Potent preclinical sensitivity to imipridone-based combination therapies in oncohistone H3K27M-mutant diffuse intrinsic pontine glioma is associated with induction of the integrated stress response, TRAIL death receptor DR5, reduced ClpX and apoptosis

Abstract

The H3K27M oncohistone mutation, identified in approximately 80% of diffuse intrinsic pontine gliomas (DIPG), is a potential target for therapy. Imipridone ONC201/TIC10 (TRAIL-Inducing Compound #10) induces apoptosis of cancer cells, and has clinical efficacy against H3K27M-mutant DIPG. We demonstrate synergy between ONC201, ONC206 and ONC212, and targeted therapies with known preclinical activity against DIPG. We hypothesized that imipridone combinations with HDAC or proteasome inhibitors may be superior to single agent ONC201 treatment in H3K27M mutant DIPG. Six patient-derived DIPG cell lines (SU-DIPG-IV, SU-DIPG-13, SU-DIPG-25, SU-DIPG-27, SU-DIPG-29, SU-DIPG-36) were exposed to imipridones alone or combinations with histone de-acetylase inhibitors [HDACi], marizomib, etoposide, and temozolomide. Dose-dependent response to imipridones was observed in DIPG cells with half-maximal inhibitory concentration (IC₅₀) of 1.46 µM, 0.11 µM, and 0.03 µM, for ONC201, ONC206, and ONC212, respectively. Upon treatment with the imipridones, DIPG cell lines engaged CLpP/CLPX, the integrated stress response with ATF4 activation, and TRAIL death receptor 5 (DR5) induction. Strong synergy was identified between ONC201 and HDACi panobinostat (combination index [CI] 0.01), romidepsin (CI 0.08) and proteasome inhibitor marizomib (CI 0.19). Synergy was demonstrated between ONC201 and etoposide (CI 0.54), although to a lesser degree than with panobinostat, romidepsin, and marizomib. ONC206 and ONC212 showed similar synergistic effects with panobinostat, romidepsin, and marizomib. Induction of apoptosis was demonstrated with imipridones and panobinostat or romidepsin combinations. Our results suggest increased sensitivity of H3K27M-mutant DIPG cell lines to second generation imipridone therapies, as compared to ONC201. Additionally, there is synergistic cell death with combination of imipridones and panobinostat, romidepsin, or marizomib, which may be further tested in vivo and in clinical trials.

Introduction

First described in 1926 by Dr. Wilfred Harris [1], diffuse intrinsic pontine glioma (DIPG), represents 10% of all pediatric central nervous system (CNS) tumors. The median age of patients affected is 6-7 years [2,3]. Despite aggressive therapeutic approaches, pediatric malignant gliomas remain the leading cause of cancer related deaths in children [4,5], with a greater than 90% mortality rate by 18-24 months from diagnosis [6].

Until recently, there has been a lack of progress in therapeutic approaches to this disease, mostly due to a lack of tissue and comprehensive molecular analyses. However, with recent advances in molecular sequencing, the oncohistone H3K27M mutation has been identified in approximately 80% of pediatric DIPG [7], and 50% of other midline gliomas [8], representing a potential new therapeutic target.

Over the past two decades there has been an increasing number of clinical trials focused on chemotherapeutics targeting DIPG. Focal radiotherapy was found to be just as effective as whole brain irradiation, and therefore became the standard of care, resulting in an improvement in symptoms for many patients, but with ultimately no change in prognosis [9]. Regrettably, chemotherapeutic agents have been evaluated with administration concurrent to radiotherapy, with no improvement in outcomes [9].

Newer innovative approaches have included intra-arterial chemotherapy, intra-nasal chemotherapy, and convection enhanced delivery, in which chemotherapy is injected directly into the tumor to bypass the blood brain barrier (BBB) [9]. More recently, trials have included targeted agents such as tyrosine kinase inhibitors, PI3K/mTOR pathway inhibitors, and antiangiogenetic agents [9]. Despite these attempts, the prognosis of this disease remains fatal, and better treatments are urgently needed.

