Clinical epigenetics settings for cancer and cardiovascular diseases: real-life applications of network medicine at the bedside.

Introduction

Despite advances in early detection, therapeutic strategies, and supportive care, cancer and cardiovascular diseases (CVDs) remain the leading causes of morbidity and mortality worldwide [ 6 , 140 ], WHO reveals leading causes of death and disability worldwide: 2000–2019, Cancer (who.int)). In 2015, then US President Barak Obama launched a precision medicine initiative [ 105 ] with the goal of refining the current standard of care. To achieve this goal, noninvasive biomarkers are needed that can diagnose diseases on a mechanistic level, as well as novel or repurposed drugs that are able to target precisely these mechanisms. Integration of clinical with genomics or even multi-omics datasets such as transcriptomics and proteomics [ 146 ] has revealed novel disease-specific molecular pathways in cancer [ 12 , 43 , 102 ] and CVDs [ 76 ]. However, the connection of hereditary genetic determinants (which may allow patient stratification) and the individual risk for cancer or CVDs is lacking. This "missing heritability" eludes us due to the complex interplay of multiple factors in complex diseases, including epigenetics, which has already been shown to provide new insights into oncogenesis [ 30 , 31 , 45 , 102 ] and CVD [ 98 – 101 ]. Epigenetic modifications may bridge the gap between the genome and the environment revealing early signs of disease onset even in the early phase of fetal development, where epigenetic factors have been associated with an increased risk of CVDs in later life (transgenerational effect) [ 36 , 98 – 101 ].

Epigenetics comprises regulatory mechanisms that do not affect the DNA sequence but alter chromatin compaction to modulate gene expression [ 46 ]. DNA methylation, regulation of chromatin accessibility, and histone tail modifications are key moderators of vital cellular processes, such as differentiation, survival, and response to external stimuli [ 46 ]. These mechanisms individually or in combination are often related to pathogenesis and offer an opportunity to improve disease diagnosis and to predict clinical outcomes [ 17 ]. As epigenetic modifications are partially reversible, chromatin-acting epi-drugs can be used to treat complex human diseases. DNA methyl transferase inhibitors (DNMTi), histone deacetylase inhibitors (HDACi), and enhancer of zeste 2 polycomb repressive complex 2 subunit inhibitors (EZH2i) have rapidly reached considerable clinical relevance. For the treatment of hematologic malignancies, one epigenetic biomarker [ 124 ] and eight epi-drugs Table 1 have been approved by the FDA and are currently in clinical trials for different solid malignancies Table 2 . In contrast, there are, as yet, no approved epi-drugs for CVDs, however, more than twenty Phase 3–4 randomized trials are currently ongoing to evaluate the epigenetic efficacy of the so-called “repurposed” drugs against CVDs, including metformin, statins, and apabetalone, a bromodomain inhibitor with quinazoline structure [ 98 , 99 ] Table 2 . Table 1 FDA approved epi-drug for cancer treatment Epi-drug Development Overview Characteristics Mechanism of Action Data of approval Disease approval Phase 3 Phase 2 Phase 1/2 Most Recent Events Vidaza Clinical studies began in 1980s. However, in the first studies its activity had not been maximized as the mechanism had not yet been identified. In 2004 Azacitidine was marketed as Vidaza in the USA, in subcutaneous administration, for the treatment of five subtypes of MDS. In January 2007, the FDA approved the administration for intravenous use. Azacitidine is available in the EU and several Asia–Pacific countries for the treatment of high-risk MDS, AML, and CML, in Canada and Japan for MDS high risk and a subgroup of AML patients who are not eligible for stem cell transplantation. In September 2020 the drug was approved in oral administration (ONUREG), for AML patients in first complete remission or with complete remission but incomplete blood recovery after chemotherapy (NCT0175735) Antineoplastics; Aza compounds; Pyrimidine nucleosides Antimetabolites; DNMT inhibitors 2004 for MDS 2020 for AML MDS, AML and CML TL BC; Leukemia; NC; NSCLC; Peripheral TL CC; Diffuse large B cell lymphoma; RC; Solid tumors 25 May 2020 Preregistration for AML (First-line therapy) in European Union (PO) Decitabine Decitabine is developed by Otsuka and Janssen-Cilag for MDS and AML. The compound is used in the USA and many other countries outside the EU for MDS and in Malaysia for CML and AML in the EU. The product is also undergoing regulatory review in China for MDS and AML. Clinical development is ongoing for MDS and ovarian cancer. Decitabine has shown limited efficacy against solid tumors. A combined, orally bioavailable, fixed-dose formulation of decitabine plus cedazuridine was approved in 2020 for MDS Antineoplastics; Aza compounds; Deoxyribonucleosides; Pyrimidine nucleosides DNMT inhibitors 2006 for MDS MDS, AML and CML T-ALL, T-LBL, T/M-MPAL Malignant melanoma; OC 07 Dec 2019 Pharmacodynamics data from a preclinical study in acute myeloid leukemia Vorinostat Vorinostat has been launched in the USA, Canada, Japan, Argentina, Chile, and Greece for CTCL treatment. Development of vorinostat was discontinued in mesothelioma non-small cell lung cancer, gynecological cancer, and pancreatic cancer. Vorinostat was in phase II development for colorectal cancer and sickle-cell anemia. However, no development in these indications have been reported Antineoplastic, hydroxamic acid HDAC inhibitors 2006 CTLC MM; MLC AML; B-cell lymphoma; BC; Glioma; HIV-1; MDS; NHL; ALL ALL; Lymphoma; Solid tumors 09 Apr 2019 Merck & Co completes a phase II trial in B-cell lymphoma in South-East Asia (PO) ( NCT00875056 ) Romidepsin Romidepsin developed by Celgene was discovered by Fujisawa Pharmaceutical from Chromobacterium violaceum. The intravenous formulation of romidepsin was tested in the USA for CTLC, PTCL. Clinical development is underway for breast cancer, multiple myeloma, T-cell leukemia, and T-cell lymphoma in several countries. In 2015 Romidepsin was in pre-registration phase for peripheral T-cell lymphoma and in phase II for cutaneous T-cell lymphoma, and studies for the treatment of renal cell carcinoma, prostate cancer, and pancreatic cancer were stopped Antineoplastics; Cytostatic antibiotics; Depsipeptides HDAC inhibitors 2009 CTLC, PTCL BC; MM; T-LL 11 Apr 2020 Celgene Corporation and University of California completes a phase I trial in Cutaneous T-cell lymphoma in the USA ( NCT01902225 ) 25 Mar 2020 Celgene Corporation suspends a phase-I/II clinical trials in breast cancer (combination therapy) in the USA, due to five patients’ responses pending ( NCT02393794 ) Belinostat Belinostat is developed by Onxeo and Spectrum Pharmaceuticals. Intravenous belinostat is available in the USA and is under regulatory review in Canada as a monotherapy for PTCL. Clinical trial is ongoing for B-cell lymphoma and glioblastoma, in the USA, and for NSCLC in the USA and Denmark. Development for the treatment of multiple myeloma has been stopped Acrylamides; Antineoplastics; Hydroxamic acids HDAC inhibitors 2014 PTCL B-cell lymphoma; Glioblastoma NSCLC 10 Apr 2020 National Cancer Institute plans a phase II trial for Chondrosarcoma (Combination therapy, second-line therapy or greater) ( NCT04340843 ) Panobinostat Panobinostat being developed by Novartis and Secura Bio, for the treatment of cancer and HIV infections, and after approved from the FDA in 2015 for MM Antineoplastic, hydroxamic acid HDAC inhibitors 2015 MM CMML AML; CLL; CC; Lymphoma; MDS; Myelofibrosis; Neuroendocrine tumors; NHL; T-LBL; RC; TC BC; HIV-1; Hodgkin’s disease; melanoma 11 Aug 2020 Phase-I clinical trials in Glioma (In children, In adults, Combination therapy, Second-line therapy or greater) in the USA (PO) ( NCT04341311 ) 22 Jun 2020 Phase I development is ongoing for Myelofibrosis (combination therapy) in Germany, France, UK, Ireland and Italy ( NCT01433445 ) Ivosidenib Ivosidenib was approved on July 20, 2018, in oral administration, for adult patients with relapsed or refractory AML with IDH1 mutation. From May 2019, the drug was approved also for newly diagnosed AML with a susceptible IDH1 mutation in patients who are at least 75 years old or that cannot be treated with chemotherapy (NCT0274839) Antineoplastics; Cyclobutanes; Pyridines; Pyrrolidines IDH1 inhibitors 2018 AML with IDH1 mutation Cholangiocarcinoma; MDS 30 Jul 2020 Agios pharmaceuticals announces to submit sNDA to US FDA for Cholangiocarcinoma, in second-line therapy in the 2021 Enasidenib In August 2017, Enasidenib (IDHIFA®), produced by Celgene, was approved from the FDA for the treatment of adult patients with relapsed or refractory AML with IDH2 mutation. The drug is used first for AML in the USA and then approved in Australia. Clinical developments are ongoing for AML, solid tumors, CML, and MDS Other potential indications include a genetic neurometabolic disorder (Type II D-2-HGA; 2-Hydroxyglutaricaciduria) Amines; Antineoplastics; Pyridines; Triazines IDH2 inhibitors 2017 AML with IDH2 mutation MDS Solid tumor 12 Jun 2020 Celgene reinitiates a phase II trial in adult patients AML as second-line therapy in the USA (PO) ( NCT03881735 ) Tazemetostat Tazemetostat, developed by Epizyme in orally administration, is used for MES and FL treatment in the USA. Clinical development of the oral formulation is ongoing for B-cell lymphoma, central nervous system cancer, follicular lymphoma, histiocytosis, mesothelioma, non-Hodgkin’s lymphoma, non-small cell lung cancer, and solid tumors in the USA, Australia, Japan, Canada, and the European Union Eisai and Epizyme with Roche Molecular Systems are developing complementary diagnostics to help companies identify patients with lymphoma with EZH2 gene mutations, to be used in parallel with the clinical development of tazemetostat Clinical development of the suspension formulation of tazemetostat for the treatment of solid tumors was ongoing in the USA. However, as of September 2018, no recent development reports for the formulation have been identified Antineoplastics; Biphenyl compounds; Dihydropyridines Inhibitor of wild-type and mutated forms of EZH2 and SMARCA2 and SMARCA4 2020 MES, FL CNS cancer; DLBCL; Histiocytosis; OC; Mesothelioma; NHL; Peritoneal cancer; Rhabdoid tumor; Solid tumors; Synovial sarcoma; Uterine cancer Bladder cancer; NSCLC; PC 08 Jul 2020 Preregistration for FL (Refractory metastatic disease, Second-line therapy or greater) in Japan (PO) 22 Jun 2020 Pharmacodynamics data from a preclinical trial in NSCLC Abbreviations: MDS: Myelodysplastic Syndrome; CTLC: cutaneous T cell lymphoma; PTCL: peripheral T cell lymphoma; MM: multiple myeloma; MES: metastatic epithelioid sarcoma; AML: Acute Myeloid Leukemia; CMML: Chronic Myelomonocytic Leukemia; MPC: Metastatic Pancreatic Cancer; MLC: Mesothelioma; ALL: acute lymphoblastic leukemia; LLy: acute lymphoblastic lymphoma; T-ALL: Acute T-Lymphocytic Leukemia; T-LBL: T-lymphoblastic leukemia/lymphoma; DLBCL: Diffuse large B-cell lymphoma; NHL: non-Hodgkin lymphoma; BC: Breast cancer; NSCLC: non-small cell lung cancer; CC: Colon cancer; RC: Renal cancer; PC: Prostate cancer; FS: follicular lymphoma; OC: Ovarian cancer; TC: Thyroid cancer; NC: Nasopharyngeal cancer Table 2 Epitherapy in clinical trials for cancer and CVDs Drug Conditions Study type Study title Aim Status/phase ID Drugs in clinical trial SGI-110 (guadecitabine) AML 302 participants, Interventional, Randomized, Parallel Assignment, Treatment Phase 3 Randomized, Open-Label Study of Guadecitabine vs Treatment Choice in Previously Treated Acute Myeloid Leukemia Multicenter, randomized, open-label, parallel-group study of guadecitabine vs treatment choice Active/Phase 3 NCT02920008 MDS CMML 408 participants, Interventional, Randomized, Parallel Assignment, Treatment Guadecitabine (SGI-110) vs Treatment Choice in Adults With MDS or CMML Previously Treated With HMAs To evaluate the efficacy and safety of guadecitabine in MDS or CMML who failed or relapsed after treatment with azacitidine, decitabine, or both Active/Phase 3 NCT02907359 AML 815 participants, Interventional, Randomized, Parallel Assignment, Treatment SGI-110 in Adults With Untreated Acute Myeloid Leukemia (AML), Not Considered Candidates for Intensive Remission Induction To compare efficacy and safety between SGI-110 and treatment in adults with previously untreated AML Completed/ Phase 3 NCT02348489 RX-3117 MPC 46 participants, Interventional, N/A, Single group Assignment, Treatment RX-3117 in Combination With Abraxane® in Subjects With Metastatic Pancreatic Cancer To determine the safety profile, dose modification, and pharmacokinetics of oral RX-3117 administered in combination with Abraxane® to subjects with metastatic pancreatic cancer Completed/ Phase 1/2 NCT03189914 Solid tumor Metastatic Bladder Cancer 124 participants, Interventional, N/A, Single group Assignment, Treatment Dose-Finding and Safety Study for Oral Single-Agent to Treat Advanced Malignancies To determine the maximum tolerated dose in metastatic solid tumors and to estimate anti-tumor activity in subjects with relapsed or refractory pancreatic or advanced bladder cancer Completed/ Phase 1/2 NCT02030067 INCB059872 Solid tumors Advanced Malignancies Metastatic Cancer 70 participants, Interventional, N/A, Single group Assignment, Treatment Azacitidine Combined With Pembrolizumab and Epacadostat in Subjects With Advanced Solid Tumors (ECHO-206) To evaluate the safety and tolerability of INCB059872 with pembrolizumab and epacadostat Completed Phase 1/2 NCT02959437 Solid tumors and hematologic malignancy 215 participants, Interventional, N/A, Single group Assignment, Treatment An Open-Label, Dose-Escalation/Dose-Expansion Safety Study of INCB059872 in Subjects With Advanced Malignancies To determine the safety, tolerability, efficacy, PK, PD, and the recommended dose(s) of azacitadine and all-trans retinoic acid in AML and in combination with nivolumab in SCLC Recruiting Phase 1/2 NCT02712905 CI-994 MM 6 participants, Interventional, N/A, Single group Assignment, Treatment CI-994 in Treating Patients With Advanced Myeloma To study the effectiveness of CI-994 in treating patients who have advanced myeloma Completed Phase 2 NCT00005624 Lung cancer Interventional, Randomized, Treatment Gemcitabine With or Without CI-994 in Treating Patients With Advanced Non-small Cell Lung Cancer To compare the effectiveness of gemcitabine with or without CI-994 in treating patients who have advanced non-small cell lung cancer Completed Phase 3 NCT00005093 Pancreatic cancer Interventional, Randomized, Treatment Gemcitabine With or Without CI-994 in Treating Patients With Advanced Pancreatic Cancer To compare the effectiveness of gemcitabine with or without CI-994 in treating patients who have advanced pancreatic cancer Completed Phase 2 NCT00004861 EPZ-5676 (Pinometostat) AML with 11q23 rearrangement 36 participants, Interventional, N/A, Single group Assignment, Treatment Pinometostat and Azacitidine in Treating Patients With Relapsed, Refractory, or Newly Diagnosed Acute Myeloid Leukemia With 11q23 Rearrangement To study the side effects and the dose of pinometostat with azacitidine in AML with 11q23 rearrangement Recruiting Phase 1/2 NCT03701295 AML with 11q23 rearrangement 37 participants, Interventional, N/A, Single group Assignment, Treatment Pinometostat With Standard Chemotherapy in Treating Patients With Newly Diagnosed Acute Myeloid Leukemia and MLL Gene Rearrangement To study the side effects and the dose of pinometostat in AML with 11q23 rearrangement Recruiting Phase 1/2 NCT03724084 TCP AML 16 participants, Interventional, N/A, Single group Assignment, Treatment Phase I/II Trial of ATRA and TCP in Patients With Relapsed or Refractory AML and no Intensive Treatment is Possible To study the feasibility, pharmacodynamics, and effectivity of ATRA and TCP in AML Phase 1/2 NCT02261779 AML, MDS 60 participants, Interventional, N/A, Single group Assignment, Treatment Study of Sensitization of Non-M3 AML Blasts to ATRA by Epigenetic Treatment With Tranylcypromine (TCP) To study the maximum tolerated and the efficacy dose of TCP with ATRA and Cytarabine Recruiting Phase 1/2 NCT02717884 CPI-0610 Myelofibrosis AML, MDS/MPD, MDS 271 participants, Interventional, Non-Randomized, Parallel Assignment, Treatment A Phase 2 Study of CPI-0610 With and Without Ruxolitinib in Patients With Myelofibrosis Open-label, sequential dose escalation study of CPI-0610 and open-label study of CPI-0610 with and without Ruxolitinib in patients with Myelofibrosis Recruiting Phase 1/2 NCT02158858 Rocilinostat (ACY-1215) MM 101 participants, Interventional, Single Group Assignment, Treatment ACY-1215 (Ricolinostat) in Combination With Pomalidomide and Low-dose Dex in Relapsed-and-Refractory Multiple Myeloma To evaluate the side effects, determine the best dose and the response rate of ACY-1215 in combination with Pomalidomide and low-dose dexamethasone in patients with relapsed-and-refractory multiple myeloma Active, not recruiting Phase 1b/2 NCT01997840 Mocetinostat Urothelial carcinoma 17 participants, Interventional, N/A, Single group Assignment, Treatment Study of Mocetinostat in Patients With Urothelial Carcinoma Having Inactivating Alterations of Specific Genes To evaluate whether the number of patients responding to treatment is substantially higher than other available treatments Completed/ Phase II NCT02236195 Hodgkin’s lymphoma 51 participants, Interventional, Non-Randomized, Single group Assignment, Treatment Study of MGCD0103 (MG-0103) in Patients With Relapsed or Refractory Hodgkin’s Lymphoma To study the effect of Mocetinostat on patients with relapsed and refractory Hodgkin’s lymphoma Terminated/ Phase II NCT00358982 CPI-1205 mCRPC 242 participants, Interventional, Randomized, Parallel Assignment, Treatment ProSTAR: A Study Evaluating CPI-1205 in Patients With Metastatic Castration Resistant Prostate Cancer To study the oral administration of CPI-1205 in combination with enzalutamide or abiraterone/prednisone in male patients with mCRPC Active, not recruiting/ Phase I/II NCT03480646 Advanced Solid Tumors 24 participants, Interventional, N/A, Single group Assignment, Treatment ORIOn-E: A Study Evaluating CPI-1205 in Patients With Advanced Solid Tumors To determine the maximum tolerated dose and recommended phase 2 dose of CPI-1205 + ipilimumab in patients with advanced solid tumors Active, not recruiting/ Phase I/II NCT03525795 Approved epi-drug ongoing in clinical trial for other treatments Vorinostat ALL, LLy 1000 participants, Interventional, Randomized, Parallel Assignment, Treatment Total Therapy XVII for Newly Diagnosed Patients With Acute Lymphoblastic Leukemia and Lymphoma To study a novel strategies based on inherited and acquired leukemia-specific genomic features and treatment to improve the cure rate and quality of life of ALL and LLy children Recruiting/ Phase2/3 NCT03117751 MM 637 participants, Interventional, Randomized, Single group Assignment, Treatment Study of Vorinostat (MK-0683) an HDAC Inhibitor, or Placebo in Combination with Bortezomib in Patients With Multiple Myeloma (MK-0683-088 AMN) To study of the efficacy and safety of bortezomib administered in combination with vorinostat in patients with relapsed or refractory MM Complete/ Phase 3 NCT00773747 MLC 662 participants, Interventional, Randomized, Single group Assignment, Treatment Suberoylanilide Hydroxamic Acid (Vorinostat, MK-0683) Versus Placebo in Advanced Malignant Pleural Mesothelioma (0683-014 AM5, EXT1) To study the safety, tolerability, and anti-tumor effectiveness of vorinostat in oral administration, for MLC Complete/ Phase 3 NCT00128102 Panobinostat AML, MDS 350 participants, Interventional, Randomized, Parallel Assignment Treatment Panobinostat Maintenance After HSCT for High-risk AML and MDS To compare maintenance treatment with panobinostat versus the standard approach of preemptive DLI alone in patients with poor-risk AML/MDS Recruiting/ Phase 3 NCT04326764 PMF, CIM, post-PVMF, PET-MF, SR-GVHD 356 participants, Interventional, Non-Randomized, Parallel Assignment Treatment CINC424A2X01B Rollover Protocol To study the efficacy of ruxolitinib and panobinostat in combination for patients that use the study treatment based on the parent protocol Recruiting/ Phase 4 NCT02386800 Vidaza AML 488 participants, Interventional, Randomized, Parallel Assignment, Treatment Study of Vidaza Versus Conventional Care Regimens for the Treatment of Acute Myeloid Leukemia (AML) To compare the effect of azacitidine to conventional care regimens in AML patients Complete/ Phase 3 NCT01074047 AML 47 participants, Interventional, Non-Randomized, Parallel Assignment, Treatment zacitidine Combined With Homoharringtonie in AML To validate the efficacy and safety advantages of the regimens that contain homoharringtonie and azacitidine, and to determine which regimen would receive more living benefits in AML Recruiting/ Phase 3 NCT04248595 AML, MDS, CMML 187 participants, Interventional, Randomized, Single group Assignment, Treatment Controlled Study of Post-transplant Azacitidine for Prevention of Acute Myelogenous Leukemia and Myelodysplastic Syndrome Relapse (VZ-AML-PI-0129) To study if Vidaza will help to control the disease in patients with AML, CMML, or MDS after a stem cell transplant Complete/ Phase 3 NCT00887068 AML, MDS, CML 114 participants, Interventional, Randomized, Parallel Assignment, Treatment Azacytidine + HAG Regimen vs. Azacytidine for Elderly Patients With Newly Diagnosed Myeloid Malignancy To explore the efficacy and safety of azacytidine and HAG regimen versus azacytidine for elderly patients with Newly Diagnosed MDS/AML/CMML in China Recruiting/ Phase 4 NCT03873311 Decitabine T-ALL, T-LBL, T/M-MPAL 100 participants, Interventional, N/A, Single group Assignment Treatment Study of Decitabine Combined With HAAG Regimen in Newly Diagnosed ETP-ALL/LBL, T/M-MPAL and ALL/LBL With Myeloid or Stem Cell Markers Patients To evaluate the efficacy and safety of decitabine combined with HAAG Recruiting/ Phase 3 NCT04446130 DLBCL 60 participants, Interventional, Randomized, Parallel Assignment Treatment A Clinical Trial of Decitabine in Relapse and Refractory Diffuse Large B Cell Lymphoma To evaluate the safety, tolerability, and clinical effects of decitabine with R ± DHAP Recruiting/ Phase 4 NCT03579082 Tazametostat Solid tumor 49 participants, Interventional, N/A, Single group Assignment Treatment Tazemetostat in Treating Patients With Relapsed or Refractory Advanced Solid Tumors, Non-Hodgkin Lymphoma, or Histiocytic Disorders With EZH2, SMARCB1, or SMARCA4 Gene Mutations (A Pediatric MATCH Treatment Trial) To study the action of tazemetostat in patients with brain tumors, solid tumors, non-Hodgkin lymphoma, or histiocytic disorders that have come back (relapsed) or do not respond to treatment (refractory) and have EZH2, SMARCB1, or SMARCA4 gene mutations Phase 2 NCT03213665 Drug repositioning in CVDs Trimetazidine HFpEF 25 participants, Interventional, Randomized, Placebo-controlled TrimetaziDine as a Performance-enhancING drug in Heart Failure with Preserved Ejection Fraction To assess whether trimetazidine improves LV diastolic function by improving myocardial energy metabolism in HFpEF Ongoing/ Phase 2 EU Clinical Trial Register: 2018-002170-52; NTR registration: NL7830 [van de Bovenkamp AA, PAH 26 participants, Interventional, Randomized, Placebo-controlled The Role of Trimetazidine on Right Ventricle Function in Pulmonary Arterial Hypertension (TRIMETA-PH) To evaluate the effect of trimetazidine versus placebo in addition to standard PAH regime on right ventricular function Completed/Phase 2/ 3 NCT03273387 Tocilizumab PAH 29 participants, Interventional, Single‐arm A Therapeutic Open Label Study of Tocilizumab in the Treatment of Pulmonary Arterial Hypertension (TRANSFORM-UK) To assess the safety and efficacy of tocilizumab in PAH Completed/Phase 3 NCT02676947 Metformin MetS 40 participants, Interventional, Randomized, Placebo-controlled Combination of Metformin/Inulin vs Inulin on Adiponectin in Metabolic Syndrome To compare the effect of the administration of Metformin/agave inulin vs. Agave inulin on adiponectin in patients with MetS Completed/ Phase 3 NCT02773927 T2D, HF 30 participants, Observational, Cohort, Prospective Lipid Accumulation in Heart Transplant From Non-diabetic Donors to Diabetic Recipients (DCM-AHEAD) To evaluate in the explanted diabetic heart the presence of possible cellular alterations attributable to the diabetic disease discerning from treated and non-treated with metformin Completed/NA NCT03546062 PAH 1899 participants, Observational, Cohort, Prospective Hormonal, Metabolic, and Signaling Interactions in PAH To evaluate if an optimal treatment of the dysfunctional metabolic pathways underlying PAH may improve pulmonary vascular function and consequences of the disease Recruiting/NA NCT01884051 Apabetalone (RVX 208) PAH 10 participants, Interventional, Single‐arm, 2‐center study Apabetalone for Pulmonary Arterial Hypertension: a Pilot Study To evaluate the possible efficacy of apabetalone in treating PAH Recruiting/Early Phase 1 NCT03655704 T2D-CHD 2425 participants, Interventional, Randomized, Placebo-controlled Effect of RVX000222 on Time to Major Adverse Cardiovascular Events in High-Risk T2DM Subjects With CAD (BETonMACE) To determine whether RVX000222 in high-risk type 2 diabetes mellitus patients with coronary artery disease increases the time to major adverse cardiovascular events Phase 3/Active NCT02586155 n-3 PUFA or Rosuvastatin HF 6975 participants, Interventional, Randomized, Placebo-controlled GISSI-HF- Effects of n-3 PUFA and Rosuvastatin on Mortality-Morbidity of Patients With Symptomatic CHF To demonstrate that, in HF patients treated at the best of recommended therapies, long-term administration of n-3 PUFA or rosuvastatin is more effective than the corresponding placebo Completed/ Phase 3 NCT00336336 Tavazzi et al. [ 128 ] Abbreviations: AML: Acute Myeloid Leukemia; MDS: Myelodysplastic Syndromes; CMML: Chronic Myelomonocytic Leukemia; MPC: Metastatic Pancreatic Cancer; MLC: Mesothelioma; MDS/MPD: Myelodysplastic–Myeloproliferative neoplasms;mCRPC: Metastatic Castration Resistant Prostate Cancer; ALL: acute lymphoblastic leukemia; LLy: acute lymphoblastic lymphoma; PMF: Primary myelofibrosis; CIM: Chronic Idiopathic Myelofibrosis;post-PV MF: Post-polycythemia vera myelofibrosis; PET-MF: post-essential thrombocythemia myelofibrosis; SR-GVHD: Steroid Refractory Graft Versus Host Disease; T-ALL: Acute T-Lymphocytic Leukemia; T-LBL: T-lymphoblastic leukemia/lymphoma;T/M-MPAL: T-lymphoid/myeloid mixed phenotype acute leukemia; DLBCL: Diffuse large B-cell lymphoma; CHD: Coronary Heart Disease; HF: Heart Failure; LV: Left Ventricle; MetS: Metabolic Syndrome; NA: Not Applicable; PAH: Pulmonary Arterial Hypertension; PUFA: Omega-3 Polyunsaturated Fatty Acids; T2D: Type 2 Diabetes

