`
`INTRODUCTION
`
`Partly agonistic antiestrogens, such as tamoxifen, have
`been widely used in the treatment of estrogen recep-
`tor-positive metastatic breast cancer for more than 20
`years. In addition, partly successful attempts at treat-
`ing non-mammary tumours which express the estro-
`gen receptor have been reported with tamoxifen or
`derivatives [1-7]. The use of “pure” antiestrogens,
`without partial agonistic activity, however, has been
`suggested for many years [8], because the estrogenic
`component of tamoxifen could be directly linked to
`the occurrence of secondary tumours, especially at the
`endometrial level [9]. It could also be one of the par-
`
`
`ameters involved in the escape of advanced tumours
`from tamoxifen treatment, as suggested by in vivo stu-
`dies in mice [10], and may even be directly involved
`in the stimulation of breast cancer metastasis [11]. In
`other respects, it has been postulated that mammary
`tumour growth could be stimulated by autocrine or
`paracrine growth factors which could replace the es-
`trogenic stimulation [12, 13]. Under growth factor-
`stimulating conditions, mixed antiestrogens are poor
`inhibitors of cell growth im vitro [14, 15], even if an
`indirect effect of tamoxifen on tumoral growth has
`been described in vivo, via a decrease of circulating
`growth factors [16].
`This led the ICI group to synthesize the first pure
`antiestrogens, exemplified by ICI 164,384 [17] and
`ICI 182,780 [14, 18],
`the latter being in phase IT
`clinical trials [19, 20]. We recently described a new
`pure antiestrogen, 11/f-[4-[5-((4,4,5,5,5,-pentafluoro-
`pentyl)sulfonyl] pentyloxy] phenyl!]-estra-1,3,5(10)-tri-~
`449
`
`© Pergamon
`
`¥. Steroid Biochem. Molec. Biol. Vol. 59, No. 5/6, pp. 449-457, 1996
`Copyright © 1996 Elsevier Science Ltd. All rights reserved
`Printed in Great Britain
`0960-0760/96 $15.00 + 0.00
`
`PII: $0960-0760(96)00140-9
`
`RU 58668: Further In Vitro And In Vivo
`Pharmacological Data Related to its
`Antitumoral Activity
`
`P. Van de Velde,'* F. Nique,' P. Planchon,’ G. Prévost,’+
`J. Bremaud,’ M. C. Hameau,' V. Magnien,’ D. Philibert’ and
`G. Teutsch’
`
`"Roussel Uclaf, 102 Route de Noisy, 93235, Romainville Cedex, France and *IOCMH, 129 Avenue de Stalingrad,
`93000, Bobigny, France
`
`Previous studies with the pure antiestrogen RU 58668 showed that this compound proved to be
`highly antiproliferative in vitro, and to be the only antiestrogenic compound so far known to induce
`long-term regression of MCF-7 tumours implanted into nude mice. In order to obtain more insight
`into the therapeutic potential of this molecule, we performed a new set of experiments in vitro and
`in vivo in comparison with tamoxifen and/or ICI 182,780. In vitro, 1nM RU 58668 induced an ac-
`cumulation of MCF-7 cells in G0/G1 phases of the cell cycle within 48 h and, in contrast to trans-4-
`hydroxy-tamoxifen, blocked the invasiveness of ras-transfected MCF-7 cells into the chick embryo
`heart during the three weeks of co-culture. An in vivo dose-effect relationship study showed that
`RU 58668 induced a regression of MCF-7 tumour with as low a dose as 10 mg/kg/week, and that such
`an effect can not be obtained either with a sublethal dose of adriamycin or with ICI 182,780,
`(2-250 mg/kg/week). This reduction in the tumour volumes accords with histological modifications
`of the tumours, which showed a decrease in the ratio of epithelial cells over the tumoral mass, and
`with a concomitant decrease in their regrowth potential when reimplanted into naive nude mice.
`Taken together, these results suggest a promising usefulness for RU 58668 in the treatment of meta-
`static breast cancer in women. Copyright © 1996 Elsevier Science Ltd.
`
`F. Steroid Biochem. Molec. Biol., Vol. 59, No. 5/6, pp. 449-457, 1996
`
`*Correspondence to P. Van de Velde. Tel: +33 1 49 91 59 20;
`Fax: +33 1 49 91 39 00.
`+Present address:
`Institut Henri Beaufour,
`91966 Les Ulis, France.
`Received 21 Mar. 1996; accepted 26 Jun. 1996.
`
`av. du Canada,
`
`5
`
`
`AstraZeneca Exhibit 2033 p. |
`InnoPharma Licensing LLC v. AstraZeneca AB IPR2017-00904
`
`

`

`450
`
`P. Van de Veldeetail.
