Trotabresib, an oral potent bromodomain and extraterminal inhibitor, in patients with high-grade gliomas: A phase I, “window-of-opportunity” study

Abstract

Background

The bromodomain and extraterminal protein (BET) inhibitor trotabresib has demonstrated antitumor activity in patients with advanced solid tumors, including high-grade gliomas. CC-90010-GBM-001 (NCT04047303) is a phase I study investigating the pharmacokinetics, pharmacodynamics, and CNS penetration of trotabresib in patients with recurrent high-grade gliomas scheduled for salvage resection.

Methods

Patients received trotabresib 30 mg/day on days 1–4 before surgery, followed by maintenance trotabresib 45 mg/day 4 days on/24 days off after surgery. Primary endpoints were plasma pharmacokinetics and trotabresib concentrations in resected tissue. Secondary and exploratory endpoints included safety, pharmacodynamics, and antitumor activity.

Results

Twenty patients received preoperative trotabresib and underwent resection with no delays or cancelations of surgery; 16 patients received maintenance trotabresib after recovery from surgery. Trotabresib plasma pharmacokinetics were consistent with previous data. Mean trotabresib brain tumor tissue:plasma ratio was 0.84 (estimated unbound partition coefficient [K_PUU] 0.37), and modulation of pharmacodynamic markers was observed in blood and brain tumor tissue. Trotabresib was well tolerated; the most frequent grade 3/4 treatment-related adverse event during maintenance treatment was thrombocytopenia (5/16 patients). Six-month progression-free survival was 12%. Two patients remain on treatment with stable disease at cycles 25 and 30.

Conclusions

Trotabresib penetrates the blood–brain-tumor barrier in patients with recurrent high-grade glioma and demonstrates target engagement in resected tumor tissue. Plasma pharmacokinetics, blood pharmacodynamics, and safety were comparable with previous results for trotabresib in patients with advanced solid tumors. Investigation of adjuvant trotabresib + temozolomide and concomitant trotabresib + temozolomide + radiotherapy in patients with newly diagnosed glioblastoma is ongoing (NCT04324840).

Key Points.

• Trotabresib showed notable brain tumor tissue penetration relative to plasma concentrations.

• Trotabresib demonstrated target engagement in resected brain tumor tissue.

• Safety data support further investigation of trotabresib in high-grade glioma.

Importance of the study.

Many therapies investigated in gliomas and other central nervous system malignancies have proved ineffective for various reasons, including limited brain tumor tissue penetration. The bromodomain and extraterminal (BET) inhibitor trotabresib has demonstrated single-agent activity in patients with high-grade gliomas, suggesting brain tumor tissue penetration. Here, we investigated trotabresib concentrations in brain tumor tissue following presurgical administration in patients with recurrent high-grade gliomas scheduled to undergo salvage resection. Trotabresib was detectable in resected tissue, and relative concentrations in resected brain tumor tissue and time-matched plasma samples, as well as the estimated unbound partition coefficient, suggested notable brain tumor tissue penetration. Pharmacodynamic analyses demonstrated target engagement in blood and brain tumor tissue. Safety was consistent with previous results in patients with advanced solid tumors. Based on the accumulated body of favorable safety, efficacy, pharmacokinetic, and pharmacodynamic data, a phase Ib/II clinical trial is currently investigating trotabresib in combination with adjuvant temozolomide and radiotherapy in patients with newly diagnosed glioblastoma (NCT04324840).

Gliomas are aggressive primary malignant brain tumors that account for the majority of malignant primary central nervous system (CNS) neoplasms in the US and Europe.^1,2 The overall annual incidence of gliomas is approximately 6 per 100 000 population in the United States and 5.4 per 100 000 population in Europe.^1,2 The most frequently occurring glioma is glioblastoma, making up 56% of glioma cases, with high-grade astrocytomas accounting for a further 20% of diagnoses.¹ The standard first-line treatment for high-grade gliomas is maximal surgical resection with adjuvant chemoradiotherapy; ^3,4 however, even complete resection of high-grade gliomas is unlikely to be curative due to tumor infiltration of surrounding brain tissue, and tumors eventually recur in all patients.⁵ The prognosis for patients with recurrent disease remains very poor, with most reports of median overall survival (OS) ranging from 3 to 12 months.^5–7

