The new england
`journal of medicine
`
`established in 1812
`
`september 27, 2012
`
`vol. 367 no. 13
`
`Increased Survival with Enzalutamide in Prostate Cancer
`after Chemotherapy
`Howard I. Scher, M.D., Karim Fizazi, M.D., Ph.D., Fred Saad, M.D., Mary-Ellen Taplin, M.D., Cora N. Sternberg, M.D.,
`Kurt Miller, M.D., Ronald de Wit, M.D., Peter Mulders, M.D., Ph.D., Kim N. Chi, M.D., Neal D. Shore, M.D.,
`Andrew J. Armstrong, M.D., Thomas W. Flaig, M.D., Aude Fléchon, M.D., Ph.D., Paul Mainwaring, M.D.,
`Mark Fleming, M.D., John D. Hainsworth, M.D., Mohammad Hirmand, M.D., Bryan Selby, M.S., Lynn Seely, M.D.,
`and Johann S. de Bono, M.B., Ch.B., Ph.D., for the AFFIRM Investigators*
`
`The authors’ affiliations are listed in the
`Appendix. Address reprint requests to
`Dr. Scher at Memorial Sloan-Kettering
`Cancer Center, 1275 York Ave., New York,
`NY 10065, or at scherh@mskcc.org.
`
`* The AFFIRM (A Study Evaluating the
`Efficacy and Safety of the Investigational
`Drug MDV3100) investigators are listed
`in the Supplementary Appendix, avail-
`able at NEJM.org.
`
`This article was published on August 15,
`2012, and last updated on September 13,
`2012, at NEJM.org.
`
`N Engl J Med 2012;367:1187-97.
`DOI: 10.1056/NEJMoa1207506
`Copyright © 2012 Massachusetts Medical Society.
`
`Abs tr act
`
`Background
`Enzalutamide (formerly called MDV3100) targets multiple steps in the androgen-
`receptor–signaling pathway, the major driver of prostate-cancer growth. We aimed
`to evaluate whether enzalutamide prolongs survival in men with castration-resis-
`tant prostate cancer after chemotherapy.
`
`Methods
`In our phase 3, double-blind, placebo-controlled trial, we stratified 1199 men with
`castration-resistant prostate cancer after chemotherapy according to the Eastern
`Cooperative Oncology Group performance-status score and pain intensity. We ran-
`domly assigned them, in a 2:1 ratio, to receive oral enzalutamide at a dose of 160 mg
`per day (800 patients) or placebo (399 patients). The primary end point was overall
`survival.
`
`Results
`The study was stopped after a planned interim analysis at the time of 520 deaths.
`The median overall survival was 18.4 months (95% confidence interval [CI], 17.3 to
`not yet reached) in the enzalutamide group versus 13.6 months (95% CI, 11.3 to
`15.8) in the placebo group (hazard ratio for death in the enzalutamide group, 0.63;
`95% CI, 0.53 to 0.75; P<0.001). The superiority of enzalutamide over placebo was
`shown with respect to all secondary end points: the proportion of patients with a
`reduction in the prostate-specific antigen (PSA) level by 50% or more (54% vs. 2%,
`P<0.001), the soft-tissue response rate (29% vs. 4%, P<0.001), the quality-of-life re-
`sponse rate (43% vs. 18%, P<0.001), the time to PSA progression (8.3 vs. 3.0 months;
`hazard ratio, 0.25; P<0.001), radiographic progression-free survival (8.3 vs. 2.9
`months; hazard ratio, 0.40; P<0.001), and the time to the first skeletal-related event
`(16.7 vs. 13.3 months; hazard ratio, 0.69; P<0.001). Rates of fatigue, diarrhea, and
`hot flashes were higher in the enzalutamide group. Seizures were reported in five
`patients (0.6%) receiving enzalutamide.
`
`Conclusions
`Enzalutamide significantly prolonged the survival of men with metastatic castration-
`resistant prostate cancer after chemotherapy. (Funded by Medivation and Astellas
`Pharma Global Development; AFFIRM ClinicalTrials.gov number, NCT00974311.)
`
`n engl j med 367;13 nejm.org september 27, 2012
`
`1187
`
`ARGENTUM EX1033
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org at INSTITUTE FOR CANCER RESEARCH on October 6, 2012. For personal use only. No other uses without permission.
`
` Copyright © 2012 Massachusetts Medical Society. All rights reserved.
