V O L U M E 2 7 䡠 N U M B E R 2 3 䡠 A U G U S T 1 0 2 0 0 9
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`O R I G I N A L R E P O R T
`
`From the Royal Marsden National
`Health Service Foundation Trust; The
`Institute of Cancer Research, Sutton,
`Surrey, United Kingdom; and Cougar
`Biotechnology, Los Angeles, CA.
`
`Submitted September 25, 2008;
`accepted February 13, 2009; published
`online ahead of print at www.jco.org on
`May 26, 2009.
`
`Supported by Cougar Biotechnology.
`G.A., A.H.M.R., C.M., L.R.M., G.M.,
`E.T., D.O., R.S., S.B.K., and J.S.d.B.
`are in the Section of Medicine, which
`is supported by a Cancer Research
`UK program grant and an Experimen-
`tal Cancer Medicines Centre grant
`from Cancer Research UK and the
`Department of Health (Ref: C51/
`A7401). G.A. and A.H.M.R. were also
`supported by the Royal Marsden
`Hospital Research Fund. G.A. is also
`supported by the Prostate Cancer
`Foundation, Santa Monica, CA. C.P.
`was supported by Cancer Research
`UK and the National Cancer Research
`Institute Prostate Cancer Collabora-
`tive. R.A. is in the Cancer Research
`UK Section of Clinical Trials at The
`Institute of Cancer Research, Surrey,
`UK. We also acknowledge National
`Health Service funding to the National
`Institute for Health Research Biomedi-
`cal Research Centre. Abiraterone
`acetate was developed at The Insti-
`tute of Cancer Research, which there-
`fore has a commercial interest in the
`development of this agent.
`
`Authors’ disclosures of potential con-
`flicts of interest and author contribu-
`tions are found at the end of this
`article.
`
`Clinical Trials repository link available on
`JCO.org.
`
`Corresponding author: Johann S. de
`Bono, MB ChB, FRCP, MSc, PhD,
`Section of Medicine, The Institute of
`Cancer Research, the Royal Marsden
`National Health Service Foundation
`Trust, Downs Rd, Sutton, Surrey SM2
`5PT, United Kingdom; e-mail: johann
`.de-bono@icr.ac.uk.
`
`The Acknowledgment and Appendix
`are included in the full-text version
`of this article; they are available
`online at www.jco.org. They are
`not included in the PDF version
`(via Adobe® Reader®).
`
`© 2009 by American Society of Clinical
`Oncology
`
`0732-183X/09/2723-3742/$20.00
`
`DOI: 10.1200/JCO.2008.20.0642
`
`Selective Inhibition of CYP17 With Abiraterone Acetate Is
`Highly Active in the Treatment of Castration-Resistant
`Prostate Cancer
`Gerhardt Attard, Alison H.M. Reid, Roger A’Hern, Christopher Parker, Nikhil Babu Oommen,
`Elizabeth Folkerd, Christina Messiou, L. Rhoda Molife, Gal Maier, Emilda Thompson, David Olmos,
`Rajesh Sinha, Gloria Lee, Mitch Dowsett, Stan B. Kaye, David Dearnaley, Thian Kheoh, Arturo Molina,
`and Johann S. de Bono
`
`A
`
`B
`
`S
`
`T
`
`R
`
`A
`
`C
`
`T
`
`Purpose
`It has been postulated that castration-resistant prostate cancer (CRPC) commonly remains
`hormone dependent. Abiraterone acetate is a potent, selective, and orally available inhibitor of
`CYP17, the key enzyme in androgen and estrogen biosynthesis.
`
`Patients and Methods
`This was a phase I/II study of abiraterone acetate in castrate, chemotherapy-naive CRPC patients
`(n ⫽ 54) with phase II expansion at 1,000 mg (n ⫽ 42) using a two-stage design to reject the null
`hypothesis if more than seven patients had a prostate-specific antigen (PSA) decline of ⱖ 50%
`(null hypothesis ⫽ 0.1; alternative hypothesis ⫽ 0.3; ␣ ⫽ .05; ␤ ⫽ .14). Computed tomography
`scans every 12 weeks and circulating tumor cell (CTC) enumeration were performed. Prospective
`reversal of resistance at progression by adding dexamethasone 0.5 mg/d to suppress adrenocor-
`ticotropic hormone and upstream steroids was pursued.
