Systematic Review of Epigenetic Therapies for Treatment of IDH-mutant Glioma

Abstract

BACKGROUND:

Isocitrate dehydrogenase (IDH) mutations are present in 70% of World Health Organization grade II and III gliomas. IDH mutation induces accumulation of the oncometabolite 2-hydroxyglutarate. Therefore, therapies targeting reversal of epigenetic dysregulation in gliomas have been suggested. However, the utility of epigenetic treatments in gliomas remains unclear. Here, we present the first clinical systematic review of epigenetic therapies in treatment of IDH-mutant gliomas and highlight their safety and efficacy.

METHODS:

We conducted a systematic search of electronic databases from 2000 to January 2021 following PRISMA guidelines. Articles were screened to include clinical usage of epigenetic therapies in case reports, prospective case series, or clinical trials. Primary and secondary outcomes included safety/tolerability of epigenetic therapies and progression-free survival/overall survival, respectively.

RESULTS:

A total of 133 patients across 8 clinical studies were included in our analysis. IDH inhibitors appear to have the best safety profile, with an overall grade 3/grade 4 adverse event rate of 9%. Response rates to IDH-mutant inhibitors were highest in nonenhancing gliomas (stable disease achieved in 55% of patients). In contrast, histone deacetylase inhibitors demonstrate a lower safety profile with single-study adverse events as high as 28%.

CONCLUSION:

IDH inhibitors appear promising given their benign toxicity profile and ease of monitoring. Histone deacetylase inhibitors appear to have a narrow therapeutic index, as lower concentrations do not appear effective, while increased doses can produce severe immunosuppressive effects. Preliminary data suggest that epigenetic therapies are generally well tolerated and may control disease in certain patient groups, such as those with nonenhancing lesions.

INTRODUCTION

Despite maximal safe resection and adjuvant chemoradiation, gliomas often harbor a poor prognosis due to tumor heterogeneity, treatment resistance, and infiltrating tumor cells on top of the limitations related to the blood-brain barrier.^1,2 However, recent advances in molecular phenotyping of gliomas have demonstrated that isocitrate dehydrogenase (IDH) mutation (IDHm) serves as a molecular prognosticator for certain gliomas, conferring improved outcomes compared to IDH wild-type gliomas.^1,2 Two forms of IDH are considered especially relevant to glioblastoma multiforme (GBM) classification: IDH1, which is primarily found in peroxisomes or in the cytoplasm, and IDH2, which is primarily mitochondrial. Within IDH-mutated gliomas, the IDH1 R132 mutation is also a more prevalent mutation than IDH2 mutations. IDHm is considered to be an early event in glioma pathophysiology that disrupts cellular epigenetic regulation by formation of 2-hydroxyglutarate (2-HG).³ The wild-type IDH enzyme catalyzes conversion of isocitrate to α-ketoglutarate (α-KG) during oxidative phosphorylation; however, in IDHm, α-KG is converted to an oncometabolite, 2-HG, which triggers a oncogenic cascade through inhibition of α-KG—dependent enzymes.^1,3 As a result, α-KG—dependent dioxygenases such as a histone/DNA demethylases and 5-methlycytosine hydroxylase do not properly function, thereby facilitating epigenetic dysregulation through histone/DNA hypermethylation.⁴ Thus, IDHm-mutant tumors are not synonymous with GBM but should rather be considered a unique subtype of gliomas with a distinct biochemical etiology. These characteristic epigenetic marks define IDHm gliomas and provide a potential avenue for novel treatment modalities.

Recently, several therapeutics have been developed that aim to disrupt the activity of epigenetic regulators in IDH-mutant gliomas. Currently, 5 classes of therapy have undergone or are currently undergoing clinical investigation for gliomas specifically: direct mutant IDH inhibitors, histone deacetylase inhibitors, bromodomain inhibitors, poly-ADP ribose polymerase inhibitors, and nicotinamide phosphoribosyl transferase inhibitors. However, despite these advances, there are no current systematic reviews of published clinical studies regarding the safety and efficacy of these treatment modalities for IDH-mutant gliomas. Here we present the first comprehensive systematic review of clinical studies investigating epigenetic therapies in management of IDH-mutant gliomas.

