[CANCER RESEARCH 51, 3867—3873, August l, 1991]
`
`A Potent Specific Pure Antiestrogen with Clinical Potential
`
`Alan E. Wakeling,l Michael Dukes, and Jean Bowler
`Bioscience 1 [A. E. W., M. D.] and Chemistry I [J. B.], ICI Pharmaceuticals, Mereside, Alderley Park, Macclesfield, Cheshire SK[0 4T0, United Kingdom
`
`ABSTRACT
`
`Previous studies from this laboratory have described a series of 7a-
`alkylamide analogues of estradiol with pure antiestrogenic activity, ex-
`emplified by ICI 164,384. A new compound, 7a—l9-(4,4,5,5,5-pentafluo—
`ropentylsulfinyl)nonyllestra-l,3,5(10)-triene-3,17B-diol (ICI 182,780)
`has now been identified which has significantly increased antiestrogenic
`potency and retains pure estrogen antagonist activity. The antiutero-
`trophic potency of ICI 182,780 in the immature rat was more than 10-
`fold greater than that of [CI 164,384 (50% effective doses of 0.06 and
`0.9 mg/kg, respectively). This order of magnitude increase of in viva
`potency was also reflected, in part, by intrinsic activity at the estrogen
`receptor. The relative binding affinities of ICI 182,780 and ICI 164,384
`were 0.89 and 0.19, respectively, compared with that of estradiol (1.0).
`Similarly, the in vitro growth-inhibitory potency of ICI 182,780 exceeded
`that of ICI 164,384 in MCF-7 human breast cancer cells, where 50%
`inhibitory concentrations of 0.29 and 1.3 nM, respectively, were recorded.
`ICI 182,780 was a more effective inhibitor of MCF-7 growth than 4’-
`hydroxytamoxifen, producing an 80% reduction of cell number under
`conditions where 4’-hydroxytamoxifen achieved a maximum of 50%
`inhibition. This increased efficacy was reflected by a greater reduction of
`the proportion of cells engaged in DNA synthesis in ICI 182,780-treated
`cell cultures compared with tamoxifen-treated cells.
`Sustained antiestrogenic effects, following a single parenteral dose of
`ICI 182,780 in oil suspension, were apparent in both rats and pigtail
`monkeys. In vivo, antitumor activity of ICI 182,780 was demonstrated
`with xenografts of MCF-7 and Br10 human breast cancers in nude mice.
`A single injection of ICI 182,780 provided antitumor efficacy equivalent
`to that of daily tamoxifen treatment for at least 4 weeks.
`The properties of ICI 182,780 identify this pure antiestrogen as a
`prime candidate with which to evaluate the potential therapeutic benefits
`of complete estrogen withdrawal in endocrine-responsive human breast
`cancer.
`
`Considerations of the kind outlined above led us to search
`for novel molecules which would bind ER with high affinity
`without activating any of the normal transcriptional hormone
`responses and consequent manifestations of estrogen action.
`Such molecules would be clearly distinguished from tamoxifen-
`like ligands and would be pure antiestrogens. The rationale for
`the design and testing of novel putative pure antagonists has
`been described elsewhere (7, 8), as have the first examples of
`such compounds (9—11). The prototype pure antiestrogen, ICI
`164,384, a 7a-alkylamide analogue of estradiol,
`is devoid of
`stimulatory activity and blocks completely the trophic actions
`of estrogens and of the partial antiestrogens in all estrogen-
`responsive cell and animal models examined to date (see Ref.
`12 for a review).
`In this report we describe some of the properties of a new
`pure antiestrogen, ICI 182,780. ICI 182,780 is a potent and
`specific inhibitor of estrogen action and demonstrated excellent
`growth-inhibitory effects in both cell and animal models of
`human breast cancer. Such properties identify ICI 182,780 as
`a prime candidate with which to explore the therapeutic poten-
`tial of pure antiestrogens in the treatment of breast cancer.
`
`MATERIALS AND METHODS
`
`INTRODUCTION
`
`Nonsteroidal antiestrogens, exemplified by tamoxifen [ICI
`46,474 (Nolvadex)], have been used extensively, and with great
`success, in the therapy of breast cancer (1, 2). Antiestrogens
`compete with endogenous estrogens for binding to ER2 but a
`complete description of the mode of action of these molecules
`remains elusive (3, 4). In particular, it is difficult to account for
`the diversity of biological actions which range between full
`agonist, estrogen-like trophic effects, through partial agonism
`to complete blockade of estrogen action. This diversity was first
`apparent in species differences of organ response (5) but re-
`markably extends to differential effects of tamoxifen on estro-
`gen—responsive genes within target cells (6). Because of the
`potential of nonsteroidal antiestrogens to manifest stimulatory
`activity it remains unclear whether their clinical activity is in
`any way limited compared with that which might be achieved
`by complete endocrine ablation.