Effective treatment of gliomas is challenging for a number of reasons, with limited CNS penetration of therapeutics being one of the most important contributing factors.⁸ Diffusion of drugs through the blood–brain barrier (BBB) is restricted to lipid-soluble compounds with a molecular weight <400 Da,⁹ and expression of drug efflux transporters on vascular endothelial cells provides an additional barrier to tissue penetration of molecules that have the necessary properties for diffusion.¹⁰ Gliomas and other brain tumors have, however, been suggested to physically disrupt the BBB, resulting in localized increases in BBB permeability. Such areas of increased permeability can be visualized on MRI through the use of contrast agents that selectively enhance tumor tissue.¹¹ Importantly, disruption of the BBB shows considerable intratumoral heterogeneity, and most glioblastomas also have non-enhancing areas that represent tumor infiltration of brain tissue where the BBB is sufficiently intact to prevent contrast uptake.¹¹

Bromodomain and extraterminal (BET) proteins are epigenetic readers that regulate the expression of a variety of genes involved in cancer cell proliferation, survival, and oncogenic progression.^12–17 In particular, the BET protein BRD4 is overexpressed in gliomas relative to healthy tissue, supporting its investigation as a treatment target.¹⁸ BET inhibitors have shown antitumor activity in preclinical models of glioblastoma,^18–20 and several BET inhibitors have demonstrated BBB penetration in mice.^18–21 However, the relevance of animal data to clinical efficacy in glioblastoma is unclear; the BET inhibitor OTX015 showed antitumor activity and high penetration of brain tumor tissue in a murine orthotopic glioblastoma model, but was present at markedly lower concentrations in peritumoral and distant brain tissue.²⁰ A subsequent dose-finding study in patients with glioblastoma was stopped due to lack of efficacy,²² an outcome which may suggest poor brain tumor tissue penetration of OTX015 in humans.

Trotabresib (CC-90010, BMS-986378) is an oral, potent, and reversible small-molecule BET inhibitor that has been shown to reduce tumor growth in cell line and xenograft models of glioma and other malignancies (data on file). Trotabresib monotherapy 45 mg/day 4 days on/24 days off per 28-day cycle was well tolerated and demonstrated antitumor activity in heavily pretreated patients with advanced cancers, including high-grade gliomas, with prolonged progression-free survival (PFS) up to 3 years in some patients.^23,24 Importantly, trotabresib has physicochemical properties consistent with BBB penetration, including a molecular weight of 383 Da and high lipophilicity, and trotabresib concentration–time profiles in plasma and brain were comparable in animal models (data on file). Together, these data provide sufficient supporting evidence to evaluate trotabresib brain tumor tissue penetration in patients with high-grade gliomas, the first such study for a BET inhibitor.

Here, we present the results of CC-90010-GBM-001 (NCT04047303), a phase I “window-of-opportunity” study to evaluate brain tumor tissue penetration, plasma pharmacokinetics (PK), pharmacodynamics, safety, and antitumor activity of trotabresib in patients with recurrent high-grade gliomas who were scheduled to undergo salvage resection.

Materials and Methods

Study Design

CC-90010-GBM-001 enrolled patients with recurrent high-grade gliomas who were candidates for salvage resection. Patients received trotabresib monotherapy 30 mg/day on days 1–4 prior to salvage resection, which was planned for 6–24 h after the day 4 dose of study drug. The design of the study is shown in Supplementary Figure S1. A dose of 30 mg/day was selected for the preoperative treatment period to maintain platelet count above the minimum of 100 000/mL recommended for surgery, based on previous data showing that this dose level provides a balance of target engagement with a minimal impact on platelet count.²³

Upon recovery from surgery and ≥4 weeks after the first preoperative dose of trotabresib, patients initiated maintenance treatment with trotabresib monotherapy 45 mg/day 4 days on/24 days off in each 28-day cycle, which was the recommended phase II dose and schedule for trotabresib monotherapy identified in the CC-90010-ST-001 study.^23,24 Maintenance treatment continued until disease progression, unacceptable toxicity, or withdrawal of consent.

