axitinib or bevacizumab plus folfiri or modified folfox-6 after failure of first-line therapy for...

Original Study

Axitinib or Bevacizumab Plus FOLFIRI orModified FOLFOX-6 After Failure of First-Line

Therapy for Metastatic Colorectal Cancer:A Randomized Phase II Study

Johanna C. Bendell,1 Christophe Tournigand,2 Anna Swieboda-Sadlej,3

Carlo Barone,4 Zev A. Wainberg,5 Jong Gwang Kim,6 Carles Pericay,7

Davide Pastorelli,8 Jamal Tarazi,9 Brad Rosbrook,9 Joanna Bloom,10

Alejandro D. Ricart,9,� Sinil Kim,9,� Alberto F. Sobrero11

AbstractThis randomized phase II study evaluated axitinib or bevacizumab combined with chemotherapy in previouslytreated patients with metastatic colorectal cancer. Survival was comparable with axitinib or bevacizumabwhen combined with second-line chemotherapy. Higher toxicity and treatment discontinuations due to adverseevents in axitinib arms suggest continuous dosing of axitinib plus chemotherapy may be less well toleratedthan bevacizumab plus chemotherapy.Objective: Axitinib, a potent and selective second-generation inhibitor of vascular endothelial growth factor receptors1, 2, and 3, shows activity in multiple tumor types, including those refractory to previous antiangiogenic therapy. This ran-domized,multicenter, parallel-group, open-label phase II trial compared axitinibwith bevacizumab each in combinationwith5-fluorouracil/leucovorin/oxaliplatin (FOLFOX) or 5-fluorouracil/leucovorin/irinotecan (FOLFIRI) for second-line treatment ofmetastaticcolorectal cancer.Methods: Patients were randomized 1:1 to axitinib 5 mg twice daily or bevacizumab 5 mg/kgevery 2 weeks plus modified FOLFOX-6 (if previously treated with irinotecan) or FOLFIRI (if previously treated withoxaliplatin) and were stratified by performance status and prior bevacizumab therapy. Primary endpoint was progression-free survival. Results: In 171 patients, progression-free survival was 7.6 months with axitinib/FOLFOX vs. 6.4 months withbevacizumab/FOLFOX (hazard ratio [HR], 1.04; 95% confidence interval [CI], 0.55-1.96; 1-sided P ¼ .55) and 5.7 monthswith axitinib/FOLFIRI vs. 6.9 months with bevacizumab/FOLFIRI (HR, 1.27; 95% CI, 0.77-2.11; 1-sided P ¼ .83). Overallsurvival was 17.1 vs. 14.1 months with axitinib/FOLFOX and bevacizumab/FOLFOX (HR, 0.69; 95% CI, 0.37-1.27; 1-sidedP ¼ .12) and 12.9 vs. 15.7 months with axitinib/FOLFIRI and bevacizumab/FOLFIRI (HR, 1.36; 95% CI, 0.82-2.24; 1-sidedP¼ .88). More grade� 3 adverse events (eg, diarrhea, fatigue, decreased appetite) and treatment discontinuations due toadverse events occurred with axitinib. Conclusions: Compared with bevacizumab, axitinib did not improve outcomeswhen added to second-line chemotherapy for metastatic colorectal cancer. With current dosing regimens, axitinib plusFOLFOX or FOLFIRI seems to be less well tolerated than bevacizumab-based regimens.

Clinical Colorectal Cancer, Vol. 12, No. 4, 239-47 ª 2013 Elsevier Inc. All rights reserved.Keywords: Chemotherapy, Irinotecan, Oxaliplatin, Second-line, VEGF

�Employed at Pfizer Inc at the time of the study described and development of thismanuscript.

1Sarah Cannon Research Institute, Nashville, TN2Hôpital Henri Mondor, Créteil, France3Medical University of Warsaw, Warsaw, Poland4Università Cattolica del S. Cuore, Rome, Italy5UCLA Jonsson Comprehensive Cancer Center, Los Angeles, CA6Kyungpook National University School of Medicine, Daegu, South Korea7Corporacio Sanitaria Parc Tauli, Barcelona, Spain

8Istituto Oncologico Veneto, Padua, Italy9Pfizer Oncology, La Jolla, CA10Envision Pharma Group, Southport, CT11IRCCS San Martino-IST, Genoa, Italy

Submitted: May 23, 2013; Revised: Sep 10, 2013; Accepted: Sep 16, 2013

Address for correspondence: Johanna C. Bendell, MD, Sarah Cannon ResearchInstitute, 250 25th Ave North, Nashville, TN 37203Fax: 615-329-7532; e-mail contact: [email protected]

1533-0028/$ - see frontmatter ª 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.clcc.2013.09.001 Clinical Colorectal Cancer December 2013 - 239

mailto:[email protected]

http://dx.doi.org/10.1016/j.clcc.2013.09.001

Axitinib or Bevacizumab Plus Chemotherapy

240 -

Introduction daily treatment with aspirin (> 325 mg/d); current use or anticipated

Currently recommended first-line chemotherapy regimens formetastatic colorectal cancer (mCRC) include 5-fluorouracil (5-FU)/leucovorin/oxaliplatin (FOLFOX) and 5-FU/leucovorin/irinotecan(FOLFIRI).1 FOLFOX and FOLFIRI have similar efficacy in thissetting, and second-line chemotherapy after disease progressiongenerally uses the regimen not used in the first line, yielding similarprogression-free survival (PFS) and overall survival (OS).2 In theUnited States, FOLFOX-based regimens are more commonly usedin first-line mCRC.3

Recent research in mCRC has focused on agents that inhibitangiogenesis, a process critical for tumor growth and metastasis.Bevacizumab, a monoclonal antibody that binds vascular endo-thelial growth factor (VEGF)-A, improves survival in patientswith mCRC when added to first-line chemotherapy and inbevacizumab-naïve and pretreated second-line settings.4-6 Datafrom the VELOUR study show that the addition of aflibercept, anagent that binds VEGF-A, VEGF-B, and placental growth factor,to second-line chemotherapy for mCRC improves OS in patientswho are both bevacizumab-naïve and pretreated.7 A randomizedphase III trial of single-agent regorafenib, an inhibitor of multipleangiogenic tyrosine kinase receptors, has shown survival benefit vs.placebo in patients with refractory mCRC.8 These data suggestantiangiogenic therapy is effective through multiple lines of treat-ment in patients with mCRC.

