Investigation of Cytotoxic T-lymphocyte antigen-4 polymorphisms in non-small cell lung cancer: a case-control study

The objective of this case-control study was to extensively explore the relationship of Cytotoxic T-lymphocyte antigen-4 (CTLA-4) tagging polymorphisms with susceptibility to non-small-cell lung cancer (NSCLC). We recruited 521 sporadic NSCLC cases and 1,030 non-cancer controls. The genotypes of CTLA-4 rs16840252 C>T, rs231775 G>A, rs3087243 G>A and rs733618 T>C polymorphisms were evaluated using a custom-by-design 48-Plex SNPscan Kit. Our findings revealed there was no statistically significant difference in CTLA-4 genotypes distribution among NSCLC patients and non-cancer controls. Similar findings were observed in the logistic regression analyses. However, the stratified analyses suggested CTLA-4 rs733618 vatiants were correlated with the development of NSCLC in ≥ 60 years subgroup (TC vs. TT: adjusted OR = 1.45, 95% CI = 1.04–2.02, P = 0.030) and even drinking subgroup (TC vs. TT: adjusted OR = 2.27, 95% CI = 1.11–4.60, P = 0.024 and TC/CC vs. TT: adjusted OR = 2.26, 95% CI = 1.15–4.43, P = 0.018). In conclusion, the present case-control study highlights that the CTLA-4 rs733618 T>C polymorphism was associated with the development of NSCLC in ≥ 60 years and even drinking subgroups. A fine-mapping study with functional assessment is necessary to confirm or refute our findings.


INTRODUCTION
In 2012, an estimated 1,824,700 new lung cancer (LC) cases occurred worldwide, accounting for approximately 13% of overall cancer diagnosis [1]. LC was the most common malignancy and the first leading cause of cancer-related death among males in 2012; in addition, it was the second cancer-related death among females [1]. Thus, exploration of potential heredity factors that might affect the risk of LC, especially nonsmall-cell LC (NSCLC), which was the most common subtype of cases, attracted our interest. We focused on the costimulatory molecules of immunoglobulin superfamily which regulate T-cell activation and proliferation.
Cytotoxic T-lymphocyte antigen-4 (CTLA-4), a member of the immunoglobulin superfamily, is also known as CD152. In generally, CTLA-4 is expressed on activated T cells and negatively regulate the proliferation Research Paper and the activation of T cells [2][3][4]. CTLA-4 competes with CD28 and binds to B7.1 and B7.2 which are costimulatory molecules expressed on antigen-presenting cells. In addition, the affinity between CTLA-4 and B7 molecules is higher than that of CD28 with B7 molecules [5,6]. The interaction of CTLA-4-B7 leads to the repression of T cells at the G 1 phase and the down-regulated expression of interleukin-2 (IL-2) and IL-2 receptor [7]. This interaction can also induce activated T cells to FAS-independent apoptosis, and then further restrain T lymphocytes.
CTLA-4, a immunoregulatory molecule, is encoded by a gene on chromosome 2q33. A number of singlenucleotide polymorphisms (SNPs) in CTLA-4 gene have been established. Song et al. reported that CTLA-4 +49A>G polymorphism was a prognostic predictor for advanced NSCLC [8]. In addition, Antczak et al. found that CTLA-4 expression was significantly correlated with CTLA-4 TT genotype (-318C/T). Recently, several case-control studies focused on the relationship of CTLA-4 SNPs with the risk of NSCLC [8][9][10][11][12]. However, due to the limited sample size and the number of study, the association between CTLA-4 SNPs and NSCLC susceptibility was not well understood. The objective of this case-control study was to extensively explore the relationship of CTLA-4 rs16840252 C>T, rs231775 G>A, rs3087243 G>A and rs733618 T>C polymorphisms with susceptibility to NSCLC.

