Platinum Priority – Editorial
Referring to the article published on pp. 1028–1038 of this issue

Observed Advantages of the STAMPEDE Study Design

By: Howard I. Scher lowast

European Urology, Volume 67 Issue 6, June 2015, Pages 1039-1041

Published online: 01 June 2015

Abstract Full Text Full Text PDF (0,9 MB)

Refers to article:

Survival with Newly Diagnosed Metastatic Prostate Cancer in the “Docetaxel Era”: Data from 917 Patients in the Control Arm of the STAMPEDE Trial (MRC PR08, CRUK/06/019)

Nicholas David James, Melissa R. Spears, Noel W. Clarke, David P. Dearnaley, Johann S. De Bono, Joanna Gale, John Hetherington, Peter J. Hoskin, Robert J. Jones, Robert Laing, Jason F. Lester, Duncan McLaren, Christopher C. Parker, Mahesh K.B. Parmar, Alastair W.S. Ritchie, J. Martin Russell, Räto T. Strebel, George N. Thalmann, Malcolm D. Mason and Matthew R. Sydes

Accepted 19 September 2014

June 2015 (Vol. 67, Issue 6, pages 1028 - 1038)

Recent advances in understanding of prostate cancer biology, in diagnostics, and in therapeutics have transformed the management of castration-resistant prostate cancer (CRPC): In a 4-yr period, five new systemic therapies have been proven to prolong life and approved for use. Issues now are (1) how to use these agents in practice to maximize patient benefit, (2) how to evaluate them in earlier disease sites for which they may be more beneficial, (3) how to show whether combining them and other approved agents with androgen deprivation therapy (ADT; which has remained the first-line standard systemic therapy for almost eight decades) is superior to using them sequentially, and (4) how to refine methods to prioritize the most promising approaches for further study while continuing to develop new approaches.

Many of these issues are addressed by the Systemic Therapy in Advancing or Metastatic Prostate Cancer: Evaluation of Drug Efficacy (STAMPEDE) trial, a paradigm-shifting effort that uses a multiarm, multistage design to build on the outcomes with ADT alone by exploring combinations of approved agents in the noncastrate clinical states ranging from high-risk localized disease to rising prostate-specific antigen (PSA) and metastatic disease [1] . The design ensures that, at minimum, patients receive the current standard of care (and they may have better outcomes by being treated and followed in a formal clinical trial setting). Since the evaluation of various combined androgen blockade approaches in the 1980s and 1990s, relatively few definitive trials have been conducted in the nonlocalized, noncastrate state because of the need for long follow-up and large sample sizes to show an effect on overall survival (OS), which is presently the main end point for regulatory approvals. The statistical design includes strict lack-of-sufficient-activity guidelines for pairwise comparisons between each experimental arm and the control arm so that only research arms that pass the bar can proceed and the others are discontinued [1] . Such was the case with the combination of ADT and celecoxib, which was insufficiently active: Accrual to this arm of the trial was halted [2] .

The report by James et al published in this issue ofEuropean Urologyhighlights another, perhaps underappreciated advantage of the STAMPEDE design [3] : the importance of a continuous-accrual control arm. The authors present outcomes of the 917 men (17% of the 5272 men enrolled) with newly diagnosed node-positive or metastatic noncastrate disease deemed to be fit for chemotherapy and randomized to the ADT control arm. The patients were treated with continuous ADT, consisting of either an orchiectomy or luteinizing hormone-releasing hormone agonist or antagonist with or without an antiandrogen. Treatment after progression was left to the discretion of the treating physician.

The primary outcomes were OS, defined as the time from randomization to death from any cause, and failure-free survival (FFS), defined as any one of the following: biochemical (PSA) progression (50% above the nadir for those with a ≥50% decline); clinical or radiographic progression locally, in nodes, or distant metastases; or death from disease. The results are reported by subgroups based on patient and disease factors at the time of diagnosis, and the subgroups are defined based on the pattern of disease spread at the start of treatment (similar to those defined in CRPC by the Prostate Cancer Clinical Trials Working Group [4] ) and post-treatment PSA nadir. An important note is that reporting the outcomes of the control group does not compromise the trial's primary objective, which is comparing outcomes of individual experimental arms to the control. The results of the current paper include many important and significant observations that are of immediate relevance to current and future trial design and interpretation and that identify or highlight unmet needs to enable clinical progress to continue. Four such observations are listed below.

First, noncastrate metastatic disease presents as discrete subtypes with different prognoses. As anticipated, bone was the most common site of spread, but 12% of the patients had soft tissue–only disease, predominantly in lymph nodes; these patients had a more favorable prognosis than those with bone or visceral spread. In the patients with bone and nodal spread, only nodes above the pelvic brim adversely affected prognosis. An unmet need is to understand the molecular basis for both site specificity of spread and differences in outcome.