Axitinib, a potent and selective second-generation inhibitorof VEGF receptors (VEGFRs) 1 to 3,9 has shown promising single-agent activity against a variety of tumor types.10-14 Whereas bev-acizumab specifically binds VEGF-A, inhibition of VEGFR1 to3 by axitinib may result in more complete blockade of VEGFsignaling.9 A phase I study showed that axitinib may be safelyco-administered with FOLFOX or FOLFIRI without affectingplasma drug concentrations.15 Our study sought to characterize theseregimens in second-line treatment of mCRC. We report the results ofthis multicenter, randomized, open-label phase II trial comparingaxitinib with bevacizumab each in combination with FOLFOX orFOLFIRI as second-line therapy for patients with mCRC.

Patients and MethodsPatients

Eligible patients were aged � 18 years (� 20 years in Japan) withhistologically documented mCRC and disease progression perResponse Evaluation Criteria in Solid Tumors (RECIST version1.0)16 within 6 months of the last dose of first-line irinotecan oroxaliplatin-containing therapy or adjuvant irinotecan-or oxaliplatin-containing therapy. Other eligibility criteria included EasternCooperative Oncology Group performance status (ECOG PS)0 or 1, life expectancy � 12 weeks, adequate organ function, res-olution of any adverse events (AEs) from prior cancer treatments,and no preexisting, uncontrolled hypertension (ie, blood pressure[BP] > 140/90 mm Hg despite use of antihypertensive medication).

Exclusion criteria included > 1 prior systemic chemotherapyregimen for mCRC; prior irradiation to � 25% of bone marrow;second malignancy; hemoptysis (> 0.5 teaspoon/d) within the weekbefore enrollment; major surgical procedure within previous 4 weeks;gastrointestinal abnormalities, including bleeding or prior surgicalprocedures affecting absorption; use of anticoagulant therapy; chronic

Clinical Colorectal Cancer December 2013

need for known cytochrome P450 (CYP) 3A4 inhibitors or CYP3A4or CYP1A2 inducers; or pregnancy or potential pregnancy.

Study Design and TreatmentPatients previously treated with irinotecan-containing reg-

imens were randomized 1:1 to axitinib/modified FOLFOX-6 orbevacizumab/modified FOLFOX-6; those previously treated withoxaliplatin-containing regimens were randomized 1:1 to axitinib/FOLFIRI or bevacizumab/FOLFIRI. Patients were stratified byECOG PS (0 vs. 1) and prior bevacizumab treatment.

This study was conducted in accordance with the Declaration ofHelsinki, the International Conference on Harmonization Guide-lines on Good Clinical Practice, study protocol, and applicable localregulatory requirements and laws. All participants provided writteninformed consent and agreed to comply with the study protocol.Study protocol and informed consent forms were approved by aninstitutional review board/independent ethics committee.

Axitinib was administered at a starting dose of 5 mg twice daily.Dosage could be increased to 7 mg twice daily and then to 10 mgtwice daily in patients with no grade > 2 treatment-related AEsfor � 2 weeks and with BP � 150/90 mm Hg without the use ofantihypertensive medication. The axitinib dose could be reduced to3 mg twice daily and then to 2 mg twice daily in patients whodeveloped grade 3 nonhematologic AEs or patients with 2 readingsof systolic BP > 150 mm Hg or diastolic BP > 100 mm Hg whowere receiving maximal antihypertensive therapy. Axitinib wasinterrupted in patients with grade 4 AEs, 2 readings of systolic BP >

160 mm Hg or diastolic BP > 105 mm Hg, or � 2 g protein/24hours; axitinib was resumed at 1 lower dose level when AEsimproved to grade � 2, BP was < 150/100 mm Hg, or < 2 gprotein/24 hours was present, respectively.

Bevacizumab 5 mg/kg was administered every 2 weeks. Dosingwas interrupted in patients with grade � 3 AEs and resumed whenAEs resolved to grade � 1, if � 2 g protein/24 hours was presentand resumed when protein was < 2 g/24 hours, or if BP was > 150/100 mm Hg and new or additional antihypertensive therapy wasinitiated. Bevacizumab was discontinued in patients who developedsymptomatic congestive heart failure, nephrotic syndrome, grade 4hypertension, serious bleeding or thromboembolic events, bowelperforation or anastomotic dehiscence, or an infusion-related severehypersensitivity reaction.

FOLFOX and FOLFIRI were administered in 2-week cycles.Patients receiving FOLFOX were administered oxaliplatin (85 mg/m2) and leucovorin (400 mg/m2 or 200 mg/m2 l-leucovorin)intravenous (I.V.) infusions concurrently over 120 minutes viaseparate infusion lines; 5-FU was administered as a 400 mg/m2

bolus I.V. injection after oxaliplatin/leucovorin infusion andthen as a 2400 mg/m2 continuous I.V. infusion over 46 to 48hours. Patients receiving FOLFIRI were administered irinotecan(180 mg/m2) I.V. infusions over 90 minutes, concurrent withleucovorin (400 mg/m2 or 200 mg/m2 l-leucovorin) I.V. infusionsover 120 minutes via separate infusion lines; 5-FU was adminis-tered as indicated earlier. Chemotherapy dose modifications wereallowed, depending on individual patient tolerability.

