original articles
Annals of Oncology
Annals of Oncology 23: 1803–1808, 2012 doi:10.1093/annonc/mdr555 Published online 23 November 2011
Randomized phase II trial of docetaxel plus prednisone in combination with placebo or AT-101, an oral small molecule Bcl-2 family antagonist, as first-line therapy for metastatic castration-resistant prostate cancer G. Sonpavde1*, V. Matveev2, J. M. Burke3, J. R. Caton4, M. T. Fleming5, T. E. Hutson6, M. D. Galsky7, W. R. Berry8, P. Karlov9, J. T. Holmlund10, B. A. Wood10, M. Brookes10 & L. Leopold10 1
Texas Oncology and US Oncology Research, Houston, USA; 2The N. N. Blokhin Cancer Research Center, Russian Academy of Medical Sciences, Moscow, Russia; Rocky Mountain Cancer Centers, Denver; 4Willamette Valley Cancer Center, Eugene; 5Virginia Oncology Associates, Norfolk; 6Texas Oncology, Dallas; 7Comprehensive Cancer Centers of Nevada, Los Vegas; 8Cancer Centers of North Carolina, Raleigh, USA; 9City Clinical Oncology Center, St. Petersburg, Russia; 10Ascenta Therapeutics, Malvern, USA 3
Received 18 May 2011; revised 12 October 2011; accepted 17 October 2011
Background: AT-101 (A), a small molecule oral inhibitor of the Bcl-2 family, has activity alone and in combination with docetaxel (Taxotere) and prednisone (DP) in metastatic castration-resistant prostate cancer (mCRPC). A randomized, double-blind, placebo-controlled phase II trial compared DP combined with either AT-101 (A) or placebo in chemonaive mCRPC. Patients and methods: Men with progressive mCRPC despite androgen deprivation were eligible and randomized 1 : 1. Patients received docetaxel (75 mg/m2 day 1) and prednisone 5 mg orally twice daily every 21 days with either AT-101 (40 mg) or placebo twice daily orally on days 1–3. The primary end point was overall survival (OS). Results: Two hundred and twenty-one patients were randomly assigned. Median OS for AT-101 plus docetaxel–prednisone (ADP) and placebo–DP was 18.1 versus 17.8 months [hazard ratio (HR) 1.07, 95% confidence interval 0.72–1.55, P = 0.63]. Secondary end points were also not statistically different. Grade 3/4 toxic effects for ADP versus placebo–DP were cardiac events (5% versus 2%), lymphopenia (23% versus 16%), neutropenia (47% versus 40%), ileus (2% versus 0%) and pulmonary embolism (6% versus 2%). In a subgroup of high-risk mCRPC (n = 34), outcomes appeared to favor ADP (median OS 19 versus 14 months). Conclusions: AT-101 was tolerable but did not extend OS when combined with DP in mCRPC; a potential benefit was observed in high-risk patients. Key words: AT-101, chemonaive, docetaxel, first-line therapy, metastatic castration-resistant prostate cancer, small molecule Bcl-2 inhibitor
introduction Docetaxel every 3 weeks plus prednisone (DP) is established as frontline chemotherapy for the management of metastatic castration-resistant prostate cancer (mCRPC) [1, 2]. However, median overall survival (OS) was extended modestly to ∼19 months and 18.6% attained 3-year survival [1]. Docetaxel is being employed as the chemotherapeutic platform to develop combination regimens with novel biologic agents in multiple ongoing trials. Dysregulation of apoptosis in general and up-regulation of the prosurvival Bcl-2 family appear important in the *Correspondence to: Dr G. Sonpavde, Texas Oncology and US Oncology Research, 501 Medical Center Blvd, Webster, TX 77598, USA. Tel: +1-281-332-7505; Fax: +1281-332-8429; E-mail:
[email protected]
pathogenesis of castration-resistant prostate cancer (CRPC). Amplification of prosurvival molecules, Bcl-2, Bcl-XL, Bcl-W and Mcl-1, appears to be associated with androgen independence and inhibiting their expression may sensitize to chemotherapy [3–6]. BH3 mimetics have been shown to inhibit the heterodimerization of Bcl-2 or Bcl-XL to proapoptotic BH3 family members (Bax, Bak) with attendant caspase-dependent apoptosis. AT-101 (R-(−)-gossypol acetic acid), the levorotatory enantiomer of gossypol, is a BH3 mimetic with submicromolar affinity (AT-101 Brochure, Ascenta Therapeutics, Malvern, PA). More recent studies have indicated that AT-101 induces proapoptotic Noxa and Puma independent of p53 as well as inhibits angiogenesis and has at least additive activity with chemotherapy including docetaxel against prostate cancer [7–11].