Second, nomograms designed to estimate survival times must be updated continuously but are still no substitute for a contemporary control group to inform the decision to advance a treatment regimen for definitive testing. The eligibility criteria were similar for the trials cited by Tangen et al [5] and the current study [3] , but considered individually, the proportions of patients with unfavorable factors such as extensive disease; bone pain; an Eastern Cooperative Oncology Group performance status (PS) of 2–3[6], [7], [8], and [9]; and, as available, baseline PSA and Gleason score 8–10[7], [8], and [9]covered a wide range. Median OS times in the trials cited by Tangen et al [5] ranged from 32 mo in the oldest trial and 49 mo in a more recent one; taken at face value, one could conclude that the latter was superior to the trial by James et al (median OS: 42 mo) when, in fact, both trials used the same treatment. An updated multivariate model for survival that included soft tissue only and bone only, primary tumor, Gleason sum, younger age (<60 yr), PS 1 or 2, and higher PSA was constructed and is useful for patient stratification but does not substitute for a contemporary control group.

Third, a higher percentage of the OS time of a patient with noncastrate disease at diagnosis is spent in the castration-resistant state. The median FFS observed was only 11.2 mo for the whole cohort, and OS was 42 mo, with 29% and 72%, respectively, of the whole cohort being failure-free at 2 yr. Slightly >25% of the median survival time was spent in a castration-resistant state (11 of 42), and the median OS from first FFS event was 22 mo, with 46% of the patients (95% confidence interval, 40–51%) alive at 2 yr. As such, it was somewhat surprising that the median time to the next therapy was 20 mo for the control arm and 15.4 mo for the experimental arms; this seem unduly long, given that CRPC is essentially a death sentence. Somewhat disturbing was that of the 502 FFS events, 252 patients (49%) did not report receiving any life-prolonging therapy; overall, only 198 ever received chemotherapy (median time: 16 mo), and 83 ever received abiraterone. This may be due in part to the recognized reluctance to receive cytotoxic therapy and the unavailability of abiraterone and enzalutamide at some sites. It is likely to change now that the indications for abiraterone and enzalutamide have been expanded to include all metastatic CRPC and are not restricted to the post–chemotherapy-treated space. One would predict that patients are unlikely to wait 15–20 mo to receive potentially effective treatment knowing that their disease is progressing. A key issue to be addressed prospectively is whether it affects outcomes to change therapy to one or more of the recently approved agents when the CRPC state is first documented or to defer treatment with a new agent until a later time point.

Fourth, the availability of more postprotocol therapies makes it more difficult to prove a survival benefit for a promising experimental arm in STAMPEDE. We have already seen that a significantly greater proportion of the survival time of a patient with metastatic noncastrate CRPC is longer than FFS. Already, a patient progressing on the control arm of the trial can receive abiraterone acetate, enzalutamide, sipuleucel-T, and radium-223 in addition to docetaxel. Median survival was 35 mo with abiraterone [10] and 32 mo with enzalutamide [11] . All of these options may make it more difficult to prove a survival benefit for an experimental arm, raising the possibility that an effective treatment might not become available [12] . Surrogate end points for survival are needed that can lead to accelerated drug approvals or that demonstrate sufficient clinical benefit to a patient to justify drug approval in their own right. Generating the evidence to establish surrogacy requires multiple prospective trials of similar design. The intermediate end points considered by James et al [3] are the delay or prevention of failure and death. The definition of FFS is a composite of PSA progression, clinical progression, and radiographic progression. Notably, the median time to PSA and clinical progression was 12 mo and 14 mo, respectively. Post-therapy PSA measures have been explored, and in one series of men treated with radiation therapy, post-therapy PSA nadir was shown to be a surrogate for prostate cancer–specific survival [13] ; however, to date, this end point alone has not led to regulatory approvals. Failure to achieve a nadir ≤ 0.5 ng/ml has also been shown to predict for worse outcome [13] but similarly has not been tested in a regulatory context.

By testing multiple experimental approaches against a single control arm, STAMPEDE enables patient resources to be utilized more efficiently than in traditional studies with two or three arms and independent control groups; in addition, the chance (odds) that a patient will receive a promising new experimental approach is increased. Accrual is also faster, enabling more approaches to be evaluated sooner. With a continuously accruing control arm, the next promising approach can be added as a new arm to the study and tested when initially identified, obviating the need to wait for an ongoing trial to reach its accrual targets; to design a new trial; and to go through what can be the arduous, slow, and costly process of regulatory approval, site initiation, and study launch. The recent addition of a STAMPEDE arm evaluating the combination of abiraterone acetate and enzalutamide is one example [14] . Although most of the focus remains on the comparisons of thenewversus theold, it is important that the value of the continuously accruing control arm be both recognized and appreciated. With the flexibility of this study design, we anticipate more highly pertinent results as the study progresses.