METHODS

Literature Search and Selection

A systematic literature search of the English literature from 1997 to 2020 in the PubMed and EMBASE databases was performed. The Medical Subject Headings system was queried for all articles containing the terms epigenetic therapy and glioma: (’histone deacetylase inhibitor’ OR ’bromodomain inhibitor’ OR ’DNA methyltransferase inhibitor’ OR ’IDH inhibitor’ OR ‘epigenetic therapy’) AND (’glioma’ OR ’low grade glioma’ OR ’glioblastoma’) AND ’IDH1 mutant’. The objective was to screen for articles containing series of patients with IDH-mutant gliomas who have undergone any form of epigenetic-modifying treatments.

Inclusion and Exclusion Criteria

Article search was limited to English with humans as the only study participants. All articles were specified as retrospective or prospective patient studies, clinical trials, randomized clinical trials, or post hoc analysis of clinical trials. Reviews, editorials, commentaries, and case reports were excluded. Studies evaluating chemotherapy clinical and radiological outcomes in human patients with IDH-mutant gliomas or glioblastomas were included. The exclusion criteria were as follows: 1) studies that did not involve human subjects, 2) cohorts that did not incorporate pharmacologic management, 3) studies that did not evaluate outcomes following administration of therapy, and 4) studies with data duplicated in a previous clinical study.

RESULTS

Study Selection and Characteristics

After applying the inclusion criteria, 23 studies were included for data analysis, where 2 articles were excluded because of inadequate data description, reaching the final number of 21 studies retrieved. A total of 157 published abstracts or manuscripts were identified. Through the PRISMA flow chart (Figure 1), and applying the inclusion and exclusion criteria, a total of 8 studies were included in the qualitative analysis. Selection and data extraction of the studies was performed by 3 authors (VG, AHS, and LD).

Figure 1.

PRISMA diagram summarizing systematic review process.

Eight articles met the criteria for inclusion from an initial pool of 157 articles. All articles were performed in humans and evaluated the impact of epigenetic therapies radiographically, biochemically, histologically, clinically, or a combination thereof. Given the limited data set and their heterogeneity, we did not perform any quantitative analyses with these articles. Key differences between articles are summarized in Table 1. Three studies evaluated IDH1 inhibitors, 3 studies evaluated histone deacetylase (HDAC) inhibitors, and one study investigated a bromodomain inhibitor. In our included studies, there were also several key divisions regarding tumor classification. First, several articles make distinctions between high-grade gliomas (World Health Organization [WHO] classification of grade III or IV) and low-grade gliomas (WHO classification of grade II or lower). Furthermore, several articles refer to differences in response between enhancing and nonenhancing lesions. Seven ongoing clinical trials were also included. However, these trials are not complete and have not provided any preliminary efficacy or safety data. All 7 of these trials investigate IDH1 inhibitors and are summarized in Table 2.

Table 1.