Demographic characteristics
This case-control study comprised 521 NSCLC cases and 1,030 control subjects. The NSCLC patients comprised 287 males and 234 females, while the noncancer control subjects were 588 males and 442 females. The mean age and SD in the NSCLC patient was 59.76 ± 10.71 years and that was 60.34 ± 9.11 years in controls. Gender and age were well-matched between the groups (P = 0.453 and P = 0.843, respectively, Table 1). All 521 confirmed cases of NSCLC were sporadic. The genotyping successful rates were shown in Table 2 Table 4. We found no statistically significant difference in CTLA-4 genotype distribution among NSCLC patients and non-cancer controls. The similar findings were observed in the logistic regression analyses.
In a stratified analysis by cancer type of NSCLC, logistic regression analyses indicated that there was no difference in genotype distribution of CTLA-4 rs231775 G>A, rs16840252 C>T, rs3087243 G>A and rs733618 T>C polymorphisms among different NSCLC types and controls ( Table 4).
Association of CTLA-4 rs16840252 C>T, rs231775 G>A, rs3087243 G>A and rs733618 T>C Polymorphisms with NSCLC in a stratification analysis  summarized the genotype frequencies of CTLA-4 rs3087243 G>A, rs16840252 C>T and rs231775 G>A polymorphisms in the stratified analyses by gender, age, BMI, drinking and smoking status. We found no difference in genotype distribution of CTLA-4 rs16840252C>T, rs231775 G>A and rs3087243 G>A polymorphisms among NSCLC cases and the control subjects in any subgroup.

The power of the present study (α= 0.05)
For CTLA-4 rs733618 T>C, the power value was 0.629 in additive model among ≥ 60 years subgroup, and 0.651 in additive model and 0.692 in dominant model among even drinking subgroup.

DISCUSSION
In generally, immune escape may be an important mechanism in the development of malignancies. The costimulatory signal related SNPs, common variants among individuals, may play important roles in the development of human cancers. In the present study, we explored the effect of CTLA-4 tagging SNPs in NSCLC among Eastern Chinese Han population for the first time. We found that CTLA-4 tagging polymorphisms might be not correlated with the susceptibility of overall NSCLC. The results of haplotype analysis suggested that there was no difference in haplotype distribution among NSCLC cases and the control subjects. However, in the stratified analyses by age, sex, BMI, alcohol use and smoking status, we found that CTLA-4 rs733618 T>C polymorphism was associated with the development of NSCLC in ≥ 60 years and even drinking subgroups.
NSCLC is a multifactorial disease which results from the interaction between individual's genetic backgrounds and environmental risk factors. Previous studies have demonstrated that CTLA-4 rs733618 T>C polymorphism decreases a transcription factor binding site for nuclear factor 1 and weaken CTLA-4 expression on cell surface [14,15]. Accumulating evidences suggested that CTLA-4 rs733618 T>C polymorphism might be associated  [16][17][18][19]. However, the relationship of CTLA-4 rs733618 T>C polymorphism with the development of cancer was conflicting. With an interest in the correlation of CTLA-4 rs733618 T>C polymorphism with cancer susceptibility, a case-control study explored the hypothesis that CTLA-4 rs733618 T>C polymorphism was associated with the etiology of NSCLC [10]; however, this study in an Iranian population established null association between CTLA-4 rs733618 T>C polymorphism and NSCLC. In the current study, we found that CTLA-4 rs733618 T>C polymorphism was associated with the development of     [20]. Our findings might be supported by this study. To the best of our knowledge, the present case-control study was the first study to examine the potential relationship between CTLA-4 rs733618 T>C polymorphism and the development of NSCLC in Asians. However, due to the limited sample size, these potential association should be explained with very caution. In the future, more case-control studies with detailed lifestyle and environmental factors data should be conducted to explore these potential association. We should acknowledge some limitations in this case-control study. Firstly, the NSCLC patients and control subjects were enrolled from local hospitals in Eastern China and might not fully represent the general Chinese Han population. Secondly, the moderate sample size of NSCLC cases might limit the statistical power to obtain a real assessment, especially in the stratified analyses. Further well-designed fine-mapping studies with large sample sizes are needed to confirm our findings. Thirdly, the clinical information on metastasis and survival of NSCLC could not be derived till now, which restricted further explores on the potential role of CTLA-4 tagging polymorphisms in NSCLC progression and prognosis. Fourthly, the information about family cancer history was not collected. Finally, due to lack of the information about individual's lifestyle, further determination for the interactions of gene-gene and gene-environment were not carried out. In consideration of the complex pathological process of NSCLC, these gene and environmental risk factors should not be ignored.
In summary, the present case-control study highlights that CTLA-4 rs733618 T>C polymorphism was associated with the development of NSCLC in ≥ 60 years and even drinking subgroups. In addition, a larger population-based fine-mapping study as well as detailed functional assessment are necessary to confirm or refute our findings.

