DNA methylation loss promotes immune evasion of tumours with high mutation and copy number load.
Jung H, Kim HS, Kim JY, Sun JM, Ahn JS, Ahn MJ, Park K, Esteller M, Lee SH, Choi JK. · 2019
Read DNA methylation loss promotes immune evasion of tumours with high mutation and copy number load. by Jung H, Kim HS, Kim JY, Sun JM, Ahn JS, Ahn MJ, Park K, Esteller M, Lee SH, Choi JK. (2019) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc6753140.txt.
Introduction
Jung and co-authors found that loss of DNA methylation helps tumors evade the immune system. Tumors with many mutations usually respond better to immunotherapy. Yet cell division also strips methylation from large DNA domains. Immune pathway genes cluster in those domains, and they get switched off when methylation falls and other marks rise.
The team studied cohorts of 60 and 81 lung cancer patients and 40 melanoma patients. Methylation loss tracked with immune escape, apart from mutation counts. It predicted immunotherapy resistance better than mutation burden did. The same pattern held in tumors with abnormal chromosome numbers. Tuning these marks could sharpen precision immunotherapy, the authors suggest.
Catalog record
| CREATOR | Jung H, Kim HS, Kim JY, Sun JM, Ahn JS, Ahn MJ, Park K, Esteller M, Lee SH, Choi JK. |
|---|---|
| YEAR | 2019 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 61 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
DNA methylation loss promotes immune evasion of tumours with high mutation and copy number load By Hyunchul Jung, Hong Sook Kim, Jeong Yeon Kim, Jong-Mu Sun, Jin Seok Ahn, Myung-Ju Ahn, Keunchil Park, Manel Esteller, Se-Hoon Lee, Jung Kyoon Choi Published in: Nature Communications DOI: 10.1038/s41467-019-12159-9 ABSTRACT Abstract Mitotic cell division increases tumour mutation burden and copy number load, predictive markers of the clinical benefit of immunotherapy. Cell division correlates also with genomic demethylation involving methylation loss in late-replicating partial methylation domains. Here we find that immunomodulatory pathway genes are concentrated in these domains and transcriptionally repressed in demethylated tumours with CpG island promoter hypermethylation. Global methylation loss correlated with immune evasion signatures independently of mutation burden and aneuploidy. Methylome data of our cohort (n = 60) and a published cohort (n = 81) in lung cancer and a melanoma cohort (n = 40) consistently demonstrated that genomic methylation alterations counteract the contribution of high mutation burden and increase immunotherapeutic resistance. Higher predictive power was observed for methylation loss than mutation burden. We also found that genomic hypomethylation correlates with the immune escape signatures of aneuploid tumours. Hence, DNA methylation alterations implicate epigenetic modulation in precision immunotherapy. Subject terms: Cancer immunotherapy, Epigenomics, Immunogenetics INTRODUCTION Cancer immunotherapy based on checkpoint blockade has become highly effective in a subset of patients with different types of human cancers. In particular, antibody-mediated interventions targeting cytotoxic T lymphocyte antigen-4 (CTLA-4) and programmed death receptor-1 (PD-1) on T lymphocytes and the principal ligand (PD-L1) on tumour cells can reverse tumour-induced immunosuppression and induce durable clinical responses1. A major challenge facing current immunotherapies is the identification of biomarkers that predict clinical responses to CTLA-4 and PD-1/PD-L1 blockade. Overall, the mutational or neoantigen load2–5 and pre-existing T cell infiltration6,7 are indicators of clinical benefit of checkpoint blockade. On the other hand, somatic copy number alterations (SCNAs)8–10, tumour heterogeneity11, and the genetic alteration of specific genes12 or pathways13 have been identified as resistance factors. Tumour cells produce neoantigens or antigens that the immune system never encountered without cancer. The epitopes of neoantigens are displayed on the surface of cancer cells and provoke immune response. Therefore, tumours with high mutation load are more likely to respond to anti-immunosuppressive strategies based on checkpoint blockade2–5. Mutation load increases as a result of replicative errors during cell division. Not only mutations but also methylation losses accumulate during successive rounds of cell division14. Global hypomethylation and CGI hypermethylation represent the hallmark methylation changes in cancer15. Considering its association with late replication timing, progressive methylation loss may occur due to the failure of methylation maintenance machinery to remethylate newly synthesized daughter stands during DNA replication14,16. However, the influence of genomic methylation loss through cell divisions has never been investigated in the context of cancer immunotherapy. Meanwhile, SCNAs emerged as a resistance parameter8–10. A pan-cancer analysis discovered the association of SCNAs with molecular signatures of cytotoxic immune activity across diverse tumour types10. Particularly, highly