`
`Received 4/10/91; accepted 5/21/91.
`The costs of publication of this article were defrayed in part by the payment
`of page charges. This article must therefore be hereby marked advertisement in
`accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
`‘ To whom requests for reprints should be addressed.
`2 The abbreviations used are: ER, estrogen receptor(s); ICI 164,384, N—n-butyl-
`N—methyl—ll—(3,l7B-dihydroxyestra-l,3,S(l0)—triene—7a»yl)undecanamide:
`lCl
`182,780. 7a-[9-(4,4,5,5,S-pentafluoropentylsulfinyl)nonyl]estra-1,3,5(10)»triene—
`3,17fi-diol; OVX, ovariectomized; leo, 50% inhibitory concentration; ED”, 50%
`effective dose.
`
`trans-l-(4-fl-
`Reagents. The antiestrogens tamoxifen (ICI 46474;
`dimethylaminoethoxyphenyl)—l,2-diphenylbut—1—ene), the trans-4’-hy-
`droxy metabolite of tamoxifen, ICI 164,384, and ICI 182,780 were
`synthesized in Chemistry Department I, ICI Pharmaceuticals. Struc-
`tures of ICI 164,384 and [CI 182,780 are illustrated in Fig. 1. Stock
`solutions of these agents were prepared in ethanol, stored at 4°C, and
`diluted as required. 17B—[JH]estradiol, 85—110 Ci/mmol, and sodium
`‘zsl-iodide, IMS 30, were obtained from the Radiochemical Centre,
`Amersham, England. Materials for gonadotropin assays were obtained
`from the National Institute for Arthritis, Metabolism, and Digestive
`Diseases, Bethesda, MD except for ovine luteinizing hormone for
`iodination, supplied by L. E. Reichert, Emory University, Atlanta, GA.
`l7B-Estradiol benzoate, insulin, and materials for flow cytometry (pro—
`pidium iodide, bromodeoxyuridine, anti-mouse IgG fluorescein isothio-
`cyanate conjugate; F 0257) were obtained from Sigma Chemical Com-
`pany, Poole, Dorset, England, except for purified mouse anti-bromo-
`deoxyuridine monoclonal antibody (No. 7580) which was from Becton
`Dickinson, Mountain View, CA. All materials for cell culture were
`from Gibco, Paisley, Scotland, with the exception of Costar flasks
`which were from Northumbria Biologicals, Cramlington, England.
`MCF—7 cells were obtained from Dr. C. M. McGrath, Michigan Cancer
`Foundation, Detroit, MI, and BT20 cells were from Dr. J. Taylor,
`Imperial Cancer Research Fund, London, England.
`Estrogen Receptor Binding Assay. The method used for competitive
`binding assays to measure the relative affinity of antiestrogens for rat
`uterine ER has been described elsewhere (13) except that competitor
`dilutions were prepared in triszdimethylformamide (1:1) (14) and mixed
`together with 175-[3H1estradiol and then with cytosol at a ratio of 1:20.
`Cell Proliferation and Flow Cytometry Studies. The methods used for
`MCF-7 cell culture and growth inhibition assays have been detailed
`elsewhere (15). Briefly, cells were cultured in multiwell plates (24-well,
`seeding density 4 x 10‘) in minimal essential medium containing phenol
`red,
`insulin (10 pg/ml), and 5% charcoal-stripped fetal calf serum
`without estradiol. Antiestrogens and/or estradiol were added at 1000-
`fold dilution from ethanol stock, in fresh medium 2 days after seeding.
`Cultures were maintained for 5 days with one further medium change
`and growth was assessed by measurement of total cell protein at the
`3867
`
`Downloaded from cancerresaacrjournals.org on December 28, 2015. © 1991 American Association for Cancer
`Research,
`
`InnoPharma Exhibit 1031.0001
`
`

`

`OH
`
`
`
`""'(CH2>.ocorla(CH2)3CHa
`CH3
`
`ICI
`
`1 64,384
`
`OH
`
`
`
`m"(CH2)9SO(CH2)SCF2CF3
`
`ICI 182,780
`Fig. 1. Structure of pure antiestrogens.
