Folate And DNA Methylation Patterns Within The Breast Cancer EPIC Study

Flavie PERRIER, nternational Agency for Research on Cancer, France
AMBATIPUDI S. Mechanisms of Carcinogenesis Section, International Agency for Research on Cancer, Lyon, France , HERNANDEZ VARGAS H. Mechanisms of Carcinogenesis Section, International Agency for Research on Cancer, Lyon, France , FERRARI P. Nutrition and Metabolism Section, International Agency for Research on Cancer, Lyon, France , CHAJES V. Nutrition and Metabolism Section, International Agency for Research on Cancer, Lyon, France , MORENO H. Universidad Autonoma Metropolitana, Mexico City, Mexico , MATEJCIC M. Nutrition and Metabolism Section, International Agency for Research on Cancer, Lyon, France , HERCEG Z. Mechanisms of Carcinogenesis Section, International Agency for Research on Cancer, Lyon, France , ROMIEU I. Nutrition and Metabolism Section, International Agency for Research on Cancer, Lyon, France

1 Nutrition and Metabolism Section, International Agency for Research on Cancer, Lyon, France

Purpose: Experimental evidence supports a mechanistic basis for the effect of specific nutrients on breast cancer (BC) development. Among these, there is increasing evidence showing that folate is a relevant candidate for modulation of the epigenome. However, the epidemiological evidence linking nutrition, epigenome and BC development is poorly documented. We aimed to assess the relationship between biomarkers of folate intake and DNA methylation patterns in the European Prospective Investigation into Cancer and nutrition (EPIC) study.
 
Methods: 451 BC cases and as many matched controls were analysed as part of a nested case-control study on BC within the EPIC cohort. Genome-wide DNA profiling in white blood cells was measured among cancer free women on about 450,000 CpG sites. Beta regression adjusted for laboratory variables (e.g batch) were conducted for each CpG site to determine methylation patterns related to blood folate level. Models were also adjusted for alcohol consumption and age at recruitment, menopausal status, level of different lymphocyte subtypes and BC status. False discovery rate (FDR) was used to control for multiple-testing.
 
Results: Folate blood levels (median: 12.9 hmol/L, range: 1.2-75.2 hmol/L) were significantly associated with methylation levels in 8 CpG sites after FDR correction. For 7 of these CpGs a high level of folate was associated with an increased methylation level while one CpG was inversely associated. Most of these CpG sites (7) were located in island regions on the chromosome and one in an open sea region. DNA methylation levels in these sites were not correlated (maximum correlation of 0.51).
 
Conclusions: A list of 8 CpG sites having a methylation level significantly influenced by the blood folate level was identified. Folate can either increase or decrease the level of methylation in these CpG sites.
 
Funding sources: Institut National du Cancer, Fondation de France and World Cancer Research Fund International