Patients continued to receive study drugs until disease progres-sion or unacceptable toxicity. Primary endpoint was PFS. Secondary

Johanna C. Bendell et al

endpoints included OS, objective response rates (ORRs), andduration of response.

AssessmentsPatients were evaluated by physical examination at screening, day

1 of each cycle, and end of treatment. Tumors were radiologicallyassessed at screening and every 8 weeks (RECIST version 1.0).16

Safety was monitored throughout the study; AEs were gradedaccording to Common Terminology Criteria for Adverse Eventsversion 3.0.17 BP was measured twice (separated by 1 hour) at eachclinic visit and twice daily by patients. Patients were instructed tocontact their physician if BP was > 150/100 mm Hg.

Statistical AnalysisPrimary and secondary efficacy analyses were conducted com-

paring the axitinib/FOLFIRI arm with the bevacizumab/FOLFIRI

Figure 1 Patient Disposition (CONSORT Diagram)

N = 1

Screened and

Axitinib/FOLFIRI

n = 49

Analyzed for efficacy

(ITT population)

Bevacizumab/FOLFIRI

n = 51


(ITT population)

Did not receive

treatment

n = 3

Received treatment

n = 46

Analyzed for safety

(Safety population)

Received treatment

n = 51

Analyzed for safety

(Safety population)

Discontinued study

n = 43

Died (n = 31)

Disease progression (n = 0)

Refused further treatment (n = 0)

Lost to follow-up (n = 1)

Other (n = 11)

Discontinued study

n = 50

Died (n = 31)




Other (n = 15)

Prior oxaliplatin-containing regimen

n = 100

Discontinued axitinib

n = 46


Adverse events (n = 8)

Health deterioration (n = 2)

Refused treatment (n = 2)

Died (n = 0)

Other (n = 7)

Discontinued bevacizumab

n = 50




Died (n = 3)

Other (n = 8)

Abbreviation: ITT ¼ intent-to-treat.

arm and the axitinib/FOLFOX arm with the bevacizumab/FOLFOXarm. The study was powered to detect a 60% improvement in theprimary endpoint of median PFS (8 vs. 5 months) for each of the 2comparisons (axitinib/FOLFIRI vs. bevacizumab/FOLFIRI, axitinib/FOLFOX vs. bevacizumab/FOLFOX). The PFS at 5 months forbevacizumab/chemotherapy was an average estimate for a mixedpopulation of bevacizumab-pretreated and bevacizumab-naïve pa-tients. By assuming uniform accrual, a 1-sided log-rank test with afalse-positive error rate of 0.15, and a power of 0.80, a sample size ofapproximately 176 patients (44 in each of the 4 treatment arms) wasrequired.

Efficacy analyses were performed in an intent-to-treat fashion.Differences in PFS and OS between treatment arms were analyzedusing a log-rank test stratified by ECOG PS and prior bevacizumabtherapy. Estimates of PFS and OS were determined using theKaplaneMeier method. The ORR was compared using the

71

randomized

Axitinib/FOLFOX

n = 36


(ITT population)

Bevacizumab/FOLFOX

n = 35


(ITT population)

Received treatment

n = 36

Analyzed for safety

(Safety population)

Received treatment

n = 35

Analyzed for safety

(Safety population)

Discontinued study

n = 36

Died (n = 21)



Lost to follow-up (n =4)

Other (n = 10)

Discontinued study

n = 35

Died (n = 23)




Other (n = 7)

Prior irinotecan-containing regimen

n = 71

Discontinued axitinib

n = 36




Died (n = 1)

Other (n = 1)

Discontinued bevacizumab

n = 35





Died (n = 0)

Other (n = 3)

Clinical Colorectal Cancer December 2013 - 241


242 -

CochraneManteleHaenszel method. All patients receiving � 1dose of study medication were included in the safety analysis.

ResultsPatient Characteristics

Between March 2008 and August 2009, 171 patients were ran-domized to treatment with axitinib/FOLFIRI (n ¼ 49), bev-acizumab/FOLFIRI (n ¼ 51), axitinib/FOLFOX (n ¼ 36), orbevacizumab/FOLFOX (n ¼ 35) (Fig. 1). The study was stoppedearly because of slow enrollment in FOLFOX arms. As of July 29,2011 (data cutoff date), 4 patients were still receiving treatment.Patient characteristics (Table 1) were generally similar among arms,although there was a higher proportion of female patients in theaxitinib/FOLFOX arm. Approximately 40% of the patients in eacharm had previously received bevacizumab.

Treatment ExposurePatients in the axitinib/FOLFIRI arm received fewer chemo-

therapy cycles than those in the bevacizumab/FOLFIRI arm (8 and12 cycles, respectively), whereas those in the axitinib/FOLFOX andbevacizumab/FOLFOX arms received similar numbers of cycles(11.5 and 10 cycles, respectively) (Table 2). Dose reductions andinterruptions were more common with axitinib than bevacizumab(Table 2). Disease progression was the most common reason foraxitinib or bevacizumab discontinuation (Fig. 1).

Table 1 Patient Baseline Demographics

Axitinib/FOLFIRIn [ 49

Bevacizumn [

Male/female, n (%) 31 (63.3)/18 (36.7) 27 (52.9

Age, years, median (range) 59.0 (24-76) 58.0

Race, n (%)

White 36 (73.5) 41

Black 1 (2.0) 2

Asian 11 (22.4) 8

Other 1 (2.0) 0

ECOG PS, n (%)

0 36 (73.5) 36

1 13 (26.5) 15

Prior Surgery, n (%)

Yes 47 (95.9) 43

No 2 (4.1) 7

NR 0 (0) 1

Prior Radiation Therapy, n (%)

Yes 6 (12.2) 6

No 41 (83.7) 42

NR 2 (4.1) 3

Prior Adjuvant Therapy, n (%)

Yes 11 (22.4) 9

No 38 (77.6) 42

Prior bevacizumab therapy, n (%) 19 (38.8) 20

Abbreviations: ECOG PS ¼ Eastern Cooperative Oncology Group performance status; FOLFIRI ¼reported.


