[DNA damage associated genetic variants contribute to lung cancer susceptibility in a Han Chinese population].

Objective: To explore the association between DNA damage-related genetic variants and lung cancer susceptibility in a Han Chinese population.
Methods: This case-control study enrolled patients from the Cancer Hospital of Jiangsu Province and Jiangsu Province Hospital from 2003 to 2009. Controls were randomly selected from individuals who visited the same hospital or a community-based health examination program during the same time period. A 5 ml venous blood sample was obtained from each participant and epidemiological information was collected on a standard questionnaire. Illumina Infinium(®) BeadChip was used for genotyping of 35 DNA damage-related single nucleotide variations (SNVs), which were identified in our previous study. Multivariate and binary logistic regressions were used to calculate the OR and 95%CI for lung cancer risk. HaploReg V4.1 and Regulome DB were used to understand functional annotation on important SNV.
Results: The distributions of age (61.06±10.15) vs. (61.32±11.07) years; t=-0.72, P=0.473) and sex (χ(2)=1.81, P=0.179) were similar between cases and controls. However, the case group had a higher frequency of smokers (61.08% vs. 48.54%; χ(2)=50.04, P<0.001) and heavy smokers (42.28% vs. 24.07%; χ(2)=122.32, P<0.001). Among the 34 SNVs that passed quality control, two SNVs were significantly associated with lung cancer risk after adjustments for age, sex and cumulative smoking dose: rs9267576 C>A (CA genotype/CC genotype, OR=1.56, 95% CI: 1.01-2.40) and rs3130683 A>G (AG genotype/AA genotype, OR=1.87, 95%CI: 1.13-3.09). After step-wise logistic regression analysis, only the rs3130683 SNV was retained in the model, indicating that the association between rs9267576 and lung cancer may be due to the effect of rs3130683. Functional annotation indicated that rs3130683 was located in the promoter and enhancer regions, and was an expression quantitative trait loci of HLA. The Cancer Genome Atlas indicated that expression of HLA-C, DQB1, DRB1 and DRB5 in lung cancer tissue was significantly lower than in paired normal tumor-adjacent tissue, with down-regulation of the four respective genes in 81.3%, 88.8%, 90.7% and 90.7% of lung cancer tissues (P-values were 6.68×10(-15), 2.21×10(-13), 2.20×10(-16), 2.58×10(-13), respectively).
Conclusions: The SNV rs3130683 (A>G) was associated with the risk of lung cancer in a Han Chinese population. This SNV may affect the risk of lung cancer by regulating HLA expression.