Association between IL‐4 tagging single nucleotide polymorphisms and the risk of lung cancer in China

Abstract Background In China, lung cancer is also the most commonly diagnosed cancer with a lower 5‐year survival rate, leading to high social burdens. Recently, many studies highlighted the importance of inflammation in the initiation and progression of cancer. The goal of this study was to investigate the association between interleukin‐4 (IL‐4, OMIM#147780) single nucleotide polymorphisms (SNPs) and lung cancer susceptibility. Methods A case‐control study was conducted in a Chinese population including 199 male patients with lung cancer and 266 healthy men. Six SNPs selected from the HapMap database were genotyped using Agena MassARRAY. Genetic models and haplotype analyses were utilized to evaluate the association between SNPs and lung cancer risk. Results In our findings, rs2243250 was associated with a decreased lung cancer risk under the log‐additive model (odds ratio, OR = 0.71, 95% confidence interval, CI = 0.51–0.97, p = 0.030), and the G/G genotype of rs2227284 conferred a negative effect; the risk of lung cancer under the codominant (OR = 0.19, 95% CI = 0.04–0.87, p = 0.040) and recessive models (OR = 0.20, 95% CI = 0.04–0.88, p = 0.012) after adjusted by age. Conclusions These data indicated potential associations between IL‐4 polymorphisms and lung cancer susceptibility. That may help to improve the understanding of the relationship between inflammation and lung cancer in the future.

Many definite risk factors have been found to be linked to lung cancer risk, such as tobacco use, environmental pollution, food, genetics, and chronic obstructive pulmonary disease (Hong et al., 2015). Also, researchers revealed that genetic factors have an important role in the etiology of lung cancer, such as TP53 (OMIM#191170), RB1 (OMIM# 614041), and so forth (Granville & Dennis, 2005;Yin et al., 2016). Recently, much attention of genetic susceptibility studies of cancer has been focused on single nucleotide polymorphisms (SNPs) in candidate genes. In 2010, Timofeeva et al. reported that polymorphisms in genes involved in Phase I and Phase II of xenobiotic metabolism are related to the risk of early-onset lung cancer (Timofeeva et al., 2010). In 2015, Leng et al. conducted a genome-wide association study, and they reported that a functional 15q12 variant was identified as a risk factor for lung cancer (Leng et al., 2015).
Many clinical and epidemiologic studies have suggested strong association between inflammation and lung cancer (Barreiro et al., 2013;Sohal, Mahmood, & Walters, 2014). Tumor cells could produce several cytokines and chemokines, some of which have been implicated to mediate different steps in the pathway leading to carcinogenesis (Gomes, Teixeira, Coelho, Araújo, & Rui, 2014). Interleukin-4 (IL-4) is an anti-inflammatory cytokine related to the growth of some tumors, such as colon, breast, and lung (Gomes et al., 2012;Toi, Bicknell, & Harris, 1992). In lung cancer, IL-4 (OMIM#147780) polymorphisms have been associated with a reduced risk of non-small cell lung cancer (NSCLC) among the Portuguese and Chinese populations (Gomes et al., 2012;Gu, Shen, & Zhang, 2014). However, the overall information about association between IL-4 polymorphisms and lung cancer risk is poor.
In this study, we performed a case-control study to evaluate the association between SNPs distributed in the IL-4 gene and lung cancer susceptibility in a Chinese population. Six SNPs were genotyped and analyzed in an attempt to identify new susceptibility locus that may provide guidance for early diagnosis for lung cancer.

| Ethical compliance
The study protocol was abided by the Declaration of Helsinki and its later amendments or comparable ethical standards.
Additionally, this study was approved by the Clinical Research Ethics of Hainan General Hospital for Approval of Research Involving Human Subjects and written informed consent was obtained from all individuals included in the study.

| Study subjects and sample collection
A total of 199 male patients with lung cancer and 266 healthy men were included in this study. The demographic characteristics of the lung cancer patients recruited from the First Affiliated Hospital of Xi'an Jiao tong University were shown in Table 1. The inclusions of the case groups were: (a) males; (b) diagnosed with confirmed lung cancer patients; (c) did not have history of any other cancers; (d) never received chemotherapy or radiotherapy treatment before. Healthy controls with no evidence of lung or other cancers were selected from healthy men who did medical examination during the same period.
Five milliliters of venous blood samples were collected from each subject into tubes containing EDTA, then centrifuged and stored at −80°C.

