Tumour necrosis factor-α(-308) polymorphism and the risk of gastric cancer: A meta- analysis and trial sequential analysis

Background: Numerous studies have reported that polymorphisms in the tumour necrosis factor-alpha (TNF-a-308) gene are implicated in susceptibility to gastric cancer. However, individual genetic association studies that assessed the relationship between TNF-a- 308 and the risk of gastric cancer showed inconclusive results. The objective of this study was to synthesis evidence on the association between polymorphisms in the TNF-a-308 and gastric cancer risk. Methods: This is a meta-analysis of genetic association studies. We searched relevant case-control studies, assessing TNF- 308 polymorphisms and gastric cancer in health-related electronic databases. The methodological quality of included studies was assessed by the Newcastle-Ottawa quality assessment scale. The strength of association was calculated as odds ratios (ORs) with its 95% confidence intervals (CIs). Pooled ORs and 95 % CIs were estimated using random-effect model or fixed effect model, based on between-study heterogeneity. We analysed the strength of association under dominant, recessive, additive and allele models. Multiple subgroup analyses including ethnic groups, HWE status, study quality were done for robustness of the estimates. Publication bias was detected by inspection of funnel plot asymmetry. Results: A total of 33 studies, comprising 7695 patients and 12327 controls were identified. Based on the studies that met HWE, significant association was found between this polymorphisms and gastric cancer risk under dominant model (OR 1.2, 95%CI 1.1-1.3, I 2 :37%), recessive model OR 1.27, 95%CI 1.0-1.62, I 2 :0%) and additive model (OR 1.31, 95%CI 1.08-1.32, I 2 :0%). The TSA plot indicated the analyses was with the required information size. There was no publication bias. In the subgroup analysis by ethnicity, the ethnic groups and the quality of studies had impact on the estimates. Conclusions: The findings suggest that TNF-α-308 gene polymorphism plays an important role as host genetic factor predisposing to gastric carcinogenesis, and it would be useful for a screening marker.


Background
Gastric cancer (GC) is globally the fifth most common cancer with the highest prevalence in both sexes in Asia (74.5%) and the third leading cause of cancer death In 2018 alone, the estimated number of deaths attributed to GC was 782,685. On genderspecific, GC is the fourth most common types of cancer occurred in men and the seventh most common cancer in women [1]. As such, a greater understanding of the risk factors that play a role in the early and late stages of gastric carcinogenesis is important to strengthen and improve the preventive and therapeutic interventions [2].
Epidemiologic studies showed that GC is a multifactorial in aetiology [3,4], in which dietary factors and Helicobacter pylori infection may contributed to the development of GC [3]. Of note is that a high prevalence of these risk factors do not always correspond to a high incidence of GC, suggesting that other susceptible factors such as genetic variations and environmental differences may play a role in the gastric carcinogenesis [5]. For instance, H. pylori infection is observed in more than half of the world's population, however there is only 1-2% of the infected population will develop GC in their lifetime [6].
Gastric neoplasms are composed not only with cancer cells but also other "noncancer" compartments (including immune cells), which are the major players in GC disease progression and aggressiveness [7]. Tumour necrotic factor (TNF) is a proinflammatory cytokine, which is produced mainly by the immune cells such as macrophages, dendritic cells, lymphocytes and mast cells [8].
The tumour microenvironment, which is mainly composed of inflammatory cells, is a crucial player in the neoplastic process, fostering proliferation, survival and migration [9]. Thus, TNF-α may through inflammation, act as a tumour promoter. Individual studies reported the significant relationship between TNF-α -308 (rs 1800629) and the risk of GC [10,11].
However, other studies reported differently [12,13,14]. On the whole, the objective of this study was to summarize the evidence of association between TNF-α-308 and the risk of GC.

Study search
The search of relevant studies was performed in the health-related databases of PubMed, Ovid Medline, google scholar and web of science, using the terms ("gastric cancer" OR "gastric carcinoma" OR "stomach cancer" OR gastric adenocarcinoma AND "tumour necrosis factor-alpha-308" OR "rs1800629" OR "TNF-α-308 G>A" OR "TNF-α-307 G>A"). The search strategy in PubMed database is provided (Additional File 1: Table S1).
The search was limited to the publications in English until June 2019. Moreover, we searched manually the references of included studies and relevant systematic reviews for any additional studies.

