V O L U M E 2 6 䡠 N U M B E R 2 0 䡠 J U L Y 1 0 2 0 0 8
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`O R I G I N A L R E P O R T
`
`From the Dana-Farber Cancer Institute;
`Massachusetts General Hospital; Beth
`Israel Deaconess Medical Center,
`Boston, MA; University of Southern
`California Norris Comprehensive Cancer
`Center, Los Angeles; University of Cali-
`fornia, San Francisco Comprehensive
`Cancer Center, San Francisco; Pfizer
`Global Research and Development, La
`Jolla, CA; Fox Chase Cancer Center, Phila-
`delphia, PA; University of Alabama,
`Birmingham, AL; and Washington Univer-
`sity, St Louis, MO.
`
`Submitted December 20, 2007;
`accepted April 7, 2008.
`
`Supported by Pfizer Inc. M.H.K.
`acknowledges additional support from
`the Caring for Carcinoid Foundation, Dr
`Raymond and Beverly Sackler, and the
`Stephen and Caroline Kaufer fund for
`neuroendocrine tumor research.
`
`Sources of preliminary data are listed in
`the Appendix available online.
`
`Authors’ disclosures of potential con-
`flicts of interest and author contribu-
`tions are found at the end of this
`article.
`
`Corresponding author: Matthew Kulke,
`MD, Dana-Farber Cancer Institute, 44
`Binney St, Boston, MA 02115, e-mail:
`Matthew_Kulke@dfci.harvard.edu.
`
`© 2008 by American Society of Clinical
`Oncology
`
`0732-183X/08/2620-3403/$20.00
`
`DOI: 10.1200/JCO.2007.15.9020
`
`Activity of Sunitinib in Patients With Advanced
`Neuroendocrine Tumors
`Matthew H. Kulke, Heinz-Josef Lenz, Neal J. Meropol, James Posey, David P. Ryan, Joel Picus, Emily Bergsland,
`Keith Stuart, Lesley Tye, Xin Huang, Jim Z. Li, Charles M. Baum, and Charles S. Fuchs
`
`A
`
`B
`
`S
`
`T
`
`R
`
`A
`
`C
`
`T
`
`Purpose
`Standard cytotoxic chemotherapy has limited efficacy in metastatic neuroendocrine tumor
`patients. Neuroendocrine tumors express vascular endothelial growth factor (VEGF) and its
`receptor (VEGFR). Sunitinib malate, an oral tyrosine kinase inhibitor, has activity against VEGFRs
`as well as platelet-derived growth factor receptors, stem-cell factor receptor, glial cell line–derived
`neurotrophic factor, and FMS-like tyrosine kinase-3. We evaluated the efficacy of sunitinib in a
`two-cohort, phase II study of advanced carcinoid and pancreatic neuroendocrine tumor patients.
`Patients and Methods
`Patients were treated with repeated 6-week cycles of oral sunitinib (50 mg/d for 4 weeks, followed
`by 2 weeks off treatment). Patients were observed for response, survival, and adverse events.
`Patient-reported outcomes were assessed.
`Results
`Among 109 enrolled patients, 107 received sunitinib (carcinoid, n ⫽ 41; pancreatic endocrine
`tumor, n ⫽ 66). Overall objective response rate (ORR) in pancreatic endocrine tumor patients was
`16.7% (11 of 66 patients), and 68% (45 of 66 patients) had stable disease (SD). Among carcinoid
`patients, ORR was 2.4% (one of 41 patients), and 83% (34 of 41 patients) had SD. Median time
`to tumor progression was 7.7 months in pancreatic neuroendocrine tumor patients and 10.2
`months in carcinoid patients. One-year survival rate was 81.1% in pancreatic neuroendocrine
`tumor patients and 83.4% in carcinoid patients. No significant differences from baseline in
`patient-reported quality of life or fatigue were observed during treatment.
`Conclusion
`Sunitinib has antitumor activity in pancreatic neuroendocrine tumors; its activity against carcinoid
`tumors could not be definitively determined in this nonrandomized study. Randomized trials of
`sunitinib in patients with neuroendocrine tumors are warranted.
