Annals of Oncology Advance Access published February 24, 2016 1
A phase III trial of exemestane plus bevacizumab maintenance therapy in patients with metastatic breast cancer after first-line taxane and bevacizumab: a GINECO group study
O. Trédan1,2, P. Follana3, I. Moullet4, C. Cropet5, S. Trager-Maury6, J. Dauba7, S. LavauDenes8, V. Diéras9, D. Béal-Ardisson10, M. Gouttebel11, H. Orfeuvre12, L. Stefani13, C. Jouannaud14, F. Bürki15, T. Petit16, E. Guardiola17, C. Becuwe18, E. Blot19, E. PujadeLauraine20, T. Bachelot1,2
Léon Bérard, 28 rue Laennec, 69008 Lyon, France
2CNRS
UMR5286, Cancer Research Center of Lyon, 69008 Lyon, France
3Centre
Antoine Lacassagne, 33 Avenue de Valombrose, 06189 Nice, France
4Clinique
de la Sauvegarde, Avenue David Ben Gourion, BP 309, 69337 Lyon, France
5Direction
of Clinical Research and Innovation, Centre Léon Bérard, 28 rue Laennec, 69008
Lyon, France 6Centre
Hospitalier de Senlis, Avenue Paul Rouge, 60309 Senlis, France
7Hôpital
Mont-De-Marsan, Avenue Pierre de Coubertin, 40024 Mont-De-Marsan, France
8CHU
Dupuytren, 2 Avenue Martin Luther King, 87042 Limoges, France
9Institut
Curie, 26 Rue d'Ulm, 75005 Paris, France
10Hôpital
Privé Jean Mermoz, 55 Avenue Jean Mermoz, 69373 Lyon, France
11Hôpitaux
Drôme Nord - Site de Romans, Avenue Geneviève de Gaulle-Anthonioz, 26102
Romans-sur-Isère, France 12Hôpital
Fleyriat, 900 Route de Paris, 01012 Bourg-en Bresse, France
13Centre
Hospitalier de la Région d'Annecy, 1 Avenue de l'Hôpital - BP 90074 - Metz-Tessy
74374 Pringy 14Institut
Jean Godinot, 1 Avenue du Général Koenig - BP 171, 51056 Reims
15Clinique
de l'Union, Boulevard de Ratelens, 31240 Saint-Jean, France
© The Author 2016. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
1Centre
2 16Centre
Paul Strauss, 3 Rue de la Porte de l'Hôpital - BP 30042, 67065 Strasbourg
17Centre
Hospitalier la Dracénie, Route de Montferrat - BP 249, 83307 Draguignan
18ORACLE 19Centre
- Centre d'Oncologie de Gentilly, 2 Rue Marie Marvingt, 54100 Nancy
Hospitalier Bretagne Atlantique, 20 Boulevard GI Maurice Guillaudot - BP 70555,
56017 Vannes 20Université
Paris Descartes, AP-HP, Hôpitaux Universitaires Paris Centre, Site Hôtel-Dieu, 1
Parvis Notre-Dame - Place Jean-Paul II, 75181 Paris
Bérard, Centre de Recherche en Cancérologie de Lyon, 28 rue Laennec, 69008, Lyon, France. Tel: (+33) 4 78 78 26 54, Fax: (+33) 4 78 78 27 16, Email: [email protected]
Abstract Background: Maintenance strategies beyond response or tumor stabilization with first line chemotherapy in metastatic breast cancer (MBC) have not been extensively studied. Endocrine therapy combined with continued bevacizumab may be a helpful option for estrogen receptor (ER)-positive MBC. Patients and Methods: In this prospective, open label, phase III study, patients with histologically confirmed ER-positive, HER2-negative MBC and non-progressive disease after 1624 weeks of taxane plus bevacizumab (T+BEV) were randomized to continuation of T+BEV or maintenance bevacizumab plus exemestane (E+BEV). The primary endpoint was progressionfree survival (PFS) from randomization. To have 80% power to detect an improvement in the 6month PFS (PFS6m) from 50% to 65%, 186 evaluable patients were needed for a total of 141 PFS events. An interim analysis was planned after 40% of the required events. Results: The interim analysis with 98 patients showed that the probability of reaching a statistically significant improvement in PFS by the end of the study was only 7%. This led the
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Corresponding author: Dr. Thomas Bachelot, Department of Medical Oncology, Centre Léon
3
Independent Data and Monitoring Committee to recommend termination of patient enrollment. After a median 21-months follow up of all randomized patients (117 in total), PFS6m from randomization was 67.2% (95% confidence interval (CI) 53.6, 77.7) with T+BEV and 55.2% (95% CI 41.5, 66.9) with E+BEV (HR: 1.0, 95% CI 0.7, 1.5, p=0.998). Median PFS from BEV initiation was 12.5 and 12.3 months in the T+ BEV and E + BEV arms, respectively. In the T+BEV arm, taxane was prematurely stopped for the majority of patients (94.9%), mainly due to toxicity (49.2%). Updated data after 35-months’ median follow-up showed death rates of 44%
Conclusion: In this trial, maintenance therapy with E+BEV in ER-positive, HER2-negative MBC patients with no evidence of progression after first-line T+BEV did not achieve longer PFS compared to continuation of T+BEV.
Key Message: "In this phase III study, metastatic breast cancer patients who had had a response on firstline taxane and bevacizumab treatment were randomized to continue on this regimen or to receive endocrine therapy plus bevacizumab. We failed to show superior efficacy of the endocrine treatment maintenance strategy compared with the continuation of the treatment that have achieved initial tumor control."
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
and 55% in T+BEV and E+BEV arms, respectively.
4
Introduction The optimal treatment strategy for estrogen receptor (ER)-positive, human epidermal growth factor receptor 2 (HER2)-negative metastatic breast cancer (MBC) remains uncertain mainly due to lack of evidence for different sequencing strategies. For front-line treatment, several international guidelines recommend endocrine therapy, even in the presence of visceral metastases [1,2]. Unless there is immediate visceral crisis, in which case chemotherapy must be
inhibitors (AIs). The optimal treatment following stabilization/progression on endocrine therapy is unknown. Chemotherapy remains the treatment of choice for rapidly progressive disease with severe organ dysfunction or after progression on endocrine therapies [1,2]. In this setting, combining bevacizumab with taxane chemotherapy significantly improves progression-free survival (PFS) but does not significantly prolong overall survival (OS) [3,4]. Improved PFS was seen across taxane trials but the absolute improvement varied with the chemotherapy regimen [3,5,6]. Taxane was given until disease progression or unacceptable toxicity. However, long term continuation of the taxane was challenging given cumulative adverse events such as hematological toxicities, sensory neuropathies and nail disorders. In the E2100 trial, 21% of patients stopped paclitaxel prematurely due to toxicity (23% of patients had grade 3/4 neuropathy and 9% grade 3/4 fatigue) [3]. In the AVADO trial, the median duration of docetaxel treatment was only 5.5 months [6]. In routine clinical practice, patients or physicians tend to discontinue taxane before evidence of disease progression or the emergence of toxicity; and for ER-positive tumors the addition of endocrine therapy is often suggested. In a few retrospective studies involving small numbers of patients, there is some evidence that endocrine maintenance therapy after first line chemotherapy may prolong survival [7-9].
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
considered, most patients receive endocrine therapy such as third generation aromatase
5
Targeting Vascular Endothelial Growth Factor (VEGF) may improve the efficacy of endocrine therapy. In preclinical models [10] as well as in patient data [11,12], levels of hormones and VEGF are correlated. Estrogen is known to cause sustained low-level VEGF expression by cancer cells and VEGF level may be modified by endocrine therapy. Phase II studies showed that the combination of endocrine therapy plus bevacizumab was safe and active in women with locally advanced or MBC [13-15]. Consequently, starting or resuming endocrine therapy for patients on bevacizumab after chemotherapy-induced antitumor response
et al reported that combined endocrine therapy and bevacizumab maintenance after taxanebased chemotherapy produced a longer median PFS than bevacizumab alone [16]. Given this background, the AROBASE trial was designed to prospectively compare the efficacy of combining endocrine therapy with bevacizumab against continuation of the initial regimen in patients who had had a response or disease stabilization with first-line taxane and bevacizumab. We hypothesized that even for patients whose disease recurred or progressed during or after nonsteroidal AIs, the addition of exemestane (a steroidal AI) to bevacizumab may have a benefit, because (i) exemestane may be effective for some patients in this situation [17], and (ii) the combination of bevacizumab and endocrine therapy may reverse hormone resistance [10].
Methods Patients Eligible patients had histologically or cytologically confirmed locally advanced nonresectable breast cancer and/or MBC. Tumors had to be estrogen receptor-positive (progesterone positive or negative) and HER2-negative. At least one evaluable target lesion by RECIST criteria was required for tumor response. Previous neoadjuvant or adjuvant anthracycline- and taxane-based chemotherapy was permitted. Prior endocrine therapy with
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
may be an effective strategy to improve PFS. In an observational study of twenty patients, Fabi
6
non-steroidal AIs (but not exemestane) was allowed in the adjuvant or metastatic setting. Patients had to have non-progressive disease (ie complete or partial response or stable disease) 16-24 weeks after starting first-line taxane plus bevacizumab. Additional eligibility requirements included an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1; life expectancy ≥3 months; no grade 3/4 peripheral neuropathy; no proteinuria (urine dipstick for proteinuria greater than 2+, unless 24-h urine collection showed ≤1 g protein); and adequate bone marrow, kidney, and liver function. All patients were required to provide written informed
Patients were ineligible if they had received any chemotherapy other than the taxane and bevacizumab combination in the first line setting. Additional exclusion criteria included: presence of symptomatic central nervous system metastases; history of another active malignancy within 5 years; symptomatic cardiovascular disease; uncontrolled hypertension; history of thrombotic events within 6 months; history or evidence of coagulopathy with a risk of bleeding; major surgical procedure in the 28 days before randomization; history of abdominal fistula, gastrointestinal perforation or intra-abdominal abscess within 6 months; pregnancy or lactation.
Treatment and study procedures This was a multicenter, randomized, open-label phase III trial. Enrollment was initiated at 40 French sites. Patients who had been treated with a taxane (either paclitaxel 80-90 mg/m² for 3 weeks of a 4-week cycle, or docetaxel 75 mg/m² every 3 weeks) plus bevacizumab were randomized 1:1 to continue on the same regimen with the same taxane plus bevacizumab (15 mg/kg every 3 weeks or 10 mg/kg every 2 weeks) (T+BEV arm), or to receive exemestane (25 mg daily, PO) in combination with bevacizumab (15 mg/kg every 3 weeks) (E+BEV arm). In this treatment group, premenopausal patients had to be treated with a GnRH analog. Treatment was continued until disease progression, unacceptable toxicity, or withdrawal of consent. Dose reductions or discontinuations of taxane were allowed according to investigators’ usual practice.
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
consent before entering the study.
