Drugging the epigenome in the age of precision medicine

In the 70 years since the term “epigenetics” was first coined, the field has yet to fulfill its true therapeutic potential, but has nonetheless proven a boon to basic researchers, to understand how cells process genetic information, differentiate, and respond to external stimuli [1]. At its core, epigenetics is focused on how cells control gene activity without changing the DNA sequence. This involves the modification of chemical signatures on DNA and its structures to alter the means by which transcription factors and other machinery interpret genetic information to control gene expression. Epigenetic modifications can induce changes in the accessibility of DNA as it is wound around histones, cause regulatory sequences to become refractory or amenable to transcription factor binding, or drive compartmentalization to activate or inactivate whole genomic loci. This complex system has been referred to as the “epigenetic code,” [2, 3] and describes the fundamental information layer that cells rely on to integrate and process the impact of external stimuli, in the context of past stimuli and cell-type determination despite a fixed genetic sequence (Fig. 1). With the advent of high-throughput sequencing and methodologies to interrogate chromatin state and DNA/RNA/protein interactions, an integrated understanding of epigenetics, functional genomics, and chromatin biology has blossomed into the current field of epigenomics.

Fig. 1figure 1

Precision epigenomic therapies have the potential to improve efficacy and tolerability. Early epigenomic therapies are limited by a lack of specificity, leading to off-target effects and more narrow therapeutic utility, as well as more limited tunability and durability. Precision therapies that act at discrete, specific loci should mitigate these challenges while delivering on the therapeutic promise of epigenomic modulation

Epigenetic effectors, enzymes capable of inducing changes in chromatin state, are varied and are often divided into three major categories—writers, erasers, and readers [1]. At the most basic level, writers create new epigenetic marks while erasers eliminate epigenetic marks. Finally, readers interpret the marks to change the conformation of DNA or histones and recruit additional machinery when needed.

The promise of epigenetic therapies

The principal promise of epigenetic-based therapies is the ability to control gene expression directly at the pre-transcriptional level and thus correct gene dysregulation at its source. Perhaps the greatest benefit of this approach lies in being able to turn gene expression up or down in a durable but (typically) not permanent manner, without making any changes to the underlying genomic sequence. This capability aids in the study of cellular differentiation, lineage specification, and programming as well as enabling this understanding to be harnessed to treat disease.

The ability to leverage the endogenous mechanisms by which cells control gene expression seemed like a new key to unlock therapeutic avenues for a variety of diseases. Cancer has been, and remains, an exemplar for the utility of epigenetic modulation as a therapeutic approach [4]. In many cancers, critical tumor suppressor genes are deactivated by hypermethylation or oncogenes are activated by demethylation, leading to dysregulated gene expression and unchecked growth. Oncology is not the only therapeutic area that could benefit from such intervention, however, as a number of inflammatory and neurological disorders, as well as rare monogenic conditions, degenerative diseases and diseases of aging, have also been shown to be linked to epigenetic dysregulation [5,6,7]. In fact, most diseases, irrespective of etiology, occur due to gene dysregulation and should be amenable to corrections. Being able to correct these defects represents a vast opportunity to improve patient outcomes in a variety of indications.

Unfortunately, translating these advances in our understanding of epigenetics into medicines has proven more challenging than initially anticipated. While there are eight FDA-approved and marketed epigenetic therapies with six to treat hematologic malignancies and two approved for use in solid tumors (Table 1), trials of current epigenetic therapies have shown greater toxicity than expected, likely due to low specificity. Even in cases where there is activity, toxicity driven by the broad impact of global inhibition of these effectors, due to lack of cell-type and genomic specificity, can drastically limit the utility of these compounds; global changes in methylation and acetylation patterns and/or interference in large, macromolecular complexes can have unintended consequences. The relatively greater success in hematologic cancers may also be related to inherently higher sensitivity of hematopoietic lineages to epigenetic modulations relative to other cell types due to greater plasticity of cellular programs, allowing for efficacy with a narrower therapeutic window or at lower and less toxic doses. Increasing the specificity of epigenetic approaches, at both the cellular and molecular levels, as well as their durability could help bridge the gap between the promise of these therapies and the current realities of bench-to-bedside translation.

Table 1 Summary of epigenetic approaches and molecules

There has also been limited success in applications of epigenetics outside of oncology. Although there is strong evidence that epigenetic dysregulation plays a role in other areas like autoimmunity and hemoglobinopathies, there has been minimal efficacy leveraging available compounds. Well-known HDAC inhibitors like butyrate have shown some proof-of-concept efficacy in certain settings, like sickle cell disease and beta-thalassemia [8], but not enough to outweigh the challenges of tolerability and dosing. Advances in the field that can improve specificity and therapeutic index would ideally help expand the application of epigenetic therapies to a broader range of indications.

