VOLUME
`
`26
`
`NUMBER 10
`
`APRIL
`
`1.
`
`2008
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`NAL REPORT
`
`
`
`From the Division of Medical Oncology,
`British Colurnbia Cancer Agency,
`University of British Columbia, Vancou-
`ver, British Columbia, Canada; Division
`of Hematology and Medical Oncology,
`Feinberg School of Medicine, North-
`westemUniversity, Chicago, IL; Depart-
`ment of Medical Oncology, Institut
`Bergonié, Bordeaux; Department of
`Medical Oncology, CRLC Val d’ Aurelle-
`Paul Lamarque, Montpellier, France;
`Medical Oncology, Instituto National de
`Cancer, Rio de Janeiro; Departmerito
`de Radiologia, Faculdade de Medicina
`da USP, University of Sao Paulo, Sao
`Paulo, Brazil; Division of Gynecologic
`Oncology, Department of Obstetrics
`and Gynecology, University Hospitals
`Leuven, Leuven, Medicine Department,
`Jules Bordet Institute, Brussels,
`Belgium, Investigacion Clinica, Centro
`Oncologico de Rosario, Rosario, Argen-
`tina; The M.D. Anderson Cancer Center,
`University of Texas, Houston, TX;
`Professorial Unit of Surgery, Notting-
`am City Hospital, Nottingharn; Clinical
`Development, AstraZeneca Pharmaceu-
`ticals, Macclesfield, Cheshire, United
`ingdom, Division of Hematology/
`Oncology, University of Pittsburgh
`School of Medicine, Pittsburgh, PA;
`Department of Oncology and Haematol-
`ogy, Charite Campus Mitte, Universita-
`tsmedizin, Berlin, Germany; and Clinical
`Development, AstraZeneca Pharimaceu-
`ticals, Wilraington, DE
`Submitted July 26, 2007; accepted
`December 5, 2007; published online
`ahead of print at www_jco.org on
`March 3, 2008
`
`Presented in part at the 29th Annual
`San Antonio Breast Cancer Syrnposium
`December 14-17, 2006, San Antonio,
`Texas.
`
`Authors’ disclosures of potential con-
`flicts of interest and author contribu-
`tions are found at the end of this
`article
`
`Corresponding author: Stephen Chia,
`MD, Division of Medical Oncology, Brit
`ish Columbia Cancer Agency, University
`of British Columbia, 600 West 10"
`Avenue, Vancouver, B.C. Canada, V5Z
`AEG; email schia@becancer.be.ca
`
`© 2008 by American Society of Clinical
`Oncology
`0732-183X/08/26 10-1664/$20.00
`DOL 10.1200/JCO.2007.13.5822
`
`Double-Blind, Randomized Placebo Controlled Trial of
`Fulvestrant Compared With Exemestane After Prior
`Nonsteroidal Aromatase Inhibitor Therapy in
`Postmenopausal Women With Hormone
`Receptor—Positive, Advanced Breast Cancer: Results
`From EFECT
`
`Stephen Chia, William Gradishar, Louis Mauriac, Jose Bines, Frederic Amant, Miriam Federico, Luis Fein,
`Gilles Romieu, Aman Buzdar, John F.R. Robertson, Adam Brufsky, Kurt Possinger, Pamela Rennie,
`Francisco Sapunar, Elizabeth Lowe, and Martine Piccart
`
`A
`
`B
`
`S$
`
`T
`
`R
`
`A
`
`T
`
`Purpose
`
`Thethird-generation nonsteroidal aromatase inhibitors (Als) are increasingly used as adjuvant and
`
`
`first-line advanced therapy for postmenopausal, hormone receptor—positive (HR+) breast cancer.
`Because many patients subsequently experience progression or relapse, it is important to Identify
`agents with efficacy after Al failure.
`Materials and Methods
`Evaluation of Faslodex versus Exemestane Clinical Trial (EFECT) is a randomized, double-blind,
`placebo controlled, multicenter phase III trial of fulvestrant versus exemestane in postmenopausal
`women with HR+ advanced breast cancer (ABC) progressing or recurring after nonsteroidal Al,
`The primary end point was time to progression (TTP). A fulvestrant loading-dose (LD) regimen was
`used: 500 mg intramuscularly on day 0, 250 mg on days 14, 28, and 250 mg every 28 days
`thereafter. Exemestane 25 mg orally was administered once daily.