`
`MATERIALS AND METHODS
`
`Chemicals
`
`Bromodeoxyuridine (BrdUrd), propidium iodide,
`trisodium citrate and 3,3’-diaminobenzidine were pur-
`chased
`from Sigma.
`Fluorescein
`isothiocyanate
`(FITC) was from Becton Dickinson and fetal calf
`serum (FCS) from Boehringer Mannheim, Germany.
`Unless otherwise stated, all
`the culture media were
`obtained from Gibco.
`
`
`
`[15, 21-23],
`ene-3,17f-diol: RU 58668 (scheme 1)
`quantify the DNA content, cells were centrifuged and
`which displayed improved im vitro antiproliferative ac-
`resuspended in a staining solution containing 7.6 uM
`tivities when compared to ICI 182,780 and similar
`propidium iodide and 4mMtrisodium citrate for
`10 min at 37°C. Doublie-stained cells were analysed in
`antiuterotrophic activities
`in mice or
`rats. When
`tested for
`their
`activities
`on MCF-7 tumours
`the FAC Scan flow cytometer (Becton Dickinson).
`Dot-plots obtained after cytofluorometric analysis
`implanted in nude mice, RU 58668 was the only
`allowed the determination of labelled cells (S-phase)
`compound able to induce a long-lasting reduction (at
`least 25 weeks) of the tumour volume [23]. When
`from unlabelled cells (GO/G1 and G2 — M phase).
`administered monthly at at dose of 250 mg/kg as sub-
`Invasiveness of tumorigenic cells. The human breast
`cutaneousoily injections, tamoxifen and ICI 182,780
`cancer cell-line MCF-7ras, a generous gift of C.
`Sommers (Georgetown University, Washington, DC,
`slowed down, or at the best stopped, tumoral growth
`without inducing a tumour regression [15]. In order
`U.S.A.) was established after the transfection of the
`to get more insight
`into the causes of this striking
`Hras oncogene in the MCF-—7 ductal adenocarcinoma
`difference between the im vivo antitumoral activities of
`cell-line [25]. A very highly tumorigenic MCF-—7 line
`the two “‘pure antiestrogens’’, we performed a set of
`(MCF-7vht) was obtained by injecting monolayers of
`MCF-7rascells (2 x 10°/0.1 ml) subcutaneously near
`new experiments. First, the dose/antitumoral activity
`the mammary gland of 5-week-old female Swiss nude
`relationships of RU 58668 and ICI 182,780 were stu-
`mice (Iffa Crédo, Les Oncins, France). After
`six
`died in order to compare the doses of the two com-
`weeks, one mouse waskilled, the tumour was cutout,
`pounds which would induce the maximal effect.
`cut into pieces of 1-2 mm? and cultured in a 25 cm?
`Second,
`experiments on the regrowth ability of
`flask with DMEM supplemented with 10% FCS.
`tumour fragments excised from animals previously
`treated with these two compounds and tamoxifen
`This process was performed twice.
`were carried out, along with histological examination
`ICI
`The dose effect
`relationship of RU 58668,
`
`
`
`
`
`of the tumoral and_trans-4-hydroxy-tamoxifentissues. In addition, im vitro experi- 182,780 (4-OH-
`ments were undertaken to study the effect of RU
`tamoxifen) on the growth ofthese cells, cultured as a
`58668 on the kinetics and intensity of cell cycle modi-
`monolayer in DMEM without phenolred, in the pre-
`fications. Invasiveness of the very highly tumorigenic
`sence of 5% charcoal-stripped serum, has been carried
`out in 24 multiwells. The cell growth was evaluated by
`ras-transfected MCF-7
`(MCF-7vht)
`in
`chick
`a fluorimetric DNAassayas previously described [15].
`embryonic heart was also studied to evaluate the in-
`fluence of this compound on the metastatic potential
`Furthermore, invasiveness properties of the MCF-
`of MCF-7 nodules into non-tumoraltissues.
`7vbt were evaluated, using the following protocol:
`cells were scraped from the plastic flask surface, the
`pellet was harvested in 1 ml of DMEM without phe-
`nol red, with 10% charcoal-stripped serum (FCS),
`1% glutamine and 1% sodium pyruvate and placed
`on top of a semi-solid medium composed of 1%
`Noble agar (Difco) in distilled water at a 1:1 ratio
`and 2 x DME medium supplemented with 20% FCS
`in 60mm Petri dishes,
`in order
`to obtain solid
`nodules. These nodules were subcultured once a
`week by cutting small pieces with microsurgical scis-
`sors. Fresh heart fragments were dissected from a 8-
`day embryonic chick and closely joined side by side to
`MCF-—7vht nodules in culture medium with or with-
`out 0.1nM RU 58668 or 4-OH-tamoxifen in the
`absence of estradiol. After
`a
`three-week culture,
`nodule-heart fragment complexes were fixed overnight
`in Bouin’s fixative, embedded in paraffin and pro-
`cessed for histological examination. The presence of
`epithelial cells in chick embryonic heart was revealed
`by the three-step indirect method using murine
`monoclonal antibodies to cytokeratin gp 56 kD (KLI,
`Immunotech, France)
`as
`first antibody (dilution
`1:400). The second antibody was a rabbit peroxidase-
`conjugated anti-mouse IgGs
`(Dakopatts Co.
