Association of Genetic Polymorphisms in Tumour Necrosis Factor-alpha (TNF-α-308G/A) and Interleukin-6 (IL-6-174G/C and IL-6-634C/G) with lung cancer risk: a meta-analysis

Background: Several studies have indicated an association between tumor necrosis factor-alpha (TNF-α) or interleukin(IL)-6 gene polymorphisms and the risk of lung cancer. However, the conclusions remain controversial.The aim of this study is to examine the association of tumour necrosis factor-alpha (TNF-α) and interleukin-6 （ IL-6 ） gene polymorphisms with the risk of developing lung cancer. Methods: We searched several electronic databases, including PubMed and Excerpt Medica Database (EMBASE). The effects of three polymorphisms, TNF-α-308G/A, IL-6-174G/C and IL-6-634C/G, were evaluated. The pooled odds ratio (OR) with 95% confidence interval (CI) was calculated by RevMan software. Heterogeneity was also assessed. Results: For a total of 17 studies involving 4,094 cases and 4,988 controls, were identified in this meta-analysis. Based on our results, we found an association between the TNF-α-308G/A polymorphism and lung cancer risk under the dominant model (GG+GA vs. AA, OR = 0.60, 95% CI: 0.40 to 0.89). For the IL-6-174G/C polymorphisms, the pooled ORs (95% CI) of GG/GC vs. CC, GG vs. GC/CC, GC vs. CC, and GG vs. CC were 1.22 (1.02 to 1.46), 1.22 (1.01 to 1.48), 1.22 (1.01 to 1.48), and 1.12 (0.87 to 1.44), respectively. For the IL-6-634C/G polymorphisms, the pooled ORs (95% CI) of CC/CG vs. GG, CC vs. CG/GG, and C vs. G were 1.04 (0.68 to 1.58), 0.69 (0.57 to 0.85), and 0.79 (0.67 to 0.93) respectively. Conclusions: The results of our analysis of these IL-6 polymorphisms revealed an association between IL-6 and lung cancer risk. This association, however, was not as strong as the association between TNF-α-308G/A polymorphisms and lung cancer risk. Because the current study was limited in sample size, further studies are needed to reveal more precise associations.

cancer [3].Tumour necrosis factor-alpha is a major inflammatory growth factor that plays a vital role in the regulation of immune and inflammatory responses. Genetic variants of the TNF-α gene may affect the host immune system and correlate with lung cancer risk. As the 308(G/A) polymorphism in the TNF-α promoter region is associated with altered protein levels and transcription rates [4], we explored the relationship between this polymorphism and lung cancer risk. IL-6 is a key pro-inflammatory cytokine that may also play a critical role in carcinogenesis and regulate the expression of several genes involved in inflammation, It is expressed in tumor-infiltrating cells [5]. Hence, we also analysed two polymorphisms in the IL-6 promoter region, 174(G/C) and 634(C/G). Genetic variants, especially functional polymorphisms located in the promoter regions of candidate genes, are known to quantitatively alter gene expression. Although several studies have reported on these IL-6 and TNF-α gene polymorphisms, the results have been inconsistent and even controversial. We have also not been able to draw conclusions that agree with the existing literature. These discrepancies may arise from a variety of factors, including differences in study populations or the use of small sample sizes. A meta-analysis would thus aid in revealing a more precise association between these gene polymorphisms and lung cancer risk. A stable and reliable conclusion is necessary to unify these inconsistences and provide more conclusive results.

Search strategy
We conducted a systematic literature search of the PubMed database and Excerpta Medica Database (EMBASE) from the earliest available date through October 2015 using the terms "Tumor necrosis factor alpha-308 or TNF-alpha-308 or Interleukin-6 or IL-6" and "Single Nucleotide Gene Polymorphism or SNPs or Nucleotide Gene Polymorphism, Single" and" Lung Cancer or Pulmonary Cancer or Lung Neoplasm" without any language restrictions. In addition, we included a manual search step to add more studies to our reference list. All distinctly irrelevant studies, case reports, editorial comments, and review articles were excluded.

Inclusion criteria and exclusion criteria
Eligible studies from our literature search had to fulfil the following inclusion criteria: (1) published studies based on case-control designs assessing the association between these gene(TNF-α-308G/A or IL-6-174G/C or IL-6-634C/G) polymorphisms and lung cancer risk, (2) case-control studies based on unrelated individuals, and (3) studies reporting sufficient sample sizes, distributions of alleles, genotype frequency data, and other information. Studies were excluded for the following reasons: (1) the study did not pertain to gene polymorphisms, (2) the study had insufficient information for data extraction and (3) the study was not conducted in humans. If a study reported results from different ethnicities, we treated those ethnicities independently.

Quality Assessment
Two authors independently assessed the quality of every imported article according to a trial quality tool, which was adapted from existing quality tools, was used to divide the trials into four quality categories : A. low risk of bias in the randomization process; B. moderate risk of bias; C. high risk of bias; D. insufficient information to score allocation concealment. Quality assessments were conducted and evaluated by 5 reviewers.

Data extraction
Data were extracted independently by two reviewers based on the inclusion criteria listed above. Any inconsistencies were discussed, and consensus was reached. The following data were obtained: name of first author, year of publication, site of original research, ethnicity of the population, number of patient cases, number of controls, type of controls, method of genotyping, and whether the gene distribution of the controls was in compliance with Hardy-Weinberg Equilibrium (HWE). This information is reported in Tables 1 and 2.

