Access the full text.
Sign up today, get DeepDyve free for 14 days.
short review memo (2021) 14:150–153 https://doi.org/10.1007/s12254-021-00696-3 DNA methylation as predictive marker of response to immunotherapy? Gerwin Heller Received: 20 January 2021 / Accepted: 24 February 2021 / Published online: 31 March 2021 © The Author(s) 2021 Summary Immunotherapy is one of the major break- Introduction throughs in cancer treatment. However, many pa- tients do not beneﬁt from this type of therapy. Thus, Inhibition of immune checkpoint molecules—inclu- there is an urgent need for a strategy to predict treat- ding cytotoxic T-lymphocyte-associated protein 4 ment efﬁcacy before start of therapy. The role of cer- (CTLA4), programmed cell death protein 1 (PD-1) tain genetic and epigenetic factors as potential pre- and programmed cell death ligand 1 (PD-L1)—has dictive markers for response to immunotherapy is dis- become a promising treatment option for a number cussed in this short review. of advanced cancers over the past few years [1–6]. However, a signiﬁcant proportion of patients will not Keywords Epigenetics · Biomarkers · Prediction · beneﬁt from this type of therapy. Thus, there is huge Oncology · Immune checkpoint inhibitors interest to identify molecular changes which may be used as markers to predict tumor response to immune Abbreviations checkpoint inhibitors (ICI). CGI CpG island Based on the mechanism of PD-1 and PD-L1 inter- CpG Cytosine-guanine action, large effort has been taken to investigate the CTLA4 Cytotoxic T-lymphocyte-associated protein 4 impact of PD-L1 expression on the efﬁcacy of PD-1 FFPE Formalin-ﬁxed parafﬁn-embedded inhibition . Davis et al. retrospectively analyzed FOXP1 Forkhead box P1 data from 45 clinical trials of anti-PD-1 or anti-PD- ICI Immune checkpoint inhibition L1 antibodies in 15 tumor entities and reported a pre- MMR DNA mismatch repair dictive value of PD-L1 in 28.9% of these studies (not MSI Microsatellite instability predictive in 53.3% and not tested in 17.8%) . Thus, NSCLC Non-small cell lung cancer although it seems that PD-L1 expression has some ORR Overall response rate impact on the patients’ outcome, it is imperfect as OS Overall survival a reliable marker to predict tumor response to ICI, PD1 Programmed cell death protein 1 and additional markers or marker combinations are PD-L1 Programmed cell death 1 ligand 1 needed. PFS Progression-free survival It is well known that human tumors may harbor SCLC Smallcelllungcancer a large number of somatic mutations with varying TMB Tumor mutational burden frequencies between tumor entities . Nonsynony- mous somatic mutations lead to altered amino acid sequences of proteins and formation of neoantigens. Thus, tumors with higher tumor mutational burden (TMB) carry a larger number of neoantigens which, in theory, increases their immunogenicity and their G. Heller, PhD, MSc. () responsiveness to immunotherapy . Indeed, as- Division of Oncology, Department of Medicine I, sociations between high TMB and response to ICI Medical University of Vienna, Waehringer Guertel were reported in various cancer types including non- 18–20, 1090 Vienna, Austria firstname.lastname@example.org small cell lung cancer (NSCLC), small cell lung cancer 150 DNA methylation and immunotherapies K short review (SCLC), melanoma and urothelial carcinoma [10–13]. islands (CGI). In humans, these regions are found However, other studies revealed no clear correlation in approximately 60% of gene promoter regions and between TMB and response to ICI [14, 15]. Thus, al- less frequently in gene bodies or in intergenic regions though the potential of TMB as predictive marker for . While 70–80% of all non-CGI CpG dinucleotides response to ICI is high, additional studies focused on in the human genome are methylated, cytosines of the deﬁnition of a predictive TMB cut-off, the estab- CpG dinucleotides in CGIs usually remain unmethy- lishment of a sequencing