Introducing HDAC-Targeting Radiopharmaceuticals for Glioblastoma Imaging and Therapy

Despite recent advances in multimodality therapy for glioblastoma (GB) incorporating surgery, radiotherapy, chemotherapy and targeted therapy, the overall prognosis remains poor. One of the interesting targets for GB therapy is the histone deacetylase family (HDAC). Due to their pleiotropic effects on, e.g., DNA repair, cell proliferation, differentiation, apoptosis and cell cycle, HDAC inhibitors have gained a lot of attention in the last decade as anti-cancer agents. Despite their known underlying mechanism, their therapeutic activity is not well-defined. In this review, an extensive overview is given of the current status of HDAC inhibitors for GB therapy, followed by an overview of current HDAC-targeting radiopharmaceuticals. Imaging HDAC expression or activity could provide key insights regarding the role of HDAC enzymes in gliomagenesis, thus identifying patients likely to benefit from HDACi-targeted therapy.


Introduction
Glioblastoma multiforme (GB) is the most malignant tumor in the central nervous system (CNS). Despite recent advances in multimodality therapy for GB incorporating surgery, radiotherapy (RT), chemotherapy and targeted therapy, the overall prognosis remains poor. Almost all tumors recur with a more aggressive form, and there is no standard of care for recurrent GB. The survival rate at 5 years postdiagnosis remains at only 5.8% [1][2][3]. Novel molecular markers were identified improving GB classification and providing powerful prognostic information [4]. However, therapy resistance remains a hurdle. Precision oncology incorporating personalized targeted therapy holds much promise in developing more efficacious and tolerable therapies [3]. One of the interesting targets for GB-targeted therapy is the histone deacetylase family (HDAC). Due to their pleiotropic effects on, e.g., DNA repair, cell proliferation, differentiation, apoptosis and senescence, they have gained a lot of attention in the last decade as anti-cancer agents. In addition, HDAC inhibitors (HDACi) have been applied for the treatment of metabolic disorders and psychiatric or neurodegenerative diseases [5]. The HDAC family contains 18 family members, categorized as following: class I (HDAC1, 2,3,8), IIa (HDAC 4,5,7,9), IIb (HDAC 6,10), III (nicotinamide adenine dinucleotide (NAD+)-dependent sirtuins (SIRTs) and IV (HDAC11) [6,7]. Two groups of enzymes control the acetylation and deacetylation of histones: histone acetyltransferase (HAT) and HDACs. The transfer or removal of acetyl groups by HATs and HDACs induce a more open and accessible chromatin structure or

Current Status of HDACi for GB Therapy
An overview of the successful clinical trials investigating HDACi in high-grade glioma is given in Table 1. Studies in pediatric glioma patients were excluded. The previous reviews have focused on the mechanisms of HDACi in GB [11,38].
Depsipeptide romidepsin (Istodax ® , FR901228, FK228) is a stable prodrug isolated from Chromobacterium violaceum and a class I HDACi [49,50]. In a phase I/II clinical trial in recurrent GB, romidepsin was found to be ineffective as a single agent [49].

Current Status of HDACi for GB Therapy
An overview of the successful clinical trials investigating HDACi in high-grade glioma is given in Table 1. Studies in pediatric glioma patients were excluded. The previous reviews have focused on the mechanisms of HDACi in GB [11,38].
Depsipeptide romidepsin (Istodax ® , FR901228, FK228) is a stable prodrug isolated from Chromobacterium violaceum and a class I HDACi [49,50]. In a phase I/II clinical trial in recurrent GB, romidepsin was found to be ineffective as a single agent [49].
Panobinostat (LBH589), a pan-deacetylase inhibitor of class I/II HDAC, is an antineoplastic and antiangiogenic drug that may work synergistically with BEV [51]. However, although this combined treatment strategy was well-tolerated, PFS and OS did not signifi-cantly improve compared to BEV monotherapy in recurrent GB [52]. A phase II trial is warranted to assess the combination with fractionated stereotactic re-irradiation therapy [53]. Panobinostat does not cross the BBB, and hence intratumoral or convection-enhanced delivery (CED) administration could be necessary [54].
In the last 2 decades, an extensive amount of preclinical research on HDACi and multidrug combinations in GB has been performed (see Supplementary Table S1) [19,33,. These studies provided new insights on HDACi-associated signaling processes.
Post-HDACi therapy, multiple genes that play a role in complex signaling pathways are up-or down-regulated, as recently summarized [11]. As expected based on preclinical data, the affected genes are involved in cell cycle progression, apoptosis, invasion and progrowth or include oncogenes and GSC markers [11].
Targeted drug combinations may beneficially affect the outcome of GB therapy, with the possible induction of synthetic lethality. Preclinically, promising combinations include a mix of epigenetic modifiers [152], HDACi combined with imipridones (activation of the mitochondrial ClpP protease) or proteasome inhibitors [153,154], panobinostat combined with a dual PI3K/mTOR inhibitor BEZ235 [155] and combining HDACi with MEK inhibitors or RTKi [156,157]. A triple combination therapy, involving panobinostat, OTX015 and sorafenib also showed potential in vitro [158]. Interestingly, the R132H mutation in isocitrate dehydrogenase 1 (IDH1R132H), commonly observed and associated with better survival in GB, has been linked to resistance to the anti-cancer effect of HDACi, such as TSA, vorinostat (SAHA) and valproic acid [159].

