Sunitinib Administered Prior to Radiotherapy in Patients with Non-Resectable Glioblastoma: Results of a Phase II Study

Targ Oncol DOI 10.1007/s11523-014-0305-1

ORIGINAL RESEARCH

Sunitinib administered prior to radiotherapy in patients with non-resectable glioblastoma: results of a Phase II study

Carmen Balaña & Miguel J. Gil & Pedro Perez & Gaspar Reynes & Oscar Gallego & Teresa Ribalta & Jaume Capellades & Sofia Gonzalez & Eugenia Verger

Received: 31 May 2013 /Accepted: 3 January 2014 # Springer-Verlag France 2014

Abstract Sunitinib is a tyrosine kinase inhibitor with direct with sunitinib 37.5 mg daily for 8 weeks pre-radiotherapy, anti-tumor and anti-angiogenesis activity targeting VEGFR during radiotherapy (60 Gy, 6 weeks) and post-radiotherapy 1–2, PDGFR α–β, c-kit, bFGF, (CSF-1), FLT3 and RET. until disease progression. The primary endpoints were overall The present trial examined the activity of sunitinib in 12 response rate (ORR; RANO criteria) after 8 weeks of sunitinib patients with newly diagnosed, non-resectable glioblastoma. and patient tolerance. Secondary endpoints were percentage Patients (≤75 years of age with performance status [PS] ≥2 of patients free of neurological deterioration pre-radiotherapy, and minimental status [MMS] ≥25) were treated post-biopsy percentage of patients completing radiotherapy, progression- free survival (PFS), overall survival (OS), and 1-year survival. A Simon 2-stage design (12 →20) based on ORR was applied * C. Balaña ( ) to calculate the number of patients needed to detect at least Medical Oncology Service, Catalan Institute of Oncology (ICO), α β Hospital Germans Trias i Pujol, Carretera Canyet sn, 10 % response with error of 0.05 and error of 0.10. The 08916 Badalona, Spain trial was closed because it did not meet minimal activity e-mail: [email protected] criteria. ORR was 0 % with only 1/12 patients (8.3 %) achiev- ing stable disease after sunitinib treatment. No patient showed M. J. Gil Medical Oncology Service, Catalan Institute of Oncology (ICO), reduction in gadolinium enhancement. The most frequent IDIBELL Hospital Duran I Reynalds, Gran Via s/n Km 2.7, G3/4 toxicities were fatigue (24.9 %) and diarrhea (16.6 %); 08907 L’Hospitalet, Barcelona, Spain one patient died of a CNS hemorrhage; 10/12 patients (83.3 %) deteriorated neurologically before radiation therapy; P. Perez – Medical Oncology Service, Hospital Clinico San Carlos, Calle median PFS was 7.7 weeks (95 % CI: 7.2 8.2); median OS Profesor Martin Lagos, s/n, 28040 Madrid, Spain was 12.8 weeks (95 % CI: 0.5–23.8 weeks); 1-year survival was 0 %. Sunitinib has no activity as monotherapy in glio- G. Reynes blastoma, and further investigation of its efficacy in this Medical Oncology Service, Hospital La Fe, Bulevar del Sur, 46026 Valencia, Spain setting is unwarranted.

O. Gallego Keywords Glioblastoma . Sunitinib . Radiotherapy Medical Oncology Service, Hospital de la Santa Creu i de Pau, Calle Sant Antoni M. Claret 167, 08025 Barcelona, Spain

