Gastric cancer (GC) is a common cancer type globally. Newly diagnosed gastric cancer account for 5.7% of all cancers, with over a million cases a year. The high incidence rate and high mortality rate make GC one of the most important cancers threatening human health [1, 2]. GC is the third leading cause of cancer-related death. The incidence rate differs depending on country, and has a higher incidence in East Asia, Eastern Europe, and South America. But the incidence rate is relatively low in Oceania, Africa, and North America [3]. Previous studies have reported some risk factors for GC, the most common of which include Helicobacter pylori infection, tobacco smoking, low fiber intake, high salt or smoked food intake, and genetic factors [4].

Vascular endothelial growth factor (VEGF) is an effective mitogen in the endothelium. Its main role is to regulate angiogenesis and postnatal vascular remodeling. Its expression is up- regulated under various pathophysiological conditions [5]. Therefore, vascular endothelial growth factor is a lymphangiogenic growth factor, which plays an important role in tumor lymphangiogenesis by activating VEGF receptor [6]. VEGF gene is located on chromosome 6p21.1 and contains 9 exons [7]. Previously, several studies have reported the relationship between VEGF gene single nucleotide polymorphism (SNP) and several cancer’s risk, including bladder cancer [8], lung cancer [9], and renal cell carcinoma [10]. However, the etiology of the relationship between VEGF gene SNPs and GC risk is very complicated and yet to be completely understood. In addition, the pathogenesis of GC involves not only genetic factors but also environmental factors; no study has focused on the synergistic effect between VEGF gene and environmental factors. Therefore, the main purpose of this study is to evaluate whether there is a statistical association between VEGF gene polymorphism and GC risk in the Chinese Han population, and the impact of the interaction between VEGF gene and environmental lifestyle factors on GC risk.

A total of 1,451 participants with an average age of 64.0 ± 10.1 years were selected, made up of 483 GC patients and 968 healthy controls. Table 2 shows the general demographic characteristics and clinical indicators in GC patients and controls. The mean age, BMI, and percentage of males were not significantly different between GC patients and controls. The percentage for current smoking and alcohol drinking were significantly higher in GC patients than in controls. In addition, 67.7% of the GC patients were non-cardia and the tumor size of 81.2% GC patients was ≤5 cm.

The genotype distribution in the control group was consistent with the HWE (All p- values more than 0.05). The allele frequencies of rs10434 -G and rs833061-T were significantly higher in GC patients than that in control group (30.0% and 20.0%, 28.9% and 20.7%, respectively). Logistic regression analysis showed that the risk of GC was significantly higher in rs10434 -G allele carriers than that in AA genotype carriers (AG + GG and AA), and the adjusted OR (95% CI) = 1.64 (1.24–2.08). In addition, we found a significantly higher GC risk in subjects with rs833061-T allele than those with CC allele (CT + TT and CC), adjusted or (95% CI) = 1.43 (1.10–1.87) (Table 3).

The GMDR model was used to evaluate the synergy effect between four VEGF gene SNPs and current smoking on the susceptibility to GC (Table 4). We found a statistically significant two- locus model (p = 0.018), including rs10434 and current smoking, indicating a significant interaction between rs10434 and current smoking on the risk of GC. The cross- validation consistency of the two- locus model was 10/10, the test accuracy was 0.632, and p- value was 0.018. To obtain the odds ratios and 95%CI for the joint effects of rs10434 polymorphism and current smoking on GC susceptibility, we conducted hierarchical analysis for interaction between rs10434 and current smoking on GC risk by using logistic regression. We found that current smokers with AG or GG genotype have the highest GC risk, compared to never- smokers with AA genotype, OR (95%CI) = 2.43 (1.64–3.28), after adjustment for gender, age, alcohol consumption status, and BMI (Table 5).