`
`J Clin Oncol 26:3403-3410. © 2008 by American Society of Clinical Oncology
`
`INTRODUCTION
`
`Carcinoid and pancreatic neuroendocrine tumors
`are characterized by indolent behavior, characteris-
`tic well-differentiated histologic features, and the
`ability to secrete neuropeptides resulting in charac-
`teristic clinical syndromes.1 The most common of
`these is the carcinoid syndrome, which is associated
`with high serotonin levels, episodic flushing, diar-
`rhea, and right-sided valvular heart disease.2,3 When
`neuroendocrine tumors are diagnosed at an early
`stage, surgical resection is often curative.4 Unfortu-
`nately, curative surgery is rarely an option for pa-
`tients with advanced disease.5
`Palliative options for patients with advanced
`neuroendocrine tumors are limited. Approximately
`90% of neuroendocrine tumors express somatosta-
`tin receptors.6 Although somatostatin analogs are
`
`effective in ameliorating hormonal secretion symp-
`toms, they only rarely result in tumor regression.7,8
`Interferon alfa therapy has been associated with ob-
`jective tumor responses in up to 10% of patients
`with advanced neuroendocrine tumors, but may
`be associated with fatigue, myelosuppression,
`and depression.9 Approximately one third of pa-
`tients with pancreatic neuroendocrine tumors ex-
`perience objective responses after therapy with
`streptozocin- or temozolomide-based combination
`chemotherapy regimens.10-13 These regimens are
`less effective in patients with advanced carcinoid
`tumors; moreover, their prolonged use is often asso-
`ciated with toxicity.14
`The highly vascular nature of neuroendocrine
`tumors led to initial interest in angiogenesis inhibi-
`tion as a treatment modality in this disease.15 Over-
`expression of vascular endothelial growth factor
`
`© 2008 by American Society of Clinical Oncology
`
`3403
`
`Downloaded from ascopubs.org by Reprints Desk on June 14, 2017 from 216.185.156.028
`
`Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
`
`Ex. 1066-0001
`
`

`

`Kulke et al
`
`(VEGF), together with VEGF receptor (VEGFR) subtypes, has
`been observed in both carcinoid and pancreatic endocrine tumors,
`suggesting that autocrine activation of the VEGF pathway may
`promote tumor growth.16-18 Inhibition of VEGFR with function-
`blocking antibodies disrupted tumor growth in a mouse pancreatic
`neuroendocrine tumor model, providing further support for this hy-
`pothesis.19 A number of other signaling pathways have also been
`implicated in neuroendocrine tumors, which also express platelet-
`derived growth factor (PDGF), PDGF receptor (PDGFR), insulin-like
`growth factor-1, insulin-like growth factor receptor, basic fibroblast
`growth factor, transforming growth factor ␣and ␤, epidermal growth
`factor receptor, and stem-cell factor receptor.20-28
`Sunitinib malate (SUTENT; Pfizer Inc, New York, NY) is a small-
`molecule kinase inhibitor with activity against a number of tyrosine
`kinase
`receptors,
`including VEGFR-1, VEGFR-2, VEGFR-3,
`PDGFR-␣, PDGFR-␤, stem-cell factor receptor, glial cell line–
`derived neurotrophic factor receptor (rearranged during transfec-
`tion), and FMS-like tyrosine kinase-3 (Pfizer Inc, data on file).29-32
`In a phase I trial of sunitinib, antitumor activity was reported in
`patients with renal call carcinoma (RCC) and GI stromal tumors
`(GIST) and in one of four patients with neuroendocrine tumors.33
`Subsequent trials in both RCC and GIST confirmed antitumor
`activity and safety in these tumor types,34-37 leading to approval by
`the US Food and Drug Administration and the European Medi-
`
`cines Agency for use in advanced RCC patients who are treatment
`naı¨ve or who experience relapse after interleukin-2 or interferon
`alfa treatment and in GIST patients after disease progression on or
`intolerance to imatinib therapy.
`We performed a phase II, open-label, multicenter study to assess
`the safety and efficacy of sunitinib in patients with advanced neuroen-
`docrine tumors. Eligible patients with carcinoid and pancreatic endo-
`crine tumors received repeated 6-week treatment cycles of sunitinib
`administered at an oral dose of 50 mg once daily for 4 weeks, followed
`by 2 weeks off treatment. In light of the tumor regression observed in
`the phase I study, radiologic response was chosen as the primary end
`point of the present study. Patients were also observed for time to
`response/progression, survival, and toxicity. Patient-reported out-
`comes and drug exposure levels were assessed.
