original article
Annals of Oncology 20: 1264–1269, 2009 doi:10.1093/annonc/mdn784 Published online 17 March 2009
Docetaxel plus oblimersen sodium (Bcl-2 antisense oligonucleotide): an EORTC multicenter, randomized phase II study in patients with castration-resistant prostate cancer C. N. Sternberg1*, H. Dumez2, H. Van Poppel3, I. Skoneczna4, A. Sella5, G. Daugaard6, T. Gil7, J. Graham8, P. Carpentier9, F. Calabro1, L. Collette10 & D. Lacombe10 for the EORTC Genitourinary Tract Cancer Group 1
Department of Medical Oncology, San Camillo and Forlanini Hospitals, Rome, Italy; Departments of 2Oncology and 3Urology, Universitair Ziekenhuis Gasthuisberg, Leuven, Belgium; 4Chemotherapy Unit, Department of Urology, Maria Sklodowska–Curie Memorial Cancer Centre, Warsaw, Poland; 5Department of Oncology, Assaf Harofeh Medical Center, Zerifin, Israel; 6Department of Oncology, Rigshospitalet, Copenhagen, Denmark; 7Department of Chemotherapy, Institut Jules Bordet, Brussels, Belgium; 8Beatson West of Scotland Cancer Centre, Gartnavel General Hospital, Glasgow, UK; 9Department of Urology, Onze Lieve Vrouw Ziekenhuis, Aalst, Belgium and 10EORTC Headquarters, Brussels, Belgium
Received 11 September 2008; revised 3 December 2008; accepted 15 December 2008
original article
Background: This randomized, phase II study assessed the activity of oblimersen sodium, a Bcl-2 antisense oligonucleotide, administered before docetaxel (Taxotere) to patients with castration-resistant prostate cancer. Patients and methods: Chemotherapy-naive patients with prostate-specific antigen (PSA) progression and testosterone £0.5 ng/ml received docetaxel 75 mg/m2 on day 1 or oblimersen 7 mg/kg/day continuous i.v. infusion on days 1–7 with docetaxel 75 mg/m2 on day 5 every 3 weeks for £12 cycles. Primary end points were confirmed PSA response (Bubley criteria) and major toxic events. Results: Confirmed PSA response was observed in 46% and 37% of 57 and 54 patients treated with docetaxel and docetaxel–oblimersen, respectively. Partial response (RECIST) was achieved in 18% and 24%, respectively. Oblimersen added to docetaxel was associated with an increase in the incidence of grade ‡3 fatigue, mucositis, and thrombocytopenia. Major toxic events were reported in 22.8% and 40.7% of patients with docetaxel and docetaxel– oblimersen, respectively. Conclusions: The primary end points of the study were not met: a rate of confirmed PSA response >30% and a major toxic event rate 25% of marrowproducing area (although prior use of estramustine or bisphosphonates was allowed); or hormonal manipulation with PC-SPES £6 weeks before study entry. Also excluded were patients with known contraindication for any of the study drugs. All patients provided written informed consent according to the International Conference on Harmonisation/Good Clinical Practice and national regulations. The protocol was reviewed and approved by the Ethics Committee in each institution and was conducted in accordance with the Declaration of Helsinki.