Summary of Published Clinical Studies Investigating Epigenetic Therapies

Author	Drug	Number of Patients	Dose	Treatment Duration	Follow-up	Primary Outcomes	Secondary Outcomes	Adverse Events
Andronesi et al., 20189	IDH305	8 (R132H mutation) WHO grade II (n = 2), Grade III (n = 5) Grade IV (n = 1)	550 mg IDH305	IDH305 twice daily for 7 days	1 week	Decreased 2-HG levels at 1 week (P < 0.05)	Significant inverse correlation between 2-HGglutamine + glutamate ratio and cell density (P = 0.03)
Mellinghoff et al., 202010	Ivosidenib	66 (86.4% R132H) 1p19q (33.3%)	Dose escalation: 100 mg BID; 1200 mg QD; 500 mg daily for expansion cohort.	18.4 months (nonenhancing lesions); 1.9 months (enhancing lesions)	Efficacy assessed every 2 cycles (56 ± 3 days)	Median PFS 13.9 months (nonenhancing) and 1.4 months (enhancing lesions)	Stable disease in 85.7% of nonenhancing lesions, 45.2% in enhancing lesions. Tumor growth rate of −14% after treatment with ivosidenib	3% grade ≥3 and treatment-related adverse effects
Tejera et al., 202011	Ivosidenib	1 (IDH1m)	500 mg daily in continuous 28-day cycles	4.08 years	5.16 years	Stable disease for duration of follow-up	Improved seizure control
Gurbani et al., 202020	Belinostat	1 IDH1 R132H MGMT-nonco-deleted	500 mg/m2/day	Five days prior to RT followed by 2 additional 5-day cycles at 3-week intervals	1.5 years	Reduction in tumor size and mass effect. Stable disease for 16 months	Improved neurocognition
Moreno et al., 202024	CC-90010	1 IDH1m MGMT+	30 mg/day	4 days followed by 24 days without drug (28-day cycles)	10 mo.	Complete response	Disappearance of enhancing and nonenhancing lesions within 6 cycles
Gurbani et al., 201918	Belinostat	2 (IDH1m)	Control arm: Daily TMZ with RT Treatment arm: 500 or 750 mg/m2/day	42 days of TMZ (control arm), 5 consecutive days across 3 cycles, every 3 weeks	1 year/until progression	No significant difference in PFS (P = 0.09)	Statistically significant improvement in depression among belinostat patients (P = 0.04)	Hematological toxicity at 700 mg/m2 dose
Galanis et al., 201819	Vorinostat	3 (IDH1m)	Initial cycle of 300 or 400 mg/day followed by 400 mg/day	Initial cycle on days 1—5, followed by 12 cycles of vorinostat on days 1—7 and days 15—21 of 28-day cycle	16 months	Median OS of 16.1 months 55.1% survival at 15 months	Statistically significant improvement in PFS among patients with vorinostat	Leukopenia, thrombocytopenia, or neutropenia seen in 9—28% of patients
Hanna et al., 202044	Olaparib	39 (several had unknown IDHm status, unknown how many)	200 mg or dose escalation protocol: 100 mg to 150 mg	Twice daily for 7 days or 42 days per cycle for 3 56-day cycles	6 months	Tumor penetration found in all 21 tumor core samples	39% of patients remained progression-free at 6-month follow-up	7 patients experienced severe anemia, lymphopenia, thrombocytopenia, or neutropenia

RT, radiation therapy; TMZ, temozolomide; 2-HG, 2-hydroxyglutarate; PFS, progression-free survival; WHO, World Health Organization.

Table 2.

Summary of Ongoing Clinical Trials Investigating Epigenetic Therapies

Author	Recruitment Status	Intervention	Control	Randomized?	Estimated Enrollment	Primary Outcomes	Secondary Outcomes
Luke et al.	Not yet recruiting	Ivosidenib	None	No	35 participants	Dose-limiting toxicity Best overall response PFS	Adverse events OS Tumor progression
Servier Pharmaceuticals	Active, not recruiting	AG-120, AG-881 prior to surgery	No treatment prior to surgery	Yes	49 participants	2-HG concentration in surgically resected tumors	Safety, pharmacodynamics, pharmacokinetics, clinical activity
	Active, not recruiting	AG-881 postoperatively	Placebo	Yes	340 participants	Progression-free survival	Adverse events, clinical response, overall survival
	Active, not recruiting	AG-881	None	No	93 participants (52 with glioma)	Safety/tolerability, Maximum tolerated dose	Pharmacokinetics and pharmacodynamics
Novartis Pharmaceuticals	Active, not recruiting	IDH305	None	No	166 patients	Dose-limiting toxicity Maximum tolerated dose	Adverse events, pharmacodynamics, pharmacokinetics, overall response rate
Loxo Oncology, Eli Lilly	Recruiting	LY3410738	None	No	180 patients	Identification of recommended phase 2 dose	Pharmacodynamics, pharmacokinetics, safety, objective response rate
Forma Pharmaceuticals	Active, not recruiting	FT-2102	None	No	200 patients	Dose confirmation	Preliminary efficacy data

OS, overall survival; 2-HG, 2-hydroxyglutarate; PFS, progression-free survival.