T. Ribalta Introduction Department of Pathology, Hospital Clinic Provincial, Calle Villarroel 170, 08036 Barcelona, Spain The standard treatment of glioblastoma following surgical J. Capellades : S. Gonzalez resection is local irradiation with concomitant administration Radiology Service, Parc de Salut Mar., Hospital del Mar., Passeig of temozolomide and subsequent adjuvant temozolomide for Marítim 25-29, 08003 Barcelona, Spain 6 cycles. This treatment improved overall survival compared to treatment with irradiation alone (27.2 % vs. 10.9 % at E. Verger Radiotherapy Service, Hospital Clinic Provincial, Calle Villarroel 2 years, 16.0 % vs. 4.4 % at 3 years, 9.8 % vs. 1.9 % at 5 years; 170, 08036 Barcelona, Spain hazard ratio, 0.6; p<0.0001) [1]. However, in a sub-analysis Targ Oncol based on type of surgery, the differences in survival were not during and after radiotherapy. The principal objectives of the significant in non-resectable patients (median survival, 9.4 vs. study were objective response after 8 weeks of treatment in 7.9 months). One of every four patients with glioblastoma is patients without previous treatment, based on new RANO non-resectable, due to tumor site and the extent of the areas criteria [16] and patient tolerance to sunitinib. The secondary affected, and may undergo only diagnosis-confirming biopsy objectives were survival rate at 1 year, percentage of patients [2-4]. Although superior surgical techniques can improve the completing the treatment pre- and peri-radiotherapy, resection rate in operable patients, this will not benefit non- progression-free survival, and overall survival. resectable patients [5, 6], making them ideal candidates for inclusion in clinical trials of alternative therapies to temozolomide. Early data from neuro-oncology research groups like Patients and methods NABTT (New Approaches to Brain Tumor Therapy) show that investigational treatments prior to local irradiation do not Patient selection worsen survival and are the best method for evaluating the activity of antineoplastic agents [7]. The study was conducted by six hospitals of the Spanish Glioblastoma expresses vascular endothelial growth factor Neuro-Oncology Research Group (Grupo Español de (VEGF) as well as its receptors (VEGFR 1 and 2) and genes Investigación de Neurooncología [GEINO]). Criteria for pa- involved in angiogenesis (VEGFR2, PDFGR-α and Kit). tient inclusion into the study were: age ≤75 years, diagnosis Bevacizumab, a monoclonal antibody directed towards confirmed by biopsy and not amenable to surgical resection, VEGF, has demonstrated activity in glioblastoma and not having received previous chemotherapy or radiotherapy, is approved for second-line treatment in the USA and stable doses of dexamethasone over the week prior to inclu- some European countries. Cediranib is a multiple inhibitor of sion, performance status (PS) ≤2, minimental status (MMS) VEGFR1/2/3, PDFGR-α/β, and c-kit that normalizes brain ≥25/30, Barthel index (BI) >50 %, healed surgical wound, vascularization. This translates into a decrease in uptake of tumor measurable by magnetic resonance imaging (MRI) contrast medium in the TI-gadolinium sequences and appears performed within 3 weeks prior to initiating treatment (MRI to simulate a radiological response [8]. However, despite acceptable if prior to the biopsy), left ventricular ejection some promising results on radiological response and fraction (LVEF) >50 %, adequate bone marrow reserve progression-free survival at 6 months in a Phase II trial, no (neutrophils ≥2,000×109/l, platelets ≥100×109/l, hemoglobin improvement in outcomes were observed in a Phase III trial ≥10 g/dl), creatinine <1.5× upper limit of normal (ULN), compared to conventional treatment with lomustine [9]. This serum bilirrubin <1.5× ULN, SGOT and SGPT <2.5× ULN, is evidence of the difficulty inherent in evaluating new drugs alkaline phosphatase (Alk Phos) <3× ULN of the laboratory’s based on radiological images in patients with recurrent reference range, efficient contraception by the patient and disease. partner during treatment, informed consent. Like cediranib, sunitinib is an inhibitor of the receptors of Exclusion criteria were as follows: previous treatment for VEGF, PDGFR, and KIT, as well as FLT1, FLK1/KDR, FLT3 the glioblastoma (except diagnostic biopsy), any previous and the RET kinases [10], and is indicated for the treatment of infiltrating neoplasia within the last 5 years, severe cerebral renal carcinoma, gastrointestinal sarcoma and neuro- hemorrhage following the biopsy, anticonvulsant inducer/ endocrine tumors. In in vitro and in vivo orthotopic models inhibitor treatment of the CYP3A4 enzymes or treatment with in glioma cell lines, sunitinib produces anti-proliferative, other drugs that interact with the metabolism of the study drug antiangiogenic and anti-apoptosis effects and a decrease in and that could not be appropriately replaced with another drug the invasive capacity of the glioma cell lines implanted in the without possible interactions; pregnant or lactating women, animal brain [11, 12]. Furthermore, in combination with irra- active cardiovascular disease, hypertension not controlled by diation in murine models, sunitinib increases apoptosis and standard anti-hypertensive medications, unstable angina, con- decreases clonogenic survival and tumor volume in follow-up gestive heart disease (New York Heart Association [NYHA] [13, 14]. There have been isolated experiences of treatment grade), cardiac arrhythmia or prior myocardial infarction with a combination of sunitinib and irradiation that indicate <1 year prior to inclusion (treatment including therapeutic certain degree of synergism. An early Phase I trial in patients doses of coumarin-derived anti-coagulants). Low molecular with tumors of the central nervous system reported some weight heparin (LMWH) was permitted for the control of deep activity (13 % of partial responses and 60 % stable disease) vein thrombosis (DVT). Other exclusion criteria were and an acceptable toxicity profile for radiotherapy plus pulmonary thromboembolism, hemorrhagic diathesis or coag- sunitinib and recommended further investigation in a ulopathy, non-healing surgical wound, any type of ulcers, and Phase II trial [15]. recent bone fracture at the time of recruitment. In the present study, we have explored the activity of All the cases were reviewed before inclusion into the study sunitinib in patients with non-resectable glioblastoma before, by an independent pathologist (co-author: TR) to confirm the Targ Oncol diagnosis of glioblastoma. Serum samples were obtained be- chemistry, including evaluation of thyroid function, every fore treatment for potential study of circulating angiogenic 4 weeks. Physical examination, neurological status, BI, factors. The trial was approved by the Clinical Trials MMS and PS were assessed every 4 weeks. If there were Committee of the Spanish Medications Agency (EUDRACT: signs of clinical worsening before the start of radiotherapy, an 2008-006728-73) and is registered with clinicaltrials.gov MRI was performed, and the patient was allowed to leave the (NCT01100177). study and to continue conventional treatment at the discretion of the principal investigator. Treatment schedule MRI was performed at 4 and 8 weeks of radiotherapy completion. Follow-up of toxicity, neurological status, BI, The study was designed to evaluate the activity of sunitinib in MMS and PS was performed monthly and MRI was per- treatment-naïve patients prior to, during and after scheduled formed every 8 weeks, until disease progression. radiotherapy. Sunitinib at 37.5 mg/day was started a minimum of 15 days and no later than 28 days following the biopsy and Evaluation of response was continued with no break in the schedule. Continuous administration allows more flexibility and has been used The RANO criteria were adapted for the evaluation of successfully in renal cancer and gastrointestinal stromal sar- response taking into account the imaging results, the coma [17, 18]. Sunitinib dose was distributed in capsules of patient’s clinical neurological status, and the doses of 12.5 and 25 mg for oral administration. Prior prophylactic dexamethasone. Neurological progression of the disease anti-emetics were not required. was defined as irreversible neurological deterioration or Response was evaluated during the pre-radiotherapy phase worsening of PS of 1 grade. Disease progression was after 8 weeks of treatment. The patients then proceeded on to defined as a continued increase of dexamethasone use (for treatment with sunitinib at the same dose concomitantly >2 weeks) considered necessary for the prevention or reversal withlocalradiotherapy(60Gyin30sessionswitha of neurological deterioration without symptom remission. CTV:GTV visible tumor uptake in basal MRI with a margin of All MRI measurements were evaluated by the principal 2–3 cm).The treatment with sunitinib continued after the investigator and an independent assessor (co-authors: JC and concomitant phase until disease progression. SG), who were blinded with respect to clinical status. The Dexamethasone was permissible throughout the treatment sequence of the TI-gadolinium image measurements included at the lowest possible dose necessary for symptom control. In the greatest diameter and the maximum perpendicular to the case of any hematologic or non-hematologic toxicity grade ≥3 axial plane in the basal MRI. T2/FLAIR sequence changes (except lymphopenia), a decrease in sunitinib dose to were recorded as “+2” if the extension was clearly 25 mg/day was permitted following a maximum delay of greater, “+1” if the extension was greater than the 2 weeks to allow toxicity levels to return to grade ≤2. previous assessment but required to be assessed in various sections, “0” without change, “−1” in case of reduction and Evaluations during the study “−2” in case of clear reduction. Progression was also defined as the appearance of new hyper-intense foci in T2/FLAIR Toxicities were graded in the pre-radiotherapy, concomitant distinct from the initial lesion. and post-radiotherapy phases according to the National The RANO criteria were applied in the evaluation of re- Cancer Institute Common Terminology Criteria (Version 3.0; sponse. The date of progression was determined by progression http://www.ctep.cancer.gov). During the pre-radiotherapy and criteria (Table 1). If progression was observed in the first MRI concomitant phases, doses of dexamethasone and arterial following concomitant radiotherapy plus sunitinib, the progres- tension were recorded weekly, complete hematology sion needed to be confirmed in the subsequent MRI after assessments were performed every 2 weeks and serum 8 weeks so as to rule out any possibility of pseudo-progression.