`
`Results
`A decline in PSA of ⱖ 50% was observed in 28 (67%) of 42 phase II patients, and declines of
`ⱖ 90% were observed in eight (19%) of 42 patients. Independent radiologic evaluation reported
`partial responses (Response Evaluation Criteria in Solid Tumors) in nine (37.5%) of 24 phase II
`patients with measurable disease. Decreases in CTC counts were also documented. The median
`time to PSA progression (TTPP) on abiraterone acetate alone for all phase II patients was 225 days
`(95% CI, 162 to 287 days). Exploratory analyses were performed on all 54 phase I/II patients; the
`addition of dexamethasone at disease progression reversed resistance in 33% of patients
`regardless of prior treatment with dexamethasone, and pretreatment serum androgen and
`estradiol levels were associated with a probability of ⱖ 50% PSA decline and TTPP on abiraterone
`acetate and dexamethasone.
`
`Conclusion
`CYP17 blockade by abiraterone acetate results in declines in PSA and CTC counts and radiologic
`responses, confirming that CRPC commonly remains hormone driven.
`
`J Clin Oncol 27:3742-3748. © 2009 by American Society of Clinical Oncology
`
`INTRODUCTION
`
`Prostate cancer mortality is invariably a result of
`what has been described as hormone-refractory or
`androgen-independent disease. Patients who expe-
`rience relapse despite castration can respond to fur-
`ther hormonal treatments, but these have been
`modestly effective to date.1,2 Evidence is accumulat-
`ing that the more appropriately named castration-
`resistant prostate cancer (CRPC) frequently remains
`hormone driven, by using adrenal hormones or
`through intracrine synthesis.3-7
`
`CYP17 is key to androgen and estrogen synthe-
`sis. The nonspecific weak inhibitor of CYP17, keto-
`conazole, has modest antitumor activity in CRPC,8
`and its utility has been limited by its toxicities. Fur-
`thermore, an increase in androgenic steroids at dis-
`ease progression on this agent indicates incomplete
`target blockade.8 Chemists at our institution used
`testicular extracts and radiolabeled CYP17 steroid
`substrates to screen for small-molecule inhibitors of
`this enzyme.9-12 This led to the design of a potent,
`selective, and irreversible inhibitor of CYP17, abi-
`raterone,13,14 which we have shown to be safe, with
`
`3742
`
`© 2009 by American Society of Clinical Oncology
`Downloaded from jco.ascopubs.org on October 29, 2014. For personal use only. No other uses without permission.
`Copyright © 2009 American Society of Clinical Oncology. All rights reserved.
`
`JANSSEN EXHIBIT 2015
`Mylan v. Janssen IPR2016-01332
`
`

`

`Phase II Trial of Abiraterone Acetate
`
`promising antitumor activity in chemotherapy-naive patients with
`CRPC.15 In these studies, concomitant castration was continued to
`prevent a compensatory luteinizing hormone surge that can overcome
`CYP17 blockade.16 Continuous CYP17 inhibition results in raised
`levels of adrenocorticotropic hormone (ACTH) that increase steroid
`levels upstream of CYP17, including corticosterone and deoxycor-
`ticosterone. These raised upstream steroids prevent adrenocortical
`insufficiency but can result in a syndrome of secondary mineralocor-
`ticoid excess characterized by fluid retention, hypertension, and hy-
`pokalemia. This can be ameliorated by mineralocorticoid antagonists
`or low-dose glucocorticoids, which decrease ACTH and steroids up-
`stream of the CYP17 blockade. This may be of relevance to antitumor
`activity because these upstream steroids are implicated in activating a
`promiscuous androgen receptor (AR).17 To rapidly evaluate the anti-
`tumor activity of abiraterone acetate, this phase I study seamlessly
`expanded15 into a two-stage, single-arm, phase II trial; we now present
`data of the antitumor activity of abiraterone acetate administered at
`1,000 mg, once daily, continuously in chemotherapy-naive CRPC
`patients who experienced progression on multiple treatments.
`
`PATIENTS AND METHODS
`
`Patients
`All of the patients enrolled onto this study were castrate, had an Eastern
`Cooperative Oncology Group performance status of 0 or 1, and had progres-
`sive disease as defined by Prostate-Specific Antigen Working Group (PSAWG)
`I.18 This was a single-center study conducted at the Royal Marsden Hospital
`(London, United Kingdom). Patients were required to have a minimum
`washout period of 4 weeks after the use of prostate cancer therapy except
`luteinizing hormone–releasing hormone (LHRH) agonists and 6 weeks after
`stopping the antiandrogens bicalutamide or nilutamide. Patients who had
`previously received cytotoxic chemotherapy or a radiopharmaceutical for
`their prostate cancer were excluded. Other eligibility criteria included castrate
`serum testosterone levels (⬍ 50 ng/dL), normal serum potassium, and ade-
`quate bone marrow, renal, and hepatic function. Patients were excluded if they
`had brain metastases or spinal cord compression, active autoimmune disease
`requiring corticosteroid therapy, uncontrolled hypertension, a history of class
`III or IV cardiac failure, or a serious concurrent medical illness. This phase I/II
`study was approved by the Ethics Review Committees of the Royal Marsden
`Hospital, and informed consent was obtained from all patients.