The historically limited success in the discovery of epigenetic biomarkers and epi-drugs using a reductionist approach calls for a paradigm shift toward network medicine (NM), which combines big data, advanced bioinformatic tools, network science, systems biology, artificial intelligence, and clinical biometric data to investigate the pathogenesis of complex diseases such as cancer and CVDs. By considering the molecular perturbations of integrated biological pathways rather than a single molecular defect ([ 15 , 75 , 92 , 64 , 120 , 123 ]), NM can leverage molecular interaction networks for advancing diagnosis, prognosis, and treatment Fig. 1 [ 15 , 75 , 120 , 123 ]. In this way, NM paves the way toward precision medicine and personalized therapy [ 54 ]. Focusing thus far mostly on genomics and transcriptomics, NM has recently been applied to the study epigenetic mechanisms such as DNA methylation changes in the pathogenesis of cancer (mostly) and CVDs Fig. 1 . Fig. 1 Application of NM in disease. Principles of network medicine methods (top) considering the differences of DNA methylation between disease and control samples. Network reconstruction (NR) methods build a disease network de novo while network analysis methods (NA) identify disease modules in existing networks based on prior knowledge (middle). Overview of commonly employed network medicine methods, their expected input and the concept they employ (bottom)

The goal of this review is to present clinical applications of epigenetics in cancer and CVDs. Additionally, we discuss how NM can gradually be advanced from the bench to bedside via epigenetics. With this overview, we provide information and insights for both basic scientists and physicians who work at the interface of these new applications of clinical epigenetics and Network Medicine.

Epi-therapy in cancer: a new frontier for the emergence of precision medicine advanced therapies

Approved epi-treatments against hematopoietic cancers

The main classification for epigenetic mediators divides them into writers, readers, and erasers [ 18 ]. DNMTs and HATs are “writers,” as they add a methyl or acetyl group on chromatin. These residues can be removed by the “erasers” such as HDACs or KDMs. Chromatin modifications may be interpreted by “readers,” such as the CBX (chromodomain) or BRD (bromodomain) family members. Writers, readers, and erasers comprise highly studied targets, some of which have led to the identification and development of FDA-approved drugs entering into the clinics.

DNMTi and HDACi are the epi-compounds most used for treatment. DNA hypermethylation is frequently associated with cancer development, and the two DNMTis, azacitidine (Vidaza) and decitabine (Dacogen), were approved for myelodysplastic syndromes (MDS) treatment in 2004 and 2006, respectively. These hypomethylating agents are currently the first-line therapy for MDS, when stem cell transplantation is unsuitable. Administered at a daily dose of 75 mg/m 2 for 7 days (azacitidine) and 5–20 mg/m 2 for 5/7 days (decitabine), these treatments are more effective than previously investigated drugs [ 110 ]. However, azacitidine has some limitations, including a low response rate, short duration of action, and aggravation of thrombocytopenia, leading to its co-administration with other therapies (Table 2 , NCT01488565 ). Azacitidine is currently also under investigation as a treatment option in solid malignancies (see Sect. 2.4 for further information).

Decitabine is well tolerated and although the most significant effects have been observed in hematologic malignancies, it also displays a good activity against solid tumors. Different Phase I/II clinical trials are currently underway, including co-administration studies with HDACi for patients with advanced solid tumors ( NCT01023737 , NCT02453620 , NCT03925428 , NCT03590054 , NCT01281176 ) or with cisplatin for resistant ovarian cancer [ 87 ]. Decitabine has shown numerous beneficial effects in solid tumors, leading to further investigation of methylation status as a prognostic marker in solid cancers. The FDA has approved second-generation HDACi, including vorinostat (Zolinza), belinostat (Beleodaq), romidepsin (Istodax), and panobinostat (Farydak) for treatment of cutaneous T-cell lymphoma (CTCL), peripheral T-cell lymphoma (PTCL), and multiple myeloma (MM).

Mutations in R132 and R172 in isocitrate dehydrogenase 1 and 2 (IDH1/2) have been found in approximately 20% of acute myeloid leukemias (patients with a devastating prognosis [ 109 ]. These mutations produce an abnormal conversion of α-ketoglutarate (D-2-kg) in α- hydroxyglutarate (2-HG), leading to a dysfunction of the enzymes using 2-kg as a cofactor, such as TETs and KDM enzymes, associated with DNA hypermethylation, “aberrant gene expression,” increase of proliferation and cell differentiation [ 95 ]. Recently, the FDA approved the first IDH1/2 inhibitors ( NCT01915498 , NCT02074839 ) against AML. Enasidenib (IDHIFA), targeting IDH2, and ivosidenib (TIBSOVO), IDH1, are administrated orally, blocking 2-HG in the blood of AML patients with IDH mutations. The patients treated with these agents showed a complete response (CR) or CR with partial hematological recovery after 8.2 months of treatment, response was superior with ivosidenib (32.8%) superior to the enasidenib treatment (23%). Ivosidenib has been approved in 2019 for AML patients older than 75 years or for patients in whom chemotherapy cannot be used.

Clinical breakthroughs after more than one decade of HDAC inhibitors use

Vorinostat, an HDACi used as an oral agent for CTCL since 2006 and now tested in clinical trials for solid tumors, displays low toxicity and high efficacy, making it a first-line drug for the treatment of this lymphoma. The majority of adverse effects are fatigue and gastrointestinal symptoms. Hematological abnormalities are observed only at the highest dose.

Like vorinostat, romidepsin is one of the first drugs to be used for CTCL. Romidepsin is a potent class I HDACi and its ability to inhibit class II HDACs at higher concentrations suggests that it may act as a broad-spectrum HDACi [ 52 ]. Unlike other drugs, this molecule induces a change in electrocardiographic patterns, after administration, with flattening of ST and T waves and depression of the ST segment. Six patients with mild cardiovascular disorders died after romidepsin treatment in Phase II studies. This drug also shows significant activity in patients with PTCL and in CTCL patients with stage IIB or higher disease presentations [ 107 ]. To date, romidepsin has proved unsuccessful as monotherapy in squamous cell carcinoma of the head and neck (SCCHN) [ 57 ] and in metastatic castration-resistant prostate cancer (CRPC) [ 94 ] showing limited adverse effects such as fatigue, nausea, and vomiting.