Table 1 RANO response criteria Criteria CR PR SD PD

T1-Gd No Gd enhancement ≥50 ↓ >50 % ↓,<25%↑ ≥25 % ↑a T2/FLAIR =, ↓ =, ↓ =, ↓ =, ↑a a CR complete response, PR partial New lesions No No No Yes response, SD stable disease, Corticosteroids No =, ↓ =, ↓ =, ↑a PD a progressive disease, Neurological status =, better =,better =, better =, worse Gd gadolinium Conditions All All All Some a Any of these situations Targ Oncol

Statistical considerations Table 2 Baseline patient demographic and disease characteristics (n=12) n (%) The principal objective of the study was to determine activity in inducing tumor response, in accordance with the RANO Gender criteria. The Simon 2-phase design was used, with an alpha Male 7 (58.3) error of 0.05 and a beta error of 0.10. Based on responses Female 5 (41.7) achieved in previous studies, in which the treatment was Age (years) considered inactive if the response rate obtained was <10 %, Median 65 and with a maximum calculated objective response of 40 %, Range 48–70 the initial number of patients planned for recruitment in the Performance status first phase was 12. If one response was observed in the first 12 0–18(66.7) patients, the sample size was to be increased by 20 patients in 24(33.3) the second phase (a total of 32 patients). The treatment would Tumor sizea be considered effective if a minimum of four responses were Median 19.46 observed. Barthel index 90–100 8 (66.7) 60–70 4 (33.3) Results Mini mental status 30 2 (16.7) Between June 2009 and June 2011, 12 patients met the inclu- 25–29 8 (66.6) sion criteria and were recruited into the study. All patients give NFE 2 (16.7) their written informed consent. The baseline characteristics of Neurological symptoms the patients are summarized in Table 2. No 7 (58.3) Yes 5 (41.7) Treatment and dose intensity Dexamethasone No 2 (16.7) The median time-lapse between biopsy and initiation of treat- Yes 10 (83.3) ment was 24 days. All the patients commenced treatment with Anticonvulsant treatment sunitinib at a dose of 37.5 mg/day. In total, 106 weeks of No 10 (83.3) treatment were administered, with a median of 7 weeks per Yes (levetiracetam) 2 (16.7) patient (range: 1–26 weeks). In one patient, an administration error occurred and she received 112.5 mg/day for 4 days. The NFE not fully evaluable because of motor dysphasia patient suffered no additional toxicity but showed subsequent a Expressed in cm2 as the product of the two maximum diameters of the neurological progression, and she was included in the analyses tumor in T1Gd sequences of toxicity and response. In one patient, the dose had to be reduced to 25 mg/day in the fourth week of treatment due to (8.3 %). The most frequent toxicities were fatigue, in 8/12 grade 3 toxicity and subsequent clinical progression. Another patients (66.6 %), and stomatitis, in 7/12 patients (58.3 %). patient needed a dose reduction to 25 mg/day from week 9 of There was one case (8.3 %) of intra-ventricular hemorrhage the treatment due to grade 4 hypertriglyceridemia and which was the cause of death, and one case of grade 4 hypercholesterolemia. hypercholesterolemia and hypertriglyceridemia which was In seven patients, neurological worsening was observed, resolved with appropriate hypolipidemic therapy together and based on the results of their MRI, they did not complete with sunitinib dose reduction to 25 mg/d. Five of 12 patients the planned sunitinib treatment. Only five patients completed (41.6 %) had one or more ≥ grade 2 toxicities (Table 3). the pre-radiotherapy treatment with sunitinib (8 weeks), and only two patients (16.6 %) completed both the pre- Efficacy radiotherapy and the concomitant phases. No response to treatment was observed. The overall response Toxicity rate was 0 %, with one patient (8.3 %) with stable disease and 11 patients (91.6 %) with disease progression during treat- Sunitinib was relatively well tolerated by the patients, and the ment. There was radiological progression in 5/12 cases, neu- most frequent toxicities were grades 1–2. However, seven rological deterioration in 9/12, and necessary dexamethasone patients had grade 3 toxicities in the form of diarrhea dose increases in 2/12. The progression was neurological (16.6 %), stomatitis (8.3 %) and hand-foot syndrome alone in two patients, and follow-up MRI was not possible. Targ Oncol