strategy to generate compa- lated. Exceptions are CGIs associated with imprinted, rable TMB detection across different laboratories and X-linked and tissue-speciﬁc expressed genes . combination of TMB and other potential markers are DNA methylation is considered important in the needed to further implement the use of TMB in rou- pathogenesis of many solid tumors as well as hema- tine practice. tological malignancies [19, 23]. CGIs of various can- Microsatellite instability (MSI) is a rare event cer-related genes are frequently methylated in cancer in most solid tumor types (except colorectal and cells, resulting in transcriptional inactivation of these endometrial carcinomas) and may be caused by genes [19, 24]. Of note, altered DNA methylation in germline or somatic inactivation of genes involved regions outside CpG islands may be equally important in the DNA mismatch repair (MMR) pathway . in tumorigenesis, with hypomethylation as relevant as MSI represents a marker for predicting response to hypermethylation . ICI, probably explained by increased expression of Several studies revealed that DNA methylation pro- neoantigens caused by high genomic instability. In ﬁling may serve as novel tool in oncology for im- this context pembrolizumab is FDA approved for pa- proved classiﬁcation and differential diagnosis of car- tients with MSI-H/deﬁcient MMR metastatic or unre- cinomas, especially brain tumors and sarcomas [26, sectable solid tumors, and nivolumab is approved for 27]. Moreover, the potential of DNA methylation pat- patients with MSI-H/deﬁcient MMR metastatic col- tern as prognostic or predictive marker of response to orectal cancer [17, 18]. However, a variation in overall speciﬁc therapies including chemotherapy, targeted response rates (ORR) between tumor entities was ob- therapy and immunotherapy in several tumor types served. While patients with endometrial cancer had was reported previously [28–30]. an ORR of 57.1%, patients with pancreatic cancer and central nervous system (CNS) cancers had ORRs of DNA methylation as potential biomarker for 18.2% and 0%, respectively, suggesting that MSI may immunotherapy serve as a tumor-type-speciﬁc predictive marker . Besides genetic alterations epigenetic changes fre- DNA methylation in the human genome can be com- quently occur in almost every type of cancer . prehensively proﬁled using high throughput assays While the contribution of epigenetic aberrations to based either on microarrays (e.g., Inﬁnium Human- cancer development and progression was extensively Methylation450 BeadChip, MethylationEPIC Bead- studied over the past decade, knowledge about their Chip; Illumina, San Diego, CA, USA) or next genera- role as predictive markers for ICI response is very lim- tion sequencing (e.g., reduced representaion bisulﬁte ited and will be outlined here. sequencing). MethylationEPIC beadChips are the lat- est generation of Illumina’s beadarrays and are used to quantitatively analyze methylation of more than What is epigenetics? 850,000 methylation sites across the genome at single- Epigenetic mechanisms including DNA methylation nucleotide resolution. This approach is not restricted and chemical modiﬁcations of histone proteins like to fresh tissue samples but can also be applied to methylation, phosphorylation, ubiquitination, acety- formalin-ﬁxed parafﬁn-embedded (FFPE) tissue. In lation and ADP-ribosylation signiﬁcantly contribute a multicenter study, Duruisseaux et al. used to the regulation of transcriptional gene activity. this technology to analyze the methylome of stage These modiﬁcations are key events which affect tran- IV NSCLC patients who were treated with anti-PD- scription factor binding to DNA and change the chro- 1-ICI during the course of their disease. In a ﬁrst matin structure resulting either in gene activation step, the authors analyzed tumors from 34 NSCLC pa- or gene silencing . Epigenetic gene regulation is tients before they