`Versus Megestrol Acetate in Postmenopausal Women
`With Advanced Breast Cancer: Results of Overview
`Analysis of Two Phase III Trials
`
`By Aman Buzdar, Walter Jonat, Anthony Howell, Stephen E. Jones, Carl Blomqvist, Charles L Vogel,
`Wolfgang Eiermann, Janet M. Wolter, Mohammed Azab, Alan Webster, and Paul V. Plourde
`for the Arimidex Study Group
`
`Pu
`se: To compare the efficacy and tolerability of an-
`astrozole (1 and 10 mg once daily), a selective, oral, nonste-
`roidal aromatase inhibitor, and megestrol acetate (40 mg
`four times daily),
`in postmenopausal women who pro-
`gressed following tamoxifen treatment.
`Patients and Methods: Two randomized, double-blind
`for anastrozole, open-label for megestrol acetate, parallel-
`group, multicenter trials were conducted in 764 patients.
`Because both trials were identical in design, an analysis of
`the combined results was performed to strengthen interpre-
`tation of results from each trial.
`Results: The median follow-up duration was approxi-
`mately 6 months. The estimated progression hazards ratios
`were 0.97 (97.5% confidence interval [CI], 0.75 to 1.24) for
`anastrozole 1 mg versus megestrol acetate and 0.92
`(97.5% CI, 0.71 to 1.19) for anastrozole 10 mg versus meg-
`estrol acetate. The overall median time to progression was
`approximately 21 weeks. Approximately one third of pa-
`tients in each group benefited from treatment. Twenty-
`seven patients (10.3%) in the anastrozole 1-mg group, 22
`(8.9%) in the anastrozole 10-mg group, and 20 (7.9%) in
`the megestrol acetate group had a complete or partial re-
`sponse, and 66 (25.1%), 56 (22.6%), and 66 (26.1%) pa-
`
`AMOXIFEN CITRATE (Nolvadex; Zeneca Pharma-
`ceuticals, Wilmington, DE), is considered the first-
`line hormonal therapy for postmenopausal women with ad-
`vanced breast cancer.1 In the adjuvant treatment of women
`with early—stage breast cancer, tamoxifen treatment pro-
`
`From the M.D. Anderson Cancer Center, The University of Texas,
`Houston; Texas Oncology, PA and Baylor University Medical Cen—
`ter, Dallas, TX; University Womens Hospital, Eppendorjf, Hamburg;
`Womens Hospital, Munich, Germany; The Christie Hospital, Man—
`chester, United Kingdom; University Central Hospital, Helsinki,
`Finland; The Comprehensive Cancer Research Group Inc, North
`Miami Beach, FL; Rush-Presbyterian-St Luke‘s Medical Center.
`Chicago, IL; and Zeneca Pharmaceuticals, Wilmington, DE and
`Macclesfield, United Kingdom.
`Submitted October 10, I 996; accepted January 29, 1996.
`Supported by a grant from Zeneca Pharmaceuticals, Wilmington,
`DE and Macclesfield, United Kingdom.
`Address reprint requests to Aman Buzdar, MD, Department of
`Breast Medical Oncology, Box 56, The University of Texas, M.D.
`Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030;
`Email Aman—Buzdar@isqm.mda.uth.tmc.edu.
`© 1996 by American Society of Clinical Oncology.
`0732—183X/96/1407—0005$3.00/0
`
`tients, respectively, had stable disease for a 24 weeks. For
`all end points, individual trial results were similar to the
`results of the combined analysis. Anastrozole and meges-
`trol acetate were well tolerated. Gastrointestinal distur-
`bance was more common among patients in the an-
`astrozole groups than the megestrol acetate group; the
`difference between the anastrozole 10 mg and megestrol
`acetate groups was significant (P = .005). Significantly
`fewer patients in the anastrozole 1-mg (P < .0001) and 10-
`mg (P < .002) groups had weight gain than in the megestrol
`acetate group. More than 30% of megestrol acetate-
`treated patients had weight gain 2 5%, and 10% of pa-
`tients had weight gain 2 10%. Patients who received meg-
`estrol acetate continued to gain weight over time.
`Conclusion: Anastrozole,
`1 and 10 mg once daily, is
`wel tolerated and as effective as megestrol acetate in the
`treatment of postmenopausal women with advanced
`breast cancer who progressed following tamoxifen treat-
`ment. Moreover, anastrozole therapy avoids the weight
`gain associated with megestrol acetate treatment.
