original articles
Annals of Oncology
Annals of Oncology 23: 589–597, 2012 doi:10.1093/annonc/mdr256 Published online 1 June 2011
Randomized phase II non-inferiority study (NO16853) of two different doses of capecitabine in combination with docetaxel for locally advanced/metastatic breast cancer A. U. Buzdar1*, B. Xu2, R. Digumarti3, L. Goedhals4, X. Hu5, V. Semiglazov6, S. Cheporov7, E. Gotovkin8, S. Hoersch9, K. Rittweger10, D. W. Miles11, J. O’Shaughnessy12 & S. Tjulandin13 on behalf of the NO16853 trial group 1 Department of Breast Medical Oncology, University of Texas MD Anderson Cancer Center, Houston, USA; 2Department of Medical Oncology, Cancer Institute and Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China; 3Department of Medical Oncology, Nizam’s Institute of Medical Sciences, Hyderabad, India; 4Department of Oncotherapy, National Hospital, Bloemfontein, South Africa; 5Cancer Hospital, Fudan University, Shanghai, China; 6Breast Cancer Department, NN Petrov Research Institute of Oncology, St Petersburg, Russia; 7Department of Oncology, Regional Clinical Oncology Hospital, Yaroslavl, Russia; 8Department of Oncology, Regional Oncology Dispensary, Ivanovo, Russia; 9Department of Statistics, Dr Manfred Ko¨hler GmbH, Freiburg, Germany; 10Product Development Oncology Department, Hoffmann-La Roche Inc, Nutley, USA; 11Department of Medical Oncology, East and North Hertfordshire NHS Trust, Mount Vernon Cancer Centre, Middlesex, UK; 12Department of Medical Oncology, Baylor-Sammons Cancer Center, Texas Oncology, US Oncology, Dallas, USA; 13 Department of Clinical Pharmacology and Chemotherapy, Blokhin Cancer Research Center, Moscow, Russia
Received 7 December 2010; revised 7 March 2011; accepted 29 March 2011
Background: This phase II study investigated whether a lower-than-approved dose of capecitabine, plus docetaxel (XT), would improve tolerability versus standard-dose XT without compromising efficacy.
Patients and methods: Women aged ‡18 years with locally advanced/metastatic breast cancer resistant to anthracycline-based chemotherapy in the (neo)adjuvant, first- or second-line metastatic setting were eligible. Patients were randomly assigned to receive standard-dose XT (capecitabine 1250 mg/m2 twice daily, days 1–14; docetaxel 75 mg/m2, day 1 every 3 weeks) or low-dose XT (capecitabine 825 mg/m2 twice daily, days 1–14; docetaxel as above). The primary objective was to demonstrate non-inferiority of low-dose to standard-dose XT in terms of progression-free survival (PFS). Results: 470 patients were randomly allocated in a 1 : 1 ratio to standard-dose or low-dose XT. Median PFS was 7.9 versus 5.8 months [hazard ratio 1.16, 95% confidence interval (CI) 0.95–1.43] in the standard-dose and low-dose arms, respectively. The upper limit of the 95% CI was above the predefined non-inferiority margin (1.35, P = 0.078). Secondary efficacy end points were consistent with PFS. The frequency and severity of adverse events was similar in both treatment arms. Conclusions: Non-inferiority of low-dose to standard-dose XT in terms of PFS was not demonstrated; this may be due to regional subgroup effects. Key words: capecitabine, docetaxel, dose modification, metastatic breast cancer
introduction Capecitabine (Xeloda; F. Hoffmann-La Roche Ltd, Basel, Switzerland) is converted to the active moiety, 5-fluorouracil (5-FU), via a triple enzymic cascade including the enzyme thymidine phosphorylase (TP). Several cytotoxic agents, including paclitaxel, docetaxel, vinorelbine, gemcitabine and cyclophosphamide, up-regulate TP activity in tumor tissues [1–4]. Moreover, the coadministration of docetaxel and capecitabine results in synergistic antitumor activity in human *Correspondence to Dr A. U. Buzdar, Department of Breast Medical Oncology, University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Unit 1354, Houston, TX 77030, USA. Tel: +1-713-792-2817; Fax: +1-713-794-4385; E-mail:
[email protected]
cancer xenograft models (MX-1 and WiDr) without increasing toxicity [1]. Docetaxel and capecitabine have proven single-agent and combination activity in metastatic breast cancer (MBC), with non-overlapping safety profiles [5–9], thus providing the rationale for exploring these agents in combination (XT). Regulatory approval of XT for the treatment of MBC following failure of prior anthracycline-containing chemotherapy was based on the results of the phase III registration trial, SO14999, in which 511 patients were randomly assigned to receive 3-weekly cycles of XT (1250 mg/m2 twice daily, days 1–14, plus docetaxel 75 mg/m2, day 1) or docetaxel alone (100 mg/m2, day 1) [10, 11]. Compared with the docetaxel group, patients receiving XT achieved
