Published Ahead of Print on September 14, 2015 as 10.1200/JCO.2015.61.5831
`The latest version is at http://jco.ascopubs.org/cgi/doi/10.1200/JCO.2015.61.5831
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`Gig |S | A ©
`
`Matthew J. Ellis, Baylor College of
`Medicine, Houston, TX; Antonio
`Llombart-Cussac,, Hospital Arnau de
`Vilanova, Lérida, Spain; David Feltl,
`FNsP Ostrava, Ostrava-Poruba, Czech
`Republic; John A. Dewar, Ninewells
`Hospital and Medical School, Dundee;
`Nicola Hewson and Yun Rukazenkov,
`AstraZeneca Pharmaceuticals, Maccles-
`field; John F.R. Robertson, University of
`Nottingham, Derby, United Kingdom;
`and Marek Jasiowka, Instytut im Marit
`Sktodowskie}-Curie, Krakow, Poland
`Published online ahead of print at
`www jeo.org on September 14, 2015
`Supported by AstraZeneca.
`Terms in blue are defined in the glos-
`sary, found at the end ofthis article
`and online at www’ jco.org
`Presented at the 2014 San Antonio
`Breast Cancer Symposium, San Anto-
`nio, TX, December 9-13, 2014.
`
`Authors’ disclosures of potential
`conflicts of interest are found in the
`article ofline at www _jco.org. Author
`contributions are found at the end of
`this article.
`Clinical trial information: NCT00274469
`
`Corresponding author: Matthew J. Ellis,
`MD, Lester-and Sue Smith Breast
`Center, One Baylor Plaza, Baylor
`College of Medicine, Houston, TX
`77030; emai Matthew-Ellis@bern.edu
`
`© 2015 by American Society of Clinical
`Oncology. Licensed under the Creative
`CommonsAttribution 3.0 License
`
`0732-183X/15/3399-1/$20.00
`DOI: 10.1200/JCO.2015 61.5831
`
`Fulvestrant 500 mg Versus Anastrozole 1 mg for the
`First-Line Treatment of Advanced Breast Cancer: Overall
`Survival Analysis From the Phase II FIRST Study
`
`MatthewJ. Ellis,Antonio Llombart-Cussac, David Feltl, John A. Dewar, Marek Jasibwka, Nicola Hewson,
`Yuri Rukazenkov, and John FR. Robertson
`
`A
`
`B
`
`$
`
`T
`
`R
`
`A
`
`C
`
`T
`
`
`
`Purpose
`To compare overall survival (OS) for fulvestrant 500 mg versus anastrozole as first-line endocrine
`therapy for advanced breast cancer.
`
`Patients and Methods
`The Fulvestrant First-Line Study Comparing Endocrine Treatments (FIRST) was a phase Il,
`randomized, open-label, multicenter trial. Postmenopausal women with estrogen receptor—
`positive,
`locally advanced/metastatic breast cancer who had no previous therapy for advanced
`disease received either fulvestrant 500 mg (days 0, 14, 28, and every 28 days thereafter) or
`anastrozole 1 mg (daily). The primary end point (clinical benefit rate [72.5% and 67.0%]) and a
`follow-up analysis (median time to progression [23.4 months and 13.1 months]) have been
`reported previously for fulvestrant 500 mg and anastrozole,
`respectively. Subsequently,
`the
`protocol was amended to assess OS by unadjusted log-rank test after approximately 65% of
`patients had died. Treatment effect on OS across several subgroups was examined. Tolerability
`was evaluated by adverse event monitoring.
`
`Results
`In total, 205 patients were randomly assigned (fulvestrant 500 mg, n = 102: anastrozole, n = 103).
`At data cutoff, 61.8% (fulvestrant 500 mg, n = 63) and 71.8% (anastrozole, n = 74) had died. The
`hazard ratio (95% Cl) for OS with fulvestrant 500 mg versus anastrozole was 0.70 (0.50 to 0.98;
`P= 04: median OS, 54.1 months v48.4 months). Treatment effects seemed generally consistent
`
`across the subgroups analyzed. No new safety issues were observed.
`
`Conclusion
`There are several limitations of this OS analysis, including that it was not planned in the original
`protocol but instead was added after time-to-progression results were analyzed, and that not
`all patients participated in additional OS follow-up. However, the present results suggest
`fulvestrant 500 mg extends OS versus anastrozole. This finding now awaits prospective
`confirmation in the larger phase Ill FALCON (Fulvestrant and Anastrozole Compared in
`Hormonal Therapy Naive Advanced Breast Cancer)
`trial
`(ClinicalTrials.gov identifier:
`NCT01602380).
