Evaluation of a procaspase-3 activator with hydroxyurea or temozolomide against high-grade meningioma in cell culture and canine cancer patients

Abstract

Background

High-grade meningioma is an aggressive type of brain cancer that is often recalcitrant to surgery and radiotherapy, leading to poor overall survival. Currently, there are no FDA-approved drugs for meningioma, highlighting the need for new therapeutic options, but development is challenging due to the lack of predictive preclinical models.

Methods

To leverage the known overexpression of procaspase-3 in meningioma, PAC-1, a blood-brain barrier penetrant procaspase-3 activator, was evaluated for its ability to induce apoptosis in meningioma cells. To enhance the effects of PAC-1, combinations with either hydroxyurea or temozolomide were explored in cell culture. Both combinations were further investigated in small groups of canine meningioma patients and assessed by MRI, and the novel apoptosis tracer, [¹⁸F]C-SNAT4, was evaluated in patients treated with PAC-1 + HU.

Results

In meningioma cell lines in culture, PAC-1 + HU are synergistic while PAC-1 + TMZ show additive-to-synergistic effects. In canine meningioma patients, PAC-1 + HU led to stabilization of disease and no change in apoptosis within the tumor, whereas PAC-1 + TMZ reduced tumor burden in all three canine patients treated.

Conclusions

Our results suggest PAC-1 + TMZ as a potentially efficacious combination for the treatment of human meningioma, and also demonstrate the utility of including pet dogs with meningioma as a means to assess anticancer strategies for this common brain tumor.

Key Points.

• Two drug combinations are evaluated in canine meningioma patients.

• Novel apoptosis tracer, [¹⁸F]C-SNAT4, used to monitor drug-induced apoptosis in vivo.

• PAC-1 combination with temozolomide reduces tumor burden in canines with meningioma.

Importance of the Study.

This study suggests the potential of PAC-1 + TMZ for canine meningioma, supporting further evaluation in human meningioma patients. Additionally, the canine studies allowed simultaneous evaluation of PAC-1 and a second investigational agent, the novel apoptosis tracer [¹⁸F]C-SNAT4. This probe was safely administered in dogs and showed a lack of drug-induced apoptosis in canine meningiomas treated with PAC-1 + HU, correlating with the minimal changes in tumor burden as assessed by MRI. These results support the assessment of [¹⁸F]C-SNAT4 in human patients.

Arising from the membrane surrounding the brain and spinal cord, meningiomas are the most common intracranial neoplasm and account for approximately 37% of new brain cancer cases each year.^1,2 The majority of meningiomas are classified as benign (WHO grade I) and respond to surgical resection and radiotherapy.³ However, 20% of meningiomas are high grade (WHO grade II and III), characterized by a more aggressive, invasive phenotype and a higher mortality rate.³ Surgery and radiotherapy are ineffective against these tumors, resulting in a high rate of recurrence.³ There are currently no FDA-approved drugs for meningioma and while chemotherapy may be employed if the tumor is inoperable or recurrent, it is rarely efficacious.^1,3

The barren clinical landscape for meningioma has spurred efforts focused on the identification of targeted therapeutics,^1,3,4 however, these efforts are hampered by the apparent low abundance of targetable mutant oncogenes, and low mutational burden in these tumors.^1,3,5 The few potentially targetable mutations identified occur only in a small percentage of patients.^3,5 There is an ongoing clinical trial for therapeutics targeting some of these mutations (NCT02523014), but identification of a more broadly conserved target would be welcome and could be leveraged to treat a larger patient population.^1,3 One promising target is the elevated expression of procaspase-3 in meningioma.⁶ Procaspase-3 is the zymogen of caspase-3, an executioner caspase responsible for the cleavage of numerous proteins during intrinsic and extrinsic apoptosis.⁷ During activation procaspase-3 is proteolyzed to caspase-3, allowing propagation of caspase-3 activity.^7–9 A number of cancer types overexpress procaspase-3 as compared to matched normal tissues, providing a basis of selectivity for procaspase-3 targeted therapy for cancerous cells.^6,7,10,11 Procaspase-3 overexpression has been found in 70% of human and 92% of canine meningiomas as compared to normal brain tissue by immunohistochemical staining.⁶

