Abstract #3927: Genetic variation in MSH5 and lung cancer risk in the beta-carotene and retinol efficacy trial

Abstract

There is some evidence that impaired DNA mismatch repair (MMR) may be associated with increased lung cancer risk. We characterized variation in MSH5, a gene involved in MMR, and its 2500 bp flanking regions using nine tag SNPs in a case-control study nested within the Beta-Carotene and Retinol Efficacy Trial, a randomized chemoprevention trial in smokers. In the cohort, 790 individuals were diagnosed with lung cancer and provided a blood sample, and 1562 study participants who remained free of lung cancer were matched to the cases on age, sex, race, enrollment year, baseline cigarette smoking status (current versus former), and a history of occupational exposure to asbestos. Tag SNPs (r²=0.80) were selected using the HapMap-CEU I and II population data. An Illumina GoldenGate custom array was used to perform genotyping, and seven SNPs were successfully genotyped: rs3131382, rs2299851, rs3131379, rs707939, rs707938, rs707937, rs6905572. Analyses were limited to Caucasians (747 cases, 1477 controls), and logistic regression was used to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for individual SNPs and common haplotypes, with adjustment for the matching factors. All control genotype distributions were in Hardy-Weinberg equilibrium. The per-allele ORs (95% CI) for the minor allele of each SNP were 1.06 (0.82-1.37), 0.98 (0.79-1.21), 1.36 (1.12-1.64), 0.91 (0.80-1.04), 1.14 (1.00-1.30), 1.02 (0.87-1.19) and 0.95 (0.79-1.15), respectively. For rs3131379, the ORs and 95% CIs for one or two copies of the minor allele (compared to none) were 1.31 (1.06-1.62) and 2.30 (1.12-4.72). The rs3131379 minor allele was carried in a single haplotype that also contained the minor allele for rs707938 (although the minor allele for rs707938 was carried in several additional haplotypes), at a frequency of 11% in controls and 14% in cases. Compared to the haplotype containing no variant alleles, this haplotype was associated with an increased risk (OR 1.32, 95% CI 1.05-1.66). In a recent meta-analysis of three whole genome scan studies (5,095 cases, 5,200 controls), Wang et al. reported that two SNPs at 6p21.33 were associated with risk of lung cancer (Nature Genetics 2008;40:1407-9). These two SNPs, rs3117582 (in intron 1 of BAT3) and rs3131379 (in intron 10 of MSH5), were in high linkage disequilibrium (r²=0.99), and had per-allele odds ratios (ORs) and 95% confidence limits (CIs) of 1.30 (1.19-1.42) and 1.26 (1.16-1.38), respectively. Our results are in accord with this observation, and suggest that rs3131379 in MSH5, or another SNP in linkage disequilibrium with it, is associated with an increased risk of lung cancer. Future fine-mapping to identify causal variants and functional studies of this region are warranted.

Citation Information: In: Proc Am Assoc Cancer Res; 2009 Apr 18-22; Denver, CO. Philadelphia (PA): AACR; 2009. Abstract nr 3927.