Articles
`
`9+k®
`
`Fulvestrant plus anastrozole or placebo versus exemestane
`alone after progression on non-steroidal aromatase inhibitors
`in postmenopausal patients with hormone-receptor-positive
`locally advanced or metastatic breast cancer (SoFEA):
`
`a composite, multicentre, phase 3 randomised trial
`
`Stephen R Djohnston, LucyS Kilburn, Paul Ellis, David Dodwell, David Cameron, Larry Hayward, Young-Hyucklm,jeremy P Braybrooke, m
`A Murray Brunt, Kwok-Leung Cheung, Rema jyothi'rinayi, Anne Robinson, Andrew M Wardley, Duncan Wheatley, Anthony Howell, Gill Coombes,
`Nicole Sergenson, Hui-lung Sin, Elizabeth Folkerd, Mitch Dowsett, judith M Bliss, on behalf of the SoFEA investigators>9
`
`Summary
`Background The optimum endocrine treatment for postmenopausal women with advanced hormone-receptor— LancetOncol2013;14z989—98
`positive breast cancer that has progressed on non-steroidal aromatase inhibitors (NSAIs) is unclear. The aim of the
`Published Online
`SoFEA trial was to assess a maximum double endocrine targeting approach with the steroidal anti-oestrogen Juli/29,2013
`fulvestrant in combination with continued oestrogen deprivation.
`hm”/dx'dOl'orgi10'1016/
`sr47o-2o45(13)/0322—x
`.
`.
`.
`.
`,
`Thisonline ublication has
`Methods In a compOSIte, multlcentre, phase 3 randomlsed controlled tr1al done In the UK and South Korea,
`beencorrectped Thecorremd
`postmenopausal women with hormone-receptor—positive breast cancer (oestrogen receptor [ER] positive, progesterone
`version first appeared at
`receptor [PR] positive, or both) were eligible if they had relapsed or progressed with locally advanced or metastatic
`thelancet-com/oncology on
`disease on an NSAI (given as adjuvant for at least 12 months or as first-line treatment for at least 6 months).
`AUQUSt 27'2013
`Additionally, patients had to have adequate organ function and a WHO performance status of 0—2. Participants were
`599 C°'“'“e"tPage 917
`randomly assigned (1:1:1) to receive fulvestrant (500 mg intramuscular injection on day 1, followed by 250 mg doses
`CoPyrigthJOhnfitO” etal- OPen
`on days 15 and 29, and then every 28 days) plus daily oral anastrozole (1 mg); fulvestrant plus anastrozole-matched figetfnirileci':{K'KTEESAUnder
`placebo; or daily oral exemestane (25 mg). Randomisafion was done with computer-generated permuted blocks, and
`*Listed in the appendix
`stratification was by centre and previous use of an NSAI as adjuvant treatment or for locally advanced or metastatic
`R0 al Marsden NHS
`disease. Participants and investigators were aware ofassignment to fulvestrant or exemestane, but not of assignment
`Fozndationmst’London’UK
`to anastrozole or placebo. The primary endpoint was progression-free survival (PFS). Analyses were by intention to
`(Profs R DJohnston'
`treat. This trial is registered with ClinicalTrials.gov, numbers NCT00253422 (UK) and NCT00944918 (South Korea).
`EFolkerd PhD,
`ProfM Dowsett PhD); Clinical
`.
`.
`.
`.
`T.
`l
`ClSt t. if.
`U '1:
`lCR—
`Findings Between March 26, 2004, and Aug 6, 2010, 723 patlents underwent randomlsatlon: 243 were ass1gned to
`(Egfihelzsljtfi Olin“,
`receive fulvestrant plus anastrozole, 231 to fulvestrant plus placebo, and 249 to exemestane. Median PFS was
`Research,London,UK
`4-4 months (95% CI 3-4—5-4) in patients assigned to fulvestrant plus anastrozole, 4-8 months (3-6—5-5) in those
`(LSKi'bum M59
`assigned to fulvestrant plus placebo, and 3-4 months (3 - 0—4- 6) in those assigned to exemestane. No difference was Ergmfiifm’c) Kin ,5
`recorded between the patients assigned to fulvestrant plus anastrozole and fulvestrant plus placebo (hazard ratio
`Health Panners’iondgn’UK
`1-00, 95% CI 0-83—1- 21; log-rank p=0-98), or between those assigned to fulvestrant plus placebo and exemestane
`(ProfPEllis MD); LeedsTeaching
`(0-95, 0- 79—1- 14; log-rank p=0- 56). 87 serious adverse events were reported: 36 in patients assigned to fulvestrant H°§Pital5NH5TfUStStJames'S
`plus anastrozole, 22 in those assigned to fulvestrant plus placebo, and 29 in those assigned to exemestane. Grade 3—4
`#2:):eésl'3tgdcvjlpgggj 59:52:;
`adverse events were rare; the most frequent were arthralgia (three in the group assigned to fulvestrant plus anastrozole;
`of Edinburgh and Nl-IS Lothian’
`seven in that assigned to fulvestrant plus placebo; eight in that assigned to exemestane), lethargy (three; 11; 11), and Edinburgh,UK
`nausea or vomiting (five; two; eight).
`S;:::IC::1:;:;IMEIZ))¥:;T
`.
`.
`.
`.
`.