`
`J Clin Oncol © 2015 by American Society of Clinical Oncology. Licensed under the Creative
`Commons Attribution 3.0 License: http://creativecommons.org/licenses/by/3.0/
`
`
`
`De
`Tamoxifen and third-generation aromatase inhibi-
`tors (Als), such as anastrozole, exemestane, and
`letrozole are established first-line endocrine thera-
`
`pies for the treatment of postmenopausal women
`with estrogen receptor (ER) —positive, advanced
`breast cancer.’ Given the high prevalenceofresis-
`tance to Al therapy, multiple treatment options with
`distinct mechanismsofaction are desirable."
`
`Fulvestrant, a 17B-estradiol analog, is a selec-
`tive ER antagonist that suppresses estrogen signaling
`by binding to ER and inducing a conformational
`change.”® Dimerization is subsequently blocked,
`triggering accelerated degradation and downregula-
`tion of the ER protein.” Fulvestrant exhibits lack of
`cross-reactivity with tamoxifen. Consequently, pa-
`tients whose disease progresses on fulvestrant may
`retain sensitivity to treatment with further endo-
`crine therapies.”* Theclinical efficacy offulvestrant
`
`© 2015 by Arnerican Society of Clinical Oncology
`Information downloaded from jco.ascopubs.org and provided by at AZ Library on September 14, 2015 from 212.209.42.182
`Copyright © 2015 American Society of Clinical Oncology. All rights reserved.
`Copyright 2015 by American Society of Clinical Oncology
`
`1
`
`AstraZeneca Exhibit 2058 p. 1
`InnoPharma Licensing LLC v. AstraZeneca AB IPR2017-00905
`Fresenius-Kabi USA LLCv. AstraZeneca AB IPR2017-01912
`
`

`

`Ellis et al
`
`was initially demonstrated in two phase II] trials that compared ful-
`vestrant 250 mg per month with anastrozole 1 mgdaily as a second-
`line therapy for advanced breast cancer.*!° A combined analysis of
`these trials demonstrated that time to progression (TTP) with fulves-
`trant 250 mg was noninferior to anastrozole."
`Fulvestrant 250 mg wasnot proven to be superiorto tamoxifen in
`a double-blind, randomizedtrial.’? This findingwas unexpected given
`the superiority of anastrozole over tamoxifen'® and the comparable
`efficacy of anastrozole and fulvestrant 250 mg as second-line ther-
`apy.'? Pharmacokinetic modeling, as well as observations made dur-
`ing early clinical studies,’ suggested theefficacy of fulvestrant could
`be improved with use of a higher dose, whichled to the development
`of a dosage regimen of fulvestrant 500 mg, including a loading dose
`component to reduce the time to reach steady-state plasmalevels.
`Subsequently, the phase III Comparison of Faslodex in Recurrent
`or Metastatic Breast Cancer (CONFIRM) trial found that fulves-
`trant 500 mg was associated with improved progression-free sur-
`vival (PFS) and overall survival (OS) compared with the 250-mg
`dose in patients who experienced disease recurrence or progression
`after previous endocrine therapy.'*'°
`The Fulvestrant First-Line Study Comparing Endocrine Treat-
`ments (FIRST) was a phase II, randomized, open-label, multicenter
`trial that also used the fulvestrant 500-mg dose regimen, comparing
`efficacy and safety with anastrozole in the first-line setting. The pri-
`mary end pointofclinical benefit rate was noninferiorfor fulvestrant
`500 mg compared with anastrozole,’® with both treatments demon-
`strating similar, well-tolerated safety profiles. A follow-up analysis,
`performed because only 35.6% of patients experienced disease pro-
`gression at the time of the primary analysis, reported a hazard ratio
`(HR) of TTP forfulvestrant 500 mg versus anastrozole of 0.66 with a
`95% CI of 0.47 to 0.92 (P = .01; median TTP, 23.4 months v
`13.1 months). No additionalsafety issues were reported.'” Given the
`improvement in TTP observed during fulvestrant 500 mg treatment
`compared with anastrozole in this phaseII trial, a subsequent protocol
`amendment was made to address whether this apparent extension in
`disease control would translate into an improvementin OS.
