Oxaliplatin and 5-FU/Folinic Acid (Modified FOLFOX6) with Or Without Aflibercept in First-Line Treatment of Patients with Metast

Annals of Oncology original articles

Annals of Oncology 27: 1273–1279, 2016 doi:10.1093/annonc/mdw176 Published online 18 April 2016

Oxaliplatin and 5-FU/folinic acid (modified FOLFOX6) with or without aflibercept in first-line treatment of patients with metastatic colorectal cancer: the AFFIRM study G. Folprecht1*, C. Pericay2, M. P. Saunders3, A. Thomas4, R. Lopez Lopez5,J.K.Roh6, V. Chistyakov7, T. Höhler8, J.-S. Kim9, R.-D. Hofheinz10, S. P. Ackland11,12, D. Swinson13, M. Kopp14, D. Udovitsa15, M. Hall16, T. Iveson17, A. Vogel18 & J. R. Zalcberg19 1Medical Department I, University Cancer Center, University Hospital Carl Gustav Carus, Dresden, Germany; 2Hospital de Sabadell, Corporació Sanitaria Parc Taulí-Institut Universitari, Sabadell, Spain; 3Department of Radiotherapy and Oncology, The Christie NHS Foundation Trust, Manchester; 4Department of Cancer Studies, University of Leicester, Leicester, UK; 5Department of Medical Oncology, Hospital Clinico Universitario e Instituto de Investigación, Santiago de Compostela, Spain; 6Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Republic of Korea; 7Pyatigorsk Cancer Dispensary, Stavropol, Russia; 8Department I of Internal Medicine, Prosper Hospital, Recklinghausen, Germany; 9Department of Oncology and Hematology, Korea University Guro Hospital, Seoul, Republic of Korea; 10Department III of Internal Medicine, University Hospital, Mannheim, Germany; 11Department of Medical Oncology, Calvary Mater Hospital, Newcastle; 12Hunter Medical Research Institute and University of Newcastle, Callaghan, Australia; 13Department of Oncology, St James’ Hospital, Leeds, UK; 14Samara Regional Oncology Dispensary, Samara; 15Oncological Dispensary #2, Sochi, Russia; 16Cancer Services Division, Mount Vernon Cancer Centre, Middlesex; 17Department of Medical Oncology, University Hospital Southampton NHS Foundation Trust, Southampton, UK; 18Clinic of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, Hannover, Germany; 19School of Public Health and Preventive Medicine, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia

Received 19 August 2015; revised 12 February 2016 and 25 March 2016; accepted 10 April 2016

Background: The combination of aflibercept with FOLFIRI has been shown to significantly prolong overall survival in patients with metastatic colorectal cancer (mCRC) after progression on oxaliplatin-based therapy. This trial evaluated the addition of aflibercept to oxaliplatin-based first-line treatment of patients with mCRC. Patients and methods: Patients with mCRC were randomized to receive first-line therapy with mFOLFOX6 plus aflibercept (4 mg/kg) or mFOLFOX6 alone. The primary end point of this phase II study was the progression-free survival (PFS) rate at 12 months in each arm. The analysis of efficacy between the arms was a pre-planned secondary analysis. Results: Of 236 randomized patients, 227 and 235 patients were evaluable for the primary efficacy analysis and safety, respectively. The probabilities of being progression-free at 12 months were 25.8% [95% confidence interval (CI) 17.2– 34.4] for the aflibercept/mFOLFOX6 arm and 21.2% (95% CI 12.2–30.3) for the mFOLFOX6 arm. The median PFS was 8.48 months (95% CI 7.89–9.92) for the aflibercept/mFOLFOX6 arm and 8.77 months (95% CI 7.62–9.27) for the mFOLFOX6 arm; the hazard ratio of aflibercept/mFOLFOX6 versus mFOLFOX6 was 1.00 (95% CI 0.74–1.36). The response rates were 49.1% (95% CI 39.7–58.6) and 45.9% (95% CI 36.4–55.7) for patients treated with and without aflibercept, respectively. The most frequent treatment-emergent grade 3/4 adverse events (AEs) excluding laboratory abnormalities reported for aflibercept/mFOLFOX6 versus mFOLFOX6 were neuropathy (16.8% versus 17.2%) and diarrhea (13.4% versus 5.2%). Neutropenia grade 3/4 occurred in 36.1% versus 29.3%. The most common vascular endothelial growth factor inhibition class-effect grade 3/4 AEs for aflibercept/mFOLFOX6 versus mFOLFOX6 were hypertension (35.3% versus 1.7%), proteinuria (9.2% versus 0%), deep vein thrombosis (5.9% versus 0.9%) and pulmonary embolism (5.9% versus 5.2%). Conclusion: No difference in PFS rate was observed between treatment groups. Adding aflibercept to first-line mFOLFOX6 did not increase efficacy but was associated with higher toxicity. Clinical Trial Number: NCT00851084, www.clinicaltrials.gov, EudraCT 2008-004178-41. Key words: aflibercept, mFOLFOX6, angiogenesis, oxaliplatin, colorectal cancer