`
`Ho
`
`HO
`
`PURE ANTIESTROGENS
`
`female nude mice (0.1 ml/approximately 5 X 10‘ cells). Mice were
`maintained in a clean environment and were given sterile food and
`water. Estrogen supplementation was provided by ethynyl estradiol at
`l ug/ml in the water. Antiestrogen treatment was initiated when tumor
`diameter attained a minimum of 0.5 cm. The Br10 tumor at passage
`49 was obtained from Dr. N. Brunner (Copenhagen, Denmark) and
`established by implantation of 1—2-mm3 tumor fragments into the flank
`of anesthetized intact adult female nude mice. After 3 passages a
`reproducible pattern of growth was established without additional es-
`trogen supplementation. Approximately two-thirds of animals estab-
`lished progressively growing tumors which attained measurable size
`(area, :70 mm‘) after 6—7 weeks. Antiestrogen treatment was initiated
`at the time of transplantation. Tamoxifen was administered once daily
`p.o. at a dose of 10 mg/kg (1 ml/100 g body weight of aqueous
`dispersion in 0.5% Tween 80) and ICI 182,780 as a single s.c. injection
`of 5 mg/mouse (50 mg/ml in arachis oil). Tumor size was assessed
`weekly as the product of caliper measurements of the largest diameter
`and the axis perpendicular to it.
`
`RESULTS
`
`Estrogen Receptor Interaction
`
`The capacity of ICI 182,780 to compete with 175-[3H]estra-
`die] for binding to rat uterine ER was evaluated and compared
`with that of estradiol and the previously reported pure anties-
`trogen ICI 164,384 (Fig. 2). Each antiestrogen displaced 175-
`[JH]estradiol
`in a concentration-dependent manner and the
`displacement curves were parallel
`to that of estradiol.
`ICso
`values of 0.83, 0.94, and 4.48 X 10’8 M were recorded for
`estradiol, ICI 182,780, and ICI 164,384, respectively. Relative
`binding affinities calculated from these ICso values were 0.89
`and 0.19 for ICI 182,780 and ICI 164,384, respectively, com-
`pared with that of estradiol (1).
`
`Estrogenic/Antiestrogenic Effects
`
`to immature
`When administered alone, parenterally (s.c.),
`female rats ICI 182,780 was devoid of uterotropic activity and,
`when coadministered with estradiol, it effectively blocked the
`uterotropic action of estradiol
`in a dose-dependent manner
`[EDso 0.06 mg/kg/day s.c. (Fig. 3A)]. Complete antagonism of
`estrogen action was achieved with a dose of 0.5 mg ICI 182,780/
`kg/day s.c. The effects of ICI 182,780 administered p.o. were
`qualitatively similar but potency was reduced by an order of
`
`100
`
`80
`
`60
`
`40
`
`20
`
`o Estradiol
`o |C|182.78O
`
`
`
`
`I
`
`A ICI 164,384
`
`
`
`o
`
`2
`
`-9
`
`>8
`
`-7
`[Competitor] log 10
`
`-6
`
`.§
`§
`
`2.
`
`beginning and end of treatment and compared with that of controls
`treated with ethanol (0.1%) alone. BT-20 cells were treated similarly.
`The measurement of antiestrogen effects on cell cycle and population
`distribution of MCF-7 cells using two parameter flow cytometry fol-
`lowed the method described previously (16).
`Assays of Estrogenic/Antiestrogenic Effects. The rat uterine weight
`assay for the measurement of estrogenic and antiestrogenic activity has
`been described elsewhere (17). Details of doses, route of administration,
`and duration of treatment are reported in individual figure legends. ICI
`182,780 and 17fi-estradiol benzoate were prepared for administration
`by diluting an ethanol stock solution into the required volume of arachis
`oil with gentle warming (60'C). Tamoxifen was prepared for adminis-
`tration p.o. as a dispersion in aqueous 0.5% Tween 80. For immature
`and mature rats the dose volumes were 0.5 and 0.1 ml/ 100 g body
`weight, respectively. In studies with intact rats, blood samples were
`collected terminally for measurement of luteinizing hormone, follicle-
`stimulating hormone, and prolactin. Plasma gonadotropin concentra-
`tions were determined by a modification of the double antibody tech-
`nique described by Niswender et al. (18).
`In studies with OVX rats, surgical preparation was performed at
`least 2 weeks before treatment began. To measure the duration of action
`of a single large dose of ICI 182,780, OVX rats were treated with a
`daily s.c. dose of 0.5 pg of estradiol benzoate beginning on the day of
`ICI 182,780 administration and continued until vaginal smears showed
`evidence of comification. At that point the experiment was terminated
`and uterine weight was recorded. The arachis oil formulation used in
`these single dose duration of action studies contained 50 mg ICI
`182,780/ml.
`For studies of the duration of action of ICI 182,780 in monkeys,
`adult female pigtail macaques (Macaca nemesm‘na) weighing 6—8 kg
`were ovariectomized not less than 6 months before treatment. Prelim-
`inary studies established that daily s.c. treatment of OVX monkeys
`with 5 pg estradiol benzoate/kg induced perineal swelling in a repro-
`ducible manner with individual maxima being attained after 11 days.
`The magnitude of the estrogenic effect was assessed visually on an
`arbitary scale of 1—6. Groups of five monkeys were treated s.c. once
`daily for 10 days with 0.1-1.0 mg/kg ICI 182,780, or with a single dose
`Fig. 2. Competition for binding of 5 X 10" M [’H]l7B—estradiol to rat uterine
`estrogen receptor by unlabeled WEI—estradiol.