PAC-1 is a small molecule that activates procaspase-3/-7 to caspase-3/-7 by chelating inhibitory zinc ions.^7,8,12–14 This mechanism has been supported by a variety of reports with selective caspase inhibitors¹⁵ and substrates,¹³ genomic knockdown of CASP3/7¹³, and other experiments.^{7,8,12–14,16} PAC-1 induces apoptosis in a variety of cancer cell lines overexpressing procaspase-3 but not in normal cells.^{10,12–14,17–19} The selectivity found in cell culture has translated to good tolerability in a Phase I clinical trial of PAC-1 for late-stage cancer patients (NCT02355535).²⁰ As PAC-1 is also blood-brain barrier penetrant,¹⁶ active against a variety of brain cancer cell lines,^6,10 and has activity in both canine glioma¹⁰ and human brain cancer patients,²¹ current evidence supports evaluation of PAC-1 for meningioma, particularly for use in high-grade tumors that have invaded the brain.

Herein, we evaluate PAC-1 as a single agent for meningioma, as well as in combination with HU and TMZ. Single-agent PAC-1 is active against meningioma cell lines in culture, demonstrating activation of procaspase-3 as a strategy to selectively induce apoptosis in meningioma. The combination of PAC-1 and HU has synergistic effects in cell culture while PAC-1 + TMZ exhibits additive-to-synergistic effects. Given the uncertain predictive power of meningioma preclinical assessments, both combinations were evaluated in pet dogs with naturally occurring meningiomas. In these studies, PAC-1 + TMZ showed more activity than PAC-1 + HU, reducing tumor burden in all three patients. The efficacy of PAC-1 combinations against meningioma supports further evaluation for the treatment of human meningioma patients.

Materials and Methods

Determination of IC₅₀ Values

Cells were seeded in a 96-well plate in 99 µl of media (IOMM-Lee 2,500 cells/well, CH157-MN 3,000 cells/well, SF6717 4,000 cells/well). Cells were incubated overnight. Then 1 µl of DMSO stocks of the compound were added to each well. Plates were incubated for 72 hours. Biomass was assessed by sulforhodamine B (SRB) assay (see Supplementary Materials).

6 × 11 Matrices

Cells were seeded in a 96-well plate in 50 µl of media at the same densities used for determining IC₅₀ values and incubated overnight. DMSO stocks of the compound were mixed with media to desired concentrations. Then 50 µl of the media and compound mix was added to the cells. Cells were incubated for 72 hours. Biomass was assessed by SRB. Loewe and Bliss calculations were done using Combenefit software.²²

Assessment of Apoptosis by Flow Cytometry

Cells were seeded in 6-well plates and allowed to attach overnight (IOMM-Lee 100 000 cells/well, SF6717 and CH157-MN 200 000 cells/well). Cells were treated with compounds and incubated 72 hours. For Q-VD-OPh protection, cells were co-treated with Q-VD-OPh (25 µM final concentration) and drug combinations and incubated for 48 hours. For PAC-1 single-agent, cells were pre-treated with Q-VD-OPh for 2 hours prior to the addition of PAC-1. Cells were then harvested and stained with Annexin V and propidium iodide for flow cytometry analysis (see Supplementary Materials). Percent death for Annexin V and propidium iodide staining is calculated by subtracting percent population in lower left quadrant from 100%.

Western Blot

For PARP-1 Western blot, IOMM-Lee was seeded in a 6-well plate (300 000 cells/well) and allowed to attach overnight. Cells were treated with compounds and incubated for 48 hours. Cells were harvested, collected with the media and debris, and pelleted. To Western blot for MGMT, a pellet of 1 million cells was collected for each cell line and stored at –80°C until further use. Western blot was run as described in the Supplementary Materials.

Caspase-3/-7 Activity Assay

IOMM-Lee was seeded in 96-well plates (10 000 cells/well at t = 0 h and 8000 cells/well at endpoint) in 100 µl of media and allowed to attach overnight. Cells were treated with indicated concentrations of DMSO, PAC-1, HU, TMZ, or raptinal (positive control, 10 µM) by first combining the compound with media to the appropriate concentrations and then adding 50 µl of this mix to the cells. Cells were incubated for 0, or 48 hours (PAC-1, and PAC-1 + TMZ), and then caspase-3/-7 activity was assessed using Ac-DEVD-AFC substrate (see Supplementary Materials).