7
No dose reduction was recommended for bevacizumab or exemestane (though temporary discontinuation was allowed if needed). If taxane, bevacizumab or exemestane was discontinued before progression, the other drug had to be continued as monotherapy. Randomization was stratified by center, menopausal status, prior exposure to nonsteroidal AIs, previous history of hormone resistance (i.e. disease progression during adjuvant endocrine therapy or within 6 months of first-line of endocrine therapy in the metastatic setting), and study site.
exemestane and bevacizumab versus continuation of the initial regimen as measured by investigator-assessed PFS from randomization. Secondary objectives were to compare the two arms in terms of OS, PFS from the beginning of bevacizuamb treatment, safety, and quality of life (QoL). The study was conducted in accordance with the principles of Good Clinical Practice, as defined by the International Conference on Harmonization, and relevant European and French regulations. The study protocol was approved by a central ethics committee (December 6, 2011) and at each individual institution.
Assessments The primary endpoint of the study, PFS, was defined as the time from randomization to the first investigator-assessed progression, or death irrespective of the underlying cause. PFSBEV was defined as the time from the start of bevacizumab in the first-line metastatic setting to the first investigator-assessed progression or death from any cause. PFS and PFS-BEV were censored at the date of the last contact when patients were known to be alive and progressionfree. OS was defined as the time between randomization and death and censored at the last date the patient was known to be alive. Assessments were performed every 6 weeks until disease progression. Computerized tomography was the preferred method for evaluating the
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
The primary objective of the study was to compare the efficacy of maintenance with
8
target lesion. Safety information was obtained from all study patients. AEs were assessed according to the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) version 4.0. Patient-reported QoL was assessed with the European Organization for Research and Treatment of Cancer QoL Questionnaire-C30, completed at randomization, every 6 weeks, and at disease progression. Patients were followed up every 3 months for 18 months after the last cycle of study treatment.
Sample size and interim analysis: The study was designed to detect an improvement in the 6-month PFS rate (PFS6m) from 50% with T+BEV to 65% with E+BEV (translating into a treatment effect hazard ratio (HR) of 0.62). A total of 141 PFS events were required to have 90% power to show a statistically significant difference (2-sided alpha, using a log-rank test). Anticipating recruitment duration of 24 months and a 6-month follow-up for the last patient included, the estimated total duration of the study was 30 months, and 186 evaluable patients were planned for inclusion. An interim analysis was planned after 40% of required PFS events. The Lan and DeMets approach that approximates the O’Brien and Fleming spending function was used to control the type I error. Boundary significance levels of 0.001 and 0.049 were used at the interim and final analyses, respectively, to test the superiority of PFS in the experimental arm. The probability of reaching statistical significance at the end of the study given the results at interim analysis (conditional power) was also calculated. The boundaries for p value defined by the interim analysis and the conditional power were to be used by the Independent Data and Monitoring Committee (IDMC) along with other relevant trial information including safety data when deciding if it was appropriate to continue the trial as planned.
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Statistics
9
Statistical analysis: All efficacy analyses were performed on the ITT population, ie they included all randomized patients analyzed according to the randomization scheme. Survival data were estimated using the Kaplan-Meier method, and the log-rank test was used to compare the survival curves of the 2 study arms. A Cox proportional hazards model was used to estimate the HR with 95% CIs. Safety analyses were performed on all randomized patients who received at least one dose of study drug. Descriptive statistics were used to characterize QoL data in the
Results Interim analysis The interim analysis was based on data from the first 98 randomized patients. Median follow up was 9.7 months (range 0.8-28.3). The characteristics of these patients are shown in Figure 1. The PFS6m was 70.1% (95% confidence interval (CI) 54, 81.5) in the T+BEV arm and 54% (95% CI 38.5, 67.2) in patients treated with E+BEV (HR: 1.16, 95% CI 0.71, 1.89, log-rank p=0.56). Given these interim results, the probability of showing a statistically significant PFS improvement with the experimental maintenance strategy at the end of the study was only 7%. Based on the overall efficacy results available, the IDMC recommended that enrollment was stopped at this stage. However, given the better toxicity profile seen in the experimental arm, the IDMC also recommended the continuation of treatment and follow-up of all randomized patients as planned in the protocol, in order to obtain accurate survival data with additional patients randomized between the data cut-off for the interim analysis and the decision to stop the trial.
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
two treatment arms. Statistical analyses were performed using the SAS software (version 9.3).
10
Patient characteristics The final analysis (April 2014) was of complete follow-up data for 117 patients randomized between June 2010 and July 2013. Of the 117 patients, 59 were randomized to continue on their first line regimen with taxane + bevacizumab (T+BEV) and 58 were randomized to receive the combination of exemestane + bevacizumab (E+BEV). Patient demographics and baseline characteristics were well balanced between the two treatment arms (Table 1). Median age at inclusion was 56 years (range 35-86 years), and 92
had to be treated with a GnRH analog). 58 patients (51.8%) had an ECOG performance status of 0 at baseline, and 42 patients 35.9% presented with three or more metastatic sites. 73 patients (62.9%) had received chemotherapy in the neoadjuvant or adjuvant setting, and 56 patients (47.9%) had already received a nonsteroidal AI in either the adjuvant or metastatic settings (Table 1). In the T+BEV group, 39 patients (66.1%) had received adjuvant endocrine therapy, as had 33 patients (57.9%) in the E+BEV group. Median time from diagnosis to metastasis was 4.3 years (range: 0-29) with 29 patients (24.8%) presenting with de novo metastatic breast cancer.
Treatment exposure The median duration of exposure to the combination regimen after randomization was 2.3 months (range 0-121.4) in the T+BEV group and 5.1 months (range 0.03-41.8) in the E+BEV group. Patients in the T+BEV arm received a median of 6.3 months (0.5-34.4) of bevacizumab treatment in total, compared with 5.3 months (0.03-41.8) in the E+BEV arm (Table 2). In the T+BEV group, 1 patient received docetaxel and the other 58 weekly paclitaxel. Taxanes were prematurely stopped for the majority of patients (n=56, 94.9%), including 8 patients who stopped their taxane shortly after randomization (continuing on bevacizumab alone). The main reason for discontinuation was toxicity (n=29, 49.2%). In the E+BEV arm,
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
patients (78.6%) were considered postmenopausal at study inclusion (premenopausal patients
11
median duration of exposure to exemestane was 6 months (0.2-41.8); the main reason for discontinuation was progression (n=41, 70.7%). Reasons for definite discontinuation of each drug are detailed in Figure 1.
Efficacy Median follow-up at the time of the analysis was 21.4 months (range: 0.8-42.9) and 20.6 months (1.2-43.1) in the T+BEV and E+BEV arms, respectively. Based on an intent-to-treat
and E+BEV, respectively (HR: 1.0, 95% CI 0.7, 1.5, p=0.998). Median PFS-BEV (from the beginning of bevacizumab treatment) was similar between the 2 arms with a median of 12.5 months (95 %CI 10.2, 14.8) and 12.3 months (95 %CI 9.2, 14.7) for the T+BEV and the E+BEV treatment groups, respectively.
Exploratory subgroup analyses In order to identify subpopulations that may benefit from the experimental E+BEV treatment, PFS data on patients supposed to have the most hormone-sensitive disease (ie those who were AI-treatment naïve and had no history of hormone resistance) were analyzed. PFS data on patients who had not received any adjuvant chemotherapy were also analyzed. However, the number of patients in these subgroups was small and no statistically significant differences were established (Figure 3).
Safety All randomized patients were analyzed for safety. Grade 3-4 adverse events occurred in 63.2% of 117 patients included: 35 patients (59.3%) in the T+BEV group and 39 patients (67.2%) in the E+BEV group. No grade 5 adverse event was reported. There was an increased rate of grade 3/4 fatigue(13.6% vs 5.2%), peripheral sensory neuropathy (13.6% vs 0%), anemia
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
analysis, PFS6m was 67.2% (95% CI 53.6, 77.7) and 55.2% (95% CI 41.5, 66.9) with T+BEV
12
(5.1% vs 0%), and hemorrhage (n=1 vs n=0) in the T+BEV arm as compared to the E+BEV arm (Table 4). In the E+BEV arm, there was marginally more grade 3/4 arthralgia (n=2 vs n=0) and thromboembolic events (n=1 vs n=0). Hypertension was the most frequently reported AE (n=21, 35.6% and n=25, 43.1% grade 3/4 hypertension in the T+BEV and E+BEV arms, respectively). In the E+BEV group, only 10 (17.2%) treatment-related serious AEs were reported, compared to 18 (30.5%) in the T+BEV arm. No gastrointestinal perforation was reported.
The analysis of the QoL data (based on the EORTC QLQ-C30 questionnaire) did not show any difference between treatment arms. The median global scores remained stable through the study: 62.5 and 66.7 at baseline (n=104) for the T+BEV and the E+BEV groups, respectively; 66.7 and 66.7 at 6 weeks (n=64); and 66.7 and 66.7 at 12 weeks (n=63).
Updated OS results Updated data after 35-month median follow-up showed that OS from randomization was not different between the two arms (Figure 4). Death rates were 44% and 55% in the T+BEV and E+BEV arms, respectively. All deaths were disease-related.
Discussion The aim of this randomized phase III study was to assess the efficacy of exemestane plus bevacizumab as a maintenance therapy after first line taxane and bevacizumab combination. Our hypothesis was that E+BEV maintenance treatment would increase PFS when compared with continuation of taxane plus bevacizumab until progression or toxicity. PFS6m was 67.2% (95% CI 53.6, 77.7) and 55.2% (95% CI 41.5, 66.9) with T+BEV and E+BEV, respectively (HR: 1.0, 95% CI 0.7, 1.5, p=0.998), meaning that we failed to show superior efficacy with this treatment strategy. However, in the control arm, median exposure to taxane
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Quality of life
13
after randomization was only 2.3 months. Looking at subgroup analyses, patients without previous history of hormone resistance may have benefited from the E+BEV maintenance strategy, especially those who were AI treatment-naïve (HR: 0.85, 95% CI 0.47, 1.54). This trend was reversed for patients with a history of progression on previous AI treatment (HR: 1.25, 95% CI 0.70, 2.24) (Figure 3). These suggestions are based on post-hoc exploratory subgroup analyses with small numbers and low statistical power. However, they may deserve further investigation.