Patient Selection

Eligible patients were ≥18 years of age, had progressive World Health Organization grade II diffuse astrocytoma, grade III anaplastic astrocytoma, or grade IV glioblastoma in radiographically confirmed first or second recurrence, and were candidates for salvage resection. Disease progression was defined as either a >25% increase in the largest bidimensional product of enhancement or a new enhancing lesion for patients with glioblastoma, or a >25% increase in the largest bidimensional product of enhancement, a new enhancing lesion, or a >25% increase in a T2 or fluid-attenuated inversion recovery (FLAIR) non-enhancing lesion for patients with grade II or grade III astrocytoma. Patients must have had an Eastern Cooperative Oncology Group performance status (ECOG PS) of 0 or 1, previously completed a course of standard or hypofractionated radiotherapy at least 12 weeks prior to the first dose of study drug, and have archival formalin-fixed paraffin-embedded (FFPE) tumor tissue suitable for genetic testing.

Key exclusion criteria included anticancer therapy within 4 weeks (6 weeks for nitrosoureas) or 5 half-lives prior to starting trotabresib, major surgery within 4 weeks or minor surgery within 2 weeks prior to starting trotabresib, evidence of CNS hemorrhage on baseline MRI or CT scan, and requirement for increasing doses of corticosteroids to treat symptomatic cerebral edema within 7 days before study entry.

Additional exclusion criteria included mild or asymptomatic SARS-CoV-2 infection within 10 days, or severe/critical SARS-CoV-2 infection within 20 days prior to the first dose of study drug. Acute symptoms must have resolved without sequelae that would place the patient at increased risk while receiving study treatment, based on investigator assessment and consultation with the study medical monitor. Patients who had received a SARS-CoV2 vaccine within 14 days prior to starting study drug were also excluded.

All patients provided written informed consent; the study was conducted in compliance with the International Council on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use, Good Clinical Practice, and general ethical principles outlined in the Declaration of Helsinki, and was approved by the Institutional Review Board or Ethics Committee of each participating center.

Endpoints and Assessments

The primary endpoints of the study were to measure the concentration of trotabresib in resected brain tumor tissue at a single time point following oral dosing (including within contrast-enhancing and non-enhancing tumor, wherever possible) and to evaluate trotabresib plasma PK. Trotabresib concentrations in resected tumor were also compared with trotabresib concentrations in matched plasma samples collected at the same time point during surgery. Safety was assessed as a secondary endpoint. Exploratory endpoints included antitumor activity, PFS, trotabresib concentration in cerebrospinal fluid (CSF), and pharmacodynamic markers of target engagement.

Tumor tissue for PK and pharmacodynamic analysis and CSF samples for PK analysis were collected during surgery. Plasma samples for assessment of trotabresib PK and pharmacodynamics were obtained at multiple time points throughout the first treatment cycle, and at the time of tumor tissue and CSF sample collection. Trotabresib concentrations in plasma samples were determined by liquid chromatography with tandem mass spectrometry (High Assay Range). PK parameters were calculated by non-compartmental analysis using Phoenix Version 8.3.3.33. Plasma PK parameters assessed included peak plasma concentration (C_max), time to peak plasma concentration (t_max), area under the plasma concentration–time curve (AUC), and terminal half-life (t_½). Adverse events (AEs) were classified using Medical Dictionary for Regulatory Activities (MedDRA) version 24.0 and graded using the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTCAE) version 5.0.²⁵ AEs were reported separately for preoperative treatment with trotabresib 30 mg/day on days 1–4 (from initiation of preoperative treatment to initiation of postoperative maintenance) and postoperative maintenance treatment with trotabresib 45 mg/day 4 days on/24 days off (from the time of initiation of postoperative maintenance therapy onwards). Pharmacodynamic markers investigated included CCR1 expression in blood and HEXIM1 expression in blood, FFPE resected on-treatment tumor tissue and FFPE, archival tumor tissue. Gene expression was assessed using RNA-Seq. Details of blood and tumor PD biomarker analyses are listed in Supplementary Materials and Methods. Tumor assessments by MRI were performed within 24–72 h post-surgery to serve as a baseline for subsequent evaluations, and at the end of every other treatment cycle from cycle 2 onwards. Antitumor activity in terms of tumor response was assessed in patients with subtotal resection using the appropriate Response Assessment in Neuro-Oncology (RANO) criteria for high-grade or low-grade glioma.^26,27 No formal statistical power calculations were performed.

