Abstract 2388: Low grade gliomas exhibit distinct patterns of epigenetic aging associated with prognostic molecular subtype

Abstract

Abstract Background: Low grade gliomas make up a significant proportion of malignant brain tumors and possess a high degree of heterogeneity, highlighting a need for clinically useful markers for prognosis and further biologic investigation. We used an existing model of biological age based upon DNA methylation to characterize epigenetic age in low grade gliomas (LGG) according to existing molecular classifications and assessed the prognostic utility of epigenetic age. Methods: We analyzed the full TCGA LGG cohort consisting of 516 patients; 216 grade 2 and 241 grade 3 tumors. Age at diagnosis ranged from 14-87 years with median age 43. We calculated “DNA methylation age” (DNAm age) based upon 353 CpG probes according to the epigenetic clock developed by Steve Horvath, and calculated age acceleration according to the DNAm age of tumor tissue relative to reported chronological age of patient at diagnosis. DNAm age was assessed for prognostic value using Cox proportional hazards regression. Age acceleration differences were assessed across LGG molecular classification methods using the Wilcoxon Ranked Sum test. Results: DNAm age remained highly correlated with chronological age in LGG tumor tissue (cor=0.58, p<0.001), with an average age acceleration of 33.5y. DNAm age was weakly positively correlated with mutational burden (cor=0.12, p=0.01). Higher DNAm age was positively associated with survival independent of chronologic age and Karnofsky Performance Score (p=0.003), but not when IDH mutation status was included into the Cox proportional hazards model. Further investigation showed that DNAm age acceleration was associated with IDH mutation status and was significantly increased in IDH-mutant compared to IDH-wt tumors (95% CI= 18.6-28.8y) as well as MGMT-promoter methylated versus unmethylated cases (95%CI= 18.1- 28.9y). No significant DNAm changes were observed across histological subtype or grade. Conclusions: Although the regulation of DNAm age in cancers remains poorly understood, significant association of DNAm age acceleration with prognostically useful molecular subtypes in LGGs underscores biological differences that cannot be discerned in LGGs based upon histological classification. This association suggests that LGGs may provide an avenue for investigating DNAm age, molecular subtype, and their relation to tumor behavior and clinical prognosis. Citation Format: Peter Liao, Jill Barnholtz-Sloan. Low grade gliomas exhibit distinct patterns of epigenetic aging associated with prognostic molecular subtype [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 2388. doi:10.1158/1538-7445.AM2017-2388