treatment Patients were centrally randomized at the European Organisation for Research and Treatment of Cancer (EORTC) Headquarters using a minimization technique [12] stratifying for institution, presence of metastatic (M1) measurable disease according to RECIST [13], prior use of estramustine, and prior use of bisphosphonates. Protocol treatment was to be initiated £15 days after randomization. In arm A, docetaxel (Taxotere) 75 mg/m2 was given as an i.v. infusion (plus standard premedication with dexamethasone or equivalent) over 60 min on day 1 of each cycle. In arm B, oblimersen 7 mg/kg/day was given by continuous i.v. infusion on days 1–7 and docetaxel 75 mg/m2 was administered as in arm A on day 5 of each cycle, consistent with the regimen of Tolcher et al. In particular, oblimersen administration preceded docetaxel administration by 5 days to achieve down-regulation of Bcl-2 and was continued for 2 days after docetaxel administration to allow adequate coexposure of the two study agents [10]. Oblimersen was administered through a central venous catheter or peripherally inserted central catheter via an ambulatory infusion pump. Treatment was to continue every 3 weeks for £12 cycles, unless the
Volume 20 | No. 7 | July 2009
following withdrawal criteria were met: PSA progression [11], objective clinical progression (RECIST) [13], excessive toxicity, or patient refusal to continue therapy. The oblimersen dose was not to be adjusted. For patients with treatment-related hematologic or specific non-hematologic toxic effects (i.e. nausea with vomiting, skin changes, or liver function abnormalities), protocol therapy was to be delayed for £2 weeks. Upon recovery, docetaxel dose was to be lowered by 1 dose level at the start of the subsequent cycle. A maximum of 2 dose reductions (from 75 to 60 mg/m2 and from 60 to 45 mg/m2) was allowed. With other non-hematologic toxic effects, protocol therapy was to be delayed, but docetaxel dose was not to be reduced. No cycle was to be initiated unless absolute neutrophil count (ANC) was ‡1500 cells/mm3, platelet count was ‡100 000 cells/mm3, aspartate aminotransferase and alanine aminotransferase returned to £1.5· upper limit of normal (ULN), bilirubin returned to £ULN, skin toxicity resolved to grade £1, and all other toxic effects resolved to grade £2. All non-orchiectomized patients were required to continue therapy with LH–RH agonist. Growth factor support was allowed for recovery of white blood cell count in grade 4 leukopenia or neutropenia or grade 3 or 4 neutropenia with infection. Diuretic treatment was recommended, along with potassium supplementation, as appropriate, to manage docetaxelinduced edema at its early stages. Other ancillary therapy and supportive care were to be administered, as indicated.
evaluations Baseline evaluations included medical history, physical examination, performance status, laboratory tests, electrocardiography, and scans. Hematologic tests were carried out £2 days before treatment start, weekly during therapy, and every 8 weeks until objective progression during follow-up. At the end of each cycle and every 8 weeks during follow-up until objective progression, assessments were to include physical examination, performance status, serum chemistry, adverse events (scored by International Common Toxicity Criteria version 3.0, http:// ctep.info.nih.gov/reporting/ctc.html), and serum PSA (with repeated measures ‡4 weeks after the first observation of response or progression). Every three cycles and every 16 weeks during follow-up until objective progression, chest X-rays were to be carried out in all patients and bone scan and computed tomography or magnetic resonance imaging were to be carried out, as appropriate, among those patients who had measurable disease at baseline or who were presenting with signs/symptoms suggestive of metastases. After progression, patients were to be followed for survival until death or lost to follow-up.