HDAC-Targeting Radiopharmaceuticals
The association of epigenetic dysfunction with disease and the development of diagnostic or therapeutic agents for treatment are challenging [160]. Most HDACi target a relatively wide spectrum of HDACs that, on their turn, inhibit various biological pathways. Their mechanisms of action as tumor suppressors have not yet been fully elucidated [10]. HDAC-targeting radiopharmaceuticals could provide better insights regarding HDAC tissue expression, HDACi biodistribution and pharmacokinetics and therapeutic efficacy and thereby unravel new insights into the function or behavior of HDACi in vivo [161,162]. Nuclear imaging of HDAC expression in GB may improve the understanding and roleplay of HDAC enzymes within gliomagenesis, identify patients likely to benefit from HDACitargeted therapy and aid in optimizing therapeutic doses of novel HDACi for glioma treatment [163]. Importantly, there are two main strategies to consider when imaging an epigenetic target in the brain: 1) by direct observation (protein target information independent of its activity) and 2) functional observation (representative visualization of the impact of a protein or enzyme) [160]. Alternative methods to determine HDAC expression include invasive tumor biopsies and the use of peripheral lymphocytes as surrogate biomarkers for global acetylation after HDACi treatment.
An overview of HDACi-based radiopharmaceuticals is given in Figure 2, and Table 2 summarizes the preclinical development of HDAC radiopharmaceuticals. To visualize or treat GB with radiopharmaceuticals, it is particularly important to only consider those HDACi that sufficiently pass the BBB (even at sub-nanomolar concentration) and are of a small enough structure to allow their penetration into the bulky, heterogeneous tumor tissue [29]. In addition, the cellular location of the targeted HDAC needs to be considered, e.g., HDAC class I proteins are found predominantly in the nucleus, while class II proteins are primarily localized in the cytoplasm but can be shuttled between the cytoplasm and nucleus depending on their phosphorylation status [6]. HDAC brain PET has been studied for the potential detection of various neurodegenerative diseases, such as Alzheimer's and Parkinson's disease, and limited studies have investigated their potential for glioma imaging [100,103,104]. Most HDAC radiopharmaceuticals are structurally related to SAHA and include 6-([ 18  In 2006, the first 18 F-labeled SAHA analogue ([ 18 F]FAHA) was radiosynthesized by Mukhopadhyay et al. [108]. Soon thereafter, Nishii et al. confirmed PET in vivo brain uptake in rats of up to 0.44%ID/g between 5 and 60 min [109]. Moreover, blocking studies revealed a specificity similar to that of SAHA, suggesting that [ 18 F]FAHA is a clinically To our knowledge, the potential of therapeutic HDAC radiopharmaceuticals for targeted radionuclide therapy (TRT) has not yet been explored. Importantly, possible brain toxicity may be a limiting aspect for this kind of application. HDACs play distinct physiological roles in the brain, and HDACi have pleiotropic effects due to their broad targets. This suggests a higher chance of success for isoform-specific HDACi or the necessity to inject such radioactive agents via CED directly into the GB tumor or its vicinity [164]. Another option is the use of nanovectors with theranostic properties to optimize the tumor delivery of potent HDACi, which could improve their anti-GB properties in vivo [165].
Other criteria to consider for the development of GB TRT agents were recently published by our group [166]. HDAC brain PET has been studied for the potential detection of various neurodegenerative diseases, such as Alzheimer's and Parkinson's disease, and limited studies have investigated their potential for glioma imaging [160,163,164]. Most HDAC radiopharmaceuticals are structurally related to SAHA and include 6-([ 18 [162,[167][168][169][170][171][172][173]. In 2006, the first 18 F-labeled SAHA analogue ([ 18 F]FAHA) was radiosynthesized by Mukhopadhyay et al. [168]. Soon thereafter, Nishii et al. confirmed PET in vivo brain uptake in rats of up to 0.44%ID/g between 5 and 60 min [169]. Moreover, blocking studies revealed a specificity similar to that of SAHA, suggesting that [ 18 F]FAHA is a clinically relevant PET tracer capable of targeting HDAC IIa expression [170]. [ 18 F]FAHA has also shown potential to monitor alterations in HDAC activity/expression in a rat model of chemotherapy-induced brain neurotoxicity [174]. Concerns were raised about [ 18 [172]. In 2019, [ 18 F]TFAHA was finally studied in GB rat models that confirmed tumor uptake 20 min post-radiotracer administration, which significantly reduced after administration of HDACi MC1568. [ 18 F]TFAHA accumulation was also observed in normal brain structures known to overexpress HDAC class IIa: the hippocampus, nucleus accumbens, periaqueductal gray matter and cerebellum [163].
Brain uptake was reported of another radiolabeled SAHA-analogue, [ 18 F]FE-SAHA, but its metabolic instability remains a substantial obstacle (high uptake in the kidneys, liver, bone and small intestines) [176]. Kim [173]. [ 125/131 I]-iodo-SAHA maintained a comparable profile (e.g., similar toxicity and pharmacokinetics) to SAHA. However, in tumor-bearing mice, it showed no preferential tumor accumulation, rapid efflux and unspecific washout. Moreover, accumulation in the liver and kidneys was high [177]. Thus, none of the proposed SAHA-based radiopharmaceuticals have reached a clinical phase.
Recently, radiolabeling of trifluoromethyloxadiazole (TFMO)-bearing class-IIa HDACi were explored, and NT160 was identified as a potent inhibitor of class-IIa HDAC4. [ 18 F]F-NT160 was capable of BBB crossing, and binding to class-IIa HDACs was confirmed in mouse brain tissue [183].