`lCl 182,780, and 1C1 164,384.
`of 10 mg/kg, 10 days before beginning estradiol treatment. Perineal
`Percent inhibition refers to specific binding corrected by subtraction from total
`size was estimated daily and the time taken for initiation of swelling
`17fl-[3H]estradiol bound, the nonspecific component recorded in the presence of
`(mean score, 2) and attainment of maximum (score, 4-6) was recorded.
`5 X 10‘7 M unlabeled 17fl-estradiol. Paints, mean of 9 observations in three
`Tumor Growth Inhibition Assays. MCF-7 cells were suspended in
`different experiments; bars, SEM. leo values were calculated by linear regression
`analysis of percentage ofinhibition versus log.o[competitor].
`culture medium (no serum) and inoculated s.c. into the flank of adult
`3868
`
`5 §
`
`5a.
`
`Downloaded from cancerresaacrjournals.org on December 28, 2015. © 1991 American Association for Cancer
`Research.
`
`InnoPharma Exhibit 1031.0002
`
`

`

`PURE ANTIESTROGENS
`
`240
`
`100
`
`120 mmin-wooamwwn)
`
`2‘0
`
`Don-(NW)
`
`200
`
`(3)
`
`f
`g m
`g...
`i so
`3
`‘0 """I"‘"t----O-----------..
`0
`0.2
`0.5
`L0
`2.0
`5.0 1.0
`“(ml
`Fig. 3. Effects of 1C1 182,780 on uterine weight of immature rats. Animals
`received daily. a single dose of arachis oil vehicle alone (D). 0.5 ug 17fi~estradiol
`benzoatc s.c. alone (). or the indicated doses of 1C1 182.780 alone (-—-—) or
`together with estradiol (—). for 3 days. A. parenteral (s.c.) administration: 8.
`p.o. administration. Points. means for a minimum of 10 observations in at least
`2 different experiments. In this and succeeding figures bars on each point represent
`the SEM. Where no bar is present errors were smaller than the symbols.
`
`125
`
`m'...........1
`
`Q}
`3
`
`i?
`é,
`
`100
`
`75
`
` g,
`
`E 50—.E”Q
`3O
`.E
`5,
`D
`
`25
`
`0
`F——f—"fi‘_fi_——Tfi
`0.2
`0.5
`1.0
`2.0
`5.0
`10.0
`
`Dose (mg/kg)
`
`Fig. 4. Antagonism of the uterotropic effect of tamoxifen by 1C1 182.780.
`Immature rats were treated as described in the legend to Fig. 3, except that the
`represents the effect of 1 mg tamoxifen/kg alone and — is the effect of the
`indicated doses of 1C1 182.780 together with tamoxifen. Points, mean for at least
`5 observations; bars. SEM.
`
`weight of the uterus in a dose-dependent fashion (Fig. 5). At
`the highest dose in this study, 1 mg/kg/day, involution of the
`uterus after 14 days approached that following ovariectomy.
`Cyclical vaginal cornification was blocked partially (0.1 mg/
`kg/day) or completely (0.3 mg/kg/day) but body weight gain
`and serum gonadotropin concentrations were largely unaffected
`(Table 1). The p.o. antiuterotropic activity of ICI 182,780 in
`intact rats was substantially less than its parenteral potency; at
`10 mg/kg/day for 14 days an effect approximating 50% that of
`ovariectomy was recorded.
`that many steroids administered
`Following the precedent
`parenterally in oil have a sustained duration of action, the effect
`of 1C1 182,780 administered as a single s.c. bolus dose in oil
`suspension was tested in adult ovariectomized rats. The initia-
`tion of vaginal cornification and uterine growth by daily admin-
`istration of 0.5 pg of estradiol benzoate was blocked for more
`than 6 weeks by 10 mg ofICI 182,780. Uterine weights 42 days
`after 1C1 182,780 for ovariectomized controls, estrogen—treated
`controls, and 1C1 182,780 plus estrogen-treated rats were 60.6
`i 3.5 (SEM), 311 t 26, and 63.3 i 0.6 mg (n = 5), respectively.
`Similar treatment of intact adult females completely blocked
`
`300
`
`250
`
`E 200
`9(D
`i
`a)
`.E
`E:
`3
`
`150
`
`100
`
`E.
`
`50
`
`0
`
`r——'—T—T—l
`0.03
`0.1
`0.3
`1.0
`
`Dose mglkg
`
`Fig. 5. Effect of 1C1 182,780 on the uterus of intact adult rats. Groups of 5
`adult rats with regular 4-day estrous cycles were treated once daily. for 14 days.