*Correspondence to: Dr Gunnar Folprecht, Medical Department I, University Cancer Center, University Hospital Carl Gustav Carus, Fetscherstr. 74, 01307 Dresden, Germany. Tel: +49-351-458-4794; E-mail: [email protected]

Downloaded from https://academic.oup.com/annonc/article-abstract/27/7/1273/1741921 by guest on 12 February 2018 original articles Annals of Oncology

introduction Imaging for tumor assessments was carried out every 8 weeks until disease progression (according to investigator assessment) and centrally fi In patients with metastatic colorectal cancer (mCRC), rst-line reviewed by an Independent Review Committee (IRC) blinded to study treat- treatment options usually involve combination chemotherapy ment and carried out according to RECIST 1.0. All time-dependent events regimens comprising infusional fluorouracil (5-FU) and leucov- were calculated from randomization. Adverse events (AEs) were reported orin plus either irinotecan (FOLFIRI) or oxaliplatin (FOLFOX) using the Medical Dictionary for Regulatory Activities (MedDRA) version [1]. The addition of an anti-angiogenic agent is noted to mod- 14.0 and toxicity was graded according to NCI common toxicity criteria estly improve treatment outcome in the first-line setting. The (NCI-CTC) version 3.0. anti-VEGF-A (vascular endothelial growth factor-A) antibody The primary objective of the study was to estimate the PFS rate at 12 bevacizumab improves overall survival (OS) when combined months according to central review. The trial was not powered to demonstrate first line with either the irinotecan/bolus 5-FU/folinic acid IFL superiority of one treatment arm over the other. As a pre-planned exploratory regimen [2] or 5-FU/folinic acid [3]. The combination of bevaci- analysis, the Kaplan–Meier estimates for PFS and OS were computed for each zumab with an oxaliplatin-based regimen was associated with treatment group. The hazard ratios (aflibercept versus control) and 95% confi- longer progression-free survival (PFS) in the first-line setting, dence interval (CI) were estimated using the Cox proportional hazards model, fi but OS differences did not reach statistical significance, and re- after adjusting according to the above-mentioned strati cation factors. Overall sponse rate was not improved by the addition of bevacizumab response rate (ORR) was estimated in each treatment group. [4]. In the treatment of second-line mCRC, the addition of bev- The anticipated median PFS in the mFOLFOX6 arm was 8.5 months. A risk reduction of 23% was expected in the aflibercept/mFOLFOX6 arm, cor- acizumab to FOLFOX4 has been shown to improve OS in a bev- responding to a median PFS of 11 months or a PFS rate of 46.9% at 12 acizumab-naïve population previously treated with irinotecan/ months for aflibercept/mFOLFOX6 and 37.6% for mFOLFOX6. Considering 5-FU [5]. The ML18147 study has demonstrated the benefitof 9.4% precision for the 95% CI, 110 patients per arm were required. The total continuing bevacizumab, with a concomitant switch in the number of patients required was estimated to be 230 allowing for 10 patients chemotherapy backbone, in mCRC patients who had progressed being lost to follow-up. With this precision, the lower limit of the 95% CI of on a prior combination of an alternative chemotherapy back- the estimated PFS at 12 months in the aflibercept/mFOLFOX6 group was to bone plus bevacizumab [6]. be at least equal to the PFS at 12 months expected in the mFOLFOX6 arm. The fl fl A ibercept (VEGF-trap, US: ziv-a ibercept) is a recombi- final analysis was planned 1 year after the randomization of the last patient. nantly produced, fusion protein binding all isoforms of human The study was approved by each recruiting center’s ethics committee and the VEGF-A, VEGF-B and the placental growth factor (PlGF). It corresponding health authority. All patients provided signed informed consent interferes with the biological actions of VEGF by complexing to participate in the study before any study procedures were carried out. VEGF and preventing it from interacting with its receptors on endothelial cells [7]. Furthermore, clinical data from the multicenter, randomized, placebo-controlled, phase III VELOUR results trial, demonstrated a significant benefit from the addition of afli- patient population bercept to FOLFIRI in terms of median OS, PFS and response rate in patients with mCRC who had progressed on prior oxali- Between March 2009 and April 2010, 236 patients were enrolled platin-based therapy [8]. Moreover, this survival benefitwas from 36 centers in Australia, Germany, Italy, Republic of Korea, noted both in patients who were bevacizumab-naïve and those Russian Federation, Spain and UK. One hundred and nineteen fl who were previously treated with bevacizumab [8, 9]. patients were randomized to receive a ibercept/mFOLFOX6, The present randomized, phase II AFFIRM study reports on and 117 patients to receive mFOLFOX6 alone. According to the the efficacy and safety of aflibercept in combination with a protocol, 235 patients were included in the safety population, fi modified FOLFOX6 regimen (mFOLFOX6) when used in the and 227 patients were evaluable for the primary ef cacy analysis first-line setting for the treatment of patients with mCRC. (PFS according to central review, CONSORT diagram, supplementary Figure S1, available at Annals of Oncology online). Of these 227 patients, 93 (40.9%) provided samples of tumor DNA methods for genetic mutation profiling, 47 (50.5%) of whom had been fl AFFIRM was a randomized, open-label phase II study of aflibercept in add- treated with a ibercept. Thirty-one patients (33%) had tumors ition to the modified (m) FOLFOX6 regimen or mFOLFOX6 alone as first- with a KRAS mutation, three patients (3%) a BRAF and three line treatment in patients with mCRC. Further patient inclusion criteria are patients (3%) a NRAS mutation [10]. described in supplementary Methods, available at Annals of Oncology online. Overall, 97.0% of patients had an ECOG performance status of 0 Patients were centrally randomized via an interactive voice response or 1. The median age was 62.5 (29–87) years. Almost 10% of patients system using permuted block randomization stratified by Eastern Coopera- had received prior adjuvant therapy, 61.0% of patients were male tive Oncology Group (ECOG) performance status (0–1 versus >1), previous and 25.8% had liver only metastases. There were no major differ- adjuvant therapy (yes or no) and liver metastases only (yes or no). ences in patients’ characteristics across the treatment arms (Table 1). All patients received mFOLFOX6 {85 mg/m2 oxaliplatin [2 hours (h) Alkaline phosphatase levels were raised in over one-third of intravenous (i.v.)] together with 350 mg/m2 leucovorin (2 h i.v.) followed by all patients at baseline with an equal distribution between the 5-FU (400 mg/m2 as bolus and 2400 mg/m2 i.v. over 46 h)}. Patients in the two treatment arms. experimental arm received 4 mg/kg aflibercept (1 h i.v.) before chemotherapy. All drugs were given on day 1 of a 14-day cycle. Patients were to be treated until disease progression, unacceptable toxicity or patient refusal. drug exposure Patients experiencing toxicity from specific agents (i.e. neuropathy due to The median number of aflibercept cycles was 7.0 (range 1–43). oxaliplatin) were planned to continue treatment with the remaining drugs. In the mFOLFOX6 and the aflibercept/mFOLFOX6 arms, the