Results

Patients and Treatment

A total of 20 patients were enrolled between January 13, 2020 and February 5, 2021 and received preoperative trotabresib monotherapy 30 mg/day on days 1–4. Four patients did not enter the postoperative treatment period, of whom 2 died of surgical complications (described in detail in the safety section), 1 withdrew from the study, and 1 had progressive disease (PD). The remaining 16 patients initiated postoperative maintenance treatment with trotabresib monotherapy 45 mg/day 4 days on/24 days off. At the time of data cutoff (March 8, 2022), 14 patients had discontinued treatment due to PD and treatment was ongoing in 2 patients, both of whom remained on treatment as of the time of manuscript preparation (November 1, 2022).

Baseline patient demographics and characteristics are shown in Table 1. Median age was 47 years, 14 patients (70%) were male and 6 (30%) female, and 12 patients (60%) had an ECOG PS of 0. Nineteen patients (95%) had glioblastoma and 1 patient had progressive diffuse astrocytoma. The enrolled population was heavily pretreated; all patients had previously received radiotherapy and 16 (80%) had received 2 or more previous systemic therapies. One patient had previously received bevacizumab. MGMT promoter was methylated in 7 patients (35%) and unmethylated in 7 patients (35%). MGMT promoter methylation status was unknown in 6 patients (30%); some study sites did not determine this parameter prior to patient enrollment, as it was not an inclusion criterion for the study, and as all patients had previously progressed during prior treatment with temozolomide, no further treatment with temozolomide was planned. IDH mutation status was wild-type in 14 patients (70%), mutant in 5 patients (25%), and not otherwise specified in 1 patient (5%).

Table 1.

Patient Characteristic

Patient Characteristics	N = 20
Median age, years (range)	47 (33–75)
Sex, n (%)
Male	14 (70)
Female	6 (30)
ECOG PS, n (%)
0	12 (60)
1	8 (40)
Tumor type, n (%)
Glioblastoma	19 (95)
Diffuse astrocytoma	1 (5)
MGMT promoter methylation status, n (%)a
Methylated	7 (35)
Unmethylated	7 (35)
Not reported	6 (30)
IDH mutation status, n (%)a
Wild-type	14 (70)
Mutant	5 (25)
Not otherwise specified	1 (5)
Number of prior radiation therapies, n (%)
1	19 (95)
2	1 (5)
Number of prior systemic therapies, n (%)
1	4 (20)
2	15 (75)
>2	1 (5)
Median time since initial diagnosis, years (range)	1.2 (0.6–10.8)

^a MGMT promoter methylation status and IDH mutation status were determined using archival surgically resected tumor tissue and assessed per standard methodology at each institution.

Abbreviations: ECOG PS, Eastern Cooperative Oncology Group performance status; IDH, isocitrate dehydrogenase.

Pharmacokinetics (PK)

The plasma PK of trotabresib monotherapy 30 mg/day 4 days on/24 days off in CC-90010-GBM-001 was consistent with the PK of trotabresib monotherapy at this dose and schedule in the CC-90010-ST-001 study (Figure 1).²³ On cycle 1, day 1 (C1D1), geometric mean (GM) C_max was 392 ng/ml and GM AUC from 0 to 24 h (AUC_0–24) was 5083 ng·h/ml. On cycle 1, day 4 (C1D4), GM C_max was 720 ng/mL and GM AUC_0–24 was 11 250 ng·h/ml. Median t_max was 1.5 h on C1D1 and 1.9 h on C1D4, and mean t_1/2 was 46 h on C1D4.

Figure 1.

Trotabresib plasma PK. ^aPatients were excluded if missing plasma concentration data led to AUC extrapolation > 25% and an adjusted R² value < 0.8. Abbreviations: AI, accumulation index; AUC₂₄, area under the trotabresib concentration–time curve from 0 to 24 h; AUC_inf, area under the trotabresib concentration–time curve from 0 to infinity; C_max, peak trotabresib concentration; D, day; GCV, geometric coefficient of variation; GM, geometric mean; PK, pharmacokinetics; SD, standard deviation; t_½, trotabresib half-life; t_max, time to peak trotabresib concentration.