`
`PATIENTS AND METHODS
`
`Patients
`Patients were enrolled at eight centers in the United States between
`March 2003 and November 2005. All patients had histologic evidence of
`carcinoid or pancreatic endocrine tumor and were not candidates for curative
`surgery. Patients with small-cell carcinoma were excluded. All patients had
`measurable disease per the Response Evaluation Criteria in Solid Tumors
`(RECIST)38; an Eastern Cooperative Oncology Group performance status of
`
`Table 1. Baseline Characteristics of the Study Population
`
`Diagnosis Cohort
`
`Carcinoid Tumor (n ⫽ 41)
`
`Pancreatic Tumor (n ⫽ 66)
`
`Characteristic
`
`No. of Patients
`
`Sex
`Male
`Female
`Age, years
`Median
`Range
`ECOG performance status
`0
`1
`Time since initial diagnosis, range in months
`Primary diagnosis
`Carcinoid
`Foregut: lungs, stomach
`Midgut: small bowel, appendix
`Hindgut: colon, rectum
`Pancreatic neuroendocrine tumor
`Functioning
`Gastrinoma
`Insulinoma
`VIPoma
`Glucagonoma
`Other
`Nonfunctioning
`Unknown
`Receiving octreotide at baseline
`Previous surgery
`Previous radiotherapy
`Previous systemic therapy
`
`22
`19
`
`21
`20
`
`40
`14
`19
`7
`
`1
`22
`40
`6
`18
`
`58
`34-73
`
`0.8-272.5
`
`%
`
`53.7
`46.3
`
`51.2
`48.8
`
`97.6
`34.1
`46.3
`17.1
`
`2.4
`53.7
`97.6
`14.6
`43.9
`
`56
`32-81
`
`0.4-149.4
`
`No. of Patients
`
`42
`24
`
`36
`30
`
`65
`19
`5
`3
`2
`4
`5
`46
`1
`18
`65
`11
`40
`
`%
`
`63.6
`36.4
`
`54.5
`45.5
`
`98.5
`28.8
`7.6
`4.5
`3.0
`6.1
`7.6
`69.7
`1.5
`27.3
`98.5
`16.7
`60.6
`
`Abbreviations: ECOG, Eastern Cooperative Oncology Group; VIPoma, vasoactive intestinal peptide secreting tumor.
`
`3404
`
`© 2008 by American Society of Clinical Oncology
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`Downloaded from ascopubs.org by Reprints Desk on June 14, 2017 from 216.185.156.028
`
`Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
`
`Ex. 1066-0002
`
`

`

`Sunitinib for Advanced Neuroendocrine Tumors
`
`0 or 1; adequate hepatic, hematologic, and renal function; and either an
`echocardiogram or multiple-gated acquisition scan that demonstrated pre-
`served left ventricular ejection fraction. Treatment with prior chemotherapy,
`embolization, or radiotherapy was permitted. Patients receiving stable doses of
`somatostatin analogs were allowed to continue receiving these treatments.
`Patients who had prior treatment with VEGF pathway inhibitors, known brain
`metastases, a history of cardiac arrhythmias, or evidence of myocardial isch-
`emia or cerebrovascular accident within 12 months were excluded. All patients
`signed informed consent, and the study was approved by the institutional
`review boards of the participating institutions. The trial was registered (www.
`clinicaltrials.gov/ct/gui/show/NCT00056693) and performed in accordance
`with International Conference on Harmonization Good Clinical Practice
`
`guidelines, the Declaration of Helsinki (1996), and applicable local regulatory
`requirements and laws.
`
`Study Treatment
`Patients self-administered sunitinib at a starting dose of 50 mg by mouth
`daily for 4 weeks followed by 2 weeks off treatment in repeated 6-week cycles.
`In patients without significant toxicity, dose escalation to 62.5 mg and 75 mg
`was permitted (but not required) at the discretion of the investigator and after
`discussion with the study sponsor. Patients with significant grade 3 or 4
`toxicities underwent dose reduction to 37.5 mg and 25 mg. Treatment was
`continued until disease progression, unacceptable toxicity, or withdrawal of
`consent. Concomitant treatment with drugs having dysrhythmic potential
`
`Table 2. Adverse Events in Patients With Carcinoid and Pancreatic Tumors (N ⫽ 107) Based on National Cancer Institute Common Terminology Criteria for
`Adverse Events (version 2.0)
`
`Adverse Event
`
`No. of Patients
`
`%
`
`No. of Patients
`
`%
`
`No. of Patients
`
`%
`
`No. of Patients %
`
`Grade 1
`
`Grade 2
`
`Grade 3
`
`Grade 4
`
`Treatment-related, nonhematologic adverse eventsⴱ
`Fatigue
`Diarrhea
`Nausea
`Dysgeusia
`Skin discoloration
`Glossodynia
`Myalgia
`Stomatitis
`Hair color changes
`Vomiting
`Anorexia
`Rash
`Oral pain
`Headache
`Flushing
`Dyspepsia
`Paresthesia
`Hand-foot syndrome
`Hypertension
`Periorbital edema
