Phase II Study of Imatinib in Patients With Recurrent Gliomas of Various Histologies: A European Organisation for Research and Treatment of Cancer Brain Tumor Group Study

Abstract

Purpose

To evaluate the safety and the efficacy of imatinib in recurrent malignant gliomas.

Patients and Methods

This was a single-arm, phase II study. Eligible patients had recurrent glioma after prior radiotherapy with an enhancing lesion on magnetic resonance imaging. Three different histologic groups were studied: glioblastomas (GBM), pure/mixed (anaplastic) oligodendrogliomas (OD), and low-grade or anaplastic astrocytomas (A). Imatinib was started at a dose of 600 mg/d with dose escalation to 800 mg in case of no toxicity; during the trial this dose was increased to 800 mg/d with escalation to 1,000 mg/d. Trial design was one-stage Fleming; both an objective response and 6 months of progression-free survival (PFS) were considered a successful outcome to treatment.

Results

A total of 112 patients (51 patients with GBM, 25 patients with A, and 36 patients with OD) were enrolled. Imatinib was in general well tolerated. The median number of cycles was 2.0 (range, 1 to 43 cycles). Five patients had an objective partial response, including three patients with GBM; all had 6 months of PFS. The 6-month PFS rate was 16% (95% CI, 8.0% to 34.0%) in GBM, 4.0% (95% CI, 0.3% to 15.0%) in OD, and 9% (95% CI, 2.0% to 25.0%) in A. The exposure to imatinib was significantly lower in patients using enzyme-inducing antiepileptic drugs. The presence of ABCG2 point mutations were not correlated with pharmacokinetic findings. No somatic activating mutations of KIT or platelet-derived growth factor receptor–A or –B were found.

Conclusion

In the dose range of 600 to 1,000 mg/d, single-agent imatinib is well tolerated but has limited antitumor activity in patients with recurrent gliomas.

INTRODUCTION

Despite recent advances in the management of gliomas using temozolomide alone or in combination with radiotherapy for newly diagnosed tumors, treatment options for patients with recurrent astrocytomas and oligodendrogliomas remain limited. Ultimately, all patients develop progressive disease, which requires the development of more effective salvage regimens. In vivo and in vitro studies on glioma models have shown that tyrosine kinase receptors play an important role in glioma cell signal transduction. Activation of these pathways leads to increased cell proliferation, migration, and survival. The platelet derived growth factor receptor (PDGFR) pathway is one of these signaling pathways that seems to be involved in glioma. Although PDGFR mutations and amplification are rare, diffuse gliomas frequently overexpress both the ligands platelet-derived growth factor (PDGF) –A and/or PDGF-B and their receptors, which suggests the presence of autocrine loops.^1-3 Levels of expression increase with increasing tumor grade, but PDGFR overexpression is already present in low-grade gliomas. Blocking of PDGF-mediated phosphorylation caused either apoptosis or growth inhibition in glioma cell lines.^4,5 PDGF autocrine loops may be even more relevant for oligodendroglioma. Virtually all oligodendrogliomas express both PDGF ligands as well as both the PDGF-α and -β receptors.^6-8 Both O-2A progenitor cells (which are assumed to be the precursor cells of oligodendrocytes and which may be the cells of origin of oligodendroglial tumors) and oligodendrogliomas express the combination of NG2 chondroitin proteoglycan sulfate and PDGF-α, a combination which is neither observed in glioblastoma (GBM) nor in normal brain.^8,9 PDGF has a strong mitogenic effect on these glial progenitor cells.¹⁰ Moreover, induction of PDGF expression by intracranial gene transfer in mice induced the formation of tumors with an oligodendroglial phenotype in a transgenic mice model.¹¹ These findings make the PDGF pathway a potential target in nearly all glioma subtypes.

Imatinib mesylate (imatinib) is a multitargeted tyrosine kinase inhibitor that blocks KIT, Bcr-Abl, and PDGFR in several in vitro and in vivo models.^12,13 Imatinib displayed antitumor activity in PDGFR-dependent human GBM primary cell cultures.¹⁴ We therefore initiated a multicenter phase II study to evaluate the safety and the antitumor activity of imatinib in patients with progressive gliomas after prior radiotherapy. Translational research included correlative studies of PDGFR, KIT, and ABCG2 mutations/polymorphisms and imatinib pharmacokinetic (PK) measurements.

