Home > CNS & pituitary tumors > DNA methylation classification of CNS tumors

CNS & pituitary tumors

General

DNA methylation classification of CNS tumors


Chris Dampier, M.D.
P.J. Cimino, M.D., Ph.D.

Last author update: 24 July 2023
Last staff update: 7 August 2023

Copyright: 2023, PathologyOutlines.com, Inc.

PubMed Search: DNA methylation classification of CNS tumors

Chris Dampier, M.D.
P.J. Cimino, M.D., Ph.D.
Page views in 2023 to date: 270
Table of Contents
Definition / general | Essential features | Terminology | Technologies for measuring DNA methylation | Utility of whole genome DNA methylation for CNS tumors | Limitations and potential pitfalls of methylation profiling | Molecular / cytogenetics images | Sample pathology report | Additional references | Board review style question #1 | Board review style answer #1 | Board review style question #2 | Board review style answer #2
Cite this page: Dampier C, Cimino PJ. DNA methylation classification of CNS tumors. PathologyOutlines.com website. https://www.pathologyoutlines.com/topic/CNStumorDNAmethylationclass.html. Accessed September 21st, 2023.
Definition / general
Essential features
  • DNA methylation is a type of epigenetic modification in which a covalent bond in a nucleotide is enzymatically altered to produce a subtly different compound
  • A common example and one of the best studied is cytosine converted to 5 methylcytosine (5mC); this is when a DNA methyl transferase enzyme replaces the hydrogen attached to the carbon atom in position 5 of the cytosine molecule with a methyl group
  • In mammalian genomes, methylation typically occurs at the 5 prime cytosine of a cytosine guanine dinucleotide, which is connected by a phosphodiester bond (CpG) (Trends Genet 2022;38:676)
Terminology
  • DNA methylation classification
  • Genome wide DNA methylation
  • Whole genome DNA methylation
  • DNA methylation
  • Methylation classifier
  • DNA methylation profiling
Technologies for measuring DNA methylation
  • Most contemporary technologies for DNA methylation profiling rely on bisulfite conversion of DNA extracted from formalin fixed paraffin embedded tissue
  • Bisulfite treatment of DNA selectively deaminates cytosine and converts it to uracil but leaves 5 methylcytosine unaffected (Nucleic Acids Res 1980;8:4777, Proc Natl Acad Sci U S A 1992;89:1827)
  • Resultant bisulfite treated DNA can then be amplified by polymerase chain reaction, which amplifies uracil and thymine as thymine and 5 methylcytosine as cytosine (Proc Natl Acad Sci U S A 1992;89:1827)
  • Cytosine residues in the PCR products are therefore markers of 5 methylcytosine in the original input DNA (Proc Natl Acad Sci U S A 1992;89:1827)
  • As an alternative to the amplification and sequencing approach, the similar hybridization behavior of uracil and thymine can be exploited to detect cytosine to uracil / thymine polymorphisms in bisulfite treated DNA using DNA hybridization technology (Genome Res 2006;16:383)
  • Bisulfite sequencing
    • Bisulfite treated and PCR amplified DNA can be sequenced by, for example, massively parallel short read sequencing
    • All cytosines in the output sequences are interpreted as 5mC, so genome wide DNA methylation can be identified at nucleotide resolution (Proc Natl Acad Sci U S A 1992;89:1827, Nucleic Acids Res 2001;29:E65)
    • Genome wide bisulfite sequencing is still prohibitively expensive for large scale implementation
  • Early whole genome DNA methylation arrays
    • Illumina GoldenGate DNA Methylation BeadArray; first generation genome wide DNA methylation array, introduced in 2006 (Genome Res 2006;16:383)
    • Illumina Infinium HumanMethylation27K BeadChip; second generation genome wide DNA methylation array, introduced in 2009 (Epigenomics 2009;1:177)
    • Illumina Infinium HumanMethylation450K BeadChip; third generation genome wide DNA methylation array, introduced in 2011 (Genomics 2011;98:288, Epigenetics 2011;6:692)
  • Current whole genome DNA methylation array
    • Illumina Infinium MethylationEPIC BeadChip; fourth generation genome wide DNA methylation array, introduced in 2015
    • Version 1.0 includes 853,307 CpG sites, including 439,562 of the 482,421 CpGs (91%) included in the 450K microarray (Epigenomics 2016;8:389, Nucleic Acids Res 2017;45:e22)
    • Adds 333,265 CpGs located in enhancer regions identified by ENCODE and FANTOM5 (Epigenomics 2016;8:389)
    • Version 2.0 includes more than 935,000 CpG sites
    • Over 186,000 new probes to target known enhancers, super enhancers, CTCF binding domains and open regions of chromatin associated with primary tumors identified by ATAC seq and ChIP seq experiments
    • DNA input requirement: 250 ng
Utility of whole genome DNA methylation for CNS tumors
Limitations and potential pitfalls of methylation profiling
  • Like other studies, methylation profiling should be considered in conjunction with other histopathologic, genetic and clinical data in order to reach a diagnosis
  • Misleading methylation profiles can occur (i.e., methylation matches that apparently conflict with histologic, genetic, radiographic or clinical features)
  • Requires a relatively large amount of optimal input DNA (~250 ng)
  • Some methylation classes do not necessarily correspond to a CNS WHO tumor type
  • High confidence methylation classifier score may not be seen in up to 1/3 of cases (Neuro Oncol 2022;24:571, Clin Epigenetics 2019;11:185)
  • Factors that may contribute to low classifier score (no methylation match)
    • Low tumor cell content (low purity of sample)
    • Failed DNA bisulfite conversion
    • Poor quality DNA (e.g., old tissue blocks)
    • Single bioinformatic pipelines are reliant on a single reference dataset
    • Rare tumor types may not be present in reference dataset, so no class exists for them
Molecular / cytogenetics images