`
`Page 1
`
`

`
`T h e ne w e ngl a nd jou r na l o f m e dicine
`
`Prostate cancer is an androgen-
`
`dependent disease that initially responds
`but later becomes resistant to established
`therapies that reduce circulating testosterone lev-
`els or inhibit androgen binding to the androgen
`receptor.1-4 Reactivation of the disease despite cas-
`trate levels of testosterone represents a transition
`to the lethal phenotype of castration-resistant
`prostate cancer.5,6 This state was previously called
`androgen-independent or hormone-refractory pros-
`tate cancer but is now recognized to be driven by
`androgen-receptor signaling, in part due to over-
`expression of the androgen receptor itself.7,8 In
`preclinical models of prostate cancer, androgen-
`receptor overexpression shortens the period of tu-
`mor latency and confers resistance to conventional
`antiandrogen agents, such as bicalutamide.9
`Enzalutamide (formerly MDV3100) is an an-
`drogen-receptor–signaling inhibitor chosen for
`clinical development on the basis of activity in
`prostate-cancer models with overexpression of
`the androgen receptor. Enzalutamide is distinct
`from the currently available antiandrogen agents
`in that it inhibits nuclear translocation of the
`androgen receptor, DNA binding, and coactivator
`recruitment. It also has a greater affinity for the
`receptor, induces tumor shrinkage in xenograft
`models (in which conventional agents only retard
`growth), and has no known agonistic effects.10,11
`In a phase 1–2 trial enrolling men with castra-
`tion-resistant prostate cancer (some of whom had
`undergone previous chemotherapy) conducted by
`the Prostate Cancer Clinical Trials Consortium,12
`enzalutamide had significant antitumor activity
`regardless of previous chemotherapy status. On the
`basis of these findings, a dose of enzalutamide
`was identified for further study.13 In our phase
`3 trial, we evaluated whether enzalutamide would
`prolong life in men with progressive castration-
`resistant prostate cancer after chemotherapy. The
`design incorporated the recommendations of the
`Prostate Cancer Clinical Trials Working Group 2
`(PCWG2)14 to avoid premature study-drug discon-
`tinuation and to help address previously identified
`difficulties in assessing outcomes in clinical trials
`involving men with prostate cancer.
`
`Methods
`
`Study Design and Conduct
`AFFIRM (A Study Evaluating the Efficacy and
`Safety of the Investigational Drug MDV3100) was
`
`an international, phase 3, randomized, double-
`blind, placebo-controlled study of enzalutamide in
`patients with prostate cancer who had previously
`been treated with one or two chemotherapy regi-
`mens, at least one of which contained docetaxel.
`The review boards of all participating institu-
`tions approved the study, which was conducted
`according to the provisions of the Declaration of
`Helsinki and the Good Clinical Practice Guide-
`lines of the International Conference on Harmo-
`nization. All patients provided written informed
`consent to participate in the study.
`The study was designed and the protocol was
`written by the senior academic authors and rep-
`resentatives of one of the sponsors (Medivation).
`The first draft of the manuscript was written by
`the first author, and the manuscript was then com-
`pleted and approved by all the authors. All the
`authors were responsible for writing the manu-
`script and for the decision to submit the manu-
`script for publication, and all the authors assume
`responsibility for the completeness and integrity
`of the data and the fidelity of the study to the pro-
`tocol and analysis plan (available with the full
`text of this article at NEJM.org). All the authors
`or authors’ institutions had agreements with the
`sponsor regarding confidentiality of the data. No
`one who is not an author contributed to the writ-
`ing of the manuscript.
`
`Study Participants
`The study was conducted at 156 sites in 15 coun-
`tries. Patients were eligible for enrollment if they
`had a histologically or cytologically confirmed
`diagnosis of prostate cancer, castrate levels of
`testosterone (<50 ng per deciliter [1.7 nmol per
`liter]), previous treatment with docetaxel, and
`progressive disease defined according to PCWG2
`criteria (see the Study End Points section below),
`including three increasing values for prostate-
`specific antigen (PSA) or radiographically con-
`firmed progression with or without a rise in the
`PSA level.14 A complete list of inclusion and ex-
`clusion criteria is provided in the protocol.
`Patients were enrolled from September 2009
`through November 2010 and were randomly as-
`signed to a study treatment centrally by means of
`an interactive voice-response system after strati-
`fication according to the baseline Eastern Coop-
`erative Oncology Group (ECOG) performance
`status score (0 or 1 vs. 2) and the Brief Pain In-
`ventory–Short Form (BPI-SF) question 3 score ad-
`
`1188
`
`n engl j med 367;13 nejm.org september 27, 2012
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org at INSTITUTE FOR CANCER RESEARCH on October 6, 2012. For personal use only. No other uses without permission.
`
` Copyright © 2012 Massachusetts Medical Society. All rights reserved.