`Results
`A total of 693 women were randomly assigned to fulvestrant (n = 351) or exemestane (n = 342)
`Approximately 60% of patients had received at least two prior endocrine therapies. Median TTP
`was 3.7 months in both groups (hazard ratio = 0.963: 95% Cl, 0.819 to 1.133; P = .6531). The
`overall response rate (7.4% v6.7%; P = .736) and clinical benefit rate (82.2% v31.5%; P= .853)
`weresimilar between fulvestrant and exemestane respectively. Median duration ofclinical benefit
`was 9.3 and 8.3 months, respectively. Both treatments were well tolerated, with no significant
`differences in the
`incidence of adverse
`events or quality oflife. Pharmacokinetic data confirm that
`steady-state was reached within 1 month with the LD schedule of fulvestrant.
`Conclusion
`Fulvestrant LD and exemestane are equally active and well-tolerated in a meaningful proportion of
`postmenopausal women with ABC who have experienced progression or recurrence during
`treatment with a nonsteroidal Al.
`
`
`
`
`
`J Clin Oncol 26:1664-1670. © 2008 by American Society of Clinical Oncology
`
`INTRODUCTION
`
`Hormone receptor—positive (HR+) breast canceris
`the most commonpresentation of breast cancerto-
`day.’ In postmenopausal HR+ breast cancer, there
`are several hormonal therapeutic options available,
`of which the classes of selective estrogen receptor
`modulators (SERMs) and aromatase inhibitors
`(Als) have been studied extensively and are standard
`therapeutic optionsin breast cancer.
`
`The third-generation Als consists of both non-
`steroidal (anastrozole, letrozole) and steroidal (ex-
`emestane) inhibitors. The nonsteroidal inhibitors
`block the peripheral conversion of androgens to
`estrogens by inhibiting the heme porphyrin por-
`tion of aromatase. In contrast, the steroidal Als
`act by binding irreversibly to the androgen bind-
`ing site and are structurally different from the
`nonsteroidal Als. As first-line therapy in HR+,
`postmenopausal advanced breast cancer (ABC),
`
`1664
`
`© 2008 by American Society ofClinical Oncology
`Information downloaded from jco.ascopubs.org and provided by at AZ Library on September 29, 2010 from 193.132.159.169
`Copyright © 2008 American Society of Clinical Oncology. All rights reserved.
`
`AstraZeneca Exhibit 2062 p. 1
`InnoPharma Licensing LLC v. AstraZeneca AB IPR2017-00905
`
`€
`

`

`Randomized Trial of Fulvestrant v Exemestane in Advanced Breast Cancer
`
`the Als have demonstrated superiority to tamoxifen for response rates
`and time to progression.’* Furthermore, the Als, either up front or
`after tamoxifen, have been clearly established as adjuvant hormonal
`optionsin early-stage HR+ postmenopausal breast cancer.'° Unfor-
`tunately, the vast majority of patients diagnosed with ABC will even-
`tually progress during treatment with a specific therapy, and a
`significant proportion of patients with early stage-breast cancerswill
`relapse. Thus, additional therapeutic agents are required to continue
`to treat the diseaseat time of progression/relapse.
`Fulvestrant is a novel estrogen-receptor (ER) antagonist that,
`unlike tamoxifen,is devoid ofany agonist activity.'' On bindingto the
`ER, fulvestrant induces a rapid degradation and loss of ER and the
`progesterone receptor (PgR).'*"'° Several large phaseIII trials have
`demonstrated significant activity for fulvestrant in the treatment of
`HR+ ABC, with similar efficacy to that of anastrozole and
`tamoxifen.'*!° Furthermore,activityhas been seen in phaseIItrials of
`fulvestrant after progression during treatment with a nonsteroidalAl,
`with clinical benefit rates (CBRs) of 30%to 35%,!78
`Exemestaneis a steroidal-based AI, with modest androgenic ac-
`tivity.’ Exemestane has been studied in a phaseIItrial after docu-
`mented progression during treatment with a nonsteroidal AI, and
`showed a 20% clinical benefit rate.*” Because ofthe lack of random-
`ized clinicaltrial data and the prevalence of patients exposed to non-
`steroidal Als, the Evaluation of Faslodex versus Exemestane Clinical
`Trial (EFECT) was undertaken to address this specific question of
`which hormonal agentto considerfirst after progression during treat-
`ment with a nonsteroidal AI.
`
`
`
`Study Design
`EFECTis a randomized, double blind, double-dummy,phaseIII inter-
`national trial designed to compare theefficacy andtolerability of a loading-
`dose (LD) schedule of fulvestrant to exemestane in postmenopausal women
`with HR+ ABC with disease progression after prior nonsteroidal AI therapy.
`
`Patient Population
`All patients were postmenopausal women with incurable locally ad-
`vanced or metastatic breast cancer whose disease had relapsed during
`treatment with (or within 6 months of discontinuation of) an adjuvant
`nonsteroidal AI, or whose advanced disease progressed during treatment
`with a nonsteroidal AI. Patients were categorized as AI sensitive if the
`investigator determined that the patient had a complete response (CR),
`partial response (PR), or stable disease (SD) for at least 6 months during
`treatment with the AI for ABC. All other patients, including all those who
`received the Al as adjuvant therapy, were defined as AI resistant.