statistical methods and end points The trial was designed according to the one-stage Bryant and Day design [14] with two coprimary end points: confirmed PSA response [11] and the rate of major toxic events. Confirmed PSA response [11] was defined as a 50% decrease of PSA level from baseline that persisted for ‡4 weeks (as determined by two consecutive measurements). A major toxic event was defined as any of (i) grade 4 (Common Toxicity Criteria version 3.0) hematologic and non-hematologic toxicity, except asymptomatic grade 4 neutropenia (ANC < 500/mm3) lasting 5 days or any grade 4 febrile neutropenia; (iii) grade 3 or 4 sepsis; (iv) grade 3 or 4 bleeding associated with grade >2 thrombocytopenia (3· ULN or need for dialysis); (vi) discontinuation of protocol therapy due to unacceptable toxicity; or (vii) death at least possibly due to toxicity. The sample size was determined to ensure 90% power of rejecting the null hypothesis that, within the docetaxel–oblimersen group, either the PSA response rate was £30% or the major toxic event rate was ‡45% at the onesided 10% significance level, under the alternative that the PSA response
doi:10.1093/annonc/mdn784 | 1265
original article
Annals of Oncology
Table 1. Baseline characteristics
Table 2. Protocol and concomitant therapy Docetaxel (N = 57)
Age, years, n (%) ULN Unknown
Docetaxel– oblimersen (N = 54)
19 (33) 11 (19) 15 (26) 12 (21) 69 43–82
29 (54) 12 (22) 6 (11) 7 (13) 64 54–84
25 (44) 1 (2)
29 (54) 0 (0)
34 (60) 22 (39) 1 (2)
32 (59) 20 (37) 2 (4)
54 (95) 28 (52)
50 (93) 21 (42)
56 10–235
175 2–155
3 21 9 24
(5) (37) (16) (42)
4 22 6 22
(7) (41) (11) (41)
13 (23)
14 (26)
18 (32)
13 (24)
2 (4)
6 (11)
86 7–2766
99 10–2469
39 (68) 17 (30) 1 (2)
38 (70) 13 (24) 3 (6)
Number of cycles Median Range Docetaxel treatment duration, days Median Range Oblimersen treatment duration, days Median Range Docetaxel relative dose intensity, % Median Range Oblimersen relative dose intensity, % Median Range Patients with one or more treatment cycle delayed ‡4 days, n (%) Number of cycles with delays ‡4 days per patient Median Range Number of patients with at least one docetaxel schedule modification, n (%) Number of patients with docetaxel dose reduction, n (%), mg/m2 To 60 To 45 Number of patients with oblimersen schedule modificationb, n (%) Number of patients with concomitant therapy, n (%) Antiemetics Growth factors Continued with LH–RH analoguesa
Docetaxel (N = 57)
Docetaxel– oblimersen (N = 54)
8 1213
6 1212
167 20–286
133 202287
N/A
134 202288
98 67–107
N/A
93 68–107
21 (37)
97 12–112 25 (46)
1 1–3 6 (11)
2 1–4 11 (20)
12 (21) 6 (11) N/A
15 (28) 5 (9) 10 (19)
35 (61) 9 (16) 51 (90)
39 (72) 14 (26) 47 (87)
a
Data were based on 28 and 21 patients in respective treatment groups who had measurable lesions at study entry. WHO, World Health Organization; PSA, prostate-specific antigen; LDH, baseline lactate dehydrogenase; ULN, upper limit of normal.
A total of 51 and 47 patients in respective treatment groups were receiving treatment with LH–RH analogues at baseline. b Oblimersen infusion was interrupted £12 h in all 10 patients, mostly because of technical reasons. LH–RH, luteinizing hormone–releasing hormone.
rate was ‡50% and the toxicity rate was £25%. According to this design, the study required that 51 patients in each treatment group be included in the per-protocol population of eligible patients who started the allocated treatment. A treatment arm was declared sufficiently active and safe if the 80% exact confidence interval (CI) around the PSA response rate excluded
30% and that around the major toxic event rate excluded 45%. Exact 95% CIs are also presented. PSA progression was defined according to the 1999 Bubley recommendations [11] Time to progression (TTP) was counted from the day of randomization to the day of PSA progression, objective progression, or death due to progression and is presented in Kaplan–Meier curves [15].
1266 | Sternberg et al.
a
Volume 20 | No. 7 | July 2009
original article
Annals of Oncology
Figure 1. Waterfall plot of maximum prostate-specific antigen (PSA) reduction from baseline.
Figure 2. Time to progression.
results patient characteristics A total of 115 patients (57 in the docetaxel and 58 in the docetaxel–oblimersen group) were enrolled from April 2004 to January 2006 at 15 centers. Four patients in the docetaxel– oblimersen group did not initiate protocol therapy (two were ineligible and two had early progression).
Volume 20 | No. 7 | July 2009
At baseline, >90% of patients in each group had bone and/or visceral metastases (Table 1). Median Gleason score at diagnosis was 8 (range 5210) in both groups. Several differences in baseline characteristics were observed between treatment groups (Table 1). In the docetaxel–oblimersen group versus the docetaxel group, patients were younger (54% versus 33%, respectively,