© 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 Phase II clinical trials show preliminary evidence for the activity of AT-101 administered once daily on a 21 of 28 day schedule, with 2 of 23 patients with mCRPC exhibiting a confirmed ≥50% prostate-specific antigen (PSA) decline [12]. Due to grade 3 nonmechanical small intestinal obstruction (n = 5, 21.7%), a dose reduction to 20 mg/day was required and intermittent dosing of higher doses on 3 of every 21 days appeared feasible and mitigated gastrointestinal toxic effects [13]. Pharmacokinetic studies suggested that doses of 20–30 mg were able to achieve peak serum levels of 0.5–1.0 μmol/l with a half-life of 3–5 h, and these levels were adequate to inhibit Bcl-2 family proteins, induce cytochrome c release and trigger apoptosis. A phase I/II trial combining docetaxel day 1 every 21 days with AT-101 40 mg twice daily on days 1–3 in chemonaive men with mCRPC demonstrated feasibility and activity [14]. Encouragingly, 24 of 36 men (67%) achieved a PSA decline ≥50%, and 9 of 19 (45%) with measurable disease displayed a partial response. Toxic effects were manageable and no intestinal obstruction was observed. Moreover, the addition of AT-101 to docetaxel in docetaxel-resistant patients demonstrated PSA declines ≥50% in 6 of 34 (18%) patients and measurable disease responses in 5 of 21 (24%) patients [15]. One partial small bowel obstruction was the only possibly related serious adverse event (AE). Therefore, a randomized phase II trial was conducted to evaluate the combination of AT-101 with conventional frontline DP for mCRPC.
patients and methods study population Men with metastatic adenocarcinoma of the prostate with progression of disease despite androgen deprivation and antiandrogen withdrawal were eligible. In accordance with the Prostate Cancer Clinical Trials Working Group (PCWG)-2 recommendations, a washout of antiandrogen therapy was not required for patients who did not respond for >3 months after an antiandrogen was administered [16]. Patients were required to have progressive disease by RECIST, bone scan (defined as the appearance of ≥2 new lesions) or rising PSA (defined by increasing levels on two consecutive assessments ≥1 week apart) with the last PSA ≥2 ng/ml. Serum testosterone level ≤50 ng/dl and Eastern Cooperative Oncology Group (ECOG) performance status (PS) more than or equal to two were required. Exclusion criteria included prior chemotherapy (including estramustine), central nervous system metastases, prior radiation therapy to ≥30% of the bone marrow, peripheral neuropathy of ≥grade 2, gastrointestinal malabsorption, inflammatory bowel disease, small bowel obstruction and class 3–4 cardiac disease. The trial was approved by institutional review boards and conducted at 41 community and academic cancer centers in Russia and the United States. The trial was registered at clinicaltrials.gov (NCT00286793). Written informed consent was required for participation.
treatment plan and assessments Patients received docetaxel (Taxotere) 75 mg/m2 i.v. over 1 h on day 1 plus oral prednisone 5 mg twice daily with either AT-
| Sonpavde et al.
