Reliable tumor detection by whole-genome methylation sequencing of cell-free DNA in cerebrospinal fluid of pediatric medulloblastoma
Jia Li1, Sibo Zhao2,3, Minjung Lee1, Yue Yin1, Jin Li1, Yubin Zhou4, Leomar Y. Ballester5,6, Yoshua Esquenazi6, Roderick H. Dashwood1, Peter J.A. Davies4, D. William Parsons2, Xiao-Nan Li2,7,8#, Yun Huang1,8#, Deqiang Sun1,8,#
Brain tumors are the leading cause of cancer-related death in children, with medulloblastoma (MB) being the most common form of pediatric brain malignancy. Mutational analysis of circulating cell-free tumor DNA (ctDNA) by liquid biopsy offers a new approach for real-time monitoring of tumor status without subjecting patients to direct tumor biopsies. Unfortunately, pediatric brain tumors, in particular MB, are known for having relatively low frequencies of oncogenic DNA mutations compared with adult cancers, making the detection of pediatric tumor DNA via mutational analysis difficult. However, epigenetic markers, such as DNA methylation modifications, are pronouncedly different in pediatric MB. In this study, we systematically profiled the DNA methylation and hydroxymethylation landscapes of MB tumors and matched ctDNA isolated from cerebrospinal fluid (CSF). We found that the dysregulation of DNA methylation in the MB tumor was well preserved in the ctDNA. Specifically, we identified 6,598 CpGs that were consistently differentially methylated in both the tumor and the CSF. The methylation status of these CpGs (i.e., MB methylation signatures in CSF) in the initial CSF collection could be used to detect the existence of the MB tumor and determine the subtype, and serial CSF collections could be used to monitor the treatment response and tumor recurrence. We also identified MB DNA methylation signatures in ctDNA that have potential diagnostic and prognostic value. Our study not only provides an atlas of the DNA methylome and hydroxymethylome in CSF-derived MB ctDNA but also sets the stage for exploiting epigenetic markers in liquid biopsy to aid the clinical management of MB in patients.
Genome Browser datahub: https://220.127.116.11/data/jiali/CSF/MB_Hub/hub.txt