Abstract 2197: Interaction between genetic variants in one-carbon metabolism and folate biomarkers on colorectal cancer risk: The Women's Health Initiative observational cohort

Abstract

Abstract Folate-mediated one-carbon metabolism (FOCM) is important in the development of colorectal cancer (CRC), but gene-nutrient interactions in this relationship have not been comprehensively evaluated. We investigated whether folate modified the association of genetic variants of FOCM with CRC risk. Among 821 Caucasian case/control matched pairs in the Women's Health Initiative observational cohort (1993-2008), we genotyped 453 single nucleotide polymorphisms (SNPs) tagging 227 independent loci in 30 genes in FOCM pathways (AARS, ADH1B, ADH1C, BHMT, BHMT2, CBS, DHFR, DNMT1, DNMT3A, DNMT3B, EHMT1, EHMT2, FOLH1, FOLR1, GGH, GNMT, MAT1A, MAT2B, MTHFD1, MTHFD2, MTHFR, MTR, MTRR, PON1, PRDM2, SHMT1, SHMT2, SLC19A1, TCN2, and TYMS). Folate concentrations in plasma and red blood cell (RBC) were measured by radioassays. Multivariable conditional logistic regression was used to estimate odds ratios, and likelihood ratio tests were performed to test the differences in the associations of SNPs with CRC among plasma/RBC folate tertile categories. Nominal P-values were reported for non-synonymous and candidate SNPs. For tagSNPs, we used the Benjamini and Hochberg method to control false discovery rate (FDR) at the gene level. When comparing homozygous variants with wild-types, we observed higher risks of CRC associated with genetic variants in folate absorption and uptake in the presence of low (tertile 1) plasma folate concentrations: GGH (rs16930073, flanking 3′UTR, P-interaction=0.006), FOLH1 (rs10839236, intron, candidate, P=0.018; rs2299650, intron, P=0.016), and FOLR1 (rs651646, intron, P=0.021). In other FOCM pathways (methionine/S-adenosylmethionine reactions, vitamin B12 transport, and DNA methylation), plasma/RBC folate also modified important non-synonymous and candidate SNPs in relation to CRC risk: MTRR (C524T [p.S202L], P=0.045), CBS (rs1801181 [p.A360A], P=0.035), TCN2 (C776G [p.R232P], P=0.040), DNMT1 (A201G [p.I311L], P=0.016), DNMT3A (rs11695471 [intron 22 A>T], candidate, P=0.002), and DNMT3B (283T>C, 5′ near gene, p=0.043). In addition, interactions with plasma/RBC folate were observed for several variants in paraoxonase 1 (PON1), an enzyme generating homocysteine from homocysteine thiolactone: p.L55M (P=0.021), rs8491 (3′UTR, P=0.037), rs854548 (3′ near gene, P=0.033), rs854551 (3′UTR, P=0.045), and rs3917527 (intron, P=0.006). All significant interactions with tagSNPs remained as “noteworthy” at the FDR 25% level. In conclusion, our observations highlight the importance of folate status when investigating FOCM-related genetic influences on CRC risk in postmenopausal women. Citation Format: Ting-Yuan D. Cheng, Karen W. Makar, Marian L. Neuhouser, Joshua W. Miller, Xiaoling Song, Elissa C. Brown, Shirley A.A. Beresford, Yingye Zheng, David J. Duggan, Elizabeth M. Poole, Nina Habermann, Reka Toth, Lynn B. Bailey, Marie A. Caudill, Cornelia M. Ulrich. Interaction between genetic variants in one-carbon metabolism and folate biomarkers on colorectal cancer risk: The Women's Health Initiative observational cohort. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2197. doi:10.1158/1538-7445.AM2014-2197