Annals of Oncology
101 40 mg [AT-101 plus docetaxel–prednisone (ADP) arm] or placebo ( placebo–DP arm) orally twice daily on days 1–3 every 21 days (one cycle). AT-101 was supplied as a 10-mg immediate release tablet and was taken at the same time each day on an empty stomach (at least 1 h before or after a meal). Dexamethasone 8 mg orally was administered 12, 3, and 1 h before docetaxel, and antiemetics were allowed. A maximum of 17 cycles of therapy (1 year) were permitted unless unacceptable toxicity or disease progression occurred. Safety assessments including laboratory evaluations were carried out weekly during the first cycle and then once every cycle (blood cell counts, comprehensive metabolic panel, PSA and troponin-I), as well as at the end of treatment. An electrocardiogram was carried out on day 4 of every cycle. Radiologic and response assessment with computed tomography scan or magnetic resonance imaging of chest/ abdomen were obtained every three cycles or at symptomatic progression. Pain was assessed by the Present Pain Index (PPI) on a scale of 0–5 and quality of life was assessed by the Functional Assessment of Cancer Therapy-Prostate, version 4, once every cycle. Circulating tumor cells (CTCs) were measured at baseline (only in patients enrolled in the United States), with a plan to potentially profile CTCs for Bcl-2 family proteins. Disease progression was defined in accordance with the PCWG-2 criteria as the occurrence of one or more of the following: progression of soft tissue disease per RECIST, bone scan progression defined as the appearance of more than or equal to two new lesions attributable to prostate cancer, skeletal event (defined as fracture, bone pain resulting in the need for radiation therapy or surgery) or symptomatic progression defined as worsening of ECOG PS and/or increased pain [16]. Increase in pain was defined as an appearance of new pain or increase in PPI score ≥2 on two consecutive assessments at least 3 weeks apart. PSA progression alone did not constitute progression. AEs were captured using the National Cancer Institute Common Terminology Criteria for Adverse Events, version 3.0. Dose delays and modifications of docetaxel and AT-101/ placebo were made, as needed, due to treatment-related toxicity. Treatment with AT-101/placebo and docetaxel was delayed up to 2 weeks to allow for resolution of toxicity. Dose levels −1 and −2 for AT-101 were 30 and 20 mg twice daily on cycle days 1–3, respectively. The dose of docetaxel was reduced by 15 mg/m2 for each of two levels of dose reductions allowed. Unacceptable toxicity was defined as: toxicity requiring more than two dose reductions or a delay of therapy >2 weeks, a severe hypersensitivity reaction (i.e. hypotension, bronchospasm and/or generalized rash/erythema) or ≥grade 3 peripheral neuropathy.
CTC methodology From each patient enrolled in the United States, ∼20 ml (range 5–20 ml) of anticoagulated EDTA peripheral blood was collected at cycle 1 day 1 and cycle 1 day 4; the schedule was later amended to cycle 1 day 1 only. Isolated mononuclear cells were incubated with a magnetically labeled antibody (EpCAM; Miltenyi Biotec, Auburn, CA) for 30 min at 2–8°C. Slides were
Volume 23 | No. 7 | July 2012
Annals of Oncology
original articles
stained for cytokeratin (clones AE1/AE3; Biocare Medical, Concord, CA) within 2 weeks of preparation. CTCs were visualized using an alkaline phosphatase-/red chromogenbased detection kit (Biocare Medical).
statistical considerations Originally, the primary end point of the study was progression-free survival (PFS) with a planned enrollment of 106 patients. The number of events required was computed using the following specifications: proportional hazards, onesided type I error probability of 0.10, power of 80% and a hazard ratio (HR) of 0.556. This corresponds to an ∼80% increase in median PFS in the investigational arm. However, during the conduct of the study, the primary end point was amended to OS employing one-sided type I error of 0.10, power of 80% and an HR of 0.6667, leading to a requirement for 110 deaths and target accrual of 250. This corresponds to an ∼50% increase in the median OS in the investigational arm. The secondary end points were PFS, and response rates for PSA, measurable disease and pain and toxic effects. PCWG-2 criteria were employed for the evaluation of all end points [16]. Exploratory end points included CTC numbers and profiling for Bcl-2. The sponsor, investigators and patients remained blinded to treatment assignments during the conduct of the study. Patients were enrolled in a double-blind, placebo-controlled, two-arm trial with 1 : 1 randomization with stratification by ECOG PS (0–1 versus 2) and pain (absent versus present). The primary efficacy analysis was a stratified log-rank test of OS from time of randomization until death.
results patient characteristics A total of 221 of the planned 250 men were accrued from November 2007 to April 2009, of whom 110 were assessable in each arm (Figure 1). One patient did not receive any therapy after enrollment and was excluded from analysis. Patient demographic characteristics were well balanced between the arms (Table 1). Accrual ceased at 221 patients after an interim analysis revealed that additional enrollment would not meaningfully contribute to the requirement of 110 events for the primary OS analysis. A similar number of patients were enrolled in the United States (n = 116) and Russia (n = 105). Patients enrolled were typical for this population with a median PSA of 81–91.4 ng/ml and bone metastases present in 85%–90% of patients. Visceral metastases were present in 20%–25% and pain in 57%–61% of patients. The vast majority (93%–96%) had an ECOG PS of zero to one.