`J Clin Oncol 14:2000-201 I. © 1 996 by American So-
`ciety of Clinical Oncology.
`
`longs survival,2 extends disease—free survival,3 reduces the
`incidence of new primary breast tumors,3 and offers a favor-
`able long-term tolerability profile.3 Because a significant
`number of patients who receive tamoxifen for advanced or
`early—stage breast cancer will have disease progression, new
`therapies that are effective and well tolerated are needed
`to treat women with advanced breast cancer.
`
`Two classes of drugs, progestins and aromatase inhibi—
`tors, are frequently prescribed for postmenopausal women
`who have progressed following tamoxifen therapy. Pro-
`gestins are effective. but their use is associated with com-
`mon side effects such as weight gain, which has been
`reported to occur in up to 64% of patients.4 Megestrol
`acetate (Megace; Bristol—Myers Squibb, Princeton, NJ)
`the most commonly used progestin,
`is associated with
`nausea and vomiting, hot
`flashes, vaginal bleeding,
`edema, hypercalcemia, rash, heart failure, hypertension,
`thrombocytopenia, depression, and Cushingoid symp—
`toms, in addition to weight gain.5 Aminoglutethimide was
`the first aromatase inhibitor to be evaluated in clinical
`
`trials, and is commercially available for the treatment
`of breast cancer in Europe and Canada. The efficacy of
`aminoglutethimide is similar to other endocrine agents,
`
`2000
`
`Journal of Clinical Oncology, Vol 14, No 7 (July), 1996: pp 2000-201 1
`
`on October 27, 2010 from 193.132.159.169
`Information downloaded from jco.ascopubs.org and provided by at AZ Library
`Copyright © 1996 American Society of Clinical Oncology. All rights reserved.
`
`AstraZeneca Exhibit 2119 p. 1
`InnoPharma Licensing LLC V. AstraZeneca AB IPR2017-00905
`
`
`
`ANASTROZOLE v MEGESTROL ACETATE FOR BREAST CANCER
`
`2001
`
`but it is nonselective and inhibits adrenal synthesis of
`glucocorticoids and mineralocorticoids and requires con-
`current administration of hydrocortisone.
`In addition,
`aminoglutethimide is associated with frequent adverse ef—
`fects that have limited its clinical use.5’6 A selective
`
`aromatase inhibitor, formestane, which is commercially
`available in some European countries, has no effect on
`steroidogenesis, but requires administration as a deep in-
`tramuscular injection and has been associated with injec-
`tion-site reactions.7 Since the development of aminoglu-
`tethimide and formestane, a series of aromatase inhibitors
`
`has been synthesized; many of these compounds are cur-
`rently under clinical investigation. Enhanced potency and
`specificity, as well as reduced side effects, are desired
`characteristics of new aromatase inhibitors in develop—
`ment.8
`Anastrozole (Arimidex, Zeneca Pharmaceuticals), an
`achiral triazole derivative,
`is a new nonsteroidal, oral
`
`aromatase inhibitor with highly effective and selective
`activity for the aromatase enzyme? Anastrozole has been
`investigated in a clinical trial program as a therapy for
`postmenopausal women with advanced breast cancer. In
`clinical pharmacology trials, circulating serum estradiol
`concentrations, measured by a sensitive assay, were con-
`sistently suppressed to the limit of quantification of the
`assay with daily anastrozole doses of 1 mg and higher?
`Anastrozole was rapidly absorbed after oral administra-
`tion, with maximal plasma concentrations occurring
`within 2 hours,9 and it possesses a half-life that supports
`once—daily oral administration.9
`In this report, we present the results of two phase III
`trials that compared the efficacy and tolerability of an-
`astrozole
`and megestrol
`acetate in postmenopausal
`women with advanced breast cancer who progressed after
`tamoxifen treatment. Because both trials were identical
`
`in design, an analysis of the combined results was per-
`formed, which thereby strengthened interpretation of re-
`sults from each trial. Anastrozole doses of 1 and 10 mg
`once daily were chosen for evaluation in these trials. The
`l—mg daily dose was selected because it was the lowest
`dose of anastrozole to give maximum-detectable reduc—
`tion of serum estradiol concentrations. In the absence of
`
`any safety concerns, the lO—mg daily dose was evaluated
`to determine whether a higher dose would offer enhanced
`antitumor activity and increased clinical benefit.