ONC201 is a first-in-class small molecule in the imipridone family [10-12]. It has been shown to cross the blood brain barrier, and induce apoptosis/growth arrest in several different tumor types by upregulation of TNF-related apoptosis inducing ligand (TRAIL) pathway and its pro-apoptotic death receptor (DR5), independent of p53 [12,13]. Initial use in phase 1 and 2 clinical trials have thus far shown promise in the use of ONC201 in patients with diffuse midline gliomas, as well as an impressive safety profile [14]. One patient with DIPG was treated with adjuvant ONC201 demonstrated both radiographic response and a significant improvement in her facial palsies and hearing loss [15]. Another patient with GBM received ONC201 for 12 months prior to progression. The patient subsequently underwent radiotherapy and 8 more months of chemotherapy, before succumbing to the disease [15]. Finally, a nine-year-old patient with an H3K27M mutant right thalamic diffuse midline glioma demonstrated partial response to ONC201 [16].

ONC206 is a Dopamine Receptor D2 (DRD2) antagonist, which is hypothesized to target certain DRD2-dysregulated tumors which ONC201 may not significantly affect. It has a similar selective anti-tumor effect to ONC201 which has been demonstrated in multiple tumor types, including glioblastoma multiforme (GBM) cells. ONC206 and ONC212 are fluorinated ONC201 derivatives with nanomolar potency [17]. Their anti-tumor effect (as well as that of ONC201) is known to involve hyperactivation of the caseinolytic mitochondrial matrix peptidase proteolytic (ClpP) subunit, initiating tumor specific apoptosis. Thus, there is interest in further evaluation of imipridone analogs such as ONC206 and ONC212 in GBM and DIPG.

Interestingly, in the preclinical setting, an H3K27M-mutant patient-derived DIPG-derived cell line SF8628 has been noted to be less responsive to ONC201 treatment as would be expected [18]. Thus, one of our goals was to evaluate ONC201 across a larger panel of established H3K27M-mutated DIPG cell lines. In addition, we wanted to investigate second generation imipridones, ONC206 and ONC212 for preclinical efficacy in DIPG. ONC206 is currently being explored in an adult phase 1 study targeting recurrent glioblastoma at the NCI (NCT04541082) while ONC212 has not yet entered clinical trials.

To further expand our understanding of the role of imipridones, we hypothesized that the combination of ONC201 and other targeted agents may be superior to single agent treatment in H3K27M mutant DIPG, and that second generation imipridones (ONC206 and ONC212) may demonstrate superior activity to ONC201 for H3K27M DIPG in combination with other chemotherapy agents.

Materials and methods

Cell culture and reagents

All DIPG cell lines originated at Stanford university and were generously provided by Dr. M. Monje to our group. The cells were maintained in Neurobasal^TM-A Medium, enriched with Antibiotic-Antimycotic liquid, B-27 supplement minus vitamin A, sodium pyruvate solution, non-essential amino acids solution, glutaMAX, HEPES Buffer solution (all purchased from Thermo Fisher Scientific Inc., Invitrogen brand, Carlsbad, CA, USA), 0.2% Heparin (from STEMCELL^TM Technologies, Vancouver, BC, Canada), human PDGF-BB, human PDGF-AA, human FGF-basic 154 aa(FGF2), and human EGF (from Shenandoah Biotechology Inc., Warwick, PA, USA). ONC201, ONC206, and ONC212 were obtained from Oncoceutics, Inc. (Philadelphia, PA, USA). Panobinostat, romidepsin, and temozolomide were purchased from MedKoo Biosciences Inc. (Morrisville, NC, USA), marizomib from Biovision, Inc. (Milpitas, CA, USA), and etoposide from Santa Cruz Biotechology Inc. (Dallas, TX, USA).

Cell viability assay and synergy assessment

Cell viability was measured by CellTiter-Glo (CTG, Promega, Madison, WI). 5×10³ cells were seeded in neurobasal medium in a 96-well plate, and kept overnight at 37°C with 5% CO₂. Cells were subsequently treated with designated chemotherapeutic(s) at indicated concentrations. Luminescent-based cell viability was determined by CTG assay after 96 hours of incubation, following manufacturer’s instructions. Normalization of luminescence signals to control wells was used to determine percent of cell viability. Results are reported as percent viability ± standard deviation. Dose response curves were generated, and the half maximal inhibitory concentration (IC₅₀) was calculated using GraphPad Prism version 6 (San Diego, CA, USA). Compusyn software (ComboSyn, Inc.) was used to calculate combination indices (CI).

