Peripheral Blood DNA Methylation as Potential Biomarker of Malignant Pleural Mesothelioma in Asbestos-Exposed Subjects

Our plain-English summary of Peripheral Blood DNA Methylation as Potential Biomarker of Malignant Pleural Mesothelioma in Asbestos-Exposed Subjects, by S. Guarrera, C. Viberti, G. Cugliari, et al (Journal of Thoracic Oncology, 2019).

The study

• Authors: S. Guarrera, C. Viberti, G. Cugliari, A. Allione, E. Casalone, M. Betti, D. Ferrante, A. Aspesi, C. Casadio, F. Grosso, R. Libener, E. Piccolini, D. Mirabelli, I. Dianzani, C. Magnani, G. Matullo
• Journal: Journal of Thoracic Oncology
• Published: 2019;14(3):527-539
• Cited by: 29 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Introduction: Malignant pleural mesothelioma (MPM) is an aggressive tumor strongly associated with asbestos exposure. Patients are usually diagnosed when current treatments have limited benefits, highlighting the need for noninvasive early diagnostic tests to monitor asbestos-exposed people.

Methods: We used a genome-wide methylation array to identify, in asbestos-exposed subjects, novel blood DNA methylation markers of MPM in 163 MPM cases and 137 cancer-free controls (82 MPM cases and 68 controls, training set; replication in 81 MPM cases and 69 controls, test set) sampled from the same areas.

Results: Evidence of differential methylation between MPM cases and controls was found (more than 800 cytosine-guanine dinucleotide sites, false discovery rate p value (pfdr) < 0.05), mainly in immune system-related genes. Considering the top differentially methylated signals, seven single- cytosine-guanine dinucleotides and five genomic regions of coordinated methylation replicated with similar effect size in the test set (pfdr < 0.05). The top hypomethylated single-CpG (cases versus controls effect size less than -0.15, pfdr < 0.05 in both the training and test sets) was detected in FOXK1 (Forkhead-box K1) gene, an interactor of BAP1 which was found mutated in MPM tissue and as germline mutation in familial MPM. In the test set, comparison of receiver operating characteristic curves and the area under the curve (AUC) of two models, including or excluding methylation, showed a significant increase in case/control discrimination when considering DNA methylation together with asbestos exposure (AUC = 0.81 versus AUC = 0.89, DeLong's test p = 0.0013).

Conclusions: We identified signatures of differential methylation in DNA from whole blood between asbestos exposed MPM cases and controls. Our results provide the rationale to further investigate, in prospective studies, the potential use of blood DNA methylation profiles for the identification of early changes related to the MPM carcinogenic process.

Copyright © 2018 International Association for the Study of Lung Cancer. Published by Elsevier Inc. All rights reserved.

In plain English

This study compared DNA methylation, a type of chemical mark on DNA, in blood from 163 people with malignant pleural mesothelioma and 137 people without it, recruited in Casale Monferrato and Turin, Italy. Each person's asbestos exposure was estimated from interviews. Subjects were split at random into a training group (82 cases, 68 controls) and a test group (81 cases, 69 controls). Cases had significantly higher cumulative asbestos doses than controls, and analyses were adjusted for exposure, sex and age. In the training group, 884 DNA sites differed significantly between cases and controls. Of these, 868 also differed in the test group, and immune-system pathways were over-represented among them. Seven top sites met strict replication criteria. The most reduced site was in the FOXK1 gene. In the test group, adding methylation to a model with age, sex and asbestos exposure raised the area under the curve, a measure of how well cases and controls were told apart, from 0.81 to 0.89 (p = 0.0013). Because cases already had the disease when their blood was drawn, the authors could not tell whether these changes are part of the cancer process or a consequence of it. They concluded that prospective studies are warranted.