Belinostat, approved in 2014 for PTCL treatment [ 74 ], generally displays no major adverse effects, but in rare cases induces liver damage. Belinostat is currently in several clinical trials Table 1 .

Panobinostat was been approved in 2015 for multiple myeloma (MM) as a second-line therapy in patients not responding to bortezomib. The limitation of panobinostat has been the low efficacy when given as a single agent; thus, it is administered in combination with bortezomib and dexamethasone. MM patients treated with panobinostat in combination with bortezomib have a significant increase in survival ( NCT01023308 ). Additionally, when studied in a Phase III trial in non-Hodgkin lymphoma patients, panobinostat-induced adverse events in 22% of the patients with severe vomiting and diarrhea ( NCT01034163 ), and hepatic abnormalities in chronic myeloid leukemia (CML) ( NCT00449761 ), although it shows promising anticancer effects, alone or in combination [ 88 ].

Although HDACi are excellent candidates for integrative network analysis due to their extensively studied molecular mechanisms, no network-based approaches were reported for repurposing this class of drugs to other cancer types. However, by using a meta-analytical approach Rafehi et al . integrated ENCODE data with microarray expression profiles showing HDACi-mediated suppression of EP300 target genes, including genes implicated in diabetes mellitus [ 108 ].

Epi-treatments being investigated for solid cancers

Mutations and overexpression of EZH2 have been associated with prostate, breast, liver, skin, lung, and gastric cancers, as well as with lymphoma and melanoma [ 1 , 7 , 8 , 58 , 60 , 114 , 129 , 139 ]. Tazemetostat (Tazverik, Epizyme, Inc), acting as a selective competitor of S-adenosyl-L-methionine, is a first-in-class EZH2 inhibitor, administered orally against hematological and solid tumors. On June 18, 2020, the FDA granted approval to tazemetostat in adults with relapsed/ refractory (R/R) follicular lymphoma (FL) carrying an EZH2 mutation who have received at least two prior systemic therapies, or for those patients who have no alternative options. The cobas EZH2 Mutation Test (Roche Molecular Systems, Inc.) as a diagnostic test was also approved, based on two open-label, single-arm cohorts of a multicenter trial (E7438-G000-101, NCT01897571 ) in patients with FL after at least two prior systemic therapies.

Tazemetostat was also studied as a single agent in a Phase II clinical trial for patients with relapsed or refractory non-Hodgkin lymphoma [ 73 ], rare aggressive forms of lymphoma [ 85 ], follicular lymphoma in refractory patients in combination with atezolizumab (Tecentriq), diffuse large B-cell lymphoma (DLBCL), and is currently in the recruitment phase for prostate cancer in combination with enzalutamide or abitarerone/prednisol ( NCT04179864 ). In combination with R-CHOP, a standard chemotherapy regimen, tazemetostat, is in ongoing studies as a first-line treatment for newly diagnosed high-risk elderly patients with DLBCL.

Ongoing clinical trials with epi treatments against cancer

Several chromatin-acting drugs are being investigated in clinical trials; see Table 2 for a complete overview. DNMTi and HDACi administration produces a synergistic effect on methylation state, with an increase in repression of pro-oncogenic and activation of apoptotic genes [ 96 , 106 , 145 ]. Clinical trials in different phases testing the combined administration of both classes of drugs are yielding excellent results [ 19 ]. Although toxicity appears more frequently in older patients, azacitidine proved effective in MDS and in AML ( NCT01074047 ), where it represents an excellent replacement therapy for patients who are not candidates for more aggressive therapies (65–74 years) and for those with cytogenetics indicating an increased risk. The HDACi entinostat, a generally well-tolerated drug, is currently in Phase II trials for co-administration with azacitidine and nivolumab in patients with metastatic non-small lung cell cancer ( NCT01928576 ), and in melanoma and lymphoma in co-administered with pembrolizumab ( NCT03179930 , NCT03765229 ). Activity of the EZH2 inhibitor CPI-1205 in combination with the antiandrogen enzalutamide (Xtandi) is being tested in Phase I/II trials as second-line treatment of metastatic castration-resistant prostate cancer patients ( NCT03480646 ), and showing excellent tolerance and considerable anticancer activity, prompting further studies. The DOT1L inhibitor pinometostat (EPZ-5676) is part of an ongoing Phase I/II study for the treatment of AML and MLL (mixed-lineage leukemia) ( NCT03701295 , NCT03724084 ). Tranylcypromine, an LSD1 inhibitor, is under investigation in at least 26 studies. A recent interventional study testing tranylcypromine with all-trans retinoic acid (ATRA) in AML and MDS found no serious side effects ( NCT02273102 ), leading now to the more intensive evaluation study of its anticancer effects.

Osteosarcoma (OS) patients are generally treated with combination chemotherapy comprising cisplatin, doxorubicin, and high-dose methotrexate, with addition of ifosfamide [ 130 ]. However, a fundamental problem is the long duration of treatment, which leads to undesirable effects [ 118 ]. Several clinical trials in initial stages involve single or co-administration of epi-drugs [ 37 ]. EZH2 is a positive regulator of the growth of metastasis and prognosis of OS [ 38 ]. In January 2020, tazametostat (Tazverik) was quickly approved by the FDA for advanced metastatic and locally epithelioid sarcoma treatment in adults and pediatric patients (16 years), based on the successful results obtained in a phase 2 study ( NCT02601950 ).

Additionally, tazametostat is in Phase 2 clinical trial against OS with mutations of EZH2, SMARCB1 and SMARCA2 ( NCT03213665 ) in pediatric patients.

Despite the steady rise in the number of clinical studies that include epi-drugs as anticancer agents, only a few agents and combinations have proven useful for broad clinical use. Thus, a joint approach involving basic scientists and physicians for the development of clinically useful protocols is needed to improve our understanding of epigenetic mechanisms and how they can be targeted most effectively. This goal is particularly important for solid malignancies, where the use of chromatin-targeting drugs has not yet been shown to provide benefit over standard treatments. With respect to hematological malignancies, the clearly lower efficacy cannot only be attributed to limited knowledge of the oncogenic mechanism. Possible additional factors might include differences in solid tumor heterogeneity or in the 3D structures of solid malignancies where cancer cells in poorly vascularized tumor areas are difficult to reach with drugs. For all of these reasons and others to be uncovered, network medicine strategies may prove fundamental for an understanding of (epi)based treatments and for assessing responses, especially in treatment-resistant phenotypes.

Network medicine in the clinical setting of cancer prevention and diagnosis

Integrating drug discovery and pathology with network medicine analysis provides rational and efficient approach to identify novel treatments and allows repurposing clinically approved drugs for human diseases, with implications for personalized medicine. Recent examples illustrate how network medicine has been adapted in the field of epigenetics for diagnosis, precision medicine, and patient stratification in terms of prognosis Table 3 . Table 3 Examples of epigenetics and network-oriented analysis in cancer and CVD susceptibility Network-oriented analysis Sample size Sample source Aim Platform Results References CVDs WGCNA, Comb-p Discovery set 2129 women from the WHI; Replication set: 2726 subjects from the FHS Blood To construct a DNA methylation-oriented network and analyze possible relationships with incident CHD HumanMethylation450 microarray DMRs annotated to SLC9A1, SLC1A5 and TNRC6C strongly correlated with incident CHD Westerman et al. [ 137 ] Co-variation of enhancer activity and gene expression across study participants and GO enrichment 10 end-stage PAH patients at time of lung explant and 9 unused donor control subjects PAECs To construct a regulatory network based on TF-H3K27ac enhancer relationship ChIP-Seq, Illumina RNA-Seq A remodeling of active (H3K27ac) enhancers combined with differential transcription factors may guide a dysregulated angiogenesis and endothelial-to-mesenchymal-transition Reyes-Palomares et al. [ 112 ] Cancer NcADMM algorithm; LLR 562 TCGA ovarian cancer Online data To construct a DNA methylation-oriented network of TCGA ovarian cancer Illumina Infinium HumanMethylation27 platform; Affymetrix HT-HGU133A platform Identified the path associated with CCNE1, AURKA and RAB25 mediated by DNA methylation https://doi.org/10.38/s41598-019-42010-6 GSEA, MSigDB, FEM 64endometrial cancer tissue and 23 healthy control samples Endometrial tissue To study new methylated biomarker test to distinguish endometrial cancers from non-cancers Illumina Infinium HumanMethylation27K BeadChip HAND2 methylation is a common and crucial molecular alteration in endometrial cancer that could potentially be employed as a biomarker https://doi.org/10.1371/journal.pmed.1001551 WGCNA, GEPIA 201 patients of the TCGA prostate cancer TCGA database To build a network analysis correlation of RNA and DNA methylation to identify target therapy Illumina human methylation 450 platform This protocol has predicted the FOXD1 as predictor of poor prognosis https://doi.org/10.2217/epi-2019-0349 WGCNA, ssGSEA, GO and KEGG pathway enrichment 1248 breast cancer patients TCGA database To build a DNA methylation and RNA-seq network for Brest cancer stratification patients RNA-seq and DNA methylation datasets Stratify breast cancer patients into low- and high-risk groups https://doi.org/10.1186/s12967-019-2126-6 GEO, MLP 391 patients of 11 different cancer TCGA database To construct DNA methylation network and gene expression Illumina human methylation 450 k BeadChip; Illumina 450 k platform New application to classified the different cancer type based on DNA methylation levels 32384093 MCODE, K-shell method 780 samples in BRCA, 468 samples in SKCM, and 428 samples in UCEC TCGA database To make a DNA methylation data, mRNA expression data and clinical data network Illumina HumanMethylation 450 K Assay Identification of gene signatures associated with cancer prognosis https://doi.org/10.3390/genes10080571 GREAT, LOLA, ENCODE 30 glioblastoma patients Tissue To study the genomic location and abundance of 5 hmC in glioblastomas to study the disease progression IlluminaHumanMethylation450kmanifest, version 0.4.0; IlluminaHumanMethylation450kanno.ilmn12.hg19, version 0.2.1 Identification of a global loss of 5 hmC in glioblastoma compared with healthy prefrontal cortex tissues https://doi.org/10.1038/ncomms13177 Affymetrix Genome Wide SNP Arrays v6; WGS; ENCODE; HotNet 200 AML Blood To make a DNA-methylation network with RNA and microRNA to investigate the AML pathogenesis, classification, and risk stratification Affymetrix U133 Plus 2 platform; Illumina Infinium HumanMethylation450 BeadChip; Affymetrix SNP Array 6.0; Illumina HiSeq 2000; Illumina GAIIX Identification of pathway that stratified the AML patients https://doi.org/10.1056/NEJMoa1301689 WGS; ATAC-seq; WGBS; ENCODE 410 TCGA samples from 23 cancer types TCGA database To build DNA regulatory elements and gene promoter network for future integrative gene regulatory analyses Illumina MiSeq Sequencer; Identification of transcription factors and enhancers driving molecular subtypes of cancer associated with clinical prognosis 10.1126/science.aav1898 GREAT; GSEA; ATAC-seq Mammary tumors from mouse models and human patients Tissue To create a network-chromatin accessibility and transcriptional profiling during mammary development to identify factors that mediate cancer cell state interconversions Illumina HiSeq 2500 Identification of SOX10 that binds the genes that regulate neural crest https://doi.org/10.1016/j.ccell.2018.08.001 SNF; GO and KEGG pathway enrichment 185pancreatic cancers Tissue To build a mRNAs, miRNAs and DNA methylation network for pancreatic cancer patient stratification Informatic platform Identification various signaling cascade associated with different tumor subtype https://doi.org/10.1038/s41598-020-58290-2 scRRBS; NONCODE, ENCODE 26 single cells isolated from a 51-year-old male HCC patient Tissue To use a DNA methylation, RNA-seq and CNV network in HCC single cell scTrio-seq; Illumina HiSeq2000 or HiSeq 2500 Sequencer Identification of new approach to study the heterogeneity and complexity of cell populations in development and cancer interrogating in the same time the genome, methylome, and transcriptome https://doi.org/10.1038/cr.2016.23 MACs2; ENCODE; GREAT CML cells Cell culture To build an ATAC-seq and RNA-seq network in single cells Illumina HiSeq 4000; NextSeq; qRT-PCR Correlation between GATA and CD24 that induce a high genetic and epigenetic variability, and resistance to imatinib mesylate treatment 28118844 WGBS; WGS; MSigDb 100 castration-resistant prostate metastases Tissue Introduction of whole-genome, whole-methylome and whole-transcriptome sequencing network in metastatic cancer to study the regulatory role of methylation ChIP–seq; RNA-seq; Illumina Novaseq 6000 Identification of a novel epigenomic subtype associated with hypermethylation and somatic mutations in TET2, DNMT3B, IDH1, and BRAF https://doi.org/10.1038/s41588-020-0648-8 Abbreviations: CHD: Coronary Heart Disease; CVDs: Cardiovascular Diseases; EF: Ejection Fraction; FHS: Framingham Heart Study; GEO: Gene Expression Omnibus; GSNCA: Gene Set Net Correlations Analysis; HF: Heart Failure; HTx: Heart Transplantation; miRNAs: microRNAs; PAECs: Pulmonary Arterial Endothelial Cells; PAH: Pulmonary Arterial Hypertension; PCA: Principle Component Analysis; PPIs: Protein–Protein Interactions; RAC1: Ras-related C3 Botulinum Toxin Substrate 1; RT-PCR: Real Time-Polymerase Chain Reaction; SLC1A5: Sodium-Dependent Neutral Amino Acid Transporter; SLC9A1: Na + /H + Antiporter; T2D: Type 2 Diabetes; TF: Transcriptional Factor; TNRC6C: Trinucleotide Repeat Containing Adaptor 6C; WGCNA: Weighted Gene Co-Expression Network; WHI: Women’s Health Initiative;LLR:L1-regularized logistic regression; MLP:multilayer perceptron; BRCA:breast invasive carcinoma; SKCM:skin cutaneous melanoma; UCEC:uterine corpus endometrial carcinoma; SNF:Similarity Network Fusion; HCC:human hepatocellular carcinoma; CNV:Copy Number Variation Analysis; CML: chronic myeloid leukemia; WM164: metastatic human melanoma cell line established from a metastatic site in a 22-year-old male with stage IV superficial spreading melanoma; tCNNS:Convolutional Neural Network for drugs in SMILES format; GDSC:Genomics of Drugs Sensitivity in Cancer; DeepDR:deepdrug response; CCLE: Cancer Cell Line Encyclopedia; DNN:deep neural network; NcADMM: nonconvex alternating direction method of multipliers; WGS: Whole Genome Sequencing; WGBS: whole-genome bisulfite sequencing; GSEA: Gene Set Enrichment Analysis; MSigDB: Molecular Signatures Database; GEPIA: Gene Expression Profiling Interactive Analysis;GO: Gene Ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes; MLP: multilayer perceptron; MCODE: Molecular Complex Detection; ENCODE: Encyclopedia of DNA Elements; NONCODE: non-coding RNAs database

Network medicine and biomarkers in cancer prevention, diagnosis, management and prognosis