ª The Author 2011. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email:
[email protected]
original articles significantly superior time to progression (TTP), the primary end point [hazard ratio (HR) 0.652, 95% confidence interval (CI) 0.545–0.780, P = 0.0001; median 6.1 versus 4.2 months, respectively]. Overall survival (OS) was also improved with XT versus docetaxel (HR 0.775, 95% CI 0.634–0.947, P = 0.0126; median 14.5 versus 11.5 months, respectively). However, a higher proportion of patients in the XT group experienced grade 3/4 adverse events (78% versus 64%, respectively) and dose reductions (65% versus 36%, respectively) compared with those receiving docetaxel alone. Dose reductions of both capecitabine and docetaxel were effective in reducing the incidence and recurrence of grade 3/4 adverse events, thereby allowing patients to continue therapy. In a retrospective analysis of SO14999 study data, no impact on TTP or OS was observed in patients requiring XT dose reduction, while the proportion of treatment cycles with grade 3/4 treatment-related adverse events was almost halved [12]. Additional randomized studies and retrospective analyses in patients receiving capecitabine monotherapy, or XT combination therapy, have shown that dose modification of capecitabine [13, 14], and the use of a lower-than-approved starting dose [5, 6, 14–17], in patients with MBC is effective in the management of adverse events without reducing the efficacy of the drug. In fulfillment of a post-approval commitment to the USA Food and Drug Administration, the open-label, multicenter, NO16853 randomized phase II study in patients with locally advanced or metastatic breast cancer (LA/MBC) was initiated to investigate whether a lower-than-approved dose of capecitabine in the XT combination would result in improved tolerability compared with standard-dose XT, without compromising efficacy. Here, we present final efficacy and safety outcomes from the study.
patients and methods patients The study (http://ClinicalTrials.gov identifier: NCT00077857; protocol NO16853) was conducted in accordance with the principles outlined in the Guideline for Good Clinical Practice, International Conference on Harmonisation Tripartite Guideline, or with local law if it afforded greater protection to the patient. The protocol and subsequent amendments were approved by the institutional review board or ethics committee of each investigational site. Written informed consent was obtained from all patients prior to any study-related procedure. Women aged ‡18 years with histologically or cytologically confirmed LA/ MBC resistant to an anthracycline-based regimen in the (neo)adjuvant, first- or second-line metastatic setting were eligible for enrollment. Primary anthracycline resistance was defined as patients progressing on anthracycline-based chemotherapy without experiencing any transient improvement. Non-primary resistance was defined as patients whose disease remained stable after administration of four or less cycles of anthracycline-based chemotherapy, patients experiencing a brief objective response to anthracycline-based chemotherapy with subsequent progression on the same therapy or within 12 months after the last dose or patients relapsing within 2 years after completion of an anthracycline-based neoadjuvant or adjuvant chemotherapy regimen. Patients were required to have a Karnofsky performance status ‡70% and at least one measurable lesion according to RECIST. Target lesions measured solely by positron emission tomography scan were acceptable if ‡2 cm in longest diameter.
590 | Buzdar et al.
Annals of Oncology
Chemotherapy was not permitted within 4 weeks before study entry. Patients receiving more than two chemotherapy regimens in the metastatic setting were excluded, as were those receiving prior continuous (>24 h) 5FU infusion, capecitabine, other oral fluoropyrimidines, or docetaxelcontaining therapy in the advanced/metastatic setting. Prior paclitaxel was permitted. Prior docetaxel was permitted in the (neo)adjuvant setting if a minimum of 12 months had passed from the time of therapy completion to relapse.
