Abstract LB-127: Maternal folate gene polymorphisms define distinct DNA methylation clusters in children with acute lymphoblastic leukemia

Abstract BACKGROUND. Acute lymphoblastic leukemia (ALL) is the most common form of pediatric cancer and occurs as early as infancy, supporting a role of intrauterine exposures. Maternal polymorphisms in genes regulating folate metabolic pathway have recently been reported to modify ALL risk in the child; however, the mechanistic basis underlying this association remains unknown. Folate metabolic pathway is indispensible for DNA methylation and synthesis during embryogenesis and early development. Gene polymorphisms affecting the mother's folate metabolism capacity may exert influence on the offspring's DNA methylome and predispose the child in utero with an increased risk of ALL. Promotor CpG island DNA methylation is a well-recognized phenomenon for silencing tumor-suppressor genes in cancers. Recent studies have discovered subsets of cancer with high degrees of methylation. This phenomenon, termed CpG methylator phenotype (or CIMP), has been validated in different adult cancers. However, the origin of CIMP remains unidentified. Additionally, very few studies have examined the DNA methylome of pediatric ALL and the discovery of CIMP was mainly based on candidate-gene approach. Therefore, in the current study we examined genome-wide DNA methylation in pediatric ALL and determined whether maternal folate gene polymorphisms were associated with distinct clusters of DNA methylation in their offspring. METHODS. We collected bone marrow (BM) specimens from 53 cases whose mothers were genotyped previously on 67 folate SNPs and seven BM specimens from healthy donors. We classified ALL patients as high-risk or low-risk based their mother's folate genotype. Methylated CpG amplification coupled with Hi-Sequencing was performed to determine genome DNA methylation profiles. One-way ANOVA was applied to assess DNA methylation differences among the high-risk, low-risk, and normal groups. Samples were clustered hierarchically using Euclidean distance and complete linkage, and analyzed using Principal Component Analysis (PCA). RESULTS. We have sequenced and analyzed 21 samples (15 cases and 6 normal controls). A mean of 1,401,690 unique reads per sample were matched to human genome hg19. A total of 62,425 sites were identified in these samples. Of these, we identified 320 unique sites with significant differences among the three groups (FDR<0.05). Genes hyper-methylated in normal tissues were generally hypo-methylated in ALL patients, regardless of their risk groups. However, genes normally hypo-methylation showed various degrees of methylation among cancer samples. Especially, high-risk patients displayed intense hyper-methylation across all genes, when compared to the low-risk group. CONCLUSIONS. This is the first study showing that variation in the mother's folate metabolism capacity might explain the CpG methylator phenotypes in pediatric ALL and has an important public health impact in cancer prevention. Citation Format: Teresa Tingting Yiu, Spiridon Tsavachidis, Philip J. Lupo, Rabin Karen, Melissa L. Bondy, Michael E. Scheurer, Lanlan Shen. Maternal folate gene polymorphisms define distinct DNA methylation clusters in children with acute lymphoblastic leukemia. [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 LB-127. doi:10.1158/1538-7445.AM2014-LB-127