`UK (L Hayward MD); Samsung
`Interpretation After loss of response to NSAIs 1n postmenopausal women w1th hormone-receptor—posmve advanced MedicalCenter’Se°U|’5°Uth
`breast cancer, maximum double endocrine treatment with 250 mg fulvestrant combined with oestrogen deprivation
`Korea (Y—H Im MD); Bristol
`is no better than either fulvestrant alone or exemestane.
`Haematology andOncology
`Centre, Bristol, UK
`(J P Braybi'ooke MD); University
`Hospital ofNorth
`Staffordshire, Stoke-o n-Trent,
`
`.
`Funding Cancer Research UK and AstraZeneca.
`
`.
`'
`Introduction
`The optimum endocrine treatment for postmenopausal
`‘
`.
`‘
`women w1th advanced hormone-receptor—posrtrve breast
`cancer that has progressed during treatment with non-
`steroidal aromatase inhibitors (NSAls) is unclear.“ The
`.
`.
`.
`sterordal aromatase inactivator exemestanezv3 and the
`
`'
`N
`'
`u '
`.
`.
`.
`.
`UK(AMIBr”mMD)?
`steroidal oestrOgen-receptor downregulator fulvestrant‘iS
`have been recognised standards of care in this setting. Homllglam 'i'vemty
`ospitals,Nottingham,UK
`‘
`‘
`.
`‘
`‘
`The phase 3 EFECT trial6 showed no difference in clinical
`(K_LCheung MD); Kent
`efficacy between these two treatments for patients With
`0nc0|09yCentreiMaidstone,
`oestrogen receptor (ER)-positive metastatic breast cancer
`UKlRJyOthlm‘aylMDli
`.
`.
`.
`.
`Southend Hos ital,Southend,
`in the first-line and second-line settings.
`p
`
`www.thelancet.com/oncology Vo|14 September2013
`
`989
`
`AstraZeneca Exhibit 2063 p. 1
`InnoPharma Licensing LLC V. AstraZeneca AB IPR2017-00900
`Fresenius-Kabi USA LLC V. AstraZeneca AB IPR2017-01913
`
`

`

`Methods
`Study design and participants
`SoEEA was a phase 3 multicentre randomised controlled
`trial
`that was done in 82 UK centres. Additionally,
`investigators in South Korea expressed interest in joining
`the trial. To simplify governance arrangements, a parallel
`trial, sponsored by AstraZeneca and following the SoEEA
`protocol and case report forms, was initiated. Patients
`were recruited from four South Korean centres. The
`
`SoEEA trial as presented here represents a composite of
`the UK and South Korean initiatives.
`
`Postmenopausal women with hormone-receptor-
`positive breast cancer (ER positive or progesterone
`receptor [PR] positive, or both) were eligible if they
`relapsed or prOgressed with locally advanced or
`metastatic disease on an NSAl. The NSAl had to have
`
`been given as adjuvant treatment for at least 12 months,
`or as
`first-line treatment
`for
`locally advanced or
`metastatic disease for at least 6 months. Patients had to
`
`have adequate haematological, hepatic, and renal
`function, and a WHO performance status of 0—2.
`Patients
`already
`established
`on
`bisphosphonate
`treatment for at least 6 months or those who were due
`
`to start bisphosphonate treatment for bone metastases
`with other assessable sites of disease were eligible.
`Patients could have previously received tamoxifen and
`chemotherapy in the adjuvant or neoadjuvant setting or
`chemotherapy as first-line treatment
`for metastatic
`breast cancer followed by an NSAl alone for at least
`6 months. Patients were excluded if they had rapidly
`progressing visceral disease, malignancies other than
`breast cancer
`in the previous 5 years
`(except
`for
`adequately treated in-situ carcinoma of the cervix, or
`basal-cell or squamous-cell carcinoma of the skin), or
`thrombocytopenia (because of the risk of bleeding with
`intramuscular injection of fulvestrant). Additionally,
`patients who had received systemic corticosteroids for
`more than 15 days in the 4 weeks before randomisation
`were excluded.
`
`In the UK, this trial was approved by the Medicines and
`Healthcare
`products
`Regulatory Authority
`and
`South West 2 Multi-Research Ethics Committee (MREC
`03/6/77). In South Korea, the study was approved by
`Korea Food
`and Drug Administration and local
`institutional review boards. All patients provided written
`informed consent. The Institute of Cancer Research-
`
`Clinical Trials and Statistics Unit (lCR-CTSU; London,
`UK) had overall responsibility for trial management; two
`additional collaborating trials units, Cancer Clinical
`Trials Team Information Services Division (Edinburgh,
`UK) and C+R Research (Seoul, South Korea), were
`responsible for regional data management. The trial
`management group was
`responsible for day-to-day
`rurming of the trial. The trial was overseen by an
`independent trial steering committee. Emerging safety
`and efficacy data were confidentially reviewed regularly
`by the independent data monitoring committee.
`
`www.the|ancet.com/oncology Vol 14 September 2013
`
`AstraZeneca Exhibit 2063 p. 2
`
`Articles
`
`UK (A Robinson MD);The
`Christie NHS FoundationTrust,
`Manchester, UK
`(A M Wardley MD,
`ProfA Howell MD); Royal
`Cornwall Hospital, Truro, UK
`(DWheatley MD); Cancer Clinical
`TrialsTeam, Information
`Services Division, Edinburgh,
`UK (N Sergenson BSc); and
`AstraZeneca Korea, Seoul,
`South Korea (H-J Sin BSc)
`Correspondence to:
`Prof Stephen R DJohnston,
`Department of Medicine, The
`Royal Marsden NHS Foundation
`Trust, London SW3 6]], UK
`stephen.johnston@rmh.nhs.
