`© 1999 Elsevier Sciaice Ltd. All rights reserved
`Printed in Great Britain
`0959-8049/99/35 - see front matter
`
`Pergamon
`
`
`
`9 .A :‘
`(W,
`
`
`
`PII: S0959-8049(98)00388-8
`
`Original Paper
`
`Onapristone, a Progesterone Receptor Antagonist, as First-line
`Therapy in Primary Breast Cancer
`
`].F.R. Robertson} P.C. '\Willsher,1 L. \X/'interbottom,1 R.W. Blameyl and S. Thorpe2
`
`1Professorial Unit of Surgery, City Hospital, Hucknall Road, Nottingham, NG5 lPB; and 2Schering Health Care,
`Burgess Hill, West Sussex, U.K.
`
`The progesterone receptor antagonist, Onapristone, is an elfective endocrine agent in experimental
`breast cancer models. This study aimed to investigate this agent as first-line endocrine therapy in
`patients with breast cancer. However, owing to the recognition in this and other clinical studies that
`some patients on Onapristone developed liver function test abnorrnalities, the development of this
`drug and recruitment to the study stopped in 1995. 19 patients either with locally advanced breast
`cancer (n = 12) or who were elderly, unfit patients with primary breast cancer (n= 7) received Ona-
`pristone 100 mglday. Seventeen of the 19 tumours expressed oestrogen receptors (ER) whilst 12 of the
`18 tumours tested expressed progesterone receptors (PgR). Tumour remission was categorised by
`International Union Against Cancer criteria. One patient was withdrawn after 4.5 months while her
`disease was static. Of the remaining 18 patients, 10 (56%) showed a partial response and 2 (11‘Vo) dur-
`able static disease (2 6 months), giving an overall tumour remission rate of 67%. The median duration
`of remission was 70 weeks. Transient liver function test abnorrnalities developed in a number of
`patients, mainly during the first 6 weeks of treatrnent. In conclusion Onapristone can induce tumour
`responses in human breast cancer. © 1999 Elsevier Science Ltd. All rights reserved.
`
`Key words: breast cancer, progesterone antagonists, onapristone
`Eur} Cancer, Vol. 35, N0. 2, pp. 214-218, 1999
`
`INTRODUCTION
`BREAST CANCER is the most common female malignancy in the
`western world, with a lifetime risk of at least 1:12 at present.
`At some point in their disease most patients will receive sys-
`temic therapy. The endocrine therapy of choice in post-
`menopausal
`or oophorectomised patients
`is
`the
`anti-
`oestrogen, tamoxifen [1]. Tamoxifen gives a tumour remis-
`sion rate of approximately 60% in selected patients [2], e.g.
`those with tumours which are oestrogen receptor (ER) and
`progesterone receptor (PgR) positive. However,
`tamoxifen
`has a partial oestrogen agonistic activity on some tissues and
`organs. The second—line hormonal agents used in relapse
`after tamoxifen include aromatase inhibitors (e.g. Aminoglu-
`tethamide, Lentaron and, Arimidex) and high-dose proges-
`tins
`(e.g. medroxyprogesterone
`acetate
`and megestrol
`acetate). These drugs are less well tolerated than tamoxifen.
`
`Correspondence to ].F.R. Robertson,
`nottingham.ac.ul<
`Received 20 Jul. 1998, accepted 8 Oct. 1998.
`
`e—mail:
`
`iohn.robertson@
`
`Progesterone antagonists offer a new therapeutic strategy
`in the treatment of invasive breast cancer. Furthermore, if the
`molecular studies on the effects of these compounds on
`breast cancer in animals are confirmed their potential clinical
`utility may extend back into preneoplastic disease (eg. atypi-
`cal ductal hyperplasia).