 | Folprecht et al. Volume 27 | No. 7 | July 2016 Downloaded from https://academic.oup.com/annonc/article-abstract/27/7/1273/1741921 by guest on 12 February 2018 Annals of Oncology original articles

Table 1. Patients’ baseline characteristics (95% CI 17.2–34.4) in the aflibercept/mFOLFOX6 arm, and was lower than anticipated in both arms. For PFS, there mFOLFOX6 Aflibercept/ All (n = 236) were 79 (71.2%) and 93 (80.2%) patients with events in the (n = 117) mFOLFOX6 fl (n = 119) mFOLFOX6 and a ibercept/mFOLFOX6 arms, respectively. The median PFS was 8.77 months (95% CI 7.62–9.27) for the a ECOG performance status mFOLFOX6 arm and 8.48 months (95% CI 7.89–9.92) for – 0 1 114 (97.4%) 115 (96.6%) 229 (97.0%) the aflibercept/mFOLFOX6 arm [stratified HR of mFOLFOX6 2 3 (2.6%) 4 (3.4%) 7 (3.0%) versus aflibercept/mFOLFOX6 1.00 (95% CI 0.74–1.36) a Liver-only metastases (Figure 1)]. Yes 24 (20.5%) 37 (31.1%) 61 (25.8%) Partial responses were observed in 45.9% (95% CI 36.4–55.7) No 93 (79.5%) 82 (68.9%) 175 (74.2%) and 49.1% (95% CI 39.7–58.6) of patients in the mFOLFOX6 Prior adjuvant chemotherapya and aflibercept/mFOLFOX6 arms, respectively, stable disease in No 105 (89.7%) 108 (90.8%) 213 (90.3%) 35.1% and 36.2% and progressive disease in 15.3% and 11.2%. Yes 12 (10.3%) 11 (9.2%) 23 (9.7%) Primary tumor Four patients per arm were not evaluable in the central review. Colon 58 (49.6%) 59 (49.6%) 117 (49.6%) In all randomized patients, the median follow-up time for OS Recto sigmoid 25 (21.4%) 34 (28.6%) 59 (25.0%) was 17.5 months and the survival probabilities at 12 months – Rectum 34 (29.1%) 26 (21.8%) 60 (25.4%) were 73.0% (95% CI 64.7 81.3) in the mFOLFOX6 arm and – fl Age 70.5% (95% CI 62.2 78.9) in the a ibercept/mFOLFOX6 arm Median (years, range) 63.0 (37–87) 62.0 (29–79) 62.5 (29–87) [stratified HR of aflibercept/mFOLFOX6 versus mFOLFOX6 <65 years 65 (55.6%) 70 (58.8%) 135 (57.2%) 0.98 (95% CI 0.66–1.45)]. The median OS was 22.3 months ≥65 but <75 years 43 (36.8%) 45 (37.8%) 88 (37.3%) [95% CI 15.6–non-calculable (n.c.)] in the mFOLFOX6 group ≥75 years 9 (7.7%) 4 (3.4%) 13 (5.5%) and 19.5 months (95% CI 15.6–n.c.) in the aflibercept/ Gender mFOLFOX6 group, but as this is based on only 101 events Male 68 (58.1%) 76 (63.9%) 144 (61.0%) (42.8% of patients), these estimates are imprecise with relatively Female 49 (41.9%) 43 (36.1%) 92 (39.0%) large CIs. Race Caucasian 90 (76.9%) 97 (81.5%) 187 (79.2%) Other 27 (23.1%) 22 (18.5%) 49 (20.8%) safety Region Overall, 75.0% of patients treated with mFOLFOX6 and 90.8% Western Europe 74 (63.2%) 75 (63.0%) 149 (63.1%) of patients treated with aflibercept/mFOLFOX6 had an AE Other countries 43 (36.8%) 44 (37.0%) 87 (36.9%) grade ≥3 (Table 2). Laboratory abnormalities were only reported as treatment-emergent AEs if they led to study discontinuation, aUpdated data presented as per eCRF, following errors in stratification dose modification or fulfilled a seriousness criterion. Most fre- at randomization. Incidences of stratification errors were similar in the quent grade ≥3 AEs in the experimental arm excluding labora- two treatment groups and mostly related