`uk
`
`See Online for appendix
`
`setting of acquired
`in the
`Treatment options
`resistance to NSAls in ER-positive advanced breast
`cancer have changed since the results of the BOLERO-2
`trial were reported}8 This trial showed that progression-
`free survival (PES) was longer with the combination of
`exemestane and the mTOR antagonist everolimus than
`with exemestane alone.8 However, whether double
`endocrine targeting would be more effective than a
`partially non-cross-resistant endocrine agent
`in the
`setting of acquired resistance is unclear. Preclinical
`studies9le
`have
`suggested
`that
`the
`efficacy of
`fulvestrant could be increased in a low oestrogen
`environment. As a competitive antagonist
`for ER,
`oestradiol can compete with fulvestrant for receptor-
`site
`occupancy.
`ln MCE-7
`aromatase-transfected
`xenografts,
`the combination of fulvestrant and an
`arornatase inhibitor was more effective than either
`
`treatment alone?”12 Furthermore, in model systems of
`acquired resistance to long-term oestrogen deprivation,
`breast cancer cells seem to be stimulated by low
`residual amounts of oestrogens, which potentially
`could be enhanced on withdrawal of oestrogen
`suppression at the time of progression.”14
`Thus,
`a maximum double endocrine targeting
`approach in the setting of acquired resistance to NSAls
`should be investigated with fulvestrant in combination
`with continued oestrogen deprivation. The Study of
`Easlodex with or without concomitant Arimidex vs
`
` V
`
`Exemestane following progression on non-steroidal
`Aromatase
`inhibitors
`(SoEEA)
`was
`designed.
`Exemestane was the appropriate standard of care
`(control) at the time the trial was designed and was
`compared with the then accepted optimum dosing
`schedule for fulvestrant.
`
`
`
`
`723 patients randomly assigned
` V V
`
`249 assigned to exemestane
`231 assigned to fulvestrant
`243 assigned to fulvestrant
`
`
`
`
`
`
`plus anastrozole
`plus placebo
`
`2 did not start treatment
`—> 2 did not start treatment
`—> 1 did not start treatment
`—>
`
`
`
`
`
`
`V V V
`
`24 1 received assigned
`230 received assigned
`247 received assigned
`
`
`
`treatment
`treatment
`treatment
`
`
`
`
`
`
`
`
`
`237 discontinued
`213 progressec
`6 died
`9 had adverse events
`9 decision by patient
`or investigator
`
`
`
`V
`
`
`
`10 still on treatment
`
`
`
`
`
`
`
`222 discontinued
`207 progressec
`8 had adverse events
`7 decision by patient
`or investigator
`
`
`
`
`
`V
`8 sti Ion treatment
`
`
`
`238 discontinued
`221 progressec
`3 died
`7 had adverse events
`7 decision by patient
`or investigator
`
`V
`3 sti | on treatment
`
`
`Figure 1:Tria| profile
`
`
`
`
`
`990
`
`

`

`Articles
`
`Plasma oestradiol concentrations
`
`at baseline and
`
`3 months were also measured as an exploratory endpoint
`in a subset of patients who underwent randomisation
`after Nov 19, 2007, and who consented to and contributed
`at least one blood sample. Oestradiol analyses were done
`by Pharmanet (Princeton, N], USA) by gas chromatography
`tandem mass spectrometry with negative ion chemical
`ionisation after derivatisation ofthe steroid. The sensitivity
`ofthe assay was 0625 pg/mL (2 - 3 pmol/L).
`
`Statistical analysis
`The sample size was based on two primary aiIns; to detect
`an improvement in median PFS from 5-5 to 7-5 months
`in patients allocated to fulvestrant plus anastrozole
`compared with fulvestrant plus placebo, and from
`4~0 to 55 months in patients allocated to fulvestrant
`compared with exemestane. With a minimum follow-up
`of 6 months, 5% significance level (two-sided), and 90%
`power, 750 patients (250 per group) with 440 progression
`
`
`Exemestane
`Fulvestrant plus
`Fulvestrant plus
`
`(n=249)
`anastrozole(n=243) placebo (n=231)
`638670-710)
`63-4 (570-735)
`
`66-0 (59-2—75-0)
`
`Age at randomisation (years)
`Hormone-receptor status
`
`Randomisation and masking
`to receive
`Patients were randomly assigned (1:1:1)
`fulvestrant plus anastrozole, fuIvestrant plus placebo, or
`exemestane. Computer-generated permuted blocks were
`used, and stratification was by centre and previous use of
`an NSAI as adjuvant treatment or for locally advanced or
`metastatic disease. Independent randomisation was by
`telephone to ICR-CTSU and the Information Services
`Division in the UK and AstraZeneca in South Korea.
`
`Participants and investigators were aware of assignment
`to fulvestrant or exemestane, but not of assignment to
`anastrozole or placebo for patients in the groups assigned
`to fulvestrant.
