Prediction of Early-Onset Gastric Carcinoma by the Multiplicative Model: a Case-Control Study of Hereditary Polygenic Gastric Cancer in Han Chinese.

Jiajia Yuan,Jing Gao,Lin Shen

DOI: https://doi.org/10.1200/jco.2015.33.15_suppl.e12555

IF: 45.3

2015-01-01

Journal of Clinical Oncology

Abstract:e12555 Background: Recent genome-wide association studies (GWASs) have identified several germline variations associated with gastric cancer. The aim of this study was to investigate the individual and combination effects of those single nucleotide polymorphisms (SNPs), which related to the risk of early-onset gastric cancer in a Chinese Han population. Methods: This case-control study included 167 gastric cancer patients and 173 healthy unrelated controls. Polymerase chain reaction (PCR) methods were used to perform the genotyping of the SNPs PSCA rs2294008, ZBTB20 rs4072037, and MUC1 rs9841504. Assuming each locus is independent without interactions and the risk is allele-dose-dependent with a simple multiplicative (log-additive) effect, we calculated the combined effect on the relative-risk scale. Results: We confirmed that the SNPs PSCA rs2294008, ZBTB20 rs4072037, and MUC1 rs9841504 were associated with increased risk for gastric cancer in Han Chinese. However, each of those low-penetrance susceptibility polymorphisms alone was not convincing enough to predict the absolute risk of early-onset gastric cancer. Thus, the multiplicative model for polygenes was applied for our population, and groups with different risk percentiles were stratified. The risk stratification was suggested to be useful for risk prediction of early-onset gastric cancer, and cost-effective for the general population prevention. Conclusions: These findings provide evidence that, instead of single low-penetrance susceptibility variations, application of the multiplicative model for polymorphisms may contribute to the gastric cancer screening by distinguishing population who are at high risk from those who are at low risk.