DNA methylation in meningioma reveals a progression risk signal

DNA methylation in meningioma may contain information that is not always visible under the microscope. Two tumours with a similar histological classification can follow very different courses: one may remain controlled for years after surgery, while another returns. The new study identifies an epigenetic signature associated with that difference in behaviour.

Meningiomas arise from the membranes surrounding the brain and spinal cord and are the most common primary central nervous system tumours in adults. Most grow relatively slowly, but a subset recur and behave aggressively. Current classification incorporates morphology, CNS WHO grade and selected molecular alterations, yet tumours within the same category can still have markedly different outcomes.

The researchers profiled DNA methylation in 231 meningioma samples: 158 primary and 73 recurrent tumours. The cohort was deliberately enriched for higher-grade disease, with 55% classified as CNS WHO grade 2 and 10% as grade 3. Clinical follow-up was available for 219 tumours, with a median follow-up of 35 months.

Two epigenetic states

DNA methylation is the addition of small chemical groups to DNA. It does not change the DNA sequence itself, but it can influence whether nearby genes are active or silent.

When the researchers examined methylation across the genome, the tumours separated into two main epigenetic states. One had relatively low methylation across a particular set of CpG sites, while the other showed strong hypermethylation. The authors called these groups METHlow and METHhigh.

The distinction tracked clinical behaviour. METHhigh was more common among tumours with established high-risk features, while the degree of methylation across the signature formed a continuous gradient: increasing methylation was associated with increasing risk of progression.

In a model accounting for factors including CNS WHO grade, extent of resection, primary versus recurrent disease and radiotherapy, methylation-cluster assignment remained independently associated with progression. The hazard ratio was 2.17, with a 95% confidence interval of 1.34–3.46.

The finding was particularly relevant to CNS WHO grade 2 meningiomas, a category in which outcomes can vary considerably. The methylation signature further separated these tumours into groups with different progression risks. The overall methylation classification was also reproduced in an independent published cohort of 565 meningiomas.

Why protocadherins matter

The prognostic association was only part of the study. The authors next asked what biological processes might underlie the methylation pattern.

Most differentially methylated sites in METHhigh tumours were hypermethylated. Among 744 genes significantly enriched for hypermethylated CpG sites, clustered protocadherin genes on chromosome 5 stood out.

Protocadherins are cell-surface proteins involved in cell interactions and signalling. In aggressive meningiomas, methylation extended across large genomic regions containing multiple members of this gene family. For several protocadherins, including PCDHGC3, hypermethylation was accompanied by reduced gene expression.

The experiments then pointed to a potential downstream mechanism involving β-catenin. This protein can act as a transcriptional co-regulator when it accumulates in the nucleus.

In cell models derived from aggressive meningiomas, β-catenin was preferentially localized to the nucleus. When the researchers restored PCDHGC3 expression in a malignant meningioma cell line, β-catenin shifted toward the cytoplasm and its total cellular level decreased.

A related pattern was seen in human tumour tissue. METHlow meningiomas had an average of 5.1% tumour cells with nuclear β-catenin, compared with 11.4% in METHhigh tumours. In an additional cohort of 529 meningiomas, a high fraction of tumour cells containing nuclear β-catenin was also associated with poorer progression-free survival.

Together, the results support a model in which hypermethylation silences selected protocadherins, reduced protocadherin activity permits more β-catenin to accumulate in the nucleus, and this state is associated with more aggressive tumour behaviour. The cell experiments provide functional support for the mechanism, but they do not establish that this pathway alone determines clinical progression in patients.

What the study changes

The findings suggest that epigenetic profiling can add prognostic information beyond conventional meningioma classification. This may be particularly useful for borderline cases and CNS WHO grade 2 tumours, where standard features do not always provide a clear forecast.

The signature should not yet be viewed as a stand-alone clinical test that dictates treatment. The discovery cohort came largely from a single centre and was intentionally enriched for high-risk tumours. The longitudinal analysis included only 18 patients with matched primary and recurrent samples, and much of the mechanistic evidence involving protocadherins and β-catenin came from cell models. Larger prospective multicentre studies will be needed to establish how much the signature improves real-world clinical decision-making.

The broader result goes beyond a new prognostic marker. Aggressive meningiomas appear to differ not only in their DNA sequence but also in an epigenetic state capable of silencing entire groups of genes and altering intracellular signalling. That additional regulatory layer may help explain why meningiomas that look similar by conventional criteria can follow very different clinical trajectories.