Historical overview of epigenetic drugs and science

Following the elucidation of the DNA double helix structure, epigenetic markers, DNA methylation, and histone modifications were soon identified [1]. One key advance came in 1974, with the observation that DNA was packaged into nucleosomes, the fundamental subunits of chromatin containing DNA wound around histones [9]. Other discoveries including modification of histone amino-terminal tails and histone acetylation in the 1990s expanded our understanding of how chromatin and other associated proteins ultimately alter gene expression [1, 8,9,10,11,12,13]. In the past 2 decades, there has been a surge of research activity in the area of histone modifications and the enzymes that make or remove epigenetic marks on DNA and histones (Fig. 2) [1]. Concomitantly, an explosion of research in non-epigenetic modalities for controlling gene expression has occurred. Table 2 summarizes these approaches with their associated strengths and shortcomings.

Fig. 2figure 2

The evolution of epigenetic and epigenomic therapies. The first epigenetic therapeutics were first discovered in the 1960s but the first approval did not come until the early 2000s. Development in the field has accelerated markedly over the past 20 years, with epigenomic programming being the most recent advance. Green = DNMT inhibitor milestones; orange = HDAC inhibitor milestones; purple = 3rd generation epigenetic therapeutic milestones; blue = epigenomic programming milestones

Table 2 Non-epigenetic approaches to modulating gene expression1st generation epigenetic drugsDNA methyltransferase (DNMT) inhibitors

DNA methyltransferases are a class of cytosine methylases that play a key role in epigenetic regulation by depositing marks on the DNA itself. DNA methylation is important in the etiology of cancer as it epigenetically regulates the expression (or lack thereof) of cancer-related genes [14]. In 1980 [15], it was found that structural analogues of nucleo(s)tides could inhibit DNA methylation. Modifications to cytidine led to 5-azacitidine [14, 16] and decitabine [14, 17, 18]. Early work on these compounds yielded promising results in acute myeloid leukemia (AML), but the US applications for marketing authorization were not approved due to toxicity concerns [19, 20]. Subsequent studies were conducted in myelodysplastic syndrome (MDS) using lower doses, leading to FDA approvals for Bristol Myers Squibb [21, 22]. These compounds are also approved in the US to treat chronic myelomonocytic leukemia (CMML) and AML (despite the original FDA rejection), with additional label expansions occurring as recently as 2022 for juvenile myelomonocytic leukemia [21, 23]. Despite current use, the safety profile of these treatments can be difficult to manage and limits their clinical utility. In a recent Phase 3 trial of azacitidine in AML, > 20% of patients experienced Grade 3/4 thrombocytopenia and > 40% experienced Grade 3/4 neutropenia.

Various other nucleoside analogs have also demonstrated DNA hypomethylation activity, but have stalled in development due to low biological activity and/or high levels of toxicity, impacting organs like the liver and heart [16, 24,25,26]. Studies of derivatives of azacitosine and others remain in early development in various cancers, but are unlikely to represent a significant advance [27]. These early DNMTs provided insights into epigenetic mechanisms and applications in clinical practice while setting the stage for the development of more refined and effective molecules in this class.

Histone deacetylase (HDAC) inhibitors

Histone deacetylases are enzymes that remove acetyl marks from lysine residues on histones, allowing chromatin to be wound more tightly, reducing accessibility for transcription. The first epigenetic drugs approved in this class were vorinostat and romidepsin [28]. These agents were discovered through phenotypic observations without an a priori understanding of their mechanism of action as HDACs. Compared with DNMTs, HDACs currently occupy a narrower therapeutic niche [28, 29]. Vorinostat (suberoylanilide hydroxamic acid, SAHA), a pan-HDAC inhibitor developed by Merck [28, 30, 31], proved effective in early studies of several types of cancer. By following the strongest positive data, vorinostat ultimately received FDA approval for cutaneous T cell lymphoma (CTCL) in 2006, supporting the idea that cells of the hematopoietic lineage are most amenable to these broad small molecule epigenetic inhibitors [32, 33]. Potential vorinostat clinical applications extend to treatment of both neurological disorders and reactivation of latent viral infections to increase the efficacy of other antivirals, although additional studies are ongoing [5]. Romidepsin was identified using high-throughput screening studies [28]. Derived from a bacterium, it possessed a unique structure relative to HDAC inhibitors known at the time [34]. It was approved by the FDA in 2009 for the treatment of CTCL [35]. Unlike vorinostat, romidepsin exhibits selectivity between Class I HDACs and other isoforms [28].