DNA methylation is the most widely investigated epigenetic mechanism in network analysis Fig. 1 . Changes in DNA methylation in promoter or other regulatory regions can reveal associations to cancer development and drug resistance [ 2 , 26 , 47 , 82 ] and allow predictions with high accuracy (reviewed in [ 111 ]). For instance, Capper et al. [ 23 ] trained a random forest classifier to distinguish between tumor entities of the central nervous system. The resulting epigenetic signatures typically consist of many features with unknown mechanistic explanation. Moreover, given that features tend to be highly correlated, they can often be replaced in a signature without a loss of accuracy. A consequence of this often-observed lack of robustness [ 133 ] is that we cannot distinguish between surrogates and features causally involved in the disease. For functional enrichment analysis, features (e.g., CpGs or regulatory regions such as promoters, repressors, or enhancers) are typically mapped to the closest gene using tools such as GREAT [ 89 ]. Such a gene-mapping is also necessary to leverage existing networks such as protein–protein interaction (PPI) or gene-regulatory networks in epigenomics data analysis, where the putative target gene with the closest transcription start site is considered. West et al. showed that such a mapping strategy can be used to extract subnetworks of cancer-related differentially methylated genes [ 135 ]. Subsequently, Jiao et al. proposed the functional epigenetic modules (FEM) method [ 67 ], which revealed HAND2 as a methylation hotspot in the endometrium and indicator of drug response in progesterone treatment [ 69 ]. Kim and Sun [ 71 ] showed that PPI networks are beneficial in network-regularized feature selection after dimension reduction. Li et al. used principal component analysis for feature reduction and sparse canonical correlation analysis to infer edge weights for gene pairs. A network-based pathway-extending approach using DNA methylation and gene expression data identified altered pathways [ 77 ]. Rather than relying on a PPI network as prior information, disease-specific regulatory networks can be inferred directly from epigenomic data. With their epigenetic module based on differential networks (EMDN) algorithm, Ma et al. used The Cancer Genome Atlas Program (TCGA) data to construct both a co-expression and a co-methylation network [ 84 ]. Bartlett et al. [ 11 ] used canonical correspondence analysis of methylation profiles to score pairs of interacting genes to construct a gene co-regulatory network. Edges significantly associated with survival were used to extract subnetworks that exhibited functional enrichment relevant to the cancer context investigated here. Recently, a deep neural network (DNN) algorithm applied to DNA methylation data from 7339 patients across 18 different TCGA tumors classified the origin of cancer [ 143 ]. Using crosstalk between genetic modules, Cui et al. [ 35 ] created a co-methylation network based on DNA methylation data. The K-shell algorithm, applied to three types of cancer, invasive breast carcinoma, skin cutaneous melanoma, and uterine corpus endometrial carcinoma, identified the main genes in the modules that are predictive of prognosis and classification. In invasive breast carcinoma, this network method identified ten genes responsible for metastasis and tumor progression. RCHY1 was found at the junction of two modules, closely related to the histone lysine demethylase KDM1B, which plays a key role in methylation and silencing. In addition to the methylation state, modification of the methylene group on DNA can be associated with cancer diagnosis. The conversion of 5-methylcytosine (5mC) to 5-hydroxymethylcytosine (5hmC) was studied in 30 glioblastomas using the OxyBS algorithm with an enrichment analysis via Genomic Regions Enrichment of Annotations Tool (GREAT). The results implicate the depletion of 5hmC in various cancer types, which is associated with an increase in proliferation markers [ 68 ]. Despite these promising early results in the network-based integration of DNA methylation in oncology, clinical translation of epigenomics-based network medicine methods is still lacking. In particular, the integration of epigenomics data in a multi-omics context remains challenging. A weighted correlation network analysis (WGCNA) of 201 patients in a TCGA prostate cancer dataset revealed hypermethylation of FOXD1 as an unfavorable marker for survival [ 141 ]. The combination of mRNA expression and DNA methylation datasets in WGCNA and downstream gene ontology (GO) enrichment using single sample gene set enrichment analysis (ssGSEA) yielded a 13-gene epigenetic signature associated with survival of breast cancer patients. This panel of genes considered upregulation of known cancer-related pathways (e.g., mTOR signaling) to distinguish high- from low-risk cancer cases [ 9 ]. The genomes of 200 clinically annotated adult cases of de novo AML were studied using whole-genome or whole-exome sequencing as well as RNA-seq, microRNA-seq, and DNA methylation profiling. A potential driver mutation in DNA methylation and RNA-seq (e.g., in DNMT3A , NPM1 , CEPBA , IDH1/2 , and RUNX1 ) promoting AML pathogenesis in individual patients was found in all AML samples (Cancer Genome Atlas Research et al. [ 22 ]. The tool SWItchMiner (SWIM) was applied to RNA-sequencing data from TCGA [ 90 ] to characterize disease etiologies and to identify potential therapeutic targets [ 103 ]. In glioblastoma multiforme (GBM), SWIM revealed new insights into the molecular mechanism determining the stem-like phenotype of glioblastoma cells [ 49 ]. Here SWIM implicated FOSL1 as a putative master regulator of a core of four master neurodevelopmental transcription factors (i.e., SOX2, SALL2, OLIG2, POU3F2), whose induction was demonstrated to be sufficient to reprogram fully differentiated glioblastoma cells into stem-like cells [ 125 ]. More recently, SWIM was used in conjunction with network analysis to enhance disease module discovery [ 104 ]. Corces et al. [ 34 ] generated high-quality ATAC-seq data for 410 TCGA samples and identified cancer- and tissue-specific DNA regulatory elements in 23 cancer types. These chromatin accessibility profiles allowed the classification of cancer subtypes with a newly recognized prognostic importance. A gene co-expression network associated with ATAC-seq analysis was built by Dravis et al. [ 39 ] to analyze breast cancer biomarkers. Here, SOX10 was identified as the major transcription factor that binds to genes responsible for neural activity.

Network medicine in cancer patient stratification using single-cell analyses

Network medicine has enabled the identification of molecular markers defining cancer subtypes [ 81 ]. Sinkala et al. stratified 185 patients with pancreatic cancer into two groups via proteomics data. Next, they built patient-similarity networks for each of the molecular data types in TCGA which they then integrated into a joint network using similarity network fusion (SNF) [ 121 ]. SNF clustered the patients into two subtypes based on their joint molecular profiles. In a subsequent step, the K-nearest neighbor (KNN) algorithm and support vector machines (SVM) were used to identify representative biomarkers for these clusters in individual data types.

While it has not yet entered clinical practice, single-cell epigenetics is an exciting new frontier that allows for studying small and previously unrecognized cell populations such as stem cells Table 4 [ 56 , 122 ], preimplantation embryos [ 144 ], and the heterogeneity of subpopulations of human tissues [ 53 , 83 ]. Single-cell epigenetics also has the potential to be applied in the diagnosis and prognosis in cancer treatment, where identifying the mutational profile might potentially impact disease evolution and response to treatment. Several sequencing protocols are developed in Table 4 , including the single-cell bisulfite sequencing (scBS-seq) method to determine the DNA methylome of cell populations [ 122 ]. Recent single-cell approaches combine the methylome with RNA transcription profiling [ 63 ]. Moreover, single-cell chromatin accessibility profiling with scATAC-seq has provided the opportunity to recognize different populations within the same tumor and to discriminate cancer cell populations and their heterogeneity [ 21 ]. Recently, scTRIO-seq, which allows parallel profiling of genomic, transcriptomic, and DNA methylation in single cells, has led to the identification of two different hepatocellular carcinoma (HCC) subpopulations in the same patient. Based on the DNA and transcriptome of 25 HCC cells, Yu Hou et al. found two distinct populations differing in DNA copy numbers, DNA methylation, or RNA expression levels, of which the smaller tumors expressed more invasive cell markers [ 61 ]. scATAC-seq, used to identify co-varying transcription factor-cell surface marker pairs, was combined with scRNA-seq for cell surface marker expression to detect efficiently CD24-labeled chronic leukemia cells as co-varying with chromatin accessibility changes linked to GATA transcription factors [ 79 ]. A positive correlation was found between CD24 and GATA levels, characterized by high genetic and epigenetic variability, conferring resistance to imatinib mesylate treatment. Table 4 Epi-single-cells technique developed for cancer studies Technique Function and description DNA methylation in single cells scBS-seq The first technique developed for methylome analysis. It degrades 90% of DNA and does not distinguish 5mC from 5hmC scRRBS-seq Technique for methylome analysis.Compared to scBS-seq, reduces DNA loss MID-RRBS RRBS in a microfluidic support, obtaining an efficient bisulfite conversion with high DNA recovery PBAT Technique for methylome analysis.It does not involve DNA fragmentation. This goes to reduce the loss MRSE Using the restriction enzymes to recognize methylation sites coupled with PCR amplification, reduced costs, and reaction times RGM It uses a fluorescence reporter to analyze dynamic changes in the state of endogenous methylation, discriminating completely the methylated genome from the single allele scCGI-seq Endonuclease-based method, which improves coverage and efficiency scTRIO-seq scTrio-seq that can be used to simultaneously analyze the genomic copy-number variations (CNVs), DNA methylome, and transcriptome of an individual mammalian cell scMAB-seq Methylome technique that use enzymes for DNA conversion. Distinguishes 5mC and 5hmC from 5fC (5-formylcytosine) and 5-caC (5-carboxycytosine) scAba-seq Methylome technique used to specifically detect 5hmC by restriction enzyme. It has low efficiency DNA methylation analysis with gene expression scMT-seq; scM&T-seq Methylation associated with transcription scTRIO-seq Gene promoter and hypo- and hyper-methylation study scNMT-seq Nucleosome, methylation, and transcription Single-cell histone modification sequencing scChip-seq Method used to analyze protein interactions with DNA scChil-seq; scCUT&Tag-seq; scChIL-seq; scChIC-seq Technologies that isolate the DNA sequences attached to histones with specific marks or transcription factors scATAC-seq Technique that gives the possibility to improve the data obtained with DNA and RNA sequencing, with the information of chromatin accessibility scCOOL-seq Detection of chromatin status/nucleosome localization, DNA methylation, copy number variation, and ploidy ISH-PLA Technique for RNA regulation study, gives information on histone modification by imaging, cell phenotype, and cell–cell interaction

Network medicine for identification of cancer treatment response vs resistance

In the last 50 years, the identification of epigenetic molecular alterations acting as drivers of cancer development and progression have transformed the clinical practice of oncology from non-specific cancer cell elimination with nonspecific chemotherapies to a more cancer-selective approach that leverages molecular profiles [ 12 ]. Predicting drug resistance, and therefore, modifying treatment according to the phenotypic response of the tumor, remains a major challenge. Several resources, such as the Genome of Drug Sensitivity in Cancer (GDSC) database, provide information required to check tumor resistance to specific drugs ( https://www.cancerrxgene.org/ ). Technologies for large-scale genomic and epigenomic profiling allowed the full (bulk) characterization of different tumors, revealing that they share similar driver mutations and enzymatic alterations [ 117 ]. For example, upregulation of SETDB1 and SETDB2 was found in resistant cells exhibiting a loss of H3K4me3 and H3K27me3 and an increase in H3K9me3 [ 3 ]. HDAC deregulation produced an aggressive phenotype in lung cancer cells resistant to doxorubicin [ 40 ]. The resistance to epigenetic therapies for a subpopulation of leukemia stem cells found in AML models [ 50 ] was not driven by genetic evolution, but was due to epigenetic plasticity [ 72 ]. Epigenetic modifications to centrosome proteins led to tumor development and drug resistance. The Manteia [ 127 ] gene ontology analysis data system reported a correlation between lysine acetyltransferase (KAT) 2A/B alteration and serine/threonine-protein kinase PLK4 overexpression, producing resistance to tamoxifen and trastuzumab [ 51 ]. ChIP-seq and DNA methylation profiles were used to study epigenetic profiles in drug-resistant melanoma, lung, and colon cancers. Thirteen genes associated with the interferon (IFN) pathway were found to be regulated by histone modifications, including the histone methyltransferase EZH2 [ 3 ]. Recent trials found that the response to a targeted drug depends on the anatomical cancer type. For example, BRAF-mutated melanoma, NSCLC, and hairy cell leukemia satisfactorily responded to vandetanib, a drug that targets the BRAF V600E mutation, while BRAF-mutated CRC does not [ 117 ]. Falcone et al. used a network approach based on the SWIM algorithm [ 44 ] to compare pairs of BRAF-mutated cancers and found a great number of switch genes suggesting that the cancer network of each tumor is different. A number of putative targetable kinases encoded by switch genes were reported in lung adenocarcinoma and thyroid cancer, while only one was found in colorectal cancer. Interestingly, the results were in accordance with clinical trial data showing a better response rate to vemurafenib in papillary thyroid cancer patients (overall response rate of 38.5%) than in colorectal cancer patients (ORR 4.8%) [ 44 ]. These results highlight the limit of the reductionist approach where typically one gene is implicated with one disease, as this strategy does not reflect the complexity of complex diseases such as cancer and CVDs in which many potential disease genes interact.

Yildirim et al. [ 138 ] built a drug–target network, in which each drug was connected to its target proteins, and two complementary projections of it: a drug network, in which nodes are drugs and two drugs are connected to each other if they share the same targets, and a target–protein network, in which the nodes are proteins and they are connected together if are targeted by the same drug. The authors observed that new drugs tend to target the already validated target proteins and that many clinically FDA-approved drugs do not target known disease-associated genes but are rather palliative drugs. A strategy aimed to increase the efficacy and to reduce the risk of adverse effects of monotherapy, is the therapy with a combination of multiple drugs. By analyzing the network-based relationship between drug pairs, their targets, and the proteins in the disease modules Cheng et al. revealed that drugs used in combination with synergistic effect have their drug–target modules overlapping with disease modules but well separated to each other in the human interactome [ 30 , 31 ]. Additionally, Tang et al. [ 126 ] developed a logic-based network pharmacology modeling approach, called TIMMA (Target Inhibition interaction using Minimization and Maximization Averaging), based on the integration of drug–target interaction profiles and single-drug sensitivities, to predict synergistic drug combinations. By applying TIMMA to the single-drug sensitivity profiles and to the kinome-wide drug–target interaction of 41 kinase inhibitors in MDA-MB-231 cell line, the authors found a synergistic target interaction between inhibition of Aurora B, a key regulator of mitosis, and ZAK, a key regulator of p38 MAPK pathway, findings also confirmed in vitro.

Twin Convolutional Neural Network for Drugs in SMILES format (tCNNS) creates a drug correlation network using the SMILES chemical representation and information on the cell phenotype [ 80 ]. Pharmacological dose–response prediction can be also obtained by DeepDR based on mutation data from TCGA, a pre-trained expression encoder, and a predictor network for drug response. From 9.059 tumor samples, DeepDR predicted cancer drug resistance and personalized therapy [ 32 ]. For resistant tumor treatment, one of the main unanswered questions is whether the cell acquires mutations during development, or whether there is a small group of cells within the same population that survives therapy. Single-cell analysis specifically allows for studying the complexity of a population [ 70 ]; however, the use of NM in epigenetics for the prediction of response versus treatment is only its inception [ 104 ].

Epi-therapy in cancer: can Network Medicine help physicians at the bedside?

Integrating network medicine with anticancer drug discovery programs would make a significant contribution to improving patient health. Recently, whole-genome bisulfite sequencing (WGBS) integrated with whole-genome sequencing (WGS) and RNA-seq was applied to 100 metastatic prostate biopsies to sequence the methylome whole genome, and transcriptome, respectively [ 142 ]. In 22% of resistant metastases, DNA methylation analysis identified an epigenetic subtype associated with hypermethylation and mutation of TET2, DNMT3B, IDH1, and BRAD, and regions where methylation is associated with AR, MYC, and ERG expression. NM has also been used in this context to predict potential implications of therapeutics. The administration of DNMTi, such as azacitidine or decitabine, which are used for myelodysplastic syndrome (MDS) treatment, may be advantageous in resistant metastases associated with hypermethylated gene promoters. Pursuing a strategy that allows for the identification of tumor subtype targets or a specific resistance mechanism may make it possible to repurpose drugs used in the treatment of other diseases.

Epigenetics and risk stratification of CVDs: network medicine is on the horizon

Metabolic syndrome, diabetes, and foodome project

A notable expansion of the prevalence of metabolic syndrome (MS) has impact on the global risk of CVDs, in particular in aging Western world populations, which are prone to a high-fat diet and a sedentary lifestyle [ 42 ]. MS arises from complex crosstalk between genetic and epigenetic factors underlying obesity, in particular central abdominal obesity, that can strongly increase the risk for coronary heart disease (CHD), type 2 diabetes mellitus (T2D), and cancer [ 42 , 64 , 98 , 99 ]. In clinical practice, a panel of five well established cardiometabolic risk factors including abdominal obesity, increased triglycerides, decreased HDL cholesterol, hypertension, and hyperglycemia is assessed by physicians, and at least three of these must be present [ 42 ]. However, in precision medicine and personalized therapy, individual DNA methylation profiles might not only represent a promising tool for the prediction, diagnosis, and prognosis of obesity and MS, but also for improving treatments to lose body weight [ 113 ]. DNA methylation changes are key molecular drivers underlying the risk of MS upon detrimental exposures (e.g., nutritional patterns), especially during early development as well as during postnatal life, offering a possible framework by which to explain the “missing heritability” of MS [ 36 , 113 ]. Recently, a targeted DNA pyrosequencing and logistic regression analysis has revealed a significant positive relationship between DNA methylation levels at specific CpG islands in promoters of the PPARα and LPL genes and serum triglyceride levels (TG) in visceral adipose tissue samples from 53 MS patients in comparison with 55 healthy subjects [ 24 ]. In addition, a negative association linked methylation levels of the tumor necrosis factor gene with TG, glucose levels, HDL-c, and blood pressure suggesting a relevant factor potentially involved in preventing MS occurrence [ 24 ].

A study using the Illumina Methylation EPIC Beadchip on 1999 blood samples isolated in the Coronary Artery Risk Development in Young Adults (CARDIA) study has unveiled a strong positive association between accelerated epigenetic aging and both the MS severity score and the risk of developing MS, after adjusting for known risk factors [ 97 ]. This observation led to suggest that pharmacological and non-pharmacological interventions targeting the epigenetic aging process at the molecular level may potentially prevent MS.

Network analysis would be of paramount importance to understand the underlying molecular determinants and pathogenic processes that guide in the development of new predictive biomarkers and prevention tools. Thus far, network analysis has already been applied to separate MS components, mainly in T2D [ 48 , 98 , 99 , 119 ] and obese-T2D patients [ 62 ], and to construct gene regulatory networks, these analyses, however, did not consider epigenetic factors.

