Association between Polymorphisms in Inflammatory Response-Related Genes and the Susceptibility, Progression and Prognosis of the Diffuse Histological Subtype of Gastric Cancer

The chronic inflammatory microenvironment and immune cell dysfunction have been described as critical components for gastric tumor initiation and progression. The diffuse subtype is related to poor clinical outcomes, pronounced inflammation, and the worst prognosis. We investigated the association of polymorphisms in inflammatory response-related genes (COX-2, OGG1, TNFB, TNFA, HSPA1L, HSPA1B, VEGFA, IL17F, LGALS3, PHB, and TP53) with gastric cancer susceptibility, progression and prognosis in a Brazilian sample, focusing on the diffuse subtype. We also performed the analysis regarding the total sample of cases (not stratified for tumor subtypes), allowing the comparison between the findings. We further investigated the polymorphisms in linkage disequilibrium and performed haplotype association analyses. In the case-control study, rs1042522 (TP53) was associated with a stronger risk for developing gastric cancer in the sample stratified for diffuse subtype patients when compared to the risk observed for the total cases; CTC haplotype (rs699947/rs833061/rs2010963 VEGFA) was associated with risk while rs699947 was associated with protection for gastric malignancy in the total sample. Regarding the associations with the clinicopathological features of gastric cancer, for the diffuse subtype we found that rs699947 and rs833061 (VEGFA) were associated with outcomes related to a worse progression while rs5275 (COX-2), rs909253 (TNFB), and rs2227956 (HSPA1L) were associated to a better progression of the disease. In the total sample, rs699947 and rs833061 (VEGFA), rs4644 (LGALS3), and rs1042522 (TP53) were able to predict a worse progression while rs5275 (COX-2), rs2227956 (HSPA1L), and rs3025039 (VEGFA) a better progression. Besides, rs909253 (TNFB) predicted protection for the overall and disease-free survivals for gastric cancer. In conclusion, these results helped us to clarify the potential role of these polymorphisms in genes involved in the modulation of the inflammatory response in the pathogenesis of gastric cancer.

. General characteristics of the studied sample and comparison of the sociodemographic status, smoking and alcohol consumption between controls and cases (both considering the total sample and the cases stratified for the diffuse histological subtype).           Text S1: Discussion of the polymorphisms that did not present any relevant association in our study.

CONTROLS CASES
Here we discuss the selected polymorphisms included in our study that did not present any relevant association.
Among the four studied VEGFA polymorphisms, we did not find any association regarding the rs2010963 polymorphism. A previous study in breast cancer showed that this SNP was associated with several factors related to a worse progression and higher aggressiveness of the disease (increased susceptibility risk, higher VEGFA mRNA levels, tumor with bigger sizes, presence of perineural invasion, higher staging and shorter disease-free survival) [1].
Another cytokine that was selected for investigation in our study was IL17F. The rs763780 (IL17F) polymorphism was associated with gastric cancer in the Allele Model, but this significance was lost in the multivariate analysis. IL17F is part of a gene family with important involvement in tissue inflammation by inducing the expression of several other cytokines and chemokines. This polymorphism leads to a His to Arg substitution at amino acid 161 and in vitro analysis showed that the polymorphic variant loses the ability to activate the production of the mitogen-activated protein kinase pathway and certain cytokines and chemokines [2]. Our in silico analysis showed that this amino acid change is tolerated/benign for the final coded product. Although this SNP has been previously associated with risk and progression of gastric cancer [3], we did not find associations to any clinicopathological variable or prognosis in our study. The frequency of the polymorphic allele was too low (8.4% and 4.8% in cases and controls, respectively) and it also presented deviation from HWE in our sample. Therefore, this result should be reanalyzed in an independent and increased set of sample. We cannot exclude the hypothesis that this SNP has an impact on gastric cancer risk and progression because its function might be compensated by other redundant molecules inside the same pathway.
Regarding the rs1061581 (HSPA1B) polymorphism, we did not find any association neither with susceptibility, progression nor prognosis in our sample. It causes a silent substitution and has been described as able to regulate the protein expression interfering with its secondary structure and mRNA stability, affecting its anti-apoptotic effect and its function as a modulator of the immune system [4,5].
OGG1 is a DNA glycosylase that belongs to the BER (Base Excision Repair) pathway, which repairs mainly endogenous/oxidative lesions as result of the cellular metabolism [6]. We hypothesized that maybe functional polymorphisms in repair genes could influence in the capacity of the organism repair DNA damage caused by diet carcinogens, oxidative stress or inflammation induced by H. pylori, in the gastric mucosa. Also, OGG1 has been shown to be important as a modulator of the immune and inflammatory systems [7]. We selected the rs1052133 polymorphism for investigation, which is located in the exon 7 and results in a change from Ser to Cis in 326 position of the protein. The in silico analysis by Polyphen and SIFT softwares showed that this change is tolerated/benign. However, Cis variant has demonstrated to increase the genetic instability and decrease the repair rate of 8-oxoguanine in vivo [8]. Nevertheless, no association was found with this polymorphism in the present study.
Although some studies have described a functional role for the rs6917 polymorphism, located in the 3' UTR of the PHB gene [9] and that it has been associated with an increased risk for development of some types of tumors [10,11], in the present study we did not find association with gastric cancer susceptibility. No association was found regarding progression and prognosis as well. Our group has been studying the role of PHB and we have observed that the TT genotype increased the risk for melanoma in the presence of specific host risk factor [12].
Another study from our group demonstrated a possible role for this polymorphism in the transcriptional regulation of PHB in gastric cancer once T allele was associated with reduced PHB expression levels [13]. Therefore, functional studies on prohibitin polymorphism are necessary to elucidate its functional role.