`Dehydration
`Pain in extremity
`Arthralgia
`Dizziness
`Mucosal inflammation
`Insomnia
`Pulmonary embolism†
`GI hemorrhage‡
`Lipase increased
`Cardiac failure congestive
`Cerebrovascular accident
`Hyponatremia
`Treatment-emergent, hematologic adverse events
`Anemia
`Leukopenia§
`Lymphopenia
`Neutropenia储
`Thrombocytopenia¶
`
`33
`47
`36
`45
`28
`21
`17
`14
`28
`18
`20
`21
`17
`16
`20
`16
`16
`14
`3
`17
`5
`12
`10
`14
`6
`7
`0
`0
`0
`0
`0
`0
`
`61
`41
`4
`20
`39
`
`30.8
`43.9
`33.6
`42.1
`26.2
`19.6
`15.9
`13.1
`26.2
`16.8
`18.7
`19.6
`15.9
`15.0
`18.7
`15.0
`15.0
`13.1
`2.8
`15.9
`4.7
`11.2
`9.3
`13.1
`5.6
`6.5
`0
`0
`0
`0
`0
`0
`
`57.0
`38.3
`3.7
`18.7
`36.4
`
`36
`18
`15
`7
`11
`12
`16
`18
`6
`7
`7
`6
`9
`8
`1
`4
`3
`2
`3
`0
`6
`2
`4
`1
`2
`4
`0
`0
`0
`0
`0
`0
`
`23
`42
`55
`34
`25
`
`33.6
`16.8
`14.0
`6.5
`10.3
`11.2
`15.0
`16.8
`5.6
`6.5
`6.5
`5.6
`8.4
`7.5
`0.9
`3.7
`2.8
`1.9
`2.8
`0
`5.6
`1.9
`3.7
`0.9
`1.9
`3.7
`0
`0
`0
`0
`0
`0
`
`21.5
`39.3
`51.4
`31.8
`23.4
`
`26
`5
`6
`0
`0
`3
`2
`2
`0
`7
`3
`1
`0
`1
`0
`0
`0
`2
`11
`0
`5
`1
`1
`0
`3
`0
`0
`0
`0
`0
`0
`2
`
`3
`15
`28
`31
`9
`
`24.3
`4.7
`5.6
`0
`0
`2.8
`1.9
`1.9
`0
`6.5
`2.8
`0.9
`0
`0.9
`0
`0
`0
`1.9
`10.3
`0
`4.7
`0.9
`0.9
`0
`2.8
`0
`0
`0
`0
`0
`0
`1.9
`
`2.8
`14.0
`26.2
`29.0
`8.4
`
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`1
`2
`1
`1
`1
`1
`
`0
`0
`0
`5
`0
`
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0.9
`1.9
`0.9
`0.9
`0.9
`0.9
`
`0
`0
`0
`4.7
`0
`
`NOTE. One grade 5 treatment-related adverse event occurred and was a result of GI hemorrhage.
`ⴱIncidence cutoff ⱖ 10% of patients.
`†No patient experienced deep vein thrombosis or vena cava thrombosis.
`‡Includes GI hemorrhage and lower GI hemorrhage.
`§Data missing for three patients at baseline.
`储Data missing for four patients at baseline.
`¶Data missing for seven patients at baseline.
`
`www.jco.org
`
`© 2008 by American Society of Clinical Oncology
`
`3405
`
`Downloaded from ascopubs.org by Reprints Desk on June 14, 2017 from 216.185.156.028
`
`Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
`
`Ex. 1066-0003
`
`

`

`Kulke et al
`
`(terfenadine, quinidine, procainamide, disopyramide, sotalol, probucol, be-
`pridil, haloperidol, risperidone, and indapamide) was not allowed.
`Study Evaluations
`Biochemical and hematologic parameters were assessed at baseline and
`every 2 weeks during the first cycle; thereafter, hematologic parameters were
`measured every 2 weeks, and biochemical parameters were measured on the
`first and last treatment days of each cycle. Multiple-gated acquisition or echo-
`cardiogram measurement of left ventricular ejection fraction was performed at
`screening and at the end of odd-numbered cycles.
`Computed tomography or magnetic resonance imaging was performed
`at screening, at the end of cycle 1 and each odd-numbered cycle, if disease
`progression was suspected, and at study end or patient withdrawal, if a scan
`had not been performed within the previous 6 weeks. All responses were
`confirmed at least 4 to 6 weeks after initial documentation of response. Objec-
`tive tumor response was assessed using RECIST.38 Patients were observed after
`treatment discontinuation for survival status.
`Blood samples for determination of predose (day 1) and trough (day
`⬎ 1) concentrations of sunitinib and its active metabolite (SU12662) were
`collected on treatment days 1, 14, and 28 of cycles 1, 2, and 3, respectively; on
`day 1 of cycles ⱖ 4; and at end of treatment.
`General health-related quality of life was assessed using the EuroQol
`Group’s EQ-5D self-report questionnaire,39 and self-reported fatigue was as-
`sessed using the Functional Assessment of Chronic Illness Therapy (FACIT)
`–Fatigue scale.40 Patients completed the EQ-5D assessment on days 1 and 28 of
`treatment cycles 1 to 6 and at end of treatment/withdrawal visit. Patients
`completed the FACIT-Fatigue assessment at baseline (cycle 1, day 1), weekly
`through cycle 4, and at the end of treatment.
`Statistical Methods
`This study used an open-label, two-cohort, Simon two-stage design41 to
`test the null hypothesis that the true objective response rate (ORR) was ⱕ 5%
`
`versus the alternative hypothesis that the true response rate was ⱖ 15%.