`Inclusion onto thetrial required women to be postmenopausal (= 60
`years old, or age = 45 years with amenorrhea for > 12 monthsor follicle
`stimulating hormonelevels within postmenopausal range, or priorbilateral
`oophorectomy). Other inclusion criteria included HR+ (ER and/or PgR)
`disease as determinedlocally, WHO performance status of0 to 2,life expect-
`ancy of at least 3 months and the presence of at least one measurable or
`assessable (nonmeasurable) lesion. Initially, the protocol required thatall
`patients have at least one measurable lesion by Response Evaluation Criteria in
`Solid Tumors (RECIST)criteria, but subsequently the protocol was amended.
`to include patients with bone only (lytic or mixed) metastatic lesions. Up to
`one prior chemotherapy regimen for the treatment ofABC wasallowed.
`Exclusion criteria included lite threatening metastatic visceral disease,
`brain or leptomeningeal metastases, prior exposure to either fulvestrant or
`exemestane,extensive radiation or cytotoxic therapy within the last 4 weeks, or
`a history of bleeding diathesis or need for long-term anticoagulation.
`
`All women provided written informed consent before registration on
`trial. The study was conducted in accordance with the ethical principles that
`originated in the Declaration ofHelsinki and with local Research Ethics Board.
`approval at each participating center.
`Trial Treatments
`
`Fulvestrant 250 mg/5 mL (<2) asan intramuscular injection or a match-
`ing5 mL (X2) oilyexcipient placebo wasinjected into each buttock (500 mg or
`matching placebo) on day 1, followed by a single injection of 250 mg fulves-
`trant/placebo at day 14 and again on day 28. Treatmentafter day 28 was every
`28 days (+ 3 days) thereafter. Exemestane 25 mg anda matching placebo were
`to be taken orally once daily.
`Patients continued treatment until objective disease progression or other
`events that required withdrawal. There was nobuilt in crossover design in this
`trial. Thereafter, patients were followed up until death. Patients who withdrew
`from trial treatment before progression were followed up for response until
`progression and death.
`All patients were seen bya physician monthly until month 6, and every 3
`months thereafter. Tumor assessment was performed every 8 weeks from
`baseline until month 6, and then every 3 months until disease progression.
`Ina subset of60 patients (30 in each treatment group) pharmacokinetic
`samples were collected at specified time intervals to confirm whether the LD
`regimen would achieve steady-state earlier than that seen previously with a
`dose of fulvestrant 250 mg every 28 days.
`
`Statistical Analysis
`The primary end point of the study was time to disease progression
`(TTP). Secondary end points included objective response (OR) rate, CBR,
`duration of response, time to response, overall survival, and tolerability. The
`trial was designed to detect superiority of fulvestrant compared with exemes-
`tane in terms of TTP. Thefinal analysis was scheduledto take place when 580
`progression events(ie, objective disease progression or death) had occurred.
`across both treatment groups. This would provide 90% power to detect a
`hazard ratio of 1.31 or greater, or of 0.76 or less for fulvestrant treatment
`compared with exemestane treatment, at a two-sided significance level of5%.
`To achieve the required number of events, it was planned to recruit 660
`patients (330 in each treatment group). Data for the efficacy parameters were
`analyzed and summarized on an intention-to-treat basis.
`TTP
`
`TTP was defined as the numberof days from the date of randomassign-
`ment until the date of objective disease progression, as per RECISTcriteria. If
`the patient died without documented disease progression, and the date of
`death was no more than 6 months from the last disease assessment per RE-
`CIST, then death was regardedas a progression event. For patients who had
`not experienced disease progression at the timeof data cutoff, data were right
`censoredto the date ofthe last RECIST assessment.
`
`The primary analysis for TTP was the unstratified log-rank test. The
`secondary analysis used the Cox proportional hazards regression model and
`includedthe following six baseline covariates: age (< 65 v = 65 years), number
`ofprior hormonal therapies (1 v = 2), receptor status (both ER+ and PgR+ v
`only one receptorpositive), visceral involvement(yes v no), presence of mea-
`surable disease compared with nonmeasurable disease, and Al sensitive versus
`AI resistant. The treatment effect was estimated using the hazard ratio of
`fulvestrant to exemestane, together with the 95% CI and P value. A global
`interaction test using a 1% significance level was performed to determine
`whether the overall treatment benefit was consistent across each of the six
`
`covariates. TTP was also summarized using Kaplan-Meier curves for each
`treatment group and the median TTP was calculated.