`The progesterone antagonist Onapristone was developed
`by Schering AG [3, 4]. It was reported to have strong anti-
`progestational and antitumour activity. Onapristone (ZK
`98.299) showed tumour inhibitory effects in several hor-
`mone-dependent mammary tumours in animal models. Its
`antitumour activity is as potent or even more potent than that
`of tamoxifen or oophorectomy in the MXT mammary
`tumour of the mouse and DMBA- and NMU-induced
`mammary tumours of the rat [5,6]. Although binding to
`tumour progesterone receptors is a prerequisite for its anti-
`proliferative effects, there is evidence that the mechanism of
`its antitumour effects does not depend on a classical anti-
`hormonal mechanism. While the mechanism of action of this
`new antiprogestin is poorly understood, it has recently been
`reported that when Onapristone was given to mice bearing
`
`214
`
`Astrazeneca Ex. 2036 p. 1
`Mylan Pharms. Inc. v. Astrazeneca AB lPR2016-01316
`
`
`
`Onapristone as Initial Therapy in Breast Cancer
`
`215
`
`hormone—dependent MXT mammary tumours, in addition to
`causing tumour regression, the compound caused histological
`changes suggestive of differentation to a more benign and
`mature status [6]. Further in viva work has suggested that the
`antitumour action of Onapristone is a direct, progesterone-
`mediated antiproliferative effect at the cell level, probably via
`the induction of terminal differentiafion associated with cell
`death
`The aim of this study was to investigate the clinical and
`biological effects of this agent as first—line endocrine therapy
`in patients with breast cancer.
`
`PATIENTS AND METHODS
`
`In September 1994 a phase II study was commenced to
`assess Onapristone as first—line therapy in postmenopausal
`patients with primary breast cancer. Patients were eligible to
`enter the study if they had a locally advanced tumour or were
`elderly with tumours suitable for endocrine therapy. In all
`patients endocrine therapy was deemed the initial treatment
`of choice. A primary goal of the study was to assess by
`sequential
`tumour biopsies the effect of Onapristone on
`tumour biology. The recruitment
`target was 30 patients.
`However, owing to the recognifion in this and other clinical
`studies that some patients on Onapristone developed liver
`function test (LFT) abnormalities, the development of this
`drug was discontinued and recruitment to the study stopped
`in 1995. At this time 19 patients had entered the study: 12
`patients had locally advanced primary tumours and 7 were
`elderly patients in whom endocrine therapy was the initial
`treatment of choice.
`
`trial programme on Onapristone was
`When the clinical
`halted all 19 patients were informed of the new data on LFT
`abnormalities. Since these changes appeared to be transient
`in all breast cancer patients treated, the 19 patients in this
`study were offered the option of continuing Onapristone with
`increased frequency of monitoring LFT measurements or
`changing to tamoxifen therapy. All patients elected to con-
`tinue with Onapristone.
`This paper reports the clinical response rate and duration
`of remission in these 19 patients. The minimum follow-up
`since randomisation was now 24 months and the maximum
`32 months. The data on liver function tests in these 19
`patients are also reported. These data are particularly important
`since, unlike the phase II/III studies in patients with metastatic
`disease, changes in LFT can be attributed to Onapristone
`rather than to metastatic involvement of liver or bone.
`
`Oestrogen receptors
`ER were roufinely measured in all pretreatment tumour
`biopsies by oestrogen receptor immunocytochemical assay
`(ERICA), as reported previously
`Two tumours showed
`no expression of ER. Of the remaining 17 tumours, four
`showed ER expression on between 10 and 50% of tumour
`cells and 13 showed ER expression on between 70 and 100%
`of tumour cells. PgR was also measured by immunocy-
`tochemistry (Abbott Laboratories, Maidenhead, Berkshire,
`U.K.). 12 pafients had PgR-positive tumours and 6 patients
`had PgR—negative tumours. In the remaining patient PgR
`status was unavailable.
`
`Therapeutic assessment
`Patients were assessed for therapeutic remission using the
`International Union Against Cancer (UICC) criteria [8].
`
`Since all patients commenced with only a palpable breast
`tumour the assessment involved measurement of the tumour.
`
`The largest diameters of the tumour in two directions were
`then multiplied together. Similar measurements were made
`of any regional lymph nodes and the sums of primary tumour
`and lymph node(s) added. All measurements were carried out
`by two breast surgeons (IFRR and PCW).
`Complete response (CR) was regarded as complete dis-
`appearance of the tumour. Partial
`response (PR) was a
`reduction of > 50% in the sum of the product of the two lar-
`gest diameters of palpable tumour (i lymph nodes). Objec-
`tive response (OR) combines the CR and PR categories. A
`tumour was classified as stafic disease (SD) if any change in
`size ranged between < 50% reduction and < 25% increase in
`the pretreatment measurements. In all cases (CR, PR or SD)
`there had to be no new lesions either in the breast or at dis-
`tant sites. Progressive disease (PD) de novo was defined as an
`increase in tumour size > 25% of the pretreatment value or
`the appearance of new lesions, or both. PD after a period of
`response or SD was defined as an increase in tumour size
`> 25% of the smallest recorded size or the appearance of new
`lesions, or both.