[¹⁸F]C-SNAT4 Biodistribution in Healthy Beagle Dogs and Canine Meningioma Patients

[¹⁸F]C-SNAT4 physiologic whole-body biodistribution was evaluated via fused Positron Emission Tomography/Computed Tomography (PET/CT) imaging in three healthy purpose-bred laboratory dogs maintained for noninvasive studies within the NIH Division of Veterinary Resources. On the day of PET/CT imaging, dogs were placed under general anesthesia with a combination of butorphanol and midazolam premedication, propofol induction, and isoflurane mixed with oxygen maintenance. PET data were collected beginning simultaneously with intravenous injection of the tracer to establish the kinetics of uptake within the brain (Supplementary Table S1). After 45 min of kinetic tracer uptake data collection within the head, whole-body CT, and static PET images were collected. Dogs also received an IV dose of 2.2 ml/kg body weight of iohexol CT contrast material (Omnipaque 240 mg/ml, GE Healthcare).

Assessment of Efficacy in Pet Dogs With Meningioma

All canine procedures were approved by the University of Illinois, Virginia Tech, or NIH Institutional Animal Care and Use Committee. Pet owners provided written informed consent prior to beginning the study. Canines with histologically confirmed or radiologically diagnosed meningiomas were included in this study. Dogs to be treated with PAC-1 + HU started daily treatment with prednisone (1 mg/kg, P.O.) 2 weeks prior to the initial assessment of tumor burden and continued throughout treatment. Patients were treated with the combination of PAC-1 (10 mg/kg, daily, P.O.) and HU (20 mg/kg, daily, P.O.) for 28 days. Tumor burden was assessed by MRI on days 0 and 28. Apoptotic cell death within the tumors was also assessed on days 0 and 28 via injection of [¹⁸F]C-SNAT4 and PET/CT imaging following the same anesthetic protocol as described in the Supplementary Materials. Dogs also received an IV dose of gadolinium contrast material or Magnevist (gadopentetate dimeglumine, Bayer Pharmaceuticals) at a dose of 0.1 mmol/kg. Dogs treated with PAC-1 + TMZ received 1 mg/kg of prednisone (P.O.) daily, 12.5 mg/kg PAC-1 (P.O.) on days 1–21, and 100 mg/m² TMZ (P.O.) on days 8–12 for two 28-day cycles. Tumor burden was assessed by MRI on days 0 and 56.

Results

Single-Agent PAC-1 Induces Apoptosis in Meningioma Cell Lines

There are few well-characterized, established high-grade meningioma cell lines,⁴ and herein we utilize IOMM-Lee, SF6717, and CH157-MN. IOMM-Lee is a grade III meningioma with a mutation in TERT and is well-characterized.^4,23 SF6717 is a grade II meningioma but is not well-characterized and there is little information on its genetic profile.²⁴ The classification of CH157-MN is unclear, however, it is considered a high-grade meningioma.²³ CH157-MN has mutations in NF2, a frequently inactivated tumor suppressor in meningioma, and TERT.²³ The cytotoxicity of PAC-1 against these cell lines was determined with PAC-1 having an IC₅₀ value of less than 4 µM against all three cell lines as determined by sulforhodamine B (SRB) (Figure 1A, Supplementary Figure S1A). The similar IC₅₀ values across all three cell lines correlate with their similar levels of procaspase-3 expression (Supplementary Figure S1B). Growth curves were also established for each cell line for further comparison between cell lines (Supplementary Figure S1C). Importantly, in the Phase 1 clinical study, the average PAC-1 trough level (C_min) in human patients is 4.1 µM at a 750 mg once daily dose, defining the clinically relevant concentration of PAC-1 anticancer effects.^21,25 Annexin V and propidium iodide (PI) staining of meningioma cells treated with PAC-1 indicate an apoptotic mode of cell death, consistent with the established mechanism of PAC-1 (Figure 1B, C). Western blot analysis of cleaved PARP-1, a known marker of apoptosis, and inhibition of cell death by pre-treatment with the pan-caspase inhibitor Q-VD-OPh further suggest PAC-1 induces apoptosis in these cell lines (Figure 1D, E). Finally, the caspase-3/-7 activity of lysates from meningioma cells treated with PAC-1 shows a dose-dependent increase, indicating PAC-1 acts in a caspase-3/-7-dependent manner (Figure 1F).

