ONC201 in combination with paxalisib for the treatment of H3K27-altered diffuse midline glioma

Abstract

Diffuse midline gliomas (DMG), including diffuse intrinsic pontine gliomas (DIPGs), are the most lethal of childhood cancers. Palliative radiotherapy is the only established treatment, with median patient survival of 9-11 months. ONC201 is a DRD2 antagonist and ClpP agonist that has shown preclinical and emerging clinical efficacy in DMG. However, further work is needed to identify the mechanisms of response of DIPGs to ONC201 treatment and to determine whether recurring genomic features influence response. Using a systems-biological approach, we showed that ONC201 elicits potent agonism of the mitochondrial protease ClpP to drive proteolysis of electron transport chain and tricarboxylic acid cycle proteins. DIPGs harboring PIK3CA-mutations showed increased sensitivity to ONC201, while those harboring TP53-mutations were more resistant. Metabolic adaptation and reduced sensitivity to ONC201 was promoted by redox-activated PI3K/Akt signaling, which could be counteracted using the brain penetrant PI3K/Akt inhibitor, paxalisib. Together, these discoveries coupled with the powerful anti-DIPG/DMG pharmacokinetic and pharmacodynamic properties of ONC201 and paxalisib have provided the rationale for the ongoing DIPG/DMG phase II combination clinical trial NCT05009992.

Graphical abstract

Figure 1.

DIPG patient–derived cell lines show variable response to ONC201 treatment. A, Resazurin proliferation (percentage compared with untreated) after 96 hours ONC201 exposure in DIPG patient-derived cell lines; EZHIP+ (circles) = CNMC-XD-760, DIPG-VUMC10; H3.1K27M (squares) = UON-JUMP4, SU-DIPG-IV, SU-DIPG-XXXIII, SU-DIPG-XXXVI, and H3.3K27M (triangles) = HSJD-DIPG-007, SU-DIPG-VI, SU-DIPG-XIII, SU-DIPG-XVII, SU-DIPG-XXIV, SU-DIPG-XXV, SU-DIPG-XXIX. The endothelial cell line, HCMEC/D3, SV-40–dependent human microglial line, HMC3 and neural progenitor cell line, ReN cells, were used as controls (diamonds). Values shown as mean ± SEM (n = 3). B, Annexin V apoptosis assay after 96 hours exposure with 5 μmol/L ONC201 (dark gray) compared with untreated (light gray) in SU-DIPG-XXXVI, HSJD-DIPG-007, UON-JUMP4, and SU-DIPG-XIII. Unpaired t test, values shown as mean ± SEM (n = 3). C, Representative phase contrast images of biological triplicates (n = 3) of HSJD-DIPG-007 and SU-DIPG-XIII following 6 days exposure to 1.25 μmol/L ONC201. Scale bar, 0.2 mm. D, Oncoplot of somatic mutations determined using TSO500. Cell lines ordered from the least to most sensitive to ONC201 exposure (top to bottom). Larger values of MSI and TMB are associated with increased pathogenicity. E, Proliferation data were grouped by H3 status; wt-H3 (n = 5), H3.1K27M (n = 4), and H3.3K27M (n = 7), and sensitivity to ONC201 was determined by the AUC, ± SEM. Statistical analysis was performed via nonparametric unpaired one-way ANOVA. F, Resazurin proliferation, AUC, following ONC201 exposure for 96 hours in Res259 cells harboring knockin of either H3.1K27M or H3.3K27M mutations. Statistical analysis performed via parametric unpaired t test, with Welch correction. G, Western blot validation of H3K27M knockin in Res259 cells. H,TP53 status, wt- and mutant-TP53 (n = 4 vs. n = 9), and sensitivity to ONC201 were determined by the AUC, with values shown as mean ± SEM. Statistical analysis performed via nonparametric unpaired t test. I, Resazurin proliferation, AUC, following ONC201 exposure in wt-TP53 HSJD-DIPG-007 DIPG cell lines transduced with a nontargeting control (NTC) gRNA, TP53-KD (knockdown), and TP53-KO (knockout). Statistical analysis was performed via parametric unpaired one-way ANOVA with Welch correction. J, Western Blot confirmation of TP53 KO and KD in HSJD-DIPG-007 cells. K, Validation of decreased response to ONC201 in TP53-KD or TP53-KO HSJD-DIPG-007 cell lines was performed by Western blot analysis of PARP cleavage (cPARP). L, Resazurin proliferation, AUC, following Nutlin-3 exposure for 96 hours in HSJD-DIPG-007 NTC, TP53-KD, and TP53-KO. Statistical analysis was performed via parametric unpaired one-way ANOVA with Welch correction. M and N, Proliferation data were grouped by ACVR1 status; ACVR1 wild-type (n = 8) versus ACVR1 mutant (n = 5; M) and PIK3CA status; PIK3CA wild-type (n = 9) versus PIK3CA mutant (n = 4; N) and compared with AUC following ONC201 exposure. Statistical analysis performed via nonparametric unpaired t test. *, P < 0.05; **, P < 0.01; ****, P < 0.0001.