Mutant IDH Inhibitors

IDH converts isocitrate to α-KG; however, mutations in the IDH have been noted in several cancers including chondrosarcoma, cholangiocarcinoma, acute myeloid leukemia (AML), and gliomas.^4,5 Mutant IDH produces excess quantities of 2-HG, which inhibits several key demethylases, resulting in marked histone and DNA methylation.⁴ The resulting methylation pattern resembles that of progenitor cells, with inhibition of cellular differentiation and enrichment of stem cell markers.⁴ Furthermore, inhibition of mutant IDH can promote differentiation in tumorigenic cells, making IDHm a promising drug target.⁶

Currently, a variety of IDH inhibitors are under clinical investigation in gliomas. A recent study by Mellinghoff et al. using ivosidenib (an IDH1 inhibitor) demonstrated stable disease in 86% of all nonenhancing gliomas with an acceptable tolerability (3% grade 3—4 adverse event rate).⁷ Vorasidenib, a dual IDH1/2 inhibitor, possesses enhanced brain penetration relative to ivosidenib.⁷ In a phase 1 study of vorasidenib in recurrent IDH-mutant glioma, the drug was well tolerated with a median progression-free survival (PFS) of 31 months and an overall response rate of 18%, primarily in patients with nonenhancing lesions.⁷ Olutasidenib, a brain penetrant, IDH 1 inhibitor, was also well tolerated as a single agent in recurrent, mainly enhancing gliomas, with a disease control rate of 45% in an ongoing phase 1b/2 study.⁸

Overall, the median age across all studies evaluating IDH inhibitors was 43 years. The overall response rate to IDH inhibitors across all studies was 55% (primarily nonenhancing lesions), with an overall grade 3—4 adverse event rate of 9%.^9,10 The mechanism of IDHm and its inhibitors is illustrated in Figure 2A. Additionally, 2-HG has been demonstrated as a specific biomarker of IDH1 inhibitor efficacy and can be utilized to monitor treatment response.⁵

Figure 2.

(A) Mechanism of IDH1m and IDH1m inhibitors. IDH1m produces 2-hydroxyglutarate (2-HG), which in turn inhibits histone demethylases. The loss of histone demethylases reverts DNA methylation patterns to those of progenitor cells, promoting tumorigenesis. IDH1m inhibitors aim to disrupt this process by directly inhibiting IDH1m. (B) Mechanism of histone deacetylase (HDAC) and HDAC inhibitors. HDAC decondenses chromatin, thereby allowing for gene expression. Mutant HDACs are products of gain-of-function mutations that upregulate multiple genes associated with increased cell survival, growth, and division, contributing to tumor formation. (C) Mechanism of bromodomains and their inhibitors. Bromodomains bind acetylated lysine residues and subsequently act as scaffold for transcription machinery. Bromodomain inhibitors may not have direct cytotoxic effects but inhibit double-strand break (DSB) repair, predisposing tumor cells to the cytotoxic effects of other chemotherapeutics. (D) Description of 5-azacitidine (5-AZ) mechanism of action. 5-AZ inhibits DNA methyltransferase (DNMT), leading to upregulation of immunomodulatory genes, namely cytokines and interferon. These gene products may leverage the host immune response to target tumor cells. IDH, isocitrate dehydrogenase.