Table 3 Most commonly reported treatment-related adverse events deterioration and an intra-ventricular hemorrhage, which was Adverse event Grade 1/2 Grade 3 Grade 4 Any grade the cause of death (Table 4: patient 8). Another patient con- n (%) n (%) n (%) (%) sidered clinically stable at the time of the evaluation and radiologically stable according to the centralized assessment Hematologic deteriorated rapidly and died 2 weeks after the MRI evaluation Anemia 1 (8.3) 8.3 (Table 4: patient 6). In neither of these cases was there a Platelets 3 (25.0) 25 decrease in the MRI uptake of contrast medium suggestive Neutropenia 1 (8.3) 8.3 of anti-angiogenic effect, and only one patient had a decrease Lymphopenia 5 (41.6) 1 (8.3) 49.9 in edema in the T2/FLAIR sequences (Table 4: patient 6). Non-hematologic Due to these results, the trial was closed after 12 patients Fatigue 5 (41.6) 2 (16.6) 1 (8.3) 58.3 were recruited, since the minimal activity to justify continued Stomatitis 6 (50.0) 1 (8.3) 58.3 recruitment (one objective response) had not been met. Study Hypopigmentation 5 (41.6) 41.6 of angiogenic factors on serum samples was not considered Diarrhea 1(8.3) 2(16.6) 25 meaningful due to the lack of activity of the drug. Hand–foot syndrome 2 (16.6) 1 (8.3) 25 Median progression free survival was 7.7 weeks (95 % CI: Constipation 2 (16.6) 16.6 7.2–8.2). Median survival was 12.8 weeks (95 % CI: 0.5– Dysphagia 2 (16.6) 16.6 23.8 weeks). Survival was 25 % at 6 months and 0 % at 1 year Epistaxis 1 (8.3) (Fig. 1). CNS hemorrhage 1 (8.3)a 8.3 Nausea and vomiting 1 (8.3) 8.3 Hypertension 1 (8.3) 8.3 Discussion Hypercholesterolemia 1 (8.3) 8.3 Hypertriglyceridemia Sunitinib inhibits not only PDFGRα and β and KIT, but also Abdominal pain 1 (8.3) 8.3 VEGFR1 and 2, all of which are expressed in glioblastoma [19-21]. Sunitinib activity has been observed in in vitro and CNS central nervous system in vivo models, and a Phase I trial reported activity and an a The patient died due to the central CNS hemorrhage, so it was graded 5. acceptable toxicity profile for radiotherapy plus sunitinib [11, It was considered unrelated to sunitinib by the PI 12, 14, 15]. In our study, however, sunitinib did not induce radiological response or neurological improvement. Nine pa- One patient with stable MRI following response evaluation tients (75 %) worsened neurologically, five (41.6 %) showed had intra-ventricular hemorrhage at 48 h post-evaluation, disease progression on the MRI, and two (16.6 %) needed an which was the cause of death. Ten patients (83.3 %) had increased dexamethasone dose. All patients had deep tumors neurological deterioration prior to radiotherapy. After leaving in eloquent areas, one patient had a multifocal tumor, and two the study, 41.6 % of patients received local irradiation therapy: had subependymal infiltration, which may partly explain these two received temozolomide plus radiotherapy, two received poor results. sunitinib plus radiotherapy, and one received radiotherapy The design of the present study was based on the Brada alone. The remaining seven patients received only support et al. [22] study, which treated 185 biopsied malignant glioma measures. patients with two cycles of temozolomide prior to local irra- Table 4 summarizes the type of response according to the diation. For the 137 glioblastoma patients included in the RANO criteria, the principal investigator’s evaluation, and the cohort, overall response rate was 20 % (95 % CI: 11–30 %), centralized radiological assessment. The centralized radiolog- progression-free survival prior to irradiation was 63 % (95 % ical assessment was conducted according to the strict criteria CI: 54–71 %), and median survival was 9 months. One-year of “product of the two perpendicular diameters” in the same survival for all patients was 41 % [22]. Other investigators section of the T1Gd sequence and showed slightly improved have reported similar results, [23] and the data on temozolo- results over that conducted by the principal investigator; five mide provide a benchmark for the assessment of other agents. of ten patients were considered as having stable disease since Temozolomide showed superiority to the nitrosureas in early they did not fulfill the radiological assessment criterion of an studies in relapsed glioblastoma patients [24]andisstillthe “increase of 25 % in the product of the two perpendicular only drug that has demonstrated benefit as first-line treatment. diameters of the target lesion”. However, in the majority of In the present study, even though response was evaluated using cases, a slight increase in the lesion was observed, which was different criteria, overall response was 0 %, progression-free sufficient to explain the neurological deterioration of the pa- survival prior to irradiation was 7.4 %, median progression- tient. In one case, there was a decrease of 21.45 % of the free survival was 7.7 weeks, and 1-year survival was 0 %. product-of-the-perpendicular-diameters but with neurological These findings lead us to conclude that sunitinib is not Targ Oncol