`Treatment and Procedures
`Four capsules (250 mg each) of abiraterone acetate powder were admin-
`istered once daily, continuously, to fasted patients in 28-day cycles. Toxicity
`related to elevated mineralocorticoid levels was managed with a mineralo-
`corticoid receptor antagonist (eplerenone 50 to 200 mg/d); treatment with
`glucocorticoids to suppress ACTH was only used if mineralocorticoid antag-
`onism did not reverse these toxicities. Spironolactone was not used because it
`has been reported to bind and activate the AR.19
`Safety evaluations were conducted at baseline, weekly for the first two
`cycles, and every cycle thereafter. All patients had a physical examination,
`CBC, international normalized ratio, partial thromboplastin time, serum cre-
`atinine, electrolytes, AST, ALT, and bilirubin. All adverse events were graded
`according to the US National Cancer Institute Common Terminology Criteria
`of Adverse Events (version 3.0). Prostate-specific antigen (PSA), lactate dehy-
`drogenase, and alkaline phosphatase were measured at baseline and at the end
`of every cycle. High-resolution computed tomography (CT) scans were per-
`formed on all patients at baseline and every 12 weeks, and bone scans were
`performed at baseline and every 24 weeks. A 7.5-mL blood sample was col-
`lected into CellSave tubes (Immunicon, Huntingdon Valley, PA) from con-
`senting patients at baseline and on the first day of every cycle for the
`enumeration of circulating tumor cells (CTCs) using the CellSearch system
`(Immunicon) as described previously.20,21 Circulating dehydroepiandro-
`
`stenedione (DHEA), DHEA-sulfate (DHEA-S), androstenedione, testoster-
`one, and estradiol were evaluated at baseline.
`Study
`This was an open-label, single-arm study of abiraterone acetate. The
`phase I, dose-escalation portion of the study, which has been reported previ-
`ously,15 recruited 21 patients and identified 1,000 mg as the recommended
`phase II dose. The primary end point was a ⱖ 50% PSA decline at any time
`after 12 weeks of treatment confirmed by a second PSA 4 weeks later (PSAWG
`I) in patients treated at 1,000 mg. PSA progression was as defined by PSAWG
`I. Because a PSA decline of ⱖ 30% was associated with improved survival in
`patients treated with docetaxel,22 the rate of PSA declines ⱖ 30% was a
`secondary end point. Declines in PSA by ⱖ 90% were also reported. However,
`in the absence of PSA being a surrogate for survival in CRPC, we have included
`waterfall plots of PSA decrements as recommended by Prostate Cancer Clini-
`cal Trials Working Group II23 (published after the start of this study), which
`although not originally specified in the protocol, are not a violation of
`protocol design. Moreover, other intermediate end points were used,
`namely radiologic assessments and changes in CTC counts. Measurable target
`lesions were monitored by CT scans using Response Evaluation Criteria in
`Solid Tumors (RECIST)24 and reported by a radiologist blinded to clinical
`outcome. Patients with ⱖ 5 CTCs/7.5 mL before treatment were classified as
`having a decline in CTCs to less than 5 CTCs/7.5 mL and/or by ⱖ 30% or
`neither; these declines have been previously shown to correlate with an im-
`provement in overall survival.25-27
`The median time to PSA progression (TTPP) was calculated using the
`Kaplan-Meier product-limit method and was defined as the time elapsed from
`the start of therapy until PSA progression after ⱖ 12 weeks as defined by
`PSAWG I criteria. This required an increase in PSA of more than 25% greater
`than baseline or nadir for patients whose PSA did not decrease or decreased by
`less than 50%, respectively, and an increase in PSA of more than 50% above
`nadir and an absolute increase of 5 ng/mL for patients whose PSA had de-
`creased by ⱖ 50%. A second confirmatory value was required. The date of data
`cutoff was July 28, 2008. Median follow-up time was calculated using the
`reverse Kaplan-Meier method; patients were censored if they had reached the
`end of their participation in the trial but were otherwise considered as events.