received anti-PD1-ICI. Ten of these important for the regulation of many biological pro- patients were classiﬁed as responders to this type cesses, including embryogenesis, genomic imprinting of therapy and 24 patients were classiﬁed as nonre- or X chromosome inactivation [19, 20]. sponders. Biostatistical analyses of the microarray DNA methylation is the covalent addition of a data revealed a signature (referred to as EPIMMUNE methyl group (–CH )to the 5 carbon of cytosine signature) of 301 differentially methylated CpG sites bases within cytosine–guanine (CpG) dinucleotides. (81% hypermethylated; 19% hypomethylated; asso- In the mammalian genome the CpG dinucleotide is ciated with 174 unique genes) between responders generally underrepresented . However, certain re- and nonresponders. Pathway enrichment analyses gions of the genome (0.5–4 kb in length) contain CpG revealed that some of these genes are involved in dinucleotides at a high density and are called CpG DNA repair, β-catenin signaling and interferon-γ sig- K DNA methylation and immunotherapies 151 short review naling. The EPIMMUNE signature was signiﬁcantly Funding Open access funding provided by Medical University associated with progression-free survival (PFS) and of Vienna. overall survival (OS) suggesting that EPIMMUNE may Conﬂict of interest G. Heller received research funding from be a good predictor of anti-PD-1-ICI response . Bristol Myers Squibb and Merck MSD. Neither PD-L1 expression nor TMB were associated with PFS or OS in this patient cohort. In a second Open Access This article is licensed under a Creative Com- step, the authors performed similar experiments in mons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in an independent cohort of 47 patients with advanced any medium or format, as long as you give appropriate credit NSCLC. Again, the EPIMMUNE signature was as- to the original author(s) and the source, provide a link to sociated with PFS and OS supporting the predictive the Creative Commons licence, and indicate if changes were potential of this methylation signature . Finally, made. The images or other third party material in this article to increase the usability of their ﬁndings in the future are included in the article’s Creative Commons licence, unless clinical routine, the authors identiﬁed the best single indicated otherwise in a credit line to the material. If material predictive methylation marker from the EPIMMUNE is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or signature, namely forkhead box P1 (FOXP1). A third exceeds the permitted use, you will need to obtain permis- cohort of 61 NSCLC patients was tested for FOXP1 sion directly from the copyright holder. To view a copy of this methylation using pyrosequencing and hypomethy- licence, visit http://creativecommons.org/licenses/by/4.0/. lation of this gene was found to be an independent predictor of PFS and OS in a multivariable model . A similar approach to identify a set of CpG sites References predictive for response to anti-PD-1/PD-L1 therapy 1. BrahmerJ,ReckampKL,BaasP,CrinoL,EberhardtWE,Pod- in advanced NSCLC patients was published by Kim dubskaya E, et al. Nivolumab versus Docetaxel in advanced et al. . In this study, tumor samples of 60 NSCLC squamous-cell non-small-cell lung cancer. N Engl J Med. patients were analyzed using MethylationEPIC bead- 2015;373(2):123–35. Chips and 377 differentially methylated CpG sites 2. Motzer RJ, Tannir NM, McDermott DF, Aren Frontera O, between responders and nonresponders were iden- Melichar B, Choueiri TK, et al. Nivolumab plus ipilimumab tiﬁed. The vast majority of these CpG sites were versus sunitinib in advanced renal-cell carcinoma. N Engl J hypomethylated in nonresponders . Based on Med. 2018;378(14):1277–90. 