Abstract Background: Medulloblastoma is the most common malignant brain tumor diagnosed in children. Treatments for medulloblastoma are associated with neurocognitive impairment; however, there is little known about the biological factors which influence variability in neurocognitive outcomes. There is emerging evidence that epigenetic age acceleration (EAA) – the difference between DNA methylation-based predicted age and chronological age – is associated with chronic health conditions among pediatric cancer survivors. Therefore, we sought to evaluate the association between EAA and neurocognitive outcomes in pediatric medulloblastoma survivors. Methods: Neurocognitive tests were obtained between 2003 and 2012 from >1-year survivors of pediatric medulloblastoma (n=49). Assessments included the following: Full Scale Intelligence Quotient (FSIQ), Working Memory Index (WMI), and Processing Speed Index (PSI). Whole blood samples were obtained, and DNA was extracted for genome-wide DNA methylation using the Illumina HumanMethylation450K chip. DNA methylation age was calculated using the Horvath DNA Methylation Age Calculator. Associations were assessed using linear regression models with statistical significance defined at p<0.05. Results: Patients were predominantly male (n=36, 73.5%), non-Hispanic White (n=34, 69%) and diagnosed at a mean age of 7.5 years. Patients received intensity-modulated radiation (n=39, 80%) or proton beam radiation (n=10, 20%). The mean FSIQ was 82.2 (95% CI: 76.9-87.6), WMI was 82.2 (95% CI: 76.2-88.2), and PSI was 79.5 (95% CI: 73.5-85.6) a median of 3.03 years (range: 1.0 – 15.2 years) post-radiotherapy. On average, the DNA methylation age of participants was significantly higher than their chronologic age at the time of sample collection (mean difference = 2.4 years; 95% CI: 1.2-3.5; p <0.01). Time from diagnosis to sample collection was significantly (p <0.01) associated with a larger deviation between DNA methylation and chronologic age. While not significantly associated with poorer neurocognitive performance on FSIQ or PSI in unadjusted models, EAA was associated with poorer WMI (beta coefficient = -1.99, p = 0.03), which was slightly attenuated after accounting for age at radiation therapy, radiation modality, dose, and time post-radiation (beta coefficient = -1.24, p = 0.15). Conclusions: In this pilot study of pediatric medulloblastoma survivors, we report that the biologic age assessed by DNA methylation profiles often exceeds the chronologic age of survivors. We also identified suggestive associations between EAA and the degree of neurocognitive impairment in survivors of medulloblastoma. These findings highlight the potential utility of evaluating measures of biological aging to better understand variability in neurocognitive performance in larger, longitudinal cohorts of patients with pediatric medulloblastoma. Citation Format: Kevin O. Wilhelm, Philip J. Lupo, Kimberly P. Raghubar, Lisa S. Kahalley, Murali Chintagumpala, Mehmet F. Ocku, Michael E. Scheurer, Austin L. Brown. A pilot study of epigenetic age acceleration and neurocognitive outcomes among survivors of pediatric medulloblastoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 3032.
Read full abstract