Fig. 1.

PAC-1 induces apoptosis in meningioma cell lines. (A) Dose-response curves of PAC-1 in three meningioma cell lines as assessed by the sulforhodamine B assay after 72 hours of treatment. IC₅₀ values: IOMM-Lee 2.3 ± 0.0 µM, SF6717 3.1 ± 0.3 µM, CH157-MN 2.3 ± 0.2 µM. (B) Percent cell death induced by PAC-1 after 72 hours as assessed by Annexin V-FITC and propidium iodide staining. (C) Representative flow plots from (B). (D) Western blot analysis for PARP-1 cleavage as a marker of apoptosis in IOMM-Lee cells after 48-hour treatment with PAC-1 (15 µg protein loading). (E) Percent death of IOMM-Lee cells as assessed by Annexin V and propidium iodide staining after 48-hour treatment with PAC-1, with or without 2-hour pre-treatment with Q-VD-OPh (25 µM). (F) Executioner caspase activity at 48 hours in IOMM-Lee. Percent caspase-3/-7 activity is normalized to minimum and maximum observed within the assay with 10 µM raptinal as a positive control. All experiments are representative of at least three independent biological experiments, error is SEM, **P < 0.01, 2-tailed t test.

The Combination of PAC-1 and HU Induces Synergistic Death of Meningioma Cells in Culture

In the search for more efficacious therapeutics, past clinical trials for meningioma have focused on single-agent efficacy.^1,3,26 The recommended drugs for the treatment of inoperable or refractory meningiomas are all single agents, however, none provide significant benefit to patients.^1,3 Consequently, drug combinations have been hypothesized as an important avenue of exploration for meningioma,¹ and many recent clinical trials for meningioma assess drug combinations rather than single-agent treatment regimens.^1,3,27,28 PAC-1 synergizes with a number of therapeutics including doxorubicin,¹⁷ paclitaxel,¹⁸ vemurafenib,¹⁹ temozolomide¹⁰ and others.^17,29 Therefore, in an effort to enhance its anticancer effects, PAC-1 was combined with drugs having previously demonstrated single-agent activity in meningioma models.

Hydroxyurea (HU) inhibits DNA synthesis via targeting of ribonucleotide reductase,¹ and is a recommended drug for recurrent meningiomas; however, the efficacy observed to date has been insufficient for FDA approval.^1,3 HU is FDA-approved to treat some solid tumors, leukemia, and sickle cell anemia.³⁰ HU is a well-tolerated therapeutic, exemplified by many sickle cell anemia patients taking HU daily for multiple years.^1,30 Adults may safely take 20–35 mg/kg orally of HU daily, with Cmax from 200–500 µM.^30,31 Given HU’s tolerability and suggested activity in meningioma, PAC-1 + HU were explored in meningioma cells in culture.

The combination of PAC-1 and HU was evaluated in a matrix format and biomass assessed by the sulforhodamine B assay (SRB) in the three meningioma cell lines (Figure 2A). The degree of synergy was calculated by two different methods, Loewe additivity and Bliss independence³²; both models indicate the combination is synergistic in all three cell lines (Figure 2A).

Fig. 2.

The combination of PAC-1 and HU induces synergistic cell death in meningioma cell lines. (A) The combination was evaluated by the sulforhodamine B assay after 72 hours in the three meningioma cell lines. The degree of synergy was evaluated by two models (Loewe and Bliss). (B) The combination was assessed by Annexin V and propidium iodide staining after 72 hours and the degree of synergy calculated by Loewe and Bliss models. Data are the average of three independent biological experiments, error is SEM.

To further assess the interaction and elucidate the mode of synergistic cell death, meningioma cells treated with PAC-1 + HU were analyzed by Annexin V and PI staining in a matrix format (Figure 2B, Supplementary Figure S2). Loewe additivity and Bliss independence calculations were consistent with the synergy observed by SRB (Figure 2B). Annexin V and PI staining and Western blot analysis of PARP-1 cleavage are consistent with synergistic activity and apoptotic cell death (Figure 3A, B). Additionally, treatment with Q-VD-OPh inhibits cell death, consistent with apoptotic induction by the combination (Figure 3C). Treatment with PAC-1 + HU synergistically increased caspase-3/-7 activity compared to either single agent, further demonstrating the potential for the combination to induce high levels of apoptotic death in meningioma cells (Figure 3D).