The epigenetic clock and nutritional epigenomics may contribute to aging as well as age-related diseases [ 4 ]. Thus, mapping food-related chemical profiles (diet) and molecular pathways is the main goal of the network-based “Foodome” project ( https://www.barabasilab.com/projects ). Foodome emphasizes the use of digital eating patterns (“barcode”) to define the personalized exposures to nutritional–chemical compounds [ 10 ]. Merging the barcode with genetic profiles and clinical data might fuel innovative platforms that can provide insights into the molecular basis for the “diet-mutation-disease risk” axis useful for precision medicine [ 10 ]. Since epigenetics plays a important role in the diet–genome crosstalk underlying vascular damage and CV risk, even during fetal development [ 36 , 98 , 99 , 101 ], the “individual foodome” could be integrated in longitudinal cohort studies investigating how epigenetic features change over time and in response to different nutritional exposures leading to the onset of CVDs.

Coronary heart disease (CHD)

Endothelial and systemic inflammation plays a relevant role in disease pathophysiology, and changes in DNA methylation levels of targeted genes may be causal or predispose to disease, contributing to destabilization and rupture of atherosclerotic plaques leading to acute cardiovascular events [ 64 , 65 , 101 , 115 ]. A recent study has emphasized the possible role of a DNA methylation-based risk score in optimizing the traditional predictors of CVD risk [ 136 ]. In addition, there is a possible correlation between blood-based methylation levels in the unique CGI regulating the human leukocyte antigen-G (HLA-G) gene, which encodes for an anti-inflammatory molecule with immunomodulatory properties, and cardiac computed tomography angiography (CCTA) features in patients with obstructive in comparison with non-obstructive CHD [ 116 ]. Hypomethylation of a specific fragment of CGI-associated HLA-G gene positively correlated with coronary calcium score and was predictive for disease severity suggesting that methylation might not only have a critical role in disease severity but also a role as noninvasive biomarker(s) improving the prognostic value of CCTA [ 116 ]. The WGCNA and Comb-p algorithms have been applied to the identification of blood-based differentially methylated regions (DMRs) and disease modules associated with incident CHD events in two independent cohorts (discovery set: 2129 women, replication set: 2726 subjects) [ 137 ]. This study has identified two modules highly enriched for development and immune-related processes. In addition, a multivariate analysis has revealed a positive correlation with BMI, highly sensitive C reactive protein (hsCRP), and TG [ 137 ]. Three DMRs annotated to the sodium/hydrogen exchanger 1 (SLC9A1), solute carrier family 1 neutral amino acid transporter member 5 (SLC1A5), and trinucleotide repeat containing adaptor 6C (TNRC6C) genes significantly replicated across the two cohorts, providing possible useful predictive biomarkers [ 137 ]. However, the possible cause–effect relationship between methylation changes in these genes and CV risk needs to be determined.

Pulmonary arterial hypertension (PAH)

PAH is a rare and incurable disease characterized by vasoconstriction and consequent elevated pulmonary artery pressure owing to three main endophenotypes, endothelial dysfunction, cell proliferation/migration, and inflammation triggered by an interplay between genetic and epigenetic risk factors with exposure to detrimental environmental stimuli [ 25 , 86 , 100 ]. Data regarding the potential clinical relevance of differential epigenetic factors in PAH, mainly changes in DNA methylation, have been increasing in the past few years, and network-oriented approaches are helping to prioritize novel candidate genes and drug targets [ 100 ]. Recently, an integrated regulatory network has been constructed by integrating chromatin with transcriptomic and interaction profiling in pulmonary arterial endothelial cells (PAECs) obtained from end-stage PAH patients at the time of lung explant and control subjects [ 112 ]. As a result, an in-depth remodeling of active enhancers marked by H3K27ac and regulated by specific transcription factors may trigger perturbation of angiogenesis and endothelial-to-mesenchymal transition processes in PAECs in response to specific growth factor signals, as experimentally confirmed for target genes such as nitric oxide synthase 3 (NOS3) [ 112 ]. However, further studies will investigate a possible correlation between key gene regulatory networks and underlying PAH severity or responsiveness to vasodilatory therapy.

Heart failure (HF) and heart transplantation (HTx)

Heart failure (HF) affects approximately 20% of general population and contributes to 11% of deaths with an estimated incidence that will rise by 25% over the next 15 years [ 16 ]. HF can develop asymptomatically for years, and once diagnosed, the effectiveness of most drug therapy interventions are modest. Nonetheless, the recent tremendous impact, both of SGLT2 inhibitors and angiotensin–neprilysin inhibition, will likely reduce HF mortality [ 66 , 132 ]. However, the development of novel early biomarkers useful for stratification and/or as prognostic markers is a priority in the clinical setting of HF. Based on the ejection fraction (EF) value, we can classify three clinical phenotypes: HF with reduced EF (HFrEF), HF with mid-range EF (HFmrEF), and HF with preserved EF (HFpEF) [ 98 ]. An epigenetic-based phenotype mapping strategy of HFrEF, HFmrEF, and HFpEF patients seems to be a possible option to identify noninvasive biomarkers discriminating differential HF subgroups and/or novel drug targets to test in clinical trials aimed at establishing specific therapeutic strategies for each phenotypic profile [ 99 ]. However, the availability of cardiac biopsy from living patients and controls in humans is scarce; thus, most pathogenic studies are based on animal models. A pioneer multi-omics study integrated the methylome and transcriptome of left-ventricular biopsies and whole peripheral blood samples of 41 patients with HFmrEF caused by dilated cardiomyopathy and compared to 31 patients who underwent routine left-heart myocardial biopsy after receiving transplantation, as a control group. The promoter DNA hypomethylation of the natriuretic peptide A and B (NPPA and NPPB) genes regulates overexpression of these genes providing a novel putative class of HF biomarkers easily detectable in peripheral blood [ 91 ].

Graft surveillance after heart transplantation is a challenge in the management of transplanted patients, and current guidelines indicate that invasive endomyocardial biopsy is the gold standard to diagnose and monitor organ rejection. Ideally, graft rejection may be diagnosed and predicted by noninvasive biomarkers present in the peripheral blood or other biological fluids [ 99 ]. Specifically, an increasing interest in circulating epigenetic molecules is providing novel findings that may aid physicians in a more accurate risk stratification [ 99 ]. DNA methylation is a pivotal contributor to a balanced immune response toward the graft, due to regulation by both the innate and adaptive immune systems, with primarily T cells as the key players of alloreactivity and targets for immunosuppressive drugs [ 27 , 28 ]. Interestingly, FOXP3 gene expression was significantly higher in biopsy samples of rejectors collected before rejection in comparison with non-rejectors, and showed the tendency to predict rejection events [ 20 ]. Longitudinal studies could evaluate the possible role of this biomarker in the clinical setting of HTx.

Repurposed drugs and epitherapy in CVDs: network medicine can improve the current reductionist approach to reach the goal of personalized treatments

As the discovery and development of novel drugs is a highly expensive time-consuming process, the repurposing of “old” drugs to treat CVDs is increasingly becoming an attractive goal. The repurposing of both metformin and statins has been widely evaluated in large controlled clinical trials for the prevention and treatment of CHD, PAH, HF, and complications after HTx Table 2 . Metformin proved in 1994, is the first-line oral drug for treatment of T2D and obesity ( www.ncbi.nlm.nih.gov/books/NBK409379 ). Metformin has glucose-lowering effects by acting on several molecular pathways and also represents an agonist of the SIRT1 enzyme [ 98 , 99 ]. Additionally, the first statin (lovastatin) received FDA approval in 1987 and now 6 statins, including simvastatin and pravastatin (semi-synthetic), as well as fluvastatin, atorvastatin, rosuvastatin, and pitavastatin (synthetic), have been introduced to the market providing first-line of oral drugs prescribed for dyslipidemias and prevention of atherosclerotic plaque development [ 41 ]. Basically, statins block the 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase enzyme leading to lipid-lowering effects and also showing epigenetic interference (HDACi). The cardio-protective and anti-inflammatory effects of metformin and statins have also been widely demonstrated in patients affected by CVDs [ 98 , 99 , 101 ]; however, large controlled trials are needed to establish whether these beneficial effects are conferred by the glucose and lipid-lowering effects, by interference with specific epigenetic-sensitive pathways, such as inflammatory pathways or by both. Another example of drug repurposing includes tocilizumab (Actemra), a monoclonal anti-IL-6 antibody approved by FDA in 2010 for treatment of rheumatoid arthritis. In addition, there are a few examples of properly defined epi-drugs under clinical evaluation for treatment of CVDs. One of the most promising is apabetalone (RVX-208), inhibiting selectively the bromodomain and extra-terminal domain (BET) family of proteins binding to acetyl groups in order to normalize acetylation imbalance underlying cardiac dysfunction in T2D, CHD, PAH, and HF Table 2 .

Ongoing and completed clinical trials

In contrast to cancer patients, epitherapy has not yet reached clinical practice in the management of CVDs. However, exploring the website http://clinicaltrial.gov/ , we retrieved a large number both of ongoing observational and interventional clinical trials (Phase 3 and 4) of these agents in CVDs Table 2 demonstrating the great interest in translating epigenetic findings from the bench to bedside to treat patients affected by CVDs.

Diabetes and CHD

A completed double-blinded, randomized, placebo-controlled Phase 3 trial ( NCT02773927 ) compared the effect of metformin/agave inulin in comparison with agave inulin on adiponectin levels in patients with metabolic syndrome by grouping patients into in four clusters: group A, metformin plus agave inulin; group B, metformin plus placebo of agave inulin; group C, agave inulin plus placebo of metformin; group D, placebo of agave inulin plus placebo of metformin. To date, there are no published results from this trial. Metformin is also currently under investigation in a Phase 2, observational, prospective clinical trial ( NCT01884051 ) enrolling 1.899 participants, in which the primary endpoints are measures of insulin resistance, urinary, and plasma oxidant stress markers, right ventricle lipid content, oxidative metabolism, and drug safety. As secondary endpoints, the quantification of lung metabolism, [18F]‐fluorodeoxyglucose uptake, blood-based expression of the bone morphogenic protein receptor (BMPR) 2, right ventricle ejection fraction and volumes using magnetic resonance imaging, insulin resistance, and 6-min walking distance (6MWD) is evaluated. Additionally, the preliminary results from an ongoing Phase III trial ( NCT02586155 ) have demonstrated possible efficacy in preventing myocardial damage in high-risk T2D-CHD patients under high-intensity statin therapy combined with RVX 208 (daily dose 100 mg capsule plus atorvastatin and rosuvastatin).

Pulmonary arterial hypertension

At the molecular level, trimetazidine can switch the metabolic cellular state from beta-oxidation toward glucose oxidation by inhibiting synthesis or carnitine-facilitated transport of fatty acids so that cardiomyocytes can obtain more energy [ 131 ]. Recently, it has been reported that trimetazidine may exert its cardio-protective role in women by affecting the DNA methylation profile of the cyclin dependent kinase inhibitor 2b (CDKN2B) gene [ 27 , 28 ]. Trimetazidine is being studied in a randomized controlled trial ( NCT03273387 ) which aims to determine if 3 months of treatment (35 mg twice a day) combined with standard therapy can alter right ventricle function in PAH patients. The primary endpoint will be a change in right ventricular function, as quantified by cardiac magnetic resonance imaging (MRI). Secondary endpoints will contain modifications in cardiac fibrosis quantified by T1 cardiac MRI mapping, functional class, and plasma levels of lactate dehydrogenase (LDH). However, tocilizumab is being investigated in the “ Therapeutic Open‐Label Study of Tocilizumab in the Treatment of Pulmonary Arterial Hypertension study ” ( NCT02676947 ), a single‐arm study aimed at investigating whether treatment with tocilizumab in a dose of 8 mg/kg monthly for 6 months may alter pulmonary vascular resistance in PAH patients [ 59 ]. The primary endpoints will be a change in pulmonary vascular resistance (PVR) and safety, with secondary endpoints including 6MWD, N-terminal pro-B-type natriuretic peptide (NT‐pro‐BNP), symptom burden, and quality of life. This trial is still ongoing Apabetalone is under investigation in a single‐arm trial in a two‐center study enrolling PAH patients combining the standard therapy with 100 mg of apabetalone twice a day for 16 weeks ( NCT03655704 ).

Heart failure (HF) and heart transplantation (HTx)

An interesting example of drug repurposing arises from the randomized, double-blind, placebo-controlled cross-over intervention DoPING-HFpEF trial (EU Clinical Trial Register: 2018-002170-52; NTR registration: NL7830), which will evaluate the possible cardio-protective effects of trimetazidine in patients affected by HFpEF [ 131 ]. A large randomized, double-blind study enrolling 6.975 HF patients with monitored EF provided evidence for which assumption of 1 g per day of long-chain omega-3 polyunsaturated fatty acids (PUFA) is associated with a small reduction (9%) in mortality and admission to the hospital for CV events in HF patients [ 128 ]. Dietary supplementation of PUFA may aid in normalizing circulating lipid levels, exerting beneficial systemic anti-inflammatory effects, preventing cardiac injury by affecting blood-based global DNA methylation levels [ 98 , 101 ].

Which additional clinical benefits would the network medicine approach add?

The interactome may reveal differential pathogenic molecular drivers in each patient that are, in part, responsible for current limitations of the one-size-fits-all approach ([ 15 , 64 , 76 , 86 ]). Indeed, CVD patients treated with optimized standard-of-care regimens can still show high residual morbidity and mortality risk [ 5 ]). Thus, interactome-based selection before enrolling subjects in a clinical trial may aid in selecting more homogeneous study populations in which to test a specific drug therapy. NM approaches would potentially resolve this heterogeneity by combining the interactome with biochemical and molecular assays and clinical information, potentially providing noninvasive biomarkers as well as drug targets ([ 14 , 64 , 98 , 100 , 120 ]). A network-oriented paradigm of CVD pathogenesis would have the power of repurposing existing FDA-approved compounds by in silico prediction of targeted disease modules to speed up the discovery of putative novel personalized treatments [ 29 ]. Moreover, repurposing bioinformatic tools could offer new molecular mechanisms of existing drugs. As an example, the seed connector algorithm (SCA) has been applied to GWAS-derived CHD seed proteins and has suggested that the neuropilin-1 (NRP1) would be potentially a novel candidate disease gene [ 134 ]. Since NRP1 is the target of the anti-angiogenic agent pegaptanib, which is indicated for the treatment of neovascular age-related macular degeneration, the SCA algorithm approach suggested its possible repurposing for atherosclerotic diseases [ 134 ]. In addition, the use of network-based proximity measures would allow to quantify the relationship between CVD modules and drug targets helping to chart novel associations and discriminate if a candidate repurposed drug may be therapeutically effective or lead to unwanted side effects [ 55 ]. Similarly, another proximity score has highlighted a relevant role for dysregulation of the immune system in MS development, suggesting the repurposing of ibrutinib, a BTK inhibitor prescribed for hematological malignancies, to counteract the inflammatory state [ 93 ].

Concluding remarks network medicine’s role in clinical medicine, challenges and a pathway forward to the new of precision medicine

The management of hematological malignancies has already seen a major benefit from epigenetics with nine epi-drugs currently approved by the FDA. Although clinical epigenetics is far less advanced in CVDs, large clinical trials show the promising results with respect to the effectiveness of drug repositioning and epitherapy in the treatment of MS, T2D, CHD, PAH, HF, and HTx. Therefore, in areas relevant to the cause of death, such as CVDs, there need to be more expeditious ways to repurpose drugs as we look toward the era of precision medicine.

The reasons for this gap may also be due to the different procedures for approval of clinical studies in cancer versus CVD. Several epigenome-wide association studies have revealed molecular pathways involved in the pathogenesis of cancer and CVDs that may offer robust biomarkers for precision medicine and personalized therapy. Novel technological developments as well as the application of NM may change our view of the role and analysis of epigenome deregulation in disease. The development and use of bioinformatics (and tools of artificial intelligence) hints at a deep change when dealing with human health, disease identification and handling.

NM applications may improve our mechanistic understanding of tumorigenesis and the dynamics of driver and contributing (epi)mutations within the 3D structure of the cell and of tissues. The indication that a cytosine is first methylated and then hydroxymethylated suggests that the order of (some) epigenome changes should be possibly integrated mining the epigenome landscape in health and disease. Furthermore, different chromatin complexes (such as the readers and some erasers) use metabolic co-factors in their reactions, suggesting that, among the regulatory networks, the availability of those cofactors might, in turn, also regulate the prioritization of events. Indeed, the integration of genome and epigenome information together with the metabolic status of the cells might, in principle, be a necessary step forward to stratify better epi-biomarkers and epi-targets in disease.

Studying the role of spatiotemporal regulation of chromatin with integrative network approaches may represent the next step forward for personalized medicine, allowing drug-based fine-tuning of the epigenome. While a small number of computational tools already leverage epigenetic profiling data in a network context, most existing approaches are limited to promoter methylation, neglecting the influence of regulatory regions distal to promoters. Recently, the EpiRegio database established a link between cis - or trans -regulatory elements and their target genes [ 13 ]. In combination with disease-specific molecular profiling data, network enrichment [ 78 ] or disease module discovery [ 33 ] can be employed on such networks to reveal more distal and complex regulatory interactions for developing clinically relevant insights into epigenetic mechanisms in diseases in the future. However, these approaches are not yet well understood by physicians. Learning workshops on this groundbreaking field are necessary for the fields of cardiology and oncology. Interpreting the developments of epigenome deregulation to the clinic may also need a further understanding of the use of epi-drugs. For example, physicians should be aware that the reduced locus-selective specificity may lead to (epi)genome undesired effects. Novel drug discovery approaches targeting DNA mutations might be a more focused solution, as indicated for the EZH2 inhibitors discussed in the previous sections. In addition, methods such as “ the proteolysis targeting chimera (PROTAC) ” might produce specific drugs, perhaps able to (re)modulate the function of non-enzymatic chromatin complexes.