Inclusion criteria
Human studies that assessed GC, irrespective of location or histological type were included, if they (i) assessed TNF-α-308 G>A (rs 1800629) or TNF-α-307 G>A; (ii) were case-control design (retrospective or nested case-control) with an outcome of GC risk; (iii) provided the genotype frequency in cases and controls, (iv) mentioned sufficient data to compute odds ratio (OR) and its 95% confidence interval (CI) as the outcome measure. GC is as defined in the primary studies.
Studies which did not meet the inclusion criteria were excluded. Studies done on family or sibling-pairs were also excluded.

Data extraction
One investigator screened the titles and abstracts and selected the relevant full-text articles, following the inclusion criteria. Two investigators extracted the data from each study independently, by using a piloted data extraction form. Information collected included: first author, publication year, country, study setting, the number of cases/controls, ethnicity (Asian or Caucasians), method of genotyping and genotype/allele frequencies in cases/controls. If an allele frequency was zero in both case and control, we added 1 to that allele, following the Laplace approximation [15]. Any discrepancy between the two investigators were resolved by consensus.

Assessment of the methodology quality
The two investigators independently evaluated the methodological quality of studies, using the Newcastle-Ottawa Scale (NOS) [16]. The assessment is based on the three factors such as 'selection of the study groups' (4 points), 'comparability of the groups' (2 points) and 'ascertainment of the exposure' (3 points). The total score for each study spanned from 0 (the worst) to 9 (the best). We categorised study quality as good (≥7), moderate (≥5) and poor (≤4), according to scores achieved. Any discrepancy between the two investigators was resolved by consensus.

Statistical analysis
We assessed an evidence of HWE in the control populations in the included studies using the goodness-of-fit test and p>0.05 was considered to indicate consistency with HWE [17]. As described elsewhere [18], the strength of the association between TNFα-308 G>A and the risk of GC in each study was estimated using OR and its 95% CI.  Figure 1 illustrates a four-phase study selection process. The initial search yielded a total of 1322 records. After removing the duplicates and screening of abstracts, 43 full-text articles that were potentially eligible were retrieved. We included a final of 33 studies (with 7695 cases and 12327 controls) in this review [2 ,5, 10-14, 24-49]. Summary of the 10 excluded studies were provided (Additional File 2: Table S2). Table 1 shows the characteristics of the studies identified. Of 33 studies included, slightly more than half (54.5%, 18/33) were done in the Asian region. The most frequent 5 studies were conducted in China, South Korea or Brazil. Figure  Regarding the methodology, the majority (78.8%) were categorized as high (14 studies) or moderate quality (12 studies) and the remaining 7 studies were of low quality.

Study characteristics
Of total 33 studies, 8 studies (24%) were gastric adenocarcinoma stage. By anatomical locations or histological types of GC, only 5 studies (7 datasets) consistently provided data that could make pooling of analysis (2,10,13,35,45). Twelve studies (36.4%) used TaqMan method for genotyping. In total, 24 studies (72.7%) had provided information that GC cases were infected with H. pylori infection, albeit with variation in distribution. For instance, all cases (100%) were infected with H. pylori in one study [33], while this was only 46% in another study [2] ( Table 1). The remaining 9 studies did not measure H. pylori status of the participants or not reported the status explicitly.

Quantitative estimates
The genotype frequencies in individual studies are presented in Table 2 (Figure 3 A, B, C, D).

Subgroup analyses
TNF-α-308 G>A was with an increased risk of GC only with the high quality studies (Additional File 3: Figure S1) By stratification on H. pylori infection status, overall TNF-α-308 G>A had a 25% increased risk of GC in those infected cases compared to non-infected cases (OR,1.25, 95% CI; 1.08-1.45); however, this association was with a substantial heterogeneity (I 2 :69%) (Additional File 5: Figure S3).
The pooled analysis on the 27 studies that were consistent with HWE also showed that TNF-α-308 G>A polymorphism was significantly associated with an increased risk of but not in any particular ethnic group) ( Table 3). A funnel plot showed no evidence of publication bias (Additional File 6: Figure S4).

TSA plot
We performed TSA of the dominant model with the use of an overall type I error of 5% and type II error of 20%. The included total participants in this meta-analysis reached the required information size (for an expected RRR 26%). Briefly, a TSA monitoring boundary crossed with Z curve, which confirms the presence of robust evidence ( Figure   4). In such case further studies are not needed to provide sufficient information.

Discussion
The current study provides evidence on the relationship between TNF-α-308 G>A and the risk of GC, comprising 7695 cases and 12327 controls from 33 studies. The major observations are as follows;

1.
Based on 27 studies that met HWE, TNF-α-308 G>A SNP was significantly associated with the GC risk under the dominant, recessive and addictive models.

