Association between BTLA polymorphisms and susceptibility to esophageal squamous cell carcinoma in the Chinese population

Abstract Background Growing evidence suggested that B‐ and T‐lymphocyte attenuator (BTLA) polymorphisms raised the susceptibility to a wide range of cancers. This study aimed to evaluate whether BTLA variants were related to the risk of esophageal squamous cell carcinoma (ESCC). Methods A total of 721 ESCC patients and 1208 matched non‐cancer controls were included in this research, and four tagging BTLA polymorphisms (rs2171513 G > A, rs3112270 A > G, rs1982809 G > A, and rs16859629 T > C) were selected and genotyped using SNPscan™ Assays. Results In the present study, no significant relationship between BTLA polymorphisms and ESCC was observed. However, stratified analyses suggested that the variant of BTLA rs3112270 A > G reduced the risk of ESCC in the male subgroup (AG vs AA: adjusted OR = 0.78, 95% CI = 0.61‐0.99, P = .042), BMI < 24 kg/m2 subgroup (AG vs AA: adjusted OR = 0.72, 95% CI = 0.55‐0.93, P = .012; AG/GG vs AA: adjusted OR = 0.77, 95% CI = 0.60‐0.98, P = .032), and ever drinking subgroup (AG vs AA: adjusted OR = 0.61, 95% CI = 0.38‐0.97, P = .037). But when stratified by BMI ≥ 24 kg/m2, the rs3112270 A > G polymorphism increased the susceptibility to ESCC (GG vs AA: adjusted OR = 1.91, 95% CI = 1.02‐3.59, P = .045). Besides, we demonstrated that BTLA rs2171513 G > A polymorphism was protective of ESCC in the ever drinking subgroup (GA/AA vs GG: adjusted OR = 0.62, 95% CI = 0.39‐0.97, P = .037). Conclusion Taken together, our initial investigation postulated that the rs3112270 A > G and rs2171513 G > A variants in the BTLA gene are candidates for the risk of ESCC, which might be helpful for the early diagnosis and treatment of ESCC.


| INTRODUC TI ON
As stated by the global epidemiological data, esophageal cancer (EC) ranks the sixth primary cause of cancer-related death, with an approximated 477 900 new occurrences and 375 000 deaths per year in China. 1,2 Different from the fact that esophagogastric junction adenocarcinoma (EGJA) is the dominant subtype of EC for the western nations, in China, esophageal squamous cell carcinoma (ESCC) makes up more than 90% of the total cases. 3 And despite rapid progress in surgical technique and adjuvant treatment, the prognosis for patients with ESCC is extremely poor, with a 5-year overall survival rate <30%. 4 Thus, it is essential to explore new risk factors for further understanding the potential mechanism of ESCC progression.
Nowadays, the immune system plays an increasingly important role in anti-tumor therapy. 5 Cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) and programmed cell death 1 (PD-1) are the prominent representative of this field. Similar to CTLA-4 and PD-1, as a co-inhibitory regulator of the immune system, BTLA contains an extracellular domain, a transmembrane region, and a cytoplasmic region. 6 When combined with its ligand named herpesvirus entry mediator (HVEM), 7 tyrosine phosphorylation of the cytoplasmic region in BTLA gene can suppress T-cell activation by recruiting Src homology phosphatase-1 and Src homology phosphatase-2, 8 which could significantly inhibit the secretion of IL-1, IFN-γ, and IL-10. 9 And the role of BTLA-HVEM pathway has also been identified in BTLA-deficient mouse models, 10 where the absence of BTLA gene could enhance sensitivity to antigen-specific immune response and therefore develop autoimmune diseases, as well as HVEM-deficient mice, 11 which, on another level, showed the negative effect of BTLA-HVEM pathway on the immune microenvironment.
In recent years, accumulating studies have focused on the genetic polymorphisms of immune molecules with susceptibility to the various tumors, including BTLA. [12][13][14][15] Fu et al 12

genotyped five
SNPs and found that BTLA rs1844089, rs2705535, and rs2633562 polymorphisms were associated with the pathological features of breast cancer. Partyka et al 13 chose seven variants and revealed the rs1982809G allele contributed to a higher-grade stage of renal cell carcinoma. Recently, there was a basic study demonstrating that the change from T to C in the BTLA rs1982809 variant could interfere with the activity of BTLA 3'UTR and regulate BTLA expression in peripheral blood T lymphocytes, which might be considered as a potential biomarker in predicting the process of sepsis and multiple organ dysfunction syndrome. 16 In addition, Karabon et al 14 enrolled a total of ten polymorphisms and demonstrated that the presence of BTLA rs1982809 polymorphism was related to a lower level of BTLA mRNA, and the variant might be deemed as a low-risk factor for the development of chronic lymphocytic leukemia. Subsequently, Tang et al 15 reported that the BTLA rs1982809 SNP was found to be conferred to an increased risk of EGJA in smoking patients.
Nevertheless, whether the variation in the BTLA gene associates with ESCC risk remains unknown. Concerning the tremendous value of co-signaling molecules in anti-tumor therapy, and to better understand this issue, we conducted this case-control study to clarify the detailed relationship of four tagging BTLA polymorphisms with the risk of ESCC in the eastern Chinese Han population.