`
`PATIENTS AND METHODS
`
`Study Design and Participants
`FIRST was a phase II, randomized, open-label, multicenter, parallel-
`group trial comparing fulvestrant 500 mg with anastrozole 1 mg, Postmeno-
`pausal women with ER-positive locally advanced. or metastatic breast cancer
`who had not received any previous systemic therapy for locally advanced or
`metastatic disease were included. Patients were permitted to havereceived
`previous endocrine therapy for early disease, providing this had been com-
`pleted more than 12 months before random assignment. This trial was con-
`ducted in accordance with the Declaration ofHelsinki, was consistent with the
`International Conference on Harmonisation—GoodClinical Practice guide-
`lines, and is registered with Clinicaltrials.gov. All patients provided written,
`informed consent. Full detailsofthis trial have been reported previously.‘°'”
`
`Random Assignment and Procedures
`Eligible patients were randomly assigned sequentially 1:1 to either fulves-
`trant 500 mg (administered intramuscularly on days 0, 14, 28, and every
`28 days thereafter) or anastrozole 1 mg (administered orally once per day). The
`data cutoff for the primary analysis was 6 monthsafter the last patient was
`randomlyassigned. On disease progressionorafter data cutofffor the primary
`analysis,all patients entered a follow-up phase after a protocol amendment for
`
`an analysis of TTP. The TTP follow-up required a questionnaire to be com-
`pleted for each patient 12 monthsafter the patient enteredthe follow-up phase
`and every 12 monthsthereafter for patients continuing to receive randomized
`treatment. After the TTP analysis was performed, a further protocol amend-
`ment was developed to enter patients into an optional follow-up phase to
`establish OS. To ensure sufficient maturity, the OS analysis was planned for
`when approximately 65%of patients had died. Patients whodid notcontrib-
`ute additional data to the follow-up extension were right-censored at the last
`known date they werealive, and their data until this point were included in the
`analysis. Sites were invited to request written consent from patients for the
`collection of additional data. Patients were contacted every 3 months until the
`first ofthe following events: death, patientwithdrawal, data cutoffwas reached,
`or the patient was lostto follow-up. Patients with a last known survival status of
`alive were contacted within 2 weeks ofdata cutofftoensure they werestill alive.
`
`Outcomes
`The primary study end point wasclinical benefit rate; secondary end
`points included objective response rate, TTP, durationofclinical benefit, and
`duration of response. These primary and secondary end points have been
`reported previously.1°!”
`The follow-up analysis assessed OS, defined as the time from being
`randomlyassigned to death from any cause. A log-ranktest. (unadjusted model
`with treatmentfactor only) was performedfor the primary analysis ofOS. HRs
`with 95% Cls were used to compare fulvestrant 500 mg with anastrozole; no
`adjustments were made for multiplicity. A statistical significance level of .05
`was used to indicate a difference in OS between the treatment groups. For
`patients for whom follow-up responses could not be obtained, data were
`censoredat the date the patient was last knownto bealive.
`Exploratory subgroup analyses were conducted using the log-ranktest to
`compare OS for the following prespecified patient subgroups:
`less than
`65 years of age versus 65 years of age or greater; not positive for both ER and
`progesterone receptor versuspositive for both ER and progesteronereceptor;
`no visceral involvement versusvisceral involvement; no previous chemother-
`apy versus previousadjuvant chemotherapy; no measurable disease versus
`measurable disease; and no previous endocrine therapyversus previous endo-
`crine therapy.
`Twosensitivity analyses were performed to examine any potential
`impact of nonparticipation on OSresults: a Kaplan-Meier OS analysis was
`performed in which the censoring indicator was reversed; and baseline
`covariates were assessed for patients censored greater than 3 months before
`data cutoff and for those censored 3 months or less before data cutoff,
`which correspondsto patients who did notparticipate in the OS follow-up
`and to those who did, respectively.
`Tolerability was assessed by serious adverse event (SAE) monitoring.All
`SAEs were coded in compliance with the Medical Dictionary for Regulatory
`Activities and recorded in an internal AstraZeneca database for evaluation.
`
`SAEs were monitored for up to 8 weeksafter the last dose offulvestrant 500 mg
`or for30 days afterthe last dose of anastrozole.
`
`RESULTS
`
`In total, 205 patients were randomlyassigned to receive fulvestrant
`500 mg (n = 102) or anastrozole 1 mg (n = 103) at 62 centers in nine
`countries (Brazil, Bulgaria, the Czech Republic, France, Italy, Poland,
`Spain, the United Kingdom, and the United States).
`Baseline characteristics and patient demographics were similar
`between the treatment groupsas reported previously.'° The propor-
`tion of patients who had not received previous endocrine treatment
`for early disease was similar for the fulvestrant 500 mg and anastrozole
`treatment groups (71.6% and 77.7% of patients at baseline, respec-
`tively). Of those that did, almost all had received tamoxifen exclu-
`sively. Of the 205 randomlyassigned patients, 35 (16 in the fulvestrant
`500 mg group and 19 in the anastrozole group) did not participate in
`
`2 ©2015 by American Society of Clinical Oncology
`Information downloaded from jco.ascopubs.org and provided by at AZ Library on September 14, 2015 from 212.209.42.182
`Copyright © 2015 American Society of Clinical Oncology. All rights reserved.