negative breast cancer is widely used and is considered a reasonable treatment option in the Advanced Breast Cancer guidelines [1,18]. However, published data supporting this strategy are limited [7-9,19]. In a retrospective analysis of French patient cohorts, maintenance endocrine treatment after first line chemotherapy was associated with longer PFS (HR: 0.64, 95 %CI 0.57, 0.72) and OS (HR: 0.81, 95 %CI 0.71, 0.92) [9]. In multivariate analysis, the maintenance strategy improved PFS for patients who had received high-dose chemotherapy with hematopoietic progenitor cell transplant, and especially for those patients who were endocrine therapy-naïve [20]. Nevertheless, to our knowledge, only one prospective randomized study has shown the superiority of adding endocrine treatment (medroxyprogesterone acetate) to standard treatment. This study was conducted in patients who had received six cycles of anthracycline- and ifosfamide-containing chemotherapy. PFS was 4.9 months in the endocrine maintenance arm versus 3.7 months without maintenance therapy [21]. On the other hand, consistent with our results, one recent prospective randomized study did not show any difference when endocrine therapy after 4-8 cycles of docetaxel/capecitabine combination was compared to capecitabine monotherapy (10.9 months versus 11.1 months) [22]. Part of our rationale was that the E+BEV combination might reverse endocrine treatment resistance and confer efficacy even in patients having previously received nonsteroidal AIs. This
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Maintenance endocrine therapy after chemotherapy for metastatic ER-positive HER2-
14
hypothesis was not confirmed, and patients whose disease recurred or progressed during nonsteroidal AIs did not derive benefit from the addition of exemestane to continued bevacizumab. Other randomized clinical trials testing the addition of bevacizumab to endocrine therapy have shown conflicting results. The LEA trial failed to demonstrate a statistically significant increase in PFS for endocrine therapy plus bevacizumab compared to endocrine therapy alone in front line treatment [23]. On the other hand, the CALGB 40503 study showed an improvement
of bevacizumab to letrozole as first line endocrine treatment. However, in this study, 43-48% of patients were endocrine therapy-naïve and almost 80% of patients were AI treatment naïve [24]. This is in sharp contrast to the AROBASE patient population described here. Our results have to be interpreted in the light of the IMELDA trial. The design was similar to our AROBASE trial, selecting patients with at least stable disease after 9-18 weeks of docetaxel plus bevacizumab combination. However, it favored the concept of switched maintenance chemotherapy (maintenance with another type of chemotherapy, i.e. capecitabine) combined with bevacizumab maintenance. The IMELDA trial was positive, with a hazard ratio for PFS of 0.38 (95 CI 0.27, 0.55) [25]. Furthermore, OS, a secondary end-point, was significantly improved from 23 to 39 months with capecitabine/bevacizumab maintenance treatment. This is the first clinical trial to have shown significant OS benefit in this setting. One has to ask why the two “maintenance” studies had such different results. Part of the explanation may be that the benefit of maintenance treatment is greatest when the treatment modality remains the one that achieved initial tumor control: i.e., selecting patients whose disease responded to chemotherapy selects patients with chemo-sensitive tumors. If this is the case, continuing chemotherapy may be superior to modifying the treatment strategy, as explored in our AROBASE trial. This hypothesis is in line with data that have been published regarding combined endocrine therapies
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
in PFS from 16 months to 20 months (HR: 0.75, 95% CI 0.59, 0.96, p=0.016) with the addition
15
for patient with “highly hormone sensitive” disease, as well as dual HER2 targeted therapies for the front line treatment for HER2-positive MBC [26].
In conclusion, our results do not support the (re)introduction of endocrine treatment for maintenance therapy in combination with a bevacizumab-based regimen, even though our safety data confirmed the favorable toxicity profile of the E+BEV combination. There was no benefit over standard taxane and bevacizumab maintenance therapy in terms of PFS or OS.
treatment naïve, though statistical significance was not reached in this limited number of patients. Future studies of maintenance therapy should include molecular selection of patients for specific targeted therapies.
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Subgroup analysis showed a trend for better tumor control among patients who were AI-
16
Disclosure OT received honoraria from Roche, Novartis, Pfizer, and Astra-Zeneca and advisory role for Roche. VD received honoraria from Roche, Novartis, and Pfizer and advisory role for Roche, Novartis, and Pfizer. TP received honoraria from Roche. TB received honoraria from Novartis, advisory role for Roche, Novartis, and Astra-Zeneca and research funding from Roche, Novartis,
All remaining authors have declared no conflicts of interest.
Acknowledgements This work was supported by Roche France and the “Ligue Nationale Contre le Cancer” (unrestricted grant). Rob Stepney, PhD (medical writer, Charlbury UK) assisted in the editing of the manuscript. The authors thank the patients who participated in the trial. We acknowledge Nicolas Gane, Marion Gauthier and Bénédicte Votan from the GINECO study office. We also thank the following investigators and pharmacists who participated in the trial: Investigators: J.Y. PIERGA (Institut Curie - Hôpital Claudius Régaud, PARIS); A. LORTHOLARY (Centre Catherine de Sienne, NANTES); J.M. FERRERO (Centre Antoine Lacassagne, NICE); A.C. HARDY-BESSARD (Clinique Armoricaine de Radiologie, SAINT-BRIEUC); I. RAY-COQUARD (Centre Léon Bérard, LYON); E. CAROLA (Centre Hospitalier de Senlis, SENLIS); H. JOOSTEN (Centre Hospitalier de Valence, VALENCE); J. MEUNIER (Centre Hospitalier Régional d'Orléans, ORLEANS); B. HOCH (Centre Azuréen de Cancérologie, MOUGINS); C. CLIPPE (Hôpitaux Drôme Nord, ROMANS-SUR-ISERE); A.M. SAVOYE (Institut Jean Godinot, REIMS); D. SPAETH (ORACLE - Centre d'Oncologie de Gentilly, NANCY); E. ANGELLIER (Clinique
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
and Astra-Zeneca.
17
Notre-Dame de Bon Secours, CHARTRES); C.B. LEVACHE (Clinique Francheville, PERIGUEUX); E. ACHILLE (Institut d'Oncologie, STRASBOURG); E. LEGOUFFE (Polyclinique KENVAL, NIMES); E. BRAIN (Hôpital René Huguenin, SAINT-CLOUD); A. MARTI (Centre Hospitalier d'Auxerre, AUXERRE); G. DEPLANQUE (Groupe Hospitalier Saint-Joseph, PARIS); O. ARSENE (Centre Hospitalier de Blois, BLOIS); P.E. HEUDEL (Centre Hospitalier de la Région d'Annecy, PRINGY); S. HENAULT (Centre Hospitalier Intercommunal de Créteil, CRETEIL); T. L'HARIDON, F. PRIOU (Centre Hospitalier Départemental Les Oudairies, LA
Pharmacists: C. ENGEL (Centre Léon Bérard, LYON); E. BESREST (Centre Antoine Lacassagne, NICE); A. GRASS (Hôpital Privé Jean Mermoz, LYON); D. BAYLOT (Clinique de la Sauvegarde, LYON); F. BADIBOUIDI (Centre Hospitalier de Senlis, SENLIS); B. GOSSELIN (Hôpital de Mont-deMarsan, MONT- DE-MARSAN); DE VINZELLES, RENON-CARON (Centre Hospitalier Universitaire Dupuytren, LIMOGES); L. ESCALUP (Institut Curie - Hopital Claudius Régaud, PARIS); F. MOREY (Hôpital Fleyriat, BOURG-EN-BRESSE); A. MARCHAUD (Hôpitaux Drôme Nord, ROMANS-SUR-ISERE); C. LABAT (Centre Hospitalier La Dracénie, DRAGUIGNAN); R. ALFONSI (ORACLE - Centre d'Oncologie de Gentilly, NANCY); F. POIROT-LUTRIN (Centre Hospitalier de la Région d'Annecy, PRINGY); J.B. REY (Institut Jean Godinot, REIMS); P. BERNARD (Clinique de l'Union, SAINT-JEAN); E. VERGNES (Centre Paul Strauss, STRASBOURG); R. BESSARD (Centre Hospitalier Bretagne Atlantique, VANNES); J. DIMET (Centre Hospitalier Départemental Les Oudairies, LA ROCHE-SUR-YON); L. GUIVARCH (Centre Catherine de Sienne, NANTES); P. PLOCCO, V. PRIOU, S. TOLLEC (Centre Hospitalier Régional d'Orléans, ORLEANS); S. ROUSSEL (Clinique Armoricaine de Radiologie, SAINT-BRIEUC); I. DUFRENE (Centre Hospitalier de Valence, VALENCE); R. BESSARD (Centre d'Oncologie Saint-Yves, VANNES); N. VERMERIE (Centre Hospitalier d'Auxerre, AUXERRE); P. BRETON (Centre Hospitalier de Blois, BLOIS); E. OLIVIER (Clinique Notre-
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
ROCHE-SUR-YON)
18
Dame de Bon Secours, CHARTRES); S. POULAIN, A. THEBAULT (Centre Hospitalier Intercommunal de Créteil, CRETEIL); S. ROUBAUD (Centre Azuréen de Cancérologie, MOUGINS); I. CHABROLLES (Polyclinique KENVAL, NIMES); M. JARDIN (Groupe Hospitalier Saint-Joseph, PARIS); B. SOUSSELIER (Clinique Francheville, PERIGUEUX); I. FERRY (Hôpital René Huguenin, SAINT-CLOUD); L. MATHIS, F. LAURENT (Institut d'Oncologie, STRASBOURG)
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
19
References 1. Cardoso F, Costa A, Norton L, et al. ESO-ESMO 2nd international consensus guidelines
for advanced breast cancer (ABC2). Ann Oncol 2014; 25: 1871-1888. 2. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in
Oncology (NCCN Guidelines). Breast Cancer 20th annual edition. http://www.nccn.org/professionals/physician_gls/f_guidelines.asp#site (accessed Dec 15, 2015).
for metastatic breast cancer. N Engl J Med 2007; 357: 2666-2676. 4. Miles DW, Diéras V, Cortés J, et al. First-line bevacizumab in combination with
chemotherapy for HER2-negative metastatic breast cancer: pooled and subgroup analyses of data from 2447 patients. Ann Oncol 2013; 24: 2773-2780. 5. Robert NJ, Diéras V, Glaspy J, et al. RIBBON-1: randomized, double-blind, placebo-
controlled, phase III trial of chemotherapy with or without bevacizumab for first-line treatment of human epidermal growth factor receptor 2-negative, locally recurrent or metastatic breast cancer. J Clin Oncol 2011; 29: 1252-1260. 6. Miles DW, Chan A, Dirix LY, et al. Phase III study of bevacizumab plus docetaxel
compared with placebo plus docetaxel for the first-line treatment of human epidermal growth factor receptor 2-negative metastatic breast cancer. J Clin Oncol 2010; 28: 32393247. 7. Berruti A, Zola P, Buniva T, et al. Prognostic factors in metastatic breast cancer patients
obtaining objective response or disease stabilization after first-line chemotherapy with epirubicin. Evidence for a positive effect of maintenance hormonal therapy on overall survival. Anticancer Res 1997; 17: 2763-2768. 8. Bertelli G, Garrone O, Bertolotti L, et al. Maintenance hormone therapy with letrozole
after first-line chemotherapy for advanced breast cancer. Oncology 2005; 68: 364-370.
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
3. Miller K, Wang M, Gralow J, et al. Paclitaxel plus bevacizumab versus paclitaxel alone
20
9. Dufresne A, Pivot X, Tournigand C, et al. Maintenance hormonal treatment improves
progression free survival after a first line chemotherapy in patients with metastatic breast cancer. Int J Med Sci 2008; 5: 100-105. 10. Dadiani M, Seger D, Kreizman T, et al. Estrogen regulation of vascular endothelial
growth factor in breast cancer in vitro and in vivo: the role of estrogen receptor alpha and c-Myc. Endocr Relat Cancer 2009; 16: 819-834. 11. Adams J, Carder PJ, Downey S, et al. Vascular endothelial growth factor (VEGF) in
and effects of tamoxifen. Cancer Res 2000; 60: 2898-2905. 12. Holmes CE, Huang JC, Pace TR, et al. Tamoxifen and aromatase inhibitors differentially
affect vascular endothelial growth factor and endostatin levels in women with breast cancer. Clin Cancer Res 2008; 14: 3070-3076. 13. Traina TA, Rugo HS, Caravelli JF, et al. Feasibility trial of letrozole in combination with
bevacizumab in patients with metastatic breast cancer. J Clin Oncol 2010; 28: 628-633. 14. Forero-Torres A, Saleh MN, Galleshaw JA, et al. Pilot trial of preoperative (neoadjuvant)
letrozole in combination with bevacizumab in postmenopausal women with newly diagnosed estrogen receptor- or progesterone receptor-positive breast cancer. Clin Breast Cancer 2010; 10: 275-280. 15. Yardley DA, Burris HA 3rd, Clark BL, et al. Hormonal therapy plus bevacizumab in
postmenopausal patients who have hormone receptor-positive metastatic breast cancer: a phase II Trial of the Sarah Cannon Oncology Research Consortium. Clin Breast Cancer 2011; 11: 146-152. 16. Fabi A, Russillo M, Ferretti G, et al. Maintenance bevacizumab beyond first-line
paclitaxel plus bevacizumab in patients with Her2-negative hormone receptor-positive metastatic breast cancer: efficacy in combination with hormonal therapy. BMC Cancer 2012; 12: 482.
