Abstract
Background: Bevacizumab combined with chemotherapy has recently shown promising efficacy in recurrent high-grade glioma. Phosphatase and tensin homolog (PTEN) mutation in glioblastoma multiforme (GBM) patients causes abnormally high activity of the pathways of Phosphatidylinositide 3-kinases (PI3K), Protein Kinase B (AKT), and the mammalian target of rapamycin (mTOR) and is associated with unfavorable prognosis. Temsirolimus, an mTOR inhibitor, has been well-tolerated in monotherapy, but with limited effects. The combination of temsirolimus and antibodies to vascular endothelial factor (VEGF) has not yet been investigated, but with the hypothesis that temsirolimus might provide complimentary therapeutic benefit in combination with bevacizumab, we included patients with progressive GBM after bevacizumab in an open phase II study. Patients and Methods: Adult patients with GBM recurrence after standard temozolomide chemoradiotherapy and bevacizumab-containing second-line therapy, received temsirolimus (25 mg i.v.) on days 1 and 8 and bevacizumab (10 mg/kg) on day 8, every two weeks. Assessments were performed every eight weeks. Blood samples for biomarkers were collected weekly for the first eight weeks and at progression. The primary end-point was median progression-free survival (PFS) and secondary end-points were radiographic response, overall survival (OS), and safety of the bevacizumab-temsirolimus combination. Results: Thirteen patients were included, whereof three went off-study during the first four weeks and were replaced. The trial was terminated at 13 patients, according to the planned two-stage design, because 0/10 patients obtained partial remission (PR). Two out of 10 patients obtained radiological stable disease (SD). The median PFS survival was eight weeks, and OS was 15 weeks. One patient had an serious adverse event (SAE) with a hypersensitive reaction to temsirolimus; overall, side-effects were mild, and the most common grade III side-effect was hypercholesterolaemia (4/10). Other grade III side-effects included hypertriglyceridaemia (1/10), thrombocytopenia (1/10), infection (1/10), hypertension (1/10), and hyperglycemia (1/10). Conclusion: Temsirolimus can be safely administered in combination with bevacizumab. This study failed to detect activity of such a combination in patients with progressive GBM beyond bevacizumab therapy.
Glioblastoma multiforme (GBM) is the most common primary malignant neoplasm of the central nervous system in adults. Treatment outcome remains poor despite multimodal therapies, including debulking surgery, radiotherapy, concomitant and adjuvant chemotherapy (1). Median survival for patients receiving this treatment is 14.6 months.
Malignant gliomas have a high expression of vascular endothelial growth factor (VEGF), and monoclonal antibodies to VEGF inhibited growth of malignant gliomas in a mouse xenograft model (2). Bevacizumab, a monoclonal antibody to VEGF, has shown activity in high-grade glioma and was approved by the US Food and Drug Administration in May 2009 for the treatment of recurrent glioblastoma (3). However, progression during or after bevacizumab therapy is frequent as tumor cells can migrate via existing vessels and are not solely-dependent on neovascularization (4).
Phosphatase and tensin homolog (PTEN) mutation in GBM causes abnormally high activity of the pathways of Phosphatidylinositide 3-kinases (PI3K), Protein Kinase B (AKT), and the mammalian target of rapamycin (mTOR), and is associated with unfavourable prognosis (5, 6). Use of temsirolimus, an mTOR inhibitor, in vivo against myeloma xenografts, has indicated additional effects of AKT activity on apoptosis that correlate with the inhibition of angiogenesis (7). In vivo studies of an orthotopic B16 melanoma model have indicated that angiogenesis inhibitors and mTOR inhibitors may act synergistically (8, 9). However, the clinical activity of temsirolimus was modest in recurrent GBM (10).
We hypothised that temsirolimus might provide complimentary therapeutic benefit in combination with bevacizumab. We, therefore, included patients with recurrent glioblastoma and progression after bevacizumab-based therapy in an open label phase-II study.
The primary aim was to determine the efficacy of this regimen using progression-free survival (PFS) as the primary end-point. Secondary end-points included the adverse event profile and objective response rate; comparison of pre- and post-treatment measurements of biomarkers and vascular system/immune system parameters; and correlation of tumor and blood biomarkers with clinical response (VEGF, PTEN, AKT, p-AKT, p53).
Patients and Methods
This investigator initiated open label phase II study was supported by Pfizer Inc. and F. Hoffmann La Roche with study medication and carried out with local funding from Rigshospitalet and Copenhagen University (ClinicalTrials.gov Identifier: NCT00800917). The protocol was approved by the Danish Medicines Agency and the local Ethics Committee (approval no.: H-D-2008-076), and monitored according to good clinical practice (GCP) by the GCP unit of Copenhagen University.
Study eligibility criteria included: age ≥18 years; performance status (PS) 0-1; histologically proven recurrent GBM; measurable tumor by magnetic ressonance imaging (MRI); previous treatment with bevacizumab; more than four weeks since chemotherapy (six weeks for nitrosoureas or mitomycin C); and normal organ function. Furthermore standard criteria regarding contraceptive use and informed consent was applied.
Neuro-imaging with MRI, including pre-contrast T1, T2/Fluid Attenuated Inversion Recovery (FLAIR) and post-contrast T1, with two orthogonal planes and T2-weighted FLAIR, was performed at baseline, before the first treatment cycle, and every eight weeks thereafter. Tumor size was measured in two dimensions, according to RANO criteria (11).