`
`PATIENTS AND METHODS
`
`Study Design
`The two trials were randomized, double—blind for anastrozole,
`open-label for megestrol acetate, parallel-group, and multicenter
`studies. One trial was conducted at sites in North America (hereafter
`
`referred to as the North American trial), and one at sites in Europe,
`Australia, and South Africa (hereafter referred to as the European
`trial). Both trials compared the efficacy and tolerability of an—
`astrozole l and 10 mg daily with megestrol acetate 40 mg four times
`daily for the treatment of postmenopausal women with advanced
`breast cancer. The primary objectives were to compare two dosages
`of anastrozole with megestrol acetate with respect to time to disease
`progression, tumor response, and tolerability. The secondary objec-
`tives were to compare the treatment groups with respect to time to
`treatment failure, response duration, and survival.
`
`Patient Population
`To enter the trials, patients were required to have progressed
`while receiving tamoxifen or other antiestrogen therapy for advanced
`breast cancer or relapsed during or after receiving adjuvant tamoxi-
`fen treatment; be postmenopausal, defined as having nonfunctioning
`ovaries through natural menopause or surgical, radiation, or chemical
`castration (women > 50 years of age who did not menstruate during
`the preceding 12 months were considered postmenopausal, whereas
`women < 50 years of age had to have a follicle-stimulating hormone
`concentration > 40 IU/L to enter); and have a World Health Organi-
`zation (WHO) performance status score 5 2. Patients were excluded
`if they had estrogen receptor—negative breast cancer (except when
`the patient had shown a previous response to tamoxifen treatment),
`exposure to more than one previous course of cytotoxic therapy for
`advanced disease (except adjuvant chemotherapy), exposure to more
`than one previous hormonal therapy for advanced breast cancer, or
`any concurrent medical illness or laboratory abnormalities that would
`compromise safety or prevent interpretation of results. Written in-
`formed conSent was obtained from all patients, and the studies were
`approved by the appropriate institutional review board at each site.
`
`Treatment Program
`Anastrozole was supplied as film-coated, white tablets that con—
`tained either 1 or 10 mg of drug. Megestrol acetate was supplied as
`white, circular, scored tablets that contained 40 mg of drug. Patients
`were randomly allocated to one of three oral treatment regimens: 1
`mg of anastrozole once daily, 10 mg of anastrozole once daily, or
`40 mg of megestrol acetate four times daily. Treatment continued
`until disease progression or until withdrawal from treatment for any
`reason other than progression. Patients who had disease progression
`were permitted to receive either cytotoxic therapy or other hormonal
`treatments. When patients withdrew before progression, they were
`monitored for time to progression.
`Baseline screening assessments were completed within the 4
`weeks before randomization. On day 1, the date of randomization,
`eligible patients underwent a complete physical examination. Each
`patient’s disease was assessed clinically every 4 weeks for the first
`24 weeks of treatment, and then every 12 weeks until week 48.
`After week 48, assessments were performed every 3 months until
`disease progression was detected. Bone scans were repeated every
`24 weeks until disease progression or withdrawal. Radiographic ex-
`amination of confirmed metastatic lesions was repeated every 12
`weeks (or earlier when clinically indicated) during treatment and at
`withdrawal.
`Patients were withdrawn from active treatment for a serious ad-
`verse event, noncompliance with protocol procedures, unwilling or
`inability to continue the trial, withdrawal by an investigator, or clini-
`cally significant breast cancer progression. All patients who were
`withdrawn were monitored for survival.
`
`on October 27, 2010 from 193.132.159.169
`Information downloaded from jco.ascopubs.org and provided by at AZ Library
`Copyright © 1996 American Society of Clinical Oncology. All rights reserved.