Fig. 3.

The combination of PAC-1 and HU induces apoptosis in meningioma cell lines in culture. (A) Annexin V-FITC and propidium iodide staining of meningioma cell lines treated with PAC-1 and HU for 72 hours. (B) Western blot analysis for PARP-1 cleavage, an apoptosis marker, in IOMM-Lee cells after 48-hour treatment (15 µg protein loading). (C) Co-treatment of IOMM-Lee with 25 µM Q-VD-OPh protects from cell death induced by 48-hour treatment with 5 µM PAC-1 + 600 µM HU. (D) Executioner caspase activity in IOMM-Lee after 48-hour treatment with 5 µM PAC-1 and 250 µM HU. Percent caspase-3/-7 activity is normalized to minimum and maximum observed within the assay with 10 µM Raptinal as a positive control. (A,C,D) n ≥ 3 biological replicates, error is SEM, 2-tailed t test, ****P < 0.0001.

The Combination of PAC-1 and TMZ Shows Additive Activity in Cell Culture

Temozolomide (TMZ) is a DNA alkylating agent used to treat glioblastoma (GBM). The DNA methyltransferase MGMT removes the methyl lesions added by TMZ, promoting resistance.³³ TMZ has been evaluated in recurrent grade I meningioma patients in a prospective Phase II clinical trial, but there was no observable efficacy.²⁶ MGMT status of patients’ tumors in this trial was not reported; however, the majority of meningiomas are believed to express MGMT as determined by promoter methylation status,^34,35 suggesting TMZ could be ineffective at least in part due to an MGMT-mediated resistance mechanism. However, case studies have reported TMZ treatment reduced tumor burden in two recurrent meningioma patients.^36,37 While again MGMT status of these tumors was not reported, these cases indicate there is potential for TMZ to have activity in this cancer.^36,37 The combination of PAC-1 and TMZ has synergy in glioma,¹⁰ and PAC-1 + TMZ is currently being assessed in a clinical trial for GBM (NCT03332355).²¹ As TMZ appears to have some potential for activity in meningioma and prior data demonstrate the synergy of PAC-1 and TMZ in brain cancer,¹⁰ this combination was explored in meningioma cell lines.

The combination of PAC-1 and TMZ was evaluated in a matrix by SRB in all three meningioma cell lines, and Loewe additivity and Bliss independence calculations indicate overall additivity or synergistic activity (at some concentrations) across the matrix in all cell lines (Figure 4A). While the marked synergy of PAC-1 + TMZ observed in glioma¹⁰ is not fully recapitulated in meningioma cell lines in culture, the additive effects still increase overall cancer cell death.

Fig. 4.

The combination of PAC-1 and TMZ is additive-to-synergistic against meningioma cell lines in culture. (A) The combination was assessed by the sulforhodamine B assay after 72 hours in the three meningioma cell lines. The combination was analyzed using two different models (Loewe and Bliss). (B) The combination was assessed by Annexin V and propidium iodide staining after 72 hours, and synergy assessed by Loewe and Bliss models. Data is the average of at least three independent biological replicates, error is SEM.

To further investigate the additivity and assess the mode of cell death, meningioma cells were treated with the combination and stained with Annexin V and PI (Figure 4B, Supplementary Figure S3). All three cell lines demonstrate similar additive/synergistic trends to those observed by SRB (Figure 4B). Flow cytometry suggests an apoptotic mode of cell death, which is supported by Western blot analysis for cleaved PARP-1, and inhibition of death by Q-VD-OPh (Figure 5A–C). Additionally, assessment of caspase-3/-7 activity further demonstrates the combination indeed potently activates these executioner caspases (Figure 5D).

Fig. 5.

