Association between HLA-DQ Gene Polymorphisms and HBV-Related Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is one of the most common causes of cancer-related mortality worldwide. Host gene variants may influence hepatitis B virus- (HBV-) related HCC. Human leukocyte antigens (HLA) play an important role in presenting virus antigens to immune cells that are responsible for the clearance of virus-infected cells and tumor cells. Previous studies have investigated the HLA-DQ (rs2856718 and rs9275572) polymorphisms that may be associated with the development of HBV-related HCC. However, the results are controversial or inconclusive. Hence, we conducted a meta-analysis to derive a more precise estimation of the associations. A total of 6 articles were used to evaluate the effect of the two polymorphisms on the risk of HBV-related HCC. Odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated. We found that rs2856718 and rs9275572 in HLA-DQ significantly decreased HBV-related HCC in total population, especially in Chinese, but not in Saudi Arabian. Further validation of our results in larger populations and different ethnicities are required.


Introduction
Primary liver cancer is one of the six most common cancers and the second largest cause of cancer deaths worldwide (746,000 cases or 9.1% of all cancer deaths), especially in undeveloped countries [1,2]. Approximately 75% of liver cancers occur in Asia, whereas China alone accounts for more than 50% of all cases [3]. Globally, the vast majority of histologic types of primary liver cancers (approximately 80%) are hepatocellular carcinoma (HCC), a malignant tumor arising from hepatocytes, the liver's parenchymal cells [4]. It is estimated that 75%-85% of HCC patients are attributable to chronic infections with hepatitis B virus (HBV), especially in Asian populations and particularly in Chinese [5]. Besides HBV infection, other extrinsic factors, such as alcohol, smoking, physical inactivity, chemical exposure, and poor dietary habits, are also involved in developing HCC [6]. However, only a small fraction of infected patients can progress to HCC during their lifetime. So intrinsic factors, such as genetic mutations, may be vital for tumor development [7,8]. Furthermore, genetic epidemiology points out that genetic polymorphisms of genes involving in different processes of carcinogenesis may also play an important role to determine individual susceptibility to HCC and improve the prevention and treatment of this cancer [9][10][11].
Human leukocyte antigen (HLA) has been identified to be associated with regulating the immune response to HBV infection and clinical outcomes [12]. HLA-DQs are highly polymorphic especially in exon 2, which codes for antigenbinding sites. The single-nucleotide polymorphism (SNP) rs2856718 locates in the intergenic region between HLA-DQA2 and HLA-DQB1. Hu et al. 's study showed that HLA-DQ rs2856718 significantly decreased the host HCC risk [13]. The SNP rs9275572 locates between HLA-DQA and HLA-DQB on 6p21.32. A recent genome-wide association study (GWAS) indicated that the HLA-DQ rs9275572 polymorphism was significantly associated with HCVrelated HCC in Japanese population [14]. Chen et al.'s study found that HLA-DQ rs9275572 polymorphism may have a protective impact on HBV-related HCC [15].
Previous studies have assessed the association between HLA-DQ (rs2856718 and rs9275572) polymorphisms and HBV-related HCC susceptibility, but the results of previous studies are inconsistent and inconclusive [13,[15][16][17][18][19]. Therefore, we performed a comprehensive meta-analysis to derive a more precise estimation of the relationship between HLA-DQ (rs2856718 and rs9275572) polymorphisms and HBV-related HCC risk. To the best of our knowledge, this is the first meta-analysis to analyze the association of the HLA-DQ (rs2856718 and rs9275572) polymorphisms with HBV-related HCC risk.

Search Strategy.
To identify relevant studies, we systematically searched PubMed, EMBASE, Google Scholar and China National Knowledge Infrastructure (CNKI) databases. The search strategy was based on a combination of "HLA-DQ," "hepatitis B virus," or "HBV"; "Hepatocellular carcinoma," "HCC," or "liver cancer"; "polymorphism" or "SNP"; and "rs2856718" or "rs9275572" (up to March 27, 2017). The languages of the reviewed articles were limited to English and Chinese. In addition, references of retrieved articles were also screened.

Inclusion and Exclusion
Criteria. The following criteria were necessary for inclusion in the meta-analysis: (1) a case-control study that had investigated the genetic risk of HBV-related HCC in relation to HLA-DQ rs2856718 or rs9275572, (2) original papers containing independent data, (3) the study that provided the available genotype frequencies, (4) the study that provided sufficient information for calculating the pooled odds ratios (ORs) with 95% confidence intervals (CIs), and (5) the genotype distribution of a control group that are consistent with the Hardy-Weinberg equilibrium (HWE).
Exclusion criteria were as follows: (1) case-only studies, (2) review articles, (3) repetitive reports, and (4) lack of genotype frequency data. In addition, if multiple studies had overlapping data, only the most recent version was used.

Data Extraction.
The following data were independently extracted from each study by two authors (Jingzhu Lv and Tao Xu): the first author's name, year of publication, country, genotyping method, number of cases and controls, genotype, and allele frequency. After extraction, data were reviewed and compared by the two independent investigators. Any disagreements were resolved by discussion with the third investigator.

