Distinct Genome-Wide Methylation Patterns in Sporadic and Hereditary Nonfunctioning Pancreatic Neuroendocrine Tumors

Cancer 2019 AI 5 Explanations View Original
Original Paper (PDF)

Unable to display PDF. Download it here or view on PMC.

Plain-English Explanations
Page [1, 2]
When Gene Silencing Drives Cancer

Pancreatic neuroendocrine tumors (PanNETs) are a distinct type of pancreatic cancer arising from the hormone-producing cells of the pancreas, different from the more common ductal adenocarcinoma. Non-functioning PanNETs (NFPanNETs) don't produce detectable hormone excess and are often found incidentally—making their biology and optimal treatment still poorly understood.

DNA methylation is a chemical modification that can effectively silence genes without changing their sequence. Abnormal methylation patterns are known to cause cancer by silencing protective tumor suppressor genes. While methylation has been studied in many cancers, genome-wide methylation data for NFPanNETs—especially comparing those that arise spontaneously versus those linked to hereditary syndromes—was almost completely absent before this study.

TL;DR: Non-functioning pancreatic neuroendocrine tumors can arise spontaneously or through inherited syndromes; how DNA methylation silences genes differently across these groups was not previously known.
Page [2, 3]
Mapping the Methylation Landscape Across 33 Tumors

Researchers analyzed 33 tissue samples including sporadic NFPanNETs (no hereditary syndrome, n=9), tumors from patients with Multiple Endocrine Neoplasia type 1 (MEN1, n=10), tumors from patients with Von Hippel-Lindau disease (VHL, n=10), and normal islet cells (n=4) as a reference.

Genome-wide CpG methylation profiling was performed using Infinium MethylationEPIC BeadChip arrays, which measure methylation at over 850,000 locations across the genome simultaneously. The data was then analyzed using R-based bioinformatics tools, including unsupervised clustering to see how samples naturally grouped, and CpG Island Methylator Phenotype (CIMP) analysis to find genes consistently silenced across subtypes.

TL;DR: Genome-wide methylation was mapped at 850,000+ sites across 33 tumor and normal tissue samples from patients with sporadic NFPanNETs, MEN1-related, and VHL-related tumors.
Pages 4-4
MEN1 Tumors Are Hypermethylated; VHL Tumors Are Different

When the samples were grouped purely by their methylation patterns, sporadic and MEN1-related NFPanNETs clustered together, while VHL-related tumors formed a separate cluster. This suggests that VHL tumors have a fundamentally different epigenetic landscape from the other two types, even though all three look similar under the microscope.

MEN1-related tumors had a significantly higher rate of hypermethylation—abnormal silencing of gene regulatory regions—compared to sporadic and VHL tumors. When the team looked at which genes were being silenced, they found that most hypermethylated genes were also downregulated in the same samples, confirming that the methylation was functionally shutting those genes off.

TL;DR: MEN1-related tumors showed much more genome-wide gene silencing through methylation than sporadic or VHL tumors, and VHL tumors had a completely distinct methylation profile.
Page [5, 6]
Specific Silenced Genes Identified in Each Subtype

CIMP analysis identified three genes with consistent methylation-driven silencing: SFRP5 in sporadic NFPanNETs, and CDCA7L and RBM47 in MEN1-related tumors. SFRP5 is a known inhibitor of the Wnt signaling pathway, which drives cell growth—its silencing would allow uncontrolled Wnt activity in sporadic tumors.

CDCA7L and RBM47 are less well-characterized but are involved in cell division and RNA processing respectively. Their silencing in MEN1 tumors may contribute to the more aggressive behavior and higher methylation burden seen in this hereditary subtype. The integration of gene expression data with methylation data confirmed that these specific genes were both methylated and turned off.

TL;DR: Three specific genes silenced by methylation were identified—SFRP5 in sporadic tumors, and CDCA7L and RBM47 in MEN1-related tumors—providing potential subtype-specific biomarkers and targets.
Page [6, 7]
Implications for Diagnosis and Treatment

The finding that sporadic, MEN1, and VHL NFPanNETs have distinct methylation patterns suggests they are molecularly distinct diseases that may require different clinical management. Methylation patterns could serve as diagnostic classifiers to help distinguish tumor subtypes without expensive genetic testing.

The high methylation burden in MEN1-related tumors raises the possibility that drugs called demethylating agents—already used in some blood cancers—might be particularly relevant for this subgroup of patients. Additionally, the silenced genes identified here are candidate biomarkers for detecting NFPanNETs non-invasively and potential targets for restoring normal gene function.

TL;DR: Distinct methylation patterns across NFPanNET subtypes suggest they are different diseases requiring different treatments; MEN1 patients in particular might benefit from demethylating drugs already in clinical use.
Citation: Open Access, 2019. Available at: PMC8237345.