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`AstraZeneca Exhibit 2058 p. 2
`
`

`

`Fulvestrant 500 mg: Overall Survival Versus Anastrozole
`
`Enrolled
`(N = 233)
`
`Not randomly allocated
`Incorrect enrollment
`Death
`Adverseevent
`Voluntary patient discontinuation
`Other
`
`(n= 28)
`(n = 20)
`(n=1)
`(n= 1)
`(n= 4)
`(n= 2)
`
`(n = 10)
`
`Randomly allocated
`(n = 205)
`
`>,
`
`Fig 1. Study overview. (*) These patients
`were right censored at the time of their last
`known date alive, and data until this point
`were used in the overall survival (OS)analysis.
`
`Fulvestrant 500 mg
`(n = 102)
`
`Anastrozole 1 mg
`(n = 103)
`
`Datacutoff for analysis of overall survival
`Alive
`(n = 23)
`Dead
`(= 63)
`Did not contribute additional data
`(n= 16)
`during OS follow-up extension*
`Patient declined to participate
`Site declinedto participate
`
`(n= 6)
`
`Data cutoff for analysis of overall survival
`Alive
`{n = 10)
`Dead
`(n = 74)
`Did not contribute additional data
`(n = 19)
`during OSfollow-up extension®
`Patient declined to participate
`Site declined to participate
`
`(n= 9)
`(n = 10)
`
`Safety
`The occurrence of SAEs during the main study period and the
`follow-up period combined is detailed in Table 2. The majority of
`SAEs were considered by the investigator to be unrelated to the treat-
`ment. Two SAEs consideredto be treatment related were documented.
`(one case ofhypertension and onecase of pulmonary embolism,both
`in the fulvestrant 500 mg treatment group).
`
`DISCUSSION
`
`This study reports improved OS with fulvestrant 500 mg treatment
`compared. with anastrozole in the first-line setting for ER-positive
`
`= Fulvestrant 500 mg
`~~ Anastrozole 1mg
`
`the OS follow-up phase and were censoredat the date they were last
`known to be alive; for these patients, data until this time are
`included in the OS analysis, and thus all patients contributed data
`to the analysis. The majority of the nonparticipating patients (n =
`20) did not contribute additional data because they attended cen-
`ters that declined to contribute to the OS follow-up phase. An
`additional 15 individual patients from nine participating centers
`did not consent to follow-up. No patients participating in the OS
`phase werelost to follow-up, and the survival status at data cutoff
`was knownfor all patients consenting to the OS follow-up.
`
`Efficacy
`At the time ofthe follow-up analysis for OS, 63 of 102 patients in
`the fulvestrant 500 mg group (61.8%) and 74 of 103 patients in the
`anastrozole group (71.8%) were known to have died (Fig 1). The
`primary analysis of OS was improvedin the fulvestrant 500 mg group
`compared with anastrozole 1 mg; the HR was 0.70 (95% CI, 0.50 to
`0.98; log-rank test P = .04; median OS, 54.1 months v 48.4 months;
`Fig 2). The HRfor fulvestrant 500 mg versus anastrozole was found to
`be generally consistent across all subgroup analyses (Fig 3). At 3 years,
`64% (fulvestrant 500 mg) and 58% (anastrozole) of patients were
`eventfree; at 5 years, the equivalent values were 47% and 38%.
`
`Sensitivity Analyses
`There were no important differences between the treatment
`groups in time to censoring (data not shown). Furthermore, when key
`baseline covariates for patients censored within the last 3 months
`before data cutoff and for those censored more than 3 months before
`Time (months)
`No,at risk
`data cutoff were summarized, there were no important differences
`90 84 77
`By
`47
`39
`31
`24
`Fulvestrant 500 mg 102
`90 80 72
`49
`39
`29
`21
`14
`between treatment groups, indicating that the results were not caused
`Anastrozole 11mg=103
`by differences between patients who did and did not consent to OS
`follow-up (Table 1).
`
`
`
`
`
`OverallSurvival
`
`2 Cor)
`
`(proportion) 5
`
`So nN
`
`Median overall survival:
`Fulvestrant 500 mg: 54.1 months
`Anastrozole 1 mg: 48.4 months
`Hazard ratio, 0.70; 95% Cl, 0.50 to 0.98; P=.04
`
`6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96102
`
`WWW.jCO. OTe
`
`© 2015 by Arnerican Society of Clinical Oricology
`Information downloaded from jco.ascopubs.org and provided by at AZ Library on September 14, 2015 from 212.209.42.182
`Copyright © 2015 American Society of Clinical Oncology. All rights reserved.