`A f11rther criterion to be fulfilled before patients were clas-
`sified as CR, PR or SD was the British Breast Group recom-
`mendation that tumours had to be in CR, PR or SD after at
`least 6 months of treatment [9]. This requirement was intro-
`duced to prevent reporting short remissions of doubtful clin-
`ical benefit. Subsequently, studies have supported this 6-
`month figure by reporting that patients who show SD for 6
`months have a statistically similar survival to patients who
`show PR or CR. All three groups show a significantly longer
`survival than the PD group [10—12].
`
`RESULTS
`
`showed PD.
`the 3-month assessment no patient
`At
`Patients’ tumours were either in PR (11 = 3) or in SD (n= 16).
`By 6 months the results were: PD (11: 6). SD (n= 2) and PR
`(71 = 10). One patient discontinued Onapristone after 4.5
`months when the tumour was static (see below). The median
`durafion of response for the 12 patients with PR or SD was
`70 weeks, compared with 20 weeks for the patients where the
`tumours showed de novo progression. 2 patients remain on
`Onapristone at a median time of 26 months from entry into
`the study.
`Some patients during therapy developed LFT abnormal-
`ities. The number of patients with abnormalities in each test,
`either before Onapristone treatment or at any time during
`therapy, is detailed in Table 1. Changes in each of the four
`liver
`function measurements are shown over
`time, and
`presented in two ways. Figure
`1(a)—(d)
`shows values
`pretreatment and regular 6-weekly values, as planned in the
`original protocol. Figure 2(a)—(d) show values pretreatment
`and at each time point at which blood was obtained
`during the patient’s first year of treatment. Figure 2 reflects
`the increased frequency of blood sampling in the patients
`recruited later in the study, when liver dysfunction became
`known.
`function abnormalities appeared mainly during
`Liver
`the first 6 weeks of follow-up. In only 3 out of 19 patients did
`the abnormality in LFT start after 6 weeks with a slight
`elevation of bilirubin (n= 1)
`(weeks 18 and 24), a 4-fold
`rise in gamma—glutaryltransferase (GGT) at week 18 (n: 1)
`and
`a
`slight
`elevation
`of
`alkaline
`phosphatase
`and
`
`Astrazeneca Ex. 2036 p. 2
`
`
`
`216
`
`].F.R. Robertson at al.
`
`GGT (n= 1) (week 15 onwards). One patient discontinued
`Onapristone at 4.5 months owing to rising LFT abnor-
`malities. At the point of discontinuing the drug a further
`blood test was taken, which subsequently showed that the
`
`LFT were starting to fall. Nevertheless, Onapristone was not
`restarted.
`
`In the 12 patients who had OR or SD 8 were both ER and
`PgR positive, 3 were ER positive and PgR negative and one
`
`Table 1. Number ofpatients with liver function test abnormalities
`
`Normal range
`
`Pretreatment
`
`During treatment
`
`No. of patients with elevated tests
`
`5—17um0l/l
`540 U/1
`80—28O U/1
`10—50 U/1
`
`0/18
`0/19
`1 1/19
`4/19
`
`4/19
`1 2/ 1 9
`18/19
`14/ 1 9
`
`Bilirubin
`Alkaline transferase
`Alkaline phosphatase
`Gamma—glutamyltransferase
`
`(b) 200
`A
`:1
`E’
`<5
`H >-
`I
`I
`1 8
`24
`
`1 50
`1 oo
`
`50
`0 I
`O
`
`13
`
`13
`
`(n) 19
`
`I
`6
`
`18
`
`I
`12
`
`13
`Weeks
`
`I
`18
`
`13
`
`I
`24
`
`13
`i
`
`(a)
`:7 so
`g
`5 40
`5
`-5
`E
`
`20
`0 I
`0
`
`(n) 18
`
`(C)
`
`80
`
`Q 60
`3 40
`3;‘
`20
`
`I
`6
`
`18
`
`I
`1 2
`
`13
`Weeks
`
`0 I
`0
`
`I
`6
`
`I
`12
`
`T“
`1 8
`
`(d) 500 s
`
`g 400 —
`E 300 —
`E 200 --
`5< 100 -
`
`O
`
`0
`
`(I7) 19
`
`‘I
`24
`
`13
`
`I
`6
`
`18
`
`,
`1 2
`
`,
`1 8
`
`13
`
`I
`24
`
`13
`
`(n)19
`
`15
`
`13
`Weeks
`
`13
`
`13
`Weeks
`
`Figure 1. LFT measurements (6 weekly). ALT, alkaline transferase; GT, gamma-glutamyltransferase; Alk, alkaline phosphatase.