`Overall Survival
`
`Timeto death was to be analyzed when more than 50% ofthe patients
`had died across both treatment groups. At the time of data analysis, only
`34% of patients had died, and therefore no formal statistical analyses
`were conducted.
`
`Best OR and CBR
`
`An ORwas defined as a patient having a best overall response of either
`CR or PR with confirmation criteria as per RECIST. A patient with clinical
`
`www.jco.org
`
`© 2008 by American Society of Clinical Oncology
`Information downloaded from jco.ascopubs.org and provided by at AZ Library on September 29, 2010 from 193.132.159.169
`Copyright © 2008 American Society of Clinical Oncology. All rights reserved.
`
`1665
`
`AstraZeneca Exhibit 2062 p. 2
`
`

`

`Chia et al
`
`benefit (CB) was defined asa patient havinga best overall response ofa CR, PR,
`or SD forat least 24 weeks. SD was defined, as per RECISTcriteria, as neither
`achieving a PR nor progressive disease at week 24 orlater.
`
`Duration of Response
`Duration ofresponse (DOR)was evaluated only for patients who had an
`OR,and wasdefined as the numberofdays from date of random assignment
`until the dayon which disease progression or death resulting from any cause
`was first observed.
`
`Quality ofLife
`Quality of life (QOL) was assessed using the Functional Assessment of
`Cancer Therapy—Endocrine Symptom (FACT-ES) instrument. The analysis
`was undertaken using both the FACT-ESand Trial Outcome Index (TOI). The
`difference between the twotreatment groups in FACT-ES and TOIovertime
`was compared using a generalized linear mixed model, with the Restricted.
`Maximum Likelihood option, including the samesix covariates as for TTP.
`
`Tolerability
`All safety data were listed and summarized according to the treatment
`received. Adverse events (AEs) were presented using MedDRA terminology.
`Eight AE categories considered relevant to endocrine therapy were predefined
`forstatistical analysis. The analysis ofthe predefined AEs was performed using
`a two-sided Fisher’s exact test at the 5% significance level.
`
`RESULTS
`
`Patients
`A total of 693 womenacross 138 centers worldwide were randomly
`assignedto either fulvestrant (n = 351) or exemestane (n = 342)from
`August 2003 to November 2005. The accountability ofall patients
`randomly assigned is seen in Figure Al (online only). Baseline char-
`acteristics between the two randomlyassigned treatments are outlined
`in Table 1. Overall, the groups were well balanced, except that the
`fulvestrant cohort had a slightly greater number of women with ER+
`and PgR+ tumors (67.5%) versus the exemestane cohort (56.4%).
`Approximately 60% of participants had two or more prior lines of
`hormonaltherapy. Approximately 60% ofpatients in both groups had
`either a response (CR or PR) or SD lasting at least 6 months during
`treatment with the prior nonsteroidal AI for ABC (termed AI sensi-
`tive) as determined by the individual investigator. Only 10% of
`women enrolled received their previous AI as adjuvant therapy. The
`median follow-upfor all patients alive is approximately 13 months.
`
`Efficacy
`The primary end point of this study was TTP. At the time of analy-
`sis, 82.1% (n = 288) of the fulvestrant group and 87.4%(n = 299) of
`the exemestane group had experienced a defined progression event.
`The median time to progression(Fig 1) in both groups was 3.7 months
`(P = .65) with a hazard ratio of 0.93 (95% CI, 0.819 to 1.133). The
`adjusted hazardratio for the specified covariates was 0.968 (P = .70)
`with the 95% CI at 0.822 to 1.141. In an investigation of the consis-
`tency oftreatmenteffect across the predefined covariates, there were
`nostatistically significant differences (Fig 2).
`
`OR Rate and CBR
`A total of 540 patients (270 in each arm) had measurable disease
`by RECISTcriteria at trial entry. Overall, 20 patients in the fulvestrant
`arm (7.4%) and 18 patients in the exemestane arm (6.7%) had a doc-
`umented response (odds ratio = 1.12; 95% CI, 0.578 to 2.186;
`P = .736). The CBR was 32.2% and 31.5% in the fulvestrant and
`
`exemestane arms, respectively (odds ratio = 1.03; 95% CI,0.72 to
`1.487; P = .853). Of note, in the cohort of patients with visceral
`involvement, the CBR was 29% and 27% in the fulvestrant and ex-
`emestane arms, respectively.
`The median DOR,as measured from the date of random assign-
`ment, was 13.5 monthsin the fulvestrant group and 9.8 monthsin the
`exemestane group (Fig 3); median DOR as measured from thedate of
`first response was 7.5 months for fulvestrant compared with 5.5
`monthsfor exemestane.