`Sixty-three patients were required in each cohort (carcinoid and pancreatic
`neuroendocrine tumor) to detect the response difference with 85% power and
`a significance level of 5%. In each cohort, 38 patients were treated in stage 1.
`If ⱕ one objective tumor response was observed in the first 38 treated patients
`in either cohort, then enrollment in that cohort was to be terminated. How-
`ever, if ⱖ two objective tumor responses were observed in the first 38 treated
`patients, then the cohort was to be expanded to enroll 63 treated patients.
`Secondary end points included time to objective tumor response (time
`from the first sunitinib dose to the first documentation of objective tumor
`response), time to tumor progression (TTP; time from the first sunitinib dose
`to the first documentation of objective tumor progression, initiation of other/
`additional anticancer therapy, or patient withdrawal as a result of unknown
`reasons), and overall survival (time from the first sunitinib dose to the date of
`death as a result of any cause). All time-to-event data (time to objective tumor
`response, TTP, and overall survival) were described using the Kaplan-
`Meier method.
`
`RESULTS
`
`Patient Characteristics and Treatment
`Of the 109 enrolled patients, 107 (41 carcinoid patients and
`66 pancreatic neuroendocrine tumor patients) were treated with
`sunitinib. Characteristics of treated patients are listed in Table 1. The
`median sunitinib doses administered in the carcinoid and pancreatic
`neuroendocrine cohorts were 50.0 mg (range, 28.0 to 53.9 mg) and
`49.6 mg (range, 28.3 to 58.3 mg), respectively. Sixty-seven patients
`(62.6%) had at least one dosing interruption, 51 patients (47.7%) had
`
`Table 3. Efficacy of Sunitinib by Diagnosis Cohort
`
`Carcinoid Tumor (n ⫽ 41)
`
`Pancreatic Tumor (n ⫽ 66)
`
`Parameter
`
`No. of Patients
`
`Best overall response
`Complete response
`Partial response
`Stable disease
`ⱖ 6 months
`Progressive disease
`Not assessable
`Missing
`Overall objective response rate
`95% CI, %ⴱ
`Time to tumor response, months
`No. of patients†
`Median‡
`Range
`95% CI for median
`Time to tumor progression, months
`No. of patients
`Patients censored
`Median‡
`Range
`95% CI for median
`1-year survival
`95% CI, %
`
`0
`1
`34
`23
`1
`4
`1
`1
`
`1
`
`41
`23
`
`41
`
`%
`
`0.0
`2.4
`82.9
`56.1
`2.4
`9.8
`2.4
`2.4
`
`56.1
`
`83.4
`
`0.1 to 12.9
`
`3.7
`
`NA
`
`10.2
`0.02-19.0§
`9.2 to 17.5
`
`66.7 to 92.3
`
`No. of Patients
`
`0
`11
`45
`37
`5
`3
`2
`11
`
`11
`
`66
`34
`
`66
`
`%
`
`0.0
`16.7
`68.2
`56.1
`7.6
`4.5
`3.0
`16.7
`
`51.5
`
`81.1
`
`8.6 to 27.9
`
`4.0
`2.6-6.7
`3.7 to 6.5
`
`7.7
`0.02-15.3§
`6.5 to 12.5
`
`67.3 to 89.5
`
`Abbreviation: NA, not applicable.
`ⴱUsing exact method based on F distribution.
`†For time to tumor response, only patients with confirmed responses are included.
`‡From the Kaplan-Meier estimate.
`§Indicates a censored observation.
`
`3406
`
`© 2008 by American Society of Clinical Oncology
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`Downloaded from ascopubs.org by Reprints Desk on June 14, 2017 from 216.185.156.028
`
`Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
`
`Ex. 1066-0004
`
`

`

`Sunitinib for Advanced Neuroendocrine Tumors
`
`sponse. The lack of at least two confirmed partial responses among
`carcinoid tumor patients in the first enrollment stage precluded
`further enrollment of carcinoid patients. Overall, 43.9% of carci-
`noid patients and 62.1% of the pancreatic neuroendocrine tumor
`patients seemed to demonstrate some degree of tumor shrinkage
`(Figs 1A and 1B). The majority of these patients had stable dis-
`ease (SD) by RECIST. The overall rate of SD was 68.2% among
`pancreatic neuroendocrine tumor patients and 82.9% among car-
`cinoid patients.
`The median TTP was 10.2 months for patients with carcinoid
`tumors and 7.7 months for patients with pancreatic neuroendocrine
`tumors (Fig 2A). One-year survival rate was 83.4% in carcinoid pa-
`tients and 81.1% in pancreatic neuroendocrine tumor patients (Table
`3 and Fig 2B).