`
`Procedures
`
`Fulvestrant was given with a loading dose schedule of a
`500 mg intramuscular
`injection into the gluteus
`maximus on day 1, followed by 250 mg injections on
`days
`15 and 29. Thereafter, 250 mg intramuscular
`injections were done every 28 days. Injections were given
`slowly, over the course of at least 2 min. Anastrozole
`(1 mg), matched placebo, and exemestane (25 mg) were
`given orally once daily. All treatments were given until
`disease progression or withdrawal.
`Data for treatment compliance were obtained for
`fulvestrant only, for which a delay was allowed for
`recovery from toxic elfects. Dose reductions are not
`standard for the treatments investigated in this trial.
`Timing of and reasons for treatment discontinuation
`were
`recorded. Fulvestrant,
`anastrozole,
`and the
`anastrozole-matched
`placebo were
`supplied
`by
`AstraZeneca. Exemestane was dispensed from hospital
`pharmacies or via the patient’s primary-care physician.
`Clinical assessment and toxicity reporting occurred
`monthly during the first 6 months, and every 3 months
`thereafter while treatment continued. Tumour assessment
`with Response Evaluation Criteria in Solid Tumors
`(RECIST; Version 10) was done every 3 months and at
`discontinuation or Withdrawal from treatment. Adverse
`events were. graded according to National Cancer Institute
`Common Toxicity Criteria (version 3 .0) and coded with the
`Medical Dictionary for Regulatory Activities (MedDRA;
`version 14.0), with central clinical review by SRD].
`The primary endpomt was PFS’ Wthh was defined as
`time from randomisation to progression of existing
`disease, new sites of disease, second primary cancer if
`change in systemic treatment was necessary, or death
`from any cause. Secondary endpoints were overall
`survival (time from randomisation to death from any
`cause),
`objective
`response
`(proportion
`achieving
`complete or partial response on trial treatment), clinical
`benefit
`(proportion achieving complete or partial
`response, or stable disease for at least 6 months on trial
`treatment), duration of response or clinical benefit (PFS
`in patients who had an objective response or clinical
`benefit), time to treatment failure (not reported here),
`and tolerability and safety.
`
`
`
`
`
`120(49%)
`38(16%)
`838%)
`2(1%)
`0
`
`124(54")
`33(14%)
`71(31%)
`1(<1%)
`20%)
`
`132(53%)
`231%)
`91 {37%)
`211%)
`1f<1%)
`
`170%)
`122 (50%)
`104 (43%),
`171 (70%)
`5-0 (23-100)
`
`14(6%)
`141 (61%)
`76 (33%,)
`170 (74%)
`5-1 (24—97)
`
`171796)
`142 (57%)
`90(36%)
`166(67%)
`5-2 (2-0-10-2)
`
`215(13-4—34-0)
`350940—447)
`201(12-6—29-2)
`
`212(12-0—34-5)
`249(174—41'9)
`18-6 (11.7—33.1)
`
`20-1 (12.9—32.9)
`242(18'5—41'9)
`19-3 (12.1—31.0)
`
`42 (17%)
`44 (18%)
`
`87 (36%)
`
`70 (29%)
`
`50 (22%)
`49 (21%)
`
`61 (26%)
`
`42 (17%)
`51 (20%)
`
`88 {35%)
`
`71 (31%)
`
`68(27%)
`
`éRPOSltlve’PRPOSltlve
`:RPOSltlve'PRnegatlve
`ERPOSlthQ'PR unknown
`ER negativeorunknow“:PRPOSi’EiVe
`ER ””k'IOW'I'PR ””k'ww”
`HERZ Status
`>°Sltlve
`Negative
`Jnknown
`PVEViOUStamOXife'Win adjuvant setting
`Time from Primary diagnosisto first
`re apselyears)
`(lmn;::lils)\lSAl before randomisation
`Adjuvant
`Locally advanced or metastatic breast
`cancer
`NSAI setting and time on NSAI
`Adjuvant
`Locally advanced or metastatic breast
`cancer; <1year
`Locallyadvanced or metastaticbreast
`cancettlto <2 Years
`Locally advancedor mEtaStatiC breast
`cancerlflyears
`Siteo‘crelapsel
`145 (58%)
`143 (62%)
`138 (57%)
`Visceral
`719996)
`50(22%)
`68(28%)
`SOfitlssue °rn°de
`32 l13%)
`37(16%)
`37(15%)
`Bone
`Data are n (9%) or median (IQR), ER=oestrogen receptor. PR=progesterone receptor, NSA|=non-steroida| aromatase
`inhibitor. *Data missing for one patient assigned to fulvestrant plus placebo and one assigned to exemestane.
`Table 1: Baseline characteristics
`
`
`
`
`www.thelancet.com/oncology Vo|14 September2013
`
`991
`
`AstraZeneca Exhibit 2063 p. 3
`
`

`

`Articles
`
`events in the two fulvestrant groups were needed for the
`principal analysis. Because ofa long period ofrecruitment,
`in 2010,
`the independent data monitoring committee
`agreed that
`the data were sufficiently mature for
`723 enrolled patients to answer the principal questions
`with the same number of events, but in a smaller total
`number ofpatients who had been followed up for a longer
`period than originally anticipated.
`The principal efficacy analyses included all patients who
`underwent randomisation on an intention-to-treat basis.