`
`E 50
`40
`30
`
`
`
`Bilirubin(UMo|/L)
`
`20
`10
`
`0IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
`0
`4
`8
`12
`16
`20
`25
`29
`41
`
`YGT(U/L)
`
`400
`
`300
`
`200
`
`100 0
`
`IIIIIIIIIIIIIIIIIIIIII
`4
`8
`1216 20
`25
`
`0
`
`I
`
`I
`
`I
`
`29
`
`41
`
`(n) 18
`
`4
`
`6
`
`13
`
`5
`Weeks
`
`3
`
`1
`
`7
`
`3
`
`(n) 19
`
`4
`
`6
`
`13
`
`5
`Weeks
`
`3
`
`2
`
`1
`
`3
`
`E 250
`200
`150
`
`ALT(U/L)
`
`1 00
`50
`
`600
`
`400
`
`200
`
`
`
`Alkphos(U/L)
`
`0IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
`0
`4
`8
`12
`16
`20
`25
`29
`41
`
`0IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
`O
`4
`B
`12
`16
`20
`25
`29
`41
`
`(n)19461353213
`Weeks
`
`(n)19461353213
`Weeks
`
`Figure 2. LFT measurements (weekly). For abbreviations, see Figure 1.
`
`Astrazeneca Ex. 2036 p. 3
`
`
`
`Onapristone as Initial Therapy in Breast Cancer
`
`217
`
`PgR unknown was weakly ER positive. In the 6 patients who
`showed de novo progression, 3 were ER and PgR positive, one
`was ER positive and PgR negative, one was ER negative and
`PgR positive and one was negative for both receptors.
`
`DISCUSSION
`
`Two-thirds of patients obtained a clinically relevant
`tumour remission: 10 out of 18 (56%) showed a PR and 2
`out of 18 (11%) SD. These figures are at least as good as
`published remission rates
`for
`the
`anti—oestrogen agent
`tamoxifen [12—14], synthetic progestogen, megestrol acetate
`[13—15] and aromatase inhibitors such as aminoglutethamide
`[16—18] or Lentaron [19] used as first—line endocrine thera-
`pies. The high remission rate is in part related to the fact that
`17 out of 19 patients were ER and or PgR positive. Similarly,
`most of the studies referenced above entered only patients
`with ER-positive tumours or ER status unknown.
`It is known that the remission rate is lower to second-line
`than to first—line endocrine therapy. Recently, 225 patients
`were reviewed who received both first—line and second-line
`endocrine therapies,
`the remission rates were 72% (10%
`CR+ 22% PR+ 40% SD) and 53% (3% CR+ 8% PR+42%
`SD), respectively. In this latter review over 90% of patients
`received tamoxifen as first—line and megestrol acetate as sec-
`ond-line therapies [20]. Overall, Onapristone appears an
`effective first—line endocrine agent, with tumour remission
`rates similar to more established therapies. However, with
`only 19 pafients in the study there were insuflficient numbers
`to draw conclusions about equivalence with other agents.
`One or more LFT was elevated in the majority of patients
`on Onapristone. The abnormalities became apparent in the
`first 6 weeks of treatment and usually declined thereafter at a
`steady rate. In 1 patient Onapristone was discontinued and
`even in this patient the LFT had started to show a downward
`turn in the sample taken on the day on which therapy was
`stopped.