to incorrect liver metastasis categorization. tory abnormalities were neutropenia (36.1% versus 29.3%), ECOG, Eastern Cooperative Oncology Group; mFOLFOX6, modified hypertension (35.3% versus 1.7%), peripheral neuropathy infusional fluorouracil and leucovorin plus oxaliplatin. (16.8% versus 17.2%) and diarrhea (13.4% versus 5.2%). The frequency of febrile neutropenia was less for patients receiving mFOLFOX6 alone (3.4%) than for those receiving aflibercept/ mFOLFOX6 (5.9%). Laboratory results showed a higher inci- – – median number of oxaliplatin cycles was 10.0 (1 31) and 9.0 (1 dence of grade 3/4 neutropenia in the aflibercept/mFOLFOX6 – 40), and the median number of 5-FU cycles was 11.0 (1 43) and group than in the mFOLFOX6 group (42.7% versus 31.0%, sup- – 10.0 (1 44), respectively (supplementary Figure S2, available at plementary Table S1, available at Annals of Oncology online). Annals of Oncology online). The median duration of exposure to One patient in the mFOLFOX6 arm and five patients treated fl – a ibercept was 17.1 weeks (range 2 94). In the mFOLFOX6 and with aflibercept/mFOLFOX6 died within 30 days of their last fl the a ibercept/mFOLFOX6 arms, the median duration of ex- dose of treatment due to AEs not regarded as disease progres- – – posure to oxaliplatin was 23.2 (2 77) and 22.0 (2 84), and to sion. One patient had an unrelated intracranial hemorrhage – – 5-FU was 25.9 (2 95) and 24.1 (2 106) weeks, respectively. At after 16 cycles (aflibercept had been discontinued after cycle 7); the time of analysis, all patients had discontinued study medi- all remaining patients died due to infections. All but one of fl cation. The most frequent reasons for mFOLFOX6 and a i- these (staphylococcal sepsis during cycle 14, aflibercept had bercept/mFOLFOX6 discontinuation were disease progression been discontinued after cycle 2) were judged by the investigator ’ (44.4% and 39.5%), AEs (22.2% and 30.3%), investigator sde- as not related to study treatment. ’ cision (11.1% and 11.8%), patient s request (9.4% and 10.1%) Grade ≥3 toxicities in the experimental arm considered to and metastatic surgery (5.1% and 5.0%), and for other reasons be a class effect of VEGF inhibition were hypertension (35.3% in 6.8% and 3.4% of patients, respectively. versus 1.7%), proteinuria (9.2% versus 0%), deep vein thrombosis (5.9% versus 0.9%), pulmonary embolism (5.9% versus fi treatment ef cacy 5.2%) and epistaxis (1.7% versus 0%). Other AEs of note The probability of being progression-free at 12 months was were rectal hemorrhage (0.8% versus 0%) and dysphonia 21.2% (95% CI 12.2–30.3) in the mFOLFOX6 arm and 25.8% grade 1–2 which was reported in only 2.6% of patients in the

Volume 27 | No. 7 | July 2016 doi:10.1093/annonc/mdw176 |  Downloaded from https://academic.oup.com/annonc/article-abstract/27/7/1273/1741921 by guest on 12 February 2018 original articles Annals of Oncology

A 1.0 Symbol = Censor 0.9 mFOLFOX6 0.8 Aflibercept/mFOLFOX6

0.7

0.6

0.5

0.4

Kaplan-meier estimate 0.3

0.2

0.1

0.0 03691215 18 21 24 Time (months) Number at risk mFOLFOX6 111 97 71 39 13 6 2 Aflib 116 106 73 49 22 7 4