The median time from the day 4 dose of trotabresib to resection was 23 h (range, 4.6–31.3). Trotabresib was found to penetrate brain tumor tissue; GM trotabresib concentrations in plasma samples collected at the time of surgery and resected brain tumor tissue were 1.02 μM and 0.74 μM, with a mean tumor tissue:plasma ratio of 0.84 (Figure 2). GM free trotabresib concentration in brain tissue was calculated to be 0.028 μM. Based on the plasma protein binding and the tissue binding of trotabresib, the estimated unbound partition coefficient (K_PUU) value²⁸ for trotabresib is 0.37. GM trotabresib concentration in CSF was 0.14 μM, and the mean CSF:plasma ratio was 0.17.

Figure 2.

Trotabresib brain tumor tissue and CSF. penetration. (A) individual patient concentrations and (B) summary statistics. Abbreviations: CSF, cerebrospinal fluid; GCV, geometric coefficient of variation; GM, geometric mean; SD, standard deviation. Blue data points indicate individual patient trotabresib concentrations. Black data points indicate median trotabresib concentrations.

Pharmacodynamics

Trotabresib monotherapy showed encouraging target engagement based on modulation of two pharmacodynamic markers of BET inhibition, HEXIM1 and CCR1 expression.^29,30 BET inhibition has been shown to decrease CCR1 expression in blood.²⁹HEXIM1 expression, which is mechanistically linked to BET function, has been shown to increase in response to BET inhibition in blood and tumor tissue.³⁰ In CC-90010-GBM-001, blood levels of CCR1 mRNA decreased to 45.8% (SD ± 28.5) of baseline levels after the fourth preoperative dose of trotabresib (Figure 3A), consistent with data from patients treated with trotabresib on the same dose and schedule in the CC-90010-ST-001 study.²³ Blood levels of HEXIM1 mRNA were increased from baseline in all patients at 72–96 h after the first preoperative dose of trotabresib (Figure 3B). Analysis of HEXIM1 expression in blood samples obtained at the time of resection showed a relationship between the magnitude of increase in HEXIM1 expression in blood and trotabresib plasma concentration (Figure 3C). A total of 18 patients had sufficient archival tumor tissue available for RNAseq analysis. Comparison of HEXIM1 expression in FFPE resected brain tumor tissue with expression in archival tissue showed that HEXIM1 mRNA was increased in 15 of 18 patients (group P = .00093) (Figure 3D).

Figure 3.

Trotabresib pharmacodynamics: modulation of (A) mean ± SD CCR1 and (B) individual patient HEXIM1 mRNA in blood, (C) association between blood HEXIM1 mRNA level modulation and plasma trotabresib concentration at the time of surgery, and (D) modulation of HEXIM1 expression in brain tumor tissue. Abbreviation: SD, standard deviation.

Safety

The safety profile of trotabresib was consistent with previous reports of trotabresib monotherapy.²³ Any-grade treatment-related AEs (TRAEs) were reported in 12 patients (60%) following preoperative treatment with trotabresib 30 mg/day on days 1–4, of whom 2 (10%) had grade 3/4 TRAEs (ALT increased in 1 patient and lymphopenia in 1 patient; Figure 4A). Grade 1/2 thrombocytopenia was reported in 5 (25%) patients during the preoperative period. TRAEs were reported in all 16 (100%) patients who received postoperative maintenance trotabresib 45 mg/day 4 days on/24 days off, with grade 3/4 TRAEs reported in 6 (38%) patients (thrombocytopenia in 5 patients and asthenia in 1 patient; Figure 4B). No serious TRAEs were reported.

Figure 4.

TRAEs reported in ≥ 10% of patients at any grade or at grade 3/4 severity in ≥ 1 patient during the overall study by treatment period. (A) TRAEs reported from initiation of preoperative trotabresib 30 mg/day on days 1–4 until initiation of postoperative maintenance therapy (n = 20), and (B) TRAEs reported after initiation of postoperative maintenance trotabresib 45 mg/day 4 days on/24 days off (n = 16). Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; TRAE, treatment-related adverse event.

No patients had surgery delayed or canceled due to an AE, and no patients required dose modifications for the management of AEs during the preoperative period. During the postoperative adjuvant treatment period, 3 patients had dose interruptions due to AEs (thrombocytopenia in 2 patients and pancreatitis in 1 patient), and 1 patient had a dose reduction due to thrombocytopenia. No patients discontinued treatment due to AEs.