`The comparison between PgR status and remission needs
`to be interpreted with caution for the following reasons. In
`this group of patients,
`the majority were ER and/or PgR
`positive. If the same group were treated with an anti-oestro-
`gen such as tamoxifen, not all would respond, despite the
`presence of ER. In fact, one would expect
`tamoxifen to
`induce OR and SD in around 60-70% of ER-positive
`tumours. In this study 9 out of 14 (64%) patients with PgR-
`positive tumours showed OR or SD to Onapristone as first-
`line endocrine therapy.
`A recent publication on the progesterone antagonist Mife-
`pristone reported an objective response rate (CR or PR) of
`10.7% with a stable disease (duration range 2-17 months)
`rate of 39.3% [21]. The authors stated that these results were
`in patients with untreated metastatic breast cancer. However,
`the majority of patients had received adjuvant hormone ther-
`apy (43%) or chemotherapy (32%). This would have influ-
`enced the response rates they reported. As noted above, the
`long-term results have been reported of 250 patients where
`first- and second-line endocrine therapies were their initial
`systemic therapies for measurable disease [20], i.e. none had
`received adjuvant endocrine therapy. Patients received endo-
`crine therapy either
`for metastatic disease or
`for
`local
`tumours (e.g. elderly, unfit patients or locally advanced dis-
`ease). The results in the study by Perrault and colleagues
`appear more refiective of response rates that
`the present
`authors have reported to second-line therapy,
`i.e. 11%
`
`OR+42% SD [20]. These results are also similar to a recent
`randomised study of megestrol acetate 40 mg four times daily
`versus anastrozole 1mg daily given as second-line endocrine
`therapy. The OR rates were 12% for both agents and the SD
`rates were 28% and 30% for megestrol acetate and anas1:ro-
`zole, respectively.
`A further point to be considered when comparing results of
`this study with Onapristone and the study reported by Per-
`rault and colleagues with Mifepristone is the effect of these
`drugs on the level of oestradiol. In a study by Klijn and col-
`leagues, Mifepristone was reported to cause up to a 5-fold
`increase in serum oestradiol [22]. Mifepristone in that study
`was given as second-line therapy and induced an objective
`response rate of 9% and SD rate of 54%. Onapristone in the
`present study resulted in no increase in serum oestradiol
`(data not shown). These latter data are the subject of a
`separate publication. However,
`if a 50-90% reduction in
`serum oestradiol by aromatase inhibitors induces clinical
`responses, a 5-fold increase by Mifepristone may have an
`adverse effect on tumour growth. This difference between
`Onapristone and Mifepristone may be clinically important.
`Future studies on the effect of pure progesterone antagonists
`should include an assessment of their effect on sex hormone
`levels.
`The results of this small study of Onapristone as first—line
`therapy support the preclinical data that Onapristone is a new
`class of endocrine agent that could have a significant impact
`on endocrine treatment of breast cancer. From a clinical
`
`viewpoint the results of this study support the development of
`second-generation progesterone receptor antagonists for use
`in the treatment of breast cancer. Studies are starting to
`assess whether the sequential tumour biopsies from this study
`during Onapristone treatment show evidence of the occur-
`rence of tumour differentiation. If confirmed, this would fur-
`ther support preclinical data indicating that progesterone
`antagonists exert a differentiation effect
`through a novel
`mechanism of action. This would further support a develop-
`ment programme of new progesterone antagonists.
`
`1. Mouridsen HT, Paridaens R. Advanced breast cancer—new
`approaches to treatment: workshop report. Eur j’ Cancer Clin
`Oncol 1988, 24, 99—105.
`2. Robertson JFR, Cannon P, Nicholson RI, et al. Oestrogen and
`progesterone receptors as prognostic variables in hormonally
`treated breast cancer. Intern}Biological Markers 1996, 11, 29—33.
`3. Neef G, Beier S, Elger W, et al. New steroids with antiprogesta—
`tional and antiglucocorticoid activities. Steroids 1984, 44, 349—372.
`4. Henderson D, Antiprogestational and antiglucocorticoid activ-
`ities of some novel 11—aryl substitute steroids. In Furr BJA,
`Wakeling AE, eds. Pharmacology and Clinical Use: of Inhibitors of
`Hormone Secretion and Action. London, Bailliere Tindal, 1987,
`1 84—21 1 .
`5. Michna H, Schneider MR, Nishino Y, et al. Antitumour activity
`of the antiprogestins ZK 98.299 and RU 38.486 in hormone
`dependent rat and mouse mammary tumours: mechanistic stu-
`dies. Breast Cancer Res Treat 1989, 14, 275—288.