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
breast cancer: comparison of plasma, serum, and tissue VEGF and microvessel density
21
17. Chia S, Gradishar W, Mauriac L, et al. Double-blind, randomized placebo controlled trial
of fulvestrant compared with exemestane after prior nonsteroidal aromatase inhibitor therapy in postmenopausal women with hormone receptor-positive, advanced breast cancer: results from EFECT. J Clin Oncol 2008; 26: 1664-1670. 18. Cardoso F, Costa A, Norton L, et al. 1st International consensus guidelines for advanced
breast cancer (ABC 1). Breast 2012; 21: 242-252. 19. Redondo A, Martínez V, Zamora P, et al. Continuation of bevacizumab and addition of
cancer. Onco Targets Ther 2014; 7: 2175-2181. 20. Montemurro F, Rondón G, Ueno NT, et al. Factors affecting progression-free survival in
hormone-dependent metastatic breast cancer patients receiving high-dose chemotherapy and hematopoietic progenitor cell transplantation: role of maintenance endocrine therapy. Bone Marrow Transplant 2002; 29: 861-866. 21. Kloke O, Klaassen U, Oberhoff C, et al. Maintenance treatment with
medroxyprogesterone acetate in patients with advanced breast cancer responding to chemotherapy: results of a randomized trial. Essen Breast Cancer Study Group. Breast Cancer Res Treat 1999; 55: 51-59. 22. Liang X, Yan Y, Wang L, et al. First-line chemotherapy with docetaxel plus capecitabine
followed by capecitabine or hormone maintenance therapy for the treatment of metastatic breast cancer patients. Oncol Lett 2015; 9: 987-993. 23. Martín M, Loibl S, von Minckwitz G, et al. Phase III Trial Evaluating the Addition of
Bevacizumab to Endocrine Therapy As First-Line Treatment for Advanced Breast Cancer: The Letrozole/Fulvestrant and Avastin (LEA) Study. J Clin Oncol 2015; 33: 1045-1052. 24. Dickler MN, Barry WT, Cirrincione CT, et al. Phase III trial evaluating the addition of
bevacizumab to letrozole as first-line endocrine therapy for treatment of hormone-
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
hormone therapy following weekly paclitaxel therapy in HER2-negative metastatic breast
22
receptor positive advanced breast cancer: CALGB 40503 (Alliance). J Clin Oncol 2015; 33 (suppl; abstr 501). 25. Gligorov J, Doval D, Bines J, et al. Maintenance capecitabine and bevacizumab versus
bevacizumab alone after initial first-line bevacizumab and docetaxel for patients with HER2-negative metastatic breast cancer (IMELDA): a randomised, open-label, phase 3 trial. Lancet Oncol 2014; 15: 1351-1360. 26. Bachelot T, Tredan O. Targeted treatments for breast cancer: a step forward. Lancet Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Oncol 2013; 14: 438-439.
23
Table 1: Patient demographics and baseline characteristics Table 2: Chemotherapy regimens prior to randomization and after randomization Table 3: Summary of efficacy results (all randomized patients) Table 4: AEs occurring in ≥10 % of patients in either treatment group
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
24
Figure Caption Figure 1: Distribution of patients
117 patients provided consent
117 patients evaluable for efficacy
59 received taxane and bevacizumab as continuation of first line chemotherapy
58 received exemestane and bevacizumab as maintenance therapy after chemotherapy
56 discontinued taxane
53 discontinued bevacizumab
47 discontinued exemestane
48 discontinued bevacizumab
12 disease progressions 21 adverse events or toxic effects after randomisation 8 adverse events or toxic effects before randomisation 13 patient or investigator decisions 2 other
41 disease progressions 7 adverse events or toxic effects 4 patient or investigator decisions 1 other
41 disease progressions 5 adverse events or toxic effects 1 other
36 disease progressions 9 adverse events or toxic effects 1 patient or investigator decisions 2 other
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
117 patients evaluable for safety
25
Figure 2: Kaplan-Meier plot for PFS (randomized subjects) Abbreviations: CI, confidence interval; PFS, progression-free survival; BEV = bevacizumab; E = Exemestane; T = taxanes.
Figure 3: Forest plot for the PFS effect of the treatment in (pre)defined subgroups
= Exemestane; TAX = taxanes.
Figure 4: Updated OS with 35-month follow-up Abbreviations: CI, confidence interval; OS, overall survival; BEV = bevacizumab; E = Exemestane; T = taxanes.
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Abbreviations: CI, confidence interval; PFS, progression-free survival; BEV = bevacizumab; EXE
26
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
27
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
28
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Table 1: Patient demographics and baseline characteristics Number (%) of patients Patients included in the interim
Patients included in the final analysis
analysis (n=98) Taxane
Exemestane
+ BEV
+ BEV
(n=59)
(n=58)
55 [35-86]
55 [35-86]
56 [35-77]
0
46 (48.9)
29 (51.8)
29 (51.8)
1
46 (48.9)
26 (46.4)
26 (46.4)
2
2 (2.1)
1 (1.8)
1 (1.8)
4
3
2
4.4 (0-29)
4.3 (0-29)
4.6 (0-34)
≤6 months
13 (22.0)
16 (27.6)
>6 months
46 (78.0)
42 (72.4)
11 (18.6)
10 (17.2)
Age [Years] median (range) ECOG PS
NR Median time (range) from diagnosis to metastatic disease [years]
Metastatic sites bone mets only bone
61 (62.2)
39 (66.1)
36 (62.1)
lung
36 (36.7)
13 (22.0)
28 (48.3)
liver
44 (44.9)
31 (52.5)
25 (43.1)
≥3
37 (37.8)
20 (33.9)
22 (37.9)
ER+ / PR +
58 (59.8)
36 (61.0)
34 (59.6)
ER+ / PR -
39 (40.2)
23 (39.0)
23 (40.4)
(neo)adjuvant chemo
64 (65.3)
38 (64.4)
35 (61.4)
adjuvant endocrine
63 (64.3)
39 (66.1)
33 (57.9)
metastatic endocrine.
23 (23.5)
18 (30.5)
12 (20.7)
8 (8.2)
12 (20.3)
9 (15.5)
post-menopausal at inclusion
76 (77.6)
45 (76.3)
47 (81.0)
prior AIs
46 (46.9)
29 (49.2)
27 (46.6)
endocrine resistance
42 (42.9)
26 (44.1)
23 (39.7)
ER and PR status
Setting of prior therapies
metastatic AIs .
Stratification factors
Abbreviations: BEV = bevacizumab; NR = not reported; ECOG PS = Eastern Cooperative Group Performance Status; mets = metastases; ER = estrogen receptor; PR = progesterone receptor; AIs = aromatase inhibitors
Table 2: Chemotherapy regimens prior to randomization and after randomization Number (%) of patients
Treatment before randomization (months) paclitaxel + BEV docetaxel + BEV median duration [range]
Treatment after randomization (months) BEV median duration [range] taxane median duration [range] exemestane median duration [range] Abbreviations: BEV = bevacizumab
Taxane + BEV
Exemestane + BEV
(n=59)
(n=58)
58 (98.3)
57 (98.3)
1 (1.7)
1 (1.7)
3.7 [2.3-5.7]
3.4 [2.07-6]
6.3 [0.5-34.4]
5.3 [0.03-41.8]
2.3 [0.0-121.4]
-
-
6.0 [0.2-41.8]
Table 3: Summary of efficacy results (all randomized patients)
Median follow up [range]
Taxane + BEV
Exemestane + BEV
(n=59)
(n=58)
21.4 months [0.8-42.9]
20.6 months [1.2-43.1]
48 (81)
44 (76)
8.1 (6.5-10.7)
7.6 (5.4-10.9)
67.2% (53.6-77.7)
55.2% (41.5-66.9)
Primary endpoint PFS: number of events (%) median, months (95%CI) 6-month PFS rate (95%CI)
Log-rank test p=0.998 HR: 1.00 (0.66-1.51)
Secondary endpoints OS after 35-month median follow-up: number of events (%)
26 (44)
32 (55)
PFS from the beginning of bevacizumab: number of events (%)
47 (81)
44 (76)
12.5 (10.2-14.8)
12.3 (9.2-14.7)
median, months (95%CI)
Abbreviations: BEV = bevacizumab; CI = confidence interval; PFS = progression-free survival; TTP = Time To Progression from the beginning of the taxane plus bevacizuamb treatment
Table 4: AEs occurring in ≥10 % of patients in either treatment group Number (%) of patients Taxane + BEV
Exemestane + BEV
(n=59)
(n=58)
any grade / grade 3-4
any grade / grade 3-4
47 (79.7%) / 8 (13.6%)
30 (51.7%) / 3 (5.2%)
Hypertension
54 (91.5%) / 21 (35.6%)
53 (91.4%) / 25 (43.1%)
Hemorrhage
23 (39.0%) / 1 (1.7%)
14 (24.1%) / 0
3 (5.1%) / 0
3 (5.2%) / 1 (1.7%)
Vomiting
3 (5.1%) / 0
11 (19.0%) / 0
Nausea
18 (30.5%) / 0
14 (24.1%) / 0
12 (20.3%) / 1 (1.7%)
6 (10.3%) / 0
10 (16.9%) / 0
10 (17.2%) / 0
39 (66.1%) / 8 (13.6%)
19 (32.8%) / 0
9 (15.3%) / 0
5 (8.6%) / 0
Alopecia
35 (59.3%) / 0
17 (29.3%) / 0
Mucositis
9 (15.3%) / 0
4 (6.9%) / 0
Nail disorders
26 (44.1%) / 0
9 (15.5%) / 0
16 (27.1%) / 1 (1.7%)
10 (17.2%) / 0
Myalgia
16 (27.1%) / 0
7 (12.1%) /0
Arthralgia
18 (30.5%) / 0
25 (43.1%) / 2 (3.4%)
General disorders Fatigue Cardiovascular disorders
Thromboembolic events Gastrointestinal disorders
Constipation Diarrhea Nervous system disorders Peripheral sensory neuropathy Peripheral motor neuropathy Skin/mucosa disorders
Edema Musculoskeletal disorders
Blood disorders Neutropenia
35 (59.3%) / 7 (11.9%)
15 (25.9%) / 0
1 (1.7%) / 0
0
Anemia
34 (57.6%) / 3 (5.1%)
22 (37.9%) / 0
Alanine transaminase elevated
30 (50.8%) / 2 (3.4%)
25 (43.1%) / 1 (1.7%)
Aspartate transaminase elevated
38 (64.4%) / 2 (3.4%)
35 (60.3%) / 3 (5.2%)
Febrile neutropenia
Abbreviation: AE, adverse event.