Two patients died of surgical complications unrelated to study drug (intracranial hemorrhage). The first patient, a 40-year-old man, had an uneventful surgery and initial recovery period. On study day 7, 2 days after surgery, a postoperative hematoma was identified that required reoperation. The hematoma was accompanied by a suspected TRAE of grade 1 thrombocytopenia, with a decrease in platelet count to ~100 000/ml. Reoperation was complicated by excessive bleeding and hydrocephalus requiring external ventricular drainage; the patient subsequently developed a gram-negative infection that was unresponsive to antibiotics on the third day after reoperation and died on study day 15. The second patient had a distant intracranial hemorrhage on study day 5; however, his platelet count and coagulation studies were normal. The patient’s recovery was complicated by confirmed SARS-CoV-2 infection on study day 9; the patient had not been vaccinated against SARS-CoV-2 and died on study day 12 following severe respiratory deterioration.

Efficacy

At the March 8, 2022 data cutoff, median PFS was 1.9 months (95% CI 1.4–3.4) and the 6-month PFS rate was 12% (95% CI 2.1–31.5). Subgroup analysis found that patients with IDH wild-type disease had a median PFS of 3.0 months (95% CI 1.4–3.6) and a 6-month PFS rate of 16.3% (95% CI 2.7–40.4), while patients with IDH-mutant disease had a median PFS of 1.2 months (95% CI 0.4–1.9) and a 6-month PFS rate of 0% (95% CI 0–0). Treatment was ongoing in 2 patients at the time of manuscript submission, both of whom had IDH wild-type disease and a methylated MGMT promoter. The patients remained on treatment with stable disease as of November 1, 2022 at cycles 25 and 30, representing approximately 23 and 27 months, respectively (Figure 5). Of the 17 patients who were evaluable for response, 7 (41%) had stable disease and 10 (59%) had PD per RANO criteria.

Figure 5.

Treatment duration as of November 1, 2022. Abbreviations: M, methylated; Mut, mutant; NA, not available; NOS, not otherwise specified; U, unmethylated; WT, wild-type. Dotted line indicates approximate time of surgery.

Discussion and Conclusions

Despite considerable research to identify new therapies for the treatment of glioma, translating antitumor activity in preclinical models of glioma to clinical efficacy has proved challenging, with poor penetration of brain tumor tissue by anticancer agents being a key limitation.³¹ The BET protein BRD4 is frequently overexpressed in gliomas and represents a potential target for anticancer therapy due to the broad-ranging role of BET proteins in tumor growth and survival.^12–18

The present “window-of-opportunity” study in patients undergoing surgical resection for recurrent high-grade gliomas was a rare opportunity to evaluate drug penetration and changes in pharmacodynamic markers of target engagement in brain tumor tissue in a clinical setting. Our results, which confirm the presence of pharmacologically active drug concentrations in resected glioma tissue, are the first such clinical data for a BET inhibitor. Combined with results from the CC-90010-ST-001 study, which showed that trotabresib monotherapy was well tolerated and had durable antitumor activity in patients with recurrent high-grade gliomas, these findings support further evaluation of trotabresib in this tumor type.^23,24

The plasma PK profile of trotabresib in the current study was consistent with that seen with the same dose and schedule in CC-90010-ST-001, with comparable C_max, t_max, and AUC.^23,24 These data confirm that the PK profile of trotabresib in patients with high-grade gliomas is comparable to that seen in patients with advanced solid tumors. Trotabresib was detected in resected brain tumor tissue, and relative concentrations in time-matched brain tumor tissue and plasma samples were suggestive of notable brain tumor tissue penetration, with a mean tissue:plasma ratio of 0.84. The median time from the last dose of trotabresib to surgery was 23 h (range, 4.6–31.3), and it may be reasonable to suggest that tissue concentrations of trotabresib would be higher at plasma t_max (1.9 h postdose on day 4) if the brain serves as a rapidly equilibrating compartment. Based on the plasma protein binding and the tissue binding of trotabresib, the estimated unbound partition coefficient (K_PUU) value²⁸ for trotabresib is 0.37. Taken together with the absolute concentration in tumor tissue, the relative concentrations in tissue and plasma, the observed biomarker activity in tumor tissue, and the potency of trotabresib, this K_PUU value supports that trotabresib has sufficient penetration into brain tumor tissue to drive pharmacological activity, potentially leading to clinical effects.