`
`AstraZeneca Exhibit 2119 p. 2
`
`
`
`2002
`
`BUZDAR ET AL
`
`Efi‘icacy Assessments
`Assessments of tumor response included the evaluation of both
`measurable and nonmeasurable disease. Measurable disease was de-
`fined as the presence of metastatic lesions measurable in one or
`two dimensions using physical or radiographic methods (including
`computed tomography scan) and osteolytic bone lesions. For measur-
`able lesions, only physical or radiologic measurements were re-
`corded. To ensure consistency and objectivity in the assignment of
`response categories, a computerized algorithm was used to assign
`responses based on the measurements. The program strictly applied
`the protocol definition of response based on Union Internationale
`Contre 1e Cancer (UICC) criteria.10 Nonmeasurable disease was de-
`fined as single metastatic lesions smaller than 0.5 cm, malignant
`pleural effusion or ascites, positive bone scan, and osteoblastic bone
`lesions. For nonmeasurable lesions, partial responses were not per-
`mitted to be assigned, in accordance with the strict criteria for assess-
`ment. Therefore, responses were assigned only in the categories of
`complete response, stable disease, or progressive disease.
`The best objective response over time was determined on the basis
`of objective responses at each visit. Complete or partial responses
`were assigned only when noted on successive visits at least 4 weeks
`apart. Measurable lesions of bone, chest, and abdomen were assessed
`at 12-week intervals. A best response of stable disease was assigned
`when responses of stable disease or better were obsewed for at least
`24 weeks. If such responses had been observed for less than 24
`weeks because a patient did not have measurements for 24 weeks
`at the time of data cutoff, then a best response of stable disease for
`less than 24 weeks was recorded.
`Time to progression, time to treatment failure, time to death, and
`duration of response were calculated from the date of randomization.
`Time to progression represented the time to objective disease pro—
`gression or death, whichever occurred first. Patients who had not
`reached progression at the time of data cutoff were right—censored
`in the analysis at the time of their latest visit. Time to treatment
`failure was the time to earliest occurrence of progression, death, or
`withdrawal. Time to death represented the number of days until
`death from any cause. Duration of response, which was recorded
`for those with either a complete or partial response, was the time to
`objective progression or death.
`
`Quality-of-Life Assessments
`The primary quality—of—life assessment was the validated Rotter-
`dam Symptom Checklist.H Other quality—of—life variables that were
`scored and recorded were the types of analgesics used, severity of
`bone pain, and performance status or level of daily activity. Because
`of differences between the two trials in the frequency at which the
`Rotterdam dimensions were collected and differences in the relative
`time frame in the wording of subjective scores, data from quality—
`of—life assessments from the two trials were not combined.
`
`Tolerability Assessments
`Any detrimental change in a patient’s condition after the trial
`began and during any follow-up period, unless related to disease
`progression, was considered an adverse event. Patients were solicited
`indirectly for adverse events; prompted by a question, each patient
`described anything that had bothered her. In addition to monitoring
`for adverse events, routine laboratory tests results were performed
`at baseline, at selected times during therapy, and at withdrawal.
`The results of clinical laboratory tests were reviewed for clinically
`
`relevant changes. Physical examinations were performed and weight,
`blood pressure, and pulse were recorded at baseline, at selected times
`during therapy, and at withdrawal.
`
`Statistical Analysis
`The trials were designed to compare anastrozole and megestrol
`acetate using time to progression as the primary end point. A popula-
`tion of 300 patients (100 in each treatment group) in each trial was
`deemed sufficient to detect a treatment difference of approximately
`14 weeks in median time to progression with 80% power and a two—
`sided alpha level of 0.05, assuming a median time to progression of
`26 weeks and a minimum follow-up time of 6 months.
`To protect against an imbalance in treatment allocation across
`centers, the randomization scheme was stratified for center in each
`trial. In addition, treatments were allocated in blocks of size three
`in the North American trial and six in the European trial, such that
`treatment groups were balanced after every three or six patients at
`each center.
`Efficacy analyses were analyzed on the basis of the treatment to
`which the patients were randomly assigned (intention-to—treat basis).
`Cox’s proportional hazards model was used to analyze time to dis—
`ease progression, time to treatment failure, and time to death. Logis-
`tic regression was used to analyze response data. All efficacy analy-
`ses were adjusted for the covariates of previous treatment status
`(adjuvant or for advanced disease) and hormone receptor status. The
`combined estimate of the treatment effect for a time—to-event variable
`for either dose of anastrozole compared with megestrol acetate was
`derived by fitting a Cox proportional hazards model with trial and
`treatment as covariates and then testing for significance of treatment.