Table 4 Patient response according to the centralized radiological evaluation, RANO criteria and the principal investigator (PI)

Centralized radiological assessment RANO by PI Response evaluation by PI

Pt Basal product Product of diameters Change (%) T2/FLAIR MRI response MRI Neurological Dexamethasone of diameters at first follow-up status dose modification

1 19.97 31.46 +68.69 0 PD PD PD WORSE + 2 21.7 22.32 +2.86 0 SD PD SD WORSE + 3 15.38 25.20 +63.89 −1PDPDPDSD< 4 7.07 8.09 +14.4 +2 PD FLAIR PD SD WORSE < 529.34 – PD – WORSE < 6 20.83 19.98 −4.08 −2SDPDPDSD< 77.37 – PD – WORSE = 8 34.33 26.97 −21.45 −1 SD PD SD WORSE = 9 18.96 19.53 +2.98 +1 SD PD SD WORSE = 10 7.38 17.72 +140 +2 PD PD PD WORSE = 11 17.03 18.64 +9.45 NE SD PD PD WORSE = 12 20.71 17.56 −15.19 −1SDSDSDSD=

Patients 5 and 7 could not be evaluated with MRI due to neurological impairment. Patient 6 died 10 days after MRI due to neurological progression; a CT scan ruled out hemorrhage as a complication. Since the principal investigator classified the patient as having PD, this has not been reported as SD although the central assessment showed stable disease with a minimal reduction of the tumor diameter product Patient 8 was reported as PD by the principal investigator at the time of the MRI. The central assessor found a reduction in 21.45 % of the product of the two diameters, but the patient deteriorated rapidly after evaluation and a CT scan showed a CNS hemorrhage with intraventricular involvement that was the final cause of death Patients 2, 4, 9 and 11 were evaluated as SD by the central assessment due to minimal increments ≤25 %, but they had neurological progression PI principal investigator, SD stable disease, PD disease progression, Dex dexamethasone, NE not evaluable comparable to temozolomide as monotherapy in the treatment anti-angiogenic profile [25]. These preclinical results indicate of glioblastoma. that a combination of sunitinib and standard therapy for Nevertheless, our findings do not rule out the possibility MGMT-positive tumors may be a potential alternative strategy that sunitinib may well have a role in combination therapy. In for non-methylated glioblastoma patients. vitro studies have revealed that sunitinib preferentially inhibits In the present study, two patients needed to have a dose the proliferation and survival of cells that express O6-methyl- reduction to 25 mg/day due to toxicity. The toxicity profile guanine-methyltransferase (MGMT), leading to an increase in was similar to that reported in patients with other tumors the soluble VEGFR-1/VEGFA ratio and a shift towards an treatedwithcontinuoussunitinib,especiallyinrelationtothe

PFS: 7.7 weeks OS: 12.8 weeks 95%CI: 7.2-8.2 95%CI: 0.5-23.8 Overall survival probability Progression free survival probability free survival Progression

