Can CpG methylation serve as surrogate markers for immune infiltration in cancer?

Our plain-English summary of Can CpG methylation serve as surrogate markers for immune infiltration in cancer?, by M.D. Bacolod, F. Barany, P.B. Fisher (Advances in Cancer Research, 2019).

The study

• Authors: M.D. Bacolod, F. Barany, P.B. Fisher
• Journal: Advances in Cancer Research
• Published: 2019;143:351-384
• Type of article: Review
• Cited by: 22 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Recent reports describe how genome-wide transcriptional analysis of cancer tissues can be exploited to identify molecular signatures of immune infiltration in cancer. We hypothesize that immune infiltration in cancer may also be defined by changes in certain epigenetic signatures. In this context, a primary objective is to identify site-specific CpG markers whose levels of methylation may be highly indicative of known transcriptional markers of immune infiltration such as GZMA, PRF1, T cell receptor genes, PDCD1, and CTLA4. This has been accomplished by integrating genome-wide transcriptional expression and methylation data for different types of cancer (melanoma, kidney cancers, lung cancers, gliomas, head and neck cancer). Our findings establish that cancers of related histology also have a high degree of similarity in immune-infiltration CpG markers. For example, the epigenetic immune infiltration signatures in lung adenocarcinoma (LUAD), mesothelioma (MESO), lung squamous cell carcinoma (LUSC), and head and neck squamous cell carcinoma (HNSC) are distinctly similar. So are glioblastoma multiforme (GBM) and brain lower grade glioma (LGG); and kidney renal papillary cell carcinoma (KIRP) and kidney renal clear cell carcinoma (KIRC). Kidney chromophobe (KICH), on the other hand has markers that are unique to this cohort. The strong relationships between immune infiltration and CpG methylation (for certain sites) in cancer tissues were not observed upon integrated analysis of publicly available cancer cell line datasets. Results from comparative pathways analyses offer further justification to methylation at certain CpG sites as being indicators of cancer immune infiltration, and possibly of predicting patient response to immunotherapeutic drugs. Achieving this target objective would significantly enhance therapeutic outcomes employing immunotherapy through focused patient-centric personalized medicine.

© 2019 Elsevier Inc. All rights reserved.

In plain English

This was a computer analysis of public data on DNA methylation and gene activity in tumors. It did not study asbestos exposure. The authors used tumor tissue data from 11 cancer types, including mesothelioma, lung, kidney, brain, and head and neck cancers, and melanoma. They looked for CpG sites, DNA spots whose methylation (a chemical tag) tracked activity of two genes, GZMA and PRF1. Their combined expression is used as a measure of the cancer-killing capability of certain immune cells. Sites near the DIABLO, NMRAL1 and NCOR2 genes were correlated in several cancer types. Mesothelioma markers resembled those for lung, head and neck, two kidney cancers and melanoma. Kidney chromophobe markers were distinct. These sites also tracked T cell receptor genes and the checkpoint genes CTLA4 and PDCD1. In melanoma, kidney, lung, glioma and head and neck cancer cell lines, methylation at these three sites did not correlate highly with GZMA and PRF1 expression. The authors concluded that methylation at specific CpG sites may be another option for assessing immune infiltration and may possibly predict response to checkpoint inhibitors. They said the findings need validation in tumor tissue samples. Two authors reported owning equity in AcuamarkDx.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Bioinformatic analysis of publicly available tumor tissue and cancer cell line genomic datasets (DNA methylation and mRNA expression), using correlation and pathway analyses
• Who was studied: Tumor tissue samples from 11 TCGA cancer cohorts: glioblastoma, head and neck squamous cell carcinoma, three kidney cancers, lower grade glioma, lung adenocarcinoma, lung squamous cell carcinoma, mesothelioma, pheochromocytoma and paraganglioma, and skin melanoma. The article does not give patient numbers per cohort. Also included were 1036 cancer cell lines (gene expression) and 1001 cancer cell lines (methylation).
• Exposure: No asbestos or other environmental exposure was studied. The authors compared methylation at each CpG site on the Illumina 450K array with the average expression of the immune genes GZMA and PRF1, then with T cell receptor genes, CTLA4 and PDCD1.
• Main results:
  • Certain CpG sites recurred across several cancer types. Sites at NCOR2 (cg05948411), GPR56 and GRB7 were positively correlated with immune cytolytic activity. Sites at DIABLO, NMRAL1, TBCD and IRF1 were negatively correlated.
  • Potential markers were highly similar among melanoma, two kidney cancers (KIRC, KIRP), lung adenocarcinoma, mesothelioma, lung squamous cell carcinoma, and head and neck cancer.
  • In the mesothelioma (MESO) cohort, the CpG site most positively correlated with GZMA and PRF1 expression was cg14479136 (LOC26102, GPSM1), with R of 0.792.
  • Markers for the kidney cancers KIRC and KIRP were highly concordant (correlation of their R values 0.968). The values for KIRC versus KICH and KIRP versus KICH were 0.77 and 0.75.
  • The brain tumor cohorts GBM and LGG had markers highly similar to each other but distinct from the other solid tumors.
  • The DIABLO site cg14094409 showed similar correlations with T cell receptor genes, PDCD1 and CTLA4 in seven cohorts, including mesothelioma.
  • In cancer cell lines classified as melanoma, kidney cancer, non-small cell lung carcinoma, glioma, and head and neck cancer, methylation at the DIABLO, NMRAL1 and NCOR2 sites did not correlate highly with GZMA and PRF1 expression.
  • Pathway analysis in selected kidney, melanoma and glioblastoma subsets found that pathways linked to these methylation levels were mostly related to T cell infiltration and activation, B cell infiltration, cytokines and inflammation.
  • Methylation at the DIABLO site was not highly correlated with DIABLO's own gene expression in several cohorts.
• The authors' conclusion: The authors concluded that measuring methylation at specific CpG sites may be another option for assessing immune infiltration in tumors and may possibly predict response to checkpoint inhibitors.
• Limitations the authors state:
  • Not every marker identified by computer analysis can become a clinically reliable assay, and the findings need validation in tumor tissue samples.
  • For some marker sites, methylation did not track the gene's own expression, which makes the biological role of those genes in immune infiltration hard to explain.

Topics

Medical terms in this study, explained in our Lexicon: Lung Cancer, Histology, Lung Adenocarcinoma, Mesothelioma, Squamous Cell Carcinoma, Carcinoma, Immunotherapy.

Related studies

• The relationship between tumor MSLN methylation and serum mesothelin (SMRP) in mesothelioma — H.H. Nelson, L.M. Almquist, et al. (2011)
• Priming the tumor immune microenvironment with chemo(radio)therapy: A systematic review across tumor types — T. van den Ende, H.G. van den Boorn, et al. (2020)
• Analysis of DNA methylation patterns in the tumor immune microenvironment of metastatic melanoma — S. Mitra, M. Lauss, et al. (2020)
• DNA methylation profile of 28 potential marker loci in malignant mesothelioma — J.A. Tsou, J.S. Galler, et al. (2007)
• Asbestos-associated genome-wide DNA methylation changes in lung cancer — E. Kettunen, H. Hernandez-Vargas, et al. (2017)
• DNA methylation status of REIC/Dkk-3 gene in human malignancies — T. Hayashi, H. Asano, et al. (2012)

Browse return to medical research.