`Addition of Corticosteroids at Progression on
`Abiraterone Acetate
`Preclinical models report that hormones upstream of CYP17 that are
`increased as a result of high ACTH levels in patients receiving abiraterone
`acetate could activate a promiscuous AR.17,28 Promiscuous AR activation by
`antiandrogens is observed in up to 30% of CRPC patients and manifests
`clinically as a withdrawal response.8 This study was prospectively designed to
`allow addition of dexamethasone 0.5 mg daily to abiraterone acetate to all
`patients at progression. Dexamethasone 0.5 mg/d is the standard corticoste-
`roid treatment for CRPC at our institution29 and was thus used in preference
`to another corticosteroid. In the phase I study, addition of dexamethasone
`suppressed ACTH and steroids upstream of CYP17 and reversed resistance to
`abiraterone acetate.15 To confirm these preliminary observations, tumor re-
`sponses (defined earlier) on addition of dexamethasone in the 1,000-mg abi-
`raterone acetate population were assessed and reported (separate to the
`abiraterone acetate alone response data) based on whether patients had previ-
`ously experienced progression on the same dose and schedule of dexametha-
`sone (group I) or not (group II). The median TTPP from start of abiraterone
`acetate until PSA progression on dexamethasone and abiraterone acetate (or
`on abiraterone acetate alone if dexamethasone was not added), as defined by
`PSAWG I, was calculated using the Kaplan-Meier method. The median TTPP
`from start of dexamethasone until PSA progression on dexamethasone and
`abiraterone acetate is also reported. In keeping with the CONSORT statement
`on transparency in clinical trials, all relevant additional details of these patients
`are listed in Appendix Table A1 (online only).
`Statistical Considerations
`The primary objective was to determine the rate of patients demonstrat-
`ing a ⱖ 50% decline in PSA after 12 weeks of treatment with single-agent
`abiraterone acetate 1,000 mg. A two-stage, attained phase II trial design was
`used30 in the phase II component of the study. The null hypothesis was a
`
`www.jco.org
`
`© 2009 by American Society of Clinical Oncology
`Downloaded from jco.ascopubs.org on October 29, 2014. For personal use only. No other uses without permission.
`Copyright © 2009 American Society of Clinical Oncology. All rights reserved.
`
`3743
`
`

`

`Attard et al
`
`ⱖ 50% PSA decline rate in 10% of patients; a decline in 30% of patients was the
`alternative hypothesis. With 42 assessable phase II patients, this study had an
`86% power to detect a ⱖ 50% decline in PSA in 30% of patients, with a
`one-sided ␣ level of 5%. The first stage ended after the evaluation of 20
`patients, with continued accrual to the second stage if more than one patient
`had a ⱖ 50% decline in PSA.
`
`RESULTS
`
`Patient Characteristics
`Fifty-four CRPC patients were recruited to this phase I/II trial
`between December 13, 2005 and November 28, 2007,15 and six pa-
`tients continue on single-agent abiraterone acetate. Forty-two patients
`were treated at 1,000 mg (including nine patients in the phase I
`portion of the study treated at 1,000 mg); the other 12 patients were
`treated at 250, 500, 750, and 2,000 mg (three patients at each dose
`level). As defined a priori in this phase I/II trial protocol, all patients
`treated at 1,000 mg were included in this phase II antitumor activity
`analysis. Table 1 lists the demographics and clinical characteristics of
`
`Table 1. Baseline Patient Demographics and Clinical Characteristics
`of Patients Treated With Abiraterone Acetate 1,000 mg
`
`No. of Patients
`
`%
`
`Patient Demographics and
`Clinical Characteristics
`
`Age, years
`Median
`Range
`Gleason score (diagnostic biopsy)
`6
`7
`8
`9
`10
`Unknown
`Baseline PSA, ng/mL
`Median
`Range
`ECOG performance status
`0
`1
`Site of baseline metastases
`Increasing PSA only
`Lymph nodes only
`Visceral lesions only
`Lymph nodes ⫹ visceral
`Bone lesions only
`Bone ⫹ lymph nodes
`Bone ⫹ lymph nodes ⫹ visceral
`LDH
`Raised
`Normal
`Prior hormone therapy
`LHRH analogues
`Antiandrogens
`Dexamethasone
`Diethylstilboestrol
`Ketoconazole
`
`the 42 patients treated at 1,000 mg. The median number of lines of
`prior hormonal treatments was three. All patients had experienced
`progression on LHRH analogs, and 41 of 42 patients had experienced
`progression on antiandrogens; one patient had experienced progres-
`sion on LHRH analogs and diethylstilboestrol but had not received an
`antiandrogen. Fourteen (33%) of 42 patients had experienced pro-
`gression on dexamethasone, and 20 (48%) of 42 patients had experi-
`enced progression on diethylstilboestrol, including seven (17%) of 42
`patients on both. Two patients had experienced progression on keto-
`conazole. Twenty-four patients (57%) had measurable disease on CT
`scan at baseline, and 32 (76%) of 42 patients had bone metastasis on
`bone scan; 17 (40%) of 42 patients had ⱖ 5 CTCs/7.5 mL at baseline.
