Association between Paraoxonase 1 polymorphisms and risk of esophagogastric junction adenocarcinoma: a case-control study involving 2,740 subjects

Esophagogastric junction adenocarcinoma (EGJA) is a serious public health problem with high mortality in China. In this study, we assessed the association between Paraoxonase 1 (PON-1) rs662 C>T, rs854560 A>T polymorphisms and EGJA risk. This case-control study enrolled 2,740 participants of Asians origin from the Eastern Chinese Han populations. SNPscanTM genotyping assay was harnessed to determine the genotyping of PON1 polymorphisms. The PON-1 rs854560 A>T and rs662 C>T genotypes distribution accorded with Hardy–Weinberg equilibrium. We found that there was no difference in the frequency of PON-1 rs662 C>T, rs854560 A>T genotypes between the overall EGJA cases and controls. In the subgroup analyses, the results indicated that PON-1 rs662 C>T polymorphism might be associated with a significantly decreased risk of EGJA in ever smoking group (TT vs. CC/CT: adjusted OR = 0.58, 95% CI 0.35–0.95, P = 0.029). In conclusion, our study highlights PON-1 rs662 C>T polymorphism may decrease the risk of EGJA, which interacts with the tobacco using. In the future, a fine-mapping case-control study with detailed gene-environmental data is needed to further assess these potential relationship.


INTRODUCTION
Esophagogastric junction adenocarcinoma (EGJA) is a common malignancy in North America, Europe and Eastern Asian and is considered that it is different from distal GC [1][2][3]. In the United States, the data of cancer registry program showed the incidence of EGJA increased 2.5-fold from 1973-1992, and it was stable in the last two decades with 2 per 100,000 morbidity [4]. In Caucasians, the 5-year relative survival rate was 16-17% [4,5]. It is Research Paper also reported that EGJA is a serious public health problem with high mortality in China. Liu et al. reported that the incidence of EGJA was 34.1% in all gastric and esophageal adenocarcinomas patients [6]. And 5-year survival rate of EGJA was [29][30][31][32][33][34][35].5% [7,8]. The pathological process of EGJA is very complex. The etiology of EGJA is unclear. It is believed that food preserved by salting, chronic gastroesophageal reflux disease, smoking, obesity and etc. may contribute to the development of EGJA. Nowadays, accumulating evidences highlighted that individual's genetic background could play a vital role in the carcinogenesis of EGJA. Single nucleotide polymorphisms (SNPs), which could alter the activity of some detoxifying carcinogenic substances, may conduce to the development of EGJA.
Chronic inflammation may influence the susceptibility of malignancy. The potential molecular mechanisms underlying the relationship have been studied and it is identified that a number of inflammation-related cells gathers and secretes some chemical mediators, in particular reactive oxygen species (ROS) [9]. ROS levels within cells and inflammatory tissues are regulated by many free-radical scavenging systems. The excessive ROS damage intracellular macromolecules, including proteins and nucleic acids. Paraoxonase 1 (PON1) gene maps to the long arm of chromosome 7. PON1, an antioxidant enzyme, keeps the balance of antioxidant-oxidant [10][11][12][13]. Enzyme Commission of the International Union of Biochemistry and Molecular Biology classifies PON1 as an aryldialkylphosphatase (EC 3.1.8.1) [14]. PON1, a glycoprotein, is a molecular mass of 43KDa. Oxidative stress have been found to be correlated with an increased susceptibility of many malignancies [9]. In vivo, body possesses several free-radical scavenging systems including paraoxonase. PON1 prevents the oxidation of LDL and cell membrane, and therefore it is thought to be atheroprotective. Furthermore, PON1 was found to play an important role in the scavenging of carcinogenic lipidsoluble radicals [15]. In addition, PON1 is versatile and it may contribute to innate immunity, putative new roles in malignancy and the promotion of healthy aging [16]. It was reported that expression or activity of PON1 decreased in lung cancer, multiple myeloma and papillary thyroid cancer [17][18][19]. Therefore, it is reasonable to believe that the decreased activity of PON1 may be associate with the development of cancer.
PON1 rs662 C>T and rs854560 A>T, two functional SNPs, were identified to be associated with the risk of multiple cancers [20][21][22]. However, most of the epidemiologic and molecular studies focused on the relationship of PON1 polymorphisms with the risk of cancer in Caucasians. Thus, the results might remain inconclusive, especially in Asians. Therefore, we conducted this casecontrol study to determine the association between PON1 rs662 C>T, rs854560 A>T polymorphisms and EGJA risk with a large sample size in Eastern Chinese Han populations.