3. Gandhi L, Rodriguez-Abreu D, Gadgeel S, Esteban E, Felip E, these data, the authors calculated a predictive methy- De Angelis F, et al. Pembrolizumab plus chemotherapy lation model consisting of 8 genes (IRF6, CTSD, GRN, in metastatic non-small-cell lung cancer. N Engl J Med. LTBR, TRIM36, EVL, CD3E and LCP1). Unfortunately, 2018;378(22):2078–92. the authors did not provide information about the 4. Paz-Ares L, Luft A, Vicente D, Tafreshi A, Gumus M, predictive value of FOXP1 methylation in their pa- Mazieres J, et al. Pembrolizumab plus chemotherapy tient cohort. Noteworthy, the clinical beneﬁt in this for squamous non-small-cell lung cancer. N Engl J Med. 2018;379(21):2040–51. cohort was not associated with TMB, neo-antigen 5. Ascierto PA, Del Vecchio M, Mandala M, Gogas H, load, aneuploidy level or PD-L1 expression . Arance AM, Dalle S, et al. Adjuvant nivolumab versus An example of a single DNA methylation event as ipilimumabin resectedstageIIIB-CandstageIV melanoma potential predictive marker for immunotherapies was (CheckMate 238): 4-year results from a multicentre, dou- reported by Goltz et al. in melanoma patients . ble-blind, randomised, controlled, phase 3 trial. Lancet The authors analyzed methylation of the CTLA4 pro- Oncol. 2020;21(11):1465–77. moter in a set of 50 anti-PD-1/CTLA-4-ICI treated 6. Bellmunt J, de Wit R, Vaughn DJ, Fradet Y, Lee JL, Fong L, et al. Pembrolizumab as second-line therapy for advanced patients with metastasized malignant melanoma by urothelial carcinoma. N Engl J Med. 2017;376(11):1015–26. methylation-speciﬁc quantitative real-time PCR. Low 7. Davis AA, Patel VG. The role of PD-L1 expression as a pre- CTLA4 methylation was signiﬁcantly correlated with dictive biomarker: an analysis of all US Food and Drug response to therapy and OS . Administration (FDA) approvals of immune checkpoint In summary, the above mentioned studies suggest inhibitors. J Immunother Cancer. 2019;7(1):278. that DNA methylation pattern may be of predictive 8. Martincorena I, Campbell PJ. Somatic mutation in cancer relevance for response to immunotherapies in the andnormal cells. Science. 2015;349(6255):1483–9. 9. Choucair K, Morand S, Stanbery L, Edelman G, Dworkin L, future; however, available data are still very limited Nemunaitis J. TMB: a promising immune-response and a multitude of additional studies are needed to biomarker, and potential spearhead in advancing targeted strengthen this concept. therapy trials. Cancer GeneTher. 2020;27(12):841–53. 10. Rosenberg JE, Hoffman-Censits J, Powles T, van der Heij- Take home message den MS, Balar AV, Necchi A, et al. Atezolizumab in patients with locally advanced and metastatic urothelial carcinoma who have progressed following treatment with platinum- Treatment with ICI can generate durable responses in based chemotherapy: a single-arm, multicentre, phase 2 a subset of cancer patients. Promising data about TMB, trial. Lancet. 2016;387(10031):1909–20. MSI and DNA methylation as future markers to predict 11. Hellmann MD, Ciuleanu TE, Pluzanski A, Lee JS, Otter- response to ICI are available. son GA, Audigier-Valette C, et al. Nivolumab plus Ip- 152 DNA methylation and immunotherapies K short review ilimumab in lung cancer with a high tumor mutational analyses in non-small cell lung cancer patients. Carcino- burden. N Engl J Med. 2018;378(22):2093–104. genesis. 2013;34(3):513–21. 12. Hellmann MD, Callahan MK, Awad MM, Calvo E, 25. Irizarry RA,Ladd-Acosta C, Wen B, WuZ,Montano C, Ascierto PA, Atmaca A, et al. Tumor mutational burden and Onyango P, et al. The human colon cancer methy- efﬁcacy of nivolumab monotherapy and in combination lome shows similar hypo- and hypermethylation at con- with Ipilimumab in small-cell lung cancer. Cancer Cell. served tissue-speciﬁc CpG island shores. Nat Genet. 2019;35(2):329. 2009;41(2):178–86. 13. Goodman AM, Kato S, Bazhenova L, Patel SP, Frampton GM, 26. Capper D,Jones DTW, Sill M, Hovestadt V,SchrimpfD, Miller V, et al. Tumor mutational burden as an independent Sturm D, et al. DNA methylation-based classiﬁ- predictorofresponsetoimmunotherapyindiversecancers. cation of central nervous system tumours. Nature. Mol Cancer Ther. 2017;16(11):2598–608. 