`
`Page 2
`
`

`
`Enzalutamide in Prostate Cancer
`
`dressing the average pain over the 7 days before
`randomization (0 to 3 [no pain to mild pain] vs.
`4 to 10 [moderate-to-severe pain]).
`ECOG performance scores range from 0 to 5,
`with 0 indicating full activity, 1 indicating a re-
`striction in strenuous activity but the ability to
`be ambulatory and do light work, and 2 indicat-
`ing an ability to be ambulatory but an inability
`to work.15 Scores on BPI-SF question 3, which asks
`about the worst pain in the previous 24 hours,
`range from 0 to 10, with higher scores reflecting
`a greater severity of pain.16
`Patients were randomly assigned in a 2:1 ratio
`to receive enzalutamide (160 mg orally once
`daily as four 40-mg capsules) or matched place-
`bo capsules. Permuted-block randomization was
`used. The use of prednisone or other glucocorti-
`coids was permitted but not required, and the
`study drug was given without regard to food in-
`take. Investigators were encouraged to continue
`study treatment until radiographically confirmed
`disease progression requiring initiation of new
`systemic antineoplastic therapy. The safety and ef-
`ficacy data that were collected are described in the
`Supplementary Appendix, available at NEJM.org.
`
`For the analysis of progression-free survival,
`we used the following measures of progression
`(as indicated by the results of computed tomog-
`raphy or magnetic resonance imaging of soft tis-
`sue and of radionuclide bone scanning): progres-
`sion of soft-tissue disease according to RECIST,
`version 1.118; progression of osseous disease ac-
`cording to bone scans showing two or more new
`lesions per PCWG2; and death from any cause.
`Progression in bone at the first scheduled assess-
`ment, at week 13, required a confirmatory scan
`performed 6 or more weeks later showing addi-
`tional new lesions.14 The times to PSA progression
`and the first skeletal-related event were also re-
`corded. PSA progression was defined as an in-
`crease by a factor of 1.25 over the baseline level
`(for patients in whom the PSA level had not de-
`creased) or over the nadir level (for patients in
`whom the PSA level had decreased) and an in-
`crease in the absolute PSA level by at least 2 ng per
`milliliter, which was confirmed by a repeat mea-
`surement.14 A skeletal-related event was defined as
`radiation therapy or surgery to bone, pathologic
`bone fracture, spinal cord compression, or change
`of antineoplastic therapy to treat bone pain.17
`
`Study End Points
`The primary end point was overall survival,
`which was defined as the time from randomiza-
`tion to death from any cause. Secondary end points
`included measures of response (in the PSA level,
`in soft tissue, and in the quality-of-life score) and
`measures of progression (time to PSA progression,
`radiographic progression-free survival, and time
`to the first skeletal-related event17).
`We used the following definitions of the sec-
`ondary end points (as detailed in Table 1S in the
`Supplementary Appendix): PSA-level response was
`defined as a reduction in the PSA level from base-
`line by 50% or more or 90% or more, as con-
`firmed on an additional PSA evaluation performed
`3 or more weeks later.14 Objective soft-tissue re-
`sponse was defined by the Response Evaluation
`Criteria in Solid Tumors (RECIST), version 1.1.18
`Quality-of-life response was defined as a 10-point
`improvement in the global score on the Func-
`tional Assessment of Cancer Therapy-Prostate
`(FACT-P) questionnaire, as compared with base-
`line, on two consecutive measurements obtained
`at least 3 weeks apart.19,20 The FACT-P is a 39-item
`questionnaire on which the score for each item
`can range from 0 to 4, with higher scores indi-
`cating a better quality of life.
`
`Statistical Analysis
`All analyses were performed by the sponsor us-
`ing data obtained as of the cutoff date of Septem-
`ber 25, 2011. The primary efficacy end point was
`a between-group comparison of the time from
`randomization to death from any cause (overall
`survival) in the intention-to-treat population (all
`randomly assigned patients). The study was de-
`signed to have a power of 90% to detect a hazard
`ratio of 0.76 for death in the enzalutamide group,
`as compared with the placebo group, with a two-
`sided type I error rate of 0.05. We planned to
`enroll approximately 1170 patients, assuming a
`median survival of 15.7 months in the enzaluta-
`mide group and 12.0 months in the placebo group,
`an accrual period of approximately 12 months,
`and a total study duration of approximately 30
`months to observe the required 650 events.