2.
On stratification, HWE status of the controls, the ethnicity or study quality had an impact on the effect estimates.

3.
The TSA plot revealed that the required information size for evidence of effect was sufficient. Any future studies in this field will less likely to change the direction of estimates.
Our findings were comparable with earlier reviews, in which the significant association was limited to the Caucasians [50,51] in the dominate models [11,51]. The association was statistically significant only for the Caucasians, indicating a dominance of racial specific factors. This difference may be explained partly due to variations in the frequency of the A allele between the different ethnic groups that could contribute to the diverse results. Moreover, it might also be related to difference in environmental factors such as smoking and diet between these two major ethnic groups. This was indirectly supported by an individual study in Poland, in which 72% of the GC cases were smokers (ex-smokers or current smokers) [39]. Due to paucity of data, we were not able to perform subgroup analysis with the smoking status of participants in the studies identified. The effects of inflammatory polymorphisms might have been masked by smoking [35]. Moreover, there was no significant differences of the gene frequencies between the anatomical sites or the histological types of GC. A silico analysis of TNF expression, using the bioinformatics web GEPLA had reported that the expression of TNF in GC tissue was higher than in normal stomach (p < 0.01) [52]. A published metaanalyses [50] focused on this polymorphism in the risk of GC reported that H. pylori infected cases had higher risk of developing GC. This was also observed in the present review.
The current findings suggest that H. pylori infection had contributed to a 30%  [53]. TNF-a has been shown to inhibit the gastric acid secretion which is important in inducing cell apoptosis and promoting epithelial cell damage [54]. The cellular immune response to H. pylori infection is a main factor which contribute to the damage in gastric mucosa. Studies reported that the hypochlorhydric milieu plays a role in promoting the growth of non-H. pylori bacteria, which may cause the damage to mucosal wall and/or the production of carcinogenic Nnitroso compounds [55]. Moreover, H. pylori infection activates the cytokines production in the lining of the stomach including inflammatory-related genes such as TNF-a in the present analysis.
Our findings of overall or sensitivity analyses had some similarities to earlier reviews, in which the significant associations were also found with the dominant, allelic, additive models and recessive models, despite variations in number of studies [11,50,51]. Although there are more recent studies in this analysis, the results, in general, retained the evidence of association. Moreover, we introduced TSA for confirmation of the estimates to assess a required information size. TSA plots indicated that there was sufficient information to provide conclusive results.

Public health Implications
The difference in association between the ethnic groups observed in the current analysis has implications. Studies had reported that the regulation of tumor immunity factors at the genetic and gene expression level may be different in the Asian and non-Asian GC populations, and this can affect the region-specific effects on therapy outcome and prognosis [56].

Limitations
We acknowledge the study limitations. Only 36% of the studies in this review used TaqMan SNP genotyping assays, which is the preferred technology due to its high throughput and is highly accurate [57] compared to other methods. Hence, accuracy of genotype frequency is of concern as some studies included in this analysis were small studies with small sample sizes. Hence, there might likely have type II statistical error.
However, meta-analysis is a retrospective pooling of published studies, and type II errors Numerous studies had highlighted the issue of deviations from HWE in genetic association studies such as genotyping error, population admixture/substructure, among others [58][59][60]. Furthermore, for robustness of the findings, we have attempted several subgroup analyses. Additionally, there are some strengths in this meta-analysis compared with published reviews in this field [11,50,51,[58][59][60]. To be comprehensive, we have attempted the TSA technique, which is useful to adjust random-error risk. An add-on TSA approach to this field will highlight to researchers about the optimal sample size to make judgement of the effect estimates. This will help the researchers and policy makers to consider on the need of future similar studies. This will further help to save the resources for more important future studies, rather than repeating a study that has been already confirmed.

Conclusions
The current findings suggest that TNF-α-308 gene polymorphism plays an important role as host genetic factor predisposing to gastric carcinogenesis, and it would be useful for a screening marker. As the relationship of GC risk is ethnic specific, the consideration as a biomarker should be tailored to specific population group. To substantiate this, studies from the Asian regions, using more reliable genotyping technique are recommended. The need for approval was waived as this study solely used published human data.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analyzed during this study are included in this article and its supplementary information files.

Competing interests
The authors declare that they have no competing interests.   Significant association is indicated in bold. Figure 1 Study selection process