PATIENTS AND METHODS

Patient Population

Eligibility criteria were age ≥ 18 years, a progressive glioma after prior radiotherapy and documented by magnetic resonance imaging (MRI) or computed tomography (CT) made within 2 weeks from the start of treatment, stable or decreasing doses of corticosteroids for at least 2 weeks before baseline scan, Eastern Cooperative Oncology Group performance status ≤ 2, at least one bidimensionally measurable contrast-enhancing target lesion with at least one diameter of at least 2 cm, no brain irradiation in the last 3 months and no dosage of radiotherapy more than 65 Gy, no brain surgery within the last 3 months, and adequate bone marrow, renal, hepatic, and cardiac function. Patients requiring treatment with warfarin or any other anticoagulants and patients with unstable or active uncontrolled infections were not eligible. The study design was approved by the local and/or national ethic committee of all participating institutions, and all patients provided written informed consent according to International Congress on Harmonisation/Good Clinical Practice standards and national/local regulations.

Eligible Histologies and Specific Study Requirements

Patients were enrolled onto the study according to the local pathologic diagnosis. Three different histologic subgroups were defined: recurrent GBM, recurrent low-grade astrocytoma and anaplastic astrocytoma (A), and recurrent low-grade or anaplastic oligodendroglioma or mixed (anaplastic) oligoastrocytoma (OD). Patients with GBM were eligible if they had received either no or no more than one prior chemotherapy regimen, either adjuvant or at administered at disease recurrence. Patients with A or OD were eligible if they had received one prior chemotherapy regimen, either adjuvant or administered at recurrence.

Imatinib Administration

At the start of the study, treatment consisted of imatinib at a once daily oral dose of 600 mg/d escalated to 800 mg/d (400 mg twice a day) in the absence of grade 2 or worse toxicity during the first 8 weeks of treatment. On the basis of initial safety data showing that imatinib was well tolerated in this patient population, the protocol was amended to increase the dose of imatinib to 800 mg/d (400 mg twice a day), with dose escalation to 1,000 mg/d (500 mg twice a day) after 8 weeks in the absence of grade 2 or worse toxicity. One cycle was defined as 4 weeks of imatinib. Treatment was continued until unacceptable toxicity or evidence of tumor progression.

Evaluation of Toxicity and Antitumor Activity

Patients were monitored for toxicity weekly for the first 8 weeks (in case of dose escalation, the first 8 weeks at the higher dose also required weekly monitoring). Toxicity was graded according to National Cancer Institute Common Toxicity Criteria version 2.0. Response to treatment was assessed every two cycles using MRI or CT scan according to Macdonald's criteria.¹⁵ The scans of all patients reportedly having an objective response (partial or complete response) or 6 months of progression-free survival (PFS) were centrally reviewed.

Study Design

This trial was an open-label, multicenter, phase II study. Because both an objective response (even if short lasting) and prolonged disease stabilization were considered to suggest activity, a combined primary end point (labeled success) was used, defined as either an objective response (partial or complete response) and/or 6 months of PFS. Secondary end points were duration of response, survival, and toxicity, as well as PK and pharmacodynamic analysis. The sample size was determined according to a one-stage Fleming design that was applied separately to each pathologic stratum.¹⁶ Imatinib was considered worthy of further investigations if success rates were obtained in 25%, 30%, and 30% for patients with GBM, A, and OD, respectively. With an α = 0.10 and β = 0.10, respectively 29, 24, and 24 eligible patients with GBM, A, and OD were required. When the dose was amended to a starting dose of 800 mg, the study was restarted.