B 1.0 Symbol = Censor 0.9 mFOLFOX6 0.8 Aflibercept/mFOLFOX6

0.7

0.6

0.5

0.4

Kaplan-meier estimate 0.3

0.2

0.1

0.0 03691215 18 2124 27 Time (months) Number at risk mFOLFOX6 117 115 106 89 80 52 30 13 3 Aflib 119 114 103 92 80 48 32 15 2

Figure 1. Progression-free and overall survival according to treatment. (A) Progression-free survival in the evaluable patient population and (B) overall survival in all randomized patients, according to treatment arm (mFOLFOX6 arm—black continuous, aflibercept/mFOLFOX6 arm—grey dotted. Aflib, aflibercept; mFOLFOX6, modified infusional fluorouracil and leucovorin plus oxaliplatin.

mFOLFOX6 arm compared with 18.5% of patients in the afli- In the VELOUR study, compared with chemotherapy alone, bercept/mFOLFOX6 arm. adding aflibercept to FOLFIRI significantly improved OS in patients in whom an oxaliplatin-containing regimen had previ- discussion ously failed (median 13.5 versus 12.1 months, HR 0.82; 95% CI 0.71–0.94) [8]. Aflibercept and other anti-angiogenic treatments have been shown In the present AFFIRM study, PFS rate at 12 months was the to prolong OS in pretreated patients with mCRC [5, 6, 11, 12]. primary end point. This PFS rate was lower than expected for

 | Folprecht et al. Volume 27 | No. 7 | July 2016 Downloaded from https://academic.oup.com/annonc/article-abstract/27/7/1273/1741921 by guest on 12 February 2018 Annals of Oncology original articles

Table 2. Treatment-emergent adverse events (grade ≥3 occurring in ≥3 patients or of special interest) Primary system organ class mFOLFOX6 (n = 116) Aflibercept/mFOLFOX6 (n = 119) Preferred term n (%) All grades Grades ≥3 All grades Grades ≥3

Any adverse event 115 (99.1) 87 (75.0) 119 (100) 108 (90.8) Infections and infestations Neutropenia 64 (55.2) 34 (29.3) 53 (44.5) 43 (36.1) Thrombocytopenia 25 (21.6) 2 (1.7) 14 (11.8) 4 (3.4) Anemia 11 (9.5) 4 (3.4) 11 (9.2) 3 (2.5) Leukopenia 10 (8.6) 3 (2.6) 11 (9.2) 6 (5.0) Febrile neutropenia 4 (3.4) 4 (3.4) 7 (5.9) 7 (5.9) Immune system disorders Drug hypersensitivity 6 (5.2) 3 (2.6) 7 (5.9) 4 (3.4) Gastrointestinal disorders Nausea 62 (53.4) 1 (0.9) 62 (52.1) 2 (1.7) Vomiting 29 (25.0) 4 (3.4) 35 (29.4) 3 (2.5) Diarrhea 51 (44.0) 6 (5.2) 69 (58.0) 16 (13.4) Stomatitis 44 (37.9) 4 (3.4) 60 (50.4) 6 (5.0) Constipation 31 (26.7) 1 (0.9) 37 (31.1) 3 (2.5) Abdominal pain 25 (21.6) 3 (2.6) 23 (19.3) 6 (5.0) Rectal hemorrhage 2 (1.7) 0 6 (5.0) 1 (0.8) General disorders and administration site conditions Fatigue 31 (26.7) 6 (5.2) 42 (35.3) 8 (6.7) Asthenia 30 (25.9) 6 (5.2) 24 (20.2) 9 (7.6) Nervous system disorders Hand–foot syndrome 6 (5.2) 0 20 (16.8) 3 (2.5) Peripheral sensory neuropathya 89 (76.7) 20 (17.2) 72 (60.5) 20 (16.8) Renal and urinary disorders Proteinuria 1 (0.9) 0 28 (23.5) 11 (9.2) Respiratory, thoracic and mediastinal disorders Epistaxis 15 (12.9) 0 34 (28.6) 2 (1.7) Dyspnea 8 (6.9) 0 22 (18.5) 3 (2.5) Dysphonia 3 (2.6) 0 22 (18.5) 0 Pulmonary embolism 6 (5.2) 6 (5.2) 7 (5.9) 7 (5.9) Vascular disorders Hypertension 8 (6.9) 2 (1.7) 64 (53.8) 42 (35.3) Deep vein thrombosis 2 (1.7) 1 (0.9) 7 (5.9) 7 (5.9)

aAlso including the terms neuropathy (peripheral), paresthesia and polyneuropathy. mFOLFOX6, modified infusional fluorouracil and leucovorin plus oxaliplatin.