A future challenge will, thus, be to integrate such notions in a network context together with other modalities such as transcriptomics, proteomics, and miRNA expression as well as with clinical information. We expect that integrative network analysis will reveal regulatory patterns that can be exploited for diagnosis, prognosis, and treatment selection. We need to encourage the scientific integration of basic scientists and physicians in the field of clinical epigenetics. A relevant role is also played by innovative technology assets applied to the clinical setting and management of patient. None of NM approaches has yet reached clinical application, but some have been validated with excellent results in ex-vivo patients Table 5 . Therefore, the final step will be to evaluate the potential of novel NM platforms in large clinical trials to test their reliability value in diagnosing, prognosticate, and treating cancer and CVDs, as two of the major causes of death worldwide. Table 5 Network medicine approach in cancer and CVDs Network medicine System validation Data obtained FEM Endometrial tumor samples and 23 healthy controls DNA methylation of HAND2 gene in 90% of tumors Jiao et al. [ 67 ], Jones et al. [ 69 ] OxyBS Fresh frozen glioblastoma specimens from 30 subjects diagnosed between 2004 and 2012 5-Hydroxymethylcytosine patterns are strongly related to transcription, localizes to disease-critical genes and are associated with patient prognosis Johnson et al. [ 68 ] WGBS + WGS Biopsy 100 castration-resistant prostate metastases Discovery of association between hypermethylation and somatic mutations in TET2, DNMT3B, IDH1 and BRAF in epigenomic subtype Zhao et al. [ 142 ] WGCNA Blood samples isolated from 2627 subjects enrolled in FHS cohort Differences in circulating DNA methylation signatures located in regulatory regions of SLC9A1, SLC1A5, TNRC6C genes may be useful biomarkers to predict incident CHD Westerman et al. [ 137 ] Abbreviations: FEM: Functional Epigenetic Modules; OxyBS: oxidative bisulfite and bisulfite; WGBS: whole-genome bisulfite sequencing; WGS: whole-genome sequencing; WGCNA: weighted correlation network analysis; FHS: Framingham Heart Study

Abbreviations

α-Hydroxyglutarate

α-Ketoglutarate

5-Hydroxymethylcytosine

5-Methylcytosine

Acute myeloid leukemia

Bromodomain and extra-terminal domain

Body mass index

Bone morphogenic protein receptor

Bruton’s tyrosine kinase

Cardiac computed tomography angiography

Cyclin-dependent kinase inhibitor 2b

CCAAT enhancer binding protein alpha

Coronary heart disease

Chronic myeloid leukemia

Cytosine-phosphate-guanine dinucleotide

Complete response

Castration-resistant prostate cancer

Cutaneous T-cell lymphoma

Cardiovascular disease

Diffuse large B-cell lymphoma

Epigenetic module based on differential networks

Enhancer of zeste 2 polycomb repressive complex 2 subunit

Food and drug administration

Finite element method

Follicular lymphoma

Forkhead Box D1

Forkhead box P3

Glioblastoma multiforme

Genome of drug sensitivity in cancer

Gene ontology

Genomic regions enrichment of annotations tool

DNA methyl transferase

Deep neural network

Histone 3 lysine 4 methylation

Histone 3 lysine 9 methylation

Histone 3 lysine 27 acetylation

Heart and neural crest derivatives expressed 2

Histone acetyl transferase

Hepatocellular carcinoma

Histone deacetylases

High-density lipoprotein

Heart failure

Human leukocyte antigen-G

3-Hydroxy-3-methylglutaryl coenzyme A

High sensitive C reactive protein

Heart transplantation

Isocitrate dehydrogenase

Interferon

Lysine acetyl transferase

Lysine demethylase

K-nearest neighbor

Lactate dehydrogenase

Lipoprotein lipase

Mitogen-activated protein kinase

Myelodysplastic syndrome

Mixed-lineage leukemia

Multiple myeloma

Metabolic syndrome

Network medicine

Nitric oxide synthase 3

Natriuretic peptide A

Natriuretic peptide B

Neuropilin-1

N-terminal pro-B-type natriuretic peptide

Pulmonary arterial endothelial cells

Pulmonary arterial hypertension

Polo-like kinase 4

Peroxisome proliferator-activated receptor alpha

POU class 3 homeobox 2

Protein–protein interactions

Peripheral T-cell lymphoma

Long-chain omega-3 polyunsaturated fatty acids

Oligodendrocyte transcription factor

Osteosarcoma

Oxidative bisulfite and bisulfite

Relapsed/refractory

RUNX family transcription factor 1

Spalt-like transcription factor 2

Squamous cell carcinoma of the head and neck

SET domain bifurcated histone lysine methyltransferase 1/2

Sodium/glucose cotransporter 2

Silent information regulator 1

Sodium/hydrogen exchanger 1

Solute carrier family 1 neutral amino acid transporter member 5

SWI/SNF related, matrix associated, actin-dependent regulator of chromatin, subfamily A member 2

SWI/SNF-related matrix-associated actin-dependent regulator of chromatin subfamily B member 1

SRY-Box transcription factor 10

Single sample gene set enrichment analysis

Support vector machines

SWItchMiner

Type 2 diabetes

The Cancer Genome Atlas

Twin convolutional neural network for drugs in SMILES format

Ten–eleven translocation

Target inhibition interaction using minimization and maximization averaging

Triglyceride

Tumor necrosis factor

Trinucleotide repeat containing adaptor 6C

Whole-genome bisulfite sequencing

Weighted correlation network analysis

Whole-genome sequencing

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Federica Sarno, Giuditta Benincasa, and Markus List are co-first authors

Lucia Altucci and Claudio Napoli are co-last authors.

Acknowledgements

We thank Dr. C. Fisher for linguistic editing. All authors are affiliated with the the International Network Medicine Consortium.

Authors’ contributions

Claudio Napoli and Lucia Altucci designed the study in the context of the International Network Medicine Consortium; Federica Sarno, Giuditta Benincasa, Markus List, Lucia Altucci, Claudio Napoli, Sebastiano Filetti, Antonella Verrienti, and Paola Paci drafted and edited the manuscript; Jan Baumbach, Albert-László Barabási, Fortunato Ciardiello, Kimberly Glass, Joseph Loscalzo, Cinzia Marchese, Bradley M. Maron, Paolo Parini, Enrico Petrillo, and Edwin K. Silverman defined the content and edited the manuscript. All authors read and approved the final manuscript.

Funding

Lucia Altucci’s research was funded by AIRC-17217; VALERE: Vanvitelli per la ricerca; Campania Regional Government Technology Platform lotta alle patologie oncologiche: ICURE-b21c17000030007; Campania Regional Government FASE2: IDEAL-b63d18000560007; MIUR, proof of concept POC01_00043-EPICUREPOC01_00043-b64i19000290008; POR Campania FSE 2014/2020 asseiii-b27d18001070006. Jan Baumbach project has received funding from the European Union’s Horizon 2020 research and innovation programme (777111) and supported by the German Federal Ministry of Education and Research (BMBF) within the framework of the *e:Med *research and funding concept (grant 01ZX1910D). Claudio Napoli research group is supported by PRIN2017F8ZB89 from Italian Ministry of University and Research (MIUR); Ricerca Finalizzata and Ricerca Corrente from the Italian Ministry of Health. This publication reflects only the authors’ view and the European Commission is not responsible for any use that may be made of the information it contains.

Availability of data and materials

Not applicable.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

Federica Sarno is a member of EPI-C srl; Albert-László Barabási: is the co-founder of Scipher, Foodome, and founder of Foodome and Nomix, companies that bring network science to health care; Joseph Loscalzo: is co-founder and member of the scientific advisory board of Scipher (Network Medicine Company) since 2013; Edwin K. Silverman: institutional grant support from GSK and Bayer; Lucia Altucci: is a co-founder of EPI-C srl and is a consultant of Merck Serono Italy; Giuditta Benincasa, Markus List, Jan Baumbach, Fortunato Ciardiello, Sebastiano Filetti, Kimberly Glass, Cinzia Marchese, Bradley M. Maron, Paola Paci, Paolo Parini, Enrico Petrillo, Antonella Verrenti, Claudio Napoli: none.

References

1. Adhikary G Grun D Balasubramanian S Kerr C Huang JM Eckert RL Survival of skin cancer stem cells requires the Ezh2 polycomb group protein Carcinogenesis 2015 36 7 800 810 10.1093/carcin/bgv064 25969142 PMC4580538

2. Akhavan-Niaki H Samadani AA DNA methylation and cancer development: molecular mechanism Cell Biochem Biophys 2013 67 2 501 513 10.1007/s12013-013-9555-2 23508887

3. Al Emran A Marzese DM Menon DR Stark MS Torrano J Hammerlindl H Zhang G Distinct histone modifications denote early stress-induced drug tolerance in cancer Oncotarget 2018 9 9 8206 8222 10.18632/oncotarget.23654 29492189 PMC5823586

4. Amenyah SD Ward M Strain JJ Nutritional epigenomics and age-related disease Curr Dev Nutr 2020 4 7 97 10.1093/cdn/nzaa097 PMC7335360 32666030

5. Arnett DK Blumenthal RS Albert MA Buroker AB Goldberger ZD Hahn EJ Himmelfarb CD Khera A Lloyd-Jones D McEvoy JW Michos ED Miedema MD Muñoz D Smith SC Jr Virani SS Williams KA Sr Yeboah J Ziaeian B Correction to: 2019 ACC/AHA Guideline on the Primary Prevention of Cardiovascular Disease: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines Circulation 2020 141 16 e773 10.1161/CIR.0000000000000770 32310698

6. Arnold M Rutherford MJ Bardot A Ferlay J Andersson TM Myklebust TÅ Tervonen H Thursfield V Ransom D Shack L Woods RR Turner D Leonfellner S Ryan S Saint-Jacques N De P McClure C Ramanakumar AV Stuart-Panko H Engholm G Walsh PM Jackson C Vernon S Morgan E Gavin A Morrison DS Huws DW Porter G Butler J Bryant H Currow DC Hiom S Parkin DM Sasieni P Lambert PC Møller B Soerjomataram I Bray F Progress in cancer survival, mortality, and incidence in seven high-income countries 1995–2014 (ICBP SURVMARK-2): a population-based study Lancet Oncol 2019 20 11 1493 1505 10.1016/S1470-2045(19)30456-5 31521509 PMC6838671

7. Bachmann IM Halvorsen OJ Collett K Stefansson IM Straume O Haukaas SA Salvesen HB Otte AP Akslen LA EZH2 expression is associated with high proliferation rate and aggressive tumor subgroups in cutaneous melanoma and cancers of the endometrium, prostate, and breast J Clin Oncol 2006 24 2 268 273 10.1200/JCO.2005.01.5180 16330673

8. Banerjee R Mani RS Russo N Scanlon CS Tsodikov A Jing X Cao Q Palanisamy N Metwally T Inglehart RC Tomlins S Bradford C Carey T Wolf G Kalyana-Sundaram S Chinnaiyan AM Varambally S D’Silva NJ The tumor suppressor gene rap1GAP is silenced by miR-101-mediated EZH2 overexpression in invasive squamous cell carcinoma Oncogene 2011 30 42 4339 4349 10.1038/onc.2011.141 21532618 PMC3154567

9. Bao X Anastasov N Wang Y Rosemann M A novel epigenetic signature for overall survival prediction in patients with breast cancer J Transl Med 2019 17 1 380 10.1186/s12967-019-2126-6 31747912 PMC6889649

10. Barabási A Menichetti G Loscalzo J The unmapped chemical complexity of our diet Nat Food 2020 1 33 37 10.1038/s43016-019-0005-1

11. Bartlett TE Olhede SC Zaikin A A DNA methylation network interaction measure, and detection of network oncomarkers PLoS ONE 2014 9 1 e84573 10.1371/journal.pone.0084573 24400102 PMC3882261

12. Bates SE Epigenetic therapies for cancer N Engl J Med 2020 383 7 650 663 10.1056/NEJMra1805035 32786190

13. Baumgarten N Hecker D Karunanithi S Schmidt F List M Schulz MH EpiRegio: analysis and retrieval of regulatory elements linked to genes Nucleic Acids Res 2020 10.1093/nar/gkaa382 32459338 PMC7319550

14. Benincasa G, Costa D, Infante T, Lucchese R, Donatelli F, Napoli C. Interplay between genetics and epigenetics in modulating the risk of venous thromboembolism: a new challenge for personalized therapy. Thromb Res. 2019;177:145–53. 10.1016/j.thromres.2019.03.008. 10.1016/j.thromres.2019.03.008 30903874

15. Benincasa G Marfella R Della Mura N Schiano C Napoli C Strengths and opportunities of network medicine in cardiovascular diseases Circ J 2020 84 144 152 10.1253/circj.CJ-19-0879 31866635

16. Benjamin EJ Blaha MJ Chiuve SE Cushman M Das SR Deo R de Ferranti SD Floyd J Fornage M Gillespie C Isasi CR Jiménez MC Jordan LC Judd SE Lackland D Lichtman JH Lisabeth L Liu S Longenecker CT Mackey RH Matsushita K Mozaffarian D Mussolino ME Nasir K Neumar RW Palaniappan L Pandey DK Thiagarajan RR Reeves MJ Ritchey M Rodriguez CJ Roth GA Rosamond WD Sasson C Towfighi A Tsao CW Turner MB Virani SS Voeks JH Willey JZ Wilkins JT Wu JH Alger HM Wong SS Muntner P American Heart Association Statistics Committee and Stroke Statistics Subcommittee Heart disease and stroke statistics–2017 update: a report from the American Heart Association Circulation 2017 135 e146 e603 10.1161/CIR.0000000000000485 28122885 PMC5408160

17. Berdasco M Esteller M Clinical epigenetics: seizing opportunities for translation Nat Rev Genet 2019 20 2 109 127 10.1038/s41576-018-0074-2 30479381

18. Biswas S Rao CM Epigenetic tools (The Writers, The Readers and The Erasers) and their implications in cancer therapy Eur J Pharmacol 2018 837 8 24 10.1016/j.ejphar.2018.08.021 30125562

19. Blagitko-Dorfs N Schlosser P Greve G Pfeifer D Meier R Baude A Brocks D Plass C Lübbert M Combination treatment of acute myeloid leukemia cells with DNMT and HDAC inhibitors: predominant synergistic gene downregulation associated with gene body demethylation Leukemia 2019 33 4 945 956 10.1038/s41375-018-0293-8 30470836

20. Boer K Caliskan K Peeters AM Thymus-derived regulatory T cells infiltrate the cardiac allograft before rejection Transplantation 2015 99 9 1839 1846 10.1097/TP.0000000000000730 25886917

21. Buenrostro JD Corces MR Lareau CA Wu B Schep AN Aryee MJ Majeti R Chang HY Greenleaf WJ Integrated single-cell analysis maps the continuous regulatory landscape of human hematopoietic differentiation Cell 2018 173 6 1535 1548.e16 10.1016/j.cell.2018.03.074 29706549 PMC5989727

22. Cancer Genome Atlas Research Network Ley TJ Miller C Ding L Raphael BJ Mungall AJ Robertson A Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia N Engl J Med 2013 368 22 2059 2074 10.1056/NEJMoa1301689 23634996 PMC3767041

23. Capper D Jones DTW Sill M Hovestadt V Schrimpf D Sturm D Koelsche C Sahm F Chavez L Reuss DE Kratz A Wefers AK Huang K Pajtler KW Schweizer L Stichel D Olar A Engel NW Lindenberg K Pfister SM DNA methylation-based classification of central nervous system tumours Nature 2018 555 7697 469 474 10.1038/nature26000 29539639 PMC6093218

24. Castellano-Castillo D Moreno-Indias I Sanchez-Alcoholado L Ramos-Molina B Alcaide-Torres J Morcillo S Ocaña-Wilhelmi L Tinahones F Queipo-Ortuño MI Cardona F Altered adipose tissue DNA methylation status in metabolic syndrome: relationships between global DNA methylation and specific methylation at adipogenic, lipid metabolism and inflammatory candidate genes and metabolic variables J Clin Med. 2019 8 1 87 10.3390/jcm8010087 PMC6352101 30642114

25. Chelladurai P Dabral S Basineni SR Isoform-specific characterization of class I histone deacetylases and their therapeutic modulation in pulmonary hypertension Sci Rep 2020 10 1 12864 10.1038/s41598-020-69737-x 32733053 PMC7393135