Pharmacokinetic Analysis

A limited sampling strategy was used to investigate the exposure to imatinib and its main metabolite CGP74588 in patients with gliomas. At cycle 1, 8-mL blood samples were drawn on day 1, before the first imatinib intake, between 1 and 3 and 6 and 9 hours after first drug administration, then before the first intake at day 2. One additional sample was taken at day 1 of cycle 2 before imatinib to evaluate the steady-state. Samples were centrifuged at 4,000 × g for 15 minutes and stored at −20°C until analysis. Imatinib and CGP74588 concentrations were determined as described elsewhere using liquid chromatography coupled with tandem mass spectrometry. Maximum plasma concentration (C_max), area under the serum concentration-time curve (AUC_0-∞), and concentration ratio at steady-state are reported.

Molecular Diagnosis

Paraffin-embedded tumor samples were retrieved for pathology review (J.M.K.) and mutational analysis (M.C.H.) for mutations of KIT (exons 9, 11, 13, and 17) and PDGFRα (exons 12, 14, and 18) and PDGFRβ (exons 11 and 17) as previously described.¹⁷ Genotyping for the hypomorphic Q141K ABCG2 polymorphism was performed using tumor DNA and a Fluorescent Resonance Energy Transfer–based hybridization probes and melting curve analysis using a Roche LC480 instrument (Roche, Neuilly sur Seine, France; primer, probe, and polymerase chain reaction conditions available on request).¹⁸

Statistical Analysis

Demographic, response rate, safety, laboratory, PK, and pharmacodynamic data were analyzed using descriptive statistics. Kaplan-Meier technique was used to estimate medians and rates of PFS and overall survival at 6 months. PFS was measured from the day of registration to objective disease progression or death, whichever occurred first. Patients were considered as censored if they never experienced disease progression or death. The log-rank test was used for survival curves comparisons. Continuous measurements were summarized by means and coefficient of variation and compared using the Kruskall-Wallis test.

RESULTS

Patient Characteristics

From March 2002 to August 2004, 112 patients (51 patients with GBM, 25 patients with A, and 36 patients with OD) were entered onto this study. Thirty patients were treated at the 600 mg/d starting dose level, and 82 patients were treated at the 800 mg/d starting dose level. Two patients were not eligible (one patient at the dose of 600 mg, one patient at 800 mg) because of a too-small baseline lesion; in one patient, baseline data were not made available. Baseline patient characteristics are listed in Table 1. No tumor material was received from 18 patients. The diagnosis of GBM was confirmed at review in 44 (92%) of 48 patients, diagnosis of an OD was confirmed in 13 (50%) of 26 patients, and diagnosis of an A was confirmed in nine (45%) of 20 patients. In case of discrepancy at review, usually another eligible histologic diagnosis was made. The median number of cycles was 2.0 (range, one to 43 cycles; Table 1). Dose reduction and cycle delay (mainly for nonhematologic toxicity) were performed in 11 (9.8%) and 36 patients (32.1%), respectively. At the 600-mg and 800-mg dose level, the dosage was increased in 15 and 16 patients (to 800 and 1,000 mg, respectively).

Table 1.

Patient Characteristics and Protocol Treatment Information

Characteristic	No. of Patients
	Imatinib 600 mg	Imatinib 800 mg
Sex
Male	10	55
Female	20	24
Age, years
Median	52	46
Range	23-66	19-72
ECOG performance status
0, 1	24	70
2	6	10
Prior chemotherapy
Adjuvant	3	21
For recurrence	3	49
Imatinib treatment
Patients with dose reduction	5	6
Patients with cycle delay	12	24
Patients with dose increase	15	16
Mean imatinib dose-intensity, cycle 1-2
≤ 70%	2	0
> 70% to ≤ 90%	5	14
> 90% to ≤ 110%	22	66
> 110% to ≤ 120%	1	0

Abbreviation: ECOG, Eastern Cooperative Oncology Group.

Efficacy

Table 2 lists the observed responses, PFS, and overall survival for all patients and the patients treated at 800 mg. At the dose level of 800 mg, and with the diagnosis according to the local pathologist, PFS at 6 months was observed in five (16%) of 31 patients with GBM, one (4%) of 27 patients with OD, and two (9%) of 22 patients with A. A partial response was observed in only two of the 80 patients treated at 800 mg. In some patients, decreased contrast enhancement was observed despite clinical deterioration, with T2-weighted images suggestive of tumor progression (Fig 1).