`
`3
`
`AstraZeneca Exhibit 2058 p. 3
`
`Fig 2. Kaplan-Meier plot of overall survival.
`
`

`

`Ellis et al
`
`Fulvestrant Anastrozole
`Hazardratio
`500 mg
`1mg
`
`events (n) (95% Cl) events (n)
`Hazard ratio and 95% Cl
`
`0.70 (0.50 to 0.98)
`All patients
`63 (102)
`74 (103)
`Age, years
`< 65
`265
`
`0.73 (0.44 to 1.24)
`0.68 (0.44 to 1.06)
`
`29 (45)
`34 (57)
`
`Favors fulvestrant 500 mg
`
`Both ER+ and PgR+
`No
`Yes
`
`Visceral involvement
`No
`Yes
`
`Prior chemotherapy
`
`Measurable disease
`No
`Yes
`
`Prior endocrine therapy
`No
`Yes
`
`14 (24)
`49 (78)
`
`29 (54)
`34 (48)
`
`43 (73)
`20 (29)
`
`11 (13)
`52 (89)
`
`44 (73)
`19 (29)
`
`26 (45)
`48 (58)
`
`57 (78)
`17 (25)
`
`7 (10)
`67 (93)
`
`59 (80)
`15 (23)
`
`0.25
`
`0.66 (0.33 to 1.32)
`0.72 (0.49 to 1.06)
`
`0.68 (0.40 to 1.18)
`0.86 (0.56 to 1.34)
`
`0.63 (0.43 to 0.94)
`0.93 (0.48 to 1.78)
`
`NC
`0.67 (0.46 to 0.96)
`
`NC
`——_e——_ }
`
`é
`0.63 (0.42 to 0.93)
`—__—___+—_———— 1.01 (0.51 to 1.99)
`
`0.50
`
`1.00
`
`2.00
`
`Favors anastrozole
`
`Fig 3. Overall survival subgroup analy-
`sis. ER+, estrogen receptor positive;
`NC, not calculable: PgR+, progesterone
`receptor positive.
`
`advanced breast cancer, with an approximately 30% reduction in
`mortality risk. The previously reported improvements in TTP have
`translated into an improvement in OS of approximately 6 months
`with fulvestrant 500 mg (54.1 months) compared with anastrozole
`(48.4 months). This OS advantage is consistent with the OS benefit for
`fulvestrant 500 mg versus 250 mg in the second-line setting in the
`CONFIRM trial.'° Theeffect offulvestrant 500 mg on OS was gener-
`ally consistent acrossall prespecified subgroups (Fig 3). Furthermore,
`
`no new safety or tolerability issues were reported from the OS
`follow-up phase of this study, consistent with previously reported
`safety data,'°'7
`The improved OSwith fulvestrant 500 mg (54.1 months) relative
`to anastrozole (48.4 months) was observed although the median OS
`for the anastrozole group in this study washigher than has previously
`been reported. For example, OS of 39.2 months was reported for
`anastrozole asfirst-line endocrine therapy for advanced breast cancer
`
`
`
`Table 1. Baseline Covariates and Subgroups by Patients Censored = 3 Months and = 3 Months Before DCO
`No, of Patients (%)
`Censored > 3 Months Before DCO Censored = 3 Months Before DCO
`
`
`
`
`Subgroup
`Fulvestrant 500 rng (n = 16)
`Anastrozole 1 mg (n = 19)
`Fulvestrant 500 mg (n = 23)
`Anastrozole 1 mg (n = 10)
`
`Age, years
`< 65
`= 65
`Receptor status at diagnosis
`ot both ER+ and PgR+
`Both ER+ and PgR+
`Visceral involvement
`°
`Yes
`Previous chemotherapy
`°
`Yes
`Measurable disease at diagnosis
`°
`Yes
`Previous endocrine therapy
`8 (80.0)
`18 (78.3)
`3 (68.4)
`11 (68.8)
`3
`
`Yes 2 (20,0) 5 (81,3) 6 (31.6) 6 (21.7)
`
`
`
`
`5 31.3)
`11 (68.8)
`
`6 (37.5)
`10 (62.5)
`
`9 (66.3)
`7 (43.8)
`
`11 (68.8)
`5 (31.3)
`
`1 (6.3)
`18 193.8)
`
`7 (36.8)
`2 (63,2)
`
`5 (26.3)
`4 (73.7)
`
`1 (67.9)
`8 (42.1)
`
`3 (68.4)
`6 (31.6)
`
`3 (15.8)
`6 (84.2)
`
`
`
`11 (47.8)
`12 (52.2)
`
`4 (17,4)
`19 (82.6)
`
`16 (69,6)
`7 (30.4)
`
`19 (82.6)
`4 (17.4)
`
`1 (4,3)
`22 (95.7)
`
`4 (40.0)
`6 (60.0)
`
`2 (20.0)
`8 (80.0)
`
`8 (80.0)
`2 (20.0)
`
`8 (80.0)
`2 (20,0)
`
`0
`10 (100.0)
`
`
`
`
`
`Abbreviations: DCO, data cutoff; ER+, estrogen receptor—positive; PgR+, progesterone receptor—positive.