`
`Pharmacokineties
`The pharmacokinetic (PK) substudy results mirrored those from
`modeling studies and demonstrated a muchfaster time to steady-state
`levels with the LD schedule of fulvestrant, compared to prior PK
`studies of the 250 mg monthly dose. Median time to steady state was
`achieved within 28 days with the LD regimen, compared with 3 to 6
`months with the 250-mg monthly dose”* (Fig 4).
`
`Tolerability
`Both fulvestrant and exemestane were well tolerated in this study
`(Table 2), with only 2% of fulvestrant-treated patients and 2.6% of
`exemestane-treated patients withdrawing because of an adverse event
`(AE). Drug-related serious AEs (SAEs) were rare, occurring in 1.1%
`and 0.6% of each arm,respectively. No patient died as a result of a
`drug-related AE. The incidence of venous thromboembolic events in
`the fulvestrant and exemestane arms was 1.1% and 0.9%,respectively.
`
`QOL
`QOL was measured with two instruments in this study, the
`FACT-ES and TOIA graph of the mean TOI over time is shown in
`Figure A2 (online only). The mean difference across both instruments
`wasnotsignificant, demonstrating that QOL wasnotstatistically dif-
`ferent between either treatment arms.
`
`DISCUSSION
`
`EFECTis notonly one ofthe largest publishedtrials to date comparing
`hormonaltherapies in HR+ ABC,but also oneofthefirst to specifi-
`cally address the optimal agent to use in sequence immediately after
`progression of a nonsteroidal Al. EFECT confirmedefficacy for both
`fulvestrant and exemestanein this setting, with clinical benefit rates of
`approximately 32% and a median TTP of 3.7 monthsfor both agents.
`The observed durations of response with fulvestrant and exemestane
`(13.5 v 9.8 months, respectively) and durations ofclinical benefit (9.3
`v 8.3 months, respectively), are encouraging for a population of pa-
`tients with relapsed disease after AI treatment. Furthermore,results
`from EFECT support the concept that patients achieving SDlasting at
`least 24 weeks have similar outcomes compared with patients obtain-
`ing a response (Fig A3, online only), even in this previously hormon-
`ally treated population.
`Itis interesting, that for more than 60% ofwomen in EFECT,the
`treating oncologist identified the patient as AI sensitive, but this was
`neither confirmed centrally or by RECISTcriteria. Yet by 6 months,
`approximately 70% oftrial subjects had experienced disease progres-
`sion. This indicates that approximately two thirds of patients did not
`benefit from either hormonal agent, implying that the majority of
`
`1666
`
`© 2008 by American Society ofClinical Oncology
`Information downloaded from jco.ascopubs.org and provided by at AZ Library on September 29, 2010 from 193.132.159.169
`Copyright © 2008 American Society of Clinical Oncology. All rights reserved.
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`AstraZeneca Exhibit 2062 p. 3
`
`

`

`Randomized Trial of Fulvestrant v Exemestane in Advanced Breast Cancer
`
`
`
`Table 1. Baseline Patient and Disease Characteristics
`
`Exemestane (n = 342)
`Fulvestrant (n = 351)
`
`
`
`
`Characteristic No. No.% %
`
`
`
`
`
`189
`162
`
`Dia
`ol3
`145
`206
`147
`87
`190
`128
`25
`224
`127
`
`8
`342
`
`236
`121
`109
`104
`71
`48
`
`197
`154
`
`345
`237
`6
`
`194
`133
`24
`
`
`
`63
`38-88
`
`53.8
`46.2
`
`61.8
`89.2
`41.3
`58.7
`419
`24.8
`54.1
`36.8
`10.0
`63.8
`36.2
`
`23
`97.4
`
`67.2
`34.5
`ciel
`20.6
`20.2
`13.4
`
`56.1
`43.5
`
`98.3
`67.5
`LF
`
`65.3
`3/9
`6.8
`
`Age, years
`Median
`Range
`Age group, years
`< 65 (adult)
`= 65 (elderly)
`Prior treatments
`Adjuvant endocrine therapy*
`Endocrine therapy for advanced diseaset
`1 prior endocrine therapy
`> 1 prior endocrine therapy
`Adjuvant chemotherapy
`Chemotherapy for advanced disease
`Adjuvant radiotherapy
`Radiotherapy for advanced disease
`Other breast cancer treatment
`Al-sensitive disease
`Al-resistant disease
`Disease staget
`Locally advanced
`Metastatic
`Sites of metastases$
`Bone
`Lung
`Liver
`Lymph nodes
`Skin/soft tissue
`Other
`Visceral involvement
`Yes
`No
`Hormone receptor status
`ER+ and/or PgR+
`ER+ and PgR+
`Other
`WHO performance status
`0 (normal activity)
`1 (restricted activity)
`2 (in bed = 50% of the time)
`Measurable disease
`78.9
`270
`76.9
`270
`Yes
`
`
`
`
`81 23,1 72No 21,1
`
`63
`32-91
`
`94
`48
`
`
`
`oo
`294
`47
`95
`68
`74
`71
`42
`29
`210
`eZ
`
`10
`832
`
`227
`124
`110
`We
`58
`56
`
`198
`144
`
`336
`193
`6
`
`181
`149
`12
`
`56.7
`43.3
`
`58.2
`86.0
`43.0
`57.0
`49.1
`21:6
`50.0
`41.5
`8.5
`61.4
`38.6
`
`29
`97.1
`
`66.4
`36.3
`o2.2
`34.2
`17.0
`16.4
`
`57.9
`42.1
`
`98.2
`56.4
`1.8
`
`52.9
`43.6
`3.5
`
`Abbreviations: ER, estrogen receptor; PgR, progesterone receptor, Al, aromatase inhibitor.