HDAC Inhibitors

Currently, 3 clinical studies have investigated the utility of HDAC inhibitors in the treatment of IDH-mutant brain tumors (median age: 50 years, n = 6 IDHm patients). In one study, Gurbani et al. were unable to demonstrate a difference in the PFS between patients who received the HDAC inhibitor belinostat (P = 0.09), though only 2 patients had IDHm gliomas.⁹ However, a similar study by Galanis et al. demonstrated a median overall survival (OS) of 16.1 months among patients with IDH1-mutant glioblastoma receiving vorinostat (n = 3).¹⁰ Furthermore, the authors demonstrated that patients with specific vorinostat sensitivity signatures (collections of putatively connected genes that mediate cellular response to chemotherapeutics) had significantly higher PFS (PFS, P < 0.0001).¹⁰ Lastly, belinostat, another HDAC inhibitor, had demonstrated some efficacy for IDH1-mutant glioblastoma with disease stabilization at the 16-month follow-up.^9,11 Several adverse events have been recorded across the available studies, mostly hematologic toxicities (leukopenia, thrombocytopenia, or neutropenia).^9,10 The mechanism of HDAC and its inhibitors is illustrated in Figure 2B.

Bromodomain Inhibitors

Bromodomains are highly conserved structures that are associated with histone acetylation. Primarily, bromodomains “read” and bind acetylated lysine residues on histones which then mark these regions for transcription.¹² Bromodomains then form a scaffold for transcription factors and requisite transcription machinery, thus serving as a critical link between epigenetic modifications and gene expression.^13,14 Only one clinical study investigating the effects of a bromodomain inhibitor has been conducted. Moreno et al. investigated the utility of the bromodomain inhibitor CC-90010 in a series of 69 patients, one of whom had an IDH1-mutant glioma.¹⁵ The patient demonstrated a complete clinical response to treatment, with loss of enhancing and nonenhancing areas of the lesion within 6 cycles of CC-90010.¹⁵ A description of bromodomains and their inhibitors can be found in Figure 2C.

5-Azacitidine

5-azacitidine (5-AZ) is a DNA methyltransferase (DNMT) inhibitor (DNMTi). Members of this chemotherapy class act by direct, irreversible inhibition of DNMT, which prevents further methylation of downstream DNA.¹⁶ One study by Federici et al. investigated the utility of 5-AZ in 12 patients with IDHm glioma. All patients received 6 cycles of 5-AZ, and at a median follow-up period of 20 months, all patients had demonstrated disease progression; however, 3 (25%) patients had prolonged disease stabilization on treatment.¹⁶ The median PFS for this cohort was 4.7 months, with a median OS of 25.2 months. Furthermore, 7 patients experienced grade 3 or grade 4 leukopenia and 9 patients experienced grade 3/4 neutropenia.¹⁶ The mechanism of 5-AZ is displayed in Figure 2D. Previously, it has been demonstrated that systemic 5-AZ can penetrate the blood-cerebrospinal fluid barrier and achieve adequate deposition in the cerebrospinal fluid¹⁶; however, systematic analyses of 5-AZ penetration into tumors have yet to be performed.

Poly-ADP Ribose Polymerase Inhibitors

Poly-ADP ribose polymerase (PARP) has been identified as a key regulator of several DNA repair processes, including both single-strand break and double-strand break (DSB) repair, homologous end-joining, and nonhomologous end-joining.¹⁷ PARP inhibitors (PARPIs) induce accumulation of DSBs, which have demonstrated cytotoxicity in the absence of complementary repair mechanisms.¹⁸ This feature of PARPIs is particularly useful in treating cancers with BRCA1/2 mutations or mutations in other homologous recombination (HR)—associated genes.¹⁸ However, PARPIs have also demonstrated efficacy in tumors with wild-type HR genes when combined with DNA-damaging agents.¹⁹ A summary of PARP and its inhibitor’s mechanisms of action is shown in Figure 3A. Currently, the PARPI olaparib is being investigated in a phase 2 study to determine its use in gliomas and other solid tumors (NCT03212274).

Figure 3.