`
`PSA, CTC Count, and Radiologic Evidence for
`Antitumor Activity
`A decline in PSA of ⱖ 50% was observed in 28 (67%) of 42
`patients, and 30 (71%) of 42 and eight (19%) of 42 patients had a
`decline in PSA of ⱖ 30% and ⱖ 90%, respectively. The change in PSA
`from baseline at 12 weeks and the maximal change at any time point
`after 12 weeks are presented in Figure 1. One of the two patients who
`had previously experienced progression on ketoconazole and hydro-
`cortisone had a decline in PSA from 79 to 10.9 ng/mL, which is
`ongoing after 230 days of treatment. By RECIST criteria, 24 patients
`had measurable disease on CT scan, with nine patients (37.5%) having
`tumor regression that constituted a partial response (Fig 2); overall, 16
`patients (66%) had no evidence of progression at 6 months. Further-
`more, 10 (59%) of 17 patients had a decline in CTC count from ⱖ 5 to
`less than 5/7.5 mL, and 12 (70%) of 17 patients had a decline of ⱖ 30%
`after starting treatment with abiraterone acetate.
`
`20
`10
`0
`-10
`-20
`-30
`-40
`-50
`-60
`-70
`-80
`-90
`-100
`
`20
`10
`0
`-10
`-20
`-30
`-40
`-50
`-60
`-70
`-80
`-90
`-100
`
`A
`
`PSA Change at 12 Weeks (%)
`
`B
`
`PSA Change (maximal %)
`
`100
`98
`33
`48
`5
`
`70
`50-84
`
`110
`9.7-964
`
`8
`10
`16
`4
`1
`3
`
`20
`22
`
`4
`4
`1
`1
`14
`12
`6
`
`13
`29
`
`42
`41
`14
`20
`2
`
`Abbreviations: ECOG, Eastern Cooperative Oncology Group; PSA, prostate-
`specific antigen; LDH, lactate dehydrogenase; LHRH, luteinizing hormone–
`releasing hormone.
`
`Fig 1. Changes in prostate-specific antigen (PSA) with abiraterone acetate.
`Waterfall plot showing maximal percentage change in PSA from baseline with
`single-agent abiraterone acetate 1,000 mg at (A) 12 weeks and (B) any time point
`after 12 weeks. The dashed lines indicate PSA declines of 30%, 50%, and 90%.
`
`3744
`
`JOURNAL OF CLINICAL ONCOLOGY
`© 2009 by American Society of Clinical Oncology
`Downloaded from jco.ascopubs.org on October 29, 2014. For personal use only. No other uses without permission.
`Copyright © 2009 American Society of Clinical Oncology. All rights reserved.
`
`

`

`Phase II Trial of Abiraterone Acetate
`
`A
`
`C
`
`E
`
`B
`
`D
`
`F
`
`(A) Patient 203 had
`Fig 2. Radiologic responses to abiraterone acetate.
`previously experienced progression on luteinizing hormone–releasing hormone
`(LHRH) agonists, bicalutamide, diethylstilboestrol, and the histone deacetylase
`inhibitor FK228. Before starting treatment with abiraterone acetate, he had a
`prostate-specific antigen (PSA) level of 200 ng/mL and multiple liver metastases
`but no bone metastasis on bone scan. The largest lesion in the liver was 102 mm.
`(B) After 12 months of treatment, his PSA had declined to 11.2 ng/mL, and his
`index lesion measured 46 mm. (C) Patient 214 had previously experienced
`progression on LHRH agonists and bicalutamide. Before starting treatment, he
`had a PSA of 129 ng/mL, mediastinal lymphadenopathy measuring 21 mm, and
`bone metastasis on bone scan. He also had bilateral pleural effusions indicated by
`the red arrows. (D) After 12 months of treatment, his PSA was 7.4 ng/mL, and his
`mediastinal
`lymphadenopathy measured 11 mm. His bone disease remained
`stable on bone scan, and the pleural effusions had resolved. (E) Patient 227 had
`previously experienced progression on LHRH agonists and bicalutamide. Before
`treatment, he had a PSA of 870 ng/mL and retroperitoneal lymphadenopathy
`measuring 15 mm. (F) After 9 months of treatment, his retroperitoneal lymph-
`adenopathy measured 6 mm. His PSA had declined to 471 ng/mL but then
`increased to 1,334 ng/mL, necessitating addition of dexamethasone 0.5 mg daily.