Table 2.

Outcome of Imatinib Treatment in All 110 Patients With Measurable Disease

Variable	800 mg			All (600 mg and 800 mg)
	GBM	AOD/mixed OA	AA/rec LGA	GBM	AOD/mixed OA	AA/rec LGA
All patients, n	31	27	22	50	35	25
Success, n			20
No	26	26	2	42	31	22
Yes, 6 months of PFS or CR/PR	5	1		8	4	3
6-month PFS	5	1	2	8	4	3
PFS, months
Median	1.7	1.8	1.7	1.8	1.9	1.8
95% CI	1.6 to 2.3	1.7 to 2.3	1.2 to 2.1	1.7 to 2.3	1.8 to 3.2	1.3 to 2.1
OS, months
Median	5.2	4.8	4.9	5.9	5.3	5.0
95% CI	3.8 to 7.8	2.8 to 7.7	3.1 to 9.7	4.2 to 7.8	3.7 to 10.9	3.1 to 9.7
6-month OS rate, %
Median	48	32	45	50.0	42.9	48.0
95% CI	30 to 64	17 to 51	24 to 64	35.6 to 62.8	26.4 to 58.3	27.8 to 65.6
Best response
PR
No.	1	0	1	3	1	1
%	3	0	5	6	3	4
SD
No.	8	4	3	13	9	5
%	26	15	14	26	25	20

Abbreviations: GBM, glioblastoma; AOD, anaplastic oligodendroglioma; OA, oligoastrocytoma; AA, anaplastic astrocytoma; rec LGA, recurrent low-grade astrocytoma; PFS, progression-free survival; CR, complete response; PR, partial response; OS, overall survival; SD, stable disease.

Fig 1.

A patient started treatment with imatinib in October 2002 for a progressive oligoastrocytoma after receiving 60 Gy of radiotherapy in 2001 and procarbazine, cyclophosphamide, and vincristine chemotherapy in 2002. At the first response assessment after two cycles, despite a slight increase of sensory deficits, T1-weighted contrast enhanced magnetic resonance imaging (MRI) suggested some decrease in enhancement and treatment was continued. One and a half weeks later, the patient was admitted because of increasing headache and further neurologic deterioration; T1-weighted contrast-enhanced MRI showed no new findings. However, close examination of T2-weighted images showed an increase of the tumor volume. Upper row: contrast-enhanced T1-weighted MRI; lower row, T2-weighted MRI on October 16, 2002 (baseline), December 18, 2002 (first response evaluation), and January 3, 2003 (at the time of admission).

Safety

Although the incidence of toxic events was slightly higher in patients receiving the dose of 800 mg/d as compared with those receiving 600 mg/d, treatment was well tolerated (Table 3). Hematologic toxicity mainly consisted of grade 1 to 2 neutropenia, with only five patients developing febrile grade 3 to 4 neutropenia at the dose of 800 mg/d. Mild to moderate increases in renal and hepatic functions were frequently reported in patients receiving imatinib concomitant with antiepileptic drugs. Grade 1 to 2 edema, primarily infra-orbital, was reported in approximately 40% of patients, regardless of the dose of imatinib.

Table 3.

Toxicity of Imatinib in Patients With Malignant Gliomas

Toxicity	Imatinib 600 mg				Imatinib 800 mg
	Grade 1-2		Grade 3-4		Grade 1-2		Grade 3-4
	No. of Patients	%	No. of Patients	%	No. of Patients	%	No. of Patients	%
Neutropenia	6	20	2	6.6	13	16	9	11
Febrile neutropenia	—		—		—		5	6.2
Thrombocytopenia	2	6.6	—		8	9.8	—
Anemia	18	60	1	3.3	36	44.4	1	1.2
Fatigue	4	13.3	2	6.6	26	32.1	2	2.4
Fever without neutropenia	—		—		2	2.4	—
Creatinine	3	10	—		9	11	—
Total bilirubin	3	10	—		12	14.8	—
ALT	9	30	2	6.6	33	40.7	2	2.4
AST	11	36.6	—		19	23.4	—
Alkaline phosphatase	10	33.3	—		22	27.1	1	1.2
Albumin	13	43.3	1	3.3	74	91.3	1	1.2
Edema	12	40	—		31	38.2	1	1.2
Hypotension	—		—		—		1	1.2
Skin toxicity*	6	20	—		23	28.4	1	1.2
Diarrhea	3	10	—		16	19.7	—
Nausea	15	50	—		23	28.4	—
Vomiting	11	36.6	1	3.3	13	16	1	1.2
Stomatitis	—		—		4	4.9	—
Arthralgia and myalgia	1	3.3	—		3	3.7	—

^*Skin toxicity included rash and dry skin.