aneuploid tumours with extensive chromosome- or arm-level SCNAs showed a lower expression of markers indicating infiltrating immune cells. In contrast, focal SCNAs mainly correlated with cell proliferation markers instead of immune activity signatures. However, the mechanism by which aneuploidy affects immune cell infiltration remains unknown. Global demethylation in cancer promotes chromosomal instability17–20, particularly involving large-scale alterations leading to aneuploidy21–23. Therefore, we investigated the relationships between methylation changes and aneuploidy. Here, we performed large-scale systematic analyses of the molecular data of TCGA samples across a variety of tumour types. We examined the relationships of global methylation levels with markers of cell proliferation, mutation burden, SCNA levels, markers of infiltrating immune cells, and activity of immune-response genes. Importantly, we tested our hypotheses developed from the molecular analyses by using our lung cancer cohort. This is the first study that inspected DNA methylation patterns in the molecular and clinical data with regards to cancer immunotherapy. As a result, we suggest that as an important predictive marker in immunotherapy, genomic demethylation implicates epigenetic modulation as a combination regimen for precision immunotherapy. RESULTS Global methylation correlates with immune signatures Our pan-cancer analyses of TCGA data demonstrated that markers of cell proliferation tightly correlate with mutation burden and aneuploidy across cancer types and among samples within each cancer type (Supplementary Fig. 1). Our measure of genomic demethylation based on long interspersed nuclear element-1 (LINE-1 or L1)24,25 probes (Supplementary Fig. 2) also strongly correlated with cell proliferation markers (Fig. 1a and Supplementary Data 1). Global methylation loss was also associated with an increase in mutation burden (Fig. 1b) and chromosomal SCNA load (Fig. 1c), two types of genomic aberrations that accumulate through cell division (Supplementary Fig. 5). [Figure: Correlates of global DNA methylation level. Correlation between genomic methylation levels and cell proliferation markers a, mutation burden b, aneuploidy level c, and tumour-infiltrating CD8 + T cell markers d across and within 21 cancer types. a–d The median values were obtained for each cancer type, and statistical significance was evaluated using Spearman’s correlation (upper scatterplots). Three outlier cancer types (ESCA, STAD, and UCEC) are not shown for the CD8 + T cell correlation, but included when evaluating Spearman’s correlation. For the correlation within each cancer type (lower bar graphs), Spearman’s partial correlation was used to adjust for tumour purity. Tumour types showing significant partial correlation coefficient (P < 0.05) were shaded in darker colours] Notably, we found a correlation between the global L1 methylation levels and immune signatures such as markers of tumour-infiltrating CD8 + T cells (Fig. 1d, Supplementary Data 1, and Supplementary Fig. 3). However, immune cell markers are expected to correlate with mutation burden and also are known to be associated with aneuploidy10. To disentangle this intercorrelation, we performed multiple regression of the expression level of each gene on sample-level features, namely, global L1 methylation, mutation burden, aneuploidy, tumour purity, age, and tumour stage. In this manner, we were able to determine that immune infiltrates are associated with the global methylation levels independently of mutation burden and aneuploidy when purity, age, and tumour stage are adjusted (Fig. 2a). [Figure: Genomic methylation loss correlates with immune evasion signatures. a Heatmap of Gene Set Enrichment Analysis (GSEA)48 normalized enrichment scores (NESs) for gene sets representing various types of innate and adaptive immune cells (rows). For each gene per cancer type, a linear regression model was fit using mRNA expression level as response variable and global methylation, mutation burden, aneuploidy, tumour purity, age, and tumour stage as predictors. For each of three predictors (global methylation level, mutation burden, aneuploidy level), GSEA was performed on genes with significant regression coefficients. Cells with significant NES (FDR <0.25) are colour-scaled. b Heatmap of GSEA NESs for hallmark immune and proliferation gene sets and genes involved in antigen processing and presentation, MHC, or cytokine-cytokine receptor interaction] Significant correlations with genomic demethylation were observed also for immunomodulatory pathways that should include genes expressed in tumour cells. These include antigen processing and presentation, major histocompatibility complex (MHC), cytokine–cytokine receptor interaction, interferon or other cytokine signaling, and complement and coagulation (Fig. 2b). There is an emerging role for the complement system in regulating the antitumour immune response26. The correlation of the cell proliferation markers was in the opposite direction to that of the immune cell markers or