`Log hazards ratios and standard errors were estimated and were
`used to calculate confidence intervals on the hazards ratio. Upper
`confidence limits 2 1.25 for a hazards ratio of either dose of an-
`astrozole to megestrol acetate would allow an inference that the
`effects of anastrozole were not substantially inferior to the effects
`of megestrol acetate (ie, an upper confidence limit of 1.25 was
`considered to represent equivalence between anastrozole and meges-
`trol acetate).
`Additional analyses were performed to assess the effects of the
`prognostic factors of prior hormonal treatment history, presence or
`absence of measurable disease, and presence or absence of visceral
`disease on time to progression and time to treatment failure. Likeli-
`hood ratio tests were performed to rule out qualitative interactions
`when treatment by prognostic factor interactions existed. Because
`the two anastrozole groups were compared with the megestrol acetate
`group, Bonferonni adjustments were made for the analyses of each
`end point. For tumor response data, an approach similar to the
`method outlined earlier was used.
`
`Safety analyses were performed according to the treatment actu-
`ally received. Adverse events that might be expected to occur on
`the basis of the pharmacology of anastrozole and megestrol acetate
`were prospectively identified and analyzed;
`these adverse events
`included weight gain, edema, thromboembolic disease, gastrointesti—
`nal disturbance, hot flushes, and vaginal dryness. Objective measure—
`ments of weight gain were also analyzed; the proportion of patients
`who had weight gain of greater than 5% and greater than 10% during
`the study compared with pretherapy were assessed. The incidence
`of adverse events was compared between patients treated with an—
`astrozole l mg and those treated with megestrol acetate and between
`those treated with anastrozole 10 mg and those given megestrol
`acetate. Fisher’s exact test was used for the statistical comparisons;
`
`on October 27, 2010 from 193.132.159.169
`Information downloaded from jco.ascopubs.org and provided by at AZ Library
`Copyright © 1996 American Society of Clinical Oncology. All rights reserved.
`
`AstraZeneca Exhibit 2119 p. 3
`
`
`
`ANASTROZOLE v MEGESTROL ACETATE FOR BREAST CANCER
`
`2003
`
`a two-sided alpha level of 0.01 was used to allow for multiple
`comparisons.
`Interim analyses of each trial were performed in 1994 to enable
`independent data-monitoring committees to evaluate periodically ef—
`ficacy and safety data from the two trials and recommend that the
`trials be continued or stopped, or recommend a change to the study
`design. In the North American trial, two interim analyses of objective
`response and time to progression were performed, whereas in the
`European trial one interim analysis was performed. In each trial, the
`O’Brien and Fleming adjustment was used in the analysis of both
`objective response and time to progression; the significance level
`for all end points was adjusted using the Bonferroni method. After
`reviewing interim results, the independent committees monitoring
`the two trials recommended that each of the trials be continued.
`
`RESULTS
`
`Patient Characteristics
`
`Seven hundred sixty-four patients from 49 centers in
`North America and 73 centers in Europe, Australia, and
`South Africa were entered into the two tlials and random-
`
`ized to one of the three treatment groups. The groups
`formed by randomization were well balanced with respect
`to demographic and pretreatment characteristics (Table
`1). Although there was an apparent imbalance in treat-
`ment allocation for the three groups, it was believed to
`be an artifact related to the large proportion of centers in
`the European trial in which the total number of patients
`recruited was not divisible by six. (Treatments were allo—
`cated in blocks of six in the European trial, compared
`with blocks of three in the North American trial.) Three
`patients did not receive therapy, and one patient who was
`randomized to 1 mg of anastrozole received 10 mg of
`anastrozole. A11 764 patients were included in the efficacy
`analyses in accordance with the intention-to-treat ap-
`proach. The median follow-up duration was approxi-
`mately 6 months.
`
`Time to Progression
`
`Results for time to disease progression showed both
`doses of anastrozole to be equivalent to megestrol acetate.
`The individual trial results were similar to the combined
`
`results. Progression of disease occurred in 159 patients
`(60%) in the anastrozole 1-mg group, 146 (59%) patients
`in the anastrozole 10-mg group, and 163 (64%) patients
`in the megestrol acetate group. The comparisons of the
`l-mg and 10—mg of anastrozole versus megestrol acetate
`groups did not differ significantly with respect to time to
`progression. The estimated progression hazards ratio for
`anastrozole 1 mg versus megestrol acetate was 0.97
`(97.5% confidence interval [CI], 0.75 to 1.24). Similarly,
`the estimated progression hazards ratio for anastrozole 10
`mg versus megestrol acetate was 0.92 (97.5% CI, 0.71
`to 1.19). No differences could be detected between the
`
`two anastrozole doses and megestrol acetate. The overall
`median time to progression was approximately 21 weeks.