(A) Mechanism of PARP and PARP inhibitors. PARP normally repairs several mechanisms of DNA damage, including double-strand breaks (DSBs). In tumor cells that lack other forms of DNA repair (nonhomologous end-joining, homologous recombination, etc.), PARP is critical for their continued survival and proliferation. PARP inhibitors (PARPIs) take advantage of this fact, leading to cytotoxic accumulation of DSBs, ultimately leading to tumor cell death. (B) Characterized by a high metabolic rate, cancer cells often rely on. PARP, poly-ADP ribose polymerase.

Nicotinamide Phosphoribosyl Transferase Inhibitors

Cancers are generally characterized by high rates of metabolic activity and therefore require large quantities of NAD to meet their metabolic demands (Figure 3B). Nicotinamide phosphoribosyl transferase (NAMPT) appears to be particularly relevant in brain tumorigenesis, as NAMPT overexpression has been previously associated with brain tumor formation and may potentially serve as a biomarker.^20,21 Furthermore, it has been previously established that NAMPT prevents cell death in neural lineage cells, which are also characterized by high metabolic activity.²² NAMPT facilitates this process by augmenting NAD regeneration, thus making NAMPT a prime target for novel chemotherapeutics. NAMPT inhibitors (NAMPTIs) prevent NAD regeneration, ultimately depleting ATP and inducing cell death.²⁰ Initially, 3 NAMPTIs underwent phase I clinical trials; however, all were discontinued due to drug toxicity concerns.²⁰

DISCUSSION

Considerations of the 2021 WHO Guidelines

While IDH mutation status has been a longstanding means of differentiating high- and low-grade gliomas, recent changes to the WHO Classification of Tumors of the Central Nervous System have further increased the nuance of GBM classification. For example, recent scholarship has suggested that there exists a group of low-grade IDHwt gliomas that lack conventional GBM signatures and are quite similar to neoplasms commonly found in the pediatric population.²³ These recent results and guideline changes may require a deeper analysis of low-grade versus high-grade gliomas, namely, in addition to IDH mutations, other significant molecular events must be analyzed, which may offer additional therapeutic targets. For example, Ahrendsen et al. recently demonstrated alterations in FGFR, BRAF, and NTRK in patients with IDHm liomas, suggesting that IDH assessment alone may be insufficient to properly characterize gliomas.^13,15,24

Assessment of Existing Epigenetic Therapies

With respect to safety, across all current studies, IDH inhibitors demonstrated an overall aceptable safety profile.^4,6,25 The best responders to IDH inhibitors seem to be patients with slowly growing or nonenhancing gliomas, suggesting that timing of therapy remains vital for clinical decision-making. IDH inhibitors also offer a means of noninvasively and objectively monitoring efficacy via measurement of 2-HG.²⁶ The benign safety profile coupled with convenient monitoring may make IDH inhibitors promising candidates for combination therapy with other chemotherapeutics, especially given their ease of monitoring via serum sampling.

HDAC Inhibitors

Histone deacetylation is a key process in cellular proliferation and growth. Namely, histone deacetylation decondenses chromatin, thereby allowing for the expression of genes responsible for cellular growth and development; therefore, HDACs may act as drivers of oncogenesis.^27-31 As such, HDAC inhibitors have been developed as a class of chemotherapeutics aimed at inducing cell cycle arrest and tumor cell apoptosis.³²

HDAC inhibitors are one of the few medications that have been investigated in both IDH wild-type and IDH-mutant glioma and glioblastoma, and its efficacy in IDH wild-type tumors may give insight into their efficacy in IDH-mutant tumors. We performed a brief meta-analysis that demonstrated that patients receiving HDAC inhibitors had lower PFS at 6 months, though this was not significantly significant (P = 0.07, OR = 0.53, 95% confidence interval [0.27—1.05]) Figure 4. Additionally, HDAC inhibitors must be titrated carefully due to their pronounced side-effect profile, with some studies quoting adverse event rates up to 28%.^10,11 This potentially poor survival benefit coupled with the relatively severe side-effect profile may limit the use of these medications in the future. Furthermore, these documented toxicities may severely limit their potential use in combination chemotherapy regimens.

Figure 4.

