A two‐stage genome‐wide association study identified four potential early‐onset nonsmall cell lung cancer risk loci based on 26,652 participants in Chinese population

Abstract

AbstractEarly‐onset lung cancer is rare with an increasing incidence rate. Although several genetic variants have been identified for it with candidate gene approaches, no genome‐wide association study (GWAS) has been reported. In this study, a two‐stage strategy was adopted: firstly we performed a GWAS to identify variants associated with early‐onset nonsmall‐cell lung cancer (NSCLC) risk using 2556 cases (age ≤ 50 years) and 13,327 controls by logistic regression model. To further discriminate younger cases from older ones, we took a case–case analysis for the promising variants with above early‐onset cases and 10,769 cases (age > 50 years) by Cox regression model. After combining these results, we identified four early‐onset NSCLC susceptibility loci at 5p15.33 (rs2853677, odds ratio [OR] = 1.48, 95% confidence interval [CI]: 1.36–1.60, Pcase‐control = 3.58 × 10−21; hazard ratio [HR] = 1.10, 95% CI: 1.04–1.16, Pcase–case = 6.77 × 10−4), 5p15.1 (rs2055817, OR = 1.24, 95% CI: 1.15–1.35, Pcase–control = 1.39 × 10−7; HR = 1.08, 95% CI: 1.02–1.14, Pcase–case = 6.90 × 10−3), 6q24.2 (rs9403497, OR = 1.24, 95% CI: 1.15–1.35, Pcase–control = 1.61 × 10−7; HR = 1.11, 95% CI: 1.05–1.17, Pcase–case = 3.60 × 10−4) and 12q14.3 (rs4762093, OR = 1.31, 95% CI: 1.18–1.45, Pcase–control = 1.90 × 10−7; HR = 1.10, 95% CI: 1.03–1.18, Pcase–case = 7.49 × 10−3). Except for 5p15.33, other loci were found to be associated with NSCLC risk for the first time. All of them had stronger effects in younger patients than in older ones. These results provide a promising overview for early‐onset NSCLC genetics.