weeks weeks Fig. 1 Progression free survival (PFS) and overall survival (OS) Targ Oncol most commonly reported adverse effects of stomatitis, response is simpler in treatment-naïve patients since this asthenia, hand–foot syndrome and diarrhea [17, 18]. Since avoids confounding factors secondary to the irradiation which five patients (41.6 %) had grade ≥3 toxicities, we ruled out the can complicate the evaluation of response based on the image, possibility of increasing the dose of sunitinib in order to including “pseudo-progression” [33]. In addition, the thera- increase the availability of the drug to the central nervous peutic targets of the drug under study can be evaluated in the system. primary tumor just prior to treatment, in contrast to trials in Our study was not accompanied by a pharmacokinetic relapsed patients in which it is difficult to obtain a biopsy prior evaluation. However, since sunitinib is metabolized in the to initiating the treatment under trial. Further investigation of liver via the cytochrome P450-3 A4 (CYP3A4), it cannot be an apparently active drug based on the results of a trial in administered either with inhibitors or inducers of CYP3A4. relapsed patients has resulted in a not inconsiderable number For this reason, such treatment was an exclusion criterion in of failures in the past few years, including trials with the trial protocol so as to avoid any interaction with the cediranib, imatinib, erlotinib and enzastaurine [9, 34-36]. metabolism of sunitinib at therapeutic doses for other tumors. The economic cost of these failures has been high, and no In the initial study with cediranib in relapsed patients and in new drug has been identified in the past 15 years, with the a later study with bevacizumab, an alleviation of edema exception of temozolomide and bevacizumab. secondary to the normalization of the vascularization was In the present study, we have demonstrated the inefficacy observed [8, 26]. In fact, as soon as 24 h after the first of sunitinib to restrict tumor growth. The absence of objective administration of cediranib, a decrease in contrast uptake with responses, together with clear neurological progression, was TI-gadolinium and in infiltration in the T2/FLAIR images sufficient grounds to discontinue the trial. We, therefore, could be seen, which enabled the dose of corticoids to be conclude that sunitinib is ineffective as monotherapy in the decreased, although subsequent clinical trials did not confirm treatment of glioblastoma and that further investigation of its cediranib efficacy. Although cediranib shares a mechanism of efficacy is not warranted in this setting. action with sunitinib, none of our patients showed a decrease in contrast uptake, and in only one case was there a decrease in Acknowledgements The study was made possible by a grant from infiltration in the T2/FLAIR images. Pfizer who provided the drug sunitinib and logistical support for In two recent Phase II studies with sunitinib in relapsed adequate monitoring of the trial. We thank Anna del Prado from MFAR (Marketing Farmacéutico & Investigación Clínica, S.L.) for her expert patients, no activity was observed in any of the patients [27, technical assistance. Statistical analyses were performed by Juan José de la 28], although there was a decrease in the dynamic suscepti- Cruz Troca of the Department of Preventive Medicine and Public Health, bility contrast in four of 14 patients (29 %), evaluated with Faculty of Medicine, Autonomous University of Madrid. perfusion techniques [27]. In the centralized radiological assessment, response criteria Conflict of interest C.B. declares no conflict of interest. M.J. declares no conflict of interest. P.P. declares no conflict of interest. G.R. declares were based on the product of the two major perpendicular no conflict of interest. O.G. declares no conflict of interest. T.R. declares diameters of the target lesion, as specified in the RANO no conflict of interest. She received a grant from GEINO supported by criteria. Some patients were classified as having stable disease Pfizer for her work on centralized review of pathological samples. J.C. because they did not have an increase of >25 % in the size of declares no conflict of interest. He received a Grant from GEINO supported by Pfizer for his work on centralized review of images. S.G. the lesion (patients 2, 8, 9 and 11; Table 4). However, an declares no conflict of interest. E.V. declares no conflict of interest. increase of <25 % predisposes to a substantial neurological worsening, and these patients had clinical progression. These findings suggest that allowing an increase of nearly 25 % in References the may well be too generous in the context of brain tumors, as patients can deteriorate neurologically with even smaller increments of size. 1. Stupp R, Hegi ME, Mason WP, van den Bent MJ, Taphoorn MJ, Exposing patients to drugs that are eventually considered Janzer RC, Ludwin SK, Allgeier A, Fisher B, Belanger K, Hau P, Brandes AA, Gijtenbeek J, Marosi C, Vecht CJ, Mokhtari K, inactive does not affect subsequent data on life expectancy, Wesseling P, Villa S, Eisenhauer E, Gorlia T, Weller M, but it does preclude further investigation of the drug in sub- Lacombe D, Cairncross JG, Mirimanoff RO (2009) Effects of radio- sequent phases of clinical trials, which are progressively more therapy with concomitant and adjuvant temozolomide versus radio- expensive and which require a greater number of patients. In therapy alone on survival in glioblastoma in a randomised phase III study: 5-year analysis of the EORTC-NCIC trial. Lancet Oncol 10(5): fact, the NABTT CNS Consortium ruled out several drugs 459–466 that were shown to be inactive in clinical studies of medical 2. Bauchet L, Mathieu-Daude H, Fabbro-Peray P, Rigau V, Fabbro M, treatment prior to irradiation [29, 30]. However, trials of novel Chinot O, Pallusseau L, Carnin C, Laine K, Schlama A, Thiebaut A, neoadjuvant therapies may be more useful in patients without Patru MC, Bauchet F, Lionnet M, Wager M, Faillot T, Taillandier L, Figarella-Branger D, Capelle L, Loiseau H, Frappaz D, Campello C, MGMT methylation, where the benefit of temozolomide is Kerr C, Duffau H, Reme-Saumon M, Tretarre B, Daures JP,Henin D, not as clear [31, 32]. Moreover, the evaluation of radiological Labrousse F, Menei P, Honnorat J, Societe Francaise de N, Targ Oncol