Immunoblotting

A total of 6×10⁶ cells were seeded in neurobasal medium in a 6-well plate and incubated overnight at 37°C with 5% CO₂. Cells were subsequently treated with either vehicle or designated chemotherapeutic(s) at designated concentrations and returned to incubation. After 72 hours, adherent cells were mechanically detached, washed with phosphate-buffered saline (PBS), and lysed with RIPA buffer (Sigma-Aldrich, St. Louis, MO, USA). Protein concentration was determined using BCA protein Assay Kit (Thermo Fisher Scientific, Carlsbad, CA, USA). Equal amounts of proteins were loaded onto NuPAGE 4-12% Bis-Tris gel (Thermo Fisher Scientific), and subsequently transferred to a polyvinylidene difluoride membrane (PVDF) membrane. Membranes were blocked with Tris-buffered saline with Tween 20 (TBST) with 5% w/v instant nonfat dry milk and incubated with primary antibody overnight at 4°C. Antibodies include PARP/cPARP (Cell Signaling 9546), ATF4 (Cell Signaling 11815), CLpP (Cell Signaling 14181S), CLPX (Sigma Aldrich HPA040262), DR5 (Cell Signaling 36965), RAN (Biosciences 610341), and Actin (Sigma Aldrich P68133). Membranes were washed with TBST and incubated for an hour at room temperature with the appropriate secondary antibodies-anti-mouse IgG (Thermo Scientific 31430) or anti-rabbit IgG (Cell Signaling 51275). Chemiluminescence reaction was detected using the ECL Reagent (Thermo Fisher Scientific). Ran and actin were used as loading controls.

Flow cytometry

A total of 6×10⁶ cells were seeded in neurobasal medium in a 6-well plate and incubated overnight at 37°C with 5% CO₂. Cells were then treated with either vehicle or chemotherapeutic(s) at designated concentrations and returned to incubation. After 96 hours, cells were detached, washed with PBS/1% fetal bovine serum (FBS), and resuspended in PBS. Cold ethanol alcohol was added, and cells were incubated at 4°C for 20 minutes. Cells were washed with PBS/1% FBS, and resuspended in PBS. Phosphate-citric acid buffer was added, and cells were incubated at room temperature for 5 minutes. Cells were resuspended in propidium iodide/RNase staining solution, incubated at room temperature for 30 minutes, and subsequently analyzed by flow cytometry.

Statistics

Statistical analyses for cell viability assays were conducted using GraphPad Prism 6. Synergy analyses were conducted with Compusyn software.

Results

Imipridone analogs ONC206 and ONC212 are more potent cytotoxic agents than ONC201 against H3K27M mutant DIPG

To evaluate for dose-dependent response to imipridone therapy, we used single treatment viability assays (CellTiterGlo). Each of our six DIPG cell lines (Supplementary Table 1) was treated with ONC201 and imipridone analogs, ONC206 and ONC212. Half maximal inhibitory concentration (IC₅₀) was calculated for each drug on all cell lines. Our results demonstrate a higher IC₅₀ in all cell lines when exposed to ONC201, as compared to treatment with imipridone analogs (Figure 1). Furthermore, treatment with ONC212 demonstrated lower IC₅₀ in each DIPG cell line, indicating increased potency of this drug. Mean IC₅₀ were found to be: 1.46 µM in ONC201, 0.11 µM in ONC206, and 0.03 µM in ONC212. A summary of the IC₅₀ results with the 3 imipridones against 6 DIPG cell lines in shown in Table 1.

Table 1.

IC₅₀ values of ONC201, ONC206, and ONC212 in each of the six human derived DIPG cell lines

Drug (µM)	SU-DIPG-IV	SU-DIPG-13	SU-DIPG-25	SU-DIPG-27	SU-DIPG-29
ONC201	1.358-1.787	1.469-1.648	1.058	1.173	1.563
ONC206	0.095-0.173	0.062-0.157	0.05	0.058	0.081
ONC212	0.018-0.036	0.023-0.042	0.026	0.014	0.015

Figure 1.

Dose dependent response demonstrates that imipridone analogs ONC206 and ONC212 are more potent cytotoxic agents than ONC201 against H3K27M mutant DMG. A. Single treatment viability assay as measured by CellTiter-Glo (CTG) of ONC201, ONC206, and ONC212 in cell line SU-DIPG-IV. B. Graphic representation and table of IC₅₀ values of imipridones effect on cell line SU-DIPG-IV. C. Single treatment viability assay as measured by CTG of ONC201, ONC206, and ONC212 in cell line SU-DIPG-13. D. Graphic representation and table of IC₅₀ values of the effect of imipridones on cell line SU-DIPG-13. Cells were treated for 96 hours before assessment of cell viability.

