Demographic and clinical characteristics
A case control study containing 535 HNC patients with a mean age of 46.80 ± 15.08 and 538 healthy controls with a mean age of 46.94 ± 13.79 was performed. Table 2 listed the demographics and clinical information of the cases and controls enrolled in the present study. Gender, age, nasopharyngeal cancer and thyroid cancer were not statistically significant in the cases and controls (P > 0.05), while lymph node metastasis and staging were significantly different between the cases and controls (P < 0.05).
Hardy-Weinberg equilibrium and allelic frequency analyses
Table 3 summarizes the basic characteristics of the SNPs in the study population, and shows that all the six SNPs from controls were in accordance with HWE (P > 0.05). The average SNPs call rate was more than 95% in cases and controls. We compared the differences in frequency distributions of alleles between cases and controls compared by Pearson’s χ 2 test and found only the minor allele “C” of rs2243250 from IL4 was significantly associated with an increased risk of HNC in the study population (P = 0.044, OR = 1.23, 95% CI = 1.01 - 1.50).
Genetic model analysis of the association between selected SNPs and HNC risk
Furthermore, we hypothesized that harboring the minor allele of each SNP was a risk factor, compared with possessing the wild-type allele. Four genetic models (codominant, dominant, recessive and log-additive) were applied to assess the relationships between the SNPs and risk of HNC using an unconditional logistic regression analysis with adjustments for age and gender (Table 4). Our analyses showed that rs2243250 was associated with an increased risk of HNC both before and after adjusted with age and gender under the codominant model (crude OR = 2.10, 95% CI: 1.16 - 3.79, P = 0.014; adjust OR = 2.10, 95% CI: 1.16 - 3.79, P = 0.014), the recessive model (crude OR = 2.02, 95% CI: 1.13 - 3.62, P = 0.018; adjust OR = 2.02, 95% CI: 1.13 - 3.62, P = 0.018) and the log- additive model (crude OR = 1.24, 95% CI: 1.01 - 1.52), P = 0.042; adjust OR = 1.24, 95% CI: 1.01 - 1.52, P = 0.042). Rs2243267 was found play a harmful role with the HNC risk both before and after adjusted with age and gender based on the codominant model (crude OR = 2.12, 95% CI: 1.18 - 3.81, P = 0.012; adjust OR = 2.12, 95% CI: 1.18 - 3.81, P = 0.012) and the recessive model (crude OR = 2.09, 95% CI: 1.17 - 3.73, P = 0.013; adjust OR = 2.09, 95% CI: 1.17 - 3.73, P = 0.013). Rs2243270 was also found to be significantly associated with an increased the risk of HNC both before and after adjusted with age and gender under the codominant model (crude OR = 2.06, 95% CI: 1.14 - 3.71, P = 0.016; adjust OR = 2.06, 95% CI: 1.14 - 3.71, P = 0.016) and the recessive model (crude OR = 2.02, 95% CI: 1.13 - 3.62, P = 0.018; adjust OR = 2.02, 95% CI: 1.13 - 3.62, P = 0.018). The three other SNPs (rs2227284, rs2243283 and rs2243289) were not found to be associated with HNC risk under the four genetic models (P > 0.05).
Stratification analysis of IL-4 polymorphisms and HNC risk
To further explore the gender influence on the potential HNC susceptibility of selected polymorphisms in IL-4 gene, we performed the same statistical analysis in males and females, separately (Table 5). We found that rs2243250 was associated with an increased risk of HNC in male under codominant model (OR = 6.66, 95% CI: 1.44 - 30.87, P = 0.015), recessive model (OR = 6.06, 95% CI: 1.32 - 27.88, P = 0.021), log-additive model (OR = 1.52, 95% CI: 1.07 - 2.17, P = 0.020) and allele model (OR = 1.46, 95% CI: 1.04 - 2.03, P = 0.027). Rs2243267 was observed to be associated with the increased the susceptibility of HNC in male under codominant model (OR = 6.99, 95% CI: 1.52 - 32.06, P = 0.012), recessive model (OR = 6.68, 95% CI: 1.47 - 30.42, P = 0.014) and log-additive model (OR = 1.42, 95% CI: 1.00 - 2.01, P = 0.049). Rs2243270 was significantly associated with increasing the HNC risk among male in codominant model (OR = 6.99, 95% CI: 1.52 - 32.06, P = 0.012), recessive model (OR = 6.68, 95% CI: 1.47 - 30.42, P = 0.014) and log-additive model (OR = 1.42, 95% CI: 1.00 - 2.01, P = 0.049). Rs2243289 also related to increase the risk of HNC among male in codominant model (OR = 6.25, 95% CI: 1.35 - 28.96, P = 0.019) and recessive model (OR = 6.09, 95% CI: 1.32 - 28.00, P = 0.020). On the contrary, the rs2243283 in IL-4 showed a decreased risk to the HNC susceptibility among male in recessive model (OR = 0.27, 95% CI: 0.07 - 0.98, P = 0.046). In addition, we have not found any association between six SNPs in IL-4 gene and the risk of HNC in females.
Stages stratification analysis showed that only rs2243283 was associated with an increased risk of HNC under codominant model (OR = 2.64, 95% CI: 1.14 - 6.13, P = 0.023), dominant model (OR = 2.78, 95% CI: 1.22 - 6.37, P = 0.016), log-additive model (OR = 2.07, 95% CI: 1.27 - 5.90, P = 0.010) and allele model (OR = 2.07, 95% CI: 1.10 - 3.89, P = 0.022) (Table 6).
LD and haplotype analysis
One haplotype block including five SNPs (rs2243250, rs2227284, rs2243267, rs2243270 and rs2243283) with D’ = 1 was detected in IL-4 SNPs by haplotype analyses, as shown in Figure 1. Haplotypes with frequencies equaling to 1% were selected for the analysis. The five SNPs constructed four haplotypes (TTCGG, CGGAC, CTGAC, and TTCGC). The results of the association between the IL4 haplotype and the risk of HNC are listed in Table 7. The haplotype CTGAC was significantly associated with decreased risk of HNC both before and after adjusted with age and gender (crude OR = 0.67, 95% CI: 0.48 - 0.94, P = 0.020; adjust OR = 0.67, 95% CI: 0.47 - 0.94, P = 0.019).