study design and treatment Study NO16853 was an open-label, multicenter, multinational, randomized phase II non-inferiority study of two doses of capecitabine in combination with docetaxel for LA/MBC. Initially, the study was to be conducted solely in the United States, but because of slow patient recruitment the protocol was amended to include non-USA sites to fulfill the post-marketing commitment in a timely manner. The primary objective was to demonstrate non-inferiority of low-dose XT to standard-dose XT in terms of progression-free survival (PFS), which was defined as the time from randomization to documented disease progression or death. Secondary objectives were to evaluate safety [according to National Cancer Institute—Common Terminology Criteria for Adverse Events (NCI–CTCAE) v3.0], overall response rate (ORR; according to RECIST v1.0), time to response and duration of overall response, time to treatment failure (TTF) and OS in the two study arms. Predefined exploratory subgroup analyses were conducted for PFS, ORR and OS for levels of prognostic and stratification factors expected to be associated with these end points. Additionally, subgroups were analyzed separately for PFS for USA and non-USA patients. The initial dose of capecitabine selected for the low-dose XT arm was 950 mg/m2 twice daily, which corresponds to an 25% reduction from the registered dose of 1250 mg/m2 twice daily, and is close to the median delivered dose (960 mg/m2 twice daily) [12] for capecitabine in combination with docetaxel in the registration study SO14999. However, study NO17629, which utilized the same regimen in the adjuvant setting, reported a large proportion of patients discontinuing treatment early due to adverse events and was amended to use a lower dose of capecitabine (825 mg/m2 twice daily). For this reason, study NO16853 was also amended such that the capecitabine dose in the low-dose XT arm was 825 mg/m2 twice daily. Patients were randomly assigned to one of two treatment arms in a 1 : 1 ratio and stratified according to geographic region (USA or non-USA), age (3 weeks.
Volume 23 | No. 3 | March 2012
Annals of Oncology
original articles
statistical analysis Three study populations were defined: the intent-to-treat (ITT) population comprised all randomized patients; the per-protocol (PP) population excluded randomly assigned patients who did not receive at least one dose of study drug and those with major protocol violations; the safety population comprised all randomly assigned patients who received at least one dose of study drug. Rather than being analyzed in the treatment group to which they were originally assigned, patients in the safety population were analyzed according to their initial dose of capecitabine, either >1037.5 mg/m2 twice daily (standard-dose group) or £1037.5 mg/m2 twice daily (low-dose group). The cut-off of 1037.5 mg/m2 was chosen as it is half way between 825 and 1250 mg/m2. For all comparisons of non-inferiority, the PP population formed the primary analysis population; key analyses were repeated for the ITT population. The primary analysis of PFS (and OS) was conducted using data from a clinical cut-off date of 31 October 2008, after 350 progression events or deaths had occurred in the PP population. At the time of the primary analysis, some patients were still in the primary treatment phase of the study so safety data were incomplete. A follow-up analysis was therefore carried out when all patients had completed the primary treatment phase (using data from a cut-off date of 30 October 2009), which included 28 days following last dose of study drug in the primary treatment phase. The observed HR for PFS (also termed TTP) with standard-dose XT compared with docetaxel alone in study SO14999 was 0.652 (all randomly assigned patients). Testing for retention of ‡25% of the contribution of capecitabine to the XT combination translated into a non-inferiority margin of 1.35 for the HR (low-dose XT versus standard-dose XT) and required 350 events in total across both treatment arms to achieve 80% power with a one-sided type I error of 0.025. Assuming median PFS to be 6.1 months for standard-dose XT, as in study SO14999, recruitment of 372 patients was expected to lead to the required number of 350 events based on recruitment and follow-up times of 18 months each. Estimating that 15% of all patients randomly assigned would be excluded from the PP population, the target accrual was 440 patients.
Figure 1. CONSORT diagram of patient flow through the study. ITT, intent-to-treat; PP, per-protocol; X, capecitabine; T, docetaxel; CNS, central nervous system.
patient disposition Between 2003 and 2008, 470 patients were randomly assigned at 63 centers across the United States, Russia, China, India, South Africa, Bosnia-Herzegovina, Thailand and Poland in a 1 : 1 ratio to standard-dose XT (n = 235) or low-dose XT (n = 235). Five patients were excluded from the PP population in the standard-dose group and six from the low-dose group (Figure 1). Of the 235 patients initially randomly assigned to the low-dose XT arm, 14 received a capecitabine starting dose of 950 mg/m2 before the protocol amendment to 825 mg/m2. Among 231 patients randomly assigned to standard-dose XT who received at least one dose of capecitabine or docetaxel, 14 patients started treatment on a capecitabine dose £1037.5 mg/m2 twice daily due to decreased creatinine clearance and were therefore included in the low-dose group in the safety population. Following treatment with XT, docetaxel was discontinued and capecitabine monotherapy continued in 61 patients in the standard-dose XT group (for a median of 3 cycles) and in 50 patients in the lowdose XT group (for a median of 3.5 cycles).