26. Chen CC Lee KD Pai MY Chu PY Hsu CC Chiu CC Chen LT Chang JY Hsiao SH Leu YW Changes in DNA methylation are associated with the development of drug resistance in cervical cancer cells Cancer Cell Int 2015 15 98 10.1186/s12935-015-0248-3 26464562 PMC4604021

27. Chen D Wu H He B Five hub genes can be the potential DNA methylation biomarkers for cholangiocarcinoma using bioinformatics analysis Onco Targets Ther 2019 12 8355 8365 10.2147/OTT.S203342 31632083 PMC6793468

28. Chen X Jiang D Xu L Han L Hu H Huang Y Lu D Ji H Li B Yang Y Zhou C Xu X Wu N Xu X Xu Y Shen Y Li J Duan S Elevated methylation of cyclin dependent kinase inhibitor 2B contributes to the risk of coronary heart disease in women Exp Ther Med 2019 17 1 205 213 10.3892/etm.2018.6920 30651784 PMC6307461

29. Cheng F Desai RJ Handy DE Wang R Schneeweiss S Barabási AL Loscalzo J Network-based approach to prediction and population-based validation of in silico drug repurposing Nat Commun 2018 9 2691 10.1038/s41467-018-05116-5 30002366 PMC6043492

30. Cheng F Kovács IA Barabási AL Publisher Correction: Network-based prediction of drug combinations Nat Commun 2019 10 1 1806 10.1038/s41467-019-09692-y 30988295 PMC6465348

31. Cheng Y He C Wang M Ma X Mo F Yang S Han J Wei X Targeting epigenetic regulators for cancer therapy: mechanisms and advances in clinical trials Signal Transduct Target Ther 2019 4 62 10.1038/s41392-019-0095-0 31871779 PMC6915746

32. Chiu YC Chen HH Zhang T Zhang S Gorthi A Wang LJ Huang Y Chen Y Predicting drug response of tumors from integrated genomic profiles by deep neural networks BMC Med Genom 2019 12 Suppl 1 18 10.1186/s12920-018-0460-9 PMC6357352 30704458

33. Choobdar S Ahsen ME Crawford J Tomasoni M Fang T Lamparter D Lin J Hescott B Hu X Mercer J Natoli T Narayan R The DREAM Module Identification Challenge Consortium Subramanian A Zhang JD Stolovitzky G Kutalik Z Lage K Slonim DK Marbach D Assessment of network module identification across complex diseases 2019 Nat Methods 10.1038/s41592-019-0509-5 PMC6719725 31471613

34. Corces MR Granja JM Shams S Louie BH Seoane JA Zhou W Silva TC The chromatin accessibility landscape of primary human cancers Science 2018 10.1126/science.aav1898 30361341 PMC6408149

35. Cui ZJ Zhou XH Zhang HY DNA methylation module network-based prognosis and molecular typing of cancer Genes (Basel) 2019 10.3390/genes10080571 PMC6722866 31357729

36. de Nigris F Cacciatore F Mancini FP Vitale DF Mansueto G D’Armiento FP Epigenetic hallmarks of fetal early atherosclerotic lesions in humans JAMA Cardiol 2018 3 1184 1191 10.1001/jamacardio.2018.3546 30419118 PMC6583096

37. de Nigris F Ruosi C Napoli C Clinical efficiency of epigenetic drugs therapy in bone malignancies Bone 2020 20 115605 10.1016/j.bone.2020.115605 32829036

38. de Nigris F Zanella L Cacciatore F De Chiara A Fazioli F Chiappetta G Apice G Infante T Monaco M Rossiello R De Rosa G Alberghini M Napoli C YY1 overexpression is associated with poor prognosis and metastasis-free survival in patients suffering osteosarcoma BMC Cancer 2011 11 472 10.1186/1471-2407-11-472 22047406 PMC3240836

39. Dravis C Chung CY Lytle NK Herrera-Valdez J Luna G Trejo CL Reya T Wahl GM Epigenetic and transcriptomic profiling of mammary gland development and tumor models disclose regulators of cell state plasticity Cancer Cell 2018 34 3 466 482 10.1016/j.ccell.2018.08.001 30174241 PMC6152943

40. El-Awady RA Hersi F Al-Tunaiji H Saleh EM Abdel-Wahab AH Al Homssi A Suhail M El-Serafi A Al-Tel T Epigenetics and miRNA as predictive markers and targets for lung cancer chemotherapy Cancer Biol Ther 2015 16 7 1056 1070 10.1080/15384047.2015.1046023 25962089 PMC4622527

41. Endo A A historical perspective on the discovery of statins Proc Jpn Acad Ser B Phys Biol Sci 2010 86 5 484 493 10.2183/pjab.86.484 20467214 PMC3108295

42. Ervin RB Prevalence of metabolic syndrome among adults 20 years of age and over, by sex, age, race and ethnicity, and body mass index: United States, 2003–2006 Natl Health Stat Rep 2009 13 1 7 19634296

43. Esteller M Epigenetics in cancer 2008 N Engl J Med 2008 358 11 1148 1159 10.1056/NEJMra072067 18337604

44. Falcone R Conte F Fiscon G Pecce V Sponziello M Durante C Farina L Filetti S Paci P Verrienti A BRAFV600E-mutant cancers display a variety of networks by SWIM analysis: prediction of vemurafenib clinical response Endocrine 2019 64 2 406 413 10.1007/s12020-019-01890-4 30850937

45. Fardi M Solali S Farshdousti Hagh M Epigenetic mechanisms as a new approach in cancer treatment: an updated review Genes Dis 2018 5 4 304 311 10.1016/j.gendis.2018.06.003 30591931 PMC6303480

46. Feinberg AP The key role of epigenetics in human disease prevention and mitigation N Engl J Med 2018 378 14 1323 1334 10.1056/NEJMra1402513 29617578 PMC11567374

47. Feng Q Hawes SE Stern JE Wiens L Lu H Dong ZM Jordan CD Kiviat NB Vesselle H DNA methylation in tumor and matched normal tissues from non-small cell lung cancer patients Cancer Epidemiol Biomark Prev 2008 17 3 645 654 10.1158/1055-9965.EPI-07-2518 PMC2798850 18349282

48. Feng T Li K Zheng P Wang Y Lv Y Shen L Chen Y Xue Z Li B Jin L Yao Y Weighted gene coexpression network analysis identified microrna coexpression modules and related pathways in type 2 diabetes mellitus Oxid Med Cell Longev 2019 2019 9567641 10.1155/2019/9567641 31915515 PMC6935443

49. Fiscon G Conte F Licursi V Nasi S Paci P Computational identification of specific genes for glioblastoma stem-like cells identity Sci Rep 2018 8 1 7769 10.1038/s41598-018-26081-5 29773872 PMC5958093

50. Fong CY Gilan O Lam EY Rubin AF Ftouni S Tyler D Stanley K BET inhibitor resistance emerges from leukaemia stem cells Nature 2015 525 7570 538 542 10.1038/nature14888 26367796 PMC6069604

51. Fournier M Orpinell M Grauffel C Scheer E Garnier JM Ye T Chavant V KAT2A/KAT2B-targeted acetylome reveals a role for PLK4 acetylation in preventing centrosome amplification Nat Commun 2016 7 13227 10.1038/ncomms13227 27796307 PMC5095585

52. Furumai R Matsuyama A Kobashi N Lee KH Nishiyama M Nakajima H Tanaka A FK228 (depsipeptide) as a natural prodrug that inhibits class I histone deacetylases Cancer Res 2002 62 17 4916 4921 12208741

53. Gravina S Dong X Yu B Vijg J Single-cell genome-wide bisulfite sequencing uncovers extensive heterogeneity in the mouse liver methylome Genome Biol 2016 17 1 150 10.1186/s13059-016-1011-3 27380908 PMC4934005

54. Greene JA Loscalzo J Putting the patient back together—social medicine, network medicine, and the limits of reductionism N Engl J Med 2017 377 25 2493 2499 10.1056/NEJMms1706744 29262277

55. Guney E Menche J Vidal M Barábasi AL Network-based in silico drug efficacy screening Nat Commun 2016 7 10331 10.1038/ncomms10331 26831545 PMC4740350

56. Guo H Zhu P Yan L Li R Hu B Lian Y Yan J The DNA methylation landscape of human early embryos Nature 2014 511 7511 606 610 10.1038/nature13544 25079557

57. Haigentz M Jr Kim M Sarta C Lin J Keresztes RS Culliney B Gaba AG Smith RV Shapiro GI Chirieac LR Mariadason JM Belbin TJ Greally JM Wright JJ Haddad RI Phase II trial of the histone deacetylase inhibitor romidepsin in patients with recurrent/metastatic head and neck cancer Oral Oncol 2012 48 12 1281 1288 10.1016/j.oraloncology.2012.05.024 22748449 PMC3465519

58. He LJ Cai MY Xu GL Li JJ Weng ZJ Xu DZ Luo GY Zhu SL Xie D Prognostic significance of overexpression of EZH2 and H3k27me3 proteins in gastric cancer Asian Pac J Cancer Prev 2012 13 7 3173 3178 10.7314/apjcp.2012.13.7.3173 22994729

59. Hernández-Sánchez J Harlow L Church C Gaine S Knightbridge E Bunclark K Gor D Bedding A Morrell N Corris P Toshner M Clinical trial protocol for TRANSFORM-UK: A therapeutic open-label study of tocilizumab in the treatment of pulmonary arterial hypertension Pulm Circ 2018 8 1 2045893217735820 10.1177/2045893217735820 28956500 PMC6852369

60. Hirukawa A Smith HW Zuo D Targeting EZH2 reactivates a breast cancer subtype-specific anti-metastatic transcriptional program Nat Commun 2018 9 2547 10.1038/s41467-018-04864-8 29959321 PMC6026192

61. Hou Y Guo H Cao C Li X Hu B Zhu P Wu X Single-cell triple omics sequencing reveals genetic, epigenetic, and transcriptomic heterogeneity in hepatocellular carcinomas Cell Res 2016 26 3 304 319 10.1038/cr.2016.23 26902283 PMC4783472

62. Hu W Ding Y Wang S Xu L Yu H The construction and analysis of the aberrant lncRNA-miRNA-mRNA network in adipose tissue from type 2 diabetes individuals with obesity J Diabetes Res 2020 2020 3980742 10.1155/2020/3980742 32337289 PMC7168724

63. Hu Y Huang K An Q Simultaneous profiling of transcriptome and DNA methylome from a single cell Genome Biol 2016 17 88 10.1186/s13059-016-0950-z 27150361 PMC4858893

64. Infante T Del Viscovo L De Rimini ML Padula S Caso P Napoli C Network medicine: a clinical approach for precision medicine and personalized therapy in coronary heart disease J AtherosclerThromb 2020 27 279 302 10.5551/jat.52407 PMC7192819 31723086

65. Infante T Forte E Schiano C Punzo B Cademartiri F Cavaliere C Salvatore M Napoli C Evidence of association of circulating epigenetic-sensitive biomarkers with suspected coronary heart disease evaluated by Cardiac Computed Tomography PLoS ONE 2019 14 1 e0210909 10.1371/journal.pone.0210909 30673762 PMC6343931

66. Jarcho JA More evidence for SGLT2 inhibitors in heart failure N Engl J Med 2020 10.1056/NEJMe2027915 32865378

67. Jiao Y, Widschwendter M, Teschendorff AE. A systems-level integrative framework for genome-wide DNA methylation and gene expression data identifies differential gene expression modules under epigenetic control. In: Bioinformatics, vol. 30, issue 16, pp. 2360–2366. 2014. 10.1093/bioinformatics/btu316. 10.1093/bioinformatics/btu316 24794928

68. Johnson KC Houseman EA King JE von Herrmann KM Fadul CE Christensen BC 5-Hydroxymethylcytosine localizes to enhancer elements and is associated with survival in glioblastoma patients Nat Commun 2016 7 13177 10.1038/ncomms13177 27886174 PMC5133638

69. Jones A Teschendorff AE Li Q Hayward JD Kannan A Mould T West J Role of DNA methylation and epigenetic silencing of HAND2 in endometrial cancer development PLoS Med 2013 10 11 e1001551 10.1371/journal.pmed.1001551 24265601 PMC3825654

70. Kim C Gao R Sei E Brandt R Hartman J Hatschek T Crosetto N Foukakis T Navin NE Chemoresistance evolution in triple-negative breast cancer delineated by single-cell sequencing Cell 2018 173 4 879 893 10.1016/j.cell.2018.03.041 29681456 PMC6132060

71. Kim K Sun H Incorporating genetic networks into case-control association studies with high-dimensional DNA methylation data BMC Bioinform 2019 20 1 510 10.1186/s12859-019-3040-x PMC6805595 31640538

72. Knoechel B Roderick JE Williamson KE Zhu J Lohr JG Cotton MJ Gillespie SM An epigenetic mechanism of resistance to targeted therapy in T cell acute lymphoblastic leukemia Nat Genet 2014 46 4 364 370 10.1038/ng.2913 24584072 PMC4086945

73. Knutson SK Kawano S Minoshima Y Warholic NM Huang KC Xiao Y Kadowaki T Selective inhibition of EZH2 by EPZ-6438 leads to potent antitumor activity in EZH2-mutant non-Hodgkin lymphoma Mol Cancer Ther 2014 13 4 842 854 10.1158/1535-7163.MCT-13-0773 24563539

74. Lee HZ Kwitkowski VE Del Valle PL Ricci MS Saber H Habtemariam BA Bullock J FDA approval: belinostat for the treatment of patients with relapsed or refractory peripheral T-cell lymphoma Clin Cancer Res 2015 21 12 2666 2670 10.1158/1078-0432.CCR-14-3119 25802282

75. Lee LY-H Loscalzo J Network medicine in pathobiology Am J Pathol 2019 189 7 1311 1326 10.1016/j.ajpath.2019.03.009 31014954 PMC6616999

76. Leopold JA Maron BA Loscalzo J The application of big data to cardiovascular disease: paths to precision medicine J Clin Investig 2020 130 29 38 10.1172/JCI129203 31895052 PMC6934200

77. Li J Zhang Q Chen Z Xu D Wang Y A network-based pathway-extending approach using DNA methylation and gene expression data to identify altered pathways Sci Rep 2019 9 1 11853 10.1038/s41598-019-48372-1 31413306 PMC6694157

78. List M Alcaraz N Batra R Canzar S Ringeling FR De Novo pathway-based classification of breast cancer subtypes Protein–protein interaction networks: methods and protocols 2020 New York Springer 201 213 10.1007/978-1-4939-9873-9_15 31583640

79. Litzenburger UM Buenrostro JD Wu B Shen Y Sheffield NC Kathiria A Greenleaf WJ Chang HY Single-cell epigenomic variability reveals functional cancer heterogeneity Genome Biol 2017 18 1 15 10.1186/s13059-016-1133-7 28118844 PMC5259890

80. Liu P Li H Li S Leung KS Improving prediction of phenotypic drug response on cancer cell lines using deep convolutional network BMC Bioinform 2019 20 1 408 10.1186/s12859-019-2910-6 PMC6664725 31357929

81. Lomberk G Blum Y Nicolle R Nair A Gaonkar KS Marisa L Mathison A Distinct epigenetic landscapes underlie the pathobiology of pancreatic cancer subtypes Nat Commun 2018 9 1 1978 10.1038/s41467-018-04383-6 29773832 PMC5958058

82. Lund RJ Huhtinen K Salmi J Rantala J Nguyen EV Moulder R Goodlett DR Lahesmaa R Carpén O DNA methylation and transcriptome changes associated with cisplatin resistance in ovarian cancer Sci Rep 2017 7 1 1469 10.1038/s41598-017-01624-4 28473707 PMC5431431

83. Luo C Keown CL Kurihara L Zhou J He Y Li J Castanon R Single-cell methylomes identify neuronal subtypes and regulatory elements in mammalian cortex Science 2017 357 6351 600 604 10.1126/science.aan3351 28798132 PMC5570439

84. Ma X Liu Z Zhang Z Huang X Tang W Multiple network algorithm for epigenetic modules via the integration of genome-wide DNA methylation and gene expression data BMC Bioinform 2017 18 1 72 10.1186/s12859-017-1490-6 PMC5282853 28137264

85. Manasanch EE Han G Mathur R Qing Y Zhang Z Lee H Weber DM A pilot study of pembrolizumab in smoldering myeloma: report of the clinical, immune, and genomic analysis Blood Adv 2019 3 15 2400 2408 10.1182/bloodadvances.2019000300 31405950 PMC6693011

86. Maron BJ Maron MS Maron BA Loscalzo J Moving beyond the sarcomere to explain heterogeneity in hypertrophic cardiomyopathy: JACC review topic of the week J Am Coll Cardiol 2019 73 1978 1986 10.1016/j.jacc.2019.01.061 31000001 PMC6550351

87. Matei D Fang F Shen C Schilder J Arnold A Zeng Y Berry WA Huang T Nephew KP Epigenetic resensitization to platinum in ovarian cancer Cancer Res 2012 72 9 2197 2205 10.1158/0008-5472.CAN-11-3909 22549947 PMC3700422

