Sep 28, 2014 - calculated according to Cockcroft-Gault equation, urine dipstick for ...... Daniel Lawson), and the biostatisticians (F Hoffmann-La Roche:.
Articles
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 Joseph Gligorov, Dinesh Doval, José Bines, Emilio Alba, Paulo Cortes, Jean-Yves Pierga, Vineet Gupta, Rômulo Costa, Stefanie Srock, Sabine de Ducla, Ulrich Freudensprung, Giorgio Mustacchi
Summary Background Longer duration of first-line chemotherapy for patients with metastatic breast cancer is associated with prolonged overall survival and improved progression-free survival. We investigated capecitabine added to maintenance bevacizumab after initial treatment with bevacizumab and docetaxel in this setting. Methods We did this open-label randomised phase 3 trial at 54 hospitals in Brazil, China, Egypt, France, Hong Kong, India, Italy, Poland, Spain, and Turkey. We enrolled patients with HER2-negative measurable metastatic breast cancer; each received three to six cycles of first-line bevacizumab (15 mg/kg) and docetaxel (75–100 mg/m²) every 3 weeks. Progression-free patients were randomly assigned with an interactive voice-response system by block (size four) randomisation (1:1) to receive either bevacizumab and capecitabine or bevacizumab only (bevacizumab 15 mg/kg on day 1; capecitabine 1000 mg/m² twice per day on days 1–14, every 3 weeks) until progression, stratified by oestrogen receptor status (positive vs negative), visceral metastases (present vs absent), response status (stable disease vs response vs non-measurable), and lactate dehydrogenase concentration (≤1·5 vs >1·5 × upper limit of normal). Neither patients nor investigators were masked to allocation. The primary endpoint was progression-free survival (from randomisation) in the intention-to-treat population. This trial is registered with ClinicalTrials.gov, NCT00929240. Findings Between July 16, 2009, and March 7, 2011 (when enrolment was prematurely terminated), 284 patients received initial bevacizumab and docetaxel; 185 (65%) were randomly assigned (91 to bevacizumab and capecitabine versus 94 to bevacizumab only). Progression-free survival was significantly longer in the bevacizumab and capecitabine group than in the bevacizumab only group (median 11·9 months [95% CI 9·8–15·4] vs 4·3 months [3·9–6·8]; stratified hazard ratio 0·38 [95% CI 0·27–0·55]; two-sided log-rank p2·5 × ULN, or >5 × ULN for patients with liver metastases, or >10 × ULN for patients with bone metastases), or renal function (serum creatinine concentration >1·5 × ULN, creatinine clearance 1·5 except for patients receiving anticoagulation treatment at a stable dose for ≥2 weeks and with coagulation monitoring within local therapeutic limits, activated partial thromboplastin time >1·5 × ULN within 7 days before the first study dose). www.thelancet.com/oncology Vol 15 November 2014
Articles
All patients provided written informed consent. The study was done in accordance with the guidelines for good clinical practice and the Declaration of Helsinki. An independent ethics committee at each participating site approved the protocol and all modifications. In February, 2011, regulatory approval of combined bevacizumab and docetaxel for patients with metastatic breast cancer was withdrawn. Some centres subsequently stopped accrual or withdrew patients from study treatment, or both. As a result, recruitment was prematurely terminated. An independent data monitoring committee was formed to monitor patients’ safety. Patients remaining in the study were followed up as initially planned (during treatment: every cycle for safety and every 9 weeks for efficacy; after treatment: every 3 months until disease progression) except that, after a protocol amendment, follow-up duration was extended to continue until 2 years after the last patient was randomly assigned (or until lost to follow-up, or consent withdrawal, if earlier).
Randomisation and masking All enrolled patients were initially treated with bevacizumab and docetaxel; those achieving complete or partial response, or stable disease, were then randomly assigned to treatment with bevacizumab either alone or in combination with capecitabine. Patients were randomly assigned (1:1) with an interactive voice-response system, and a block-design randomisation procedure (block size four). Stratification factors were oestrogen receptor status (positive vs negative), visceral metastases (present vs absent), response status (stable disease vs response vs non-measurable), and lactate dehydrogenase concentration (≤1·5 vs >1·5 × upper limit of normal). The randomisation scheme was done by Parexel Informatics (Nottingham, UK). Neither participants nor investigators were masked to treatment allocation.