2018;555(7697):469–74. 14. MiaoD,MargolisCA,GaoW,VossMH,LiW,MartiniDJ,etal. 27. Koelsche C, Kriegsmann M, Kommoss FKF, Stichel D, Genomic correlates of response to immune checkpoint Kriegsmann K, Vokuhl C, et al. DNA methylation proﬁl- therapies in clear cell renal cell carcinoma. Science. ing distinguishes Ewing-like sarcoma with EWSR1-NFATc2 2018;359(6377):801–6. fusion from Ewing sarcoma. J Cancer Res Clin Oncol. 15. Klempner SJ, Fabrizio D, Bane S, Reinhart M, Peoples T, 2019;145(5):1273–81. Ali SM, et al. Tumor mutational burden as a pre- 28. Cancer Genome Atlas Research Network, Electronic ad- dictive biomarker for response to immune checkpoint dress edsc, Cancer Genome Atlas Research N. Comprehen- inhibitors: a review of current evidence. Oncologist. sive and Integrated Genomic Characterization of Adult Soft 2020;25(1):e147–e59. TissueSarcomas. Cell. 2017;171(4):950–965e28. e28. 16. Trabucco SE, Gowen K, Maund SL, Sanford E, Fabrizio DA, 29. Sigin VO, Kalinkin AI, Kuznetsova EB, Simonova OA, Ches- HallMJ,etal. Anovelnext-generationsequencingapproach nokova GG, Litviakov NV, et al. DNA methylation markers to detecting microsatellite instability and pan-tumor char- panel can improve prediction of response to neoadju- acterization of 1000 microsatellite instability-high cases in vant chemotherapy in luminal B breast cancer. Sci Rep. 67,000 patientsamples. J Mol Diagn. 2019;21(6):1053–66. 2020;10(1):9239. 17. Le DT,Kim TW,Van Cutsem E, Geva R, JagerD,HaraH,et 30. Duruisseaux M, Martinez-Cardus A, Calleja-Cervantes ME, al. Phase II open-label study of pembrolizumab in treat- MoranS,CastrodeMouraM,DavalosV,etal. Epigeneticpre- ment-refractory, microsatellite instability-high/mismatch diction of response to anti-PD-1 treatment in non-small- repair-deﬁcient metastatic colorectal cancer: KEYNOTE- cell lung cancer: a multicentre, retrospective analysis. 164. J Clin Oncol. 2020;38(1):11–9. LancetRespir Med. 2018;6(10):771–81. 18. Marabelle A, Le DT, Ascierto PA, Di Giacomo AM, De Jesus- 31. KimJY,ChoiJK,JungH.Genome-widemethylationpatterns Acosta A, Delord JP, et al. Efﬁcacy of pembrolizumab in predict clinical beneﬁt of immunotherapy in lung cancer. patients with noncolorectal high microsatellite instability/ Clin Epigenetics. 2020;12(1):119. mismatch repair-deﬁcient cancer: results from the phase II 32. Goltz D, Gevensleben H, Vogt TJ, Dietrich J, Golletz C, Bootz KEYNOTE-158 study. J Clin Oncol. 2020;38(1):1–10. F, et al. CTLA4 methylation predicts response to anti-PD-1 19. Sandoval J, EstellerM. Cancerepigenomics: beyond ge- and anti-CTLA-4 immunotherapy in melanoma patients. nomics. Curr Opin GenetDev. 2012;22(1):50–5. JCI Insight. 2018;3(13). https://doi.org/10.1172/jci.insight. 20. Zhang L, Lu Q, Chang C. Epigenetics in health and disease. 96793 Adv Exp MedBiol. 2020;1253:3–55. Publisher’s Note Springer Nature remains neutral with regard 21. Deaton AM, Bird A. CpG islands and the regulation of to jurisdictional claims in published maps and institutional transcription. Genes Dev. 2011;25(10):1010–22. afﬁliations. 22. JonesPA.FunctionsofDNA methylation: islands,startsites, genebodiesandbeyond. NatRevGenet. 2012;13(7):484–92. 23. Heller G, Topakian T, Altenberger C, Cerny-Reiterer S, Herndlhofer S, Ziegler B, et al. Next-generation sequencing For latest news from interna- identiﬁes major DNA methylation changes during pro- gression of Ph+ chronic myeloid leukemia. Leukemia. tional oncology congresses see: 2016;30(9):1861–8. http://www.springermedizin.at/ 24. Heller G, Babinsky VN, Ziegler B, Weinzierl M, Noll C, Al- memo-inoncology tenberger C, et al. Genome-wide CpG island methylation K DNA methylation and immunotherapies 153
memo - Magazine of European Medical Oncology – Springer Journals
Published: Jun 1, 2021
Keywords: oncology; medicine/public health, general
Access the full text.
Sign up today, get DeepDyve free for 14 days.