both arms. Although the trial was not powered to demonstrate versus 22.2%) and may reflect the opinion of investigators that superiority, no difference was observed between the two treat- treatment beyond 6 months would be of limited benefit[13]. ment groups for PFS at 12 months (25.8% versus 21.2%), These relatively short treatment durations have been observed in median PFS (8.48 months versus 8.77 months) or response rate several recent colorectal cancer trials, including those in which (49.1% versus 45.9%) for patients treated with and without afli- the trial protocol recommended treatment until tumor progres- bercept, respectively. The study was not designed to detect dif- sion, e.g. FIRE-3, comparing the FOLFIRI combination with ferences in OS, and the follow-up time for OS was probably too cetuximab or bevacizumab (median duration of treatment for short to draw firm conclusions. all study agents 5.3 or 4.8 months, respectively) [14] and in the In AFFIRM, most patients were not treated until disease NO16966 trial (oxaliplatin-based chemotherapy with or without progression; the median number of cycles was 11 and 10 for bevacizumab, treatment duration 190 and 176 days, respectively) the mFOLFOX6 and aflibercept/mFOLFOX6-treated arms, re- [4]. The interval between the median treatment duration and spectively—shorter than the median PFS of 8.77 and 8.48 the median PFS is similar to the intervals reported for the treat- months. Only 44% and 39% of patients in the mFOLFOX6 and ment holidays (planned treatment pause until progression) in aflibercept/mFOLFOX6 arms discontinued treatment due to the CAIRO3 and AIO-KRK 0207 trials [15, 16]. In the NO16966 progressive disease, and a further 5% of patients due to metasta- trial (oxaliplatin-based chemotherapy plus/minus bevacizumab) sectomy. The remaining percentage of patients who discontin- [4], it was calculated, that the difference in PFS between the ued treatment is only partly explained by AEs (more frequent treatment arms might have been larger if all patients in with aflibercept/mFOLFOX6 than with mFOLFOX6, 30.3% both arms had been treated until progression [4]. The median

Volume 27 | No. 7 | July 2016 doi:10.1093/annonc/mdw176 |  Downloaded from https://academic.oup.com/annonc/article-abstract/27/7/1273/1741921 by guest on 12 February 2018 original articles Annals of Oncology

Table 3. First-line trials with FOLFOX and drugs interfering with the vascular endothelial growth factor pathway Study Study drug n Response rate PFS OS

AFFIRM mFOLFOX6/aflibercept 119 49% 8.5 n.r. mFOLFOX6 116 46% 8.8 n.r. HR 1.00 (0.74–1.36) HR 0.98 (0.66–1.45) NO16966 FOLFOX4/bevacizumab 349 38%a 9.4 21.2 Saltz et al. [4] FOLFOX4 351 38%a 8.6 20.3 HR 0.89 (0.73–1.08) HR 0.94 (0.75–1.16) CONFIRM-1 FOLFOX/vatalanib (PTK/ZK) 585 46% 7.7 21.4 Hecht et al. [18, 19] FOLFOX 583 42% 7.6 20.5 HR 0.88 (0.74–1.03) HR 1.08 (0.94–1.24) Van Cutsem [20] FOLFOX/BIBF112 85 64% 10.5 n.r. FOLFOX/bevacizumab 41 56% 15.4 n.r. HORIZON II FOLFOX or CAPOX/cediranib 502 51% 9.9 19.7 Hoff et al. [21] FOLFOX or CAPOX 358 50% 10.3 18.9 HR 0.84 (0.73–0.98) HR 0.94 (0.79–1.12) HORIZON III FOLFOX/cediranib 709 46% 9.9 22.8 Schmoll et al. [22] FOLFOX/bevacizumab 713 47% 10.3 21.3 HR 1.10 (0.97–1.25) HR 0.94 (0.79–1.12)

aReflects the numbers of FOLFOX/CAPOX with or without bevacizumab; numbers for FOLFOX only not fully published. CAPOX, capecitabine plus oxaliplatin; (m)FOLFOX, (modified) infusional fluorouracil and leucovorin plus oxaliplatin; HR, hazard ratio; n.r., not reported; PFS, progression-free survival; OS, overall survival; VEGF, vascular endothelial growth factor.