88. Matsuda Y Yamauchi T Hosono N Uzui K Negoro E Morinaga K Nishi R Yoshida A Kimura S Maekawa T Ueda T Combination of panobinostat with ponatinib synergistically overcomes imatinib-resistant CML cells Cancer Sci 2016 107 7 1029 1038 10.1111/cas.12965 27166836 PMC4946706

89. McLean CY Bristor D Hiller M Clarke SL Schaar BT Lowe CB Wenger AM Bejerano G GREAT improves functional interpretation of cis-regulatory regions Nat Biotechnol 2010 28 5 495 501 10.1038/nbt.1630 20436461 PMC4840234

90. McLendon R Cancer Genome Atlas Research Network. Comprehensive genomic characterization defines human glioblastoma genes and core pathways Nature 2008 455 7216 1061 1068 10.1038/nature07385 18772890 PMC2671642

91. Meder B Haas J Sedaghat-Hamedani F Kayvanpour E Frese K Lai A Nietsch R Scheiner C Mester S Bordalo DM Amr A Dietrich C Pils D Siede D Hund H Bauer A Holzer DB Ruhparwar A Mueller-Hennessen M Weichenhan D Plass C Weis T Backs J Wuerstle M Keller A Katus HA Posch AE Epigenome-wide association study identifies cardiac gene patterning and a novel class of biomarkers for heart failure Circulation 2017 136 16 1528 1544 10.1161/CIRCULATIONAHA.117.027355 28838933

92. Menche J Sharma A Kitsak M Disease networks. Uncovering disease-disease relationships through the incomplete interactome Science 2015 347 6224 1257601 10.1126/science.1257601 25700523 PMC4435741

93. Misselbeck K Parolo S Lorenzini F Savoca V Leonardelli L Bora P A network-based approach to identify deregulated pathways and drug effects in metabolic syndrome Nat Commun 2019 10 5215 10.1038/s41467-019-13208-z 31740673 PMC6861239

94. Molife LR Attard G Fong PC Karavasilis V Reid AH Patterson S Riggs CE Jr Higano C Stadler WM McCulloch W Dearnaley D Parker C de Bono JS Phase II, two-stage, single-arm trial of the histone deacetylase inhibitor (HDACi) romidepsin in metastatic castration-resistant prostate cancer (CRPC) Ann Oncol 2009 21 1 109 113 10.1093/annonc/mdp270 19608618

95. Montalban-Bravo G DiNardo CD The role of IDH mutations in acute myeloid leukemia Future Oncol 2018 14 10 979 993 10.2217/fon-2017-0523 29543066

96. Moufarrij S Srivastava A Gomez S Combining DNMT and HDAC6 inhibitors increases anti-tumor immune signaling and decreases tumor burden in ovarian cancer Sci Rep 2020 10 3470 10.1038/s41598-020-60409-4 32103105 PMC7044433

97. Nannini DR Joyce BT Zheng Y Gao T Liu L Yoon G Huan T Ma J Jacobs DR Jr Wilkins JT Ren J Zhang K Khan SS Allen NB Horvath S Lloyd-Jones DM Greenland P Hou L Epigenetic age acceleration and metabolic syndrome in the coronary artery risk development in young adults study Clin Epigenet 2019 11 1 160 10.1186/s13148-019-0767-1 PMC6858654 31730017

98. Napoli C Benincasa G Schiano C Salvatore M Differential epigenetic factors in the prediction of cardiovascular risk in diabetic patients Eur Heart J Cardiovasc Pharmacother 2020 6 239 247 10.1093/ehjcvp/pvz062 31665258 PMC7363021

99. Napoli C Benincasa G Donatelli F Ambrosio G Precision medicine in distinct heart failure phenotypes: focus on clinical epigenetics Am Heart J 2020 224 113 128 10.1016/j.ahj.2020.03.007 32361531

100. Napoli C Benincasa G Loscalzo J Epigenetic inheritance underlying pulmonary arterial hypertension Arterioscler Thromb Vasc Biol 2019 39 653 664 10.1161/ATVBAHA.118.312262 30727752 PMC6436974

101. Napoli C Crudele V Soricelli A Al-Omran M Vitale N Infante T Primary prevention of atherosclerosis: a clinical challenge for the reversal of epigenetic mechanisms? Circulation 2012 125 2363 2373 10.1161/CIRCULATIONAHA.111.085787 22586291

102. Nebbioso A Tambaro FP Dell’Aversana C Altucci L Cancer epigenetics: moving forward PLoS Genet 2018 14 6 e1007362 10.1371/journal.pgen.1007362 29879107 PMC5991666

103. Paci P Colombo T Fiscon G Gurtner A Pavesi G Farina L SWIM: a computational tool to unveiling crucial nodes in complex biological networks Sci Rep 2017 7 44797 10.1038/srep44797.Erratum.In:SciRep.2017Jun,16(7),pp.46843 28317894 PMC5357943

104. Paci P Fiscon G Conte F Licursi V Morrow J Hersh C Cho M Castaldi P Glass K Silverman EK Farina L Integrated transcriptomic correlation network analysis identifies COPD molecular determinants Sci Rep 2020 10 1 3361 10.1038/s41598-020-60228-7 32099002 PMC7042269

105. Parikh RB Schwartz JS Navathe AS Beyond genes and molecules—a precision delivery initiative for precision medicine N Engl J Med 2017 376 17 1609 1612 10.1056/NEJMp1613224 28445664

106. Pathania R Ramachandran S Mariappan G Thakur P Shi H Choi JH Combined inhibition of DNMT and HDAC blocks the tumorigenicity of cancer stem-like cells and attenuates mammary tumor growth Cancer Res 2016 76 3224 3235 10.1158/0008-5472.can-15-2249 27197203 PMC4891240

107. Piekarz RL Frye R Turner M Wright JJ Allen SL Kirschbaum MH Zain J Phase II multi-institutional trial of the histone deacetylase inhibitor romidepsin as monotherapy for patients with cutaneous T-cell lymphoma J Clin Oncol 2009 27 32 5410 5417 10.1200/JCO.2008.21.6150 19826128 PMC2773225

108. Rafehi H Kaspi A Ziemann M Okabe J Karagiannis TC El-Osta A Systems approach to the pharmacological actions of HDAC inhibitors reveals EP300 activities and convergent mechanisms of regulation in diabetes Epigenetics 2017 12 11 991 1003 10.1080/15592294.2017.1371892 28886276 PMC5788408

109. Ragon BK DiNardo CD Targeting IDH1 and IDH2 mutations in acute myeloid leukemia Curr Hematol Malig Rep 2017 12 6 537 546 10.1007/s11899-017-0418-6 29064021

110. Raj K Mufti GJ Azacytidine (Vidaza(R)) in the treatment of myelodysplastic syndromes Ther Clin Risk Manag 2006 2 4 377 388 10.2147/tcrm.2006.2.4.377 18360650 PMC1936359

111. Rauschert S Raubenheimer K Melton PE Huang RC Machine learning and clinical epigenetics: a review of challenges for diagnosis and classification Clin Epigenet 2020 12 1 51 10.1186/s13148-020-00842-4 PMC7118917 32245523

112. Reyes-Palomares A Gu M Grubert F Berest I Sa S Kasowski M Remodeling of active endothelial enhancers is associated with aberrant gene-regulatory networks in pulmonary arterial hypertension Nat Commun 2020 11 1673 10.1038/s41467-020-15463-x 32245974 PMC7125148

113. Samblas M Milagro FI Martínez A DNA methylation markers in obesity, metabolic syndrome, and weight loss Epigenetics 2019 14 5 421 444 10.1080/15592294.2019.1595297 30915894 PMC6557553

114. Sasaki M Ikeda H Itatsu K Yamaguchi J Sawada S Minato H Ohta T Nakanuma Y The overexpression of polycomb group proteins Bmi1 and EZH2 is associated with the progression and aggressive biological behavior of hepatocellular carcinoma Lab Invest 2008 88 8 873 882 10.1038/labinvest.2008.52 18591938

115. Schiano C Benincasa G Franzese M Epigenetic-sensitive pathways in personalized therapy of major cardiovascular diseases Pharmacol Ther 2020 210 107514 10.1016/j.pharmthera.2020.107514 32105674

116. Schiano C Benincasa G Infante T Integrated analysis of DNA methylation profile of HLA-G gene and imaging in coronary heart disease: pilot study PLoS ONE 2020 15 8 e0236951 10.1371/journal.pone.0236951 32790754 PMC7425923

117. Schneider G Schmidt-Supprian M Rad R Saur D Tissue-specific tumorigenesis: context matters Nat Rev Cancer 2017 17 4 239 253 10.1038/nrc.2017.5 28256574 PMC5823237

118. Scotlandi K Hattinger CM Pellegrini E Gambarotti M Serra M Genomics and therapeutic vulnerabilities of primary bone tumors Cells 2020 9 4 968 10.3390/cells9040968 PMC7227002 32295254

119. Sharma A Halu A Decano JL Controllability in an islet specific regulatory network identifies the transcriptional factor NFATC4, which regulates type 2 diabetes associated genes NPJ Syst Biol Appl 2018 4 25 10.1038/s41540-018-0057-0 29977601 PMC6028434

120. Silverman EK Schmidt HHHW Anastasiadou E Altucci L Angelini M Badimon L Balligand JL Molecular networks in network medicine: development and applications Wiley Interdiscip Rev Syst Biol Med 2020 10.1002/wsbm.1489 32307915 PMC7955589

121. Sinkala M Mulder N Martin D Machine learning and network analyses reveal disease subtypes of pancreatic cancer and their molecular characteristics Sci Rep 2020 10 1 1212 10.1038/s41598-020-58290-2 31988390 PMC6985164

122. Smallwood SA Lee HJ Angermueller C Krueger F Saadeh H Peat J Andrews SR Stegle O Reik W Kelsey G Single-cell genome-wide bisulfite sequencing for assessing epigenetic heterogeneity Nat Methods 2014 11 8 817 820 10.1038/nmeth.3035 25042786 PMC4117646

123. Sonawane AR Weiss ST Glass K Sharma A Network medicine in the age of biomedical big data Front Genet 2019 10 294 10.3389/fgene.2019.00294 31031797 PMC6470635

124. Song L Wang J Wang H Chen Y Jia J Guo S Liu H Peng X Xiao W Gong Y Yang B Lu Y Li Y The quantitative profiling of blood mSEPT9 determines the detection performance on colorectal tumors Epigenomics 2018 10 12 1569 1583 10.2217/epi-2017-0154 30426784

125. Suvà ML Rheinbay E Gillespie SM Patel AP Wakimoto H Rabkin SD Riggi N Chi AS Cahill DP Nahed BV Curry WT Martuza RL Rivera MN Rossetti N Kasif S Beik S Kadri S Tirosh I Wortman I Shalek AK Rozenblatt-Rosen O Regev A Louis DN Bernstein BE Reconstructing and reprogramming the tumor-propagating potential of glioblastoma stem-like cells Cell 2014 157 3 580 594 10.1016/j.cell.2014.02.030 24726434 PMC4004670

126. Tang J Gautam P Gupta A He L Timonen S Akimov Y Wang W Szwajda A Jaiswal A Turei D Yadav B Kankainen M Saarela J Saez-Rodriguez J Wennerberg K Aittokallio T Network pharmacology modeling identifies synergistic Aurora B and ZAK interaction in triple-negative breast cancer NPJ Syst Biol Appl 2019 5 20 10.1038/s41540-019-0098-z 31312514 PMC6614366

127. Tassy O Pourquié O Manteia, a predictive data mining system for vertebrate genes and its applications to human genetic diseases Nucleic Acids Res 2014 10.1093/nar/gkt807 24038354 PMC3964984

128. Tavazzi L Maggioni AP Marchioli R Barlera S Franzosi MG Latini R Lucci D Nicolosi GL Porcu M Tognoni G Gissi-HF Investigators Effect of n-3 polyunsaturated fatty acids in patients with chronic heart failure (the GISSI-HF trial): a randomised, double-blind, placebo-controlled trial Lancet 2008 372 9645 1223 1230 10.1016/S0140-6736(08)61239-8 18757090

129. Tremblay-LeMay R Rastgoo N Pourabdollah M EZH2 as a therapeutic target for multiple myeloma and other haematological malignancies Biomark Res 2018 6 34 10.1186/s40364-018-0148-5 30555699 PMC6286605

130. Tsukamoto S Errani C Angelini A Mavrogenis A Current treatment considerations for osteosarcoma metastatic at presentation Orthopedics 2020 43 e345 e358 10.3928/01477447-20200721-05 32745218

131. van de Bovenkamp AA Bakermans AJ Allaart CP TrimetaziDine as a Performance-enhancING drug in heart failure with preserved ejection fraction (DoPING-HFpEF): rationale and design of a placebo-controlled cross-over intervention study Neth Heart J 2020 28 6 312 319 10.1007/s12471-020-01407-z 32162204 PMC7270414

132. Velazquez EJ Morrow DA DeVore AD Duffy CI Ambrosy AP McCague K Rocha R Braunwald E PIONEER-HF Investigators Angiotensin-neprilysin inhibition in acute decompensated heart failure N Engl J Med. 2019 380 539 548 10.1056/NEJMoa1812851 30415601

133. Venet D Dumont JE Detours V Most random gene expression signatures are significantly associated with breast cancer outcome PLoS Comput Biol 2011 7 10 e1002240 10.1371/journal.pcbi.1002240 22028643 PMC3197658

134. Wang RS Loscalzo J Network-based disease module discovery by a novel seed connector algorithm with pathobiological implications J Mol Biol 2018 430 2939 2950 10.1016/j.jmb.2018.05.016 29791871 PMC6097931

135. West J Beck S Wang X Teschendorff AE An integrative network algorithm identifies age-associated differential methylation interactome hotspots targeting stem-cell differentiation pathways Sci Rep 2013 10.1038/srep01630 23568264 PMC3620664

136. Westerman K Fernández-Sanlés A Patil P Sebastiani P Jacques P Starr JM Epigenomic assessment of cardiovascular disease risk and interactions with traditional risk metrics J Am Heart Assoc 2020 9 e015299 10.1161/JAHA.119.015299 32308120 PMC7428544

137. Westerman K Sebastiani P Jacques P Liu S DeMeo D Ordovás JM DNA methylation modules associate with incident cardiovascular disease and cumulative risk factor exposure Clin Epigenet 2019 11 142 10.1186/s13148-019-0705-2 PMC6792327 31615550

138. Yildirim MA Goh KI Cusick ME Barabási AL Vidal M Drug-target network Nat Biotechnol 2007 25 10 1119 1126 10.1038/nbt1338 17921997

139. Yomtoubian S Lee SB Verma A Izzo F Markowitz G Choi H Cerchietti L Vahdat L Brown KA Andreopoulou E Elemento O Chang J Inghirami G Gao D Ryu S Mittal V Inhibition of EZH2 catalytic activity selectively targets a metastatic subpopulation in triple-negative breast cancer Cell Rep 2020 30 3 755 770 10.1016/j.celrep.2019.12.056 31968251

140. Yusuf S Joseph P Rangarajan S Islam S Mente A Hystad P Brauer M Kutty VR Gupta R Wielgosz A AlHabib KF Dans A Lopez-Jaramillo P Avezum A Lanas F Oguz A Kruger IM Diaz R Yusoff K Mony P Chifamba J Yeates K Kelishadi R Yusufali A Khatib R Rahman O Zatonska K Iqbal R Wei L Bo H Rosengren A Kaur M Mohan V Lear SA Teo KK Leong D O’Donnell M McKee M Dagenais G Modifiable risk factors, cardiovascular disease, and mortality in 155 722 individuals from 21 high-income, middle-income, and low-income countries (PURE): a prospective cohort study Lancet 2020 395 10226 795 808 10.1016/S0140-6736(19)32008-2 31492503 PMC8006904

141. Zhang E Hou X Hou B Zhang M Song Y A risk prediction model of DNA methylation improves prognosis evaluation and indicates gene targets in prostate cancer Epigenomics 2020 12 4 333 352 10.2217/epi-2019-0349 32027524

142. Zhao SG Chen WS Li H Foye A Zhang M Sjostrom M Aggarwal R The DNA methylation landscape of advanced prostate cancer Nat Genet 2020 10.1038/s41588-020-0648-8 32661416 PMC7454228

143. Zheng C Xu R Predicting cancer origins with a DNA methylation-based deep neural network model PLoS ONE 2020 15 5 e0226461 10.1371/journal.pone.0226461 32384093 PMC7209244

144. Zhu P Guo H Ren Y Hou Y Dong J Li R Lian Y Single-cell DNA methylome sequencing of human preimplantation embryos Nat Genet 2018 50 1 12 19 10.1038/s41588-017-0007-6 29255258

145. Zhu WG Otterson GA The interaction of histone deacetylase inhibitors and DNA methyltransferase inhibitors in the treatment of human cancer cells Curr Med Chem Anticancer Agents 2003 3 3 187 199 10.2174/1568011033482440 12769777

146. Zierer J Pallister T Tsai PC Krumsiek J Bell JT Lauc G Spector TD Menni C Kastenmüller G Exploring the molecular basis of age-related disease comorbidities using a multi-omics graphical model Sci Rep. 2016 6 37646 10.1038/srep37646 27886242 PMC5122881