`
`Patient-Reported Outcomes
`Assessable EQ-5D questionnaires were received from 90% to
`100% of available patients at each assessment on days 1 and 28 of
`treatment cycles 1 to 6. No significant changes in the EQ-5D index or
`EQ-5D visual analog scale scores were evident during the first six
`cycles of treatment (Figs 3A and 3B). Completion rates for FACIT-
`Fatigue questionnaires during the first four cycles of treatment were
`84% to 99% on days 1, 7, 14, 21, and 28 and 47% to 78% on day 35.
`Although the mean FACIT-Fatigue score remained relatively stable
`for each treatment cycle, a pattern of modest increases in patient-
`reported fatigue during the dosing period with recovery during the
`off-treatment periods seemed evident (Fig 3C).
`
`Patient
`
`Patient
`
`Progressive Disease
`
`Partial Response
`
`Progressive Disease
`
`Partial Response
`
`70
`
`50
`
`20
`
`0
`
`-30
`
`-50
`
`A
`
`Change (%)
`
`-100
`
`70
`
`50
`
`20
`
`0
`
`-30
`
`-50
`
`-100
`
`B
`
`Change (%)
`
`Fig 1. Investigator-assessed maximum percent reduction in tumor lesions in
`patients with (A) carcinoid and (B) pancreatic neuroendocrine tumors. Patients
`delineated in gold (wider bars) were classified as having partial responses by
`Response Evaluation Criteria in Solid Tumors.
`
`a dose reduction, and three patients had dose increases to 62.5 mg
`daily. Of the 107 patients who began treatment, 94 (37 with carcinoid
`tumors and 57 with pancreatic tumors) received more than one treat-
`ment cycle, and 65 patients (24 with carcinoid tumors and 41 with
`pancreatic tumors) began ⱖ five cycles of treatment (approximately 6
`months). The median time on treatment was 7.2 months (range, 0.9 to
`19.4 months). The median follow-up duration was 13.4 months, with
`a median duration of 15.1 months for patients with carcinoid tumors
`and 12.5 months for patients with pancreatic tumors. The most com-
`mon reasons for treatment discontinuation were disease progression
`(n ⫽ 45), completion of therapy (n ⫽ 30; all 30 patients completed six
`cycles of therapy and continued therapy on a continuation protocol),
`withdrawal of consent (n ⫽ 20), and adverse events (AEs; n ⫽ 11). Of
`the patients who withdrew consent, 11 withdrew consent before initi-
`ation of their fifth treatment cycle, and nine withdrew consent after
`initiation of cycle 5.
`
`Pharmacokinetics
`In patients with carcinoid tumors, median trough concentra-
`tions of sunitinib, SU12662, and total drug (sunitinib ⫹ SU12662)
`were 49, 21, and 71 ng/mL, respectively, on days 14, 21, and/or 28 of
`cycles 1 to 3; corresponding values for patients with pancreatic neu-
`roendocrine tumors were 37, 20, and 60 ng/mL, respectively, on days
`14, 21, and/or 28 of cycles 1 to 3. These values approximated the
`preclinically determined therapeutic total drug concentration of more
`than 50 ng/mL needed to inhibit receptor phosphorylation and cause
`tumor regression.29 Median total drug concentrations on day 1 of
`cycle 2 and of all subsequent cycles were less than 3 ng/mL in both
`cohorts, suggesting nearly complete drug washout between cycles.
`
`AEs
`
`Fatigue was the most common treatment-related AE and devel-
`oped in 95 patients (88.8%) overall (Table 2). In most patients, fatigue
`was mild; 26 patients experienced grade 3 fatigue, and none experi-
`enced grade 4 fatigue. Other common AEs included diarrhea,
`nausea, dysgeusia, glossodynia, and skin discoloration. Hyperten-
`sion, a toxicity also observed with other VEGF pathway inhibitors,
`was observed in 15.9% of the patient population and was more
`common among carcinoid patients (19.7%) than among patients
`with pancreatic neuroendocrine tumors (9.8%). Grade 3 hyperten-
`sion was reported in 10.3% of patients; no incidents of grade 4
`hypertension were reported. Treatment-related grade 4 AEs were
`reported infrequently and included GI hemorrhage (1.9%), pulmo-
`nary embolism (0.9%), increased lipase (0.9%), cardiac congestive
`failure (0.9%), cerebrovascular accident (0.9%), and hyponatremia
`(0.9%; Table 2). A single treatment-related death was caused by GI
`hemorrhage. Grade 3 or 4 neutropenia and thrombocytopenia
`developed in 33.7% and 8.4% of patients, respectively (Table 2). A
`higher incidence of grade 3 leukopenia was observed in patients
`with pancreatic tumors versus carcinoid cancers (18.2% v 7.3%,
`respectively; no grade 4 leukopenia was reported).
`
`Efficacy
`The ORR in the pancreatic endocrine tumor cohort was
`16.7% (Table 3). Responders included one patient with gastri-
`noma, one patient with a vasoactive intestinal peptide tumor, and
`nine patients with nonfunctioning pancreatic endocrine tumors.