Statistical Analysis.
All statistical analyses were performed using Stata software version 12.0 (Stata Corporation, College Station, TX). The Hardy-Weinberg equilibrium test in the control group was undertaken using the χ 2 -test (P < 0 05 was considered as significant disequilibrium in the control group). Odds ratios (ORs) with 95% confidence intervals (95% CIs) were used to estimate the strength of the association between HLA-DQ (rs2856718 and rs9275572) polymorphisms and HBV-related HCC. The pooled ORs were obtained from combination of single studies by homozygote comparison, heterozygote comparison, dominant and recessive models, and allele comparison, respectively. The significance of pooled ORs was assessed by the Z test, and P Z < 0 05 was considered as the significance threshold. Based on the heterogeneity test, the pooled OR was calculated using the fixed (P H ≥ 0 05, I 2 ≤ 50%) or   random (P H < 0 05, I 2 > 50%) effect models. Sensitivity analysis was carried out to identify the effect of data from each study on the pooled ORs. Finally, publication bias was assessed using Egger's test, and P E < 0 05 was considered statistically significant.

Characteristics of the Included Studies.
A total of 47 potentially relevant articles published up to March 27, 2017 were systematically identified in PubMed, EMBASE, Google Scholar, and CNKI. The flow chart that summarized the literature review process and the specific reasons for any exclusion from the meta-analysis are shown in Figure 1. In the end, 6 articles that met the inclusion criteria were included in the meta-analysis of 1 study for HLA-DQ rs2856718, 2 studies for HLA-DQ rs9275572, and 3 studies for HLA-DQ rs2856718 and HLA-DQ rs9275572 [13,[15][16][17][18][19]. Characteristics of all eligible articles are shown in Table 1. Of the 6 articles, 4 articles including 1820 cases, 2266 CHB, and 2601 controls evaluated the association between HLA-DQ rs2856718 and HBV-related HCC risk, while 5 studies evaluated the association between the HLA-DQ rs9275572 and HBV-related HCC risk (1092 cases, 1920 CHB, and 1646 controls).

3.2.
Meta-Analysis of the Association between HLA-DQ rs2856718 and HBV-Related HCC Risk. The potential association of the HLA-DQ rs2856718 and HBV-related HCC was investigated in four articles; among which, three articles were in Chinese population and one in Saudi Arabian population. The meta-analysis of a possible association between the HLA-DQ rs2856718 and risk of HBV-related HCC is summarized in  (Figure 2(a)). Additionally, this risk was more significant in the Chinese population, but there was no association between the HLA-DQ rs2856718 and HBV-   related HCC susceptibility in Saudi Arabian population. When the controls were CHB patients, only GG versus AA and G versus A of HLA-DQ rs2856718 were significantly associated with the risk of HBV-related HCC. However, subgroup analysis by ethnicity showed that HLA-DQ rs2856718 AG genotype had a significantly increased risk of HBVrelated HCC among Saudi Arabian population (AG versus AA: OR = 2.00, 95% CI: 1.12-3.58, P Z = 0 019) (Figure 2(b)).

Sensitivity Analysis.
The sensitivity analysis was performed to assess the influence of an individual study on the overall OR, and the corresponding pooled ORs were not materially altered (Figures 4 and 5).

Publication Bias.
Publication bias of the included articles was assessed using Egger's test. The results of Egger's test indicated that no evidence of publication bias was observed in HLA-DQ (rs2856718 and rs9275572) polymorphisms (Tables 2 and 3).

Discussion
HLA-DQs belong to HLA class II molecules, which are expressed as cell-surface glycoproteins that present viral  peptides to CD4 + T cells resulting in generating immunity against infection [20]. A study showed that the CD4 + T cells were significantly increased in tumor, ascites, and peripheral blood of patients with HCC and showed that HBV-specific and HLA class II-restricted CD4 + T cell responses may be related to HBV-related HCC development [21]. Recently, several studies have shown that genetic variations in HLA genes influence disease progression in HBV infection [22][23][24][25]. Accumulating evidence indicated the associations between HLA-DQ (rs2856718 and rs9275572) polymorphisms and HBV-related HCC, but the results are inconclusive or inconsistent. Hu   the HBV-related HCC [17]. Meanwhile, a study observed that there was no association between HLA-DQ (rs2856718 and rs9275572) polymorphisms and HBV-related HCC in Saudi Arabian patients [16].
In order to resolve this conflict, we conducted a metaanalysis on the association between HLA-DQ (rs2856718 and rs9275572) polymorphisms and HBV-related HCC. Meta-analysis has been recognized as an important tool to more precisely define the effect of selected genetic polymorphisms on the risk for disease and to identify potentially important sources of between-study heterogeneity. There are variations between human populations, so a SNP allele that is common in one geographical or ethnic group may be much rarer in another. So we also analyzed the distribution of HLA-DQ (rs2856718 and rs9275572) polymorphisms in different ethnic groups. Our results indicated that HLA-DQ (rs2856718 and rs9275572) polymorphisms were associated with the decreased risk of HBV-related HCC in Chinese population, but not in Saudi Arabian population. However, the sample size in Saudi Arabian population is too small to conclude a negative association. The comparison between Chinese population and Saudi Arabian population may be imbalanced as a result of the sample size. Therefore, the findings of this study should be validated in the future through a population-based study. As a meta-analysis of observational studies, there are some limitations. Firstly, we did not have original data for all studies to adjust estimates and perform a more precise analysis, including gender, age, drinking, smoking, lifestyle, body mass index, and so on. Secondly, the number of published studies was not sufficiently large for a comprehensive analysis. Thirdly, the interaction of gene-gene and of geneenvironment has not been evaluated owing to the absence of original data. Therefore, more studies are needed to get more reliable results.
In conclusion, the current meta-analysis suggested that HLA-DQ (rs2856718 and rs9275572) polymorphisms were associated with HBV-related HCC risk among Chinese population. Taken together, our study suggested that HLA-DQ loci are candidate susceptibility regions that have some marker SNPs (rs2856718 and rs9275572) for HBV-related HCC in Chinese population.

Conflicts of Interest
The authors declare that they have no conflicts of interest.