`This resulted in a secondary decline in PSA to 820 ng/mL, and he continues on
`treatment after 12 months.
`
`Safety Profile
`A syndrome of secondary mineralocorticoid excess characterized
`by hypokalemia (37 of 42 patients; 88%), hypertension (17 of 42
`patients; 40%), and fluid overload (13 of 42 patients; 31%) was re-
`ported. This was managed by eplerenone 50 to 200 mg daily except in
`three patients who required glucocorticoid replacement for symp-
`tomatic fluid overload; in two patients, this was associated with mi-
`grainous headaches. Magnetic resonance imaging of the brain was
`normal in both patients. Another patient with a prior history of
`asthma required high-dose corticosteroids for worsening asthma and
`was subsequently maintained on dexamethasone 0.5 mg daily. Hot
`flushes developed in four of 42 patients; venlafaxine was required to
`control symptoms in two patients. Two patients (5%) developed
`asymptomatic grade 3 transaminase elevation 10 weeks and 27 weeks
`
`after starting abiraterone acetate. Interruption of treatment resulted in
`complete resolution. One patient was rechallenged with abiraterone
`acetate 750 mg, and the transaminase elevation recurred, suggesting
`that this event was secondary to study treatment. Another patient
`suffered grade 2 asymptomatic transaminase elevation 16 weeks after
`starting abiraterone acetate 1,000 mg, but this resolved after tempo-
`rary discontinuation of treatment, and a dose of 750 mg was then
`administered uneventfully. Four patients developed grade 1 head-
`aches, and five patients complained of grade 1 joint aches; these symp-
`toms occurred intermittently on abiraterone acetate and did not
`necessitate treatment interruption. No other adverse events consid-
`ered related to abiraterone acetate of grade ⱖ 2 severity or occurring in
`two or more patients were reported.
`
`TTPP
`The median follow-up time for the 42 patients treated at 1,000
`mg was 505 days (95% CI, 457 to 552 days), with a median TTPP on
`abiraterone acetate alone of 225 days (95% CI, 162 to 287 days; Fig
`3A). The median TTPP for the patients treated at 1,000 mg who had
`a ⱖ 50% decline in PSA was 253 days (95% CI, 122 to 383 days), and
`the median TTPP for the patients who had a ⱖ 90% decline in PSA
`was 393 days (95% CI, 252 to 533 days). Fourteen patients were
`censored for this analysis; five patients continued on single-agent
`abiraterone acetate at the date of data cutoff, and nine patients started
`dexamethasone or stopped abiraterone acetate without confirmed
`PSA progression after 12 weeks (Appendix Table A1). The median
`follow-up time for all 54 phase I/II patients who received any amount
`of study medication was 588 days (95% CI, 409 to 767 days), with a
`median TTPP on abiraterone acetate alone of 229 days (95% CI, 157 to
`301 days) and median TTPP for patients who had a ⱖ 50% and ⱖ 90%
`decline in PSA of 339 days (95% CI, 136 to 542 days) and 477 days
`(95% CI, 350 to 604 days), respectively. Sixteen patients were censored
`for this analysis; six patients continued on single-agent abiraterone
`acetate at the date of data cutoff, and 10 patients started dexametha-
`sone or stopped abiraterone acetate without confirmed PSA progres-
`sion after 12 weeks (Appendix Table A1).