Combinations of ONC201 and HDAC inhibitors panobinostat, romidepsin, and entinostat show synergistic activity against H3K27M mutant DIPG

Given the importance of the epigenetic mutation H3K27M to clinical efficacy of ONC201, we explored various combinations of epigenetic drugs to identify those which demonstrate promising preclinical activity. We chose to evaluate pan-HDAC inhibitors panobinostat, romidepsin, and entinostat in combination with imipridone agents against our panel of DIPG cell lines. Synergy was identified via combination analyses (Figure 2 and Supplementary Figure 1), with combination indices of <1 indicating synergy, and <0.5 indicating very strong synergy [19]. Values that meet this definition are highlighted in yellow in Figure 2 and Supplementary Figure 1. All combinations were tested for 96 hours before assessment of cell viability. We observed potent synergy between panobinostat and ONC201 across two DIPG cell lines, SU-DIPG-13 and SU-DIPG-IV, with the best CIs of 0.01 and 0.22 respectively (Figure 2A, 2B). The combination of ONC201 and romidepsin similarly demonstrated robust synergy in cell lines SU-DIPG-13 and SU-DIPG-29, with the best combination index (CI) of 0.02 in SU-DIPG-13 and 0.18 in SU-DIPG-29 (Figure 2C, 2D). Synergy analyses between entinostat and ONC201, revealed a lesser degree of synergy, with the best CI of 0.71 (Supplementary Figure 1).

Figure 2.

Combination of ONC201 and HDAC inhibitors reveal synergistic activity against H3K27M mutant DIPG. Synergy analyses as measured by CTG, demonstrating the combination of (A) ONC201 and panobinostat in cell line SU-DIPG-13, with best combination index (CI) of 0.01 (B) ONC201 and panobinostat in cell line SU-DIPG-IV, with best CI of 0.01 (C) ONC201 and romidepsin in cell line SU-DIPG-13, with best CI of 0.02 (D) ONC201 and romidepsin in cell line SU-DIPG-29, with best CI of 0.18. Combinations were tested for 96 hours before assessment of cell viability.

Combination of ONC201 and proteasome inhibitors show synergy in treatment of H3K27M mutant DIPG

Prior preclinical experiments suggested that proteasome inhibitors and in particular marizomib, appear promising as therapeutics for H3K27M mutant DIPG. Marizomib crosses the blood brain barrier as has been demonstrated in preclinical studies. Previous studies also showed that TRAIL receptor DR5 could be upregulated by proteasome inhibitors [20]. Thus, since ONC201 can upregulate TRAIL as well as DR5, we hypothesized that it would be reasonable to evaluate the effects of the combination in DIPG cell lines. The combination of ONC201 and proteasome inhibitor marizomib at 96 hours, demonstrated promising synergy in cell line SU-DIPG-IV, with the best combination index of 0.19 (Figure 3). Synergy was evident at all doses of ONC201 tested (0.3125-10 µM) and from 0.03125-0.5 µM of marizomib.

Figure 3.

Combination of ONC201 and proteasome inhibitor marizomib demonstrates promising synergy against H3K27M mutant DIPG. Synergy analysis of ONC201 and marizomib in cell line SU-DIPG-IV as measured by CTG, with best combination index of 0.19. Combinations were tested for 96 hours before assessment of cell viability.

Combinations of ONC201 and temozolomide or etoposide are synergistic against H3K27M mutant DIPG cell lines

Chemotherapeutic agents used to treat malignant gliomas include temozolomide and etoposide. In considering the potential for future clinical trials, temozolomide and etoposide were tested for synergy with ONC201 (Figure 4). Combination of etoposide and ONC201 demonstrated mild synergism with the best CI of 0.46 (Figure 4A). DIPG cell lines treated with temozolomide, both as monotherapy and in combination, appeared to be resistant to this chemotherapeutic (Figure 4B). There is nevertheless evidence of synergy between temozolomide and ONC201 in two cell lines (SU-DIPG-IV, SU-DIPG-13) at both low (0.3125 µM) and high (2.5 µM, 5 µM) doses of ONC201 with the best combination index of 0.4 in both lines.