Pharmacokinetic Parameters

Pharmacokinetic data were obtained for 80 patients, allowing an accurate determination of C_max, AUC, and the steady-state concentration ratio for imatinib and its main metabolite CGP74588. Overall, there was considerable interpatient variability. At the doses of 600 and 800 mg, C_max for imatinib was 1,366 ng/mL (49%) and 1,335 ng/mL, respectively; C_max of CGP74588 was 300 ng/mL and 234 ng/mL, respectively. At the doses of 600 and 800 mg, AUC_0-∞ of imatinib were 771 ng · h/mL and 1,089 ng · h/mL, respectively; and AUC_0-∞ of CGP74588 were 172 ng · h/mL and 199 ng · h/mL, respectively. Concentration ratios at steady-state were 3.9 and 2.5 for imatinib at the dose of 600 and 800 mg, respectively. Concentration ratios at steady-state were 4.4 and 3.6 for CGP74588 at the dose of 600 and 800 mg, respectively. C_max and AUC_0-∞ were significantly reduced (P = .0019 and P ≤ .0001) in patients taking enzyme-inducing antiepileptic drugs (EIAEDs; Table 4). No effect of EIAEDs was observed on concentration ratio at steady-state and on the PK parameters of CGP74588. PFS was not affected by dosing of imatinib nor by any PK parameters.

Table 4.

Pharmacokinetic Parameters for STI571 and its Main Metabolite, CGP74588, in Patients Not Taking Antiepileptic Drugs and Patients Taking Either EIAEDs or Non-EIAEDs

Parameter	No.	Median	Minimum Observed Value	Maximum Observed Value	P (df = 2) All Patients
Cmax, STI571, ng/mL
No antiepileptic drugs	18	1595	533	3930	.0019
EIAEDs	37	861	35	2960
Non-EIAEDs	25	1270	78	4620
Cmax, CGP74588, ng/mL
No antiepileptic drugs	18	225	49	1170	.35
EIAEDs	37	198	18	892
Non-EIAEDs	25	166	11	570
AUC, STI571, ng·h/mL
No antiepileptic drugs	18	1263	314	2574	< .0001
EIAEDs	37	614	247	1611
Non-EIAEDs	25	1139	192	3021
AUC, CGP74588, ng·h/mL
No antiepileptic drugs	18	174	34	942	.31
EIAEDs	37	152	59	529
Non-EIAEDs	25	160	22	552
RAP29, STI571
No antiepileptic drugs	10	1.87	0.68	16.44	.99
EIAEDs	28	1.85	0.62	6.58
Non-EIAEDs	15	1.80	0.88	24.38
RAP29, CGP74588
No antiepileptic drugs	10	2.75	1.48	9.70	.87
EIAEDs	28	2.65	1.23	6.46
Non-EIAEDs	15	2.25	0.92	39.60

Abbreviations: EIAEDs, enzyme-inducing antiepileptic drugs; Non-EIAEDs, non–enzyme-inducing AEDs; C_max, maximum concentration; AUC, area under the curve; RAP, concentration at steady state (day 29 before taking imatinib).

Molecular Analysis

Tumor samples allowing molecular analysis were available for 70 patients. Among 69 assessable specimens, no somatic activating mutations of KIT, PDGFRα, or PDGFRβ were detectable. PDGFRα single nucleotide polymorphisms (SNPs) were observed in the intron of exon 11 in one patient, in exon 14 in one patient (K666K), and in exon 18 in seven patients (SNP V824V). PFS was not correlated with PDGFRα SNPs. In 69 patients, material was available for the assessment of ABCG2 point mutations. Twelve patients had the Q141K exon 5 SNP. No correlation was found between the presence of ABCG2 point mutations and any of the pharmacokinetic parameters (data not shown).