`
`Survival endpoints were shown graphically with Kaplan-
`Meier plots, and treatment comparisons made with the
`log-rank test. Hazard ratios (H Rs) were obtained from Cox
`
`100 —
`
`90—
`
`80—
`
`70—
`
`60—
`
`50—
`
`40—
`
`— Fulvestrant plus anastrozole (median 4-4 months, 95% CI 34—54)
`— Fulvestrant plus placebo (median 48 months, 95% CI 36—55)
`
`
`20— 10—
`
`
`Progression-freesurvival(We) 30—
`
`HR 1-00 (95% CIO-83—1-21);log-rank p=O-98
`
`0
`
`0
`
`Number at risk
`Fulvestrant plus 243
`anastrozole
`Fulvestrant plus 231
`placebo
`
`100 —
`
`El,
`
`148
`
`149
`
`0
`
`89
`
`90
`
`9
`
`67
`
`55
`
`1l2
`
`51
`
`44
`
`1l5
`
`34
`
`29
`
`1l8
`
`23
`
`18
`
`2l1
`
`17
`
`12
`
`2l4
`
`13
`
`11
`
`— Fulvestrant plus placebo (median 48 months, 95% CI 36—55)
`— Exemestane (median 3-4 months, 95% CI 30—46)
`
`90—
`
`80—
`
`70—
`
`60—
`
`50—
`
`40—
`
`Numberat risk
`Fulvestrantplus 231
`placebo
`13
`17
`21
`30
`42
`64
`88
`137
`Exemestane 249
`
`
`
`
`proportional hazards regression models, with HRs of less
`than 1 favouring fulvestrant plus anastrozole in the
`comparison of fulvestrant plus placebo and fulvestrant
`plus anastrozole, and fulvestrant plus placebo in the
`comparison of fulvestrant plus placebo and exemestane.
`The proportionality assumption of the Cox model was
`tested with Schoenfeld residuals, and was shown to hold.
`Subgroup analyses were reported with forest plots for
`age at randomisation, ER and PR status, HER2 status,
`time from diagnosis to first relapse, dominant site of
`relapse, and NSAI setting and time on NSAI combined.
`In view of the absence of standard prognostic factors in
`this setting, and to avoid overparameterisation of a
`were
`multivariable model, baseline
`characteristics
`assessed for prognostic ability, irrespective of treatment
`effect. Variables shown to be significant were combined
`in a multivariable model with a forward stepwise method.
`Treatment was then added to the model to obtain the
`
`effect. Proportions of
`treatment
`adjusted HR for
`responses were compared with Fisher’s exact tests.
`Safety analyses were done for all patients who received at
`least one dose of trial treatment (as treated population).
`The worst grade of adverse event during trial treatment
`was reported and compared with Fisher’s exact
`tests.
`All prespecified toxic effects and any MedDRA—coded event
`satisfying predefined criteria (ie, 210% frequency, p<0~01,
`or >1% difference in frequency between treatment groups)
`are presented. A significance level of <0-01 allowed some
`adjustment for multiple testing of toxicity endpoints.
`Geometric mean oestradiol concentrations were calculated
`
`by treatment group at each timepoint.
`This analysis includes all data received and processed
`by )an 3, 2012. Data were collated at ICR-CTSU, where all
`interim and final analyses were done. Central statistical
`monitoring was done by ICR-CTSU and was supple-
`mented by selected on-site source document verification.
`All analyses were done in Stata (version 10.1).
`This trial is registered as an International Standard
`Randomised Controlled Trial, number ISRCTN4419574-7,
`and with ClinicalTrialsgov, numbers NCT00253422 (UK)
`and NCT00944918 (South Korea).
`
`Role ofthe funding source
`The trial was cosponsored by The Royal Marsden NHS
`Foundation Trust and The Institute of Cancer Research in
`
`the UK; AstraZeneca sponsored the trial in South Korea.
`The funders had no role in data collection, data analysis,
`data interpretation, or writing of the report. The study
`design was peer-reviewed by Cancer Research UK and the
`protocol was
`approved by the trial
`sponsors and
`AstraZeneca. SRD), LSK, and )MB had full access to all
`the data in the study, and SRD] had final responsibility for
`the decision to submit for publication.
`
`Results
`Between March 26, 2004, and Aug 6, 2010, 723 patients
`underwent randomisation (figure 1): 698 from the UK
`
`www.thelancet.com/oncology Vol 14 September 2013
`
`AstraZeneca Exhibit 2063 p. 4
`
`
`
`HR 0-95 (95% Cl 079—114); log-rank p=O-56
`20— 10—
`
`Progression-freesurvival(%) 30—
`
`0
`
`0
`
`3
`I
`
`149
`
`é
`
`90
`
`15
`12
`9
`I
`I
`I
`Time from randomisation (months)
`
`18
`I
`
`21
`I
`
`24
`I
`
`55
`
`44
`
`29
`
`18
`
`12
`
`11
`
`FigureZ: Progression-free survival
`(A) Fulvestrant plus anastrozole vs fulvestrant plus placebo. (B) Fulvestrant plus placebo vs exemestane.
`-lR=hazard ratio.