Comparison of percent progression-free patients at 6 months (PFS6) in patients receiving HDAC inhibitors relative to historical controls. Patients receiving HDAC inhibitors were less likely to be progression-free at 6 months post treatment though this was not significant (P = 0.07, OR = 0.53, 95% CI [0.27–1.05]). CI, confidence interval; HDAC, histone deacetylase; OR, odds ratio.

Bromodomain Inhibitors

Bromodomain inhibitors are currently the least well-studied epigenetic chemotherapy in management of IDH1-mutant gliomas. There have been several preliminary studies that have investigated its use in other solid tumors, with varying results.^33,34 Given their relatively new arrival to the pharmaceutical market, there is a general paucity of clinical data using bromodomain inhibitors in gliomas.^35,36 Therefore, more studies are required before any judgments regarding their utility can be made.

DNMT Inhibitors

5-AZ is a DNMTi that was formerly used as a salvage therapy for myelodysplastic syndromes and AML.³⁷ 5-AZ primarily exerts its effects by upregulating several immunomodulatory genes, namely cytokines, interferon, and proinflammatory genes, and has shown efficacy in both solid tumors and AML.³⁸ Given the relative lack of large systematic studies using 5-AZ for gliomas, its safety and effiacacy cannot be determined at this time, although there are some data to suggest that 5-AZ may improve outcomes in a subset of patients.^37,39 Systemic 5-AZ must be titrated carefully due to its risk of developing severe leukopenia. Additional investigations using intracavitary or intraventricular 5-AZ are also currently underway for select CNS tumors.

Limitations of Current Therapies

Despite the development of these novel therapies, several limitations exist that may explain their relatively modest efficacy. Primarily, epigenetic therapies must be designed to efficiently cross the blood-brain barrier with minimal neurotoxicity.⁴⁰ Some HDAC inhibitors may have poor BBB permeability due to hydroxyamic acid—, amino, and zinc-binding groups.⁴⁰ Therefore, utilizing native BBB transport proteins may be a promising strategy to augment BBB penetration of novel HDAC inhibitors.⁴¹

Another issue surrounding IDHm tumors is that the sequence of acquired mutations directly impacts the utility of any direct epigenetic therapies. It has been previously demonstrated that IDHms are drivers of further mutation, producing DNA damage and methylation by producing 2-HG.^4,5 Thus, early inhibition of IDHm is critical in preventing the acquisition of further mutations, thereby mitigating tumor invasion. Furthermore, it has been previously demonstrated that any epigenetic changes produced by IDHm are preserved and transmitted through successive divisions of tumor cell lines.⁴²

Overall Assessment and Future Directions

We have reviewed the most common epigenetic chemotherapeutics used in the treatment of IDH-mutant glioma, as summarized in Table 3. In summary, most available data are limited to small samples, which greatly limits our ability to perceive possible benefits. Thus, there is currently little consensus regarding the true benefits of these agents. However, there are multiple ongoing multicenter clinical trials that can potentially demonstrate the true utility of several of these agents.

Table 3.

Summary of Different Mechanisms, Strengths, and Weaknesses of Current Epigenetic Therapies

Therapeutic	Mechanism of Action	Strengths	Weaknesses
IDHm inhibitors	Promotes differentiation in tumorigenic cells Indirectly downregulates 2-HG production	Low adverse event rate Can be effectively monitored via measurement of serum 2-HG	Can only be used in IDHm tumors Further mutations in IDHm tumors may impact the efficacy of IDHm inhibitors Poor BBB permeability
HDAC inhibitors	Induces cell cycle arrest and apoptosis in tumor cells	Can be used in both IDHw and IDHm tumors Limited efficacy	High adverse event rate Poor BBB permeability
Bromodomain ininhibitors	Inhibition of DSB repair in tumor cells Amplifies the cytotoxic effects of other chemotherapies	Can be used in a wide variety of malignancies	Limited studies
5-Azacitidine	Upregulates immunomodulatory genes	Synergistic association with TMZ	Side-effect profile shows frequent leukopenia and neutropenia
PARP inhibitors	Leads to accumulation of double-strand breaks, leading to cytotoxicity	May have potential for long-term disease stabilization High degree of brain penetration	Typically only effective in tumors with poor homologous recombination, may require concomitant DNAtargeted therapy
NAMPT inhibitors	Prevents regeneration of NAD, leading to ATP depletion and tumor cell death	Unknown, all prospective trials were ceased	High levels of toxicity, leading to cessation of all trials