`A single interim analysis was planned to be
`performed after 520 deaths (80% of the 650 total
`events) had occurred. The analysis was done ac-
`cording to a group sequential design with the
`use of a Lan–DeMets implementation of the
`O’Brien–Fleming stopping boundary (P<0.02). In
`the primary analysis, we used a log-rank test to
`evaluate overall survival, with stratification ac-
`cording to the ECOG performance-status score
`
`n engl j med 367;13 nejm.org september 27, 2012
`
`1189
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org at INSTITUTE FOR CANCER RESEARCH on October 6, 2012. For personal use only. No other uses without permission.
`
` Copyright © 2012 Massachusetts Medical Society. All rights reserved.
`
`Page 3
`
`

`
`The NEW ENGLAND IOURNAL of MEDICINE
`
`Enzalutamide OverallSurvival(96)
`
`Hazard ratio, 0.63 (95% CI, 0.53-0.75)
`p<o.oo1
`
`A Overall Survival
`
`1°°
`90
`80
`70
`60
`50
`40
`30
`
`20
`10
`
`
`
`No. at Risk
`Enzalutamide
`Placebo
`
`800
`399
`
`775
`376
`
`701
`317
`
`627
`263
`
`400
`167
`
`211
`81
`
`72
`33
`
`Months
`
`B TimetoP$AProgression
`100
`90
`3°
`7°
`°°
`5°
`:3
`
`oE:
`
`1.
`
`Hm“, mm 015 (95% c,’°_2o_o_3o)
`
`Enzalutamide
`
`
`
`Progression(96)
`
`‘S
`.8.1:
`3M
`
`E_
`
`Figure 1. Kaplan—Meier Estimates of Primary and Sec-
`ondary End Points in the Intention-to-Treat Population.
`Shown are data for overall survival, the primary end
`point (Panel A), and for two secondary end points. the
`time to prostate-specific antigen (PSA) progression
`(Panel B) and radiographic progression-free survival
`(Panel C), in the enzalutamide group, as compared
`with the placebo group. Cl denotes confidence interval.
`
`all survival were performed with the use of the
`unstratified log-rank test and Cox proportional-
`hazards models. Subgroup analyses were con-
`ducted to determine whether treatment effects
`
`were consistent across patient subgroups. A mul-
`tivariate analysis was also performed.
`Only if the overall survival analysis showed
`statistical superiority of enzalutamide over pla-
`cebo was the testing of the key secondary end
`points to be undertaken, in the rank-prioritized
`order — the time to PSA progression, radiograph-
`ic progression-free survival, and the time to the
`first skeletal-related event — with the signifi-
`cance of the previous end point gating further
`testing. These end points were tested by means
`of the stratified log-rank test in a protected hier-
`archical manner, each at the two-sided signifi-
`cance level of 0.05.
`
`RESULTS
`
`PATIENTS AND TREATM ENT
`
`The study enrolled 1199 patients who were ran-
`domly assigned to receive either enzalutamide
`(800 patients) or placebo (399 patients). The en-
`rollment, follow-up, and data analysis of patients
`are shown in Figure 1S in the Supplementary Ap-
`pendix. Baseline characteristics were well matched
`between groups in terms of demographic charac-
`teristics, previous treatment history, and extent of
`disease (Table 28 in the Supplementary Appendix).
`At the time of the interim analysis, the median
`time on treatment was 8.3 months in the enzalu-
`tamide group and 3.0 months in the placebo
`group. The median duration offollow-up to ascer-
`tain survival status was 14.4 months.
`
`EFFICACY
`
`The median overall survival was 18.4 months (95%
`confidence interval [CI], 17.3 to not yet reached)
`among patients receiving enzalutamide and 13.6
`months (95% CI, 11.3 to 15.8) among patients
`receiving placebo (Fig. 1A). At the time of the
`prespecified interim analysis, the use of enzalu-
`
`::
`
`No. at Risk
`Enzalutamide
`Placebo
`
`800
`399
`
`603
`107
`
`287
`12
`
`145
`
`(96)
`
`C Radiographic Progression-free Survival
`mo
`90
`so
`70
`60
`5°
`‘°
`3°
`2°
`1°
`
`
`
`RadiographicProgresion-freeSunn'val
`
`Hm“, mm,’ M, (95% c,,o_35_M7)
`P<0.001
`
`.
`Emlmmlde
`
`'
`
`'
`
`'
`' 1‘;
`Months
`
`800
`399
`
`533
`176
`
`447
`86
`
`237
`46
`
`140
`20
`
`and the baseline mean pain score (as measured
`by the BPI-SF score); the results are presented as
`Kaplan—Meier curves. Supportive analyses ofover-
`
`1190
`
`N ENGLJ MED 367:1; NEjM.ORG
`
`SEPTEMBER 27, 2012
`
`The New England Journal of Medicine
`Downloaded from nejrnorg at INSTITUTE FOR CANCER RESEARCH on October 6, 2012. For personal use only. No other use
`Copyright © 2012 Massachusetts Medical Society. All rights reserved.