`A Kaplan—Meier plot of time to progression is presented
`in Fig 1.
`
`Tumor Response
`
`The best objective tumor response rates for the 1- and
`10—mg anastrozole groups did not differ significantly from
`that for the megestrol acetate group. The individual trial
`results were similar to the combined results. Approxi-
`mately one third of patients in each treatment benefited
`from therapy. Twenty-seven patients (10.3%) in the an-
`astrozole 1-mg group, 22 (8.9%) in the anastrozole 10—
`mg group, and 20 (7.9%) in the megestrol acetate group
`had either a complete or partial response to treatment,
`and 66 (25.1%), 56 (22.6%), and 66 (26.1%) patients in
`the respective groups had stable disease for 2 24 weeks
`(Table 2). The estimated odds ratio for response for an-
`astrozole 1 mg versus megestrol acetate was 1.32, with
`a 97.5% CI of 0.66 to 2.65. Similarly, the estimated odds
`ratio for response for anastrozole 10 mg versus megestrol
`acetate was 1.15, with a 97.5% CI of 0.55 to 2.36. Tumor
`
`responses were observed in all sites of disease, with the
`best responses in patients with soft tissue disease (com-
`plete or partial responses of 34% anastrozole 1 mg, 28%
`anastrozole 10 mg, and 27% megestrol acetate). The per—
`centage of patients with soft tissue disease at entry was
`low: 12% and 15% of patients in the anastrozole 1- and
`10-mg groups and 16% of patients in the megestrol ace-
`tate group. Responses were observed in patients who pro-
`gressed after receiving adjuvant tamoxifen, as well as in
`patients who received tamoxifen for advanced disease.
`The duration of response ranged from approximately
`3 to 18 months in the three treatment groups. Among all
`responders, 2 70% were without progression for at least
`24 weeks (74% anastrozole 1 mg, 81% anastrozole 10
`mg, and 70% megestrol acetate). These percentages repre—
`sented a conservative estimate, as many of the patients
`were continuing treatment and were censored at the time
`of data cutoff.
`
`Treatment Failure
`
`Treatment failure occurred in 169 patients (64%) in
`the anastrozole l—mg group, 160 (65%) in the anastrozole
`10-mg group, and 174 (69%) in the megestrol acetate
`group. For the majority of patients who reached treatment
`failure in each treatment group (51% anastrozole 1 mg,
`53% anastrozole 10 mg, and 55% megestrol acetate), the
`reason for treatment failure was disease progression.
`Therefore, the inferences for time to treatment failure are
`the same as those generated for time to progression, and
`
`on October 27, 2010 from 193.132.159.169
`Information downloaded from jco.ascopubs.org and provided by at AZ Library
`Copyright © 1996 American Society of Clinical Oncology. All rights reserved.
`
`AstraZeneca Exhibit 2119 p. 4
`
`
`
`Megestrol Acetate
`(n = 253)
`
`No.
`
`%
`
`65
`39-90
`
`67
`42-130
`
`89
`
`165
`
`237
`89
`156
`
`1 19
`34
`20
`5
`1 1
`64
`
`148
`
`102
`
`151
`16
`30
`60
`16
`29
`
`1 16
`103
`32
`1
`1
`180
`73
`
`92
`156
`1 33
`41
`212
`3
`
`41
`77
`38
`94
`3
`
`94
`35
`62
`
`47
`13
`8
`2
`4
`25
`
`1 1
`20
`40
`1 1
`19
`
`46
`41
`13
`<1
`<1
`71
`29
`
`36
`62
`45
`16
`84
`1
`
`16
`30
`15
`37
`1
`
`Table 1. Demographic and Pretreatment Characteristics
`Anastrozole
`
`Parameter
`
`1 mg (n = 263)
`No.
`
`96
`
`10 mg (n = 248)
`No.