Median patient age was 51 years (range 22–75 years). The majority of patients had node-positive, stage T1–2, ductal breast cancer. Approximately 57% of patients received XT in the first-line setting. Baseline disease characteristics were also generally well balanced within the subgroup of non-USA patients (n = 433; Table 2). However, differences in baseline demographic data and breast cancer disease characteristics were noted within the USA subgroup (n = 37). Specifically, in the standard-dose group versus the low-dose group, respectively, more patients were white (89% versus 44%), with node-negative disease (42% versus 17%), and a longer median time from diagnosis to randomization (1080 versus 682 days) or from last anthracycline dose to randomization (553 versus 422 days). Furthermore, fewer patients in the standard-dose group versus the low-dose group, respectively, had T3/4 stage tumors (16% versus 24%), prior taxane exposure (47% versus 61%), a disease-free interval (DFI) £12 months if receiving XT first line (17% versus 46%) and estrogen/progesterone receptornegative (47% versus 61%) or triple-negative disease (37% versus 50%). Thus, it appears that the standard-dose XT group included patients with baseline characteristics and breast cancer characteristics that impart a better prognosis compared with those in the low-dose XT group. Comparison of the USA subgroup with the non-USA subgroup also revealed differences in baseline disease characteristics. Greater proportions of patients in the USA subgroup than in the non-USA subgroup received prior taxanes, received XT as first-line therapy and had triplenegative breast cancer (Table 2).
demographic data The treatment groups were generally well balanced with regard to baseline demographic and disease characteristics (Table 1).
efficacy Median observation time at database closure (September 2009) was 16.4 months at which time 419 PFS events had been
results
Volume 23 | No. 3 | March 2012
doi:10.1093/annonc/mdr256 | 591
original articles
Annals of Oncology
Table 1. Baseline demographic and disease characteristics: intent-totreat population Standard-dose XT (n = 235) Median age, years (range) Race, n (%) Asian or Pacific Islander Black White Breast cancer subtype, n (%) Ductal Lobular Inflammatory Other Unknown Nodal status, n (%) N0 N+ Unknown Tumor stage, n (%) T1–2 T3–4 TX Prior taxane use, n (%) Chemotherapy line, n (%) First With DFI £12 months With DFI >12 months Second/third ER/PgR status, n (%) Positive Negative Unknown HER2 status, n (%) Positive Negative Unknown Triple negative, n (%)
51 (22–75)
Low-dose XT (n = 235) 51 (28–75)
104 (44) 9 (4) 121 (51)
110 (47) 12 (5) 108 (46)
168 20 4 29 13
190 17 1 20 12
(71) (9) (2) (12) (6)
(81) (7) (12 months Second/third Median time from diagnosis to randomization, days (range) Median time from last anthracycline dose to randomization, days (range) ER/PgR status, n (%) Positive Negative Unknown HER2 status, n (%) Positive Negative Unknown Triple negative, n (%)
56 (32–71) – 2 (11) 17 (89) –
Low-dose XT (n = 18) 51 (32–72) 1 6 8 3
(6) (33) (44) (17)
Non-USA patients Standard-dose XT (n = 216) 51 (22–75)
Low-dose XT (n = 217) 51 (28–75)
104 (48) 7 (3) 104 (48) –
109 (50) 6 (3) 100 (46) – 172 16 1 17 12
16 (84) 4 (21) 1 (5) – –
18 (100) 1 (6) – 3 (17) –
152 16 3 29 13
8 (42) 11 (58) –
3 (17) 14 (78) 1 (6)
41 (19) 132 (61) 43 (20)
47 (22) 127 (59) 43 (20)
16 (84) 3 (16) – 9 (47)
13 (76) 4 (24) – 11 (61)
111 92 13 45
127 73 17 40
12 2/12 10/12 7 1080
(63) (17) (83) (37) (180–5053)
553 (30–2174)
9 (47) 9 (47) 1 (5) 2 16 1 7
(11) (84) (5) (37)
13 6/13 7/13 5 682
(72) (46) (54) (28) (327–2514)
422 (117–1156)
7 (39) 11 (61) – 3 (17) 15 (83) – 9 (50)
126 77/125 48/125 90 617
(70) (7) (1) (13) (6)
(51) (43) (6) (21) (58) (62) (38) (42) (77–9248)
245 (25–1784)
79 (37) 92 (43) 45 (21) 30 83 103 42
(14) (38) (48) (19)
117 67/117 50/117 100 675
(79) (7) (