`
`Rmgt fiermission.
`
`Page 4
`
`

`
`ENZALUTAMIDE IN PROSTATE CANCER
`
`0.63 (0.53—0.75)
`
`0.63 (0.46—0.87)
`0.63 (0.5l—0.78)
`
`0.62 (052-015)
`0.65 (0.39—1.07)
`
`0.59 (o.47—o.74)
`0.71 (054-094)
`
`0.63 (0.47—0.83)
`0.64 (051-030)
`
`0.59 (0.46—0.75)
`0.68 (0.53—o.s3)
`
`0.59 (0.48—0.73)
`0.74 (0.54—l.03)
`
`0.62 (0.46-0.83)
`0.64 (o.52—o.so)
`
`0.59 (046-075)
`0.67 (o.52—o.s7)
`
`0.56 (o.46—0.69)
`
`0.78 (0.56—1.09)
`
`0.67 (050-039)
`0.62 (o.50—o.7a)
`
`0.63 (0.46—0.86)
`0.61 (0.50—0.76)
`
`Placebo
`Ermdntamide
`median overall survival (mo)
`18.4
`13.6
`
`Hazard Ratio (95% Cl)
`III
`
`I-C—I
`
`I I
`
`I
`E
`I
`II
`
`|—0j|
`I-Oil
`
`I-O—I
`
`I E
`
`I I E
`
`I
`
`E
`
`IIIII
`IIIIII
`
`I
`|—ojI-|
`
`l—o::-I
`:
`I
`:
`
`5IIII
`
`l
`
`12.4
`13.9
`
`-
`
`_
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`Subgroup
`
`No. ofPatients
`
`1199
`
`362
`837
`
`1097
`102
`
`All patients
`Age
`<65 yr
`:65 yr
`Baseline ECOG performance status score
`0-1
`2
`Baseline mean pain score on BPI—SF
`(question no. 3)
`
`<4
`24
`Geographic region
`North Ameria
`Other
`No. of previous hormonal treatments
`s2
`>2
`No. of previous dlemotherapy regimens
`l
`22
`Type of progression at study entry
`PSA progression only
`Radiographic progression with or
`without PSA progression
`No. of bone lesions
`520
`>20
`
`Visceral (liver or lung) disease at baseline
`No
`
`Yes
`Baseline PSA level
`sMedian
`>Median
`Baselirle LDH level
`sMedian
`>Median
`
`Figure 2. Subgroup Analyses of Hazard Ratios for Death ‘II the Two Study Groups.
`Hazard ratios are based on a nonstratified proportional-hazards model. Dashes indicate that the median time to death had not been
`reached for the indicated subgroup. The size ofthe circles is proportional to the size ofthe subgroup. The horizontal bars represent
`95% confidence intervals. The Eastern Cooperative Oncology Group (ECOG) grades the performance status of patients with respect to
`activities of daily living. with 0 indicating that the patient is fully active and able to carry out all predisease activities without restriction;
`1 indicating that the patient is restricted in physically strenuous activity but is ambulatory and able to carry out work of a light or seden-
`tary nature; and 2 indicating that the patient is ambulatory and up and about for more than 50% ofwaking hours and is capable of self-
`care but unable to carry out work activities. Scores on the Brief Pain |nventory—Short Form (BPI-SF) range from 0 to 10. with scores of
`0 to 3 indicating that clinically significant pain is absent and scores of 4 to 10 indicating that clinically significant pain is present, and
`with higher scores indicating greater pain. LDH denotes lactate dehydrogenase, and PSA prostate-specific antigen.
`
`tamide resulted in a 37% reduction in the risk of mended that the study be halted and unblinded,
`death, as compared with placebo (hazard ratio with eligible patients in the placebo group of-
`for death, 0.63; 95% CI, 053 to 0.75; P<0.001).
`fered treatment with enzalutamide. These results
`On the basis of these results, an independent were confirmed at the time that the database was
`data and safety monitoring committee rec0m-
`locked and are presented here.
`
`N ENGLJ urn 367:1; NE]M.OltG SEPTEMBER 27, 2012
`
`1191
`
`The New England Journal ofMedicine
`Downloaded fiom nejm.org at lNS'I'ITU'I‘E FOR CANCER RESEARCH on October 6, 2012. For personal use only. No other ImP grmission
`Copyright 0 2012 Massachmets Medical Society. All rights reserved
`
`Page 5
`
`

`
`T h e ne w e ngl a nd jou r na l o f m e dicine
`
`In the intention-to-treat population, 308 of
`800 patients (39%) died in the enzalutamide
`group and 212 of 399 patients (53%) died in the
`placebo group. When the study was unblinded,
`231 patients (29%) in the enzalutamide group
`were receiving the study drug, as compared with
`only 19 patients (5%) in the placebo group.