‘on-treatment PFS’ for the NO16966 trial was 10.4 months Moiseyenko, Sabe Sabesan, Javier Sastre, Pilar Garcia Alfonso, with chemotherapy plus bevacizumab versus 7.9 months with Jin Ho Baek, Libero Ciuffreda, Gabriella Farina, Joan Maurel, chemotherapy plus placebo (HR, 0.63; 97.5% CI 0.52–0.75; Chang Hak Sohn, Francesco Di Costanzo, Andrea Kerkhoff for P = 0.0001). However, treatment until progression in all patients enrolling patients into the trial and the study teams at all sites. does obviously not reflect clinical practice even though it was planned, e.g. in the AFFIRM, NO16966 and FIRE3 trials. funding The absence of significant differences in response rate and OS is consistent with all trials investigating anti-VEGF strategies in This study was funded by Sanofi, in collaboration with combination with first-line oxaliplatin-based schedules (Table 3). Regeneron Pharmaceuticals, Inc. (no grant number). Employees Furthermore, a significant survival benefit from the addition of of Sanofi were involved in the study design, the collection, ana- an anti-VEGF drug has only been demonstrated first line in lysis and interpretation of data. The Medical Affairs Depart- mCRC with fluoropyrimidine monotherapy [3, 16] and the IFL ments at Sanofi and Regeneron were allowed several reviews by regimen [2], the latter of which is rarely if ever used in the the authors for scientific accuracy, and provided feedback to the current era. Unfortunately, unlike KRAS/NRAS for EGFR-tar- authors for their consideration. Editorial assistance in the prep- geting antibodies, no predictive biomarkers for the benefitof aration of this manuscript was provided by Dr George Xinaria- VEGF inhibitors with any chemotherapy backbone have yet nos (Cancer Communications Ltd, Knutsford, UK), and funded been identified [14, 17]. In AFFIRM, treatment outcomes with by Sanofi (Cambridge, MA), in collaboration with Regeneron aflibercept appear to be independent of (K)RAS mutation status Pharmaceuticals, Inc. (Tarrytown, NY). Sanofi and Regeneron and other biomarkers [10]. were not involved in the writing or editing of this manuscript, Aflibercept, in combination with FOLFIRI, has shown efficacy and were not permitted to censor any content from the authors. for patients with mCRC that is resistant to or has progressed following an oxaliplatin-containing regimen. In the AFFIRM disclosure trial, adding aflibercept to FOLFOX-based first-line therapy did not increase efficacy but was associated with higher toxicity. GF received study grants (Merck KGaA, Pfizer), honoraria for This is consistent with the observations made with other anti- advisory boards (Merck KGaA, Roche/Genentech, Bristol- angiogenic agents in combination with FOLFOX in the first-line Myers-Squibb, Sanofi-Aventis) and lecture honoraria (Merck setting. KGaA, Roche/Genentech, Sanofi-Aventis, Novartis). MPS has received honoraria for an advisory board (Sanofi). RLL received fi acknowledgements study grants (Roche, Lilly, Sano ), honoraria for advisory boards (Merck, Novartis, Otsuka) and lecture honoraria (Merck, The authors would like to thank the patients and their families Otsuka). R-DH received study grants and personal fees (Sanofi) for participating in the trial, the investigators Jorge Aparicio, during the course of the study and study grants and personal Giuseppe Colucci, Francesco Ferrau, Jong-Gwang Kim, Sam fees (Sanofi, Merck, Amgen, Bayer and Roche) outside the sub- Yong Kim, Seung Taik Kim, Sun Young Kim, Vladimir mitted work. SPA has received honoraria and travel grants from

 | Folprecht et al. Volume 27 | No. 7 | July 2016 Downloaded from https://academic.oup.com/annonc/article-abstract/27/7/1273/1741921 by guest on 12 February 2018 Annals of Oncology original articles