EfficacyNo significant difference was noted in PFS between axitinib/

FOLFIRI and bevacizumab/FOLFIRI (hazard ratio [HR], 1.27; 95%confidence interval [CI], 0.77-2.11; 1-sided P ¼ .83) or betweenaxitinib/FOLFOX and bevacizumab/FOLFOX (HR, 1.04; 95%CI, 0.55-1.96; 1-sided P ¼ .55) (Table 3, Fig. 2). Median PFS was5.7 months (95% CI, 4.5-7.4) with axitinib/FOLFIRI, 6.9 months(95% CI, 5.0-9.1) with bevacizumab/FOLFIRI, 7.6 months (95%CI, 5.6-11.1) with axitinib/FOLFOX, and 6.4 months (95% CI,5.5-13.1) with bevacizumab/FOLFOX (Table 3).

Median OS was 12.9 months (95% CI, 10.2-16.6) with axitinib/FOLFIRI vs. 15.7 months (95% CI, 12.1-23.0) with bevacizumab/FOLFIRI (HR, 1.36; 95% CI, 0.82-2.24; 1-sided P ¼ .88) and17.1 months (95% CI, 10.1-23.0) with axitinib/FOLFOX vs. 14.1months (95% CI, 9.0-16.4) with bevacizumab/FOLFOX (HR,0.69; 95% CI, 0.37-1.27; 1-sided P ¼ .12) (Table 3, Fig. 2).

The ORR was similar in all study arms: 24.5% of patients re-ceiving axitinib/FOLFIRI, 23.5% of patients receiving bevacizumab/FOLFIRI, 19.4% of patients receiving axitinib/FOLFOX, and20.0% of patients receiving bevacizumab/FOLFOX. Two completeresponses were reported (1 each in the axitinib/FOLFIRI and bev-acizumab/FOLFIRI arms).

In a subgroup analysis of a small number of patients, those withprior bevacizumab exposure had numerically shorter PFS thanbevacizumab-naïve patients across all arms (Table 3). Patients with

ab/FOLFIRI51

Axitinib/FOLFOXn [ 36

Bevacizumab/FOLFOXn [ 35

)/24 (47.1) 16 (44.4)/20 (55.6) 24 (68.6)/11 (31.4)

(34-80) 58.5 (25-75) 60.0 (41-77)

(80.4) 28 (77.8) 27 (77.1)

(3.9) 2 (5.6) 0 (0)

(15.7) 6 (16.7) 7 (20.0)

(0) 0 (0) 1 (2.9)

(70.6) 24 (66.7) 25 (71.4)

(29.4) 12 (33.3) 10 (28.6)

(84.3) 34 (94.4) 33 (94.3)

(13.7) 2 (5.6) 2 (5.7)

(2.0) 0 (0) 0 (0)

(11.8) 4 (11.1) 5 (14.3)

(82.4) 30 (83.3) 30 (85.7)

(5.9) 2 (5.6) 0 (0)

(17.6) 9 (25.0) 13 (37.1)

(82.4) 27 (75.0) 22 (62.9)

(39.2) 15 (41.7) 15 (42.9)

5-fluorouracil/leucovorin/irinotecan; FOLFOX ¼ 5-fluorouracil/leucovorin/oxaliplatin; NR ¼ not

Table 2 Treatment Exposure


Bevacizumab/FOLFIRIn [ 51



Chemotherapy

No. of cycles, median (range) 8 (1-33) 12 (1-65) 11.5 (2-36) 10 (1-45)

Dose per cycle, mg, mean (SD)

5-FU infusion 3856 (719.3) 3976 (835.3) 3936 (808.1) 4238 (539.2)

5-FU bolus 643.5 (124.5) 663.7 (136.4) 659.7 (132.1) 720 (95.2)

Leucovorin 633.4 (187.1) 577 (220.0) 622.5 (183.0) 684.7 (300.7)

Irinotecan 290.6 (68.4) 298.4 (61.6) e e

Oxaliplatin e e 139.9 (27.9) 150.2 (20.4)

Axitinib

Days on drug, median (range) 125 (8-599) e 178 (12-590) e

Daily dose, mg, median (range) 9.2 (4.5-17.6) e 9.7 (5.9-19.4) e

Relative dose intensity, median % 86.1 e 92.1 e

Dose increase, n (%) 7 (15.2) e 9 (25.0) e

Dose interruption, n (%) 34 (73.9) e 32 (88.9) e

Dose reduction, n (%) 20 (43.5) e 8 (22.2) e

Bevacizumab

No. of cycles, median (range) e 12 (1-65) e 10 (1-45)

Dose per cycle, mg/kg, median (range) e 5.0 (4.0-5.6) e 5.0 (5.0-5.6)

Relative dose intensity, median % e 100 e 100

Dose increase, n (%) e 3 (5.9) e 3 (8.6)

Dose interruption, n (%) e 2 (3.9) e 3 (8.6)

Dose reduction, n (%) e 6 (11.8) e 0

Abbreviations: 5-FU ¼ 5-fluorouracil; FOLFIRI ¼ 5-fluorouracil/leucovorin/irinotecan; FOLFOX ¼ 5-fluorouracil/leucovorin/oxaliplatin; SD ¼ standard deviation.


prior bevacizumab exposure had numerically longer median PFSwith axitinib/FOLFOX vs. bevacizumab/FOLFOX (6.1 vs. 5.5months). Bevacizumab-naïve patients had numerically longermedian PFS with bevacizumab/FOLFIRI vs. axitinib/FOLFIRI (9.0vs. 7.4 months).