Study population and patient selection
A total of 521 Eastern Chinese Han population with NSCLC were included in this study. These NSCLC  During recruitment, all study subjects signed the written informed consents following the Declaration of Helsinki. The information of risk factors and demographics was obtained by a pre-structured questionnaire. NSCLC cases and controls were well-matched in terms of age and sex (Table 1). Subjects who smoked at least one cigarette per day over 1 year were considered as 'ever smokers' [21], and those who drinked no less than three times a week for more than 6 months were defined as 'ever drinkers' [21]. The Ethical Committee of Fujian Medical University approved the study protocols (No. 2017KY019).

DNA extraction and genotyping
Two milliliters blood sample was donated by each enrolled subject and stored in Ethylenediamine tetraacetic acid (EDTA)-anticoagulation tube. DNA was elaborately extracted from lymphocytes by using the Promega kit (Promega, Madison, USA). We extracted DNA according to the manufacturer's instruction (www.promega.com/ protocols/). The genotypes of CTLA-4 rs16840252 C>T, rs231775 G>A, rs3087243 G>A and rs733618 T>C polymorphisms were evaluated using a custom-by-design 48-Plex SNPscan Kit (Genesky Biotechnologies Inc., Shanghai, China) as presented in previous studies [23,24]. Briefly, we first denatured 150ng DNA sample at 98°C for 5 min. The ligation reaction was performed in an ABI 2720 thermal cycler. A 48-plex fluorescence PCR reaction  was carried out for each ligation product. Then, in an ABI 3730XL sequencer, the obtained PCR products were separated and detected by using capillary electrophoresis. The raw data were analyzed by GeneMapper 4.1 software (Applied Biosystems, USA). A 4% randomly selected DNA sample was reciprocally verified by another laboratory technician, and the reproducibility was 100%.

Statistical analysis
Statistical analysis of this case-control study was done using the SAS 9.4 Statistical Package for Windows (SAS Institute, Cary, NC). A P < 0.05 (two-tailed) was accepted as the level of significance. The results of continuous variables were presented as mean ± standard deviation (SD). The mean values of age between NSCLC patients and non-cancer controls were calculated using the Student's t-test. The deviation of HWE in controls was analyzed using an online Pearson's two-sided χ 2 test (http://ihg.gsf.de/cgi-bin/hw/hwa1.pl) [25]. The differences in smoking, drinking, demographic variables and the frequencies of genotypes between NSCLC cases and controls were also determined by χ 2 test. The crude/ adjusted ORs and the corresponding 95% CIs were harnessed to assess the relationship of CTLA-4 rs16840252 C>T, rs231775 G>A, rs3087243 G>A and rs733618 T>C polymorphism genotypes with NSCLC risk. SHESIS online haplotype construction software [(http://analysis. bio-x.cn/myAnalysis.php), Bio-X Inc., Shanghai, China] was used to obtain the haplotypes [13]. The power value of this study was calculated by Power and Sample Size software (http://biostat.mc.vanderbilt.edu/twiki/bin/view/ Main/PowerSampleSize) [26].