Prognostic Value Of Tumor Size In Gastric Cancer:an Analysis Of 548 Postoperative Patients

Objective: Gastric cancer is one of the most common malignant tumors in China and also zhe forth common cancer in the world.Gastric cancer is the second most common cause of cancer-related deaths across the globe. In China, the situation is even worse.According to the statistics of 2011, gastric canceris estimated to be 23.8 deaths per million which make gastric cancer rank the top of cancer deaths. Similar to the common types such as lung and breast cancer, the tumornode-metastasis(TNM) staging system is also demonstrated as the most important prognostic index for gastric cancer.Submucosal invasion has been included into the staging systems of gastric cancer. wheather the T stageing reflec to the biology character or precisely estimate the prognosis of the primary tumor is unknown.In many malignant tumors,such as lung cancer,breast cancer and liver cancer,T staing is tumor size not the invated depth and toumor size is thought to be the independent factor for evaluating prognosis.The measurement of tumor size can be done preoperation or intraoperation without any special tools and is useful for judgement of the safe incisal margin and the necessity of extended resection.The convenience of the evaluation is help for clinical staging and confirm the range of radical gastrictomy and lymph node dissection.However the impacts of tumor size in the prognosis of gastric cancer are rarely reported in our country.Its prognostic value for gastric cancer needs to be investigated in this paper.Method: We enrolled a total of 548 patients who received radical resection for histopathologically confirmed gastric adenocarcinoma from February 2009 to February 2014 at the Fourth Hospital of Hebei Medical University through the case management system.According to the selection criterias,all the patientswere followded up by phone or letter.The staring time was the date of surgery and the deadline was January 1th 2015.The withdraw patients during the time were defined as censored data.The tumor size was devided into small size gastric cancer group(SGC) and large size gastric cancer group(LGC) by an optimal cutoff point using Cox proportional hazards models. Except for tumor size,the research also involved in other clinical pathology factors,such as sex,age,tumor location, macroscopic type, Lauren,vessel invasion,invaded depth, status of regional lymph nodes,TNM staging and chemotherapy.Kaplan-Meier was used to draw survival curves and the survival difference was compared by using the log-rand test. The independent prognostic factors was determined by Cox proportional hazards model. P<0.05 was considered significant.All analyses were performed using SPSS 16.0 statistical ackages(SPSS, Inc., Chicago, IL, USA).Consequence: 524 patients had completed follow-up data in this research.The media follow-up time was 30 months(range from 10 months to 67 months).Acording to the postoperative pathological diagnosis,the tumor size ranged from 0.5cm to 25 cm and the most sounded cut-off point was 5cm. There was no significant difference of different size in OS(P>0.05).The median PFS of SGC was 25 months.The median PFS of LGC was 19 months. The median PFS of SGC was superior to LGC(P<0.05).SGC had a better survival than LGC determined by univariate analysis,mainly for IA,IIB and IIIA stage.Cox regression model showed tumor size,age,tumor location,vascular invation,depth of tumor invasion,status of regional lymph nodes and TNM stage were independent predictors of PFS for gastric cancer patients(P<0.05). Macroscopic type and chemotherapy were independent predictors for SGC and sex was independent predictors for LGC.Conclusion: 1 The optimal cutoff point for tumor size was 5cm by Cox regression model and devide all the patients into SGC and LGC group. 2 524 patients had unbroken follow-up meterials in this study.The mediafollow-up time was 30 months(10-67months). There was no significant difference of different size in OS(P>0.05).The median PFS of SGC was 25 months.The median PFS of LGC was 19 months. The median PFS of SGC was superior to LGC(P<0.05).SGC had a better survival than LGC determined by univariate analysis,mainly for IA,IIB and IIIA stage. 3 Cox regression model showed tumor size,age,tumor location,vascular invation,depth of tumor invasion,status of regional lymph nodes and TNM stage were independent predictors of PFS for gastric cancer patients. 4 Tumor size was an independent prognostic factor for patients with gastric cancer and it would be helpful for the clinical staging and therapy.