Abstract Anomalous DNA methylation contributes to carcinogenesis by regulating gene expression and may be important in germ cell tumor (GCT) development as germ cells undergo tremendous epigenetic reprogramming during embryogenesis. Histologic subtypes of GCTs show differences in global methylation patterns with less differentiated tumors (e.g. embryonal carcinoma) experiencing lower levels of global methylation relative to more differentiated tumors (e.g. teratoma). Differences in promoter methylation have also been reported in GCTs. In particular, developmental signaling pathways and the BMP/TGF beta pathway have been consistently found to be differentially methylated. In this analysis, we conducted an epigenome wide association study to identify gene-specific methylation differences by GCT histology. Using the Illumina HumanMethylation450K array we determined methylation profiles for 154 pediatric GCTs (21 germinomas, 54 mixed, 9 teratomas, 70 yolk sac tumors [YST]) from fresh frozen and formalin fixed paraffin embedded tumor tissue samples. Fisher's exact tests were used to assess differences in the distributions of age group, sex, and tumor location by histologic subtype (alpha=0.05 for a two-sided test of significance [SAS v9.4]). Bioconductor package minfi was used to identify differential methylation between GCT histologic subtypes by comparing the methylation beta values. To identify differential methylation of genes predictive of tumor histology, a generalized linear model (glm) with LASSO was fit (R v3.4.1) to the quantile normalized beta values for all probes from all samples. There were significant differences in age, sex, and tumor location by histologic subtype (all p<0.01). We observed unique clusters for both germinoma and YST, but a single cluster for mixed tumors and teratoma. The methylation clusters did not differ by age, sex or FFPE status after accounting for histologic subtype. When stratifying by age group (0-10 vs. 11-19 years), we observed 2,043 differentially methylated regions (DMRs) (FWER <0.05); however, these differences did not persist once we adjusted for tumor histology. Differential methylation tests across histologic subtypes resulted in 8,049 DMRs (FWER < 0.05). Germinoma clustered most consistently and displayed a unique methylation pattern relative to other tumor types, which were more similar to one another. Fitting a glm with LASSO to the methylation data identified 21 probes that were predictive of histologic subtype with misclassification error < 20%: ANKRD24, ASL, ATG13, CCDC53, CKB, FAM224B, FZD10, KCNMB3, LINC00939, LOC101927248, MIR378C, MMP14, MUC1, OGFOD3, PLP2, PLPP3, PRR5, SERPINA6¸ SGSM2, VARS, and YWHAZ. In our analyses, we identified a large number of genes with differential methylation by tumor histology. These data may be informative in understanding GCT etiology and the timing of GCT initiation. Citation Format: Lindsay A. Williams, Lauren Mills, Anthony J. Hooten, Erica Langer, Michelle Roesler, A Lindsay Frazier, James Amatruda, Jenny N. Poynter, Jenny N. Poynter. Differences in methylation patterns by pediatric germ cell tumor histologic subtype [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2064.
Read full abstract