Club de Neuro-Oncologie of the Societe Francaise de N, Societe 16. Wen PY,Macdonald DR, Reardon DA, Cloughesy TF, Sorensen AG, Francaise de N, Association des Neuro-Oncologues d'Expression F Galanis E, Degroot J, Wick W, Gilbert MR, Lassman AB, Tsien C, (2010) Oncological patterns of care and outcome for 952 patients Mikkelsen T, Wong ET, Chamberlain MC, Stupp R, Lamborn KR, with newly diagnosed glioblastoma in 2004. Neuro Oncol 12(7): Vogelbaum MA, van den Bent MJ, Chang SM (2010) Updated 725–735. doi:10.1093/neuonc/noq030 response assessment criteria for high-grade gliomas: response 3. Balaña C, Capellades J, Teixidor P, Roussos I, Ballester R, Cuello M, assessment in neuro-oncology working group. J Clin Oncol Arellano A, Florensa R, Rosell R (2008) Clinical course of high- 28(11):1963–1972 grade glioma patients with a “biopsy-only” surgical approach: a need 17. Escudier B, Roigas J, Gillessen S, Harmenberg U, Srinivas S, Mulder for individualised treatment. Clin Transl Oncol 9(12):797–803. doi: SF, Fountzilas G, Peschel C, Flodgren P, Maneval EC, Chen I, 10.1007/s12094-007-0142-0 Vogelzang NJ (2009) Phase II study of sunitinib administered in a 4. Chang SM, Parney IF, Huang W, Anderson FA Jr, Asher AL, continuous once-daily dosing regimen in patients with cytokine- Bernstein M, Lillehei KO, Brem H, Berger MS, Laws ER (2005) refractory metastatic renal cell carcinoma. J Clin Oncol 27(25): Patterns of care for adults with newly diagnosed malignant glioma. 4068–4075 JAMA 293(5):557–564 18. George S, Blay JY, Casali PG, Le Cesne A, Stephenson P, Deprimo 5. Castellano A, Bello L, Michelozzi C, Gallucci M, Fava E, Iadanza A, SE, Harmon CS, Law CN, Morgan JA, Ray-Coquard I, Tassell V, Riva M, Casaceli G, Falini A (2012) Role of diffusion tensor Cohen DP, Demetri GD (2009) Clinical evaluation of continuous magnetic resonance tractography in predicting the extent of resection daily dosing of sunitinib malate in patients with advanced gastroin- in glioma surgery. Neuro Oncol 14(2):192–202 testinal stromal tumour after imatinib failure. Eur J Cancer 45(11): 6. Kubben PL, ter Meulen KJ, Schijns OE, ter Laak-Poort MP, van 1959–1968 Overbeeke JJ, van Santbrink H (2011) Intraoperative MRI-guided 19. Nobusawa S, Stawski R, Kim YH, Nakazato Y, Ohgaki H (2011) resection of glioblastoma multiforme: a systematic review. Lancet Amplification of the PDGFRA, KITand KDR genes in glioblastoma: Oncol 12(11):1062–1070 a population-based study. Neuropathology 31(6):583–588 7. Grossman SA, Carson K, Piantadosi S, Fisher J, The NABTT CNS 20. Puputti M, Tynninen O, Sihto H, Blom T, Maenpaa H, Isola J, Paetau Consortium, Baltimore, MD (2002) Survival of adults with newly A, Joensuu H, Nupponen NN (2006) Amplification of KIT, diagnosed glioblastoma multiforme (GBM) treated with experimen- PDGFRA, VEGFR2, and EGFR in gliomas. Mol Cancer Res tal agents and delayed radiation therapy (RT) versus experimental 4(12):927–934 agents and immediate radiation: the experience of the NABTT CNS 21. Plate KH, Risau W (1995) Angiogenesis in malignant gliomas. Glia consortium. J Clin Oncol Proc Am Soc Clin Oncol (21):Abstr 282 15(3):339–347 8. Batchelor TT, Sorensen AG, di Tomaso E, Zhang WT, Duda DG, 22. Brada M, Ashley S, Dowe A, Gonsalves A, Huchet A, Pesce G, Reni Cohen KS, Kozak KR, Cahill DP, Chen PJ, Zhu M, Ancukiewicz M, M, Saran F, Wharram B, Wilkins M, Wilkins P (2005) Neoadjuvant Mrugala MM, Plotkin S, Drappatz J, Louis DN, Ivy P, Scadden DT, phase II multicentre study of new agents in patients with malignant Benner T, Loeffler JS, Wen PY, Jain RK (2007) AZD2171, a glioma after minimal surgery. Report of a cohort of 187 patients pan-VEGF receptor tyrosine kinase inhibitor, normalizes tumor treated with temozolomide. Ann Oncol 16(6):942–949 vasculature and alleviates edema in glioblastoma patients. Cancer 23. Chinot OL, Barrie M, Fuentes S, Eudes N, Lancelot S, Metellus P, Cell 11(1):83–95 Muracciole X, Braguer D, OuafikL,MartinPM,DufourH, 9. Batchelor TT, Mulholland P, Neyns B, Nabors LB, Campone M, Figarella-Branger D (2007) Correlation between O6-methylguanine- Wick A, Mason W, Mikkelsen T, Phuphanich S, Ashby LS, Degroot DNA methyltransferase and survival in inoperable newly diagnosed J, Gattamaneni R, Cher L, Rosenthal M, Payer F, Jurgensmeier JM, glioblastoma patients treated with neoadjuvant temozolomide. J Clin Jain RK, Sorensen AG, Xu J, Liu Q, van den Bent M (2013) Phase III Oncol 25(12):1470–1475 randomized trial comparing the efficacy of cediranib as monotherapy, 24. Yung WK, Albright RE, Olson J, Fredericks R, Fink K, Prados MD, and in combination with lomustine, versus lomustine alone in pa- Brada M, Spence A, Hohl RJ, Shapiro W, Glantz M, Greenberg H, tients with recurrent glioblastoma. J Clin Oncol 31(26):3212–3218. Selker RG, Vick NA, Rampling R, Friedman H, Phillips P, doi:10.1200/JCO.2012.47.2464 Bruner J, Yue N, Osoba D, Zaknoen S, Levin VA (2000) A 10. Chow LQ, Eckhardt SG (2007) Sunitinib: from rational design to phase II study of temozolomide vs. procarbazine in patients clinical efficacy. J Clin Oncol 25(7):884–896 with glioblastoma multiforme at first relapse. Br J Cancer 83(5): 11. de Bouard S, Herlin P, Christensen JG, Lemoisson E, Gauduchon P, 588–593 Raymond E, Guillamo JS (2007) Antiangiogenic and anti-invasive 25. Chahal M, Xu Y,Lesniak D, Graham K, Famulski K, Christensen JG, effects of sunitinib on experimental human glioblastoma. Neuro Aghi M, Jacques A, Murray D, Sabri S, Abdulkarim B (2010) Oncol 9(4):412–423 MGMT modulates glioblastoma angiogenesis and response to the 12. Giannopoulou E, Dimitropoulos K, Argyriou AA, Koutras AK, tyrosine kinase inhibitor sunitinib. Neuro Oncol 12(8):822–833 Dimitrakopoulos F, Kalofonos HP (2010) An in vitro study, evaluat- 26. Ananthnarayan S, Bahng J, Roring J, Nghiemphu P, Lai A, ing the effect of sunitinib and/or lapatinib on two glioma cell lines. Cloughesy T, Pope WB (2008) Time course of imaging changes of Invest New Drugs 28(5):554–560 GBM during extended bevacizumab treatment. J Neurooncol 88(3): 13. Schueneman AJ, Himmelfarb E, Geng L, Tan J, Donnelly E, Mendel 339–347 D, McMahon G, Hallahan DE (2003) SU11248 maintenance therapy 27. Neyns B, Sadones J, Chaskis C, Dujardin M, Everaert H, Lv S, prevents tumor regrowth after fractionated irradiation of murine Duerinck J, Tynninen O, Nupponen N, Michotte A, De Greve J tumor models. Cancer Res 63(14):4009–4016 (2011) Phase II study of sunitinib malate in patients with recurrent 14. Zhang HP, Takayama K, Su B, Jiao XD, Li R, Wang JJ (2011) Effect high-grade glioma. J Neurooncol 103(3):491–501 of sunitinib combined with ionizing radiation on endothelial 28. Kreisl TN, Smith P, Sul J, Salgado C, Iwamoto FM, Shih JH, cells. J Radiat Res 52(1):1–8 Fine HA (2013) Continuous daily sunitinib for recurrent glioblasto- 15. Wuthrick EJ, Kamrava M, Curran WJ Jr, Werner-Wasik M, ma. J Neurooncol 111(1):41–48. doi:10.1007/s11060-012-0988-z Camphausen KA, Hyslop T, Axelrod R, Andrews DW, Glass J, 29. Fetell MR, Grossman SA, Fisher JD, Erlanger B, Rowinsky E, Machtay M, Dicker AP (2011) A phase 1b trial of the combination Stockel J, Piantadosi S (1997) Preirradiation paclitaxel in glioblastoma of the antiangiogenic agent sunitinib and radiation therapy for multiforme: efficacy, pharmacology, and drug interactions. New patients with primary and metastatic central nervous system malig- Approaches to Brain Tumor Therapy Central Nervous System nancies. Cancer 117(24):5548–5559 Consortium. J Clin Oncol 15(9):3121–3128 Targ Oncol