Conflicts of interest

The author has a consultancy or is on an advisory board for Astellas, BIND Pharmaceuticals, Bristol-Myers Squibb, Celgene, Exelixis, Ferring Pharmaceuticals, Foundation Medicine, Genentech, Janssen Pharmaceuticals, Medivation, OncologySTAT, Palmetto GBA, LLC, Pfizer (Biologic), Sanofi Aventis, Takeda Millennium, and WCG Oncology and has received honoraria from the Chugai Academy for Advanced Oncology. He has received research funding through Memorial Sloan Kettering Cancer Center from BIND Pharmaceuticals, Exelixis, Janssen Pharmaceuticals, Medivation, and Janssen Diagnostics.


  • [1] M.R. Sydes, M.K. Parmar, M.D. Mason, et al. Flexible trial design in practice - stopping arms for lack-of-benefit and adding research arms mid-trial in STAMPEDE: a multi-arm multi-stage randomized controlled trial. Trials. 2012;13:168
  • [2] N.D. James, M.R. Sydes, M.D. Mason, et al. Celecoxib plus hormone therapy versus hormone therapy alone for hormone-sensitive prostate cancer: first results from the STAMPEDE multiarm, multistage, randomised controlled trial. Lancet Oncol. 2012;13:549-558
  • [3] N.D. James, M.R. Spears, N.W. Clarke, et al. Survival with newly diagnosed metastatic prostate cancer in the “docetaxel era”: data from 917 patients in the control arm of the STAMPEDE Trial (MRC PR08 CRUK/06/019). Eur Urol. 2015;67:1028-1038
  • [4] H.I. Scher, S. Halabi, I. Tannock, et al. Design and end points of clinical trials for patients with progressive prostate cancer and castrate levels of testosterone: recommendations of the Prostate Cancer Clinical Trials Working Group. J Clin Oncol. 2008;26:1148-1159
  • [5] C.M. Tangen, M.H. Hussain, C.S. Higano, et al. Improved overall survival trends of men with newly diagnosed M1 prostate cancer: a SWOG phase III trial experience (S8494, S8894 and S9346). J Urol. 2012;188:1164-1169
  • [6] E.D. Crawford, M.A. Eisenberger, D.G. McLeod, et al. A controlled trial of leuprolide with and without flutamide in prostatic carcinoma. N Engl J Med. 1989;321:419-424
  • [7] M.A. Eisenberger, B.A. Blumenstein, E.D. Crawford, et al. Bilateral orchiectomy with or without flutamide for metastatic prostate cancer. N Engl J Med. 1998;339:1036-1042
  • [8] M. Hussain, B. Goldman, C. Tangen, et al. Prostate-specific antigen progression predicts overall survival in patients with metastatic prostate cancer: data from Southwest Oncology Group Trials 9346 (Intergroup Study 0162) and 9916. J Clin Oncol. 2009;27:2450-2456
  • [9] M. Hussain, C.M. Tangen, D.L. Berry, et al. Intermittent versus continuous androgen deprivation in prostate cancer. N Engl J Med. 2013;368:1314-1325
  • [10] D.E. Rathkopf, M.R. Smith, J.S. de Bono, et al. Updated interim efficacy analysis and long-term safety of abiraterone acetate in metastatic castration-resistant prostate cancer patients without prior chemotherapy (COU-AA-302). Eur Urol. 2014;66:815-825
  • [11] T.M. Beer, A.J. Armstrong, D.E. Rathkopf, et al. Enzalutamide in metastatic prostate cancer before chemotherapy. N Engl J Med. 2014;371:424-433
  • [12] E.L. Korn, B. Freidlin, J.S. Abrams. Overall survival as the outcome for randomized clinical trials with effective subsequent therapies. J Clin Oncol. 2011;29:2439-2442
  • [13] A.V. D’Amico, M.H. Chen, M. de Castro, et al. Surrogate endpoints for prostate cancer-specific mortality after radiotherapy and androgen suppression therapy in men with localised or locally advanced prostate cancer: an analysis of two randomised trials. Lancet Oncol. 2012;13:189-195
  • [14] G. Attard, M.R. Sydes, M.D. Mason, et al. Combining enzalutamide with abiraterone, prednisone, and androgen deprivation therapy in the STAMPEDE trial. Eur Urol. 2014;66:799-802


Genitourinary Oncology Service, Memorial Sloan-Kettering Cancer Center and Weill Cornell Medical College, New York, NY, USA

lowast Genitourinary Oncology Service, Memorial Sloan Kettering Cancer Center and Weill Cornell Medical College, New York, NY 10065, USA. Tel. +1 646 422 4323: Fax: +1 212 988-0851.

Place a comment

Your comment *

max length: 5000