`Inc.,
`Denmark)
`(dilution 1:20). The third antibody was
`a
`pig
` peroxidase-conjugated
`anti-rabbit
`IgGs
`(Dakopatts)
`(dilution 1:20). The 3,3’-diaminobenzi-
`
`(HTB 22 from
`Cell cycle analysis. MCF-7 cells
`ATCC) were seeded (5x 10° cells)
`in Dulbecco
`modified essential medium (DMEM) without phenol
`red supplemented with 10% dextran charcoal-treated
`FCS into 25 cm? flasks. Concentrations of RU 58668
`(0.01—1 nM) were addedat the cell seeding time. Cell
`cycle analysis was performed as described by Khoch-
`bin et al.
`[24]. To isolate cells in S-phase, BrdUrd
`was added to the culture medium for 15 min at 37°C,
`cells were then trypsinized, washed in phosphate-buf-
`fered saline (PBS) and fixed with 70% ethanol. Incor-
`porated BrdUrd was
`revealed with a monoclonal
`antibody anti-BrdUrd conjugated with FITC. To
`
`In vitro studies
`
`
`
`AstraZeneca Exhibit 2033 p. 2
`
`

`

`Further Pharmacological Studies of RU 58668
`
`451
`
`dine was used as peroxidase-substrate-chromogen and
`Harris haematoxylin as counterstain.
`
`In vivo studies
`
`tumour evolution was calculated as the ratio: tumour
`volume on week five/tumour volume on week 0.
`Incidence of a pretreatment with RU 58668 on tumour
`regrowth. After a five-week induction of the tumours,
`mice were randomized into four groups of five ani-
`mals which received a weekly administration of
`0.25 mg/kg E2 alone or along with twos.c. injections
`of 250 mg/kg RU 58668, ICI 182,780 or tamoxifen
`(one on week 0 and one on weekfive). On week 10,
`the animals were killed and each tumour was cut into
`1-2mm pieces which were reimplanted into five
`naive nude miceleading to groups of 25 animals bear-
`ing tumours previously treated by E2, E2+ RU
`58668, E2+ ICI 182,780 or E2+ tamoxifen. The
`growth of these tumours was stimulated by a weekly
`administration of 5 mg/kg E2 for six weeks. The
`tumour growth was checked every week in order to
`evaluate the incidence of the previous treatment on
`their evolution into non-treated mice.
`the
`correlate
`Histological
`studies.
`In order
`to
`regrowth profiles with histological parameters, a 14-
`week experiment, in which animals received 250 mg/
`kg test compounds on weeks 0, 5 and 10 along with
`0.25 mg/kg E2 weekly, was carried out. On week 14,
`mice were killed and the tumours were removed and
`fixed in Bouin’s solution, embedded in paraffin and
`processed for histological examination. Preparations
`were stained with Masson’s trichrome technique in
`which collagen appears in green, cell nuclei in crim-
`son red and cytoplasm in light brown.
`
`RESULTS
`
`In vitro studies
`
`
`
`MCF-7 cell culture and tumour settlement. MCF-7
`cells from ATCC (HTB 22) were routinely cultured
`and
`subpassaged
`in minimal
`essential medium
`(MEM) with phenol red in the presence of 5% FCS
`as previously described [15]
`in 75 cm? flasks. When
`subconfluent, the MCF-7 cells were trypsinized and
`resuspended in the above-mentioned medium at a
`density of 5x10’ cells/ml. One hundred microlitres
`of this suspension was injected subcutaneously into
`the right mammary fat pad of four to five week old
`female balb/ca nude mice (ffa Crédo). Tumour
`growth was stimulated by a weekly percutaneous ad-
`ministration of 5 mg/kg estradiol
`(E2) dissolved in
`10 yl ethanol. When tumours reached 250-500 mm?
`(calculated as width? x length/2),
`the animals were
`killed and the tumours cut into small pieces for reim-
`plantation. All the experiments were carried out with
`tumourfragments obtained from a single tumour.
`Dose activity relationship of RU 58668 on MCF-7
`tumours. Nude mice were
`subcutaneously
`(s.c.)
`implanted with 1-2 mm pieces of MCF-7 tumour,
`obtained as described above, and rtceived a weekly
`administration of 5 mg/kg E2 percutaneously (p.c.)