`
`%
`
`65
`2997
`
`68
`31-112
`
`93
`
`128
`
`346
`98
`1 53
`
`134
`45
`14
`
`62
`
`128
`
`121
`
`137
`12
`22
`64
`12
`27
`
`138
`91
`34
`
`191
`72
`
`99
`166
`124
`46
`217
`
`32
`75
`42
`1 O7
`
`94
`37
`58
`
`51
`
`NO:
`
`24
`
`16
`47
`
`20
`
`53
`35
`1 3
`
`73
`27
`
`38
`63
`47
`18
`83
`
`12
`29
`16
`41
`
`66
`41 -91
`
`69
`35-131
`
`96
`
`133
`
`230
`92
`146
`
`1 15
`34
`17
`4
`1 2
`66
`
`138
`
`92
`
`148
`17
`33
`59
`15
`24
`
`109
`101
`34
`4
`0
`168
`80
`
`95
`149
`102
`38
`210
`14
`
`36
`66
`34
`98
`14
`
`93
`37
`59
`
`46
`14
`7
`2
`5
`27
`
`12
`22
`40
`10
`16
`
`44
`41
`14
`2
`0
`68
`32
`
`38
`60
`41
`15
`85
`6
`
`15
`27
`14
`40
`6
`
`Age, years
`Mean
`
`Range
`Weight' (kg)
`Mean
`
`Range
`Previous treatment
`
`Surgery
`Cytotoxic chemotherapy
`Radiotherapy
`Receptor status
`ER+, PR+
`ER+, PR—
`ER+, PR unknown
`ER—, PR+
`ER—, PR—
`Unknown
`Duration 01 tamoxiten treatment tor advanced disease‘l’t
`No. oi patients
`Median disease-tree interval, weeks
`Relapsed during adiuvant tamoxifen treatment‘H
`No. of patients
`Median disease-tree interval, weeks
`Previous best response to tamoxifen for advanced
`disease
`
`No. ot patients
`Complete
`Partial
`Stable disease
`Progression
`Unknown
`
`WHO performance status score
`
`01 23 A
`
`Measurable disease
`No measurable disease
`Sites of metastatic disease§
`Soft tissue
`Bone
`Visceral
`Liver
`No evidence of liver involvement
`No assessable metastatic disease”
`Extent of metastatic diseasefl
`Soft tissue only
`Bone only
`Visceral only
`Mixed
`No assessable metastatic disease”
`
`Abbreviations: ER, estrogen receptor; PR, progesterone receptor.
`‘Weight was recorded for 256 patients in the anastrozole 1-mg group, 235 in the anastrozole 10-mg group, and 244 in the megestrol acetate group.
`TFor treatment at primary disease (alter mastectomy or lumpectomy) and metastatic lesions.
`*Patients who did not receive tamoxifen and For whom duration at treatment could not be calculated are not included.
`§Patients may be in > 1 category.
`“Includes patients with excised or irradiated local or distant disease at entry, patients with local or distant metastases that were excised or eradicated
`before entry, and 1 patient who had no assessable disease.
`filCategories are mutually exclusive.
`
`Information downloaded from jco.ascopubs.org and provided by at AZ Library on October 27, 2010 from 193.132.159.169
`Copyright © 1996 American Society of Clinical Oncology. All rights reserved.
`
`AstraZeneca Exhibit 2119 p. 5
`
`
`
`ANASTROZOLE v MEGESTROL ACETATE FOR BREAST CANCER
`
`2005
`
`1.0
`
`0.9
`
`
`Fig l. Kaplan-Meier proba-
`bility of time to progression in
`patients in the anastrozole l-mg,
`anastrozole lO-mg, and meges-
`trol acetate groups.
`
`
`
`PROPORTIONNOTPROGRESSlNG
`
`0.8
`
`0.7
`
`0.6
`
`0.5
`
`0.4
`
`0.3
`
`0.2
`
`0.1
`
`-———’|elel<—— Anastrozole, 1 mg
`
`—l—H—— Anastrozole, 10 mg
`
`—o+o—-—~ Megestrol Acetate
`
`
`
`do'
`TIME (DAYS):
`0
`
`50
`
`AT RISK:
`
`764
`
`684
`
`100
`
`413
`
`150
`
`305
`
`200
`
`177
`
`250
`
`146
`
`300
`
`63
`
`350
`
`46
`
`400
`
`25
`
`450
`
`13
`
`500
`
`550
`
`600
`
`2
`
`conclusions drawn from these data were identical to those
`
`for time to progression. The individual trial results for
`time to treatment failure were similar to the combined
`results.