Carboxylic acid is another zinc-binding motif that has been studied for its HDAC inhibiting properties [28]. The sodium salt of butyric acid was the first compound shown to inhibit histone deacetylation [36]. Due to rapid excretion, however, and modest clinical activity to date across rare diseases, epilepsy, and cancer, carboxylic acid HDAC inhibitors continue to serve predominantly as research tools. [28, 37]

2nd generation epigenetic drugsSecond-generation DNMT inhibitors

Assays for DNMT and HDAC activity were available by the early 1990s [28]. Given the limitations of azacitidine and decitabine, several new drugs were developed leveraging these new experimental capabilities. Second-generation DNMTs employ the bi-substrate strategy where a methyl group donor and cytosine are linked together resulting in some of the most potent DNMTi compounds available [28]. These agents can reactivate genes through promoter demethylation in cancerous cells [38]. To date, several of these compounds like guadecitabine (SGI-110), a degradation-resistant hypomethylating CpG dinucleotide mimic, and fluorocyclopentenylcytosine (RX-3117), an oral cytidine analog, have been tested in a range of cancers but none have been approved by the FDA for clinical use due to limited efficacy [4, 39,40,41,42]. Other non-nucleoside small molecule DNMTi have been used as preclinical tools and are being evaluated for clinical utility in neoplastic disease [43, 44].

Second-generation HDAC inhibitors

With the second-generation HDACs, applications have broadened to include non-hematological cancers [28]. These molecules tend to have limited efficacy as single agents but have demonstrated clinical utility in combination therapy. Given the efficacy seen with vorinostat, numerous synthetic analogues were developed, leading to the identification of belinostat, which was approved by the FDA in 2014 for the treatment of peripheral T cell lymphoma (PTCL) [28, 32, 45]. Panobinostat gained accelerated approval in combination with dexamethasone and bortezomib from the FDA in 2015 for relapsed or refractory multiple myeloma [32, 46, 47]; however, the approval was withdrawn in 2019. As with the first-generation HDACis, the pharmacokinetic profile of these drugs is not ideal and they can cause off-target effects due to non-selective metal binding [28].

Another successful structural class of compounds are the benzamides, which demonstrate selectivity toward Class I HDACs [28]. One example, entinostat, has been evaluated in clinical trials for multiple solid tumors in combination with hormone therapy and immune checkpoint therapy; however, a lack of robust efficacy data has stalled development [48]. Tucidinostat, a benzamide containing an alkenyl linker, inhibits Class 1 HDACs 1, 2, 3, and class II HDACs and is the first HDACi developed wholly in China, receiving approval from the Chinese FDA in 2015 [32, 49].

3rd generation epigenetic drugs

With multiple DNMT and HDAC inhibitors approved for clinical use, the fundamental hypothesis that epigenetics can be harnessed for therapeutic use has been borne out [28]. With improvements in the understanding of epigenetics, though, and the desire to improve the therapeutic window and safety profile of these therapies, efforts have expanded to identify new drugs that target other readers, writers, and erasers.

The third wave of epigenetic drug discovery has identified three new targets: lysine histone methyltransferases (KMTs), lysine demethylases (KDMs), and bromodomain inhibitors [28]. Agents targeting these epigenetic effectors have quickly advanced to clinical trials and regulatory approvals are anticipated in the near future. Unlike the earlier generations, where discovery was serendipitous and the epigenetic effect was unknown, many of these more recent compounds have been identified using prospective knowledge of their mechanisms of action.

Histone methyltransferase inhibitors

Histone methyltransferases, either KMTs or protein arginine methyltransferases (PRMTs), post-translationally add between one and three methyl groups to lysine or arginine residues on histone proteins, which can have a range of important effects [28]. Depending on the specific lysine residue being methylated, it can silence or activate gene transcription [50]. Pinometostat (EPZ-5676), developed by Epizyme, was the first KMT inhibitor studied for the treatment of leukemia [51]. Efficacy was, as with previous generations of epigenetic therapies, modest although tolerability was somewhat improved; however, there was a risk of increased infections observed with this agent. Subsequent targeting of the KMT enzyme EZH2 with tazemetostat yielded success for Epizyme in epithelioid sarcoma and follicular lymphoma with FDA approvals in 2020 [52]. Studies in other heme malignancies including diffuse large B-cell lymphoma (DLBCL) are ongoing [32, 53,54,55]. The first PRMT inhibitor to undergo evaluation in clinical trials was GSK3326595, targeting PRMT5 [56]. Other PRMT inhibitors for PRMT1 and 5 are in clinical trials for the treatment of solid tumors, non-Hodgkin lymphoma, MDS, and DLBCL [57, 58].

Lysine demethylase inhibitors

Lysine demethylases reverse lysine methylation on either DNA or histone proteins (among other molecules) which can alter either the transcription of genes at the promoter or via changes in chromatin state. One family of KDMs, lysine-specific demethylases (LSDs) are homologous to monoamine oxidases in their mechanism; thus, MAOIs have been repurposed as epigenetic therapies [59]. Tranylcypromine is an MAOI originally approved in 1961 as an antidepressant but is now in clinical trials as a potential therapy for AML and MDS [31, 60,61,62]. ORY-1001 and GSK2879552, LSD inhibitors created to improve tranylcypromine’s modest activity and reduce off-target effects, are being tested in Phase I/II trials [32,

留言 (0)

沒有登入
gif