efficacy The primary end point, survival, was not statistically different for ADP compared with placebo–DP in the intention to treat population [median OS 18.1 versus 17.8 months, HR 1.07, 95% confidence interval (CI) 0.72–1.55, P = 0.63] (Figure 2). The median PFS was also similar in both arms (11.0 versus 10.3 months, HR 0.88, 95% CI 0.63–1.22, P = 0.53) (supplemental Figure S1, available at Annals of Oncology online). Confirmed
Volume 23 | No. 7 | July 2012
Figure 1. Consort diagram. DP, docetaxel–prednisone. Table 1. Demographic data and baseline disease characteristics ADP, N = 110 Age (years) Median (min–max) 69.0 (46–86) Weight (kg) Median (min–max) 88.8 (60–122) Ethnic origin, n (%) African American 5 (4.5) White 102 (92.7) ECOG status, n (%) 0 42 (38.2) 1 64 (58.2) 2 4 (3.6) Baseline PSA Median (min–max) ng/ml 90.38 (0.6–3631) Median velocity (min–max) 2.49 (1.08–4.39) log ng/ml/year Histopath type and grade at diagnosis, n (%) Adenocarcinoma 108 (98.2) G1: well differentiated 0 G2: moderately differentiated 7 (6.4) G3–4: poorly differentiated 47 (42.7) Unknown 53 (48.2) Disease characteristics, n (%) Bone only metastases 31 (28.2) RECIST measurable disease 56 (50.9) Sites of disease, n (%) Lymph nodes 49 (44.5) Bone 94 (85.5) Visceral (lung, liver, etc) 27 (24.5) Number of prior regimens, n (%) 1 16 (14.5) 2 39 (35.5) >2 54 (49.1) Prior ketoconazole 19 (17.3) Prior antiandrogen Flutamide 10 (9.1) Bicalutamide 90 (81.8) Nilutamide 2 (1.8)
Placebo + DP, N = 110 69.5 (51–88) 88.4 (47–132) 4 (3.6) 103 (93.6) 37 (33.6) 65 (59.1) 8 (7.3) 83.25 (0–13056) 2.53 (0.90–4.32)
106 (96.4) 1 (0.9) 16 (14.5) 35 (31.8) 58 (52.7) 36 (32.7) 50 (45.5) 56 (50.9) 99 (90.0) 21 (19.1) 23 (20.9) 29 (26.4) 56 (50.9) 21 (19.1) 21 (19.1) 85 (77.3) 0 (0)
ADP, AT-101 plus docetaxel–prednisone; DP, docetaxel–prednisone; ECOG, Eastern Cooperative Oncology Group; PSA, prostate-specific antigen.
doi:10.1093/annonc/mdr555 |
original articles
Annals of Oncology
PSA reductions of ≥30% were seen in 66% of ADP-treated patients compared with 54% of controls (P = 0.099) and confirmed PSA reductions of ≥50% were seen in 54% of ADP patients versus 46% in control patients (P = 0.28). Measurable disease control rates (93% versus 80%), objective response rates (26% versus 27%) and pain response rates (29% versus 28%) in the respective assessable populations for ADP and placebo–DP were similar. In a subgroup of patients with high-risk mCRPC (n = 34) as defined by a previous report (more than or equal to three of the following factors: pain, anemia, visceral metastasis, bone scan progression), efficacy end points numerically favored ADP for median OS (19 versus14 months) as well as other secondary end points (Figure 2) [17].
toxic effects Grade 3/4 AEs that occurred with higher incidence in the ADP arm of the trial compared with placebo included cardiac AEs (5% versus 2%), lymphopenia (23% versus 16%), neutropenia (47% versus 40%), ileus (2% versus 0%) and pulmonary embolism (6% versus 2%). The incidence of grade 1/2 peripheral neuropathy was 24% versus 13%. However, neutropenic fevers were uncommon and occurred in 2.7% of patients in both arms. Grade 3/4 AEs with an incidence of 2% or greater, by treatment arm, are summarized in supplemental Table S1 (available at Annals of Oncology online). The median number of cycles delivered was eight in the ADP arm and nine in the placebo–DP arm. A higher proportion of patients discontinued therapy (27.3% versus 16.4%) or underwent dose reductions (21.8% versus 13.6%) due to AEs in the ADP arm.
Figure 2. Overall survival. A0DP, AT-101 plus docetaxel–prednisone.
CTC analysis CTCs were evaluable in 37 men enrolled in the United States, of whom 12 men had CTCs
