Jian-Xian Lin1,2,3, Wei Wang4, Jun-Peng Lin1,2, Jian-Wei Xie1,2,3, Jia-Bin Wang1,2,3, Jun Lu1,2, Qi-Yue Chen1,2, Long-Long Cao1,2, Mi Lin1,2, Ruhong Tu1,2, Chao-Hui Zheng1,2,3, Chang-Ming Huang5,6,7, Zhi-Wei Zhou8, Ping Li9,10,11. 1. Department of Gastric Surgery, Fujian Medical University Union Hospital, Fuzhou, Fujian Province, China. 2. Department of General Surgery, Fujian Medical University Union Hospital, Fuzhou, Fujian Province, China. 3. Key Laboratory of Ministry of Education of Gastrointestinal Cancer, Fujian Medical University, Fuzhou, Fujian Province, China. 4. Department of Gastric and Pancreatic Surgery, Sun Yat-sen University Cancer Center, Guangzhou, Guangdong Province, China. 5. Department of Gastric Surgery, Fujian Medical University Union Hospital, Fuzhou, Fujian Province, China. hcmlr2002@163.com. 6. Department of General Surgery, Fujian Medical University Union Hospital, Fuzhou, Fujian Province, China. hcmlr2002@163.com. 7. Key Laboratory of Ministry of Education of Gastrointestinal Cancer, Fujian Medical University, Fuzhou, Fujian Province, China. hcmlr2002@163.com. 8. Department of Gastric and Pancreatic Surgery, Sun Yat-sen University Cancer Center, Guangzhou, Guangdong Province, China. zhouzhw@sysucc.org.cn. 9. Department of Gastric Surgery, Fujian Medical University Union Hospital, Fuzhou, Fujian Province, China. Pingli811002@163.com. 10. Department of General Surgery, Fujian Medical University Union Hospital, Fuzhou, Fujian Province, China. Pingli811002@163.com. 11. Key Laboratory of Ministry of Education of Gastrointestinal Cancer, Fujian Medical University, Fuzhou, Fujian Province, China. Pingli811002@163.com.
Abstract
OBJECTIVE: The aim of this study was to determine the prognostic significance of preoperative carcinoembryonic antigen (CEA) and carbohydrate antigen (CA) 19-9 levels in patients with resectable gastric cancer (GC). PATIENTS AND METHODS: Patients who underwent a radical resection for GC at the Fujian Medical University Union Hospital between 2007 and 2014 were included in this study. The estimated area under the curve (AUC) was compared to evaluate the discriminatory ability of tumor makers. Additional external validation was performed using a dataset from Sun Yat-sen University Cancer Center. RESULTS: Preoperative CEA/CA19-9 levels were identified as an independent predictor of overall survival (OS) and disease-specific survival (DSS) (both p < 0.05) in the development group. In a subgroup analysis based on TNM stage, preoperative CEA/CA19-9 levels clearly stratified the survival rates for stage III GC (p < 0.05). A multivariate analysis revealed that preoperative CEA/CA19-9 levels were an independent prognostic factor (p < 0.05) in stage III; the AUC of the preoperative CEA/CA19-9 was equivalent to that of T stage. A prediction model (TNMC) for stage III GC was developed by incorporating preoperative CEA/CA19-9 levels into the American Joint Committee on Cancer (AJCC) staging system. The AUC of the TNMC was significantly higher than that of the TNM staging system at 1, 3, and 5 years postoperatively (all p < 0.05), with similar results also being obtained in the external validation set. CONCLUSION: Preoperative CEA/CA19-9 levels are an independent predictor of OS and DSS in stage III GC patients. The inclusion of preoperative CEA/CA19-9 levels in AJCC TNM staging provided an optimal prognosis in stage III GC.
OBJECTIVE: The aim of this study was to determine the prognostic significance of preoperative carcinoembryonic antigen (CEA) and carbohydrate antigen (CA) 19-9 levels in patients with resectable gastric cancer (GC). PATIENTS AND METHODS: Patients who underwent a radical resection for GC at the Fujian Medical University Union Hospital between 2007 and 2014 were included in this study. The estimated area under the curve (AUC) was compared to evaluate the discriminatory ability of tumor makers. Additional external validation was performed using a dataset from Sun Yat-sen University Cancer Center. RESULTS: Preoperative CEA/CA19-9 levels were identified as an independent predictor of overall survival (OS) and disease-specific survival (DSS) (both p < 0.05) in the development group. In a subgroup analysis based on TNM stage, preoperative CEA/CA19-9 levels clearly stratified the survival rates for stage III GC (p < 0.05). A multivariate analysis revealed that preoperative CEA/CA19-9 levels were an independent prognostic factor (p < 0.05) in stage III; the AUC of the preoperative CEA/CA19-9 was equivalent to that of T stage. A prediction model (TNMC) for stage III GC was developed by incorporating preoperative CEA/CA19-9 levels into the American Joint Committee on Cancer (AJCC) staging system. The AUC of the TNMC was significantly higher than that of the TNM staging system at 1, 3, and 5 years postoperatively (all p < 0.05), with similar results also being obtained in the external validation set. CONCLUSION: Preoperative CEA/CA19-9 levels are an independent predictor of OS and DSS in stage III GC patients. The inclusion of preoperative CEA/CA19-9 levels in AJCC TNM staging provided an optimal prognosis in stage III GC.