The combination of PAC-1 and TMZ induces apoptosis in meningioma cell lines. (A) Annexin V-FITC and propidium iodide staining of meningioma cell lines treated with PAC-1 and TMZ show induction of apoptosis after 72 hours. (B) Western blot analysis for apoptosis marker PARP-1 cleavage in IOMM-Lee cells after 48-hour treatment (15 µg protein loading). (C) Co-treatment of IOMM-Lee with 25 µM Q-VD-OPh protects from cell death induced by 48-hour treatment with 15 µM PAC-1 + 250 µM TMZ. (D) Executioner caspase activity in IOMM-Lee after 48-hour treatment with 5 µM PAC-1 and 5 µM TMZ. Percent caspase-3/-7 activity is normalized to minimum and maximum observed within the assay. Raptinal (10 µM) was used as a positive control. (E) Western blot of MGMT expression in meningioma cell lines (15 µg protein loading). (A,C,D) n ≥ 3 biological replicates, error is SEM, 2-tailed t test, *P < 0.05, **P < 0.01, ****P <0.0001.

Western blot analysis for MGMT in the three meningioma cell lines showed IOMM-Lee is MGMT negative, while SF6717 and CH157-MN are MGMT-positive (Figure 5E). MGMT status correlates with sensitivity to TMZ, with IOMM-Lee being the most sensitive (Figure 4). The additive-to-synergistic effects of the combination in the SF6717 and CH157-MN cell lines suggest the intriguing possibility that PAC-1 may sensitize MGMT-positive meningioma cells to TMZ to some degree.

Functional Apoptosis Imaging and Efficacy of PAC-1 Combinations in Canine Meningioma Patients

In translating these drug combinations from cell culture to an in vivo model, the lack of clinically relevant murine models of meningioma presents a challenge.⁴ To address this, the combinations were instead evaluated in canine meningioma patients. Canine cancer patients present a unique opportunity to evaluate experimental therapeutic strategies in a more clinically relevant setting, while potentially providing benefits for these companion animals.^38–40 Canines are larger and physiologically more similar to humans than laboratory mice.^38,40 Additionally, the heterogeneity of spontaneous tumors in canine patients mimics that seen in human cancer patients.^38,40 Canine meningiomas have been found to be histologically and genetically similar to human meningiomas, making them good models of the human disease.^40–42 Furthermore, high-grade meningiomas are more common in canines than humans (30–45% of all primary spontaneous CNS tumors in canines versus 13–25% in humans).⁴⁰

The pharmacokinetics,⁴³ tolerability,^43,44 and efficacy of PAC-1 have been well-studied in canine cancer patients in multiple settings.^10,17,44 The combination of PAC-1 and TMZ has also previously been shown to be safe and efficacious in dogs with glioma.¹⁰ As procaspase-3 is overexpressed in canine meningiomas,⁶ and canine meningiomas are fair representations of human meningiomas,^40–42 this comparative oncology model is valuable for preclinical assessment of drug combinations with PAC-1 for the treatment of human meningioma.

Early assessment of cell death within a tumor can have prognostic implications influencing subsequent treatment. [¹⁸F]C-SNAT4 is a novel apoptosis tracer that, once activated by caspase-3/-7 in apoptotic cells, undergoes an intramolecular cyclization followed by self-assembly into nanoaggregates (Supplementary Figure S4A).⁴⁵ This tracer allows serial tracking of apoptosis within a tumor via PET imaging.^45,46 The first-generation tracer, [¹⁸F]C-SNAT, has been directly compared favorably to established radiotracer imaging agents.⁴⁶ The treatment of pet dogs with naturally occurring tumors allows not only evaluation of the novel therapeutic, PAC-1, but also simultaneous development of [¹⁸F]C-SNAT4, which would not be possible in human patients.

Prior to the evaluation of [¹⁸F]C-SNAT4 in pet dogs with meningioma, the biodistribution, and safety of the tracer were investigated in three healthy beagle dogs. [¹⁸F]C-SNAT4 was safely administered to all three dogs with no observed side effects. Routine hematology and biochemical assessments approximately 6 months later showed no significant abnormalities in any of the three dogs. PET/CT imaging showed increased signal from the tracer in the kidneys and, to a lesser extent, the liver (Supplementary Figure S4B). In comparison, tracer uptake in the spleen, muscles, and brain was minimal (Supplementary Figure S4B). The low signal from [¹⁸F]C-SNAT4 in normal brain tissue suggested there would be the minimal background for assessment of a brain tumor.