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: Cross-sectional comparison of mesothelioma cases and controls, using a genome-wide blood DNA methylation array, with random split into a training (discovery) set and a test (replication) set
• Who was studied: 300 people: 163 malignant pleural mesothelioma cases and 137 non-mesothelioma controls, enrolled in Casale Monferrato (an area with high occupational and environmental asbestos exposure) or Turin, Piedmont, Italy. Training set: 82 cases and 68 controls. Test set: 81 cases and 69 controls. Enrollment dates are not given in the text.
• Exposure: Cumulative asbestos exposure (fibers/mL-years) assessed by an occupational epidemiologist from interview data. Sources were occupational, domestic (sharing a house with an asbestos worker, or asbestos-containing tools or materials in the house) and environmental. Fiber types are not specified. Cases had significantly higher cumulative doses than controls. Mean dose was 34.8 in training-set cases versus 5.11 in training-set controls, and 62.98 in test-set cases versus 2.13 in test-set controls.
• Main results:
  • In the training set, 884 DNA sites (CpGs) had effect size of at least 0.10 and false discovery rate p less than 0.05: 13 were more methylated in cases and 871 less methylated. Of these 884, 868 also showed significant differences in the test set, with significant enrichment for neutrophil degranulation, innate immune system and immune system pathways.
  • Seven top sites replicated under strict criteria. Three were less methylated in cases (genes FOXK1, MYB, TAF4) and four were more methylated in cases (CXCR6/FYCO1, TAP1, MORC2, LIME1).
  • Regional analysis: 35 regions were less methylated in training-set cases, and all 35 were also significantly less methylated in test-set cases. 21 regions in 21 genes met strict replication criteria. The broadest cluster was nine sites in the FOXK1 gene.
  • In the test set, how well cases and controls were told apart (area under the curve) was 0.81 for a model with age, sex and asbestos exposure, and 0.89 when methylation of the replicating sites was added (DeLong's test p = 0.0013).
  • Estimated white blood cell make-up differed significantly between cases and controls in both sets. Cases had lower estimated CD4 and CD8 T cell, NK cell and B cell percentages, and higher granulocyte percentages.
  • Methylation-based age acceleration did not differ significantly between cases and controls for the main index (p = 0.50) or the IEAA index (p = 0.49). The EEAA index differed significantly (Wilcoxon p = 0.007), which the authors interpreted as cases' white blood cells being epigenetically older.
  • In separate healthy-pleura samples from other studies, methylation at the FYCO1 site and at eight of nine FOXK1 cluster sites correlated significantly with gene expression. No significant correlation was found in blood cells from 72 healthy subjects.
• The authors' conclusion: The authors concluded that they identified differences in whole-blood DNA methylation between asbestos-exposed mesothelioma cases and controls. In their view, these results provide the rationale to investigate, in prospective studies, whether blood methylation profiles could identify early changes related to mesothelioma development.
• Limitations the authors state:
  • All cases already had mesothelioma when blood was taken. The findings may reflect disease status, and the authors could not exclude reverse causality or tell whether the changes are part of the cancer process or a consequence of the disease.
  • No mesothelioma tissue was available from the same subjects, so blood methylation could not be compared with tumor tissue or pleural effusions.
  • Only DNA was collected, so gene expression could not be measured in the cases and controls themselves.
  • Cases and controls differed significantly in asbestos exposure. The authors called this a critical issue and adjusted their analyses for individual exposure levels.

Topics

The National Library of Medicine indexes this study under subjects our site covers: Lung cancer, Mesothelioma.

Medical terms in this study, explained in our Lexicon: Biomarker, Pleural Mesothelioma, Asbestos Exposure, Discovery, BAP1, Germline Mutation, Carcinogenic.

Related studies

• The value of SHOX2 methylation test in peripheral blood samples used for the differential diagnosis of lung cancer and other lung disorders — M. Konecny, J. Markus, et al. (2016)
• Identification of miRNA-103 in the cellular fraction of human peripheral blood as a potential biomarker for malignant mesothelioma--a pilot study — D.G. Weber, G. Johnen, et al. (2012)
• DNA methylation profiling of asbestos-treated MeT5A cell line reveals novel pathways implicated in asbestos response — E. Casalone, A. Allione, et al. (2018)
• Asbestos-associated genome-wide DNA methylation changes in lung cancer — E. Kettunen, H. Hernandez-Vargas, et al. (2017)
• DNA methylation profile of 28 potential marker loci in malignant mesothelioma — J.A. Tsou, J.S. Galler, et al. (2007)
• New DNA Methylation Signals for Malignant Pleural Mesothelioma Risk Assessment — G. Cugliari, A. Allione, et al. (2021)

Browse more research in Journal of Thoracic Oncology, or return to medical research.