`
`Addition of Corticosteroids at Progression on
`Abiraterone Acetate Alone
`To date, 39 of 54 patients in this phase I/II study have received
`abiraterone acetate in combination with dexamethasone 0.5 mg daily;
`14 continued on this combination at date of data cutoff (Appendix
`Table A1). The median TTPP from addition of dexamethasone until
`stopping both abiraterone acetate and dexamethasone (n ⫽ 39) was
`151 days (95% CI, 117 to 185 days). Thirty patients received the
`combination for ⱖ 12 weeks after progression on abiraterone acetate
`alone and were thus assessable for magnitude of PSA decline; 24
`patients experienced progression on the combination (Appendix
`Table A1). Eleven of these 30 patients had previously experienced
`progression on dexamethasone (group I), and 19 of 30 were dexa-
`methasone naive (group II); four (36%) of 11 patients in group I and
`six (32%) of 19 patients in group II had ⱖ 50% declines in PSA
`(Appendix Fig A1, online only). The median TTPP for all 54 patients
`from start of abiraterone acetate until PSA progression on the combi-
`nation (n ⫽ 25) or on abiraterone acetate alone if dexamethasone was
`not used (n ⫽ 3) was 420 days (95% CI, 259 to 580 days; Fig 3B); the
`median TTPP was 372 days (95% CI, 55 to 688 days) for patients who
`previously experienced progression on dexamethasone (group I) and
`
`www.jco.org
`
`© 2009 by American Society of Clinical Oncology
`Downloaded from jco.ascopubs.org on October 29, 2014. For personal use only. No other uses without permission.
`Copyright © 2009 American Society of Clinical Oncology. All rights reserved.
`
`3745
`
`

`

`Attard et al
`
`Table 2. Association of Pretreatment Serum Investigations
`(continuous variable) in Patients Who Received Any Amount of
`Abiraterone Acetate (N ⫽ 54) With Probability of a ⱖ 50% PSA Decline
`
`Serum Marker
`
`DHEA
`DHEA-S
`Androstenedione
`Estradiol
`Alkaline phosphatase
`PSA
`Lactate dehydrogenase
`
`OR
`
`4.84
`3.6
`4.64
`5.15
`0.71
`0.83
`0.45
`
`95% CI
`
`1.47 to 15.97
`1.14 to 11.38
`1.5 to 14.35
`1.33 to 19.97
`0.27 to 1.86
`0.46 to 1.48
`0.05 to 3.99
`
`P
`
`.01
`.029
`.008
`.018
`.492
`.519
`.477
`
`NOTE. OR applies to a unit log change in value. Unacceptably high levels of
`discordance between the three super-sensitive assays used for testosterone
`measurement (radioimmunoassay ⫻1 and liquid chromatography/mass spec-
`trometry/mass spectrometry ⫻2) led to these results not being included in
`these analyses.
`Abbreviations: PSA, prostate-specific antigen; OR, odds ratio; DHEA, dehy-
`droepiandrostenedione; DHEA-S, dehydroepiandrostenedione-sulfate.
`
`DISCUSSION
`
`This study confirms that selective inhibition of CYP17 with abi-
`raterone acetate results in declines in PSA, radiologic responses, and
`declines in CTC counts in patients with hormone-dependent CRPC.
`Abiraterone acetate 1,000 mg was administered daily continuously to
`42 patients in this study, resulting in ⱖ 50% PSA declines in 67% of
`patients. The rates of ⱖ 50% PSA declines in prospective phase II
`studies of ketoconazole in chemotherapy-naive patients ranged from
`31% to 62%, and the median duration of response was up to 7.7
`months.31-35 All of the ketoconazole studies required hydrocortisone
`replacement therapy, which may have contributed to the antitumor
`activity36 and described grade ⱖ 3 toxicities in up to 30% of patients.
`Although it is difficult to make any comparisons, the patients in our
`study had received more lines of hormone treatment and had more
`advanced disease. Furthermore, abiraterone acetate administered
`without concomitant glucocorticoids can be well tolerated, although
`monitoring and immediate treatment of secondary mineralocorticoid
`excess is important. Androstenedione is the sole hormone reported to
`
`Table 3. Association of Pretreatment Serum Investigations
`(continuous variable) in Patients Who Received Any Amount of Abiraterone
`Acetate (N ⫽ 54) With TTPP on Abiraterone Acetate and Dexamethasone
`
`Serum Marker
`
`DHEA
`DHEA-S
`Androstenedione
`Estradiol
`Alkaline phosphatase
`PSA
`Lactate dehydrogenase
`
`HR
`
`0.57
`0.71
`0.5
`0.38
`0.95
`1.32
`1.1
`
`95% CI
`
`0.34 to 0.98
`0.51 to 1.0
`0.32 to 0.98
`0.2 to 0.71
`0.55 to 1.62
`0.97 to 1.8
`0.33 to 3.63
`
`P
`
`.04
`.05
`.002
`.003
`.84
`.08
`.87
`
`NOTE. HR applies to a unit log change in value. Unacceptably high levels of
`discordance between the three super-sensitive assays used for testosterone
`measurement (radioimmunoassay ⫻1 and liquid chromatography/mass spec-
`trometry/mass spectrometry ⫻2) led to these results not being included in
`these analyses.