To evaluate the feasibility and in vivo efficacy of PAC-1 + HU, as well as the safety and clinical utility of [¹⁸F]C-SNAT4 in tumor-bearing subjects, two canine patients were treated, and tumor burden and apoptotic death assessed by MRI and PET imaging respectively, pre- and post-treatment. Patients 1 and 2 were given prednisone (1 mg/kg, P.O.) daily for two weeks prior to the pre-treatment MRI and PET/CT scans. Then these patients began treatment with 10 mg/kg PAC-1 (P.O.) and 20 mg/kg HU (P.O.) daily for 28 days. Patients continued prednisone treatment throughout the 28 days (Figure 6A). While both patients successfully completed treatment with PAC-1 + HU without any intolerable adverse effects, neither dog had significant changes in tumor volume as assessed by MRI (Supplementary Table S2). The observed changes are within the range of expected variability compared to a small historical cohort of palliatively treated canine meningioma patients, where –6% to +13% changes in tumor volume were observed after 30 days.⁴⁷ [¹⁸F]C-SNAT4 was safely administered to both patients with no observed side effects. PET imaging showed uptake of the tracer in normal brain tissue was minimal in patients 1 and 2, congruent with biodistribution findings in normal beagle dogs, whereas increased uptake was observed within the tumors (Figure 6B, C). However, the tracer showed minimal changes in apoptosis in the tumors pre- and post-treatment (Figure 6C). These results are consistent with the minimal changes in tumor volume as assessed by MRI (Figure 6D).

Fig. 6.

Schematics for canine treatment regimens and canine tumor assessment for (A-D) PAC-1 + HU and for (E, F) PAC-1 + TMZ. (A) Schematic of treatment regimen for PAC-1 + HU. (B) Pre-treatment MRI, and pre- and post-treatment PET imaging of [¹⁸F]C-SNAT4 in patient 1. (C) Uptake of [¹⁸F]C-SNAT4 into normal brain tissue and into the brain tumors in patients 1 and 2 by PET/CT imaging. Ctrls 1, 2, and 3 are healthy beagle dogs. Pre 1 and Post 1 are pre- and post-treatment measurements for patient 1. Pre 2 and Post 2 are pre- and post-treaetment measurements for patient 2. SUV is standardized uptake value. (D) Quantification of tumor volume by serial MRI in patients 1 and 2, pre- and post-treatment. (E) Schematic of treatment regimen for PAC-1 + TMZ. (F) Pre- and post-treatment MRIs for patient 3 treated with PAC-1 + TMZ showed a 47% reduction in tumor burden. Tumors have been false-colored green.

As both [¹⁸F]C-SNAT4 and MRI confirmed a lack of efficacy with PAC-1 + HU, the combination of PAC-1 and TMZ was evaluated in three different canine meningioma patients for two 28-day cycles instead of one. Patients 3–5 were treated with prednisone (1 mg/kg, P.O.) daily, 12.5 mg/kg PAC-1 (P.O.) on days 1–21, and 100 mg/m² TMZ (P.O.) on days 8–12, for two 28-day cycles (Figure 6E). Tumor volume was assessed by MRI on days 0 and 56 (Figure 6E). As these patients were enrolled at a different veterinary oncology clinic than patients 1 and 2, [¹⁸F]C-SNAT4 imaging could not be performed. All patients completed treatment with no intolerable adverse effects and achieved some reduction in tumor burden. Patient 3 achieved the greatest reduction in tumor burden at 47% (Figure 6F, Supplementary Table S2) while patients 4 and 5 had 18% and 8% reductions, respectively (Supplementary Figure S5, Supplementary Table S2). All three patients had reductions greater than would be expected in canine patients treated palliatively (Supplementary Table S2), suggesting potential for the combination of PAC-1 + TMZ.

Discussion

Current therapeutic options for high-grade meningioma patients that do not respond to surgery or radiotherapy are severely limited, and as such, this cancer is associated with poor patient outcomes.³ Herein, single-agent PAC-1, and combined with HU or TMZ, was evaluated against meningioma cell lines in culture. Loewe additivity and Bliss independence calculations indicated PAC-1 + HU was the more synergistic combination in cell culture, and PAC-1 + TMZ was more efficacious in the treatment of a small number of canine meningioma patients. Evaluation in canine patients also allowed simultaneous assessment of novel apoptosis tracer, [¹⁸F]C-SNAT4. The lack of apoptosis upon treatment with PAC-1 + HU, as indicated by [¹⁸F]C-SNAT4, correlated with the lack of change in tumor burden assessed by MRI. These results support the evaluation of [¹⁸F]C-SNAT4 in human patients. A clinical trial to assess the biodistribution and safety of [¹⁸F]C-SNAT4 in human patients has already begun (NCT04017819).