DISCUSSION

Oral imatinib doses of 600 and 800 mg daily with dose escalation to 1,000 mg/d in case of no significant toxicity was well tolerated in patients with malignant gliomas, with a slightly increased incidence of mild to moderate toxicity in patients receiving higher imatinib doses. Despite the sporadic objective responses and prolonged tumor stabilization that were observed, the overall response rate and 6-month PFS rate achieved are insufficient to suggest that imatinib has clinically useful activity in any subtype of diffuse gliomas when given as a single agent. Somewhat unexpectedly, the few sustained objective responses were mainly observed in patients with GBM, with 16% PFS at 6 months. Although currently a 6-month PFS rate of less than 15% is considered as the benchmark indicating a negative trial in recurrent high-grade glioma, previous negative studies from the European Organisation for Research and Treatment of Cancer in recurrent GBM using classical cytotoxic agents and using similar inclusion criteria consistently observed 6-month PFS rates well below 10%.^19-22 Although this suggests that imatinib has some low level of activity in GBM, this was not confirmed in a North American study on 35 patients with GBM.²³ In that study, only a 3% 6-month PFS rate was noted, together with a 10% 6-month PFS rate in 15 recurrent grade 3 tumors. Recent phase II studies reported interesting activity of imatinib combined with hydroxyurea, a ribonucleoside diphosphate reductase inhibitor, in patients with recurrent GBM.^24,25 A large phase III trial, however, has not been able to confirm these findings.²⁶ The North American study observed intratumoral hemorrhages in patients with GBM treated at 800 mg, although all occurred in the setting of progressive disease.²³ A phase I study on pediatric brainstem gliomas and recurrent malignant gliomas also observed intratumoral hemorrhages.²⁷ In the present study, no clinically significant hemorrhages were observed.

The outcome of the central pathology review in our study was similar to that of other studies in glioma: a high rate of confirmation of locally diagnosed GBMs but frequent discrepancies in the diagnosis of ODs and As. However, at central review, enough patients with the various glioma subtypes were left to conclude that sufficient single-agent activity of imatinib was present in none of the glioma subtypes.

Imatinib is metabolized by the CYP4503A4, with the N-demethylation of the piperazine 4-nitrogen producing its main metabolite CGP74588. Because many patients with brain tumors receive EIAEDs that may modify the PK profile of imatinib, the exposure to imatinib and its main metabolite CGP74588 was investigated using a limited sampling strategy. Overall, exposure to imatinib was similar in patients receiving 600 and 800 mg imatinib, possibly as a result of the large interpatient variability at daily dosage greater than 600 mg. The use of EIAEDs significantly influenced the PK profile of imatinib, with an approximately 50% reduction of its C_max and AUC. Still, no correlation between exposure and toxicity was detectable nor between PK exposure or used imatinib dosage (600 v 800 mg) and antitumor activity (data not shown). In a phase I dose-escalation study, other investigators have shown that glioma patients using EIAEDs may receive up to 1,200 mg/d of imatinib without developing dose-limiting toxicity.²³ In that study, plasma exposure of imatinib was reduced by approximately 68% in patients receiving EIAEDs. The presence of ABCG2 point mutations were not correlated with PK findings. This confirms the previous findings that common genetic variants in this gene have only a limited impact on the PK of imatinib.²⁸