`One patient with a foregut carcinoid tumor had a confirmed re-
`
`www.jco.org
`
`© 2008 by American Society of Clinical Oncology
`
`3407
`
`Downloaded from ascopubs.org by Reprints Desk on June 14, 2017 from 216.185.156.028
`
`Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
`
`Ex. 1066-0005
`
`

`

`Kulke et al
`
`Carcinoid
`Pancreatic tumor
`Total
`
`C1D1 C1D28 C2D1 C2D28 C3D1 C3D28 C4D1 C4D28 C5D1 C5D28 C6D1 C6D28
`Treatment Cycle and Day
`
`Carcinoid
`Pancreatic tumor
`Total
`
`C1D1 C1D28 C2D1 C2D28 C3D1 C3D28 C4D1 C4D28 C5D1 C5D28 C6D1 C6D28
`Treatment Cycle and Day
`
`Carcinoid
`Pancreatic tumor
`Total
`
`C4D35
`C5D1
`C4D28
`C4D21
`C4D14
`C4D7
`C3D35
`C3D1
`C2D35
`C3D21
`C2D28
`C4D1
`C3D28
`C2D14
`C2D21
`C3D14
`C1D7
`C2D7
`1D1
`C2D1
`C1D35
`C3D7
`C1D28
`C1D21
`C1D14
`Treatment Cycle and Day
`
`1.0
`0.9
`0.8
`0.7
`0.6
`0.5
`0.4
`0.3
`0.2
`0.1
`0
`
`100
`90
`80
`70
`60
`50
`40
`30
`20
`10
`0
`
`52
`
`39
`
`26
`
`13
`
`0 C
`
`A
`
`Mean EQ-5D Index Score
`
`B
`
`Mean EQ-VAS Score
`
`C
`
`Mean FACIT-Fatigue Score
`
`Fig 3. Quality of life of sunitinib-treated patients. (A) EQ-5D index, ranging from
`0 (death) to 1 (perfect health). (B) EQ-5D visual analog scale (EQ-VAS), ranging
`from 0 (worst) to 100 (best imaginable self-assessed health state). (C) Functional
`Assessment of Chronic Illness Therapy (FACIT) –Fatigue scores, ranging from 0
`(highly fatigued state) to 52 (no fatigue). C, cycle; D, day.
`
`mon treatment-related toxicities were constitutional (fatigue and
`anorexia) or GI (diarrhea and nausea). Hypertension, a toxicity also
`observed with other inhibitors of the VEGF pathway, was observed in
`15.9% of the patient population. Hypertension was more common in
`carcinoid patients than in patients with pancreatic neuroendocrine
`tumors (19.7% v 9.8%, respectively), a finding possibly related to
`concurrent secretion of vasoactive neuropeptides in some carcinoid
`
`Carcinoid (N = 41)
`Median, 10.2 months
`95% CI, 9.2 to 17.4
`Pancreatic tumor (N = 66)
`Median, 7.7 months
`95% CI, 6.5 to 12.5
`
`100
`90
`80
`70
`60
`50
`40
`30
`20
`10
`
`A
`
`Time To Tumor Progression
`
`Probability (%)
`
`0
`
`2
`
`4
`
`6
`
`10
`12
`8
`Time (months)
`
`14
`
`16
`
`18
`
`20
`
`Carcinoid (N = 41)
`Median, 25.3 months
`95% CI, 18.4 to NA
`Pancreatic tumor (N = 66)
`Median, has not
`been reached
`
`100
`90
`80
`70
`60
`50
`40
`30
`20
`10
`
`B
`
`Probability (%)
`Overall Survival
`
`0
`
`2 4
`
`6 8
`
`10 12 14 16 18 20 22 24 26 28 30 32
`Time (months)
`
`(A) Time to progression and (B) overall survival of patients
`Fig 2.
`receiving sunitinib.
`
`DISCUSSION
`
`This multicenter, phase II study demonstrated that sunitinib is asso-
`ciated with antitumor activity in patients with advanced neuroendo-
`crine tumors. The ORR was greater in patients with pancreatic
`neuroendocrine tumors (16.7%) than in patients with carcinoid tu-
`mors (2.4%). In both patient populations, administration of sunitinib
`was associated with preserved quality of life.
`Our observation that sunitinib seems to be more active in pan-
`creatic neuroendocrine tumors than in carcinoid tumors mirrors sim-
`ilar observations with more traditional cytotoxic agents. Streptozocin-
`based combination chemotherapy regimens, as well as regimens
`incorporating the alkylating agent dacarbazine or the oral analog
`temozolomide, have been associated with ORRs of 33% to 45% in
`pancreatic endocrine tumors.10,11,42 ORRs with similar regimens in
`carcinoid tumors range from 7% to 16%.11,14,43 In light of the often
`indolent nature of both carcinoid and pancreatic neuroendocrine
`tumors, the cumulative toxicities of these regimens (which include
`effects on renal, cardiac, and hematologic function) have limited their
`widespread acceptance in these patient populations.