`*Thirty-eight patients (10.8%) in the fulvestrant group and 48 (14.0%) patients in the exemestane group received their last non-steroidal Al therapy as adjuvant therapy.
`tThree hundred ten patients (88.3%) in the fulvestrant group and 293 (85.7%) in the exemestane group received their last non-steroidal Al therapy for advanced disease.
`+Disease stage was unknownin one patient in the fulvestrant group (patients subsequently classed as a violator).
`§Patients could have > 1 site of metastases.
`One patient in the fulvestrant group and five patients in the exemestane group had no evidence of meeting the criterion for ER of PgR positivity at baseline and
`so were classed as violators. The remaining five patients in the fulvestrant group and one patient in the exemestane group did not meet this criterion at baseline
`but met it previously and so weren't considered violators.
`
`
`patients enrolled on EFECT had hormone-insensitive disease. In ad-
`dition, in close to 60% of women, the study hormonal agent was
`administered as third-line or greater therapy.All of these factors could
`have contributed to a less-than-optimal clinical efficacy than had been
`hopedfor, and may have undermined the power ofthe study. Indeed,
`in a retrospective analysis looking at TTP in patients who received
`fulvestrant or exemestane as second-line treatment and were deemed
`
`to be sensitive to the prior nonsteroidal AI, the curves do appear to
`
`separate in favor offulvestrant (hazard ratio = 0.73; 99.8% CI, 0.45 to
`1.19; Fig A4, online only). However, the numberofpatients contrib-
`uting to this analysis is small (n = 190), and the results are nonsignif-
`icant as well as being retrospectively derived.
`Whenusedearlier in the hormonal treatment sequence of ER+
`ABC,fulvestrant has demonstrated significantly better clinical out-
`comes than those seen here.Asfirst line therapy fulvestrant was shown
`to be similar to tamoxifen, with a clinical benefit rate of 57% and a
`
`www.jco.org
`
`© 2008 by American Society of Clinical Oncology
`Information downloaded from jco.ascopubs.org and provided by at AZ Library on September 29, 2010 from 193.132.159.169
`Copyright © 2008 American Society of Clinical Oncology. All rights reserved.
`
`1667
`
`AstraZeneca Exhibit 2062 p. 4
`
`

`

`Chia et al
`
`Receptor status
`
`Visceral involvement
`
`Al sensitive/resistant
`
`Measurable disease
`
`Age
`
`No. of prior hormonals
`
`All patients
`
`ER+/PgR+
`Not ER+/PgR+
`With
`Without
`
`Resistant
`Yes
`No
`
`< 65 years
`2 65 years
`1 prior
`22 prior
`
`Sensitive
`
`1.25 151,75
`1
`0.80
`0.60
`0.40
`Hazard ratio (fulvestrant v exemestane) and 95% Cl
`
`Fig 2. Forest plot of effect of predefined covariates on time to progression. ER,
`estrogen receptor; PgR, progesterone receptor.
`
`ER,itresults in reduced nuclear uptake ofthe ER-fulvestrant complex,
`prevention of the ER bindingto the estrogen-responsive genes, and,
`ultimately, downregulation of ERlevels.”*” Given a distinctly differ-
`ent mechanism ofaction, it was rational to assume that a substantial
`degreeofclinical activity would be seen with fulvestrant in this setting.