A phase III trial of exemestane plus bevacizumab maintenance therapy in patients with metastatic breast cancer after first-line taxane and bevacizumab: a GINECO group study
O. Trédan1,2, P. Follana3, I. Moullet4, C. Cropet5, S. Trager-Maury6, J. Dauba7, S. LavauDenes8, V. Diéras9, D. Béal-Ardisson10, M. Gouttebel11, H. Orfeuvre12, L. Stefani13, C. Jouannaud14, F. Bürki15, T. Petit16, E. Guardiola17, C. Becuwe18, E. Blot19, E. PujadeLauraine20, T. Bachelot1,2
Léon Bérard, 28 rue Laennec, 69008 Lyon, France
2CNRS
UMR5286, Cancer Research Center of Lyon, 69008 Lyon, France
3Centre
Antoine Lacassagne, 33 Avenue de Valombrose, 06189 Nice, France
4Clinique
de la Sauvegarde, Avenue David Ben Gourion, BP 309, 69337 Lyon, France
5Direction
of Clinical Research and Innovation, Centre Léon Bérard, 28 rue Laennec, 69008
Lyon, France 6Centre
Hospitalier de Senlis, Avenue Paul Rouge, 60309 Senlis, France
7Hôpital
Mont-De-Marsan, Avenue Pierre de Coubertin, 40024 Mont-De-Marsan, France
8CHU
Dupuytren, 2 Avenue Martin Luther King, 87042 Limoges, France
9Institut
Curie, 26 Rue d'Ulm, 75005 Paris, France
10Hôpital
Privé Jean Mermoz, 55 Avenue Jean Mermoz, 69373 Lyon, France
11Hôpitaux
Drôme Nord - Site de Romans, Avenue Geneviève de Gaulle-Anthonioz, 26102
Romans-sur-Isère, France 12Hôpital
Fleyriat, 900 Route de Paris, 01012 Bourg-en Bresse, France
13Centre
Hospitalier de la Région d'Annecy, 1 Avenue de l'Hôpital - BP 90074 - Metz-Tessy
74374 Pringy 14Institut
Jean Godinot, 1 Avenue du Général Koenig - BP 171, 51056 Reims
15Clinique
de l'Union, Boulevard de Ratelens, 31240 Saint-Jean, France
© The Author 2016. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
1Centre
2 16Centre
Paul Strauss, 3 Rue de la Porte de l'Hôpital - BP 30042, 67065 Strasbourg
17Centre
Hospitalier la Dracénie, Route de Montferrat - BP 249, 83307 Draguignan
18ORACLE 19Centre
- Centre d'Oncologie de Gentilly, 2 Rue Marie Marvingt, 54100 Nancy
Hospitalier Bretagne Atlantique, 20 Boulevard GI Maurice Guillaudot - BP 70555,
56017 Vannes 20Université
Paris Descartes, AP-HP, Hôpitaux Universitaires Paris Centre, Site Hôtel-Dieu, 1
Parvis Notre-Dame - Place Jean-Paul II, 75181 Paris
Bérard, Centre de Recherche en Cancérologie de Lyon, 28 rue Laennec, 69008, Lyon, France. Tel: (+33) 4 78 78 26 54, Fax: (+33) 4 78 78 27 16, Email: [email protected]
Abstract Background: Maintenance strategies beyond response or tumor stabilization with first line chemotherapy in metastatic breast cancer (MBC) have not been extensively studied. Endocrine therapy combined with continued bevacizumab may be a helpful option for estrogen receptor (ER)-positive MBC. Patients and Methods: In this prospective, open label, phase III study, patients with histologically confirmed ER-positive, HER2-negative MBC and non-progressive disease after 1624 weeks of taxane plus bevacizumab (T+BEV) were randomized to continuation of T+BEV or maintenance bevacizumab plus exemestane (E+BEV). The primary endpoint was progressionfree survival (PFS) from randomization. To have 80% power to detect an improvement in the 6month PFS (PFS6m) from 50% to 65%, 186 evaluable patients were needed for a total of 141 PFS events. An interim analysis was planned after 40% of the required events. Results: The interim analysis with 98 patients showed that the probability of reaching a statistically significant improvement in PFS by the end of the study was only 7%. This led the
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Corresponding author: Dr. Thomas Bachelot, Department of Medical Oncology, Centre Léon
3
Independent Data and Monitoring Committee to recommend termination of patient enrollment. After a median 21-months follow up of all randomized patients (117 in total), PFS6m from randomization was 67.2% (95% confidence interval (CI) 53.6, 77.7) with T+BEV and 55.2% (95% CI 41.5, 66.9) with E+BEV (HR: 1.0, 95% CI 0.7, 1.5, p=0.998). Median PFS from BEV initiation was 12.5 and 12.3 months in the T+ BEV and E + BEV arms, respectively. In the T+BEV arm, taxane was prematurely stopped for the majority of patients (94.9%), mainly due to toxicity (49.2%). Updated data after 35-months’ median follow-up showed death rates of 44%
Conclusion: In this trial, maintenance therapy with E+BEV in ER-positive, HER2-negative MBC patients with no evidence of progression after first-line T+BEV did not achieve longer PFS compared to continuation of T+BEV.
Key Message: "In this phase III study, metastatic breast cancer patients who had had a response on firstline taxane and bevacizumab treatment were randomized to continue on this regimen or to receive endocrine therapy plus bevacizumab. We failed to show superior efficacy of the endocrine treatment maintenance strategy compared with the continuation of the treatment that have achieved initial tumor control."
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
and 55% in T+BEV and E+BEV arms, respectively.
4
Introduction The optimal treatment strategy for estrogen receptor (ER)-positive, human epidermal growth factor receptor 2 (HER2)-negative metastatic breast cancer (MBC) remains uncertain mainly due to lack of evidence for different sequencing strategies. For front-line treatment, several international guidelines recommend endocrine therapy, even in the presence of visceral metastases [1,2]. Unless there is immediate visceral crisis, in which case chemotherapy must be
inhibitors (AIs). The optimal treatment following stabilization/progression on endocrine therapy is unknown. Chemotherapy remains the treatment of choice for rapidly progressive disease with severe organ dysfunction or after progression on endocrine therapies [1,2]. In this setting, combining bevacizumab with taxane chemotherapy significantly improves progression-free survival (PFS) but does not significantly prolong overall survival (OS) [3,4]. Improved PFS was seen across taxane trials but the absolute improvement varied with the chemotherapy regimen [3,5,6]. Taxane was given until disease progression or unacceptable toxicity. However, long term continuation of the taxane was challenging given cumulative adverse events such as hematological toxicities, sensory neuropathies and nail disorders. In the E2100 trial, 21% of patients stopped paclitaxel prematurely due to toxicity (23% of patients had grade 3/4 neuropathy and 9% grade 3/4 fatigue) [3]. In the AVADO trial, the median duration of docetaxel treatment was only 5.5 months [6]. In routine clinical practice, patients or physicians tend to discontinue taxane before evidence of disease progression or the emergence of toxicity; and for ER-positive tumors the addition of endocrine therapy is often suggested. In a few retrospective studies involving small numbers of patients, there is some evidence that endocrine maintenance therapy after first line chemotherapy may prolong survival [7-9].
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
considered, most patients receive endocrine therapy such as third generation aromatase
5
Targeting Vascular Endothelial Growth Factor (VEGF) may improve the efficacy of endocrine therapy. In preclinical models [10] as well as in patient data [11,12], levels of hormones and VEGF are correlated. Estrogen is known to cause sustained low-level VEGF expression by cancer cells and VEGF level may be modified by endocrine therapy. Phase II studies showed that the combination of endocrine therapy plus bevacizumab was safe and active in women with locally advanced or MBC [13-15]. Consequently, starting or resuming endocrine therapy for patients on bevacizumab after chemotherapy-induced antitumor response
et al reported that combined endocrine therapy and bevacizumab maintenance after taxanebased chemotherapy produced a longer median PFS than bevacizumab alone [16]. Given this background, the AROBASE trial was designed to prospectively compare the efficacy of combining endocrine therapy with bevacizumab against continuation of the initial regimen in patients who had had a response or disease stabilization with first-line taxane and bevacizumab. We hypothesized that even for patients whose disease recurred or progressed during or after nonsteroidal AIs, the addition of exemestane (a steroidal AI) to bevacizumab may have a benefit, because (i) exemestane may be effective for some patients in this situation [17], and (ii) the combination of bevacizumab and endocrine therapy may reverse hormone resistance [10].
Methods Patients Eligible patients had histologically or cytologically confirmed locally advanced nonresectable breast cancer and/or MBC. Tumors had to be estrogen receptor-positive (progesterone positive or negative) and HER2-negative. At least one evaluable target lesion by RECIST criteria was required for tumor response. Previous neoadjuvant or adjuvant anthracycline- and taxane-based chemotherapy was permitted. Prior endocrine therapy with
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
may be an effective strategy to improve PFS. In an observational study of twenty patients, Fabi
6
non-steroidal AIs (but not exemestane) was allowed in the adjuvant or metastatic setting. Patients had to have non-progressive disease (ie complete or partial response or stable disease) 16-24 weeks after starting first-line taxane plus bevacizumab. Additional eligibility requirements included an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1; life expectancy ≥3 months; no grade 3/4 peripheral neuropathy; no proteinuria (urine dipstick for proteinuria greater than 2+, unless 24-h urine collection showed ≤1 g protein); and adequate bone marrow, kidney, and liver function. All patients were required to provide written informed
Patients were ineligible if they had received any chemotherapy other than the taxane and bevacizumab combination in the first line setting. Additional exclusion criteria included: presence of symptomatic central nervous system metastases; history of another active malignancy within 5 years; symptomatic cardiovascular disease; uncontrolled hypertension; history of thrombotic events within 6 months; history or evidence of coagulopathy with a risk of bleeding; major surgical procedure in the 28 days before randomization; history of abdominal fistula, gastrointestinal perforation or intra-abdominal abscess within 6 months; pregnancy or lactation.
Treatment and study procedures This was a multicenter, randomized, open-label phase III trial. Enrollment was initiated at 40 French sites. Patients who had been treated with a taxane (either paclitaxel 80-90 mg/m² for 3 weeks of a 4-week cycle, or docetaxel 75 mg/m² every 3 weeks) plus bevacizumab were randomized 1:1 to continue on the same regimen with the same taxane plus bevacizumab (15 mg/kg every 3 weeks or 10 mg/kg every 2 weeks) (T+BEV arm), or to receive exemestane (25 mg daily, PO) in combination with bevacizumab (15 mg/kg every 3 weeks) (E+BEV arm). In this treatment group, premenopausal patients had to be treated with a GnRH analog. Treatment was continued until disease progression, unacceptable toxicity, or withdrawal of consent. Dose reductions or discontinuations of taxane were allowed according to investigators’ usual practice.
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
consent before entering the study.
7
No dose reduction was recommended for bevacizumab or exemestane (though temporary discontinuation was allowed if needed). If taxane, bevacizumab or exemestane was discontinued before progression, the other drug had to be continued as monotherapy. Randomization was stratified by center, menopausal status, prior exposure to nonsteroidal AIs, previous history of hormone resistance (i.e. disease progression during adjuvant endocrine therapy or within 6 months of first-line of endocrine therapy in the metastatic setting), and study site.
exemestane and bevacizumab versus continuation of the initial regimen as measured by investigator-assessed PFS from randomization. Secondary objectives were to compare the two arms in terms of OS, PFS from the beginning of bevacizuamb treatment, safety, and quality of life (QoL). The study was conducted in accordance with the principles of Good Clinical Practice, as defined by the International Conference on Harmonization, and relevant European and French regulations. The study protocol was approved by a central ethics committee (December 6, 2011) and at each individual institution.