30. Hochberg F, Grossman SA, Mikkelsen T, Glantz M, Fisher JD, glioma patients treated with chemoirradiation with temozolomide. Piantadosi S (2000) Lack of efficacy of 9-aminocamptothecin in adults Cancer 113(2):405–410 with newly diagnosed glioblastoma multiforme and recurrent high-grade 34. van den Bent MJ, Brandes AA, Rampling R, Kouwenhoven MC, astrocytoma. NABTT CNS Consortium. Neuro Oncol 2(1):29–33 Kros JM, Carpentier AF, Clement PM, Frenay M, Campone M, 31. Wick W, Platten M, Meisner C, Felsberg J, Tabatabai G, Simon M, Baurain JF, Armand JP, Taphoorn MJ, Tosoni A, Kletzl H, Nikkhah G, Papsdorf K, Steinbach JP, Sabel M, Combs SE, Vesper J, Klughammer B, Lacombe D, Gorlia T (2009) Randomized phase II Braun C, Meixensberger J, Ketter R, Mayer-Steinacker R, trial of erlotinib versus temozolomide or carmustine in recurrent Reifenberger G, Weller M (2012) Temozolomide chemotherapy glioblastoma: EORTC brain tumor group study 26034. J Clin alone versus radiotherapy alone for malignant astrocytoma in the Oncol 27(8):1268–1274 elderly: the NOA-08 randomised, phase 3 trial. Lancet Oncol 13(7): 35. Dresemann G, Weller M, Rosenthal MA, Wedding U, Wagner W, 707–715. doi:10.1016/S1470-2045(12)70164-X Engel E, Heinrich B, Mayer-Steinacker R, Karup-Hansen A, Fluge 32. Malmstrom A, Gronberg BH, Marosi C, Stupp R, Frappaz D, O, Nowak A, Mehdorn M, Schleyer E, Krex D, Olver IN, Steinbach Schultz H, Abacioglu U, Tavelin B, Lhermitte B, Hegi ME, JP, Hosius C, Sieder C, Sorenson G, Parker R, Nikolova Z (2010) Rosell J, Henriksson R (2012) Temozolomide versus standard Imatinib in combination with hydroxyurea versus hydroxyurea alone 6-week radiotherapy versus hypofractionated radiotherapy in as oral therapy in patients with progressive pretreated glioblastoma patients older than 60 years with glioblastoma: the Nordic resistant to standard dose temozolomide. J Neurooncol 96(3):393–402 randomised, phase 3 trial. Lancet Oncol 13(9):916–926. doi: 36. Wick W, Puduvalli VK, Chamberlain MC, van den Bent MJ, 10.1016/S1470-2045(12)70265-6 Carpentier AF, Cher LM, Mason W, Weller M, Hong S, Musib L, 33. Taal W, Brandsma D, de Bruin HG, Bromberg JE, Swaak- Liepa AM, Thornton DE, Fine HA (2010) Phase III study of Kragten AT, Smitt PA, van Es CA, van den Bent MJ (2008) enzastaurin compared with lomustine in the treatment of recurrent Incidence of early pseudo-progression in a cohort of malignant intracranial glioblastoma. J Clin Oncol 28(7):1168–1174