Original Study

A Phase II Trial of Sunitinib in Patients With Renal Cell Cancer and Untreated Brain Metastases Christine Chevreau,1 Alain Ravaud,2 Bernard Escudier,3 Eric Amela,4 Remy Delva,5 Frederic Rolland,6 Diego Tosi,7 Stephane Oudard,8 Ellen Blanc,9 Celine Ferlay,9 Sylvie Négrier9,10 for the French Group on Renal Cancer Abstract In this prospective trial, the efficacy and safety of sunitinib in patients with previously untreated brain metastases (BM) in metastatic renal cell cancer (RCC) were evaluated. Safety appears to be acceptable, however, its efficacy is limited, with no objective response and poor median overall survival. Background: The expanded access program and anecdotal cases suggested sunitinib is safe in RCC patients with BM and might have worthwhile activity. Patients and Methods: In a phase II trial, patients with untreated BM received the standard regimen of sunitinib. The primary end point was objective response (OR) rate in BM after 2 cycles. An OR rate of 35% was prospectively defined as the minimum needed to warrant further investigation. According to Simon’s optimal 2-stage design, at least 3 of the initial 15 patients had to have an OR for accrual to continue. Results: Among 16 evaluable patients, 1 had a complete response outside the central nervous system (CNS). CNS disease was stabilized in 5 (31%). However, no BM showed an OR. Therefore, no further accrual took place. Median time to progression was 2.3 months and overall survival was 6.3 months. There was 1 toxic death, from peritonitis with gastric perforation. Three patients experienced at least 1 treatment-related grade 3 or greater toxicity but no neurological adverse events were attributable to sunitinib. Conclusion: Although tolerability was acceptable in RCC patients with previously untreated BM, sunitinib has limited efficacy in this setting. Clinical Genitourinary Cancer, Vol. -, No. -, --- ª 2013 Elsevier Inc. All rights reserved. Keywords: Brain metastases, Clinical trial, Renal cell carcinoma, Sunitinib

Introduction Although there is wide variation in the reported incidence of brain metastases (BM) in patients with renal cell cancer (RCC), with figures ranging from 4% to 48% cited, central nervous system (CNS) involvement is undoubtedly frequent1 and in a recent European registry, RCC was second only to lung cancer as a source of cerebral metastases. In a series of more than 11,000 metastatic RCC (mRCC) patients from the Nationwide Inpatient Sample (NIS) in 1

Institut Claudius Regaud, Toulouse, France Centre Hospitalier Universitaire Saint André, Bordeaux, France Institut Gustave Roussy, Villejuif, France 4 Centre Oscar Lambret, Lille, France 5 Institut de Cancérologie de l’Ouest, Angers, France 6 Institut de Cancérologie de l’Ouest, Saint Herblain, France 7 Institut regional de Cancérologie de Montpellier, Montpellier, France 8 Hôpital Européen Georges Pompidou, University Paris Descarte, Paris, France 9 Centre Leon Bérard, Lyon, France 10 Université Lyon, Lyon, France 2 3

Submitted: Jun 6, 2013; Revised: Sep 12, 2013; Accepted: Sep 24, 2013 Address for correspondence: Christine Chevreau, MD, Medical Oncology Department, Institut Claudius Regaud, 20-24, rue du Pont Saint-Pierre, 31052 Toulouse cedex, France E-mail contact: [email protected]

1558-7673/$ - see frontmatter ª 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clgc.2013.09.008

the United States, the rate of BM in patients with exclusively abdominal metastases was 2%, but that in patients with thoracic and bone metastases was 16%.2 In the overall NIS population, 8% of patients were affected. The effective detection of BM and—if surgical and radiotherapy options are exhausted—effective systemic management is a major unmet medical need. Sunitinib (Sutent, Laboratoire Pfizer) is an oral tyrosine kinase inhibitor (TKI) which selectively blocks certain proangiogenic growth factors involved in mRCC. Among its targets are the vascular endothelial growth factor receptor (VEGFR) types 1-3, and the platelet derived growth factor receptor-a and -b. Sunitinib has proven efficacy in the first-line treatment of patients with metastatic clear-cell RCC, having been shown to more than double median progression-free survival (PFS) when compared with interferon alfa in the pivotal phase III trial.3 Sunitinib remains a standard of care in this setting.4 The TKI sorafenib is also a VEGFR inhibitor. As with sunitinib, sorafenib is a small molecule with a wide distribution in tissues including the CNS. In the pivotal phase III trial of sorafenib in mRCC, patients randomized to sorafenib were less likely to develop BM (3%) than those in the placebo arm (12%).5 However, in the