`
`992
`
`

`

`Articles
`
`
`
`A
`B
`
`
`n
`Hazard ratio (95% Cl)
`n
`Hazard ratio (95% Cl)
`
`Age at randomisation (years)
`090 (0-49—1-67)
`<50
`45 —I—
`0-92 (070—121)
`50—64
`211
`—-——
`-01 (070—144)
`65—75
`129 —I—
`275
`89 ——I— -06 (069—163)
`ER and PR status"
`
`ER positive, PR positive
`ER positive, PR negative
`ER positive, PR unknown
`HER2 status
`
`0-85 (0-66—1-10)
`_-__
`244
`71 ——I—> 30 (0-80—2-10)
`154
`——-— -17 (084—163)
`
`263 + 0-95 (075—122)
`HERZ negative
`31 ——-—> -44(0-68—3-05)
`HER2 positive
`180 —I—
`-03 (076—140)
`HERZ unknown
`Time from diagnosisto first relapse (years)
`0-90 (0-56— -46)
`<1
`72 —I—
`1-3
`73 ——l—> -34 (084—215)
`3 to <5
`88 —I——
`0-89 (058—137)
`25
`241 —l—
`-06 (0-82— -38)
`Dominant site of relapsei’
`
`-10 (0-86— -39)
`——-—
`281
`Visceral
`118 —-— 0-98 (0-67— -43)
`Softtissue 0r node
`74 —I— 0-99 (0-61- 59)
`Bone
`NSAI setting andtime on NSAI
`Adjuvant
`92 —I— 097 (0-64— -47)
`Locally advanced or metastatic breast cancer
`0-95 (0-63— -44)
`<1 year
`93 —I
`1 to <2 years
`148
`——I— 126 (0-90— -77)
`22
`141 —-——
`0-85 (0-60— 19)
`Country
`
`
`
`
`
`27 ——-—> 1-51(0-59—3-85)
`210
`—-—
`0-94 (072—125)
`135
`—.——
`0-81 (057—115)
`108
`—I
`095 (064—142)
`
`—I—
`256
`56 —I—
`162
`—I—
`
`283
`31 4—
`166
`
`_-
`—-——
`
`0-94 (0-71—1-23)
`085 (049—148)
`093 (067—129)
`
`106 (083—134)
`0-20 (0-08—0-51)
`093 (068-127)
`
`79 ——-— 118 (074—189)
`75 ——I— 21-13 (0721—1-80)
`82 —I— 0-98 (062—153)
`244
`+—
`081(0-62—1-05)
`
`288
`121
`69
`
`92
`
`100
`149
`139
`
`0-93 (0-73—1-18)
`—I—
`079 (054—116)
`—I——
`——I—> 1-37 (0383-2-25)
`
`—I—
`
`090 (059—138)
`
`——I— 127 (084—191)
`+—
`0-75 (054—106)
`—-— 106 (075—150)
`
`465
`15
`
`
`—-—
`-
`
`0-96 (080—116)
`> 0-54 (0-14—2-05)
`
`480:1:
`Ol6
`Ol-l
`4—
`Favours fulvestrant plus placebo
`
`—-—
`1-0
`
`ll2
`—>
`Favours exemestane
`
`0.92 (0.77—1.11)
`2lO
`
`
`
`
`
`
`UK
`South Korea
`
`Overall
`
`459
`15
`
`4741:
`
`
`—-—
`
`1-00 (0-83— -20)
`p 1-74 (046—662)
`
`105 (087—126)
`
`Ol6
`Ol-l
`4—
`Favours fulvestrant plus anastrozole
`
`—-—
`2-0
`l-O
`1l2
`—>
`Favours fulvestrant plus placebo
`
`Figure3: Subgroup analyses of progression-free survival
`(A) Fulvestrant plus anastrozole vs fulvestrant plus placebo. (B) Fulvestrant plus placebo vs exemestane. ER=oestrogen receptor. PR=progesterone receptor. NSAl=non-steroidal aromatase inhibitor.
`*Data for the few patients with ER-negative or unknown, and PR-positive disease, and those with unknown hormone-receptor-status not shown here. iData missing for one patient assigned to
`ulvestrant plus placebo and one assigned to exemestane. iAdjusted for time from diagnosisto first relapse, number of disease sites at baseline, and NSAI setting and time on NSAI.
`
`and 25 from South Korea. Baseline characteristics, such
`as time from diagnosis to first relapse and sites of
`dominant disease, are representative of a population of
`patients with
`hormone-receptor-positive metastatic
`breast cancer (table 1). 589 (81%) had previously received
`an NSAI in the locally advanced or metastatic setting for
`a median of 193 months (IQR 12-1—31-2;
`table 1),
`suggesting that this population had a good response to
`previous NSAI
`treatment. Four patients assigned to
`fulvestrant plus
`anastrozole missed a
`fulvestrant
`injection, and 109 patients (50 assigned to fulvestrant
`plus anastrozole, 59 assigned to fulvestrant plus placebo)
`had at least one scheduled fulvestrant dose delay.
`After a median follow-up in all patients of 37-9 months
`(IQR 23-1—50-8), 689 progression events were reported:
`235 in patients assigned to fulvestrant plus anastrozole,
`
`221 in those assigned to fulvestrant plus placebo, and 233
`in those assigned to exemestane. No difference in PFS
`was recorded between patients assigned to fulvestrant
`plus anastrozole and fulvestrant plus placebo, or between
`those assigned to fulvestrant plus placebo and exemestane
`(figure 2). A multivariable analysis with adjustment for
`time from diagnosis to first relapse, number of disease
`sites present at baseline, and NSAI setting and time on
`NSAI did not substantially affect estimates of treatment
`effect (fulvestrant plus anastrozole vs fulvestrant plus
`placebo: HR 105, 95% Cl 087—126, p=0~62; fulvestrant
`plus placebo vs exemestane: 092, 077—111, p=0~41).
