Although much is known about molecular and chromosomal characteristics that distinguish glioma histologic subtypes, DNA methylation patterns of gliomas and their association with mutation of isocitrate dehydrogenase (IDH) genes has only recently begun to be investigated. We measured DNA methylation of glioblastomas, astrocytomas, oligodendrogliomas, oligoastrocytomas, ependymomas, and pilocytic astrocytomas (n = 131) from the Brain Tumor Research Center at UCSF, as well as non-tumor brain tissues (n = 7), with the Illumina GoldenGate methylation array. Methylation data were subjected to recursively partitioned mixture modeling (RPMM) to derive methylation classes. Next, differential DNA methylation between tumor and non-tumor was assessed. RPMM was again used to model methylation data for tumors with IDH mutation data (n = 95). Associations between IDH mutation and survival were also examined. Among all gliomas (n = 131), RPMM resulted in eleven methylation classes, and there was a statistically significant association between methylation class and glioma histologic subtype (P < 2.2 × 10−16). Comparing non-tumor brain tissues to gliomas to investigate differential methylation, glioblastomas showed a low ratio of hyper- to hypomethylated loci (ratio = 1.3) compared with the ratio for astrocytomas, oligoastrocytomas, and oligodendrogliomas (ratios = 3.7, 7.6, and 9.7, respectively). Ependymomas had increased hypomethylation (ratio = 0.3). These ratios were significantly different across glioma subtypes (Permutation P < .0001). Assessing IDH1 and IDH2 mutation (IDH), 59% of gliomas had an IDH mutation. IDH mutation was more common in oligoastrocytomas, oligodendrogliomas, or astrocytomas than in glioblastomas, pilocytic astrocytomas, or ependymomas (P = 6.4 × 10−9); in lower-grade tumors (P = .01); in tumors with TP53 mutation (P = .06); and in younger patients (P = .0009). In addition, patients whose tumors harbored mutant IDH had significantly improved survival (HR = 0.27, 95% CI = 0.10 to 0.72). In tumors with available IDH mutation data, RPMM resulted in nine methylation classes, methylation class was significantly associated with IDH mutation (P = 3.0 × 10−16), and this association remained significant when controlling for patient age and tumor histology (likelihood ratio P < .0001). Only two methylation classes contained tumors with IDH mutation, and they had a homogeneous, hypermethylation-rich character compared to the methylation classes for tumors with wild-type IDH. The homogeneity of methylation classes for gliomas with IDH mutation, despite their histologic diversity, strongly suggests that IDH mutation “drives” the observed hypermethylated phenotype, irrespective of tumor histology.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 878. doi:10.1158/1538-7445.AM2011-878