`The clinical activity seen with fulvestrant in EFECTis similar to those
`in a previously published experience./??*°°
`Whatperhaps was surprising from this study was the clinical
`activity seen with exemestanein this setting: The CBR of 31.5% was
`higher than the 20% CBRreported in a phaseIItrial, even though the
`median TTP was similar.°” EFECT reinforces the notion of incom-
`plete resistance between the nonsteroidal and steroidal Als. This in-
`complete cross-resistance is likely not a result of differences in the
`degree of aromatase inhibition between the Als.*'~* It may be caused
`by the androgenic effects of exemestane.'?7°
`Some questionsstill remain unanswered today in regard to the
`optimaluse of fulvestrant in the treatment of breast cancer. A higher
`dose is currently being investigated in several trials. The combination
`
`= Fulvestrant
`==== Exemestane
`
`100
`
`200
`
`300
`
`400
`
`500
`
`600
`
`700
`
`~=800
`
`Duration of Response (days)
`
`Days
`Fulvestrantatrisk
`Exemestaneatrisk
`
`0
`20
`18
`
`50 100150 200 250 300 350 400 450 500 550 600 650 700
`20
`20
`20
`15
`13
`10
`9
`8
`3
`1
`0
`6
`0
`0O
`18
`18
`17
`15
`12
`9
`56
`5&6
`4
`8
`8
`8
`8
`
`2
`
`ProportionofPatientsResponding
`
`
`
`50 100 150 200 250 300 350 400 450 500 550 600 650 700
`0
`Days
`29
`23
`13
`10
`4
`4
`2
`=O
`351 301191127 89 67
`46
`Fulvestrant atrisk
`24
`21
`13
`10
`8
`8
`6
`2
`Exemestane atrisk 342 305184130 86 56
`37
`
`Fig 1. Kaplan-Meier estimates fortime to progression (TTP). Estimated median
`TTP for patients receiving fulvestrant was 3.7 months, compared with 3.7
`months for patients receiving exemestane (hazard ratio = 0.963; 95% Cl, 0.819
`to 1.133; P = .6531).
`
`Fig 3. Kaplan-Meier estimates for duration of response (DOR; from random
`assignment). Estimated median DOR for patients receiving fulvestrant was 13.5
`months, compared with 9.8 months for patients receiving exemestane.,
`
`1668
`
`© 2008 by American Society ofClinical Oncology
`Information downloaded from jco.ascopubs.org and provided by at AZ Library on September 29, 2010 from 193.132.159.169
`Copyright © 2008 American Society of Clinical Oncology. All rights reserved.
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`AstraZeneca Exhibit 2062 p. 5
`
`
`
`Table 2. Most Commonly Occurring Treatment-Related Adverse Events
`(> 2%incidence in either treatment group)
`Fulvestrant
`Exemestane
`(n = 351)
`(n = 340)
`
`Adverse Event
`No.
`%
`No,
`%
`
`
`
`
`8.2
`28
`g.4
`oa
`njection-site pain
`11.5
`3g
`88
`31
`Hot flashes
`72
`af
`6.8
`24
`ausea
`10.0
`34
`63
`22
`Fatigue
`41
`14
`4.0
`14
`yalgia
`5.6
`19
`Bul
`13
`Arthralgia
`22
`10
`34
`12
`Diarrhea
`2.1
`7
`3.1
`11
`Asthenia
`2.1
`a
`a0
`8
`njection-site reaction
`1.8
`5
`2.3
`8
`Alopecia
`25
`10
`20
`e
`Headache
`2.1
`Z
`2.0
`Z
`Anorexia
`2.1
`?
`09g
`3
`Dyspepsia
`24
`8
`0.3
`1
`Pain in extremity
`
`
`median TTP of 8.2 months.'® In a combinedanalysis of two multi-
`center trials as either first- or second-line therapy in ABC compared
`with anastrozole, fulvestrant demonstrated a clinical benefit rate of
`43.5% and a median TTP of 5.5 months.”Interestingly in a relatively
`small phaseII trial of fulvestrant administered immediately after pro-
`gression during treatment with an AI, in the subset of patients whose
`only prior hormonal therapy was an Al, the clinical benefit rate was
`52.4% (95% Cl, 32.8%to 71.4%).!” Of note, in EFECT, there was no
`difference in either CBR or median TTP between the predefined
`subgroup ofpatients exposed to only oneprior hormonalagentor two
`or more prior hormonalagents.