`Subgroup analyses were consistent with the overall effect
`on PFS (figure 3).
`508 patients had died: 168 (69%) assigned to fulvestrant
`plus anastrozole, 167 (72%) to fulvestrant plus placebo,
`
`www.thelancet.com/oncology Vol14 September2013
`
`993
`
`AstraZeneca Exhibit 2063 p. 5
`
`

`

`Articles
`
`
`
`— Fulvestrant plus anastrozole (median 20-2 months, 95% CI17-2—22-5)
`— Fulvestrant plus placebo (median 19-4 months, 95% CI16-8—22-8)
`
`
`
`HR 0-95 (95% Cl 076—117); log rank p=O-61
`I
`3
`
`é
`199
`
`9')
`182
`
`1'2
`165
`
`1'5
`133
`
`1'8
`112
`
`2'1
`89
`
`2'4
`77
`
`2'7
`57
`
`3'0
`46
`
`3'3
`39
`
`3'6
`37
`
`192
`
`176
`
`154
`
`133
`
`109
`
`86
`
`68
`
`57
`
`44
`
`34
`
`30
`
`— Fulvestrant plus placebo (median 19-4 months, 95% CI16-8—22-8)
`— Exemestane (median 21-6 months, 95% CI19-4—23-9)
`
` 0
`
`100
`
`90—
`
`80—
`
`70—
`
`60—
`
`50—
`
`40—
`
`30—
`
`20—
`
`10—
`
`0
`
` 0
`
`
`HR 1-05 (95% Cl 084—129); log rank p=O-68
`|
`|
`|
`l
`|
`|
`|
`I
`3
`24
`21
`18
`6
`9
`12
`15
`Time from randomisation (months)
`154
`133
`109
`86
`68
`
`225
`
`192
`
`176
`
`27
`
`3O
`
`33
`
`36
`
`T
`
`44
`
`34
`
`30
`
`Numberatrisk
`Fulvestrantplus 231
`placebo
`Exemestane 249
`31
`37
`49
`59
`79
`96
`117
`134
`158
`179
`200
`225
`
`
`100
`
`90—
`
`80—
`
`70—
`
`60—
`
`50—
`
`40—
`
`30—
`
`20—
`
`10— 0
`
`Numberatrisk
`Fulvestrantplus 243
`anastrozole
`Fulvestrantplus 231
`placebo
`
`227
`
`225
`
`
`
`
`
`Overallsurvival(%)
`
`
`
`
`
`Overallsurvival(%)
`
`
`
`Figure4: Overall survival
`(A) Fulvestrant plus anastrozole vs fulvestrant plus placebo. (B) Fulvestrant plus placebo vs exemestane.
`-lR=hazard ratio.
`
`and 173 (69%) to exemestane. Most deaths were due to
`breast cancer. Only 12 deaths were reportedly due to
`other causes: cardiovascular (one patient assigned to
`fulvestrant plus anastrozole,
`two to fulvestrant plus
`placebo), cerebrovascular (one assigned to fulvestrant
`plus placebo, one to exemestane), primary lung cancer
`(one assigned to fulvestrant plus placebo, one to
`exemestane), pneumonia (one assigned to fulvestrant
`plus anastrozole, one to exemestane), neutropenic sepsis
`(one assigned to fulvestrant plus placebo), and unknown
`(one assigned to fulvestrant plus anastrozole, one to
`exemestane). Only two of the deaths due to causes other
`
`994
`
`
`
`than breast cancer (one pneumonia and one unknown)
`occurred on trial treatment, and neither was deemed to
`be related to treatment.
`No difference in overall survival was recorded between
`
`patients assigned to fulvestrant plus anastrozole and
`fulvestrant plus placebo, or between those assigned to
`fulvestrant plus placebo and exemestane (figure 4).
`Subgroup analyses were consistent with the overall effect
`on overall survival (appendix).
`(7%) of
`18
`In the intention-to-treat population,
`243 patients assigned to fulvestrant plus anastrozole
`had objective
`tumour
`responses
`(one
`complete
`response,
`17 partial
`response), as did 16 (7%) of
`231 assigned to fulvestrant plus placebo (all partial
`response), and nine (4%) of 249 assigned to exemestane
`(two complete
`response,
`seven partial
`response;
`fulvestrant plus anastrozole vs fulvestrant plus placebo:
`p=0~88;
`fulvestrant plus placebo VS
`exemestane:
`p=0-27). 558 patients (77%) had measurable disease:
`194 (80%) assigned to fulvestrant plus anastrozole,
`178 (77%) to fulvestrant plus placebo, and 186 (75%) to
`exemestane. Of these patients,
`15
`(8%) patients
`assigned to fulvestrant plus anastrozole achieved
`objective responses (all partial response), as did 14 (8%)
`assigned to fulvestrant plus placebo (all partial
`response), and seven (4%) assigned to exemestane (one
`complete response, six partial response;
`fulvestrant
`plus anastrozole VS fulvestrant plus placebo: p=1~00;
`fulvestrant plus placebo vs exemestane: p=0~17).