| Statistical analysis
Data were analyzed using SPSS version 18.0 statistical software (SPSS Inc., Chicago, IL, United States) and Excel 20.0 (Microsoft Corp., Redmond, WA, United States) (Zhou et al., 2018). Distribution differences in age and gender between cases and controls were compared by using Welch's t test and Chi-square test, respectively (Adamec, 1964). Hardy-Weinberg Equilibrium (HWE) was assessed for the frequency of each SNP using a goodness-of-fit χ 2 test in the control subjects. Logistic regression analysis was used to determine the association between SNPs and the risk of lung cancer by calculating ORs and 95% CIs (Bland & Altman, 2000;Zheng et al., 2017). Web-based software SNPStats (https://www.snpstats.net/start.htm?q=snpstats/start.htm) was used to analyze the association under five genetic models (codominant, dominant, over-dominant, recessive, and logadditive) with an adjustment of age (Valls & Iniesta, 2006). Haploview software package (version 4.2) and the SHEsis software platform were used to conduct linkage disequilibrium analysis and haplotype-based associations (Barrett, Fry, Maller, & Daly, 2005;Shi & He, 2005). All p values in this study were two-tailed, and p < 0.05 was considered statistically significant. SNP-SNP interactions were investigated by using the MDR (Multifactor dimensionality reduction) software package (version 3.0.2). The best model with the maximization of cross-validation consistency was selected (Ritchie et al., 2001). In addition, traditional statistical methods were performed to examine the results from MDR analyses, and p < 0.05 was considered statistically significant (Lin et al., 2013).

| Demographic information of participants
A total of 199 patients with lung cancer and 266 healthy individuals were enrolled in this study, and all the participants are males. Demographic information of participants is listed in Table 1. The mean age of the participants was 48.06 ± 12.555 years in the control group and 59.31 ± 9.642 years in the case group. In addition, there were significant differences in age distribution between the case and control groups (p < 0.05).

| Association between SNPs in IL-4 and lung cancer risk
Six SNPs (rs2243250, rs2227284, rs2243267, rs2243270, rs2243283, rs2243289) selected from the HapMap were genotyped, Table 2 summarized the frequency information of tested SNPs among the individuals in the case and control groups. In the control groups, all six SNPs were conformed to HWE (p > 0.05). Then, we conducted two-sided Pearson chi-square tests to identify differences in allele frequency distributions between cases and controls. As shown in Table  3, rs2243250 (HGVS: NM_000589.3:g.132673462C>T) was related to a decreased risk of lung cancer (OR = 0.692, 95% CI = 0.500-0.958, p = 0.026), and similar results (OR = 0.722, 95% CI = 0.521-1.001, p = 0.050) were observed in rs2243267 (HGVS: NM_000589.3: g.132678194G>C). No associations were observed between the other four SNPs and lung cancer. After Bonferroni correction, none of the SNPs showed statistically significant associations.
Additionally, we assessed the association under five different genetic models (dominant, recessive, log-additive, codominant, and over-dominant) by logistic regression analysis. As shown in Table 4, rs2243250 was associated with a decreased lung cancer risk under the log-additive model (OR = 0.71, 95% CI = 0.51-0.97, p = 0.030); the G/G genotype of

IL-4 and lung cancer risk
We further performed linkage disequilibrium analysis for six SNPs in the logistic regression model. The haplotype structure of the IL-4 gene was analyzed, and single LD block consisting of all six SNPs were detected (Figure 1). Four common haplotypes were listed in Table 5. Interestingly, we found that the C rs2243250 G rs2227284 G rs2243267 A rs2243270 C rs2243283 A rs2243289 haplotype was associated with a significantly decreased lung cancer risk (OR = 0.63, 95% CI = 0.42-0.96, p = 0.034). After adjustment of age, the association was still significant (OR = 0.67, 95% CI = 0.46-0.97, p = 0.035).

| SNP-SNP interaction and lung cancer risk
MDR software package was used to detect the potential interactions between the analyzed SNPs in relation to lung cancer risk. Table 6 summarized the cross-validation consistency and the prediction error of best models. Obviously, the three-locus model (rs2243250, rs2243284, rs2243283) had a maximum testing accuracy of 56.4% and a maximum cross-validation consistency (10/10) that was significant at p < 0.05 level, after determined empirically by permutation testing.