SafetySerious AEs occurred in 37.0% and 17.6% of patients receiving

axitinib/FOLFIRI and bevacizumab/FOLFIRI, respectively, andin 27.8% and 14.3% of patients receiving axitinib/FOLFOX andbevacizumab/FOLFOX, respectively. Axitinib was discontinuedbecause of AEs in 8 patients (17.4%) receiving FOLFIRI and 6patients (16.7%) receiving FOLFOX; bevacizumab was dis-continued because of AEs in 3 patients (5.9%) receiving FOLFIRIand 2 patients (5.7%) receiving FOLFOX.

Nausea, diarrhea, neutropenia, and fatigue were frequentlyreported in all treatment arms (Table 4). In general, the incidence ofall-grade and grade � 3 AEs appeared lower with FOLFOX thanFOLFIRI in patients receiving axitinib or bevacizumab.

AEs occurring in more patients in axitinib arms vs. bevacizumabarms included diarrhea, fatigue, decreased appetite, vomiting,and hypertension. Nausea was experienced by similar propor-tions of patients receiving axitinib or bevacizumab in both FOL-FIRI and FOLFOX arms, although grade 3/4 nausea was higherwith axitinib. In patients receiving FOLFIRI, neutropenia wasreported by slightly fewer individuals receiving axitinib; incontrast, in the FOLFOX arms, neutropenia occurred in

approximately twice as many axitinib-treated vs. bevacizumab-treated patients. Proteinuria occurred in 8 patients (17.4%) (2[4.3%] grade � 3) and 5 patients (9.8%) (2 [3.9%] grade � 3) inthe axitinib/FOLFIRI and bevacizumab/FOLFIRI arms, respec-tively, and in 2 patients (5.6%) (none with grade � 3) and 3patients (8.6%) (none with grade � 3) in the axitinib/FOLFOXand bevacizumab/FOLFOX arms, respectively. Hypothyroidismoccurred in 7 patients (15.2%) (1 [2.2%] grade � 3) in the axi-tinib/FOLFIRI arm and 9 patients (25.0%) (1 [2.8%] grade � 3)in the axitinib/FOLFOX arm, with only 1 case reported inbevacizumab-treated patients.

In the axitinib/FOLFIRI and axitinib/FOLFOX arms, 32patients (69.6%) and 28 patients (77.8%), respectively, had axi-tinib dose interruptions due to AEs, and 20 patients (43.5%) and3 patients (8.3%), respectively, had axitinib dose reductions due toAEs. In the bevacizumab/FOLFIRI and bevacizumab/FOLFOXarms, 31 patients (60.8%) and 19 patients (54.3%), respectively,had bevacizumab dose interruptions due to AEs, and 4 patients(7.8%) and 0 patients, respectively, had bevacizumab dosereductions due to AEs.

DiscussionIn this phase II study, no significant differences in PFS (the

primary endpoint) were detected between axitinib and bevacizumabwhen combined with second-line FOLFOX or FOLFIRI; OS andORR results were consistent with PFS results; however, resultsshould be interpreted in the context of the open-label design, small


Table 3 Efficacy Measures





Median PFS, mo 5.7 6.9 7.6 6.4

HRa,b (95% CI) 1.27 (0.77-2.11) 1.04 (0.55-1.96)

P valueb,c .83 .55

Median OS, mo 12.9 15.7 17.1 14.1

HRa,b (95% CI) 1.36 (0.82-2.24) 0.69 (0.37-1.27)

P valueb,c .88 .12

Best response, n (%)

Complete response 1 (2.0) 1 (2.0) 0 (0) 0 (0)

Partial response 11 (22.4) 11 (21.6) 7 (19.4) 7 (20.0)

Stable disease 15 (30.6) 20 (39.2) 16 (44.4) 15 (42.9)

Progressive disease 11 (22.4) 15 (29.4) 8 (22.2) 8 (22.9)

Indeterminate 11 (22.4) 4 (7.8) 5 (13.9) 5 (14.3)

ORR, n (%) 12 (24.5) 12 (23.5) 7 (19.4) 7 (20.0)

95% CId (13.3-38.9) (12.8-37.5) (8.2-36) (8.4-36.9)

Prior bevacizumab treatment, n 19 20 15 15

Median PFS, mo 5.1 5.2 6.1 5.5

HRa (95% CI) 1.49 (0.68-3.30) 0.52 (0.17-1.60)

P valuee .84 .12

No prior bevacizumab treatment, n 30 31 21 20

Median PFS, mo 7.4 9.0 7.6 7.6

HRa (95% CI) 1.07 (0.57-2.00) 1.15 (0.55-2.43)

P valuee .59 .64

Abbreviations: CI ¼ confidence interval; ECOG PS ¼ Eastern Cooperative Oncology Group performance status; FOLFIRI ¼ 5-fluorouracil/leucovorin/irinotecan; FOLFOX ¼ 5-fluorouracil/leucovorin/oxaliplatin; HR ¼ hazard ratio; ORR ¼ objective response rate; OS ¼ overall survival; PFS ¼ progression-free survival.aAxitinib vs. bevacizumab.bStratified by ECOG PS (0 vs. 1) and prior treatment with bevacizumab (yes vs. no).c1-sided stratified log-rank test.dBased on F distribution.e1-sided unstratified log-rank test.


244 -

sample size, and inclusion of both bevacizumab-pretreated andbevacizumab-naïve patients.