`with arachis oil vehicle alone () or the indicated doses oflCl 182,780 s.c. Uterine
`weight was also recorded for vehicle~treated animals ovariectomized at the begin-
`ning of the study ([1).
`
`Table 1 Eflect of1C] 182. 780 on body weight and plasma gonadotropins ofadult
`female rats
`Values are mean : SEM; n = 5. All 1C1 182.780 values differ from correspond-
`ing OVX controls, at P < 0.001.
`Gonadotropins (ng/ml)
`Follicle-
`Body wt
`Luteinizing
`stimulating
`
`Treatment
`gain (g)
`hormone
`hormone
`Prolactin
`lntact control
`40.0 1- 2.5
`2.4 t 0.6
`3.0 t 0.4
`25.3 i 3.3
`
`OVX control
`
`64.8 t 1,9“
`
`19.7 1 2.2”
`
`24.0 t 0.5“
`
`3.7 i 0.7“
`
`[CI 182,780 (mg/kg)
`18.2 t 8.9
`2.2 t 0.2
`2.1 t 0.2
`43.6 t 2.5
`0.03
`19.1 i 6.4
`2.5 i 0.6
`1.2 t 0.1
`44.6 t 1.7
`0.1
`28.8 1 17.2
`2.5 t 0.5
`1.0 t 0.1
`45.8 1 2.0
`0.3
`
`1.0 6.0 i 2.2” 42.6 t 2.1 2.3 t 0.3 3.6 i 0.6
`
`
`
`" P < 0.001 versus intact control.
`" P < 0.05 versus intact control.
`
`magnitude compared with s.c. dosing [cf E050 0.46 and com-
`plete antagonism at 5 mg/kg/day p.o. (Fig. 38)]. Similarly, the
`uterotropic action of tamoxifen was also blocked in a dose-
`dependent manner by coadministration of [CI 182,780 (Fig. 4).
`Complete blockade of tamoxifen action required an approxi-
`mately 5-fold dose ratio. Similar studies of uterotropic and
`antiuterotropic activity in immature mice and in ovariectomized
`adult rats and mice provided confirmation of the pure antago-
`nist profile of 1C1 182,780 (data not shown).
`Chronic daily parenteral (s.c.) treatment of intact adult fe-
`male rats with increasing doses of 1C1 182,780 reduced the
`3869
`
`Downloaded from cancerres.aacrjournals.org on December 28, 2015. © 1991 American Association for Cancer
`Research.
`
`InnoPharma Exhibit 1031.0003
`
`

`

`PURE ANTIESTROGENS
`
`Table 2 Antiestrogenic action of[Cl 182, 780 in ovarieclamized estrogen-treated
`monkeys
`
`Days from start of
`estrogen treatment
`to reach perineal
`score of
`4—6
`
`Treatment
`>2
`(maximum)
`Control (estradiol benzoate alone)
`5
`11
`
`1C1 182,780 10 days: pretreatment with
`0.1 mg/kg s.c.
`0.5 mg/kg s.c.
`1.0 mg/kg s.c.
`
`10.0 mg/kg s.c.
`
`13
`18
`41
`23
`
`17
`23
`47
`33
`
`100
`
`80
`
`60
`
`40
`
`20
`
`0 8T 20
`
`O MCF -7
`
`a
`
`E 8
`
`o\"
`.Ew
`a.
`E,
`
`i 0
`
`o Fr—Tfi—r—i—rfi
`o
`-10
`-9
`-a
`~7
`-6
`-5
`
`no: 182,780] log 10
`
`Fig. 6. Effects of lCl 182,780 on the proliferation of MCF—7 and BT-20 human
`breast cancer cells. Cells were plated in 24»well dishes (4 x 10‘/well) and cultured
`for 2 days in minimal essential medium with 5% charcoal—stripped fetal calf
`serum containing phenol red and insulin but no additional estrogens. One dish
`was assayed for total protein (Lowry) as day 0 control; remaining dishes received
`fresh medium with (treated) or without (control) the indicated concentrations of
`1C] 182,780 added in ethanol (1 ug/ml medium). Cells were grown for a further
`5 days with fresh medium added after 3 days. Cell growth is represented as the
`difference between the increase of total protein in control and treated wells
`between day 0 and day 5. Points, mean of quadruplicate observations where SEM
`was less than 5%.
`
`cyclical vaginal cornification for at least 3 weeks and regressed
`the uterus (24% cf.
`intact control weight at 21 days after
`treatment).
`The antiestrogenic activity of lCl 182,780 was also measured
`in OVX pigtail macaques. Maximum swelling is attained after
`11 days of estrogen treatment (5 ug/kg/day). Pretreatment of
`monkeys with 0.1, 0.5, or 1 mg 1C1 182,780/kg/day s.c. for 10
`days prior to estrogen replacement produced an increasing delay
`in the onset of perineal swelling, of the order of 1, 2, and 5
`weeks, respectively (Table 2). Administration of a single dose
`of 10 mg [C] 182,780/kg s.c.