BBB, blood-brain barrier; TMZ, temozolomide; 2-HG, 2-hydroxyglutarate; IDHm, IDH-mutant; IDHw, IDH wild-type; NAMPT, nicotinamide phosphoribosyl transferase; PARP, poly-ADP ribose polymerase inhibitor; DSB, double-strand break.

Study Limitations

Due to the nature of the studies included in our report, this study is based on a summary of heterogeneous clinical trials and constitutes class III evidence. The trials included in this systemtatic review may be subject to selection bias and reporting bias. Additionally, the heterogenous patient cohorts in all the studies may also complicate pooled analyses and comparisons across medications. A direct comparison of HDAC inhibitors and IDH1m inhibitors among patients matched for age is necessary to resolve this potential confounder. Furthermore, most studies did not have sufficiently large cohorts of patients with IDHm tumors; thus, results of these studies may not be generalizable to this subset of patients.

Combination therapy may be an alternative to current monotherapy regimens, particularly if used early in the disease course. For example, one clinical trial investigating combination therapy in AML has demonstrated a 78% response rate among patients receiving a combination of ivosidenib and 5-AZ and a 67% response rate among patients receiving a combination of enasidenib and 5-AZ.⁴¹ Unfortunately, similar research for use in IDHm glioma and glioblastoma has largely been preclinical.⁴³ However, our own review of the current literature suggests that many of these epigenetic agents may have narrow therapeutic indices, and thus, combination therapies must be balanced by their potential toxicities.

CONCLUSIONS

We performed the first-known systematic review investigating current epigenetic therapies in management of IDH-mutant gliomas. Among current therapies demonstrated in clinical studies, IDH inhibitors appear to be the most promising given their benign safety profile and the presence of 2-HG as a potential biomarker. In contrast, HDAC inhibitors appear to be more limited as current studies have not demonstrated a clinically significant benefit. Furthermore, HDAC inhibitors appear to have a narrow therapeutic index, given the relatively high rate of adverse events documented in previous studies. Bromodomain inhibitors may be a promising third class of chemotherapeutics; however, more research is necessary to elucidate their true utility. Future research in combination epigenetic therapies may be necessary as demonstrated in several ongoing clinical trials in patients with AML; however, this strategy must be tailored to limit dose-related toxicities of the respective drugs. This may be particularly true of the newer NAMPT and PARP inhibitors that may require combination therapy to demonstrate efficacy, while limiting toxicities. Generally speaking, epigenetic therapies may have a particular value in a subset of patients with nonenhancing gliomas, though further efficacy studies in agents aside from IDHm inhibitors will be valuable in determineing the scope of epigenetic therapy utility.

Abbreviations and Acronyms

α-KG

α-ketoglutarate

2-DG

2-hydroxyglutarate

5-AZ

5-azacitidine

AML

Acute myeloid leukemia

BBB

Blood-brain barrier

DNMT

DNA methyltransferase

GBM

Glioblastoma multiforme

HDAC

Histone deacetylase

IDH

Isocitrate dehydrogenase

IGHm

Isocitrate dehydrogenase mutant

NAMPT

Nicotinamide phosphoribosyl transferase

NAMPTI

Nicotinamide phosphoribosyl transferase inhibitor

OS

Overall survival

PARP

Poly-ADP ribose polymerase

PARPI

Poly-ADP ribose polymerase inhibitor

PFS

Progression-free survival

Data Availability:

Data sharing is not applicable to this article as no data sets were generated or analysed during the current study.