Challenges and Future Outlook
Although extensive research has been performed on HDACi in glioma with clear radio-and/or chemosensitizing effects, the potential of radiolabeled HDACi has only been confirmed in the field of neurodegenerative diseases and been primarily diagnostic, with the goal of quantifying HDAC expression and/or monitoring treatment response. Their potential for GB imaging and TRT is underexplored. Whilst furthering this field of research should be recommended, one of the major issues that slowed down recent translation to clinics was poor BBB penetration, poor specificity and diverse target locations. Interestingly, adamantane-conjugated radioligands seem promising to increase brain penetrance [164]. Only two radiotracers have been investigated in healthy adults: [ 11 C]martinostat and [ 18 F]F-bavarostat, confirming the ability to quantify HDAC expression [186,190]. Both should be recommended for GB HDAC imaging as they have shown target specificity and reported brain penetrance. [ 18 F]TFAHA is the only radiopharmaceutical that has been evaluated in GB models, with uptake in GB tumors but also in normal brain structures known to overexpress HDAC class IIa [163]. Another recommendable radiopharmaceutical is [ 18 F]F-NT160 featuring potent binding to class-IIa HDACs and BBB crossing in mice [183]. However, future studies are needed to increase its tumor specific uptake while preventing damage to healthy tissues.
The potential for HDACi-based radiopharmaceuticals in GB can currently be formulated as (1) biomarkers for HDAC expression, (2) elucidate the roles of HDAC class enzymes and (3) dose optimization of cold HDACi [163]. Cancer resistance and the toxic effects of HDACi are currently an issue to translate radiolabeled HDACi for potential application in TRT. HDACi are often pan-specific towards a specific HDAC class. As their substrates are present all over the human brain, targeting HDACi in GB may cause unwanted effects on healthy tissues too. However, CED could be considered to mitigate any adverse effects and circumvent the BBB. Targeting multiple HDAC proteins could also be advantageous due to the heterogeneous nature of GB. Research should be initiated to confirm this, including optimal combinatorial strategies for HDACi that permit efficacy as well as safety in GB.