Benefits of second-line antiangiogenic therapy in combinationwith chemotherapy were originally demonstrated in the ECOG-3200 study,5 which randomized bevacizumab-naïve patients toFOLFOX/bevacizumab, FOLFOX, or bevacizumab after first-lineprogression on irinotecan-based therapy. Results showed animprovement in PFS (7.3 vs. 4.7 months) and OS (12.9 vs. 10.8months) in patients who received FOLFOX/bevacizumab vs.FOLFOX, respectively. Response rate also was improved (22.7% vs.8.6%, FOLFOX/bevacizumab vs. FOLFOX, respectively). In ourstudy, PFS, OS, and ORR in both bevacizumab-and axitinib-basedarms were similar to those seen in the bevacizumab/FOLFOX armof the ECOG-3200 study.

Other VEGFR tyrosine kinase inhibitors (TKIs) have beeninvestigated in mCRC, with only 1 showing OS benefit. In therandomized phase III CONFIRM-1 trial of FOLFOX4/vatalanib(PTK787/ZK222584) vs. FOLFOX4/placebo in patients withtreatment-naive mCRC, PFS and OS were similar among treat-ment arms.18 A first-line randomized phase III study (HORIZONIII) of FOLFOX/bevacizumab vs. FOLFOX/cediranib showed nosignificant differences in PFS (10.3 vs. 9.9 months, respectively) orOS (21.3 vs. 22.8 months, respectively).19 Another randomized


first-line trial of FOLFOX/bevacizumab vs. FOLFOX/sunitinibshowed no benefit in PFS (11.2 vs. 9.1 months, respectively), witha higher rate of AEs and resultant dose reductions in the sunitinibvs. bevacizumab arm.20 In the second-line setting, a randomizedphase II study (HORIZON I)21 evaluating cediranib (20 or 30 mg/d)/FOLFOX vs. bevacizumab/FOLFOX as second-line mCRCtherapy found no significant difference in PFS (5.8 vs. 7.2 vs. 7.8months, respectively) between these regimens. The randomizedphase III CONFIRM-2 trial22 of FOLFOX4/vatalanib vs. placeboin previously treated patients did not meet the primary endpoint ofimproved OS but showed increased PFS (5.6 vs. 4.2 months,respectively). In the refractory setting, the randomized phase 3CORRECT study of best supportive care plus single-agent regor-afenib or placebo8 showed improvements in PFS and OS (PFS 1.9vs. 1.7 months and OS 6.4 vs. 5.0 months, respectively). This wasthe first trial to show a benefit with a VEGFR TKI in patients withcolorectal cancer, perhaps because regorafenib was used as a singleagent rather than in combination with chemotherapy. Toxicitiesincluded hand-foot syndrome, fatigue, hypertension, and diarrhea,which are commonly seen with VEGFR TKIs. Overall, trials withVEGFR TKIs plus chemotherapy have not significantly improvedOS in patients with colorectal cancer, and this may be related totreatment intensity and AE profiles.

Table 4 Treatment-Emergent Adverse Events (All-Causality)a

n (%)





All Grades Grade ‡3 All Grades Grade ‡3 All Grades Grade ‡3 All Grades Grade ‡3Diarrhea 33 (71.7) 15 (32.6) 25 (49.0) 6 (11.8) 15 (41.7) 4 (11.1) 11 (31.4) 2 (5.7)

Nausea 27 (58.7) 6 (13.0) 30 (58.8) 4 (7.8) 18 (50.0) 2 (5.6) 16 (45.7) 0 (0)

Decreased appetite 24 (52.2) 5 (10.9) 16 (31.4) 0 (0) 13 (36.1) 5 (13.9) 7 (20.0) 0 (0)

Vomiting 22 (47.8) 5 (10.9) 18 (35.3) 4 (7.8) 12 (33.3) 2 (5.6) 3 (8.6) 0 (0)

Fatigue 21 (45.7) 5 (10.9) 18 (35.3) 2 (3.9) 12 (33.3) 4 (11.1) 9 (25.7) 1 (2.9)

Neutropenia 20 (43.5) 14 (30.4) 28 (54.9) 17 (33.3) 20 (55.6) 12 (33.3) 10 (28.6) 4 (11.4)

Hypertension 17 (37.0) 4 (8.7) 8 (15.7) 2 (3.9) 24 (66.7) 3 (8.3) 7 (20.0) 1 (2.9)

Asthenia 14 (30.4) 9 (19.6) 11 (21.6) 1 (2.0) 6 (16.7) 1 (2.8) 6 (17.1) 1 (2.9)

Decreased weight 13 (28.3) 3 (6.5) 4 (7.8) 0 (0) 5 (13.9) 1 (2.8) 2 (5.7) 0 (0)

Abdominal pain 12 (26.1) 2 (4.3) 11 (21.6) 2 (3.9) 7 (19.4) 1 (2.8) 9 (25.7) 0 (0)

Epistaxis 11 (23.9) 0 (0) 15 (29.4) 0 (0) 4 (11.1) 0 (0) 5 (14.3) 0 (0)

Stomatitis 11 (23.9) 2 (4.3) 7 (13.7) 0 (0) 9 (25.0) 0 (0) 7 (20.0) 0 (0)

Mucosal inflammation 7 (15.2) 4 (8.7) 11 (21.6) 3 (5.9) 4 (11.1) 0 (0) 3 (8.6) 0 (0)

Hypothyroidism 7 (15.2) 1 (2.2) 0 (0) 0 (0) 9 (25.0) 1 (2.8) 1 (2.9) 0 (0)

Constipation 7 (15.2) 0 (0) 16 (31.4) 0 (0) 3 (8.3) 0 (0) 10 (28.6) 0 (0)

Anemia 6 (13.0) 3 (6.5) 11 (21.6) 3 (5.9) 4 (11.1) 1 (2.8) 5 (14.3) 1 (2.9)

Alopecia 6 (13.0) 0 (0) 12 (23.5) 0 (0) 2 (5.6) 0 (0) 0 (0) 0 (0)

Thrombocytopenia 3 (6.5) 1 (2.2) 4 (7.8) 1 (2.0) 8 (22.2) 3 (8.3) 5 (14.3) 1 (2.9)

Peripheral neuropathy 3 (6.5) 0 (0) 4 (7.8) 1 (2.0) 7 (19.4) 0 (0) 16 (45.7) 3 (8.6)

Abbreviations: FOLFIRI ¼ 5-fluorouracil/leucovorin/irinotecan; FOLFOX ¼ 5-fluorouracil/leucovorin/oxaliplatin.aReported in � 20% of patients in any treatment group.