Clinical Genitourinary Cancer Month 2013

-1

A Phase II Trial of Sunitinib sorafenib expanded access program, among the 47 patients with BM evaluable for efficacy, a partial response was seen in only 2 (4%).6 Considering their poor prognosis, patients with known BM were excluded from the pivotal trial of sunitinib in RCC.3 In the expanded access program, in contrast, 321 patients known to have had CNS lesions at baseline were included and received a median of 3 cycles of treatment.7 Thirty-two percent of these patients with BM discontinued for lack of efficacy, and 8% because of adverse events. However, sunitinib appeared to be safe in patients with asymptomatic or previously treated BM; and cases of tumour regression were reported. Twelve percent of the 213 evaluable patients had an objective response. Median PFS was 5.6 months and median overall survival (OS) 9.2 months. Complete response in an mRCC patient with BM has also been described.8 Based on these retrospective data, sunitinib has been suggested as a good therapeutic option for RCC patients with BM. However, in the conclusion to their report on BM patients in the expanded access program, Gore et al recommended that the potential activity of sunitinib in this setting should be studied prospectively.7 It was, therefore, appropriate to carry out this controlled, prospective phase II study to assess the potential activity of sunitinib against CNS metastases from RCC.

Patients and Methods Patients Male and female patients aged 18 years or older with measurable (more than 2 cm) and inoperable BM from renal adenocarcinoma of any histology were eligible for this multicenter phase II study. Cytological or histological confirmation of RCC, including Fuhrman grade, was mandatory. To be enrolled, patients had to have an Eastern Cooperative Oncology Group Performance Status of 2 or less and adequate organ function. They had not previously been exposed to sunitinib (at least not within the 6 months before study entry) and had not been treated for BM. Patients had to be asymptomatic or have symptoms adequately controlled with steroid treatment for at least 2 weeks. Exclusion criteria included cerebral metastasis presenting as hemorrhage, presence of an isolated BM of less than 2 cm amenable to surgery or radiosurgery, previous treatment with growth factors, and uncontrolled hypertension. The study was approved by local Institutional Review Boards. All patients were fully informed and gave written consent.

Study Treatment Sunitinib was administered orally at 50 mg daily for 4 weeks followed by 2 weeks of no drug use (ie, according to a 6-week cycle). Treatment continued until either disease progression or intolerable toxicity. In individual patients, the dose was reduced to 37.5 mg or 25 mg daily when required by the nature and severity of toxicity. No other anticancer treatment was allowed for the duration of the patients’ participation in the study.

End Points

2

-

The main objective of the study was to determine the objective response rate (ORR), including complete and partial responses, in cerebral metastases after 2 cycles of sunitinib, ie, 10-12 weeks after the start of treatment. Secondary end points were response duration, ORR for lesions outside of the CNS, disease stabilization, time to

Clinical Genitourinary Cancer Month 2013

Table 1 Patient Characteristics Characteristic

Value

Median Age, Years Sex, Male More Than 1 Metastatic Site Previous Nephrectomy Clear-Cell Carcinoma ECOG PS 0-1 MSKCC Intermediate or Poor Risk Corticosteroid Therapy Median Number of CNS Metastases (Range) Median Sum of Diameters (Range), mm

13 10 7 16 14 10 12 1 23

62 (75) (59) (41) (94) (88)a (77)b (75) (1-4) (10-61)

Data are presented as n (%) except where otherwise noted. n ¼ 17 patients. Abbreviations: CNS ¼ central nervous system; ECOG PS ¼ Eastern Cooperative Oncology Group Performance Status; MSKCC ¼ Memorial Sloan-Kettering Cancer Center. a One missing value. b Four missing values.

