis.Results MST in the patients of whole group was 254 d.The univariate analysis showed that sex,Borrmann type and therapeutic mode were the related factors afecting gastric cancer prognosis,while the Cox regression model revealed that above 3 indicators were also independent factors affecting the prognosis of the patients with stage Ⅳ gastric cancer in this group(P＜0.05).Conclusion The treatment mode is an important independent factor affecting the survival of stage Ⅳ gastric cancer,the translational medicine model of palliative chemotherapy combined with palliative operation conduces to improve the prognosis in the patients with stage Ⅳ gastric cancer.