`in oil suspension delayed the
`onset of perineal swelling by 3 weeks and the attainment of
`maximum swelling by in excess of 4 weeks (Table 2).
`
`Breast Cancer Growth Inhibition
`
`182,780 on MCF—7 cells was reversed in a competitive manner
`by estradiol (Fig. 7). In the presence of 10’8 M ICI 182,780,
`coincubation with 10"0 M estradiol had no effect but growth
`inhibition was reversed partially at 10’“ M and completely by
`10'” M estradiol. For MCF-7 cells grown in medium containing
`phenol red addition of estradiol alone provided a moderate,
`concentration-dependent growth stimulus (Fig. 7).
`A comparison of the effect of ICI 182,780 with that of other
`antiestrogens on the growth of MCF-7 cells (Fig. 8) showed
`that it was significantly more potent than ICI 164,384 (leo =
`0.29 and 1.3 nM, respectively) or 4’-hydroxytamoxifen. Also,
`like ICI 164,384, the maximum growth-inhibitory effect of [CI
`182,780 exceeded that of 4’-hydroxytamoxifen [approximately
`80% cf. 50% (Fig. 8)].
`Flow cytometric analysis of cell cycle and population distri—
`bution of MCF-7 cells treated with tamoxifen or [Cl 182,780
`showed that both antiestrogens caused accumulation of cells in
`Go/G. and also reduced the proportion of cells capable of
`continued DNA synthesis (Table 3). However,
`the maximal
`efficacy of ICI 182,780 compared with that of tamoxifen, when
`both compounds were used at optimum antiestrogenic (but not
`cytotoxic concentrations), was much greater. Thus, only 7% of
`cells were still potentially capable of division after 3—5 days of
`treatment with 10 nM ICl 182,780 compared with 37% in
`cultures treated with 4 uM tamoxifen.
`Human Breast Tumors in Vivo. The effects of [CI 182,780
`were compared with those of tamoxifen in two models of human
`breast cancer grown in nude mice. The growth of xenografts of
`MCF-7 human breast cancer cells, supported by continuous
`treatment with ethynyl estradiol, was blocked completely for at
`least 4 weeks by a single s.c. injection of 5 mg of [CI 182,780
`in oil suspension (Fig. 9). The magnitude of this effect was
`comparable with that in animals treated continuously with a
`high dose of tamoxifen (10 mg/kg/day p.o.).
`The growth of transplants of the Br10 human breast tumor
`was also suppressed effectively by ICI 182,780. Mice implanted
`with a 1—2-mm3 tumor mass were given a single 5-mg s.c.
`injection of [Cl 182,780 on the day of implantation or daily
`treatment for 8 weeks with tamoxifen (10 mg/kg/day p.o.).
`Tumor measurements (Fig. 10) showed a substantial and sus-
`tained reduction of tumor growth in [CI 182,780-treated mice
`
`0 Estradiol
`.
`0 Estradlol
`
`+IC|182780
`
`
`
`
`1042
`
`10'“
`
`10"0 10'9
`
`10"
`
`[Estradiol] log 10
`
`140
`
`120
`
`A 100
`7.2
`
`so
`
`so
`
`4o
`
`§
`°\°
`v
`
`g
`5
`
`ai 0
`
`20
`
`o
`
`0
`
`Human Breast Cancer Cells in Vitro. ICI 182,780 was an
`effective inhibitor of the growth of ER-positive MCF—7 human
`breast cancer cells but was without effect on the growth of ER—
`negative BT-20 human breast cancer cells (Fig. 6). ICI 182,780
`was fully effective at 10‘9 M on MCF-7 cells grown in medium
`containing phenol red but without added estradiol. Cell death
`Fig. 7. Effects of 17B—estradiol on the growth of MCF-7 cells in the absence
`was not observed in either MCF-7 or BT—20 cells exposed to
`and presence of 10“ M [C] 182.780. The experimental procedure was as described
`for Fig. 6.
`10’5 M ICI 182,780. The growth-inhibitory action of [CI
`3870
`
`Downloaded from cancerres.aacrjournals.org on December 28, 2015. © 1991 American Association for Cancer
`Research,
`
`InnoPharma Exhibit 1031.0004
`
`

`

`PURE ANTIESTROGENS
`
`o 161182.780
`A IC|164.384
`
`
`
`.
`O 4-hydroxytamoxtfen
`
`
`
`in which the amide moiety was
`tion of the 7a side chain,
`replaced by other polar groups and the terminal alkyl function
`was fluorinated (10), produced the pentafluoropentylsulfinyl
`compound ICI 182,780.