| DISCUSSION
In the present hospital-based case-control study, we investigated the association between six single nucleotide polymorphisms in the IL-4 gene and lung cancer risk in the Chinese male populations. Among these SNPs, rs2243250 and rs2243267 were associated with a decreased risk of lung cancer. In the genetic model analysis, rs2243250 decreased lung cancer risk under the log-additive model, and the G/G genotype of rs2227284 conferred a protective effect on lung cancer susceptibility under the codominant and recessive models. Additionally, LD block was constructed with six SNPs in IL-4, and we found that the CGGACA haplotype was related to a significantly decreased lung cancer susceptibility. In the result of MDR, the potential interactions among the three SNPs (rs2243250, rs2243284, rs2243283).
IL-4 is a pleiotropic typeⅡcytokine which is encoded by the gene IL-4 located within the cytokine gene cluster on chromosome 5q31.1 (Rosenwasser et al., 1995). Multiple biological processes are regulated by IL-4, such as proliferation, differentiation, and apoptosis in various cell types (Nelms, Keegan, Zamorano, Ryan, & Paul, 1999). However, the role of IL-4 in cancer is paradoxical. Okada et al. found that direct delivery of IL-4 gene to a central nervous system tumor site prolongs the survival of animals by inhibiting tumor growth to some extent (Villa, 2001). Lee et al. found that IL-4 exhibited a strong inhibitory function in the two critical steps of angiogenesis-migration of endothelial cells to the inflamed site and differentiation of migrated cells to the organized vessel structure (Lee et al., 2002). Based on the above evidences, it is suggested that IL-4 has strong potential as a tumor therapy agent. But, there is still some evidence that IL-4 is a tumor-promoting molecule. For instance, Stremmel et al. found that IL-4 knockout mice are more resistant to tumor challenge than IL-4 competent mice, indicating that IL-4 may play a critical role in abrogation of the antitumor immune response (Stremmel, Greenfield, Howard, Freeman, & Kuchroo, 1999). In addition, the expression of IL-4 was up-regulated in many cancers, such as colon cancer, NSCLC, and cervical cancer (Al-Saleh et al., 1998;Asselin-Paturel et al., 1998;Berghella et al., 1998). Therefore, further studies are required to clarify the real function of IL-4 in cancers.
We found that rs2243250 was related to a lower risk of lung cancer. This was consistent with the previous studies. In 2012, Li et al. conducted a case-control study in Chinese population and reported that frequencies of IL-4 -590 genotype (TC and CC), and −590 C allele were significantly lower in patients with NSCLC than in healthy controls . The same year, Gomes et al. reported that increased expression of IL-4 associated with the TT genotype may contribute to immune surveillance during NSCLC development in Caucasian (Gomes et al., 2012). Besides, many studies showed that rs2243250 were associated with genetic susceptibility to other diseases, such as asthma, rheumatoid arthritis, and multiple sclerosis (Liu, Li, & Liu, 2012;Qiu et al., 2015). In our results, we revealed that rs2243267 and rs2227284, two less studied SNPs, were also found to be associated with lung cancer in this study. As well as, the relationship between rs2227284 and lung cancer risk has not been previously reported and we firstly reported it for the first time.
However, there were still some deficiencies in this study. Firstly, some bias in population selection should not be ignored. We observed the HWE in the controls, ORs, and 95% CIs on the basis of logistic regression analysis adjusted by age and gender to eliminate the bias to some extent. Secondly, the sample size was not larger and the reliability of the results needs to be further verified. Third, other risk factors (occupational exposures, physical activity etc.) were not included and should be considered to be assessed in the future and more comprehensive analyses about IL-4 polymorphisms in lung cancer are needed to confirm the results. In conclusion, we demonstrated that rs2243250, rs2243267, and rs2227284 in IL-4 gene are associated with a decreased risk of lung cancer in northwestern Chinese males. These results may help improve the understanding of relationship between lung cancer and inflammation related gene IL-4. On the other hand, the lung cancer-associated molecular markers identified here might be useful as diagnostic and prognostic markers for lung cancer in future clinical studies.