Contributed by P.J. Cimino, M.D., Ph.D. and Chris Dampier, M.D.
Methylation copy number plot

Methylation copy number plot

Dimensionality reduction of medulloblastoma

Dimensionality reduction of medulloblastoma



Images hosted on other servers:
Methylation based CNS tumor clustering

Methylation based CNS tumor clustering

Impact of tumor purity

Impact of tumor purity

Workflow in surgical neuropathology

Workflow in surgical neuropathology

Heterogeneity of histologic diagnoses

Heterogeneity of histologic diagnoses

Sample pathology report
  • Brain, cervical spinal cord, biopsy:
    • Histological classification: glial neoplasm with focal ependymal features
    • CNS WHO grade 2
    • Consensus methylation profiling class: spinal ependymoma
    • Integrated diagnosis: spinal ependymoma, CNS WHO grade 2
Board review style question #1
Brq image


A 7 year old girl undergoes stereotactic needle biopsy for diagnosis of an intracranial intrinsic right parietal lobe mass. There is scant tissue in the biopsy and nucleic acid extraction is expected to yield material for only a single genome wide assay (e.g., cytogenomic array, next generation sequencing, DNA methylation array). Which of these features are specifically detected by DNA methylation array profiling?

  1. Chromosomal copy number alterations
  2. Genomic breakpoints to suggest gene fusions
  3. Methylation class
  4. MGMT promoter methylation status
  5. Single gene mutations (e.g., specific H3 G34 mutation)
Board review style answer #1
C. Methylation class is specifically detected by DNA methylation array profiling. Answer D is incorrect because MGMT promoter methylation status can also be detected by methylation specific PCR. Answers A and B are incorrect because chromosomal copy number alterations and genomic breakpoints to suggest gene fusions can also be detected by other modalities such as cytogenetic microarrays, next generation sequencing and fluorescence in situ hybridization. Answer E is incorrect because single gene mutations may be inferred by DNA methylation array but the DNA sequence is not directly detectable.

Comment Here

Reference: DNA methylation classification of CNS tumors
Board review style question #2
A 5 year old boy undergoes resection of a medulloblastoma from the posterior fossa. DNA methylation profiling of the tumor can determine which of the following?

  1. Medulloblastoma histologic subtype
  2. Medulloblastoma molecular subtype
  3. Mitotic count
  4. Patient age
  5. TP53 mutational status in SHH activated medulloblastoma
Board review style answer #2
B. Medulloblastoma molecular subtype. DNA methylation profiling is useful for determining medulloblastoma molecular subtype. Answer D is incorrect because patient age is not discernable by methylation profiling. Answer E is incorrect because TP53 status requires DNA sequencing, not methylation. Answers A and C are incorrect because medulloblastoma histologic subtype and mitotic counts are determined by microscopic examination, not methylation.

Comment Here

Reference: DNA methylation classification of CNS tumors
Back to top
Home > CNS & pituitary tumors > DNA methylation classification of CNS tumors
Image 01 Image 02