Figure 2.

Pharmacoproteogenomic analysis identifies DRD2 and ClpP as targets of ONC201 in DIPG. A, Western blot analysis of basal DRD2, SDHA, and CLPP expression across DIPG models. B, Densitometry of protein expressions was normalized to DIPG-VUMC10 and compared with the z-AUC (median AUC) for the control cell lines (HMC3, HCMEC/D3, ReN)—AUC of DIPG cells after exposure to ONC201. Pearson linear regression, accounting for replicates, was used to determine ONC201 sensitivity correlation for DRD2, CLPP, SDHA, and the ratio of SDHA to CLPP (SDHA/CLPP). ns, not significant; n = 12. C,CLPP RNA expression from RNA-seq data publicly available through St Jude's PeCan database, normalized to FPKM (fragments per kilobase of transcript per million mapped reads). NBM (normal bone marrow CD34-positive hemopoietic stem cells/mononuclear cells), DIPG (diffuse intrinsic pontine glioma), non-BS-HGG (non-brainstem-high grade glioma, including not otherwise specified), LGG (low-grade glioma), MB (medulloblastoma), and BT other (brain tumor other—ependymoma, atypical teratoid rhabdoid tumor, choroid plexus carcinoma, cranio and CNS tumor not specified). Statistical significance determined via one-way ANOVA. D, Resazurin proliferation following ONC201 exposure (compared with untreated, 96 hours) of CRIPSR-Cas9–mediated knockdown of CLPP and DRD2 was performed in SU-DIPG-XIII (blue) and SU-DIPG-XXXVI (yellow). Values shown as mean ± SEM (n = 3). E, Western blot validation of successful knockdown of CLPP and DRD2 in SU-DIPG-XIII and SU-DIPG-XXXVI. *, P < 0.05; **, P < 0.01; ****, P < 0.0001.

Figure 3.