In our study, we investigated the relationship of four SNPs of VEGF gene with GC risk. The results indicated that the risk of GC was significantly higher in rs10434 -G allele carriers than that in AA genotype carriers. In addition, we also found a significantly higher GC risk in subjects with rs833061-T allele than those with CC allele. Nevertheless, after several covariates’ adjustment, we found that no statistically significant correlation existed between rs3025039, rs2010963, and GC risk. Previously, several studies [14–19] focused on the association between SNPs of VEGF gene and GC risk in different populations. According to the research of Zhou et al [14] in the Chinese Han population, VEGF rs10434 (+ 1612 G/A) gene polymorphism may be related to a higher risk of gastric cancer, and the difference of genotype distribution may be related to Lauren classification and the location of gastric cancer, but they also found that rs2010963 and rs833061 were not associated with GC risk. In a Japanese population, Tahara et al [15] suggested that the G1612A mutation in VEGF gene was statistically associated with the risk of GC, but the association between C936T polymorphism and GC susceptibility was not statistically significant. A meta-analysis by Hong et al [16] including these studies showed that the 1612a allele of VEGF gene is a recessive susceptibility site of gastric cancer, with a 60% increased risk. However, the meta-analysis results of Zhuang et al [17] showed that people carrying VEGF -634 G allele may increase the risk of gastric cancer, while VEGF +1612 G/A G allele may reduce the risk of gastric cancer. No association was found between + 936C/T and - 460C/T polymorphisms and susceptibility to gastric cancer. Liu et al. [18] conducted meta-analysis on Asian, European, and American populations respectively, and found that in the overall population, VEGF-634 G> C GG genotype was associated with the reduction of gastric cancer risk, VEGF- 634 G> C- C allele and GG genotype were associated with the risk in Caucasian people, while in the Asian population, VEGF + 1612G/A locus was significantly associated with the risk of GC. In our study, we also verified a significant relationship of rs10434 and rs833061 with GC susceptibility in the Chinese population. The mechanisms underlying the association between the VEGF gene and gastric cancer are not exhaustive. Previous studies have suggested that VEGF is one of the most important growth factors in the process of lymphatic angiogenesis and can play an important role in the progression of lymphangiogenesis in several tumors by activating the VEGF receptor. In the stomach, enhanced VEGF gene expression has been identified as contributing to the healing of GI lesions [20, 21]. In addition, gastric adenocarcinomas frequently show high levels of VEGF expression accompanied by an increase in intratumoral microvessel density. Higher levels of VEGF expression can also be observed in pre-cancerous gastric lesions, such as precancerous lesions, chronic atrophic gastritis, and intestinal chemosis [22].

GC susceptibility was influenced by many risk factors, including environmental and genetic factors, and the synergistic effect between gene and lifestyle factors [23]. Previously, the association between smoking habit and GC risk has been reported [24–27], and to date, no study focused on the effect of VEGF gene- smoking interaction on GC risk. In the current study, GMDR software was employed to investigate the interaction among four SNPs of VEGF gene and current smoking. We found a significant interaction between rs10434 and current smoking on the risk of GC; current smokers with AG or GG genotype have the highest GC risk, compared to never- smokers with AA genotype. The biological mechanism of the interaction between VEGF gene and smoking on the risk of gastric cancer is not very clear. Previous studies [27, 28] have confirmed that VEGF gene and smoking are related to the inflammatory response in the human body. Perhaps there is the same biological mechanism between them, resulting in a significant change to susceptibility of gastric cancer when they exist at the same time.

Some limitations existed in this study and should be explained. Firstly, there are too many SNPs in VEGF gene. In this study, only four SNPs are selected. These four SNPs could not represent all VEGF genes, but these SNPs are often mentioned in previous studies, and the results are different. Secondly, in this study, all environmental factors listed in Table 2 were included in GMDR software, but just the significant results were listed in Table 4; in the future, more lifestyle factors should be included in the interaction analysis. Lastly, we did not collect information about Helicobacter pylori infection, so the bias due to H. pylori infection may exist in this study.

In conclusion, our research shows rs10434 -G and rs833061-T alleles, gene- environment interaction between rs10434, and current smoking were all related to increased GC risk. In the future, more functional experiments are needed to verify the results obtained in our study.