Figure 2 Kaplan-Meier Plots of Progression-free and Overall Survival. KaplaneMeier Plots of PFS (A, B) and OS (C, D). (A, C) Axitinibor Bevacizumab Plus FOLFIRI. (B, D) Axitinib or Bevacizumab Plus FOLFOX

A B

DC




246 -

Two randomized clinical trials have reported benefit of continuedantiangiogenic therapy into the second-line setting. The randomizedphase III VELOUR study showed prolonged PFS and OS inpatients treated with aflibercept/FOLFIRI vs. placebo/FOLFIRI(PFS 6.90 vs. 4.67 months and OS 13.50 vs. 12.06 months,respectively).7 These benefits were maintained in patients who hadfailed prior bevacizumab. The phase III ML18147 trial randomizedpatients with mCRC who previously received bevacizumab tosecond-line chemotherapy with or without bevacizumab6 andshowed continuation of bevacizumab into the second-line settingimproved PFS and OS (PFS 5.7 vs. 4.1 months and OS 11.2 vs. 9.8months, respectively). In the present study, bevacizumab-exposedpatients had shorter PFS than bevacizumab-naïve patients in allarms, although the differences were not tested statistically.

The safety profile of axitinib in the current study was similar tothat previously reported in other tumor types.10-14 Diarrhea,decreased appetite, hypertension, and fatigue were common andhave been associated with other VEGFR TKIs plus chemotherapyin patients with mCRC.19,20 The rates of these toxicities werehigher in the axitinib arms compared with the bevacizumab arms.There was also an increased incidence of hypothyroidism in theaxitinib arms. The higher incidence of these toxicities in the axitinibarms compared with the bevacizumab arms may account for higherrates of dose modification, interruptions, and discontinuations dueto AEs, as well as lower relative dose intensity in the axitinib arms.

ConclusionsThis study did not achieve its primary endpoint of significantly

improved PFSwith axitinib vs. bevacizumab in patients with treatment-refractory mCRC receiving chemotherapy and therefore does notprovide a strong signal to proceed to phase III clinical trials of axitinibplus chemotherapy in second-line mCRC. The results may have beeninfluenced by suboptimal axitinib dosing regimens and earlier discon-tinuations in the axitinib/FOLFIRI arm, likely secondary to increasedAEs. From this and previous studies in both first- and second-linetreatment of mCRC,19-21 it seems that VEGFR TKI plus chemo-therapy is not as well tolerated as bevacizumab-based regimens andmayresult in early discontinuations and decreased dose intensity of allagents. Although their efficacy may be due, in part, to the tolerabilityissues, current dosing regimens with VEGFR TKIs plus chemotherapydo not seem to offer benefit over bevacizumab-based regimens.

Clinical Practice Points
� Combining agents that inhibit the VEGF pathway (eg, bev-acizumab) with chemotherapy has been shown to improve out-comes compared with chemotherapy alone in patients withmCRC. Thus, we compared the efficacy and safety of axitinib, aVEGFR TKI, vs. bevacizumab in combination with chemo-therapy in previously treated patients with mCRC.
� These results are consistent with those reported from otherstudies of VEGF TKIs plus chemotherapy.

DisclosureThis study was sponsored by Pfizer Inc. Pfizer Inc participated in

the design of the study and collection and analysis of the data. Theauthors thank all the patients who participated in this study, the


physicians who referred them, and the study coordinators. Editorialsupport was provided by Giles Brooke, PhD, and Joseph Ram-charan, PhD, of Engage Scientific Solutions and was funded byPfizer Inc. C. Barone received a commercial research grant (PfizerInc), honoraria from speakers bureaus (Novartis, GlaxoSmithKline,Pfizer Inc, Bayer, Amgen, Merck, Roche), and has acted as aconsultant (Roche, Amgen, Merck). A. F. Sobrero has receivedhonoraria from speakers bureaus and acted as a consultant (Sanofi,Merck, Roche, Amgen). Z. A. Wainberg has received honorariafrom speakers bureaus and acted as a consultant (Pfizer Inc, Gen-entech). J. Bloom is an employee of Engage Scientific Solutions andwas a paid consultant to Pfizer Inc in connection with the devel-opment of this manuscript. J. Tarazi and B. Rosbrook areemployees of Pfizer Inc and own Pfizer Inc stock. A. D. Ricart wasemployed at Pfizer Inc at the time of the study described anddevelopment of this manuscript and owns Pfizer Inc stock. S. Kim,employed at Pfizer Inc at the time of the study described, iscurrently employed by Mirna Therapeutics and owns stock inPfizer Inc and Mirna Therapeutics. J. C. Bendell, C. Tournigand,A. Swieboda-Sadlej, J. G. Kim, C. Pericay, and D. Pastorelli haveno conflicts of interest to disclose. Presented in part as a poster at:the Gastrointestinal Cancers Symposium (American Society ofClinical OncologyeGastrointestinal Cancers), January 20-22,2011, San Francisco, California.

References1. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in

Oncology� Colon Cancer v.3. January 2012. Available at: http://www.nccn.org/index.asp. Accessed June 12, 2012.