Quantitative proteomic profiling identifies increased PI3K/Akt signaling in resistant models. High-resolution quantitative proteomic profiling was conducted on SU-DIPG-VI, exposed to 5 μmol/L ONC201 for 24 hours. Cells were treated in low oxygen (5% O₂, 5% CO₂) and normoxic conditions (20% O₂, 5% CO₂) in biological triplicate. A and B, Major canonical pathways (A) and activated upstream regulators (determined by IPA; B) of proteins significantly altered following 5 μmol/L ONC201, regardless of oxygen tension (Student t test, P < 0.05, n = 6). C, Expression changes of proteins were calculated as log₂-fold change and grouped by mitochondrial proteins, transcription factors, and protein markers of apoptosis. Student t test of average change; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001. D, Orthogonal validation of mitochondrial changes, such as decreased SDHA, was analyzed in DIPG cell lines (HSJD-DIPG-007, SU-DIPG-XXXVI, SU-DIPG-VI, SU-DIPG-XIII, and SU-DIPG-XVII) via Western blot, exposed to 5 μmol/L ONC201 for up to 48 hours. E, Network of proteins from upregulated PI3K/Akt signaling predicted by IPA were integrated in Cytoscape StringApp. Predicted increase (orange) and predicted decrease (blue) functional networks indicated with sharp and dark lines linking proteins to indicate a higher confidence interval. Protein expression changes mapped as log₂-fold change of ONC201/untreated calculated using the right-tailed Fisher exact test with the smaller the P value, the more likely the association between proteins not to be a random event (P < 0.05). F and G, High-resolution quantitative proteomic profiling was conducted on SU-DIPG-XXXVI, SU-DIPG-VI and SU-DIPG-XXXVI, exposed to 5 μmol/L ONC201 for 24 hours. F, Heatmap and unbiased hierarchical clustering of protein expression values normalized using z-score of abundances in Perseus. G, Canonical pathways and predicted upstream regulators determined by IPA analysis of proteins altered following ONC201 exposure. Positive z-score value is predictive of pathway activation, whereas a negative z-score is predictive of inhibition. H, Orthogonal validation of protein-associated PI3K/Akt/mTOR signaling and the antioxidant-response element (ARE) axis were assessed in DIPG cell lines following ONC201 exposure.

Figure 4.

ONC201 in combination with paxalisib is synergistic across DIPG models. A, SU-DIPG-VI was treated with 5 μmol/L ONC201 for 48 hours, 20 mmol/L NAC for 24 hours, and 1 mmol/L H₂O₂ for 1 hour, and protein changes downstream PI3K/Akt and reductase signaling were validated by Western blot. B, Western blot analysis of PI3K/Akt, Erk, and antioxidant response element (ARE) signaling in SU-DIPG-XVII treated with 5 μmol/L ONC201 (48 hours) and 1 μmol/L paxalisib (24 hours). C, SU-DIPG-VI was grown in soft agarose in colony formation for 2 weeks treated with 0.5 μmol/L ONC201, 100 nmol/L paxalisib, and the combination. The number of colonies was then quantified using ImageJ. Assay was performed in biological triplicate with representative images shown. One-way ANOVA; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; values shown as mean ± SEM. D and E, DIPG cells SU-DIPG-VI, SU-DIPG-XIII, and SU-DIPG-XVII were passaged, grown in low oxygen (5% O₂, 5% CO₂) or atmospheric oxygen (20% O₂, 5% CO₂) conditions for a week, and then proliferation assays were performed using ONC201, paxalisib, or both for 96 hours (n = 3). Synergy was determined using Chou–Talalay via CompuSyn (D) or Bliss synergy (E) analysis where a combination index (CI), where <1 (dotted line) demonstrates a synergistic effect and Bliss score >10 represents a strong synergism. F–H, Parental wt-TP53 HSJD-DIPG-007 and HSJD-DIPG-007 cell lines transduced with a nontargeting control (NTC) gRNA, harboring KO or KD of TP53 were subjected to analysis following ONC201 treatment alone or in combination with paxalisib. F, Western blot confirmation of mitochondrial marker, SDHA, PI3K/Akt, Erk, and ARE signaling to ONC201 (5 μmol/L, 48 hours). G and H, Cells were treated with increasing concentrations of ONC201, paxalisib, or both for 96 hours, in biological triplicate. Synergy was determined using Chou–Talalay (G) or Bliss synergy (H) analysis. Chou–Talalayand Bliss synergy graphs are reported as mean ± SD.

Figure 5.