`The overall survival benefit was consistent
`across all subgroups, including age, baseline
`pain intensity, geographic region, and type of
`disease progression at entry (Fig. 2), and it was
`maintained in the supportive analyses of overall
`survival performed with the use of the unstrati-
`fied log-rank test and the Cox proportional-
`hazards model. The effect of enzalutamide on
`overall survival was maintained after adjustment
`for stratification factors and baseline prognostic
`factors, as shown in Table 1 in a multivariate
`analysis (hazard ratio for death, 0.58; 95% CI,
`0.49 to 0.70; P<0.001). Systemic antineoplastic
`treatments were used for prostate cancer after
`the study drug was discontinued in a large pro-
`portion of patients, more commonly in the pla-
`cebo group (in 61% of patients) than in the
`enzalutamide group (in 42% of patients). Among
`patients receiving at least one therapy after dis-
`continuation of the study drug, the agents used
`
`included abiraterone acetate in 21% of patients
`in the enzalutamide group and in 24% of those
`in the placebo group and cabazitaxel in 10% and
`14%, respectively. Both agents have been shown
`to confer a survival benefit for men with this
`disease state (see Table 3S in the Supplementary
`Appendix).
`The superiority of enzalutamide over placebo
`was shown for all secondary end points, includ-
`ing PSA-level response rate (54% vs. 2%, P<0.001),
`soft-tissue response rate (29% vs. 4%, P<0.001),
`FACT-P quality-of-life response (43% vs. 18%,
`P<0.001), the time to PSA progression (8.3 vs. 3.0
`months; hazard ratio, 0.25; P<0.001) (Fig. 1B),
`radiographic progression-free survival (8.3 vs. 2.9
`months; hazard ratio, 0.40; P<0.001) (Fig. 1C), and
`the time to the first skeletal-related event (16.7 vs.
`13.3 months; hazard ratio, 0.69; P<0.001) (Table 2).
`
`Safety
`Though the period of observation for the enzalu-
`tamide group was more than twice that for the
`placebo group, the rates of adverse events were
`similar in the two groups (Table 3). The enzalu-
`tamide group had a lower incidence of adverse
`events of grade 3 or above (45.3%, vs. 53.1% in
`the placebo group). The median time to the first
`
`Table 1. Multivariate Analysis of Hazard Ratios for Death.*
`
`Variable
`
`Study treatment (enzalutamide vs. placebo)
`
`ECOG performance score (0 or 1 vs. 2)
`
`Mean pain score on BPI-SF (question no. 3)(<4 vs. ≥4)†
`
`Progression at study entry (PSA only vs. radiographic)
`
`Visceral disease at screening (no vs. yes)
`
`Baseline serum lactate dehydrogenase (per increase
`of 1 U per liter)
`
`Measurement Estimates
`
`Hazard Ratio for
`Death (95% CI)
`
`Coefficient
`
`−0.54±0.09
`
`−0.33±0.14
`
`−0.23±0.10
`
`−0.29±0.09
`
`−0.47±0.10
`
`0.00±0.00
`
`P Value
`
`<0.001
`
`0.02
`
`0.02
`
`0.002
`
`<0.001
`
`<0.001
`
`0.58 (0.49–0.70)
`
`0.72 (0.55–0.95)
`
`0.79 (0.65–0.97)
`
`0.75 (0.62–0.90)
`
`0.63 (0.52–0.76)
`
`1.002 (1.001–1.002)
`
`Baseline hemoglobin (per increase of 1 g per liter)
`
`−0.03±0.00
`
`<0.001
`
`0.97 (0.97–0.98)
`
`* Data with respect to survival for patients who were alive at the time of analysis were censored at the date the patient
`was last known to be alive. Hazard ratios for death were calculated after adjustment for prognostic factors. Several fac-
`tors were entered into a Cox proportional-hazards model, and a stepwise selection method was used in which nonsig-
`nificant factors were eliminated at entry into the model (P≥0.10) and further eliminated after the contribution to the
`model was assessed (P≥0.25). These included the factors listed in the table as well as age (<65 years vs. ≥65 years),
` region (North America vs. other), number of previous chemotherapy regimens (1 vs. 2), and baseline serum prostate-
`specific antigen (PSA) level (per increase of 1 ng per milliliter). The Gleason score for prostate tumors was excluded
`owing to a large number of missing values. CI denotes confidence interval, and ECOG Eastern Cooperative Oncology
`Group.