`
`JOURNAL OF CLINICAL ONCOLOGY
`4 ©2015 by American Society of Clinical Oncology
`Information downloaded from jco.ascopubs.org and provided by at AZ Library on September 14, 2015 from 212.209.42.182
`Copyright © 2015 American Society of Clinical Oncology. All rights reserved.
`
`AstraZeneca Exhibit 2058 p. 4
`
`

`

`Fulvestrant 500 mg: Overall Survival Versus Anastrozole
`
`
`
`Table 2. Incidence of SAEs and Deaths
`
`Fulvestrant
`Anastrozole
`500 mg
`1 mg
`
`SAE
`(n = 101)
`(n = 103)
`
`No, of Patients (%)
`
`24 (23.8)
`3 (3.0)
`21 (20.8)
`2 (2.0)
`
`22 (21.4)
`5 (4.9)
`18(17.5)
`0
`
`Any SAE
`Any SAE related to death
`Any SAE with outcome other than death
`Any causally related SAE
`Most commonly reported (= two patients)
`SAEs
`1 (1.0)
`1 (1.0)
`Atrial fibrillation
`0
`2 (2.0)
`Cardiac failure
`2.(1.9)
`0
`Death
`0
`2 (2.0)
`Decreased appetite
`0
`2 (2.0)
`Dehydration
`0
`2 (2.0)
`Dyspnea
`2(1.9)
`1 (1.0)
`Fernur fracture
`1 (1.0)
`1 (1.0)
`Neuralgia
`
`Transient ischemic attack 2 (1,9) 0
`
`
`Abbreviation: SAE, serious adverse event.
`
`in a combined analysis of two phase III studies,'® and OS of 41.3
`months wasreported for the anastrozole monotherapy arm of a phase
`III combination study.’*In addition, corresponding median OSval-
`ues of34.0 months (letrozole)”° and 37.2 months (exemestane)”' have
`been reported for other Als. It is therefore unlikely that the present
`analysis overestimates the margin of improvement with fulvestrant
`500 mg over anastrozole, which might have been possible had the
`control arm underperformed.
`The role of fulvestrant 500 mgas first-line therapy will be further
`defined by the ongoing phase III, double-blind FALCON (Fulvestrant
`and Anastrozole Compared in Hormonal Therapy Naive Advanced
`Breast Cancer) trial (ClinicalTrials.gov identifier. NCT01602380). The
`FALCONtrialwill assess the efficacy offulvestrant 500 mg versus anastro-
`zole in womenwithlocally advanced or metastatic breast cancer with strict
`definitions of endocrine therapy—naive disease, includingrestrictions on
`exposure to hormonereplacementtherapy.
`Endocrine therapy—naive advancedbreast canceris relatively un-
`common in countries with advanced health care, but represents a
`numerically substantial patient population, given the high disease
`prevalence. Furthermore, in unscreened populations and in develop-
`ing countries, metastatic disease at presentation is a significant prob-
`lem. Recent clinical trials reporting on first-line endocrine therapy in
`patients with ER-positive breast cancer have contained a substantial
`proportion, and often a majority, of endocrine therapy—naive
`patients."°°?4 In FIRST, previous endocrine therapy had been re-
`ceived by 29 (28.4%) of the patients treated with fulvestrant 500 mg
`and 23 (22.3%) of the anastrozole-treated patients. Of these 52 pa-
`tients, only 3 had received AI previously (2 in the anastrozole group and
`1 in the fulvestrant 500 mg group); the remainder hadreceived adjuvant
`tamoxifen. Therefore, AI resistance resulting from previous AI exposure
`cannot account for the observed OS difference. Indeed, hypothetically,
`previous exposure to tamoxifen may bias against fulvestrant as both
`agents are in the same therapeutic class. Upon disease progression, pa-
`tients were treated according to the standardofcare, and therefore, there
`could potentially be imbalances between the two treatment groups that
`
`could have affected the OS analysis. However, response to subsequent
`therapies (systemic chemotherapy or endocrine therapy) has previously
`been shown to be similar between the treatment groups, demonstrating
`that patients with disease progression on fulvestrant retain sensitivity to
`subsequent treatments.’” Differential second-line response, therefore,is
`also an unlikely explanation for the observed OSeffect.