Procedures All patients received initial treatment with up to six cycles of bevacizumab (15 mg/kg) and docetaxel (75–100 mg/m²), on day 1 every 3 weeks. Patients achieving stable disease, or a complete or partial response, and still meeting the eligibility criteria were randomly assigned to continue bevacizumab either alone (15 mg/kg; day 1 every 3 weeks) or with capecitabine 1000 mg/m² twice a day on days 1–14, every 3 weeks. If a patient had responded to initial treatment by the third cycle, but had toxic effects requiring docetaxel treatment interruption, she could start the randomised phase immediately, at the investigator’s discretion. Maintenance treatment was continued until disease progression, unacceptable toxic effects, or withdrawal of consent. If either bevacizumab or capecitabine was discontinued before progression in patients receiving combination treatment, the other drug was continued as monotherapy. After progression on maintenance www.thelancet.com/oncology Vol 15 November 2014
treatment, all patients were offered standard-of-care treatment. We did not systematically collect details of postprogression treatment. Only grade 3 or worse adverse events were routinely reported (Common Terminology Criteria for Adverse Events version 3.0), although data could also be recorded for adverse events of any grade that led to withdrawal of study drug. Reduction or modification of bevacizumab dose for toxic effects was not permitted, but bevacizumab treatment was suspended or permanently discontinued in the event of grade 3 or 4 hypertension, proteinuria, thrombosis or embolism, haemorrhage or bleeding, congestive heart failure, or wound-healing complication, or any grade gastrointestinal perforation, tracheooesophageal or non-gastrointestinal fistula, reversible encephalopathy syndrome, or hypersensitivity or allergic reaction related to bevacizumab, or other serious grade 3 or 4 bevacizumab-related toxic effect (appendix). Capecitabine dose modifications or delays were made according to individual institutional guidelines, or the applicable prescribing information (appendix). Tumours were assessed at baseline and at the end of every third cycle with CT scans, MRI, radiography, bone scan, or clinical examinations. The same imaging technique was
See Online for appendix
331 patients screened 44 excluded* 287 enrolled 3 withdrew consent 99 discontinued initial treatment 41 disease progression 31 adverse events or toxic effects 13 withdrew consent or patient decision 5 inclusion or exclusion criteria or protocol violation 3 health authority or study termination 4 investigator or medical decision 2 died
284 treated in initial phase
185 eligible for randomisation
94 assigned to bevacizumab only 92 treated 2 untreated
91 assigned to bevacizumab and capecitabine 91 treated
92 discontinued maintenance treatment 73 disease progression 9 adverse events or toxic effects 2 withdrew consent or patient decision 1 inclusion or exclusion criteria or protocol violation 1 health authority or study termination 2 investigator or medical decision 4 changed treatment
81 discontinued maintenance treatment 60 disease progression 12 adverse events or toxic effects 6 withdrew consent or patient decision 2 health authority or study termination 1 investigator or medical decision 10 treatment ongoing at study closure
Figure 1: Trial profile *Reasons for exclusion not collected.
1353
Articles
used throughout the study for each patient and, as far as possible, tumour assessments were done by the same investigator or radiologist. All laboratory assessments were done locally. Bevacizumab only (n=94)
Bevacizumab and capecitabine (n=91)
54 (24–77)
49 (24–80)
≥65
13 (14%)
14 (15%)
≥70
5 (5%)
7 (8%)
0
57 (61%)
44 (48%)
1
37 (39%)
47 (52%)
At baseline Age (years)
ECOG performance status
Hormone receptor status Triple negative
21 (22%)
25 (27%)
ER or PgR positive or both
73 (78%)
66 (73%)
54 (57%)
43 (47%)
≥3 metastatic organs
At randomisation (stratification factors) Visceral metastases
65 (69%)
62 (68%)
Response
68 (72%)
68 (75%)
Stable disease
22 (23%)
20 (22%)
Response to initial phase therapy
Non-measurable disease
4 (4%)
3 (3%)
LDH concentration ≤1·5 × upper limit of normal
89 (95%)
85 (93%)
ER positive
69 (73%)
64 (70%)
Data are number (%) or median (range). ECOG=Eastern Cooperative Oncology Group. ER=oestrogen receptor. LDH=lactate dehydrogenase. PgR=progesterone receptor.
Table 1: Participant characteristics
Initial phase
Maintenance phase
Non-randomly All randomly assigned assigned patients (n=99) patients (n=185)
Bevacizumab only (n=92)
Number of cycles
3 (1–6)
6 (2–6)
6 (1–34)
Bevacizumab
3 (1–6)
6 (2–6)
6 (1–34)
Docetaxel
3 (1–6)
6 (2–6)*
Capecitabine
NA
NA
Bevacizumab and capecitabine (n=91) 12 (1–59) 9·5 (1–59)
NA
NA
NA
10 (1–59)
Treatment duration (months)†
1·6 (0·0–4·1)
3·5 (0·7–4·4)
3·5 (0·0–23·2)
8·3 (0·2–41·6)
Treatment duration from baseline (months)
3·5 (0·7–4·4)
3·5 (0·7–4·4)
7·6 (3·5–27·4)
12·5 (3·9–45·7)
Patients still on treatment at cycle 18 of maintenance treatment
NA
NA
9 (10%)
30 (33%)
Patients still on treatment at cycle 36 of maintenance treatment
NA
NA
0
11 (12%)
Data are median (range) or number (%). NA=not applicable. *Six patients were randomly assigned after fewer than four cycles of initial treatment: four to bevacizumab only (two after two cycles of bevacizumab and docetaxel, two after three cycles of bevacizumab and docetaxel) and two patients to bevacizumab and capecitabine (both after three cycles of bevacizumab and docetaxel). †Calculated from the day of first drug administration to the day of last drug administration (ie, the drug-free days in each cycle [days 15–21 for capecitabine, days 2–21 for bevacizumab and docetaxel] are not counted).