Characteristics of the included studies
The process of screening articles is shown in Figure 1

Association between the TNF-α-308G/A polymorphism and lung cancer risk
As shown in Table 3, ten studies including a total of 2013 patient cases and 2410 controls were selected to evaluate the association between TNF-α-308G/A and lung cancer risk. The results revealed that the    Table 3 Main results of the meta-analysis Test of association OR: odds ratio; CI: confidence interval; F: fixed-effects model; R: random-effects model.P A : P value for test of the association; P Q : P value for between study heterogeneity.

Association between the IL-6-174G/C polymorphism and lung cancer risk
We identified four studies that investigated the association between the IL-6-174G/C polymorphism  Table 3. The forest plot for the dominant model of IL-6-174G/C is shown in Figure 3.

Association between the IL-6-634C/G polymorphism and lung cancer risk
We identified a single article that included three studies with a total of 617 patient cases and 1,091 controls that investigated the association between the IL-6-634C/G polymorphism and lung cancer risk.
All of the studies were carried out in Europe. Our analysis is shown in Table 3   OR: odds ratio; CI: confidence interval; NA: not available.

Subgroup analysis
We conducted further subgroup analyses based on ethnicity. All of the subgroup results are shown in Table 4. Two studies on the TNF-α-308G/A polymorphism [9,10] were conducted in Asian cohorts, five [11,12,13,14,15] were conducted in European cohorts, three were conducted in [13,16,17]Caucasian cohorts, and two [13,18]were conducted in African cohorts. Our meta-analysis revealed a statistically significant association between the TNF-α-308G/A polymorphism and lung cancer risk in Asians.
Among the IL-6-174G/C studies, one [13]was conducted in Caucasian and African cohorts; one [19], in Asian cohorts; and the remaining [13,14,20], in European cohorts. Similarly, we found a significant relationship between IL-6-174G/C and increased lung cancer risk specifically in Asians. We were not able to perform an analysis for other single nucleotide polymorphisms (SNPs) as we lacked sufficient information from individual subgroups.

Sensitivity analysis
Sensitivity analysis was done for each result, reflecting the influence of the individual data set to the pooled ORs. After removing the maximum-weight study, The pooled OR remained unchanged, which indicate that high reliability in our conclusions (data not shown).

Discussion
Over the past several decades, the link between TNF-α and IL-6 variations and disease has garnered increasing attention. However, the field is lacking in consistent and reliable conclusions regarding infectious diseases [21]. As TNF-α is closely relied upon as a defence mechanism against disease, its regulation and expression after gene conversion can become increasingly uncontrollable. Therefore, we cannot confirm that our results are entirely thorough. Further studies are needed to assess multiple conditions of TNF-α expression to validate our findings.
An autocrine or paracrine factor, IL-6 may have direct effects on tumour cells to modulate their growth. IL-6 may also indirectly promote tumour cell growth by inducing the acute phase reaction, inhibiting apoptosis and angiogenesis [22]. The 5 kb IL-6 gene is located on chromosome 7p21-14 and contains 4 introns and 5 exons. The promoter region contains many different regulatory elements, including transcription factor binding sites for NF-IL6, NF-ĸB, Fos/Jun, CRBP, glucocorticoid receptor, and others [23]. Transcription factors play a critical role in the expression of IL-6. Hence, polymorphisms in the promoter region of the IL-6 gene may cause variations in transcription, causing the gene locus to strongly impact its expression and susceptibility to diseases. Specifically, in the current meta-analysis, we found that polymorphisms of IL-6-174G/C and IL-6-634C/G were associated with the occurrence and development of lung cancer, but these connections were negligible. Thus, although our results were very different from those of previous studies, we illustrate that some connections exist between IL-6 and lung cancer. This association may become more pronounced upon further investigation.
The association of cytokine gene polymorphisms in TNF-α and IL-6 with lung cancer risk has been reported in recent years. The results of our study may not be in accordance with other published findings, which may be attributed to certain limitations in the present meta-analysis. For instance, several SNPs in the TNF-α and IL-6 genes have been identified. However, because TNF-α-308G/A, IL-6-174G/C, and IL-6-634C/G gene polymorphisms are the most widely researched, our study was limited to these three polymorphisms. Further investigation into the link between lung cancer and other TNF-α and IL-6 gene polymorphisms should be conducted in future studies. Additionally, the sample sizes used in our subgroup analyses were limited, making it impossible to stratify according to different tumour types. Therefore, we performed only a subgroup analysis stratifying for ethnicity. Furthermore, we could not precisely demonstrate the interactions between gene/gene and gene/environmental factors due to a lack of specific data published in each study.

Conclusion
we illustrate that TNF-α gene polymorphisms at position 308G/A and IL-6 gene polymorphisms at positions 174G/C and 634C/G have weak associations with lung cancer. However, the mechanism(s) underlying these correlations have yet to be determined [8]. We can be almost certain, however, that such occurrences in gene exchange may lead to differences in gene expression that can have an impact on lung cancer risk. To reach a more definitive conclusion, further studies are needed to assess multiple polymorphisms in the TNF-α and IL-6 genes. Larger sample sizes in case-control studies conducted with different pathological types to investigate the interactions between TNF-α and IL-6 with other individual genes and environmental factors may further contribute to our knowledge of lung cancer pathogenesis. PA: P value for test of the association; PQ: P value for between study heterogeneity; NA: not available.