Table 2 Best Response According to RECIST Criteria Site Brain Other Sitesa

Best Response

n

SD CR SD

5 1b 5c

Abbreviation: RECIST ¼ Response Evaluation Criteria In Solid Tumors. a The other sites were: pulmonary metastases (12 patients), liver metastases (2 patients), bone (5 patients), kidney (1 patient), abdominal lymph node (3 patients), adrenal gland (4 patients), mediastinum (8 patients), pleura (1 patient), skin (1 patient), and others (2 patients). b Cutaneous metastases. c Pulmonary metastases, bone, kidney, adrenal gland, and abdominal lymph nodes.

progression (TTP), PFS, OS, the course of tumor-related neurological symptoms, and overall tolerability of treatment. A further objective was to determine whether any factors predictive of response could be identified at initial assessment or in the first 2 weeks of therapy. Brain metastases were assessed using magnetic resonance imaging (MRI) every 2 cycles and response classified according to Response Evaluation Criteria In Solid Tumors (RECIST) criteria by a central review committee. Response duration was calculated from the date of first response until the occurrence of documented progression or death. TTP was calculated as the time from inclusion in the study until progression or death from progressive RCC, and censored at the date of last assessment or death from a cause other than cancer. PFS was calculated as the time from inclusion until the date of first evidence of progression or death or date of last assessment. OS was calculated as the time from date of inclusion until death from any cause or last follow-up. The course of symptomatic neurological symptoms was assessed according to the use of corticosteroids. Toxicity was graded according to the National Cancer InstituteCommon Terminology Criteria classification (version v3.0). Particular attention was paid to the possible risk of acute, treatment-related neurological toxicities of grade 3 or greater severity, the exacerbation of existing neurological deficits, the appearance of new signs that were not rapidly reversible, or the need to increase the dose of corticosteroids. Neurological toxicity was judged in relation to the frequency of such events expected on the basis of our clinical experience. Patients

Christine Chevreau et al Figure 1 Time to Progression

were assessed clinically and standard hematological and laboratory values monitored on days 14 and 28 of the first cycle, and then during the fifth week of each subsequent cycle.

Two-Stage Trial Design An ORR of 35% or greater in BM was prospectively defined as the minimum needed to warrant further investigation of sunitinib in this setting. An ORR of 15% or less was regarded as of little interest. Using Simon’s optimal 2-stage design,9 the total number of evaluable patients required was established as 28. This number was sufficient to detect the ORR of interest with a 5% risk of alpha error and a power of 80% (on a minimax design). Among the 15 patients enrolled in the first stage, at least 3 patients had to have an objective response for accrual to continue into stage 2. Statistical analysis was undertaken by the Statistics and Therapeutic Evaluation Unit of the Centre Léon Bérard, Lyon, France.

Results The clinical and disease characteristics of the 17 patients enrolled between April 2009 and January 2011 are shown in Table 1. Patients were predominantly male, and all but 1 had clear-cell RCC. Most (77%; 10 of the 13 patients for whom data were obtained) were of poor or intermediate risk and 75% (12 patients) were taking corticosteroid therapy. In 59%, there was more than 1 site of metastatic disease but the median number of BM was 1 (range, 1-4). One patient died before the start of treatment from a cerebral hemorrhage. Sixteen patients were evaluable after 2 cycles of sunitinib.

Efficacy One patient (6%) had a complete response outside the CNS. However, no objective responses were observed in BM. Disease stabilization was the best response and seen in 5 patients (31%)

Figure 2 Overall Survival

Clinical Genitourinary Cancer Month 2013

-3

A Phase II Trial of Sunitinib Table 3 Description of Grade ‡ 3 Events Type

n

All General Alteration of Status Anorexia Epilepsy Diabetic Decompensation Pulmonary Embolism Aphasia Asthenia Digestive Hemorrhage Dyspnea Gastric Perforation Hypertension Intracranial hypertension Intraventricular Hemorrhage Left Upper Limb Motor Impairment Left Pleural Effusion Left Pulmonary Abscess Right Superior Limb Pains Lower Limb Weakness Peritonitis Rectal Bleeding Right Hemiparesis Septic Shock Neutropenia Thrombopenia Balance Disorders

14 4 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

(Table 2). Median TTP was 2.3 months (95% confidence interval [CI], 1.2-5.4 months; Fig. 1). Median OS was 6.3 months (95% CI, 2.1-7.9 months; Fig. 2).