Table 2: Treatment exposure
1354
Patient-reported outcomes were assessed with the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-C30, completed at screening (before initial bevacizumab and docetaxel treatment), at randomisation, every three cycles until disease progression, and at (but not after) disease progression.
Outcomes The primary endpoint was investigator-assessed progression-free survival from randomisation, defined as the time from randomisation until disease progression or death. Secondary endpoints in the initial (bevacizumab and docetaxel) treatment phase were the proportions of patients with an objective response (based on best overall response) or clinical benefit (documented complete or partial response, or stable disease), and safety. Secondary endpoints in the maintenance phase were overall survival from randomisation, safety, the proportions of patients achieving an objective response or clinical benefit (complete or partial response, or stable disease), time to progression, and quality of life. We did predefined subgroup analyses by clinical characteristics to assess the internal consistency of the results. We did post-hoc exploratory analyses of progression-free survival and overall survival from the first study dose of initial treatment (instead of randomisation).
Statistical analysis The sample size was based on a log-rank test assuming an increase in median progression-free survival in the maintenance phase from 5·8 months with bevacizumab alone to 8·3 months with bevacizumab and capecitabine, with a hazard ratio (HR) of 0·70. We needed to enrol 360 patients to have 290 eligible for randomisation; 244 progression-free survival events would provide 80% power at a two-sided α of 0·05. We analysed the intention-to-treat population (all randomised patients) for efficacy outcomes, and the perprotocol population (all patients who received at least one dose of maintenance treatment) for safety. There was no central review. We used a stratified Cox proportional hazards model to estimate the HR with 95% CIs. We compared treatment groups with a two-sided log-rank test stratified by stratification factors. We estimated medians by Kaplan-Meier analysis. For the primary analysis, patients who received further cancer treatment before disease progression were censored at the last tumour assessment before first non-study cancer treatment. Sensitivity analyses included an analysis without this censoring approach and an unstratified analysis.11 Adverse events were assigned to either initial or maintenance treatment phases according to the date of onset. We calculated duration of follow-up with reversed Kaplan-Meier methods.12 We did all statistical analyses with SAS (version 9.2). www.thelancet.com/oncology Vol 15 November 2014
Articles
This trial is NCT00929240.
registered
with
ClinicalTrials.gov,
A 100
Results 287 patients were enrolled from 54 hospitals between July 16, 2009, and March 7, 2011. Recruitment was terminated prematurely after withdrawal of European Commission approval of bevacizumab plus docetaxel combination treatment for metastatic breast cancer on Feb 28, 2011. Of the 284 patients treated in the initial phase, 99 (35%) discontinued initial study treatment, and were therefore ineligible for randomisation (figure 1). The remaining 185 (65%) of 284 patients completed the initial phase and were randomly assigned in the maintenance phase (two randomly assigned patients were untreated; table 1). Median exposure to initial treatment was six cycles (IQR five to six) for both docetaxel and bevacizumab. The most common reason for discontinuation of initial treatment was disease progression (figure 1). Efficacy endpoints for the initial phase have been previously presented (table 1).13 In the maintenance phase, treatment exposure was longer in the combination group than in the bevacizumab only group (table 2). At the data cutoff date (Oct 4, 2013), the median duration of follow-up in the maintenance period was 30·4 months (IQR 27·0–34·0) in the bevacizumab only group, and 31·6 months (IQR 27·1–35·4) in the bevacizumab and capecitabine group. Ten patients (all in the bevacizumab and capecitabine group) were still receiving study treatment (bevacizumab, capecitabine, or both). Because of study closure, follow-up was stopped for 31 (33%) of 94 patients receiving bevacizumab and 46 (51%) of 91 receiving bevacizumab and capecitabine. 83 (88%) of 94 patients receiving bevacizumab alone had disease progression, as did 69 (76%) of 91 patients receiving bevacizumab and capecitabine. Progression-free survival was significantly improved in the the bevacizumab and capecitabine group (median 11·9 months [95% CI 9·8–15·4]) than in the bevacizumab only group (median 4·3 months [95% CI 3·9–6·8]; stratified HR 0·38, 95% CI 0·27–0·55; two-sided log-rank p