Assessments The primary endpoint of the study, PFS, was defined as the time from randomization to the first investigator-assessed progression, or death irrespective of the underlying cause. PFSBEV was defined as the time from the start of bevacizumab in the first-line metastatic setting to the first investigator-assessed progression or death from any cause. PFS and PFS-BEV were censored at the date of the last contact when patients were known to be alive and progressionfree. OS was defined as the time between randomization and death and censored at the last date the patient was known to be alive. Assessments were performed every 6 weeks until disease progression. Computerized tomography was the preferred method for evaluating the
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
The primary objective of the study was to compare the efficacy of maintenance with
8
target lesion. Safety information was obtained from all study patients. AEs were assessed according to the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) version 4.0. Patient-reported QoL was assessed with the European Organization for Research and Treatment of Cancer QoL Questionnaire-C30, completed at randomization, every 6 weeks, and at disease progression. Patients were followed up every 3 months for 18 months after the last cycle of study treatment.
Sample size and interim analysis: The study was designed to detect an improvement in the 6-month PFS rate (PFS6m) from 50% with T+BEV to 65% with E+BEV (translating into a treatment effect hazard ratio (HR) of 0.62). A total of 141 PFS events were required to have 90% power to show a statistically significant difference (2-sided alpha, using a log-rank test). Anticipating recruitment duration of 24 months and a 6-month follow-up for the last patient included, the estimated total duration of the study was 30 months, and 186 evaluable patients were planned for inclusion. An interim analysis was planned after 40% of required PFS events. The Lan and DeMets approach that approximates the O’Brien and Fleming spending function was used to control the type I error. Boundary significance levels of 0.001 and 0.049 were used at the interim and final analyses, respectively, to test the superiority of PFS in the experimental arm. The probability of reaching statistical significance at the end of the study given the results at interim analysis (conditional power) was also calculated. The boundaries for p value defined by the interim analysis and the conditional power were to be used by the Independent Data and Monitoring Committee (IDMC) along with other relevant trial information including safety data when deciding if it was appropriate to continue the trial as planned.
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Statistics
9
Statistical analysis: All efficacy analyses were performed on the ITT population, ie they included all randomized patients analyzed according to the randomization scheme. Survival data were estimated using the Kaplan-Meier method, and the log-rank test was used to compare the survival curves of the 2 study arms. A Cox proportional hazards model was used to estimate the HR with 95% CIs. Safety analyses were performed on all randomized patients who received at least one dose of study drug. Descriptive statistics were used to characterize QoL data in the
Results Interim analysis The interim analysis was based on data from the first 98 randomized patients. Median follow up was 9.7 months (range 0.8-28.3). The characteristics of these patients are shown in Figure 1. The PFS6m was 70.1% (95% confidence interval (CI) 54, 81.5) in the T+BEV arm and 54% (95% CI 38.5, 67.2) in patients treated with E+BEV (HR: 1.16, 95% CI 0.71, 1.89, log-rank p=0.56). Given these interim results, the probability of showing a statistically significant PFS improvement with the experimental maintenance strategy at the end of the study was only 7%. Based on the overall efficacy results available, the IDMC recommended that enrollment was stopped at this stage. However, given the better toxicity profile seen in the experimental arm, the IDMC also recommended the continuation of treatment and follow-up of all randomized patients as planned in the protocol, in order to obtain accurate survival data with additional patients randomized between the data cut-off for the interim analysis and the decision to stop the trial.
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
two treatment arms. Statistical analyses were performed using the SAS software (version 9.3).
10
Patient characteristics The final analysis (April 2014) was of complete follow-up data for 117 patients randomized between June 2010 and July 2013. Of the 117 patients, 59 were randomized to continue on their first line regimen with taxane + bevacizumab (T+BEV) and 58 were randomized to receive the combination of exemestane + bevacizumab (E+BEV). Patient demographics and baseline characteristics were well balanced between the two treatment arms (Table 1). Median age at inclusion was 56 years (range 35-86 years), and 92
had to be treated with a GnRH analog). 58 patients (51.8%) had an ECOG performance status of 0 at baseline, and 42 patients 35.9% presented with three or more metastatic sites. 73 patients (62.9%) had received chemotherapy in the neoadjuvant or adjuvant setting, and 56 patients (47.9%) had already received a nonsteroidal AI in either the adjuvant or metastatic settings (Table 1). In the T+BEV group, 39 patients (66.1%) had received adjuvant endocrine therapy, as had 33 patients (57.9%) in the E+BEV group. Median time from diagnosis to metastasis was 4.3 years (range: 0-29) with 29 patients (24.8%) presenting with de novo metastatic breast cancer.
Treatment exposure The median duration of exposure to the combination regimen after randomization was 2.3 months (range 0-121.4) in the T+BEV group and 5.1 months (range 0.03-41.8) in the E+BEV group. Patients in the T+BEV arm received a median of 6.3 months (0.5-34.4) of bevacizumab treatment in total, compared with 5.3 months (0.03-41.8) in the E+BEV arm (Table 2). In the T+BEV group, 1 patient received docetaxel and the other 58 weekly paclitaxel. Taxanes were prematurely stopped for the majority of patients (n=56, 94.9%), including 8 patients who stopped their taxane shortly after randomization (continuing on bevacizumab alone). The main reason for discontinuation was toxicity (n=29, 49.2%). In the E+BEV arm,
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
patients (78.6%) were considered postmenopausal at study inclusion (premenopausal patients
11
median duration of exposure to exemestane was 6 months (0.2-41.8); the main reason for discontinuation was progression (n=41, 70.7%). Reasons for definite discontinuation of each drug are detailed in Figure 1.
Efficacy Median follow-up at the time of the analysis was 21.4 months (range: 0.8-42.9) and 20.6 months (1.2-43.1) in the T+BEV and E+BEV arms, respectively. Based on an intent-to-treat
and E+BEV, respectively (HR: 1.0, 95% CI 0.7, 1.5, p=0.998). Median PFS-BEV (from the beginning of bevacizumab treatment) was similar between the 2 arms with a median of 12.5 months (95 %CI 10.2, 14.8) and 12.3 months (95 %CI 9.2, 14.7) for the T+BEV and the E+BEV treatment groups, respectively.
Exploratory subgroup analyses In order to identify subpopulations that may benefit from the experimental E+BEV treatment, PFS data on patients supposed to have the most hormone-sensitive disease (ie those who were AI-treatment naïve and had no history of hormone resistance) were analyzed. PFS data on patients who had not received any adjuvant chemotherapy were also analyzed. However, the number of patients in these subgroups was small and no statistically significant differences were established (Figure 3).
Safety All randomized patients were analyzed for safety. Grade 3-4 adverse events occurred in 63.2% of 117 patients included: 35 patients (59.3%) in the T+BEV group and 39 patients (67.2%) in the E+BEV group. No grade 5 adverse event was reported. There was an increased rate of grade 3/4 fatigue(13.6% vs 5.2%), peripheral sensory neuropathy (13.6% vs 0%), anemia
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
analysis, PFS6m was 67.2% (95% CI 53.6, 77.7) and 55.2% (95% CI 41.5, 66.9) with T+BEV
12
(5.1% vs 0%), and hemorrhage (n=1 vs n=0) in the T+BEV arm as compared to the E+BEV arm (Table 4). In the E+BEV arm, there was marginally more grade 3/4 arthralgia (n=2 vs n=0) and thromboembolic events (n=1 vs n=0). Hypertension was the most frequently reported AE (n=21, 35.6% and n=25, 43.1% grade 3/4 hypertension in the T+BEV and E+BEV arms, respectively). In the E+BEV group, only 10 (17.2%) treatment-related serious AEs were reported, compared to 18 (30.5%) in the T+BEV arm. No gastrointestinal perforation was reported.
The analysis of the QoL data (based on the EORTC QLQ-C30 questionnaire) did not show any difference between treatment arms. The median global scores remained stable through the study: 62.5 and 66.7 at baseline (n=104) for the T+BEV and the E+BEV groups, respectively; 66.7 and 66.7 at 6 weeks (n=64); and 66.7 and 66.7 at 12 weeks (n=63).
Updated OS results Updated data after 35-month median follow-up showed that OS from randomization was not different between the two arms (Figure 4). Death rates were 44% and 55% in the T+BEV and E+BEV arms, respectively. All deaths were disease-related.
Discussion The aim of this randomized phase III study was to assess the efficacy of exemestane plus bevacizumab as a maintenance therapy after first line taxane and bevacizumab combination. Our hypothesis was that E+BEV maintenance treatment would increase PFS when compared with continuation of taxane plus bevacizumab until progression or toxicity. PFS6m was 67.2% (95% CI 53.6, 77.7) and 55.2% (95% CI 41.5, 66.9) with T+BEV and E+BEV, respectively (HR: 1.0, 95% CI 0.7, 1.5, p=0.998), meaning that we failed to show superior efficacy with this treatment strategy. However, in the control arm, median exposure to taxane
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Quality of life
13
after randomization was only 2.3 months. Looking at subgroup analyses, patients without previous history of hormone resistance may have benefited from the E+BEV maintenance strategy, especially those who were AI treatment-naïve (HR: 0.85, 95% CI 0.47, 1.54). This trend was reversed for patients with a history of progression on previous AI treatment (HR: 1.25, 95% CI 0.70, 2.24) (Figure 3). These suggestions are based on post-hoc exploratory subgroup analyses with small numbers and low statistical power. However, they may deserve further investigation.