Safety Fourteen patients experienced a total of 32 adverse events of grade 3 or greater (Table 3). There was 1 toxic death, from peritonitis with gastric perforation. Three patients had more than 1 grade  3 toxicity related to treatment. These included pulmonary embolism, neutropenia, thrombopenia, and hypertension. In addition, 1 patient had grade 3 asthenia, which was thought possibly related to sunitinib. The other toxicities were those commonly observed with sunitinib use. However, although several neurological events were observed, none were attributable specifically to sunitinib and there were no cases of cerebral hemorrhage. Of the 17 patients enrolled, sunitinib dose was reduced in 3 because of moderate toxicity. Three patients discontinued treatment; the reasons were peritonitis with gastric perforation and septic shock, pulmonary embolism, and general alteration of status.

assessment might have been inconsistent.5 In this study, response was assessed centrally according to RECIST criteria using MRI. Sunitinib appears to have an acceptable toxicity profile in patients with previously untreated BM from RCC. We found no evidence of treatment-associated cerebral hemorrhage, which had been suggested as a possible adverse event in patients with BM treated with VEGF inhibitors. However, no objective responses were seen in CNS lesions, and the median OS of 6.3 months was poor compared with the pivotal trial (which excluded BM patients) and the expanded access study (which did not). Fatigue was no more frequent than expected, and we do not believe that the absence of efficacy is attributable to a high rate of treatment-associated toxicity and consequent dose reduction and discontinuations. The lack of activity against cerebral metastases, coupled with absence of neurotoxicity, suggest that sunitinib might be less present in brain than previously thought. Poor penetration, or increased drug metabolism because of the effect of inducers such as glucocorticoids (most of our patients were taking such therapy), might be part of the explanation. Typically, mRCC patients with BM are not entered into clinical trials because they are regarded as having a relatively high risk of adverse events and generally poor prognosis. However, such patients are a major element in the everyday population of RCC patients with advanced disease, and any suggestions as to how they can be more effectively managed would be helpful. The idea that sunitinib might be active in this setting, supported by the expanded access study, was therefore welcomed and justified this prospective trial. Since the prospective study was designed and completed, data on 2 retrospective series of RCC patients with BM have been published, both from the M.D. Anderson Cancer Center. Verma et al compared the outcome in 41 patients from the pre-TKI era with 40 patients who received TKIs.10 Their data are suggestive of improved survival in the latter group. However, such secular trends must be treated with caution, especially because surgery and radiotherapy were also used. Lim et al assessed the outcome in 6 clear-cell RCC patients with small, asymptomatic, and previously untreated BM.11 Of 5 patients subsequently treated with sunitinib (with no local therapy), 2 experienced a durable near complete response. However, the number of cases involved is too small to enable a clear conclusion to be drawn.

Conclusion To our knowledge, this is the only controlled trial to have been undertaken of sunitinib in RCC patients with BM. In terms of efficacy, the outcome was not encouraging. There is, therefore, no clear consensus about its use in the treatment of BM. However, evidence about the safety of the agent would justify its use as firstline treatment in small and asymptomatic BM as part of a multidisciplinary approach and perhaps after radiotherapy in the case of stable disease. It remains possible that other TKIs will prove to be more effective in this setting.

Clinical Practice Points

Discussion

4

-

The number of patients involved in our trial was small relative to the sunitinib expanded access study. However, the conditions were well controlled. In the expanded access study, for example, the protocol did not require the reporting of response, and its

Clinical Genitourinary Cancer Month 2013

 Clinical experience had suggested that sunitinib was safe and

might have useful activity in renal cell cancer (RCC) patients with untreated brain metastases (BM). However, neither safety nor efficacy had been prospectively studied in this setting. We therefore conducted a multicentre phase II trial.