Baseline characteristics
The relevant demographics and risk factors are summarized by case/control status in Table 1. EGJA patients and non-cancer controls were similar in terms of age and sex distributions. There were more smokers and drinkers among EGJA patients than among controls. The minor allele frequency (MAF) of PON-1 rs854560 A>T and rs662 C>T polymorphisms in controls was similar to its data in the database ( Table 2). In controls, as demonstrated in Table 2, the PON-1 rs854560 A>T and rs662 C>T genotypes distribution accorded with Hardy-Weinberg equilibrium (HWE).
The frequencies of PON-1 rs662 CC, CT, and TT genotypes were 39.19%, 48.13% and 12.68% in EGJA group and 41.28%, 46.36%, and 12.37% in controls, respectively. When the frequency of rs662 CC genotype was used as reference, there was no difference in the frequency of PON-1 rs662 CT genotype between the EGJA group and the controls (crude OR = 1.04, 95% CI: 0.88-1.23, P = 0.624). When the frequency of PON-1 rs662 CC genotype was used as reference, we found there was no difference in the frequency of PON-1 rs662 TT genotype between EGJA group and the controls (crude OR = 1.03, 95% CI: 0.80-1.32, P = 0.821). In addition, when the frequency of PON-1 rs662 CC genotype was used as reference, there was no difference in the frequency of PON-1 rs662 CT/TT genotypes between EGJA group and the controls (crude OR = 1.09, 95% CI: 0.93-1.28, P = 0.282). When the frequency of PON-1 rs662 CC/CT genotypes were used as reference, no difference was found in the frequency of PON-1 rs662 TT genotype between EGJA group and the controls (crude OR = 1.03, 95% CI: 0.82-1.30, P = 0.809). Adjustment for age, sex, smoking and drinking, the similar results were also found (CT vs.  Table 4).

Association of PON-1 rs662 C>T and rs854560 A>T polymorphisms with EGJA in Different Stratification Groups
The PON-1 rs854560 A>T genotype frequencies in the different stratified analyses are summarized in Table  5. We found that PON-1 rs854560 A>T polymorphism was not associated with the risk of EGJA in any subgroup ( Table 5).
The PON-1 rs662 C>T genotype frequencies in the different stratified analyses are summarized in Table  6. After adjustment by logistic regression analysis, the results indicated that PON-1 rs662 C>T polymorphism might be associated with a significantly decreased risk of EGJA in ever smoking group [TT vs. CC/CT: adjusted OR = 0.58, 95% CI 0.35-0.95, P = 0.029 (Table 6)]. In other subgroups, we did not find any association between PON-1 rs662 C>T polymorphism and the risk of EGJA (Table 6).

