A Novel Nomogram Integrated with Inflammation-Based Factors to Predict the Prognosis of Gastric Cancer Patients

Abstract

Background: Gastric cancer (GC) is the fifth most common cancer worldwide, approximately 950,000 individuals are diagnosed with gastric cancer worldwide, causing 800,000 mortalities every year. Our study aimed to establish an effective nomogram to predict the prognosis of GC patients, which based on inflammation biomarkers. Methods: We retrospectively analysed GC patients from the Sun Yat-sen University Cancer Center between 2009 and 2017. The nomogram was developed with a primary cohort (n=1067), and 537 patients was included in the validation cohort. Univariate and multivariate Cox survival analyses were used to identify the prognostic factors. We tested the accuracy of the nomogram by discrimination and calibration tests and then performed a decision curve analysis that compared the nomogram with the AJCC TNM stage. Furthermore, according to the total points assigned by the nomogram, GC patients were stratified into three risk groups. Findings: The univariate survival analyses included 19 biomarkers. The multivariate analysis showed that tumour stage (P<0.001), metastasis stage (P<0.001), and C-reactive protein (CRP) (P<0.001), albumin (ALB) (P = 0.013), carcinoembryonic antigen (CEA) (P = 0.001) and carbohydrate antigen-199 (CA199) (P =0.001) levels as well as lymphocyte (LYM) (P = 0.002) count were independent risk factors for the prognosis of GC patients. The nomogram was based on the above factors. In the primary cohort, the nomogram had a concordance index (C-index) of 0.825 (95% CI 0.796-0.854), which was higher than the C-index of the AJCC TNM stage (0.769, 95% CI 0.741-0.797) and that the other biomarkers (CEA and CA199). The calibration plot suggested good agreement between the actual and nomogram-predicted overall survival (OS) probabilities, and the decision curve analyses showed that the nomogram model had a higher overall net benefit in predicting OS than the AJCC TNM stage. Moreover, we divided the patients were into the following three distinct risk groups for OS based on the nomogram points: a low-risk group, middle-risk group and high-risk group. The differences in OS rates were significant among the subgroups (P<0.001). Interpretation: A novel nomogram integrated with inflammatory prognostic factors was proposed, which is highly predictive of OS in GC patients. Funding Statement: This study was supported by Natural Science Foundation of Guangdong Province, China (grant no. 2018A030310260) and Medical Scientific Research Foundation of Guangdong Province, China (grant no. 2018102516469945). Declaration of Interests: The authors have declared that they have no conflict of interest. Ethics Approval Statement: Written informed consent for the use of plasma and serum samples was obtained from all HCC patients. This study was approved by the Institute Research Ethics Committee of the Sun Yat-Sen University Cancer Center, Guangzhou, China.