Figure 4.

Mild synergy revealed via combination analysis of ONC201 and etoposide or temozolomide. A. Combination of ONC201 and etoposide with best combination index of 0.46 in cell line SU-DIPG-IV, and 0.53 in cell line SU-DIPG-25. B. Combination of ONC201 and temozolomide with best combination index of 0.35 in cell line SU-DIPG-IV and 0.44 in cell line SU-DIPG-13. Combinations were tested for 96 hours before assessment of cell viability.

Second generation imipridones (ONC206 and ONC212) demonstrate potent synergy against H3K27M mutant DIPG in combination with HDACi, proteasome inhibitors, and other chemotherapeutic agents

As ONC206 and ONC212 demonstrated greater potency against DIPG cell lines, we further investigated their potential to be combined with therapeutic agents that sensitized to ONC201. Similar combinations were assessed using the imipridone analogs, with comparable results demonstrating synergy (Figure 5 and Supplementary Figure 2). Lower doses were used for both ONC206 and ONC212 as compared to ONC201 given their increased potency (Figure 5 and Supplementary Figure 2). Panobinostat and ONC206 presented a combination index of 0.54, while ONC206 and romidepsin demonstrated a combination index of 0.1. Entinostat and etoposide both showed mild synergism with ONC206 with the best combination indices of 0.59 and 0.46 respectively. Finally, robust synergy between marizomib and ONC206 was illustrated with the best CI of 0.07.

Figure 5.

Second generation imipridones synergize with HDAC and proteasome inhibitors. A. Combination analysis of ONC206 and panobinostat in cell line SU-DIPG-IV, with best combination index of 0.54. B. Synergy analysis of ONC206 and marizomib has best CI of 0.07 in cell line SU-DIPG-36. C. Synergy analysis of ONC212 and marizomib in cell line SU-DIPG-IV with best combination index of 0.29. D. Combination of ONC212 and romidepsin in cell line SU-DIPG-IV shows best combination index of 0.24. All combinations were tested for 96 hours before assessment of cell viability.

Apoptotic cell death induction following treatment of DIPG cell lines with imipridones

Given the observations with regard to drug effects on cell viability, we investigated the cell death induction following imipridone treatment of DIPG cell lines by flow cytometry. Data was analyzed for sub-G1 DNA after drug treatment. The results show higher percentage of cell death with imipridones and a lesser extent of apoptosis with both panobinostat and etoposide as single drug treatments (Figure 6). Additionally, there is a much higher rate of cell death at 96 hours as compared to 48 hours of incubation (Figure 7). The majority of the DIPG cells treated by the imipridones were found to be in sub-G1 phase after 96 hours, indicative of extremely potent pro-apoptotic effects against the H3K27M-mutated DIPGs (Figures 6, 7).

Figure 6.

Imipridone therapies induce tumor cell apoptosis. Sub-G1 DNA content of SU-DIPG-13 cells treated with ONC201, ONC206, ONC212, Panobinostat, and etoposide for 96 hours at equitoxic doses. All combinations were tested for 96 hours before assessment of cell viability.

Figure 7.

Imipridone induced apoptosis is dependent on timeframe of incubation period. Sub-G1 DNA content of SU-DIPG-13 cells treated with equitoxic doses of ONC201 and ONC206 at both 48 and 96 hours.

Upon treatment with imipridone monotherapy, DIPG cell lines engage CLpP/CLPX, the integrated stress response with ATF4 activation, and TRAIL death receptor 5 (DR5) induction

Our previous work has documented induction of the integrated stress response involving ATF4, CHOP, and TRAIL death receptor 5 (DR5) after the treatment of multiple human tumor cell lines with TIC10/ONC201 [10,21]. We therefore investigated activation of the integrated stress response in DIPG cells with H3K27M mutation after treatment with ONC201 as well as 2^nd generation imipridones ONC206 and ONC212. We also analyzed changes in mitochondrial caseinolytic protease CLpP and its regulator CLPX as the imipridones act as agonists to block mitochondrial oxidative phosphorylation [22,23].