| Ethics statement
All procedures of this research were administered in line with the Declaration of Helsinki and approved by the Institutional Review Board of Jiangsu University (NO. K-20160036-Y). Each participant provided the written informed consent to this study and was willing to donate 2 mL of peripheral blood. and alcohol consumption, were obtained by questionnaires and patients' clinical records. We defined the "ever drinkers" as the subjects with drinking no <3 times a week for longer than half a year, and the individuals who smoked at least one cigarette per day over 1 year were deemed as "ever smokers." Besides, we used the body mass index (BMI) value of 24 kg/m 2 as a threshold for distinguishing individuals at obesity. 17

| DNA genotyping
The whole blood sample was stored in an anti-coagulated tube that contained EDTA. We extracted the genomic DNA by using a Promega DNA Mini Kit (Promega) under the instruction of the manufacturer's procedure, 18 and then, the four SNPs were genotyped using the SNPscan™ Assays (Genesky Biotechnologies Inc). 19 For qualitative tests, 4% of the total DNA samples were selected at random and genotyped again by different laboratory staff, and the final results of the four BTLA genotypes were in concord with the primary findings.

| Statistical analysis
In this study, all data analyses were conducted with software SAS version 9.4 (SAS Institute). The value of continuous variable, including age, was reported as means ± standard deviation (SD) and evaluated by Student's t test.
The comparison of categorical variables between ESCC cases and controls, such as BTLA genotypes, was conducted with the chi-square test. The deviation of HWE for each SNP distribution in the controls was assessed via the online software (http://ihg.gsf.de/cgi-bin/hw/hwa1.pl). After adjusting for age, gender, smoking status, and alcohol consumption, the potential associations between BTLA variants and the risk of ESCC were examined by the multivariate logistic regression analyses and described by calculating the adjusted odds ratio (OR) with 95% confidence intervals (CIs). A twosided P value < .05 was deemed as statistically significant.

| Basic characteristics
Basic information regarding BTLA polymorphisms is revealed in Table 1. Results showed that the MAF of each BTLA SNP was in accord with the database of the Chinese population. In the control group, frequencies of the four BTLA genotypes were all reached HWE (all P > .05), and the failed genotype data for each polymorphism were <1%. Table 2 summarizes the basic features of 721 ESCC cases and 1208 controls, and the mean age of the case and control groups was 62.59 ± 8.18 and 62.92 ± 8.94 years, respectively.
The ESCC group composed of 551 males (76.42%) and 170 females (23.58%), and there involved 899 males (74.42%) and 309 females (25.58%) among the controls. There was no difference in age and sex between the study groups (both P > .05), meaning that the two above factors were well matched. However, compared with the controls, the degree of BMI and the proportion of drinking and smoking were significantly higher in those of ESCC group (all P < .05).

| BTLA polymorphisms and ESCC risk in the overall population
The detailed frequencies of BTLA genotypes and the results about the association between each selected polymorphism with the risk of ESCC are presented in Table 3. We found that BTLA rs2171513 G > A,

| BTLA polymorphisms and ESCC risk in stratification groups
Furthermore, we conducted a stratified analysis mainly relied on the enrolled parameters, including age, sex, BMI, smoking status, and alcohol consumption. As presented in Table 4, when stratified by alcoholic use, in the ever drinking subgroup, we found that the rs2171513 G > A variant in BTLA gene might be a protective variable against the progression of ESCC (GA/AA vs GG: adjusted OR = 0.62, 95% CI = 0.39-0.97, P = .037).
As exhibited in Table 5

| CON CLUS ION
Despite these limitations, our preliminary findings suggest that the two tagging variants of rs3112270 A > G and rs2171513 G > A in the BTLA gene might contribute to the progression of ESCC in the eastern Chinese population, which is the first study for the involvement of the co-inhibitory BTLA SNPs in ESCC to our knowledge. But future intensive studies with larger samples are worth to elucidate these works as well as the underlying molecular function of BTLA polymorphisms.

ACK N OWLED G M ENTS
We appreciate all subjects who participated in this study. We wish to thank

CO N FLI C T O F I NTE R E S T
The author reports no potential financial conflicts of interest in this work.