`Asapure ER antagonist, fulvestrantis in a distinct class of its own
`in regard to its mechanism of action. When fulvestrant binds to the
`
`==" Fulvestrant
`==== Exemestane
`
`100
`
`200
`
`300
`
`400
`
`500
`
`Time to Progression (days)
`
`©®&LL"
`=wnwn®Ee
`a3=
`n&=G
`‘po
`a.e
`3°
`cot3
`-e
`3a
`°=
`a
`
`aw
`
`c=
`
`

`

`Randomized Trial of Fulvestrant v Exemestane in Advanced Breast Cancer
`
` with a “U”are those for which no compensation wasreceived; those
`
`== Population predicted
`@ Observed
`
`a E=
`
`~a
`
`= <3
`
`S
`SBRS)
`-—5
`O
`oO
`S}
`<2
`=o
`@2
`Oo
`
`co
`
`ci
`
`3@_C
`
`:
`
`relationships marked with a “C” were compensated. For a detailed
`description ofthe disclosure categories, orfor more information about
`ASCO’sconflict of interest policy, please refer to the Author Disclosure
`Declarationand the Disclosures ofPotential Conflicts ofInterest sectionin
`Information for Contributors.
`Employmentor Leadership Position: Pamela Rennie, AstraZeneca (C);
`Francisco Sapunar, AstraZeneca (C); Elizabeth Lowe, AstraZeneca (C)
`Consultant or Advisory Role: Stephen Chia, Advisory board for Astra
`Zeneca (C); William Gradishar, AstraZeneca (C), Pfizer (C); Louis
`Mauriac, Novartis (C), AstraZeneca (C); John F.R. Robertson,
`AstraZeneca (C), Pfizer (C); Martine Piccart, AstraZeneca (C), Pfizer
`(C), Novartis (C) Stock Ownership: Pamela Rennie, AstraZeneca;
`Francisco Sapunar, AstraZeneca; Elizabeth Lowe, AstraZeneca
`Honoraria: Stephen Chia, Honoraria for CMEtalks by Astra Zeneca;
`Louis Mauriac, Novartis, AstraZeneca; Jose Bines, AstraZeneca; Aman
`Buzdar, AstraZeneca, Eli Lilly, Roche, Genetech, Pfizer, Taiho; John E.R.
`Robertson, AstraZeneca; Adam Brufsky, AstraZeneca; Kurt Possinger,
`AstraZeneca Research Funding: Stephen Chia,Investigator initiated
`funding with Astra Zeneca; William Gradishar, AstraZeneca; John F.R.
`Robertson, AstraZeneca; Martine Piccart, Novartis, Pfizer Expert
`Testimony: None Other Remuneration: John F.R. Robertson,
`AstraZeneca, Pfizer; Kurt Possinger, AstraZeneca
`
`UOTOMeMEOnEeS
`
`Conception and design: Stephen Chia, Aman Buzdar, Pamela Rennie,
`Francisco Sapunar
`Provision of study materials or patients: Stephen Chia, William
`Gradishar, Louis Mauriac, Jose Bines, Frederic Amant, Miriam Federico,
`Luis Fein, Gilles Romieu, Aman Buzdar, John F.R. Robertson, Adam
`Brufsky, Kurt Possinger, Martine Piccart
`Collection and assembly of data: Stephen Chia, William Gradishar,
`Louis Mauriac, Frederic Amant, John F.R. Robertson, Adam Brufsky,
`Pamela Rennie, Elizabeth Lowe
`Data analysis and interpretation: Stephen Chia, William Gradishar,
`Aman Buzdar, John F.R. Robertson, Adam Brufsky, Pamela Rennie,
`Francisco Sapunar, Elizabeth Lowe, Martine Piccart
`Manuscript writing: Stephen Chia, William Gradishar, Frederic Amant,
`Aman Buzdar, John F.R. Robertson, Adam Brufsky, Pamela Rennie,
`Elizabeth Lowe
`Final approval of manuscript: Stephen Chia, William Gradishar,
`Louis Mauriac, Jose Bines, Frederic Amant, Miriam Federico, Luis Fein,
`Gilles Romieu, Aman Buzdar, John F.R. Robertson, Adam Brufsky,
`Kurt Possinger, Pamela Rennie, Francisco Sapunar, Elizabeth Lowe,
`Martine Piccart
`
`Fig 4. Observed and population-predicted pharmacokinetic profile for the
`fulvestrant loading-dose regimen (500 mg day 0, 250 mg day 14, 250 mg day 28,
`and then 250 mg/month thereafter).
`
`of fulvestrant and an AI compared with an AI alone is another ap-
`proach being studied in several clinical trials. The premise for the
`combination is that fulvestrant may be more effective in a low-
`estrogen environment, which is supported by preclinical data.**
`In conclusion, EFECT has demonstrated clinical activity for both
`LD fulvestrant and exemestane in a meaningful proportion of post-
`menopausal HR+ ABC after progression during treatment with a
`nonsteroidal AI. Both agents were well tolerated, with a similar inc-
`dence ofreported adverse events and quality oflife. There were also no
`apparentpreliminary differences in the proportion of womenreceiv-
`ing chemotherapy (approximately 50%) as the first subsequent sys-
`temic therapy a