negative breast cancer is widely used and is considered a reasonable treatment option in the Advanced Breast Cancer guidelines [1,18]. However, published data supporting this strategy are limited [7-9,19]. In a retrospective analysis of French patient cohorts, maintenance endocrine treatment after first line chemotherapy was associated with longer PFS (HR: 0.64, 95 %CI 0.57, 0.72) and OS (HR: 0.81, 95 %CI 0.71, 0.92) [9]. In multivariate analysis, the maintenance strategy improved PFS for patients who had received high-dose chemotherapy with hematopoietic progenitor cell transplant, and especially for those patients who were endocrine therapy-naïve [20]. Nevertheless, to our knowledge, only one prospective randomized study has shown the superiority of adding endocrine treatment (medroxyprogesterone acetate) to standard treatment. This study was conducted in patients who had received six cycles of anthracycline- and ifosfamide-containing chemotherapy. PFS was 4.9 months in the endocrine maintenance arm versus 3.7 months without maintenance therapy [21]. On the other hand, consistent with our results, one recent prospective randomized study did not show any difference when endocrine therapy after 4-8 cycles of docetaxel/capecitabine combination was compared to capecitabine monotherapy (10.9 months versus 11.1 months) [22]. Part of our rationale was that the E+BEV combination might reverse endocrine treatment resistance and confer efficacy even in patients having previously received nonsteroidal AIs. This
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Maintenance endocrine therapy after chemotherapy for metastatic ER-positive HER2-
14
hypothesis was not confirmed, and patients whose disease recurred or progressed during nonsteroidal AIs did not derive benefit from the addition of exemestane to continued bevacizumab. Other randomized clinical trials testing the addition of bevacizumab to endocrine therapy have shown conflicting results. The LEA trial failed to demonstrate a statistically significant increase in PFS for endocrine therapy plus bevacizumab compared to endocrine therapy alone in front line treatment [23]. On the other hand, the CALGB 40503 study showed an improvement
of bevacizumab to letrozole as first line endocrine treatment. However, in this study, 43-48% of patients were endocrine therapy-naïve and almost 80% of patients were AI treatment naïve [24]. This is in sharp contrast to the AROBASE patient population described here. Our results have to be interpreted in the light of the IMELDA trial. The design was similar to our AROBASE trial, selecting patients with at least stable disease after 9-18 weeks of docetaxel plus bevacizumab combination. However, it favored the concept of switched maintenance chemotherapy (maintenance with another type of chemotherapy, i.e. capecitabine) combined with bevacizumab maintenance. The IMELDA trial was positive, with a hazard ratio for PFS of 0.38 (95 CI 0.27, 0.55) [25]. Furthermore, OS, a secondary end-point, was significantly improved from 23 to 39 months with capecitabine/bevacizumab maintenance treatment. This is the first clinical trial to have shown significant OS benefit in this setting. One has to ask why the two “maintenance” studies had such different results. Part of the explanation may be that the benefit of maintenance treatment is greatest when the treatment modality remains the one that achieved initial tumor control: i.e., selecting patients whose disease responded to chemotherapy selects patients with chemo-sensitive tumors. If this is the case, continuing chemotherapy may be superior to modifying the treatment strategy, as explored in our AROBASE trial. This hypothesis is in line with data that have been published regarding combined endocrine therapies
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
in PFS from 16 months to 20 months (HR: 0.75, 95% CI 0.59, 0.96, p=0.016) with the addition
15
for patient with “highly hormone sensitive” disease, as well as dual HER2 targeted therapies for the front line treatment for HER2-positive MBC [26].
In conclusion, our results do not support the (re)introduction of endocrine treatment for maintenance therapy in combination with a bevacizumab-based regimen, even though our safety data confirmed the favorable toxicity profile of the E+BEV combination. There was no benefit over standard taxane and bevacizumab maintenance therapy in terms of PFS or OS.
treatment naïve, though statistical significance was not reached in this limited number of patients. Future studies of maintenance therapy should include molecular selection of patients for specific targeted therapies.
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Subgroup analysis showed a trend for better tumor control among patients who were AI-
16
Disclosure OT received honoraria from Roche, Novartis, Pfizer, and Astra-Zeneca and advisory role for Roche. VD received honoraria from Roche, Novartis, and Pfizer and advisory role for Roche, Novartis, and Pfizer. TP received honoraria from Roche. TB received honoraria from Novartis, advisory role for Roche, Novartis, and Astra-Zeneca and research funding from Roche, Novartis,
All remaining authors have declared no conflicts of interest.
Acknowledgements This work was supported by Roche France and the “Ligue Nationale Contre le Cancer” (unrestricted grant). Rob Stepney, PhD (medical writer, Charlbury UK) assisted in the editing of the manuscript. The authors thank the patients who participated in the trial. We acknowledge Nicolas Gane, Marion Gauthier and Bénédicte Votan from the GINECO study office. We also thank the following investigators and pharmacists who participated in the trial: Investigators: J.Y. PIERGA (Institut Curie - Hôpital Claudius Régaud, PARIS); A. LORTHOLARY (Centre Catherine de Sienne, NANTES); J.M. FERRERO (Centre Antoine Lacassagne, NICE); A.C. HARDY-BESSARD (Clinique Armoricaine de Radiologie, SAINT-BRIEUC); I. RAY-COQUARD (Centre Léon Bérard, LYON); E. CAROLA (Centre Hospitalier de Senlis, SENLIS); H. JOOSTEN (Centre Hospitalier de Valence, VALENCE); J. MEUNIER (Centre Hospitalier Régional d'Orléans, ORLEANS); B. HOCH (Centre Azuréen de Cancérologie, MOUGINS); C. CLIPPE (Hôpitaux Drôme Nord, ROMANS-SUR-ISERE); A.M. SAVOYE (Institut Jean Godinot, REIMS); D. SPAETH (ORACLE - Centre d'Oncologie de Gentilly, NANCY); E. ANGELLIER (Clinique
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
and Astra-Zeneca.
17
Notre-Dame de Bon Secours, CHARTRES); C.B. LEVACHE (Clinique Francheville, PERIGUEUX); E. ACHILLE (Institut d'Oncologie, STRASBOURG); E. LEGOUFFE (Polyclinique KENVAL, NIMES); E. BRAIN (Hôpital René Huguenin, SAINT-CLOUD); A. MARTI (Centre Hospitalier d'Auxerre, AUXERRE); G. DEPLANQUE (Groupe Hospitalier Saint-Joseph, PARIS); O. ARSENE (Centre Hospitalier de Blois, BLOIS); P.E. HEUDEL (Centre Hospitalier de la Région d'Annecy, PRINGY); S. HENAULT (Centre Hospitalier Intercommunal de Créteil, CRETEIL); T. L'HARIDON, F. PRIOU (Centre Hospitalier Départemental Les Oudairies, LA
Pharmacists: C. ENGEL (Centre Léon Bérard, LYON); E. BESREST (Centre Antoine Lacassagne, NICE); A. GRASS (Hôpital Privé Jean Mermoz, LYON); D. BAYLOT (Clinique de la Sauvegarde, LYON); F. BADIBOUIDI (Centre Hospitalier de Senlis, SENLIS); B. GOSSELIN (Hôpital de Mont-deMarsan, MONT- DE-MARSAN); DE VINZELLES, RENON-CARON (Centre Hospitalier Universitaire Dupuytren, LIMOGES); L. ESCALUP (Institut Curie - Hopital Claudius Régaud, PARIS); F. MOREY (Hôpital Fleyriat, BOURG-EN-BRESSE); A. MARCHAUD (Hôpitaux Drôme Nord, ROMANS-SUR-ISERE); C. LABAT (Centre Hospitalier La Dracénie, DRAGUIGNAN); R. ALFONSI (ORACLE - Centre d'Oncologie de Gentilly, NANCY); F. POIROT-LUTRIN (Centre Hospitalier de la Région d'Annecy, PRINGY); J.B. REY (Institut Jean Godinot, REIMS); P. BERNARD (Clinique de l'Union, SAINT-JEAN); E. VERGNES (Centre Paul Strauss, STRASBOURG); R. BESSARD (Centre Hospitalier Bretagne Atlantique, VANNES); J. DIMET (Centre Hospitalier Départemental Les Oudairies, LA ROCHE-SUR-YON); L. GUIVARCH (Centre Catherine de Sienne, NANTES); P. PLOCCO, V. PRIOU, S. TOLLEC (Centre Hospitalier Régional d'Orléans, ORLEANS); S. ROUSSEL (Clinique Armoricaine de Radiologie, SAINT-BRIEUC); I. DUFRENE (Centre Hospitalier de Valence, VALENCE); R. BESSARD (Centre d'Oncologie Saint-Yves, VANNES); N. VERMERIE (Centre Hospitalier d'Auxerre, AUXERRE); P. BRETON (Centre Hospitalier de Blois, BLOIS); E. OLIVIER (Clinique Notre-
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
ROCHE-SUR-YON)
18
Dame de Bon Secours, CHARTRES); S. POULAIN, A. THEBAULT (Centre Hospitalier Intercommunal de Créteil, CRETEIL); S. ROUBAUD (Centre Azuréen de Cancérologie, MOUGINS); I. CHABROLLES (Polyclinique KENVAL, NIMES); M. JARDIN (Groupe Hospitalier Saint-Joseph, PARIS); B. SOUSSELIER (Clinique Francheville, PERIGUEUX); I. FERRY (Hôpital René Huguenin, SAINT-CLOUD); L. MATHIS, F. LAURENT (Institut d'Oncologie, STRASBOURG)
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
19
References 1. Cardoso F, Costa A, Norton L, et al. ESO-ESMO 2nd international consensus guidelines
for advanced breast cancer (ABC2). Ann Oncol 2014; 25: 1871-1888. 2. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in
Oncology (NCCN Guidelines). Breast Cancer 20th annual edition. http://www.nccn.org/professionals/physician_gls/f_guidelines.asp#site (accessed Dec 15, 2015).
for metastatic breast cancer. N Engl J Med 2007; 357: 2666-2676. 4. Miles DW, Diéras V, Cortés J, et al. First-line bevacizumab in combination with
chemotherapy for HER2-negative metastatic breast cancer: pooled and subgroup analyses of data from 2447 patients. Ann Oncol 2013; 24: 2773-2780. 5. Robert NJ, Diéras V, Glaspy J, et al. RIBBON-1: randomized, double-blind, placebo-
controlled, phase III trial of chemotherapy with or without bevacizumab for first-line treatment of human epidermal growth factor receptor 2-negative, locally recurrent or metastatic breast cancer. J Clin Oncol 2011; 29: 1252-1260. 6. Miles DW, Chan A, Dirix LY, et al. Phase III study of bevacizumab plus docetaxel
compared with placebo plus docetaxel for the first-line treatment of human epidermal growth factor receptor 2-negative metastatic breast cancer. J Clin Oncol 2010; 28: 32393247. 7. Berruti A, Zola P, Buniva T, et al. Prognostic factors in metastatic breast cancer patients
obtaining objective response or disease stabilization after first-line chemotherapy with epirubicin. Evidence for a positive effect of maintenance hormonal therapy on overall survival. Anticancer Res 1997; 17: 2763-2768. 8. Bertelli G, Garrone O, Bertolotti L, et al. Maintenance hormone therapy with letrozole
after first-line chemotherapy for advanced breast cancer. Oncology 2005; 68: 364-370.
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
3. Miller K, Wang M, Gralow J, et al. Paclitaxel plus bevacizumab versus paclitaxel alone
20
9. Dufresne A, Pivot X, Tournigand C, et al. Maintenance hormonal treatment improves
progression free survival after a first line chemotherapy in patients with metastatic breast cancer. Int J Med Sci 2008; 5: 100-105. 10. Dadiani M, Seger D, Kreizman T, et al. Estrogen regulation of vascular endothelial
growth factor in breast cancer in vitro and in vivo: the role of estrogen receptor alpha and c-Myc. Endocr Relat Cancer 2009; 16: 819-834. 11. Adams J, Carder PJ, Downey S, et al. Vascular endothelial growth factor (VEGF) in
and effects of tamoxifen. Cancer Res 2000; 60: 2898-2905. 12. Holmes CE, Huang JC, Pace TR, et al. Tamoxifen and aromatase inhibitors differentially
affect vascular endothelial growth factor and endostatin levels in women with breast cancer. Clin Cancer Res 2008; 14: 3070-3076. 13. Traina TA, Rugo HS, Caravelli JF, et al. Feasibility trial of letrozole in combination with
bevacizumab in patients with metastatic breast cancer. J Clin Oncol 2010; 28: 628-633. 14. Forero-Torres A, Saleh MN, Galleshaw JA, et al. Pilot trial of preoperative (neoadjuvant)
letrozole in combination with bevacizumab in postmenopausal women with newly diagnosed estrogen receptor- or progesterone receptor-positive breast cancer. Clin Breast Cancer 2010; 10: 275-280. 15. Yardley DA, Burris HA 3rd, Clark BL, et al. Hormonal therapy plus bevacizumab in
postmenopausal patients who have hormone receptor-positive metastatic breast cancer: a phase II Trial of the Sarah Cannon Oncology Research Consortium. Clin Breast Cancer 2011; 11: 146-152. 16. Fabi A, Russillo M, Ferretti G, et al. Maintenance bevacizumab beyond first-line
paclitaxel plus bevacizumab in patients with Her2-negative hormone receptor-positive metastatic breast cancer: efficacy in combination with hormonal therapy. BMC Cancer 2012; 12: 482.