Prior literature shows data from cell culture or murine meningioma models often fails to predict the results of subsequent clinical trials.^28,48–50 For example, Ragel et al. demonstrated the combination of HU and verapamil induced additive cell death in cell culture using IOMM-Lee and benign meningioma primary tumor samples.⁴⁸ Their results were further supported by a mouse xenograft model with IOMM-Lee.⁴⁸ However, a Phase I/II clinical trial evaluating the combination in meningioma patients whose tumors did not respond to surgery or radiation found there was no therapeutic benefit.²⁸ The combination of octreotide and everolimus also failed in a Phase II clinical trial despite promising data in cultured primary cells from patient tumors.^49,50 These results suggest the integration of canine meningioma patients early in the drug development pipeline for meningioma could be a valuable supplement to cell culture data and minimize late-stage clinical failure. Naturally occurring canine meningiomas are similar to human disease, with consistent genetic and histologic aspects, and are a better reflection of the heterogeneity observed in human tumors.^40–42 The combination of PAC-1 and HU, while promising in cell culture, showed no significant benefit for two canine patients, whereas the combination of PAC-1 and TMZ showed signs of anticancer activity in the three canine meningioma patients it was administered to.

While the canine efficacy data reported in this study suggests the potential of PAC-1 + TMZ, there are limitations to these studies. Current conclusions are based on a small patient sample (enrolled at different veterinary oncology clinics) for each drug combination, with variable dosing strategies employed. An ideal study would involve larger patient populations and standardized dosing or dose finding cohorts to enhance therapeutic efficacy. Without these explorations, it is difficult to compare directly, but initial evidence in this small study suggests PAC-1 + TMZ has efficacy against canine meningioma. Importantly, the combination was well-tolerated in these veterinary cancer patients, as previously reported in canines with glioma.¹⁰ An additional consideration in the use of TMZ is the presence of MGMT, which can result in resistance to TMZ for GBM patients.³³ Unfortunately, the MGMT status of the three canine patients herein could not be assessed, but cell culture studies of PAC-1 + TMZ in meningioma cell lines do suggest the intriguing possibility that this combination could be effective even in the presence of MGMT. There are no prior reports on MGMT status in canine meningiomas, however, MGMT expression is extremely common in human meningiomas.^34,35 Therefore, further study of the effects of MGMT on the response to the combination of PAC-1 and TMZ in meningioma is necessary. However, the promising results in the three canine patients will help guide further research on the combination of PAC-1 and TMZ for the treatment of human high-grade meningioma patients.

Funding

This work was supported by the NCI (R01CA120439) and the Canine Health Foundation (02321). This work was supported (AKL and SH) by the Intramural Program of the National Cancer Institute, NIH (Z01-BC006161). RCB was a member of the NIH Chemistry-Biology Interface Training Grant (T32-GM136629). The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.

Conflict of interest statement. The University of Illinois has filed patents on PAC-1 on which P.J.H, T.M.F, G.J.R., and R.C.B. are co-inventors. The University of Illinois has licensed the patents on PAC-1 to Vanquish Oncology, Inc, and P.J.H. and T.M.F. serve as consultants for Vanquish Oncology, Inc. Stanford University has filed a patent on [¹⁸F]C-SNAT4 on which Z.C. and J.R. are co-inventors.

Authorship statement. P.J.H., E.J.T., A.K.L., and T.M.F. project design. E.J.T., R.C.B., M.R.B., X.Z., F.B., A.K.L., S.K.J., Z.C., J.H.R., G.B.D., and T.M.F. experimentation and data collection. E.J.T., T.M.F, and P.J.H. writing the manuscript. E.J.T., R.C.B., M.R.B., X.Z. A.K.L., S.K.J., J.H.R., G.B.D., J.R., G.J.R., T.M.F., and P.J.H. data analysis and manuscript editing.