2. Tournigand C, André T, Achille E, et al. FOLFIRI followed by FOLFOX6 or thereverse sequence in advanced colorectal cancer: a randomized GERCOR study.J Clin Oncol 2004; 22:229-37.

3. Hess GP, Wang PF, Quach D, et al. Systemic therapy for metastatic colorectalcancer: patterns of chemotherapy and biologic therapy use in US medical oncologypractice. J Oncol Pract 2010; 6:301-7.

4. Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan,fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004;350:2335-42.

5. Giantonio BJ, Catalano PJ, Meropol NJ, et al. Bevacizumab in combination withoxaliplatin, fluorouracil, and leucovorin (FOLFOX4) for previously treated met-astatic colorectal cancer: results from the Eastern Cooperative Oncology GroupStudy E3200. J Clin Oncol 2007; 25:1539-44.

6. Bennouna J, Sastre J, Arnold D, et al. Continuation of bevacizumab after firstprogression in metastatic colorectal cancer (ML18147): a randomised phase 3 trial.Lancet Oncol 2013; 14:29-37.

7. Van Cutsem E, Tabernero J, Lakomy R, et al. Addition of aflibercept to fluoro-uracil, leucovorin, and irinotecan improves survival in a phase III randomized trialin patients with metastatic colorectal cancer previously treated with an oxaliplatin-based regimen. J Clin Oncol 2012; 30:3499-506.

8. Grothey A, Van Cutsem E, Sobrero A, et al. Regorafenib monotherapy for pre-viously treated metastatic colorectal cancer (CORRECT): an international, mul-ticentre, randomised, placebo-controlled, phase 3 trial. Lancet 2013; 381:303-12.

9. Hu-Lowe DD, Zou HY, Grazzini ML, et al. Nonclinical antiangiogenesis andantitumor activities of axitinib (AG-013736), an oral, potent, and selectiveinhibitor of vascular endothelial growth factor receptor tyrosine kinases 1, 2, 3.Clin Cancer Res 2008; 14:7272-83.

10. Cohen EE, Rosen LS, Vokes EE, et al. Axitinib is an active treatment for allhistologic subtypes of advanced thyroid cancer: results from a phase II study. J ClinOncol 2008; 26:4708-13.

11. Fruehauf J, Lutzky J, McDermott D, et al. Multicenter, phase II study of axitinib,a selective second-generation inhibitor of vascular endothelial growth factorreceptors 1, 2, and 3, in patients with metastatic melanoma. Clin Cancer Res 2011;17:7462-9.

12. Rini BI, Wilding G, Hudes G, et al. Phase II study of axitinib in sorafenib-refractory metastatic renal cell carcinoma. J Clin Oncol 2009; 27:4462-8.

13. Rixe O, Bukowski RM, Michaelson MD, et al. Axitinib treatment in patients withcytokine-refractory metastatic renal-cell cancer: a phase II study. Lancet Oncol2007; 8:975-84.

14. Schiller JH, Larson T, Ou SH, et al. Efficacy and safety of axitinib in patients withadvanced non-small-cell lung cancer: results from a phase II study. J Clin Oncol2009; 27:3836-41.

http://www.nccn.org/index.asp

http://www.nccn.org/index.asp

http://refhub.elsevier.com/S1533-0028(13)00089-3/sref1










































15. Sharma S, Abhyankar V, Burgess RE, et al. A phase I study of axitinib
(AG-013736) in combination with bevacizumab plus chemotherapy or chemo-therapy alone in patients with metastatic colorectal cancer and other solid tumors.Ann Oncol 2010; 21:297-304.

16. Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluatethe response to treatment in solid tumors. European Organization forResearch and Treatment of Cancer, National Cancer Institute of the UnitedStates, National Cancer Institute of Canada. J Natl Cancer Inst 2000; 92:205-16.

17. Trotti A, Colevas AD, Setser A, et al. CTCAE v3.0: development of a compre-hensive grading system for the adverse effects of cancer treatment. Semin RadiatOncol 2003; 13:176-81.

18. Hecht JR, Trarbach T, Hainsworth JD, et al. Randomized, placebo-controlled,phase III study of first-line oxaliplatin-based chemotherapy plus PTK787/ZK222584, an oral vascular endothelial growth factor receptor inhibitor, in patientswith metastatic colorectal adenocarcinoma. J Clin Oncol 2011; 29:1997-2003.

19. Schmoll HJ, Cunningham D, Sobrero A, et al. Cediranib with mFOLFOX6 versusbevacizumab with mFOLFOX6 as first-line treatment for patients with advancedcolorectal cancer: a double-blind, randomized phase III study (HORIZON III).J Clin Oncol 2012; 30:3588-95.

20. Hecht JR, Yoshino T, Mitchell EP, et al. A randomized, phase IIb study ofsunitinib plus 5-fluorouracil, leucovorin, and oxaliplatin (mFOLFOX6) versusbevacizumab plus mFOLFOX6 as first-line treatment for metastatic colorectalcancer (mCRC): interim results. Paper presented at: 46th Annual Meeting of theAmerican Society of Clinical Oncology, June 4-8, 2010, Chicago, IL.

21. Cunningham D, Wong RP, D’Haens G, et al. Cediranib with mFOLFOX6 vsbevacizumab with mFOLFOX6 in previously treated metastatic colorectal cancer.Br J Cancer 2013; 108:493-502.

22. Van Cutsem E, Bajetta E, Valle J, et al. Randomized, placebo-controlled, phase IIIstudy of oxaliplatin, fluorouracil, and leucovorin with or without PTK787/ZK222584 in patients with previously treated metastatic colorectal adenocarcinoma.J Clin Oncol 2011; 29:2004-10.





























axitinib or bevacizumab plus folfiri or modified folfox-6 after failure of first-line therapy for...

Documents