`Median duration ofobjective response was 12-3 months
`(lQR 5 -7—22 - 1) for patients assigned to fulvestrant plus
`anastrozole, 16 - 5 months (78—29 - 2) for those assigned
`to fulvestrant plus placebo, and 17 2 months (9 - 6—26 - 9)
`for those assigned to exemestane.
`82 patients
`(34%)
`assigned to fulvestrant plus
`anastrozole, 73
`(32%) assigned to fulvestrant plus
`placebo, and 67 (27%) assigned to exemestane achieved
`clinical benefit (fulvestrant plus anastrozole VS fulvestrant
`plus placebo: p=0-75;
`fulvestrant plus placebo VS
`exernestane: p=0~27).
`In patients with measurable
`disease, 63 (33%) assigned to fulvestrant plus anastrozole,
`55 (31%) assigned to fulvestrant plus placebo, and 43
`(23%) assigned to exemestane achieved clinical benefit
`(fulvestrant plus anastrozole VS fulvestrant plus placebo:
`p=0~94; fulvestrant plus placebo vs exemestane: p=0 - 16).
`Median duration of clinical benefit was 13-0 months
`
`for patients assigned to fulvestrant plus
`(89—189)
`anastrozole, 13-0 months (83—175) for those assigned
`to fulvestrant plus placebo, and 13 -0 months (93—217)
`for those assigned to exemestane.
`87 serious adverse events were reported, of which
`three were
`suspected unexpected serious
`adverse
`reactions (one in the group assigned to fulvestrant plus
`anastrozole and two in the group assigned to fulvestrant
`plus placebo) and 11 were serious adverse reactions
`(six in the group assigned to fulvestrant plus anastrozole,
`three in that assigned to fulvestrant plus placebo, and
`
`www.the|ancet.com/oncology Vol 14 September 2013
`
`AstraZeneca Exhibit 2063 p. 6
`
`

`

`
`
`Articles
`
`Fulvestrant plus placebo
`Fulvestrant plus anastrozole
`(n=241)
`(n=230)
`
`
`Exemestane (n=247)
`
`
`
`
`
`0,22
`0-89
`0-51
`0.41
`0-33
`071
`0-34
`1-00
`0-65
`0-75
`0-50
`0-84
`0.31
`023
`0:32
`0-20
`0-41
`0-36
`0-68
`0-13
`0.65
`0-09
`0-77
`1-00
`0-11
`0-36
`004
`0-69
`0-20
`0-08
`0-37
`0-57
`0-69
`1-00
`017
`0-50
`0357
`1-00
`1.00
`0-19
`0.37
`0-77
`051
`0-11
`0-05
`0-30
`0-16
`0.62
`0-51
`(Continueson next page)
`
`
`p value fulvestrant
`p value fulvestrant plus
`plus placebo vs
`anastrozole vs
`fulvestrant plus placebo
`exemestane
`
`Any grade
`Grades 3 and 4
`Any grade
`Grades 3 and 4
`Any grade
`Grades 3 and 4
`3 (1%)
`0
`1 (<1%)
`0
`5 (2%)
`1 (<1%)
`25 (10%)
`0
`31 (13%)
`0
`32 (13%)
`0
`5 (2%)
`0
`3 (1%)
`1 (<1%)
`6 (2%)
`2 (1%)
`97 (40%)
`3 (1%)
`98 (43%)
`7 (3%)
`115 (47%)
`8 (3%)
`18 (7%)
`1 (<1%)
`23 (10%)
`0
`18 (7%)
`1 (<1%)
`21 (9%)
`3 (1%)
`13 (6%)
`3 (1%)
`17 (7%)
`3 (1%)
`5 (2%)
`1 (<1%)
`3 (1%)
`0
`7 (3%)
`0
`4 (2%)
`1 (<1%)
`1 (<1%)
`0
`1 (<1%)
`0
`6 (2%)
`1 (<1%)
`8 (3%)
`4 (2%)
`11 (4%)
`0
`64 (27%)
`2 (1%)
`57 (25%)
`0
`(823%)
`1 (<1%)
`8 (3%)
`0
`20 (9%)
`1 (<1%)
`(17)
`0
`73 (30%)
`1 (<1%)
`(627%)
`3 (1%)
`708(2”26)
`3 (1%)
`40 (17%)
`1 (<1%)
`53(23%)
`2 (1%)
`47(19%)
`0
`12 (5%)
`0
`9 (4%)
`0
`16(6%)
`0
`0
`0
`6 (3%)
`0
`3 (1%)
`0
`2 (1%)
`0
`4(2%)
`0
`1 (< %)
`0
`52 (22%)
`0
`59 (26%)
`1 (<1%)
`72 (29%)
`0
`0
`0
`3 (1%)
`0
`1 (< %)
`0
`0
`O
`3 (1%)
`0
`2 (1%)
`0
`17 (7%)
`1(<1%)
`35 (15%)
`5 (2%)
`26(1 %)
`2 (1%)
`7 (3%)
`1 (<1%)
`8 (3%)
`2 (1%)
`11 (4%)
`0
`49(20%)
`1 (<1%)
`64 (28%)
`2 (1%)
`52 (2 %)
`1 (<1%)
`