Genetic variants in MGMT and risk of lung cancer in Southeastern Chinese: a haplotype-based analysis

✍ Scribed by Zhibin Hu; Haifeng Wang; Minhua Shao; Guangfu Jin; Weiwei Sun; Yi Wang; Hongliang Liu; Ying Wang; Hongxia Ma; Ji Qian; Li Jin; Qingyi Wei; Daru Lu; Wei Huang; Hongbing Shen

• Publisher: John Wiley and Sons
• Year: 2007
• Tongue: English
• Weight: 323 KB
• Volume: 28
• Category: Article
• ISSN: 1059-7794
• DOI: 10.1002/humu.20462

✦ Synopsis

O6-alkylguanine-DNA alkyltransferase (MGMT) is a universal DNA repair protein involved in the DNA direct reversal repair pathway that copes with alkylating carcinogens. Reduced MGMT expression as well as enzyme activity may result in an increased susceptibility to cancers. To elucidate the role of sequence variation in MGMT in the etiology of lung cancer, we conducted a comprehensive association study focusing on linkage disequilibrium (LD) structure of common variations across the MGMT sequence and its modification effect on smoking-related lung cancer risk. We rebuilt the LD block of MGMT by genotyping 39 SNPs and selected a subset of 10 haplotype-tagging SNPs (htSNP) and three pre- and interblock SNPs to capture variation across MGMT. By using a haplotype-based multifactor dimensionality reduction (MDR) analysis, we found that there were significant more-than-multiplicative interactions between diplotypes in block 5 and cumulative smoking and additive interaction between genotypes of preblock SNP rs1625649:C>A and smoking status in relation on lung cancer risk. Diplotypes in block 3 and block 5, genotypes of rs1625649:C>A, and trichotomized cumulative smoking are the four factors included in the MDR-defined best model on lung cancer risk. When these variables were combined and dichotomized, we found that subjects carrying the combined risk stratum had a significantly increased risk for lung cancer of 4.10-fold (odds ratio [OR]=4.10, 95% confidence interval [CI]=3.12-5.37, P=2.09 x 10(-24)). These findings suggest that genetic variants in MGMT may modulate the risk of smoking-related lung cancer. This haplotype-based interaction analysis might provide a "proof-of-principle" approach for studying candidate genes in cancer susceptibility.