Merck and Roche/Genentech. MK received study grants from 10. Lambrechts D, Thienpont B, Thuillier V et al. Evaluation of efficacy and safety Sanofi during the conduct of the study and grants outside the markers in a phase II study of metastatic colorectal cancer treated with aflibercept fi – submitted work (Amgen, F. Hoffmann-La Roche). MH has in the rst-line setting. Br J Cancer 2015; 113: 1027 1034. 11. Grothey A, Van Cutsem E, Sobrero A et al. Regorafenib monotherapy for previously received honoraria for advisory boards (Astra Zeneca, GSK, treated metastatic colorectal cancer (CORRECT): an international, multicentre, Roche/Genentech, Boehringer Ingelheim) and lecture honoraria randomised, placebo-controlled, phase 3 trial. Lancet 2013; 381: 303–312. (Roche/Genentech, GSK). TI received honoraria for advisory 12. Tabernero J, Yoshino T, Cohn AL et al. Ramucirumab versus placebo in boards (Roche/Genentech, Sanofi, Celgene, Amgen). AV re- combination with second-line FOLFIRI in patients with metastatic colorectal ceived honoraria for advisory boards (Roche/Genentech, carcinoma that progressed during or after first-line therapy with bevacizumab, Amgen, Lilly, Bayer, Sanofi) and lecture honoraria (Merck oxaliplatin, and a fluoropyrimidine (RAISE): a randomised, double-blind, KGaA, Amgen, Roche/Genentech, Lilly, Bayer, Sanofi). JRZ multicentre, phase 3 study. Lancet Oncol 2015; 16: 499–508. reports grants, personal fees and other from Bayer, grants and 13. Berry SR, Cosby R, Asmis T et al. Continuous versus intermittent chemotherapy strategies in metastatic colorectal cancer: a systematic review and meta-analysis. other from Merck Serono, grants and personal fees from Ann Oncol 2014; 26: 477–485. Amgen, grants, personal fees and other from Roche, outside the 14. Heinemann V, von Weikersthal LF, Decker T et al. FOLFIRI plus cetuximab versus submitted work. CP, AT, JKR, VC, TH, J-SK, DS and DU have FOLFIRI plus bevacizumab as first-line treatment for patients with metastatic nothing to disclose. colorectal cancer (FIRE-3): a randomised, open-label, phase 3 trial. Lancet Oncol 2014; 15: 1065–1075. 15. Simkens LH, van Tinteren H, May A et al. Maintenance treatment with capecitabine and bevacizumab in metastatic colorectal cancer (CAIRO3): a phase references 3 randomised controlled trial of the Dutch Colorectal Cancer Group. Lancet 2015; 1. Van Cutsem E, Cervantes A, Nordlinger B et al. Metastatic colorectal cancer: 385: 1843–1852. ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann 16. Hegewisch-Becker S, Graeven U, Lerchenmuller CA et al. Maintenance strategies Oncol 2014; 25(Suppl 3): iii1–iii9. after first-line oxaliplatin plus fluoropyrimidine plus bevacizumab for patients with 2. Hurwitz H, Fehrenbacher L, Novotny W et al. Bevacizumab plus irinotecan, metastatic colorectal cancer (AIO 0207): a randomised, non-inferiority, open-label, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004; phase 3 trial. Lancet Oncol 2015; 16: 1355–1369. 350: 2335–2342. 17. Schwartzberg LS, Rivera F, Karthaus M et al. PEAK: a randomized, multicenter 3. Kabbinavar FF, Hambleton J, Mass RD et al. Combined analysis of efficacy: the phase II study of panitumumab plus modified fluorouracil, leucovorin, and addition of bevacizumab to fluorouracil/leucovorin improves survival for patients oxaliplatin (mFOLFOX6) or bevacizumab plus mFOLFOX6 in patients with previously with metastatic colorectal cancer. J Clin Oncol 2005; 23: 3706–3712. untreated, unresectable, wild-type KRAS exon 2 metastatic colorectal cancer. J 4. Saltz LB, Clarke S, Diaz-Rubio E et al. Bevacizumab in combination with Clin Oncol 2014; 32: 2240–2247. oxaliplatin-based chemotherapy as first-line therapy in metastatic colorectal 18. Hecht JR, Trarbach T, Hainsworth JD et al. Randomized, placebo-controlled, phase cancer: a randomized phase III study. J Clin Oncol 2008; 26: 2013–2019. III study of first-line oxaliplatin-based chemotherapy plus PTK787/ZK 222584, an 5. Giantonio BJ, Catalano PJ, Meropol NJ et al. Bevacizumab in combination with oral vascular endothelial growth factor receptor inhibitor, in patients with oxaliplatin, fluorouracil, and leucovorin (FOLFOX4) for previously treated metastatic metastatic colorectal adenocarcinoma. J Clin Oncol 2011; 29: 1997–2003. colorectal cancer: results from the Eastern Cooperative Oncology Group Study 19. Hecht JR, Trarbach T, Jaeger E et al. A randomized, double-blind, placebo- E3200. J Clin Oncol 2007; 25: 1539–1544. controlled, phase III study in patients (Pts) with metastatic adenocarcinoma of the 6. Bennouna J, Sastre J, Arnold D et al. Continuation of bevacizumab after first colon or recturm receiving first-line chemotherapy with oxaliplatin/5-fluorouracil/ progression in metastatic colorectal cancer (ML18147): a randomised phase 3 leucovorin and PTK787/ZK 222584 or placebo (CONFIRM-1). J Clin Oncol 2005; trial. Lancet Oncol 2013; 14: 29–37. 23(16S): Abstr LBA3. 7. Papadopoulos N, Martin J, Ruan Q et al. Binding and neutralization of vascular 20. Van Cutsem E, Prenen H, D’Haens G et al. A phase I/II, open-label, randomised endothelial growth factor (VEGF) and related ligands by VEGF Trap, ranibizumab study of nintedanib plus mFOLFOX6 versus bevacizumab plus mFOLFOX6 in first- and bevacizumab. Angiogenesis 2012; 15: 171–185. line metastatic colorectal cancer patients. Ann Oncol 2015; 26(10): 2085–2091. 8. Van Cutsem E, Tabernero J, Lakomy R et al. Addition of aflibercept to fluorouracil, 21. Hoff PM, Hochhaus A, Pestalozzi BC et al. Cediranib plus FOLFOX/CAPOX versus leucovorin, and irinotecan improves survival in a phase III randomized trial in placebo plus FOLFOX/CAPOX in patients with previously untreated metastatic patients with metastatic colorectal cancer previously treated with an oxaliplatin- colorectal cancer: a randomized, double-blind, phase III study (HORIZON II). J Clin based regimen. J Clin Oncol 2012; 30: 3499–3506. Oncol 2012; 30: 3596–3603. 9. Tabernero J, Van Cutsem E, Lakomy R et al. Aflibercept versus placebo in 22. Schmoll HJ, Cunningham D, Sobrero A et al. Cediranib with mFOLFOX6 versus combination with fluorouracil, leucovorin and irinotecan in the treatment of bevacizumab with mFOLFOX6 as first-line treatment for patients with advanced previously treated metastatic colorectal cancer: prespecified subgroup analyses colorectal cancer: a double-blind, randomized phase III study (HORIZON III). J Clin from the VELOUR trial. Eur J Cancer 2014; 50: 320–331. Oncol 2012; 30: 3588–3595.

Volume 27 | No. 7 | July 2016 doi:10.1093/annonc/mdw176 |  Downloaded from https://academic.oup.com/annonc/article-abstract/27/7/1273/1741921 by guest on 12 February 2018