`There are significant limitations to this report. The sample size was
`relatively small, and the OS analysis was not specified in the original
`protocol but was added as a hypothesis in a protocol amendmentafter
`TIP results were known. Furthermore, 35 patients did not contribute
`additional data to the OS follow-up; the decision notto participate in the
`extended follow-up for OS was madesolelyby the patientor participating
`center and was known at thestart ofthe OS follow-up and before the data
`were collected and analyzed. Data from these patients until the time of
`censoring were included in the OSanalysis, and similar censoring patterns
`were seen in the two treatment groups. The sensitivity analyses support
`the main findings, thatis, the differences in OS between treatment arms
`were unrelated to differences in censoring patterns. All-cause mortality
`wasused to determine OSin this. analysis because itis regarded. as the most
`unbiased and objective end point used in oncology.”This pointis partic-
`ularly relevant to an open-label study like FIRST. A final limitation was
`that the numberofpatients within subgroupswas relatively small. There-
`fore, care should be taken when interpreting results.
`Recent results from several trials with the cyclin-dependent ki-
`nase 4/6 (CDK4/6) inhibitor palbociclib are also pertinent to the
`discussion. PALOMA-1 (Palbociclib Ongoing Trials in the Manage-
`ment of Breast Cancer), a phaseII trial of letrozole plus palbociclib
`versus letrozole alone, provided provisional US Food and Drug Ad-
`ministration approval for palbociclib in the first-line setting on the
`basis of PFS.”* No positive OS data have been reported to date; the
`results ofa phaseII] trial ofthis comparison are pending (PALOMA-2,
`NCT01740427). Data from the phase III PALOMA-3 trial, comparing
`fulvestrant 500 mgplus palbociclib versus fulvestrant 500 mg alone in
`the second-line or subsequent setting in postmenopausal women(or
`pre- or perimenopausal women receiving goserelin), reported a
`marked PFS advantage for the combination, but OS data were also
`pending at the timeof publication.*° The medianPFS for fulvestrant
`500 mg alone was shorter in PALOMA-3 than in previous studies,
`indicative of the younger, higher-risk, and more heavily pretreated
`population recruited into the PALOMA-3 trial.
`The treatmentalgorithm for ER-positive advancedbreast cancer,
`therefore, is in a state of flux. Currently, it is rational to consider
`fulvestrant 500 mg asa first-line treatment option given the potential
`for survival benefits, particularly in settings where palbociclib is not
`available or palbociclib cost or adverseeffects are a significant concern,
`and especially if these results are confirmed in FALCON. These data
`also suggestthata first-line study offulvestrant 500 mg with aCDK4/6
`inhibitor versus fulvestrant 500 mgaloneis a logical proposition that
`could lead to further prolonged TTP. Recent preclinical data on the
`efficacy of an ER degrading agent with a CDK4/6 inhibitor in ESR1-
`mutant breast cancer provides further rationale for this population,
`because improvements in TTP or OS could be caused by suppression
`of ESR1-mutant Al-resistant clones.””
`In conclusion, we report that fulvestrant 500 mg maybeassoci-
`ated with improved OSversus anastrozole in the first-line setting for
`ER-positive advanced breast cancer. To our knowledge, this repre-
`sents the first time an endocrine monotherapy has demonstrated
`
`WWW.jCO. OTe
`
`© 2015 by Arnerican Society of Clinical Oncology
`Information downloaded from jco.ascopubs.org and provided by at AZ Library on September 14, 2015 from 212.209.42.182
`Copyright © 2015 American Society of Clinical Oncology. All rights reserved.
`
`5
`
`AstraZeneca Exhibit 2058 p. 5
`
`

`

`Ellis et al
`
`improvedefficacy compared with a third-generation Al. The phaseIII
`FALCONtrial may provide confirmation for these OS results; until
`then, the findings reported here should be regarded as preliminary,
`butclinically relevant.
`
`OSS
`
`AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS
`
`Disclosures provided by the authors are available with this article at
`www.jco.org.
`
`AUTHORCONTRIBUTIONS
`
`Conception and design: Matthew J. Ellis, John F.R. Robertson
`Provisionof study materials or patients: MatthewJ. Ellis, John F.R.
`Robertson
`Collection and assembly of data: MatthewJ. Ellis, David Feltl, John F.R.