`6. Schneider MR, Michna H, Nishino Y, et al. Antitumour activity
`of the progestone receptor antagonist ZK 98.299 and RU 486 in
`the hormone—dependent mammary tumour model of the rat. Eur
`j‘ Cancer Clin Oncol1989, 25, 691—701.
`7. Robertson JFR, Bates K, Pearson D, et al. Comparison of two
`oestrogen receptor assays in the prediction of the clinical course
`of patients with advanced breast cancer. Br] Cancer 1992, 65,
`727—730.
`8. Hayward ]L, Carbone PP, Heuson JC, ez al. Assessment of
`response to therapy in advanced breast cancer. Cancer 1977, 39,
`1 289—1293.
`
`Astrazeneca Ex. 2036 p. 4
`
`
`
`218
`
`9.
`
`10.
`
`11.
`
`12.
`
`13.
`
`14.
`
`15.
`
`].F.R. Robertson at al.
`
`British Breast Group. Assessment of response to treatment in
`advanced breast cancer. Lancet 1974, ii. 38—39.
`Robertson JFR, Williams MR, Todd ], et al. Factors predicting
`the response ofpatients with advanced breast cancer to endocrine
`(megace) therapy. Eur] Cancer Clin Oncol 1989, 25, 4694175.
`Howell A, Mackintosh J, Jones M, et al. The definition of the ‘no
`change’ category in patients treated with endocrine therapy and
`chemotherapy for advanced carcinoma of the breast. Eur] Can-
`cer Clin Oncol1988, 24, 1567—l572.
`Robertson JFR, Willsher P, Cheung KL. et al. The clinical rele-
`vance of static disease (no change) category for 6 months on
`endocrine therapy in patients with breast cancer. Eur] Cancer
`1997, 33, 1774—1779.
`lV1organ LR. Megestrol acetate V Tamoxifen in advanced breast
`cancer in postmenopausal patients. Semin Oncol 1985, XII, 43417.
`Ingle JN, Creagan ET, Ahmann DL, er al. Randomised clinical
`trial of megestrol acetate versus tamoxifen in paramenopausal or
`castrated women with advanced breast cancer. Am 5‘ Clin Oncol
`1982, 5, 155—160.
`Muss HB, Wells HB, Paschold EH, et al. Megestrol acetate ver-
`sus tamoxifen in advanced breast cancer: 5 year analysis—a
`phase III trial of the Piedmont Oncology Association. 3’ Clin
`Oncol1988, 6, 1098—1106.
`
`16.
`
`17.
`
`18.
`
`19.
`
`20.
`
`21.
`
`22.
`
`Lipton A, Harvery HA, Santen R], et al. Randomised trial of
`aminoglutethamide versus tamoxifen in metastatic breast cancer.
`Cancer Res 1982, 42, 3434s—3436s.
`Gale KE, Anderson JW, Tormey DC, er al. Hormonal treatment
`for metastic breast cancer. Cancer 1994, 73, 354—361.
`Smith IE, Harris AL, Morgan M, et al. Tamoxifen versus ami-
`noglutethamide in advanced breast carcinoma: a randomised
`cross—over trial. Br Med} 1981, 283, 1432—1434.
`Carrion RP, Candel VA, Calabresi F, et al. Comparison of the
`selective aromatose inhibitor formestane with tamoxifen as first-
`line hormonal therapy in postmenopausal women with advanced
`breast cancer. Ann Oncol 1994, 5, S19—S24.
`Cheung KL, Willsher PC, Pinder SE. er al. Predictors of
`response to second—line endocrine therapy for breast cancer. Br
`Cancer Res Treat 1997, 45, 219—224.
`Perrault D, Eisenhauer EA, Pritchard KI, er al. Phase II study of
`the progesterone antagonist mifepristone in patients with
`untreated metastatic breast carcinoma. A National Cancer Insti-
`tute of Cancer Clinical Trials Group study. 3’ Clin Oncol 1996,
`14, 2709—2712.
`Klijn GM, de Jong FH, Bakker GH, et al. Antiprogestins, a new
`form of endocrine therapy for human breast cancer. Cancer Res
`1989, 49, 2851—2856.
`
`Astrazeneca Ex. 2036 p. 5