`In receptor binding studies (Fig. 2) and as a specific and
`reversible inhibitor of MCF—7 breast cancer cell growth (Figs.
`6-8), 1C1 182,780 demonstrated an approximately 5-fold in-
`crease of intrinsic potency compared with ICI 164,384. This
`increased potency was clearly manifest in viva where, in the rat
`antiuterotropic assay, ICI 182,780 [EDso = 0.06 mg/kg (Fig.
`3)] was at least an order of magnitude more potent than ICI
`
`‘00
`
`80
`
`60
`
`40
`
`20
`
`0
`
`:5
`
`0
`
`E.
`
`EG)
`§
`o.
`iO
`
`300*
`
`200‘
`
`100
`
`0
`
`752
`a
`3.3
`
`E éI
`
`VE)
`
`a5
`
`E
`3
`.—
`
`~
`
`————————————— —-
`“r
`
`
`
`1
`
`l
`1
`2 3 4
`Control
`
`l
`,
`1 2 3 4
`Tamoxifen
`
`l
`
`2 3 4
`
`lCl 182,780
`
`Weekly tumor measurement
`Fig. 9. Effect of [Cl 182,780 and tamoxifen on the growth of established MCF-
`7-derived tumors in nude mice. Columns, means (n z 5) of tumor area normalized
`by reference to initial area preceding the 4«week treatment period; bars. SEM.
`All animals received continuous ethynyl estradiol (1 ug/ml in drinking water).
`Additionally, tamoxifen-treated animals were dosed daily (10 mg/kg p.o.) and
`1C1 182,780 once (5 mg/mouse so.) at the beginning of the 4—week measurement
`period.
`
`°Control
`ATamoxifen
`0161182180
`
`14°
`
`‘2°
`A100
`N
`
`
`
`E E
`
`3
`
`[Antiestrogen] log 10
`Fig. 8. Effects of different antiestrogens on the growth of MCF—7 cells. The
`experimental procedure was as described for Fig. 6. Points, mean derived from
`three or more different experiments with quadruplicate observations in each.
`SEM was less than 4%.
`
`Table 3 Eflecrs ofantiesrrogens on population distribution ofMCF—7 human
`breast cancer cells
`% of cells
`
`Treatment
`Control
`
`Tamoxifen
`0.4x10'°M
`1 x 10“5 M
`2 x 10" M
`4 x 10" M
`1C1 182,780
`0.4 x 10” M
`1 x 10" M
`2 x 10" M
`4 x 10’9 M
`10 x 10" M
`
`Cycling
`S + G; + M
`26
`
`Go/G.
`56
`
`Noncycling
`Go/G,
`S + G; + M
`10
`8
`
`4|
`35
`30
`27
`
`47
`13
`10
`4
`6
`
`7
`ll
`8
`10
`
`22
`3
`2
`l
`1
`
`41
`44
`52
`50
`
`17
`77
`79
`84
`82
`
`12
`ll
`10
`13
`
`14
`7
`9
`10
`10
`
`similar to that of high—dose tamoxifen treatment. Note that 2
`weeks after the end of tamoxifen treatment tumor growth rate
`showed evidence of a return to control level whereas, even 3
`months after a single dose of 1C1 182,780, tumor growth rate
`remained below that of control. Ovariectomy of all animals
`after 3 months demonstrated the estrogen sensitivity of these
`tumors (Fig. 10).
`
`DISCUSSION
`
`so
`
`so
`
`40
`
`20
`
`Hj/l
`I
`l
`
`o
`so
`so
`70
`so
`90T100110120130
`
`1’
`
`0EB
`
`The disc0very of novel steroidal antiestrogens exemplified by
`1C1 164,384 (9, 10) provided for the first time pure estrogen
`antagonists which have shed new light on the physiology (1 1,
`19-22) and the molecular mode of action of estrogens and
`antiestrogens (23-26). Among the initial series of 7a-alky1am-
`ide analogues of 17B-estradiol described previously (10) none
`was of sufficient potency in vivo to merit serious consideration
`Days post implantation
`as a candidate for clinical use. We therefore sought to identify
`Fig. 10. Effect of [Cl 182,780 and tamoxifen on the growth of Brio human
`more potent compounds which retained the potentially advan-
`breast tumors in nude mice. Groups of 10 female nude mice bearing transplants
`tageous properties of 1C] 164,384. For breast cancer treatment,
`of Br10 received either no treatment (control), daily tamoxifen (10 mg/kg p.o.)
`for 8 weeks beginning on the day of transplantation, or a single dose of [Cl
`these included high affinity for ER (27), the absence of estro-
`182,780 (5 mg/mouse s.c.). Values are mean tumor area, (n = 6—8) for all tumors
`genie activity (9, 11-14), and more effective antiproliferative
`attaining measurable size by day 50 postimplantation; bars, SEM. Approximately
`action on breast cancer cells than classical tamoxifemlike par-
`3 months postimplantation all mice were ovariectomized (arrow) and tumor
`measurements were continued for 1 month further.