`† Scores on the Brief Pain Inventory–Short Form (BPI-SF) range from 0 to 10, with scores of 0 to 3 indicating that clinically
`significant pain is absent and scores of 4 to 10 indicating that clinically significant pain is present, and with higher
`scores indicating greater pain.
`
`1192
`
`n engl j med 367;13 nejm.org september 27, 2012
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org at INSTITUTE FOR CANCER RESEARCH on October 6, 2012. For personal use only. No other uses without permission.
`
` Copyright © 2012 Massachusetts Medical Society. All rights reserved.
`
`Page 6
`
`

`
`Enzalutamide in Prostate Cancer
`
`Table 2. Secondary End Points Related to Response and Disease Progression.*
`
`End Point
`
`Confirmed PSA decline†
`
`Enzalutamide
`(N = 800)
`
`Placebo
`(N = 399)
`
`Hazard Ratio
`(95% CI)
`
`P Value
`
`Patients with ≥1 postbaseline PSA assessment ―
`no. (%)
`
`731 (91)
`
`330 (83)
`
`PSA response ― no./total no. (%)
`
`Decline ≥50% from baseline
`
`Decline ≥90% from baseline
`
`Soft-tissue objective response
`
`395/731 (54)
`
`5/330 (2)
`
`181/731 (25)
`
`3/330 (1)
`
`<0.001
`
`<0.001
`
`Patients with measurable disease ― no. (%)
`
`446 (56)
`
`208 (52)
`
`Complete or partial objective response ― no./
`total no. (%)
`
`FACT-P quality-of-life response†
`
`129/446 (29)
`
`8/208 (4)
`
`<0.001
`
`Patients with ≥1 postbaseline assessment ― no. (%)
`
`651 (81)
`
`257 (64)
`
`Quality-of-life response ― no./total no. (%)‡
`
`281/651 (43)
`
`47/257 (18)
`
`<0.001
`
`Progression indicators
`
`Time to PSA progression ― mo
`
`Median
`
`95% CI
`
`Radiographic progression-free survival ― mo
`
`Median
`
`95% CI
`
`0.25 (0.20–0.30)
`
`<0.001
`
`0.40 (0.35–0.47)
`
`<0.001
`
`8.3
`
`5.8–8.3
`
`8.3
`
`8.2–9.4
`
`3.0
`
`2.9–3.7
`
`2.9
`
`2.8–3.4
`
`Time to first skeletal-related event ― mo
`
`0.69 (0.57–0.84)
`
`<0.001
`
`Median
`
`95% CI
`
`16.7
`
`13.3
`
`14.6–19.1
`
`9.9–NYR
`
`* For a complete definition of end points, see Table 1S in the Supplementary Appendix. FACT-P denotes Functional
`Assessment of Cancer Therapy–Prostate, NYR not yet reached, and PSA prostate-specific antigen.
`† Only patients with both baseline and postbaseline assessments are included.
`‡ The quality-of-life response was defined as a 10-point improvement in the global score on the FACT-P questionnaire, as
`compared with baseline, on two consecutive measurements obtained at least 3 weeks apart.
`
`such adverse event was 12.6 months in the enzalu-
`tamide group, as compared with 4.2 months in
`the placebo group (Fig. 2S in the Supplementary
`Appendix). There was a higher incidence of all
`grades of fatigue, diarrhea, hot flashes, muscu-
`loskeletal pain, and headache in the enzaluta-
`mide group than in the placebo group. Cardiac
`disorders were noted in 6% of patients receiving
`enzalutamide and in 8% of patients receiving
`placebo (with cardiac disorders of grade 3 in 1%
`and 2%, respectively). Hypertension or increased
`blood pressure was observed in 6.6% of patients
`in the enzalutamide group and 3.3% of those in
`the placebo group. There were no significant
`between-group imbalances in the rates of other
`adverse events, such as hyperglycemia, weight gain,
`
`hyperlipidemia, or glucose intolerance. There-
`fore, there was no evidence to suggest the devel-
`opment of a metabolic syndrome associated with
`enzalutamide, although the study was not de-
`signed to formally evaluate this event. Liver-func-
`tion abnormalities were reported as adverse events
`in 1% of patients receiving enzalutamide and in
`2% of those receiving placebo.