ONC201 in combination with paxalisib is a synergistic drug combination in DIPG xenograft models. A, SU-DIPG-VI/Luc and HSJD-DIPG-007 cells were injected into the brainstem of NSG mice. Treatment was started at 4 or 3 weeks, respectively, from xenograft date. ONC201 and paxalisib were administered by oral gavage. Xenografts were sacrificed for pharmacodynamics and survival was tracked where they were culled at ethical endpoints. B and C, Survival curve analysis of days after treatment start at animal sacrifice, with significance determined by survival curve comparison for SU-DIPG-VI/Luc (B) and HSJD-DIPG-007 (C). Shading indicates treatment duration. Log-rank (Mantel–Cox) test. D, Tumor tissue from SU-DIPG-VI/Luc xenografts sacrificed at 2 weeks following start of treatment analyzed by Western blot. E, SU-DIPG-VI/Luc in vitro cells exposed to 5 μmol/L ONC201 for 0, 24, 48, and 72 hours compared with in vivo SU-DIPG-VI/Luc tissue collected from the prefrontal cortex (PFC) and brainstem (BS), treated with ONC201. E, Tumor tissue was resected from HSJD-DIPG-007 xenografts following 4 weeks of treatment and analyzed by IHC. F, Sections were stained for H3K27M, Ki67, and SDHA (representative images are presented). Scale bars, 2 mm, 200 or 50 μm. G, IHC images quantified via ImageJ (measured in technical triplicate, across biological replicates, n = 3). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, not significant. b.i.d., twice daily; q.w., once a week; t.i.w., three times a week.

Figure 6.

ONC201 in combination with paxalisib drives tumor regression and increased survival in DIPG case studies. A, Six-year-old H3.1K27M, PIK3R1, ACVR1 mutant patient with DIPG underwent biopsy soon after diagnosis and received 54 Gy radiotherapy over 30 fractions. MRI was performed six weeks after the completion of radiotherapy, and compassionate access was granted for the use of paxalisib to target PIK3R1 mutations. Family of the patient sourced German ONC201 and started concurrently with paxalisib. B, Tumor size at diagnosis, following radiotherapy and throughout treatment. C, T2 and T1 after contrast MR axial scans at patient diagnosis, tumor area = 1,554 mm². D, Following radiotherapy, tumor area decreased by 38.1% to 962 mm² compared with diagnosis. E, MRI showed that tumor area was stable following radiotherapy = 1,156 mm², 20.2% progression. Following this scan, the ONC201 (15 mg/kg once a week) and paxalisib (27 mg/m² daily) combination was started. F, MRI following 8 weeks on the combination tumor area was stable (1,224 mm²), 6% increase. G, Tumor regression was seen after 20 weeks on the drug combination. Tumor area = 464 mm²; tumor reduction by 62% compared with the last scan. H, Most recent MRI. Tumor area = 306 mm²; total tumor area reduction compared with diagnosis = 80%. I, The patient continues to remain on the combination 22 months following diagnosis. J–P, Sixteen-year-old H3.3K27M, TP53, PIK3CA mutant patient with DIPG received 54 Gy over 30 fractions. Patient enrolled in the ONC201 monotherapy trial NCT03416530 and experienced stable disease for 2 months. Following radiological and clinical progression, the patient received panobinostat (45 mg daily three times a week) with ONC201 (625 mg once a week). Further progression was seen in the subsequent MRI, where the patient then received reirradiation. The patient immediately commenced ONC201 and paxalisib, both on compassionate grounds. K, Tumor area measured throughout treatment. L, T2 and T1 after contrast MR axial scans at patient diagnosis. Tumor area = 977.8 mm². M, MRI following first progression. Tumor area = 1,303.3 mm². N, Patient received panobinostat in combination with ONC201; MRI image following the combination. Tumor area = 1,814 mm². Following, this patient received reirradiation and ONC201 (15 mg/kg once a week) and paxalisib (27 mg/m² daily) O, Tumor regression was seen 8 weeks after re-RT, while receiving ONC201 in combination with paxalisib. Tumor area = 1,322.6 mm². P, Tumor regression was again seen after 20 weeks on the combination; tumor area = 1,209 mm², 20 months after diagnosis. Patient continued to receive ONC201 and paxalisib for the next 3 months and then contracted pneumonia and passed away 24 months from diagnosis.