`Abbreviations: TTPP, time to prostate-specific antigen progression; HR,
`hazard ratio; DHEA, dehydroepiandrostenedione; DHEA-S, dehydroepiandro-
`stenedione-sulfate; PSA, prostate-specific antigen.
`
`1.00
`
`0.75
`
`0.50
`
`0.25
`
`A
`
`Progression-Free Survival (%)
`
`0
`
`120
`
`240
`360
`480
`600
`Time Since Treatment (days)
`
`720
`
`42
`
`25
`
`12
`
`8
`
`1
`
`1
`
`1
`
`No. at risk:
`
`1.00
`
`0.75
`
`0.50
`
`0.25
`
`B
`
`Progression-Free Survival (%)
`
`0
`
`120
`
`240
`360
`480
`600
`720
`Time Since Start of Treatment (days)
`
`No. at risk:
`
`54
`
`42
`
`36
`
`26
`
`15
`
`5
`
`4
`
`Fig 3. Time on treatment with abiraterone acetate. (A) Kaplan-Meier (KM) plot
`showing time to prostate-specific antigen progression (TTPP) on abiraterone
`acetate alone for 42 patients treated at 1,000 mg. (B) KM plot for TTPP for all
`patients (N ⫽ 54) after progression on addition of dexamethasone (or abiraterone
`acetate alone if dexamethasone was not added).
`
`361 days (95% CI, 241 to 480 days) for dexamethasone-naive patients
`(group II). The remaining 26 patients were censored at the date of data
`cutoff either because they continued on treatment (n ⫽ 20) or they
`had stopped treatment without confirmed criteria for PSA progres-
`sion (n ⫽ 6; Appendix Table A1).
`
`Predictors of Response to Abiraterone Acetate
`All 54 patients who received any amount of single-agent abi-
`raterone acetate were included in this analysis. Pretreatment levels of
`DHEA, DHEA-S, androstenedione (continuous variable), and estra-
`diol were associated with increased probability of a ⱖ 50% PSA de-
`cline on abiraterone acetate and with median TTPP calculated from
`start of abiraterone acetate until PSA progression on dexamethasone
`and abiraterone acetate (or on abiraterone acetate alone if dexameth-
`asone was not added; Tables 2 and 3).
`
`3746
`
`JOURNAL OF CLINICAL ONCOLOGY
`© 2009 by American Society of Clinical Oncology
`Downloaded from jco.ascopubs.org on October 29, 2014. For personal use only. No other uses without permission.
`Copyright © 2009 American Society of Clinical Oncology. All rights reserved.
`
`

`

`Phase II Trial of Abiraterone Acetate
`
`significantly predict a decline in PSA by ⱖ 50% in a series of 103
`patients treated with ketoconazole.38 We report that DHEA, DHEA-S,
`and estradiol, in addition to androstenedione, were associated with
`ⱖ 50% declines in PSA and TTPP, although the relatively small size of
`this study implies that only strong relationships would be identified,
`and none of the outcomes evaluated is significant if a correction for
`multiple testing, requiring P ⬍ .006, is used.
`We have not previously observed and, to our knowledge, there
`are no published reports of secondary responses to reinstitution of
`single-agent dexamethasone in patients who had previously experi-
`enced progression on this therapy. In this trial, 33% of patients had a
`secondary ⱖ 50% PSA decline on addition of dexamethasone to
`abiraterone acetate regardless of prior dexamethasone exposure.
`These data suggest that AR may be activated by elevated hormone
`levels upstream of CYP17 and supports the future evaluation of a
`combination of abiraterone acetate with low-dose corticosteroids to
`maximize efficacy and minimize toxicity. Abiraterone acetate is now
`being evaluated in combination with corticosteroids in a 1,180-
`patient, multicenter, double-blind, randomized phase III study com-
`paring abiraterone acetate plus prednisone versus prednisone plus
`placebo in CRPC patients who have previously received docetaxel.
`The primary end point of this study is overall survival. This phase III
`study incorporates the prospective evaluation of whether CTC counts
`after treatment can serve as a robust intermediate end point for overall
`survival to accelerate new drug approval for CRPC. Phase III studies in
`the prechemotherapy setting are also planned. Overall, these data
`suggest that abiraterone acetate is an effective, well-tolerated treat-
`ment and confirm that a subgroup of CRPC patients continue to have
`hormone-driven disease.
`
`AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS
`OF INTEREST
`
`Although all authors