We found that SU-DIPG-IV cells upregulate ATF4 and DR5 after treatment with ONC206 (Figure 8A). We note with ONC206, as we have observed previously with other imipridones [24] reduced expression of CLPX consistent with engagement of CLpP to inhibit oxidative phosphorylation. In these SU-DIPG-IV cells there was a correlation between apoptotic cell death as measured by PARP cleavage, engagement of CLpP, and upregulation of ATF4 and DR5 (Figure 8A). We made similar observations in SUDIPG-36 after treatment with ONC206 with induction of death receptor 5 and modest PARP cleavage at 48 hours (Figure 8B). After ONC212 treatment of SU-DIPG-36 we observed elimination of expression of CLPX, upregulation of DR5 (Figure 8C). A modest PARP cleavage was also observed at 48 hours after treatment of SU-DIPG-36 by ONC212 (Figure 8C).

Figure 8.

Imipridone single agent therapy induces integrated stress response as demonstrated by upregulation of ATF4 and DR5. A. Western blot analysis of SU-DIPG-IV cells treated with both high and low dose ONC206 monotherapy incubated for 48 hours. B. Western blot analysis of SU-DIPG-36 cells treated with high and low dose ONC206 as single agent therapy, incubated for 48 hours. C. Western blot analysis of SU-DIPG-36 cells treated with high and low dose ONC212, incubated for 48 hours.

Induction of the integrated stress response and apoptotic cell death markers is modulated by imipridone combinations with HDAC inhibitors panobinostat and romidepsin

We further explored the effects of imipridones on induction of the integrated stress response, mediators or inhibitors of cell death and markers of apoptosis. We examined DIPG cell line SU-DIPG-13 upon treatment with increasing doses of ONC201, ONC206, and ONC212 as monotherapies after 96 hours (Figure 9). All three imipridone drugs upregulated ATF4 in the DIPG cells at equitoxic doses. We combined the imipridones with HDAC inhibitors panobinostat and romidepsin and observed increased induction of poly (ADP-ribose) polymerase (PARP) cleavage with the combination therapies as compared to single agents. Notably, cleaved PARP was increased with use of lower (equitoxic) doses of ONC206 and ONC212 as compared to ONC201 (Supplementary Figure 3). Thus, we have shown that all three imipridones (ONC201, ONC206, and ONC212) activate the integrated stress response to DIPG cells and consistent with the earlier observed synergies with HDAC inhibitors, the combinations induced PARP cleavage indicative of apoptosis. We did not observe suppression of anti-apoptotic proteins Bcl-XL or XIAP in SU-DIPG-13 after treatment with ONC212 (data not shown).

Figure 9.

Integrated stress response and apoptosis induced by imipridones as monotherapy and in combination with other therapeutic agents. Western blot analysis of SU-DIPG-13 cells treated with (A) ONC201 at varying doses, both alone and in combination with panobinostat or romidepsin (B) ONC206 at both low and high doses, alone and in combination with panobinostat or romidepsin (C) Low and high dose ONC212, as single agent therapy and in combination with panobinostat or romidepsin.

Discussion

There is an urgent need to explore new treatment options for pediatric diffuse intrinsic pontine glioma and other malignant gliomas as they remain the most common cause of childhood cancer-related mortality [5]. Novel imipridone therapies, ONC201, ONC206, and ONC212, represent a promising new approach. We compared for the first time the relative effects of these novel therapeutics alone or in combination with other chemotherapy agents such as HDAC inhibitors, proteasome inhibitors or other chemotherapeutics in pediatric DIPG cells lines. We observed synergies between the imipridones and HDAC inhibitors, proteasome inhibitors, or etoposide in killing of DIPG cells. These effects on cell viability were associated with induction of the integrated stress response as demonstrated by increased ATF4 in DIPG cells treated with any of the three imipridones.