`Robertson
`
`Data analysis and interpretation: Matthew J. Ellis, Antonio
`Llombart-Cussac, John A. Dewar, Marek Jasi6wka, Nicola Hewson, Yuri
`Rukazenkoy, John E.R. Robertson
`Manuscript writing: All authors
`Final approval of manuscript: All authors
`
`
`
`
`
`
`
`endocrine treatment. J Clin Oncol 20:3396-3403,
`20. Mouridsen H, Gershanovich M, Sun Y, ét al:
`2002
`Phase Ill study of
`letrozole versus tamoxifen as
`11. Robertson JFR, Osborne CK, Howell A,et al:
`first4ine therapy of advanced breast cancer in post-
`Fulvestrant versus anastrozole for the treatrnent of
`1. Cardoso F, Costa A, Norton L, et al: ESO-
`menopausal women: Analysis of survival and update
`ESMO 2nd International Consensus Guidelines for
`advanced breast carcinoma
`in postmenopausal
`of efficacy from the International Letrozole Breast
`advanced breast cancer (ABC2). Ann Oncol 25:1871-
`women: A prospective combined analysis of two
`Cancer Group. J Clin Oncol 21:2101-2109, 2003
`1888, 2014
`multicenter trials. Cancer 98:229-238, 2003
`21. Paridaens RJ, Dirix LY, Beex LV, et al: Phase
`12. Howell A, Robertson JFR, Abram P, et al:
`2. Cardoso F, Costa A, Norton L, et al: ESO-
`Ill study comparing exemestane with tamoxifen as
`ESMO 2nd International Consensus Guidelines for
`Comparison of fulvestrant versus tamoxifen for the
`first-line hormonal treatment of metastatic breast
`advanced breast cancer (ABC2). Breast J 23:489-
`treatment of advanced breast cancer in postmeno-
`cancer in postmenopausal women: The European
`502, 2014
`pausal women previously untreated with endocrine
`Organisation for Research and Treatment of Cancer
`3. Burstein HJ, Temin S, Anderson H, et al:
`therapy: A multinational, double-blind, randornized
`Breast Cancer Cooperative Group. J Clin Oncol
`trial. J Clin Oncol 22:1605-1613, 2004
`Adjuvant endocrine therapy for women with hor-
`26:4883-4890, 2008
`13. Howell A, Cuzick J, Baurn M, et al: Results of
`mone receptor—positive breast cancer: American
`22. Bergh J, Jonsson PE, Lidbrink EK, et al: FACT:
`the ATAC (Arimidex, Tamoxifen, Alone or in Combi-
`Society of Clinical Oncology Clinical Practice Guide-
`An open-label randomized phaseIII study of fulves-
`line Focused Update. J Clin Oncol 32:2255-2269,
`nation) trial after completion of 5 years’ adjuvant
`trant and anastrozole in combination compared with
`2014
`treatment for breast cancer. Lancet 365:60-62, 2005
`anastrozole alone as first-line therapy for patients
`14. Di Leo A, Jerusalem G, Petruzelka L, et al:
`4. Ma CX, Reinert T, Chmielewska |, et al: Mech-
`with receptor-positive postmenopausal breast can-
`anisms of aromatase inhibitor resistance. Nat Rev
`Results of the CONFIRM
`PhaseIII trial comparing
`cer, J Clin Oncol 30:1919-1925, 2012
`Cancer 15:261-275, 2015
`fulvestrant 250 mg with fulvestrant 500 mg in
`23. Finn RS, Crown JP, Lang |, et al: The cyclin-
`postmenopausal women with estrogen receptor—
`5. Wakeling AE, Dukes M, Bowler J: A potent
`dependent kinase 4/6 inhibitor palbociclib in combi-
`positive advanced breast cancer. J Clin Oncol 28:
`specific pure antiestrogen with clinical potential.
`nation with letrozole versus letrozole aloneas first-
`4594-4600, 2010
`Cancer Res 51:3867-3873, 1991
`line
`treatment
`of oestrogen receptor—positive,
`15. Di Leo A, Jerusalem G, Petruzelka L, et al:
`6. Wakeling AE: Similarities and distinctions in
`HER2-negative, advanced breast cancer (PALOMA-
`the mode of action of different classes of antioes-
`Final overall survival: Fulvestrant 500mg vs 250mg
`1/TRIO-18): A randomised phase 2 study, Lancet
`in the randomized CONFIRMtrial. J Natl Cancer Inst
`trogens. Endocr Relat Cancer 7:17-28, 2000
`Oncol 16:25-35, 2015
`106:djt337, 2014
`7. Osborne CK, Wakeling A, Nicholson RI: Ful-
`24. Martin M, Loibl S, Von Minckwitz G, et al:
`16. Robertson JF, Llombart-Cussac A, RolskiJ, et
`vestrant: An oestrogen receptor antagonist with a
`PhaseIII trial evaluating the addition of bevacizumab
`novel mechanism of action. Br J Cancer 90:S2-S6,
`al: Activity of fulvestrant 500 mgversus anastrozole
`to endocrine therapy as fi