`tial agonist antiestrogens (22—24). Further synthetic modifica-
`3871
`
`Downloaded from cancerresaacrjournals.org on December 28, 2015. © 1991 American Association for Cancer
`Research,
`
`InnoPharma Exhibit 1031.0005
`
`

`

`PURE ANTI ESTROGENS
`
`164,384 [EDSO = 0.9 mg/kg; see Ref. 15]. The apparent 2-fold
`difference in potency ratio improvement between in vitro and
`in viva assays for the two compounds is likely a reflection of
`differences in distribution and metabolism. Both in vitro and in
`
`vivo studies were consistent with a competitive interaction
`between ICI 182,780 and estradiol for binding to ER. The
`absence of a significant estrogenic activity of ICI 182,780 was
`clearly apparent in rodent uterotropic assays (e.g., Fig. 3) and
`in its capacity to block completely the stimulatory action of
`tamoxifen (Fig. 4).
`Of particular relevance to the therapeutic potential of ICI
`182,780 are two observations reported here: (a) the enhanced
`efficacy compared with 4’-hydroxytamoxifen (or tamoxifen) on
`breast tumor cells (Fig. 8; Table 3); and (b) the excellent
`antiuterotropic action (Figs. 3—5; Table 2) achieved without
`affecting body weight and gonadotropin secretion (Table 1).
`The castration—like uterine involution achieved in intact animals
`
`with a cytostatic rather than a cytotoxic action. Others, using
`the same model system, have reported tumor involution during
`tamoxifen treatment to the extent that tumors almost disap-
`peared after 3—4 weeks (35). This has been attributed to an
`alteration in cell death rate (35). In the current study (Fig. 9),
`3—4 weeks of continuous high dose tamoxifen treatment pro-
`duced a significant decrease (P < 0.05) of mean tumor volume,
`compared with that at the start of treatment, but not tumor
`disappearance. A recent study of the effects of estrogen with-
`drawal on MCF-7 tumors in nude mice has also demonstrated
`tumor regression associated with both cessation of cell prolif-
`eration and the activation of programmed cell death (36).
`Kyprianou et al. (36) attribute interlaboratory differences in
`apparent response to hormone withdrawal
`to variations in
`MCF-7 cell phenotype where some but not all sublines have
`lost the capacity to initiate apoptosis. Whether differences exist
`between pure and partial agonist antiestrogens in their capacity
`to promote apoptosis, as well as the apparent differences in
`sensitivity between different strains of MCF-7 cells, remains to
`be determined.
`
`in the absence of an effect on the latter indices of hypothalamic-
`pituitary function indicates that ICI 182,780 might be differ-
`entially active against peripheral and central targets of estrogen
`action, a property shared with ICI 164,384 (15). If translated
`to the clinical setting, this peripheral selectivity of action would
`obviate blockade of central negative estrogen feedback and
`consequent increases of estrogen production in the premeno-
`pausal patient. With respect to the enhanced efficacy of pure
`antiestrogens against tumor cell growth in vitro we have shown
`previously for ICI 164,384 (16, 28—30) and here for ICI 182,780
`(Table 3) that fewer of the cells remain in the actively prolifer-
`ating fraction than is the case when partial agonists like tamox-
`ifen, 4’-hydroxytamoxifen, or hydroxyclomiphene are used.
`This has been attributed to a residual stimulatory estrogenic
`effect of the partial agonists which, although small (16, 31), is
`amplified synergistically by the concurrent presence of other
`breast cell mitogens like insulin (16) and insulin-like growth
`factor 1 (32). The pure antiestrogens obviate such effects. The
`corollary of these data in the clinical setting is the possibility
`that differences of antitumor efficacy between tamoxifen and
`pure antiestrogens may be greater than otherwise anticipated.
`The order of magnitude lower potency between the p.o. and
`parenteral routes of administration (Fig. 3) suggests strongly
`that the p.o. bioavailability of ICI 182,780 is relatively low. A
`common means of circumventing the practical constraints con—
`sequent on the poor p.o. bioavailability of steroids is to use
`parenteral depot formulations with an extended duration of
`action. The utility of this approach was demonstrated with ICI
`182,780 dispersed in arachis oil. Thus, single s.c. injections of
`ICI 182,780 in ovariectomized, estrogen-treated rats and mon-
`keys (Table 2) provided extended antiestrogenic activity.
`The potential efficacy of “oil depot” formulations of ICI
`182,780 was demonstrated in nude mouse antitumor studies.
`The antitumor action of a single parenteral dose of ICI 182,780
`on MCF—7 xenografts was similar to that achieved by daily
`administration of a high dose of tamoxifen ove