Christine Chevreau et al  The primary endpoint was objective response (OR) in BM











after two cycles of the standard regimen of sunitinib. Response was centrally assessed using MRI and according to RECIST criteria. Using a two-stage trial design, an OR rate of 35% was prospectively defined as the minimum needed to justify further investigation. Accrual to the trial was stopped after 16 evaluable patients since this target OR rate had not been reached. Although CNS disease was stabilised in 5 patients, no BM showed an OR. Median time to progression was 2.3 months and overall survival 6.3 months. Three patients had at least one treatment-related grade 3 or greater toxicity. However, no neurological adverse events were attributable to sunitinib and there were no cases of cerebral hemorrhage. Sunitinib has acceptable tolerability in RCC patients with untreated BM but its efficacy is limited. This may be due to poor penetration into the CNS or enhanced drug metabolism accompanying use of glucocorticoids. In contrast to case reports and retrospective data from the expanded access study, our data do not suggest that sunitinib is a good option in this setting. The effective systemic management of BM in RCC remains a major unmet need.

Acknowledgments The study was supported by Pfizer. Editorial assistance in the preparation of the report was provided by Rob Stepney, medical writer, Charlbury, UK.

Disclosure Dr Chevreau has received honoraria from Pfizer, Novartis, Bayer, GSK. Dr Ravaud has received honoraria from Pfizer, Novartis, Bayer, GSK, Aveo, Astellas, BMS, Roche. Dr Escudier has received honoraria from Pfizer, Novartis, GSK, Bayer, Genentech. Dr Oudard has received honoraria from Sanofi, Roche, Pfizer, Novartis, Bayer, BMS. Dr Négrier has received honoraria from Pfizer, Novartis, Astellas and rest of the authors declare no conflicts of interest.

References 1. Remon J, Lianes P, Martínez S. Brain metastases from renal cell carcinoma. Should we change the current standard? Cancer Treat Rev 2012; 38:249-57. 2. Bianchi M, Sun M, Jeldres C, et al. Distribution of metastatic sites in renal cell carcinoma: a population-based analysis. Ann Oncol 2012; 23:973-80. 3. Motzer RJ, Hutson TE, Tomczak P, et al. Sunitinib versus interferon alfa in metastatic renal-cell carcinoma. New Eng J Med 2007; 356:115-24. 4. NCCN Kidney Cancer Guidelines v2. 2012. Available at: http://www.nccn.org/ professionals/physician_gls/f_guidelines.asp#site. Accessed September 17, 2012. 5. Massard C, Zonierek J, Gross-Goupil M, et al. Incidence of brain metastases in renal cell carcinoma treated with sorafenib. Ann Oncol 2010; 21:1027-31. 6. Henderson CA, Bukowski RM, Stadler WM, et al. The Advanced Renal Cell Carcinoma Sorafenib (ARCCS) expanded access trial: subset analysis of patients with brain metastases (abstract). J Clin Oncol 2007; ASCO Annual Meeting; 25(suppl 18s):15506. 7. Gore ME, Hariharan S, Porta C, et al. Sunitinib in metastatic renal cell carcinoma patients with brain metastases. Cancer 2011; 117:501-9. 8. Medioni J, Cojocarasu O, Belcaceres JL, et al. Complete cerebral response with sunitinib for metastatic renal cell carcinoma. Ann Oncol 2007; 18:1282-3. 9. Simon R. Design for efficient clinical trials. Oncology (Williston Park) 1989; 3:43-9. 10. Verma J, Jonasch E, Allen PK, et al. The impact of tyrosine kinase inhibitors on the multimodality treatment of brain metastases from renal cell carcinoma. Am J Clin Oncol doi:10.1097/COC.0b013e31825d59db, Published online August 13, 2012. 11. Lim ZD, Mahajan A, Weinberg J, et al. Outcome of patients with renal cell carcinoma metastatic to the brain treated with sunitinib without local therapy. Am J Clin Oncol 2013; 36:258-60.

Clinical Genitourinary Cancer Month 2013

-5