`A comprehensive evaluation of electrocardio-
`graphic data, including the QT interval and the
`QT interval corrected for heart rate (QTc), revealed
`no clinically relevant changes in heart rate, atrio-
`ventricular conduction, cardiac depolarization, or
`effect on cardiac repolarization as determined
`by means of the QTc according to Fridericia’s
`formula.21
`
`n engl j med 367;13 nejm.org september 27, 2012
`
`1193
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org at INSTITUTE FOR CANCER RESEARCH on October 6, 2012. For personal use only. No other uses without permission.
`
` Copyright © 2012 Massachusetts Medical Society. All rights reserved.
`
`Page 7
`
`

`
`T h e ne w e ngl a nd jou r na l o f m e dicine
`
`Table 3. Adverse Events, According to Grade.
`
`Adverse Event
`
`Enzalutamide (N = 800)
`
`Placebo (N = 399)
`
`Any Grade
`
`Grade ≥3
`
`Any Grade
`
`Grade ≥3
`
`number of patients (percent)
`
`≥1 Adverse event
`
`Any serious adverse event
`
`Discontinuation owing to adverse event
`
`Adverse event leading to death
`
`Frequent adverse events more common with
`enzalutamide*
`
`Fatigue
`
`Diarrhea
`
`Hot flash
`
`Musculoskeletal pain
`
`Headache
`
`Clinically significant adverse events
`
`Cardiac disorder
`
`Any
`
`Myocardial infarction
`
`Abnormality on liver-function testing†
`
`Seizure
`
`785 (98)
`
`268 (34)
`
`61 (8)
`
`23 (3)
`
`269 (34)
`
`171 (21)
`
`162 (20)
`
`109 (14)
`
`93 (12)
`
`49 (6)
`
`2 (<1)
`
`8 (1)
`
`5 (<1)
`
`362 (45)
`
`227 (28)
`
`37 (5)
`
`23 (3)
`
`50 (6)
`
`9 (1)
`
`0
`
`8 (1)
`
`6 (<1)
`
`7 (1)
`
`2 (<1)
`
`3 (<1)
`
`5 (<1)
`
`390 (98)
`
`154 (39)
`
`39 (10)
`
`14 (4)
`
`116 (29)
`
`70 (18)
`
`41 (10)
`
`40 (10)
`
`22 (6)
`
`30 (8)
`
`2 (<1)
`
`6 (2)
`
`0
`
`212 (53)
`
`134 (34)
`
`28 (7)
`
`14 (4)
`
`29 (7)
`
`1 (<1)
`
`0
`
`1 (<1)
`
`0
`
`8 (2)
`
`2 (<1)
`
`3 (<1)
`
`0
`
`* Included in this category are adverse events that occurred in more than 10% of patients in the enzalutamide group and
`that occurred in the enzalutamide group at a rate that was at least 2 percentage points higher than that in the placebo
`group.
`† Abnormalities on liver-function testing included hyperbilirubinemia and increased levels of aspartate aminotransferase
`or alanine aminotransferase.
`
`Five of the 800 patients in the enzalutamide
`group (0.6%) were reported by the investigators
`to have had a seizure; no seizures were reported
`in the placebo group. One case of status epilep-
`ticus (confusion associated with partial complex-
`status epilepticus) required medical intervention;
`the four other seizures were self-limited and did
`not recur after study-drug discontinuation. Four
`of the seizures were witnessed. Potentially predis-
`posing factors were present in several patients.
`Two patients had brain metastases, 1 of whom
`had a seizure reported 26 days after the last dose
`of enzalutamide. One patient had inadvertently
`been administered lidocaine intravenously im-
`mediately before the seizure, and 1 patient with
`brain atrophy had an unwitnessed event classi-
`fied as a seizure, in the context of a history of
`heavy alcohol use, after initiation of haloperidol
`7 days beforehand. One additional adverse event
`reported by the investigator as a syncope had
`several features suggestive of seizure.
`
`Discussion
`
`In this phase 3 study, we found that enzaluta-
`mide, an oral androgen-receptor–signaling in-
`hibitor, significantly prolonged the survival of
`men with metastatic castration-resistant prostate
`cancer after chemotherapy by a median of 4.8
`months and reduced the risk of death from any
`cause by 37% versus placebo. In a multivariate
`analysis, the survival benefit was seen in all pa-
`tient subgroups, including those stratified ac-
`cording to age and ECOG performance status,
`the geographic location of the study center, the
`extent of disease on diagnostic imaging, and bio-
`chemical measurements that included PSA and
`lactate dehydrogenase, even after adjustment for
`baseline factors.
`These data confirm the central role of the an-
`drogen receptor and androgen-receptor signaling
`in the progression of prostate cancer throughout
`the spectrum of disease. Castration-resistant
`
`1194
`
`n engl j med 367;13 nejm.org september 27, 2012
`
`The New England