Preclinical high-throughput combination drug screening exploring >2000 agents and >9000 drug combinations identified several potential drugs with potent effects on DIPG and classified them by potency and brain penetration. These included marizomib, romidepsin, and panobinostat, providing a framework on which we based our decision to trial these medications in combination with imipridones [25]. Lin et al. (2019) demonstrate that proteasome inhibitors and HDACi have potent synergy when combined, however, here we were interested in testing each one in combination with imipridone therapy. Proteasome inhibitors are known to induce apoptosis in cultured cancer cell lines as well as murine models. Discovered in the 1980’s, the ubiquitin proteasome pathway (UPP) has been identified as a key element of protein degradation in eukaryotic cells. It is the central pathway of protein regulation, supporting DNA repair, cell survival, cell cycle progression, and apoptosis. Cancer cells are well known to dysregulate this pathway leading to uninhibited cell proliferation. Marizomib targets the proteasome more broadly than other drugs in this class and has shown increased activity in combination therapies [26,27]. HDACs are known to alter the balance of acetylation of the histone terminal lysine group, thereby regulating transcription and translation and activating gene expression. HDAC inhibitors are a powerful group of chemotherapeutics which initiate changes in histone methylation and acetylation. They not only lead to apoptosis, but can inhibit angiogenesis and disrupt cellular growth and differentiation [28].

Our study provides evidence that DIPG cell lines have a significant dose-dependent response to imipridone therapies. We found that ONC206 and ONC212 were more potent in killing DIPG cell lines as compared to ONC201, with ONC212 showing greater potency compared to ONC206. Second generation imipridones ONC206 and ONC212 were more potent at apoptosis induction of DIPG cell lines. All of the imipridones induced the integrated stress response with induction of ATF4, TRAIL death receptor 5 (DR5), and reduced CLPX expression in all DIPG cell lines that were tested. It is noteworthy that ONC212, which has not been implicated as acting through blockade of dopamine receptors DRD2 and DRD3 but rather is believed to in part act as an agonist of GPR132, is more potent than either ONC201 or ONC206 at killing DIPG cell lines in culture. ONC212, like the other imipridones activates the ISR and DR5 signaling to induce apoptosis in DIPG cells.

Imipridones synergize well in vitro with HDAC inhibitors, panobinostat and romidepsin, as well as proteasome inhibitor marizomib in the treatment of DIPG. Immunoblotting showed increased integrated stress response signaling induced by imipridone combinations with panobinostat or romidepsin, further supporting their synergistic properties.

Notably, an international multi-center preclinical trial using both zebrafish and murine models of DIPG was able to demonstrate increased survival in ONC201 and ONC206 as mono-therapies, as well as with a combination of these two imipridones. Additionally, they demonstrated significant survival benefit with the use of a combination of ONC201 and panobinostat. Treatment with marizomib was attempted, however, there was toxicity demonstrated in both animal models [29].

Temozolomide and ONC201 did not appear to show combinational efficacy against H3K27M mutant DIPG cell lines SU-DIPG-IV and SUDIPG-13. Previous studies have not demonstrated efficacy of temozolomide against DIPG [30,31]. Despite some evidence of synergy at both high and low doses of ONC201, there is a lack of cell death at doses as high as 2000 µM temozolomide. ONC206 demonstrated promising synergy with both romidepsin and marizomib across a wide range of doses. Both of these medications also revealed a strong, though not as ubiquitous, synergistic relationship with ONC212. Synergy was further validated via immunoblotting, with evidence of increased integrated stress response and increased apoptosis.

As historically there has been scarce tissue obtained from these tumors, the availability of cell lines proved to be limited. As such, we were unable to obtain a pediatric H3K27M wild-type DIPG cell line for comparison.

Our results indicate increased sensitivity to imipridone analogs, relative to ONC201 therapy, in H3K27M mutant DIPG cells, both as monotherapy and in combination therapies with other agents. Our data suggests that a combination of second generation imipridones with panobinostat, marizomib, or romidepsin could be further explored in the treatment of pediatric DIPG. Caution will need to be taken with use of marizomib as a proteasome inhibitor in combination studies given the previous group’s experience with toxicity in their animal models [29]. Based on the findings we report, ongoing clinical studies should investigate more closely biomarkers of the integrated stress response, CLPX, and TRAIL death receptor DR5 expression in pre- and post-treatment biopsies with imipridone therapeutics administered to patients with midline gliomas. Our results provide insight into potential strategies for new therapeutic approaches to pediatric DIPG, which should be explored in in vivo models for translation into future clinical trials.

Disclosure of conflict of interest

W.S.E-D. is a co-founder and shareholder of Oncoceutics, Inc., a subsidiary of Chimerix. Dr. El-Deiry has disclosed his relationships with Oncoceutics, Inc., Chimerix, and potential conflict of interest to his academic institution/employer and is compliant with the institutional policy that is managing this potential conflict of interest. V.V.P. is an employee and shareholder of Chimerix.