Head-and-neck squamous cell carcinoma risk in smokers: no association detected between phenotype and AHR, CYP1A1, CYP1A2, or CYP1B1 genotype. Issue 1 (December 2016)

• Record Type: Journal Article
• Authors: Jorge-Nebert, Lucia; Zhang, Ge; Wilson, Keith; Jiang, Zhengwen; Butler, Randall; Gluckman, Jack; Pinney, Susan; Nebert, Daniel
• Is Part Of: Human genomics. Volume 10:Issue 1(2016)
• Page Start: 1
• Page End: 17
• Publication Date: 2016-12
• Subjects: AHR gene -- CYP1A1 -- CYP1A2 -- CYP1B1 genes -- Tag-SNPs (single nucleotide polymorphisms) -- Head-and-neck squamous cell carcinoma (HNSCC) -- Cigarette smoking -- Extreme discordant phenotype method -- Population stratification -- Candidate-gene approach to genotype-phenotype association; Genomics -- Periodicals; Human genome -- Periodicals; Genetic Research -- Periodicals; Pharmacogenetics -- Periodicals; 611.01816
• DOI: 10.1186/s40246-016-0094-y
• Languages: English
• ISSNs: 1479-7364

Abstract

Abstract Background Head-and-neck squamous cell carcinoma (HNSCC) differs between smokers and nonsmokers in etiology and clinical presentation. Because of demonstrated unequivocal involvement in smoking-induced cancer in laboratory animals, four candidate genes––AHR, CYP1A1, CYP1A2, andCYP1B1 ––were selected for a clinical genotype-phenotype association study of HNSCC risk in smokers. Thirty-six single-nucleotide variants (mostly tag-SNPs) within and near these four genes [16 (AHR ), 4 (CYP1A1 ), 4 (CYP1A2 ), and 12 (CYP1B1 )] were chosen. Methods Extreme discordant phenotype (EDP) method of analysis was used to increase statistical power. HNSCC patients––having smoked 1–40 cigarette pack-years––represented the "highly-sensitive" (HS) population; heavy smokers having smoked ≥80 cigarette-pack-years without any type of cancer comprised the "highly-resistant" (HR) group. The vast majority of smokers were intermediate and discarded from consideration. Statistical tests were performed onN = 112 HS andN = 99 HR DNA samples from whole blood. Conclusions Among the four genes and flanking regions––one haploblock, ACTTGATC in the 5′ portion ofCYP1B1, retained statistical significance after 100, 000 permutations (P = 0.0042); among our study population, this haploblock was found in 36.4% of African-American, but only 1.49% of Caucasian, HNSCC chromosomes. Interestingly, in the 1000 Genomes Project database, frequency of this haplotype (in 1322 African and 1006 Caucasian chromosomes) is 0.356 and 0.003, respectively. This study represents an excellent example of "spurious association by population stratification". Considering the cohort size, we therefore conclude that the variant alleles chosen for these four genes, alone or in combinations, are not statistically significantly associated with risk of cigarette-smoking-induced HNSCC.