`The toxicity profile of sunitinib in our study was similar to that
`observed in trials of sunitinib in other disease types. The most com-
`
`3408
`
`© 2008 by American Society of Clinical Oncology
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`Downloaded from ascopubs.org by Reprints Desk on June 14, 2017 from 216.185.156.028
`
`Copyright © 2017 American Society of Clinical Oncology. All rights reserved.
`
`Ex. 1066-0006
`
`

`

`Sunitinib for Advanced Neuroendocrine Tumors
`
`patients. A higher incidence of grade 3 leukopenia in pancreatic neu-
`roendocrine tumor patients than in patients with carcinoid cancers
`(18.2% v 7.3%, respectively) may be attributable to the greater num-
`ber of pancreatic neuroendocrine tumor patients who had received
`prior systemic therapy, including cytoxic chemotherapy (Table 1).
`The modest ORRs observed in our trial highlight the challenge of
`assessing the efficacy of an agent that may be largely cytostatic, partic-
`ularly in a disease that is naturally indolent. In a randomized trial of
`sunitinib versus placebo in patients with imatinib-refractory GIST,
`for example, the ORR associated with sunitinib treatment was only
`7%, yet treatment with sunitinib was associated with a significant
`improvement in TTP.34 Our observation that the majority of pa-
`tients with both carcinoid and pancreatic neuroendocrine tumors
`seemed to experience minor responses to therapy, classified as
`SD by RECIST, suggests that sunitinib may also delay TTP in
`neuroendocrine tumors. The median TTP in our trial exceeds 7
`months in both patient cohorts, compared with values ranging
`from 3.2 to 7.6 months reported in other recent therapeutic trials in
`neuroendocrine tumor patients.14,44,45 However, differences in pa-
`tient selection and inherent variability in the natural history of
`neuroendocrine tumors make direct comparisons between these
`different trials difficult. The uncertainty surrounding the interpre-
`tation of tumor stability in this setting highlights the need for
`appropriately stratified randomized trials or, alternatively, vali-
`dated surrogate markers of antitumor activity in this disease.
`The mechanism of action of sunitinib in neuroendocrine tu-
`mors remains unclear. Inhibitors of the VEGF pathway are gener-
`ally thought to exert their antitumor effects indirectly by targeting
`endothelial cells and inhibiting tumor angiogenesis.46 In preclini-
`cal studies, VEGF inhibition alone has been associated with inhi-
`bition of tumor growth and metastases, although not necessarily
`tumor shrinkage.47 For this reason, VEGF pathway inhibitors have
`generally been combined with standard cytotoxic chemotherapy in
`clinical trials.48
`In RCC, treatment with sunitinib was associated with an ORR
`of 40%, suggesting that direct inhibition of VEGF signaling in
`tumor cells may be an alternative mechanism of action in specific
`tumor type.35,49 Like RCC, neuroendocrine tumors are both highly
`vascular and associated with high levels of VEGF and VEGFR
`expression. Additional mechanisms,
`including inhibition of
`PDGFR or other cell signaling pathways, may also have contrib-
`uted to the antitumor activity in neuroendocrine tumors observed
`in this study.
`Secondary end points in our study included quality-of-life assess-
`ment and assessment of fatigue using the EQ-5D and FACIT-Fatigue
`questionnaires, respectively. The relatively high number of patients
`who withdrew consent to participate in the study raises the possibility
`that treatment may have been associated with undue toxicity. How-
`ever, the formal assessments suggest that treatment with sunitinib did
`not seem to cause significant changes from baseline in either quality-
`of-life or fatigue scores during the treatment period. These findings are
`consistent with a high incidence of SD but are also intriguing in light of
`
`the high incidence of fatigue reported as an AE. A similarly high
`incidence of fatigue was reported in a randomized trial of sunitinib
`versus placebo in patients with GIST.34 In this study, the incidence of
`fatigue was similar in both the treatment and placebo groups, suggest-
`ing that a large proportion of reported fatigue may be attributable to
`tumor burden. It is possible that, in our study, tumor burden also
`accounted for the high incidence of reported fatigue. Alternatively,
`drug toxicities may have been, to some extent, balanced by treatment-
`related clinical benefit.
`In conclusion, treatment with sunitinib resulted in objective
`tumor responses in patients with pancreatic neuroendocrine tu-
`mors. Whether sunitinib may also be associated with an antitumor
`effect in carcinoid tumors could not be clearly determined in this
`nonrandomized study. Further investigation of sunitinib in the
`randomized setting or in combination with other agents is war-
`ranted in these diseases.
`
`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