DISCUSSION
EGJA is thought to be an independent malignancy entirety of upper digestive tract tumors [23]. It is reported that the incidence of EGJA is increasing worldwide [1][2][3]24]. A number of previous studies indicated that dietary habits, lifestyle (e.g. smoking and drinking et al.), oxidative and carbonyl stresses, and estrogens might play important roles in carcinogenesis [25][26][27][28][29][30]. In vivo, there are many antioxidant enzyme which may prevent genotoxic damage. PON1, an antioxidant enzyme, may play a vital role in keeping the balance of antioxidantoxidant balance [11,31]. Several studies reported that expression or activity of PON1 decreased in several cancers [17][18][19]. Considering the potential role of PON1 in carcinogenesis, we selected two most common PON1 polymorphisms (rs662 C>T and rs854560 A>T) and assessed their susceptibility to EGJA. In this study, we identified that PON1 rs662 C>T polymorphism was associated with the decrease the risk of EGJA in ever smoking subgroup.
It was found that expression or activity of PON1 was lower in cancer patients than controls [17][18][19]. Delimaris et al. reported that oxidative stress might contribute to pathogenesis of cancer involving the proliferation and malignancy conversion [32]. Previous   [34]. Recently, the association between the decreased risk of cancer and PON1 rs662 C>T polymorphism was also found in Asians [35]. Our findings were similar to the results of these studies. However, there were only a few casecontrol studies with small sample size conducted in Asians. The evidence might be limited. In the future, more case-control studies focusing on the relationship of PON1 rs662 C>T polymorphism with cancer risk should be performed to confirm these potential associations.
There are some limitations which may be interpreted. Firstly, although the number of the participants was relatively large, when the stratification analyses were carried out for the age, sex, smoking and drinking status, resulting in insufficient capacity which limited the power of this study. Secondly, the enrolled participants were from local hospitals, which might lead to the bias. Thirdly, there were only two missense SNPs in PON1 gene included in this case-control study. In the future, a fine-mapping study should be performed to further determine the potential association between the functional SNPs in PON1 gene and EGJA risk. Fourthly, only the information of drinking and smoking was collected as major risk factor. Other potential risk factors [e.g. gastroesophageal reflux disease, obesity, Helicobacter pylori infection status and dietary behavior (low intake of fruit and veggies, pickled food consumption, and drinking beverages at high temperatures etc.)] were not considered. Fifthly, biomarkers for oxidative stress were not measured in our study. Finally, for lack of raw data from other lifestyles, we did not further assess the relationships for the potential interactions of gene-gene or gene-environment factors.
In conclusion, our study highlights PON1 rs662 C>T polymorphism may be correlated with the decreased risk of EGJA which interacted with the tobacco using. In the future, a fine-mapping case-control study are needed to further assess the potential relationship between PON1 SNPs and EGJA risk.

Subjects
This case-control study enrolled 2,740 participants of Asians origin from the Chinese Han population. Cases (n = 1,063) were newly diagnosed EGJA patients at Fujian Medical University Union Hospital and Fujian Medical University Cancer Hospital from January 2014 to May 2016, and at Affiliated People's Hospital of Jiangsu University from January 2008 to November 2016. EGJA patients were included consecutively. Two experienced pathologists confirmed the diagnosis of EGJA for all cases. The major selection criterion for EGJA cases were: (a) all EGJA cases were Siewert II subtype; (b) Patients were Eastern Chinese Han population, and (c) EGJA was confirmed via histopathological examinations. The major exclusion criteria for EGJA cases were: (1) had a history of personal autoimmune disease, (2) EGJA cases who received prior chemoradiotherapy and (3) had a history of another malignancy. At the same time, non-cancer controls were recruited from the Physical Examination Center of these local hospitals. The control subjects had no history of autoimmune disorder or personal malignancy, and were frequency matched to EGJA patients by sex and age. These subjects have been reported in our previous study [36].
Two trained personnels interviewed each participant. The demographic and lifestyle characteristics were obtained by using a questionnaire. Information on smoking, drinking, age and sex was collected for the present study. All participants were informed and signed written consent to allow their blood samples to be genetically tested. Approval was given by the ethical committees of Jiangsu University and Fujian Medical University, in accordance with the Declaration of Helsinki.

DNA extraction and genotyping
DNA from the participants was extracted from (EDTA)-anticoagulated blood samples with Promega DNA Kit (Promega, Madison, USA). The obtained DNA sample was stored at −80°C. Genotyping was carried out by using SNPscan ™ assay (Genesky Biotechologies Inc., Shanghai, China) to determine the genotypes of PON-1 rs854560 A>T and rs662 C>T polymorphisms. For quality control, 110 DNA samples randomly selected from 2,740 specimens were reanalyzed. The genotypes of PON-1 rs854560 A>T and rs662 C>T polymorphisms were confirmed by another laboratory technicians. As shown in Table 2, the success rate of PON-1 genotyping was both more than 99%.