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
breast cancer: comparison of plasma, serum, and tissue VEGF and microvessel density
21
17. Chia S, Gradishar W, Mauriac L, et al. Double-blind, randomized placebo controlled trial
of fulvestrant compared with exemestane after prior nonsteroidal aromatase inhibitor therapy in postmenopausal women with hormone receptor-positive, advanced breast cancer: results from EFECT. J Clin Oncol 2008; 26: 1664-1670. 18. Cardoso F, Costa A, Norton L, et al. 1st International consensus guidelines for advanced
breast cancer (ABC 1). Breast 2012; 21: 242-252. 19. Redondo A, Martínez V, Zamora P, et al. Continuation of bevacizumab and addition of
cancer. Onco Targets Ther 2014; 7: 2175-2181. 20. Montemurro F, Rondón G, Ueno NT, et al. Factors affecting progression-free survival in
hormone-dependent metastatic breast cancer patients receiving high-dose chemotherapy and hematopoietic progenitor cell transplantation: role of maintenance endocrine therapy. Bone Marrow Transplant 2002; 29: 861-866. 21. Kloke O, Klaassen U, Oberhoff C, et al. Maintenance treatment with
medroxyprogesterone acetate in patients with advanced breast cancer responding to chemotherapy: results of a randomized trial. Essen Breast Cancer Study Group. Breast Cancer Res Treat 1999; 55: 51-59. 22. Liang X, Yan Y, Wang L, et al. First-line chemotherapy with docetaxel plus capecitabine
followed by capecitabine or hormone maintenance therapy for the treatment of metastatic breast cancer patients. Oncol Lett 2015; 9: 987-993. 23. Martín M, Loibl S, von Minckwitz G, et al. Phase III Trial Evaluating the Addition of
Bevacizumab to Endocrine Therapy As First-Line Treatment for Advanced Breast Cancer: The Letrozole/Fulvestrant and Avastin (LEA) Study. J Clin Oncol 2015; 33: 1045-1052. 24. Dickler MN, Barry WT, Cirrincione CT, et al. Phase III trial evaluating the addition of
bevacizumab to letrozole as first-line endocrine therapy for treatment of hormone-
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
hormone therapy following weekly paclitaxel therapy in HER2-negative metastatic breast
22
receptor positive advanced breast cancer: CALGB 40503 (Alliance). J Clin Oncol 2015; 33 (suppl; abstr 501). 25. Gligorov J, Doval D, Bines J, et al. Maintenance capecitabine and bevacizumab versus
bevacizumab alone after initial first-line bevacizumab and docetaxel for patients with HER2-negative metastatic breast cancer (IMELDA): a randomised, open-label, phase 3 trial. Lancet Oncol 2014; 15: 1351-1360. 26. Bachelot T, Tredan O. Targeted treatments for breast cancer: a step forward. Lancet Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Oncol 2013; 14: 438-439.
23
Table 1: Patient demographics and baseline characteristics Table 2: Chemotherapy regimens prior to randomization and after randomization Table 3: Summary of efficacy results (all randomized patients) Table 4: AEs occurring in ≥10 % of patients in either treatment group
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
24
Figure Caption Figure 1: Distribution of patients
117 patients provided consent
117 patients evaluable for efficacy
59 received taxane and bevacizumab as continuation of first line chemotherapy
58 received exemestane and bevacizumab as maintenance therapy after chemotherapy
56 discontinued taxane
53 discontinued bevacizumab
47 discontinued exemestane
48 discontinued bevacizumab
12 disease progressions 21 adverse events or toxic effects after randomisation 8 adverse events or toxic effects before randomisation 13 patient or investigator decisions 2 other
41 disease progressions 7 adverse events or toxic effects 4 patient or investigator decisions 1 other
41 disease progressions 5 adverse events or toxic effects 1 other
36 disease progressions 9 adverse events or toxic effects 1 patient or investigator decisions 2 other
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
117 patients evaluable for safety
25
Figure 2: Kaplan-Meier plot for PFS (randomized subjects) Abbreviations: CI, confidence interval; PFS, progression-free survival; BEV = bevacizumab; E = Exemestane; T = taxanes.
Figure 3: Forest plot for the PFS effect of the treatment in (pre)defined subgroups
= Exemestane; TAX = taxanes.
Figure 4: Updated OS with 35-month follow-up Abbreviations: CI, confidence interval; OS, overall survival; BEV = bevacizumab; E = Exemestane; T = taxanes.
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Abbreviations: CI, confidence interval; PFS, progression-free survival; BEV = bevacizumab; EXE
26
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
27
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
28
Downloaded from http://annonc.oxfordjournals.org/ at University of Newcastle on March 1, 2016
Table 1: Patient demographics and baseline characteristics Number (%) of patients Patients included in the interim
Patients included in the final analysis
analysis (n=98) Taxane
Exemestane
+ BEV
+ BEV
(n=59)
(n=58)
55 [35-86]
55 [35-86]
56 [35-77]
0
46 (48.9)
29 (51.8)
29 (51.8)
1
46 (48.9)
26 (46.4)
26 (46.4)
2
2 (2.1)
1 (1.8)
1 (1.8)
4
3
2
4.4 (0-29)
4.3 (0-29)
4.6 (0-34)
≤6 months
13 (22.0)
16 (27.6)
>6 months
46 (78.0)
42 (72.4)
11 (18.6)
10 (17.2)
Age [Years] median (range) ECOG PS
NR Median time (range) from diagnosis to metastatic disease [years]
Metastatic sites bone mets only bone
61 (62.2)
39 (66.1)
36 (62.1)
lung
36 (36.7)
13 (22.0)
28 (48.3)
liver
44 (44.9)
31 (52.5)
25 (43.1)
≥3
37 (37.8)
20 (33.9)
22 (37.9)
ER+ / PR +
58 (59.8)
36 (61.0)
34 (59.6)
ER+ / PR -
39 (40.2)
23 (39.0)
23 (40.4)
(neo)adjuvant chemo
64 (65.3)
38 (64.4)
35 (61.4)
adjuvant endocrine
63 (64.3)
39 (66.1)
33 (57.9)
metastatic endocrine.
23 (23.5)
18 (30.5)
12 (20.7)
8 (8.2)
12 (20.3)
9 (15.5)
post-menopausal at inclusion
76 (77.6)
45 (76.3)
47 (81.0)
prior AIs
46 (46.9)
29 (49.2)
27 (46.6)
endocrine resistance
42 (42.9)
26 (44.1)
23 (39.7)
ER and PR status
Setting of prior therapies
metastatic AIs .
Stratification factors
Abbreviations: BEV = bevacizumab; NR = not reported; ECOG PS = Eastern Cooperative Group Performance Status; mets = metastases; ER = estrogen receptor; PR = progesterone receptor; AIs = aromatase inhibitors
Table 2: Chemotherapy regimens prior to randomization and after randomization Number (%) of patients
Treatment before randomization (months) paclitaxel + BEV docetaxel + BEV median duration [range]
Treatment after randomization (months) BEV median duration [range] taxane median duration [range] exemestane median duration [range] Abbreviations: BEV = bevacizumab
Taxane + BEV
Exemestane + BEV
(n=59)
(n=58)
58 (98.3)
57 (98.3)
1 (1.7)
1 (1.7)
3.7 [2.3-5.7]
3.4 [2.07-6]
6.3 [0.5-34.4]
5.3 [0.03-41.8]
2.3 [0.0-121.4]
-
-
6.0 [0.2-41.8]
Table 3: Summary of efficacy results (all randomized patients)
Median follow up [range]
Taxane + BEV
Exemestane + BEV
(n=59)
(n=58)
21.4 months [0.8-42.9]
20.6 months [1.2-43.1]
48 (81)
44 (76)
8.1 (6.5-10.7)
7.6 (5.4-10.9)
67.2% (53.6-77.7)
55.2% (41.5-66.9)
Primary endpoint PFS: number of events (%) median, months (95%CI) 6-month PFS rate (95%CI)
Log-rank test p=0.998 HR: 1.00 (0.66-1.51)
Secondary endpoints OS after 35-month median follow-up: number of events (%)
26 (44)
32 (55)
PFS from the beginning of bevacizumab: number of events (%)
47 (81)
44 (76)
12.5 (10.2-14.8)
12.3 (9.2-14.7)
median, months (95%CI)
Abbreviations: BEV = bevacizumab; CI = confidence interval; PFS = progression-free survival; TTP = Time To Progression from the beginning of the taxane plus bevacizuamb treatment
Table 4: AEs occurring in ≥10 % of patients in either treatment group Number (%) of patients Taxane + BEV
Exemestane + BEV
(n=59)
(n=58)
any grade / grade 3-4
any grade / grade 3-4
47 (79.7%) / 8 (13.6%)
30 (51.7%) / 3 (5.2%)
Hypertension
54 (91.5%) / 21 (35.6%)
53 (91.4%) / 25 (43.1%)
Hemorrhage
23 (39.0%) / 1 (1.7%)
14 (24.1%) / 0
3 (5.1%) / 0
3 (5.2%) / 1 (1.7%)
Vomiting
3 (5.1%) / 0
11 (19.0%) / 0
Nausea
18 (30.5%) / 0
14 (24.1%) / 0
12 (20.3%) / 1 (1.7%)
6 (10.3%) / 0
10 (16.9%) / 0
10 (17.2%) / 0
39 (66.1%) / 8 (13.6%)
19 (32.8%) / 0
9 (15.3%) / 0
5 (8.6%) / 0
Alopecia
35 (59.3%) / 0
17 (29.3%) / 0
Mucositis
9 (15.3%) / 0
4 (6.9%) / 0
Nail disorders
26 (44.1%) / 0
9 (15.5%) / 0
16 (27.1%) / 1 (1.7%)
10 (17.2%) / 0
Myalgia
16 (27.1%) / 0
7 (12.1%) /0
Arthralgia
18 (30.5%) / 0
25 (43.1%) / 2 (3.4%)
General disorders Fatigue Cardiovascular disorders
Thromboembolic events Gastrointestinal disorders
Constipation Diarrhea Nervous system disorders Peripheral sensory neuropathy Peripheral motor neuropathy Skin/mucosa disorders
Edema Musculoskeletal disorders
Blood disorders Neutropenia
35 (59.3%) / 7 (11.9%)
15 (25.9%) / 0
1 (1.7%) / 0
0
Anemia
34 (57.6%) / 3 (5.1%)
22 (37.9%) / 0
Alanine transaminase elevated
30 (50.8%) / 2 (3.4%)
25 (43.1%) / 1 (1.7%)
Aspartate transaminase elevated
38 (64.4%) / 2 (3.4%)
35 (60.3%) / 3 (5.2%)
Febrile neutropenia
Abbreviation: AE, adverse event.