Selection of the 45 mg/day 4 days on/24 days off dose and schedule for trotabresib monotherapy in the CC-90010-ST-001 first-in-human study has been described previously.^23,24 Due to the long half-life of trotabresib (mean t_1/2 was 46 h in this present study), the 4 days of consecutive dosing leads to drug accumulation in circulation, which in turn is expected to drive drug penetration into tumor tissue as observed in this study, while the 24-day dosing holiday allows patients to recover. In addition, the 4 days on/24 days off schedule allows trotabresib dosing to be aligned with the standard dosing schedule for temozolomide (days 1–5 of each 28-day cycle).^32–35

Trotabresib was shown to modulate pharmacodynamic markers of target engagement in blood, based on expression of CCR1 and HEXIM1. Blood levels of CCR1 mRNA were decreased to 45.8% (SD ± 28.5) of baseline levels after the fourth dose of trotabresib 30 mg/day, consistent with results at the same dose level in the CC-90010-ST-001 study.^23,24 Reductions in blood CCR1 mRNA to ≤ 50% of baseline levels have previously been associated with plasma BET inhibitor concentrations and response in patients with lymphoma.³⁶ Trotabresib also increased blood levels of HEXIM1 mRNA, a marker of target engagement that has a mechanistic link with BET inhibition,³⁰ with increases observed in all 19 evaluable patients from 72 to 96 h. Analysis of HEXIM1 levels in blood samples obtained at the time of surgery showed that the magnitude of increase in HEXIM1 mRNA expression was associated with trotabresib concentration in time-matched plasma samples. Comparison of HEXIM1 expression in FFPE on-treatment resected tissue versus archival tumor tissue showed increased HEXIM1 expression in 15 of 18 samples, indicating that trotabresib reaches pharmacologically active tumor tissue concentrations after 4 days of treatment at the 30 mg/day dose level. Although CCR1 mRNA levels were measured in brain tumor tissue, CCR1 expression has not been validated as a marker of BET inhibitor target engagement in tumor tissue. CCR1 mRNA levels in tumor tissue were found to be much lower than those of HEXIM1 mRNA, and therefore, interpretation of the effects of trotabresib on modulation of CCR1 expression in brain tumor tissue was not reliable.

The safety profile of trotabresib monotherapy was also comparable to that seen in the CC-90010-ST-001 study,^23,24 with hematological and gastrointestinal TRAEs the most frequently reported. The majority of TRAEs were mild or moderate in severity, and most individual TRAEs were reported during maintenance treatment with trotabresib 45 mg/day 4 days on/24 days off. Grade 3/4 TRAEs were reported in 2 of 20 (10%) patients following preoperative treatment (ALT increase in 1 patient and lymphopenia in 1 patient), and 6 of 16 (38%) patients during maintenance treatment (thrombocytopenia in 5 patients and asthenia in 1 patient). No patients had surgery delayed or canceled, and no patients discontinued postoperative maintenance treatment due to AEs. Two patients died of surgical complications unrelated to study drug. The first patient had a postoperative hematoma that required reoperation, and subsequently developed a postoperative gram-negative infection that was unresponsive to antibiotic therapy. The second patient had confirmed SARS-CoV-2 infection that resulted in severe respiratory deterioration; of note, SARS-CoV-2 vaccines had not been approved in Europe at the time of the patient’s participation in the study, and thus the patient had not been vaccinated. Although the patient had a postoperative intracranial hemorrhage, their platelet count was normal throughout the study period and the hemorrhage was not considered to be related to treatment with trotabresib.

The 6-month PFS rate in CC-90010-GBM-001 was 12%, with a median PFS of 1.9 months; this result reflects the poor prognosis for the population enrolled in our study, with the evaluable population comprising patients with heavily pretreated recurrent glioblastoma. Although a subgroup analysis based on IDH mutation status showed longer PFS in patients with wild-type vs. mutant tumors, this result must be interpreted with caution due to the small number of patients—only 5 patients with IDH-mutant tumors were enrolled, of whom 2 died of postoperative complications and 1 withdrew from the study shortly after surgery. Outcomes in subgroups based on MGMT promoter methylation status were not evaluated due to the small number of patients with confirmed promoter methylation status (n = 14). Two patients with stable disease remained on treatment with trotabresib maintenance therapy at cycles 25 and 30 as of November 1, 2022, both of whom had IDH wild-type tumors and a methylated MGMT promoter.