Treatment plan. Flat dose temsirolimus was administered weekly as a 30-minute i.v. infusion of 25 mg on days 1, 8, 15 and 22, and bevacizumab was administered at 10 mg/kg i.v. over 30-90 minutes on day 8 and 22. Treatment was repeated every 28 days for a maximum of 12 courses in the absence of disease progression or unacceptable toxicity. Premedication with i.v. clemastine 2 mg was administered 30 minutes before temsirolimus administration. Concomitant use of inducers or inhibitors of Cytochrome P450 3A4 (CYP3A4) was not allowed.
A safety analysis was planned when the first 10 patients had received a minimum of four cycles (eight weeks). The study was to be stopped if one of the following occurred: dose limiting toxicity (DLT), with standard criteria were applied, observed in >2/10 patients; occurrence of any serious adverse events not described in the summary of product characteristics (SPC) of each agent, or if partial remission (PR) not observed in at least 1/10 patients (RANO criteria).
Therapy was discontinued if any of the following occurred: patient's consent withdrawn; grade 4 hypertension or haemorrhage, grade 4 venous thromboembolic event; any grade arterial thromboembolic event; venous thomboembolic event requiring anticoagulant therapy; gastrointestinal perforation; wound dehiscence requiring medical or surgical intervention; progressive disease; grade 2 or greater CNS haemorrhage; grade 4 proteinuria; non-compliance by the patient.
A patient could continue on-study with temsirolimus after discontinuation of bevacizumab due to occurrence of DLT or serious treatment related adverse events (AE) from bevacizumab. If a patient only received one cycle of treatment this patient would be replaced in order to fulfil the safety and futility analyses, unless treatment was discontinued due to adverse events.
Endpoints and statistics. The primary endpoint was median PFS and secondary endpoints were radiographic response, overall survival (OS) and safety. Yung et al. (12) reported a median PFS of 3.1 months among patients with first relapse GBM who were treated with temozolamide. Recent data have indicated that a PFS of 4.6 months can be achieved with bevacizumab and irinotecan (13). The expected PFS after bevacizumab-based therapy is even lower, as we anticipated that most patients had discontinued prior regimens with bevacizumab due to disease progression. We planned to include 32 evaluable patients over 12-24 months with at least 12 months of follow-up for every patient. With a median PFS of historical controls of three months, there would be an approximate 80% power to detect an improvement of two months, and an approximate 60% power to detect an improvement of 1.4 months.
Results
Thirteen patients were included between November 2008 and December 2009. Three patients went off-study due to disease progression during the first cycle and were replaced according to protocol. The patients were evenly distributed by gender (5 males, 8 females) with a median age of 48 years (range=24-72 years). The trial was terminated early because no patients obtained PR out of the 10 evaluable patients. The median treatment duration was seven weeks (range=2-16 weeks). Two out of 10 patients obtained radiological SD for four months. Median PFS was eight weeks and OS 15 weeks. One patient had a grade 2 hypersensitivity reaction, and one patient developed grade 3 deep-vein thrombosis after eight weeks on study. Overall, adverse events were mild, with the most common treatment-related grade 3 event being hypercholesterolemia (4/10). Other grade 3 adverse effects included hypertriglyceridaemia (1/10), thrombocytopenia (1/10), infection (1/10), hypertension (1/10) and hyperglycaemia (1/10). Related grade 1 and 2 events included stomatitis, fatigue, proteinuria, hypertension, alopecia, cough, gastritis, nausea, vomiting, hyperglycaemia and skin rash. Biomarker assessments were not performed as clinical activity was not observed.
Discussion
Recent in vivo studies have shown synergy between everolimus and sunitinib in gastric cancer (14). In addition, mTOR inhibition has shown activity in renal cancer after failure with VEGF inhibitors (15). Therefore, the combination of bevacizumab and temsirolimus was explored in this study, despite modest activity with mTOR inhibitors in recurrent glioma (10, 16). Combinations of mTOR inhibitors and receptor tyrosine kinase inhibitors of both VEGF and EGFR have resulted in extensive skin toxicity (17, 18), whereas combinations of mTOR inhibitors and VEGF antibodies have only recently been combined in clinical studies (19, 20).
Various doses of temsirolimus have been evaluated, and most studies of patients with high-grade gliomas have used a weekly dose of 250 mg temsirolimus intravenously (10, 16). However, this dose has been associated with significant adverse events (21, 22). In renal cancer, several doses of temsirolimus have been compared, and no difference in activity or adverse events was observed between weekly doses of 25 mg and 250 mg (1). Consequently, the FDA-approved dose in renal cancer is 25 mg weekly. The brain tumor concentration of temsirolimus is comparable to blood concentrations, indicating adequate distribution in the CNS (21). A recent study compared temsirolimus and bevacizumab with interferon and bevacizumab in renal cancer (24). The combination was safe although grade 3 stomatitis and and hypercholesterolemia was frequently observed. In the present study, temsirolimus and bevacizumab was well-tolerated without toxicity exceeding the frequency or grade of already known adverse events from either agent. No activity was observed, and the study was prematurely terminated after a pre-planned futility analysis, as no response was observed among the first 10 evaluable patients.
This regimen has limited activity for recurrent GBM, but may provide therapeutic benefits in other cancer types.
- Received January 28, 2013.
- Revision received March 11, 2013.
- Accepted March 12, 2013.
- Copyright© 2013 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