`
`Survival
`
`Thirty-eight patients (14%) in the anastrozole l-mg
`group, 32 (13%) in the anastrozole lO—mg group, and 47
`(19%) in the megestrol acetate group died by the time of
`data cutoff. The comparisons of the anastrozole 1- and
`lO—mg groups versus the megestrol acetate treatment
`group did not differ significantly with respect to overall
`survival. The individual trial results were similar to the
`
`combined results. When adjusted for trial differences with
`a Cox covariate analysis, the estimated death hazards ratio
`for anastrozole 1 mg versus megestrol acetate was 0.80,
`with a 97.5% CI of 0.49 to 1.30. Similarly,
`the death
`
`Table 2. Obiective Tumor Responses in Patients in the Anastrozole
`l-mg, Anastrozole lO-mg, and Megestrol Acetate Groups
`Anastrozcle
`
`Obiective Response
`
`1 mg
`(n = 263)
`No.
`%
`
`10 mg
`(n = 248)
`No.
`%
`
`Megestrol
`Acetate
`(n = 253]
`No.
`%
`
`2
`5
`2
`4
`2
`6
`Complete response
`6
`15
`7
`18
`8
`21
`Partial response
`26
`66
`23
`56
`25
`66
`Stable disease 2 24 weeks
`12
`31
`18
`45
`l l
`29
`Stable disease < 24 weeks‘
`
`Progression 54 141 54 l 25 50 1 36
`
`
`
`
`
`‘Represents patients who had stable disease but had tumor assessments
`taken at < 24 weeks at the time at data cutol't.
`
`hazards ratio for anastrozole 10 mg versus megestrol ace-
`tate was 0.71, with a 97.5% CI of 0.42 to 1.18. Because
`
`few deaths occurred by the date of data cutoff, Kaplan—
`Meier estimates of median survival could not be calcu-
`lated.
`
`Efi‘ects of Prognostic Factors
`
`There was no evidence that patients from any of the
`subgroups (patients who received prior adjuvant therapy
`with tamoxifen v those who received tamoxifen for ad—
`
`vanced disease, patients with measurable disease v those
`with nonmeasurable disease, and patients with visceral
`disease v those with no visceral disease) did not receive
`similar benefit from either 1 or 10 mg of anastrozole
`compared with megestrol acetate. The results of the anal—
`ysis for time to progression and time to treatment failure
`were similar for these subgroups.
`
`Tolerability
`
`Table 3 lists common adverse events for the three treat-
`
`ment groups. The most common events with anastrozole
`1 mg were asthenia, nausea, headache, hot flushes, pain,
`and back pain. Other than one episode of dyspnea, none
`of these events led to withdrawal of therapy. The range
`of adverse events in patient treated with anastrozole 10
`mg was similar to that
`in patients who received an-
`astrozole 1 mg. Nausea and vomiting were more common
`in patients who received anastrozole 10 mg, which sug-
`gests the possibility that these effects were dose-related.
`Episodes of nausea and vomiting in anastrozole-treated
`patients were generally of mild or moderate severity and
`
`on October 27, 2010 from 193.132.159.169
`Information downloaded from jco.ascopubs.org and provided by at AZ Library
`Copyright © 1996 American Society of Clinical Oncology. All rights reserved.
`
`AstraZeneca Exhibit 2119 p. 6
`
`
`
`2006
`
`BUZDAR ET AL
`
`Table 3. Common Adverse Events, Regardless of Causality, Reported
`in 10% of Patients in the Anastrozole 1-mg, Anastrozole IO-mg,
`and Megestrol Acetate Groups
`Anastrozole
`
`over time (Fig 2). None of the patients in the megestrol
`acetate group withdrew from therapy because of weight
`gain.
`Mean blood pressure and pulse were similar in all three
`treatment groups at entry; during the two trials, there were
`only minor changes in these parameters in each treatment
`group.
`
`DISCUSSION
`
`The objectives of the two trials were to compare the
`efficacy and tolerability of anastrozole 1 and 10 mg once
`daily and megestrol acetate 40 mg four times daily in
`postmenopausal women with breast cancer who had re-
`ceived prior therapy with tamoxifen. To strengthen the
`interpretation of results from each trial, an overview anal-
`yses of results was p