In conclusion, trotabresib demonstrated brain tumor tissue penetration in patients with recurrent high-grade glioma, and a robust PK profile consistent with a previous study in patients with advanced malignancies.^23,24 Trotabresib was also well tolerated and showed modulation of pharmacodynamic markers of target engagement in blood, consistent with previous results.^23,24 Modulation of HEXIM1 expression in FFPE on-treatment resected tissue suggests that trotabresib concentrations in glioma tissue were sufficient for pharmacological activity. Two patients remain on treatment with prolonged stable disease. Based on these findings and the efficacy of trotabresib in patients with recurrent high-grade gliomas in the CC-90010-ST-001 study,^23,24 a phase Ib/II study, CC-90010-GBM-002 (NCT04324840), is investigating the safety and efficacy of concomitant trotabresib in combination with temozolomide and radiotherapy and adjuvant trotabresib plus temozolomide in patients with newly diagnosed glioblastoma.^37,38

Conflict of Interest:

V.M. has received consulting fees from Basilea, Bayer, Bristol Myers Squibb, Janssen, and Roche, and has received principal investigator institutional funding from AbbVie, ACEA Bio, Adaptimmune, ADC Therapeutics, Aduro, Agenus, amcure, Amgen, Astellas, AstraZeneca, Bayer, BeiGene, BioInvent International AB, Bristol Myers Squibb, Boehringer, Boston, Celgene, Daiichi Sankyo, Debiopharm, Eisai, e-Therapeutics, Exelixis, Forma Therapeutics, Genmab, GSK, Harpoon, Hutchison, Immutep, Incyte, Inovio, Iovance, Janssen, Kyowa Kirin, Lilly, Loxo Oncology, MEDSIR, Menarini, Merck, Merus, Millennium, MSD, Nanobiotix, Nektar, Novartis, Odonate Therapeutics, Pfizer, PharmaMar, Principia, PsiOxus, Puma, Regeneron, Rigontec, Roche, Sanofi, Sierra Oncology, Synthon, Taiho, Takeda, Tesaro, Transgene, Turning Point Therapeutics, and Upsher-Smith. D.A.R. has received advisory/consultation fees from Bristol Myers Squibb, Agios, AnHeart, Boston Biomedical, Del Mar Pharma, DNAtrix, Hoffman-LaRoche, Ltd, Imvax, Kiyatec, Medicenna, Neuvogen, Novartis, Pyramid, and Y-mabs, and his institution has received research support from Bristol Myers Squibb, Acerta Phamaceuticals, Agenus, Celldex, EMD Serono, Enterome, Epitopoietic Research Corporation, Incyte, Inovio, Insightec, Novartis, Omniox, and Tragara. J.M.S. has received research funding from IDP Pharma and Pfizer, and has received consulting fees from GSK, MSD, and UCB Pharma. A.P.-N. and P.G.L. have no conflicts of interest to declare. M.A.V. has participated in advisory boards for Chimerix, has received grant funding to his institution from Celgene, National Cancer Institute, Infuseon Therapeutics, Oncosynergy, and Denovo Therapeutics, and is named on a patent for Infuseon Therapeutics. B.A., C.M., T.S.-P., and Z.N. are employed by Centre for Innovation and Translational Research Europe, a Bristol Myers Squibb Company. B.H., D.S., E.F., and H.C.. are employed by Bristol Myers Squibb. B.H., C.M., D.S., E.F., H.C., M.Z., T.S.-P., and Z.N. hold Bristol Myers Squibb stock. I.A. and M.Z. were employed by Bristol Myers Squibb at the time of the study.

Funding

This work was supported by Celgene, a Bristol Myers Squibb company.

Authorship

Conception and design: Z.N., V.M., M.A.V. Patient recruitment: V.M., J.M.S., D.A.R., A.P.-N., P.G.L., M.A.V. Data analysis and interpretation: All authors. Drafting and critical review of the manuscript: All authors.