As expected, all analyzed tumors were shown to express wild-type PDGFRs at the time of initial surgery. A drawback of our approach is that no samples collected at the time of recurrence were investigated. Our data confirm that somatic mutations of PDGFRs are infrequent in gliomas. Expression of PDGFR and its ligands using immunohistochemistry was not attempted because of its modest reliability in archival paraffin-embedded tissue samples. Despite the upregulation of PDGF signaling pathways in many gliomas, the lack of significant imatinib activity in our study and those of others may reflect a lack of importance of PDGFR signaling for the survival and growth of human gliomas. An insufficient exposure of the tumor cells to imatinib could be another explanation for the lack of activity in glioma. Imatinib is a substrate for the P-glycoprotein efflux pump, limiting the passage of imatinib over an intact blood-brain barrier.²⁹ Alternatively, sufficient levels of target inhibition may not have been achieved. These latter considerations are a problem for nearly all recently investigated targeted agents in gliomas. Most of these trials failed, but only a few studies explored whether the intended intratumoral target inhibition was indeed achieved. A recent study that investigated GBM samples before and after exposure to imatinib demonstrated inhibition of AKT and MAPK activity, but no significant effect on PDGFRA/B or KIT. In some patients, a persistent phosphorylation level of PDGFRB was detected despite imatinib treatment, which may imply an imatinib-resistant activation mechanism of PDGFRB.³⁰ To answer these pivotal questions on drug penetration and target inhibition, investigations of new targeted treatments in gliomas need to integrate molecular analyses into the design of early trials (phase 0 trials).

In this study, we also observed pseudo improvements, characterized by a decrease of gadolinium enhancement on MRI, despite the fact that the patient was clinically deteriorating. This may be due to the expression of PDGFR on both endothelial cells and pericytes, leading to a normalization of abnormal vessel permeability or to changes in regional cerebral blood volume without a real antitumor effect. Similar though more pronounced observations have recently been made in studies on agents interfering with vascular endothelial growth factor signaling pathways (bevacizumab, AZD2171).^31,32 This suggests that classical T1-weighted and contrast-enhanced MRI alone may not be the best way to monitor responsiveness to agents interfering with the PDGF and vascular endothelial growth factor signaling pathways.

In summary, using doses ranging from 600 to 1,000 mg/d, imatinib had an acceptable safety profile but no clinically significant activity in patients with recurrent diffuse gliomas, although marginal activity was observed in GBM. On the basis of these results, further studies using single-agent imatinib at doses of 600 to 800 mg/d are not warranted in patients with malignant gliomas.

AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.

Employment or Leadership Position: None Consultant or Advisory Role: Roger Stupp, Novartis Pharmaceuticals (C); Michael C. Heinrich, Novartis Pharmaceuticals (C), Molecular MD (U); Martin J. van den Bent, Novartis Pharmaceuticals (C) Stock Ownership: Michael C. Heinrich, Molecular MD Honoraria: Roger Stupp, Novartis Pharmaceuticals; Michael C. Heinrich, Novartis Pharmaceuticals; Martin J. van den Bent, Novartis Pharmaceuticals Research Funding: Michael C. Heinrich, Novartis Pharmaceuticals; Martin J. van den Bent, Novartis Pharmaceuticals Expert Testimony: None Other Remuneration: None

AUTHOR CONTRIBUTIONS

Conception and design: Eric Raymond, Denis Lacombe, Martin J. van den Bent

Administrative support: Roger Stupp, Denis Lacombe, Martin J. van den Bent

Provision of study materials or patients: Eric Raymond, Alba A. Brandes, Christian Dittrich, Pierre Fumoleau, Bruno Coudert, Paul M.J. Clement, Marc Frenay, Roy Rampling, Roger Stupp, Johan M. Kros, Martin J. van den Bent

Collection and assembly of data: Eric Raymond, Alba A. Brandes, Christian Dittrich, Roy Rampling, Roger Stupp, Michael C. Heinrich, Martin J. van den Bent

Data analysis and interpretation: Eric Raymond, Michael C. Heinrich, Thierry Gorlia, Denis Lacombe, Martin J. van den Bent

Manuscript writing: Eric Raymond, Alba A. Brandes, Pierre Fumoleau, Paul M.J. Clement, Roger Stupp, Michael C. Heinrich, Thierry Gorlia, Martin J. van den Bent

Final approval of manuscript: Eric Raymond, Alba A. Brandes, Christian Dittrich, Pierre Fumoleau, Bruno Coudert, Paul M.J. Clement, Marc Frenay, Roy Rampling, Roger Stupp, Johan M. Kros, Michael C. Heinrich, Thierry Gorlia, Denis Lacombe, Martin J. van den Bent