`for five weeks. Animals were then randomized accord-
`ing to their tumour volume (week 0) and treated
`weekly with 0.25 mg/kg E2 alone, or with a weekly
`subcutaneous injection of 2, 10, 50 or 250 mg/kg of
`test compounds, suspended in arachis oil. A control
`group received E2+ the sublethal dose of 10 mg/kg
`adriamycin by intraperitoneal route in saline on weeks
`0 and 3. The tumour volume was checked every week
`and, at the end of the experiment (week five), the ani-
`Effect on the MCF-7cell cycle. Table 1 presents the
`mals were killed by cervical dislocation and the
`compared fractions of MCF-—7 cells in G0/G1, in S
`
`tumours and uteri were removed and weighed. The and in G2+M_phases, either untreated or treated
`
`Table 1. Distribution of MCF-7 cells in the various phases of the cell cycle
`24h
`
`Control
`
`4-OH-tamoxifen
`
`ICI 182,780
`
`RU58668
`
`10°
`10°
`10°”
`10°°
`107)
`107?°
`10°
`107"?
`1971°
`10°
`
`G0/G1(%)
`
`45
`
`44
`45
`45
`46
`49
`50
`
`S(%)
`
`35
`
`34
`34
`35
`32
`31
`29
`
`G2/M(%)
`
`GO/G1(%)
`
`19
`
`22
`20
`19
`23
`21
`21
`
`52
`41
`48
`49
`55
`53
`54
`57
`51
`54
`59
`
`48h
`
`S(%)
`
`27
`35
`31
`28
`23
`24
`25
`18
`24
`22
`16
`
`G2/M(%)
`
`21
`23
`21
`23
`22
`22
`21
`25
`24
`24
`25
`
`Cells were untreated or treated with 0.0] to 1nM of the pure antiestogens for 24 or 48h or with 1nM to 1 yM of 4-OH-tamoxifen for
`48 h. At appropriate times, bromodeoxyuridine (BrdUrd; Sigma) was incorporated in cells in the S-phase. Incorporated BrdUrd was
`revealed with a monoclonal antibody anti-BrdUrd conjugated with fluorescein isothiocyanate (FITC; Becton Dickinson). To quantify
`DNAcontent, cells were resuspended in a staining solution containing 7.6 1M propidium iodide. Double-stained cells were analysed in
`the FAC Scan flow cytometer (Becton Dickinson). Dot-plots obtained after cytofluorometric analysis allow the determination of GO/
`G1, S-phase and G2 — M phase.
`
`
`
`AstraZeneca Exhibit 2033 p. 3
`
`

`

`
`
`
`
`AstraZeneca Exhibit 2033 p. 4
`
`452
`
`P. Van de Veldeezail.
`
`with 4-OH-tamoxifen, ICI 182,780 or RU 58668.
`Treatment of MCF-7 cells with 10°° M RU 58668
`for 48h resulted in a 40% decrease of the number of
`cells in S-phase with a corresponding increase in the
`number of cells in G0O/G1 phase. The experiment
`with RU 58668 was carried out
`three times. It
`is
`interesting to note that a reproducible and significant
`(P< 0.05) effect of RU 58668 on the cell cycle
`appeared after only 24h. Only the presented exper-
`iment compared the three compoundsat 24 and 48 h.
`In this experiment 4-OH-tamoxifen and ICI 182,780
`seem to be inactive at 24h. Low concentration
`(1 nM) of 4-OH-tamoxifen increased S-phase at 48 h.
`One micromole of that compound was required for
`reducing the cell number in S-phase. After a five-day
`treatment no difference could be observed between
`treated and untreated groups, all cells being accumu-
`lated in GO/G1 phases. This effect
`in the control
`group is related to the cell confluence.
`Effect on in vitro proliferation of MCF—7vht. The ras-
`transfected MCF-—7vht cell line, cultured in the pre-
`sence of 5% stripped serum, showed a two to three
`times higher ‘“‘spontaneous”’ proliferation rate than
`standard MCF-7 (data not shown). This allowed us
`to study the effect of test compounds on the growth
`of these cells, without any exogenous stimulating
`agent. Fig.
`1 shows the activity of RU 58668, ICI
`182,780 and 4-OH-tamoxifen after a seven-day cul-
`ture of MCF-7vht cells. RU 58668 was 2.5 times
`more potent than ICI 182,780 and 40 times more
`potent than 4-OH-tamoxifen to inhibit the growth of
`these
`cells,|
`with
`respective
`ICs59’s
`of
`
`EFFECT ON ras-TRANSFECTED MCF-7 CELL LINE
`
`_2o
`
`oo
`
`60
`
`40
`
`20
`
`DNA(%ofcontrol)
`
`
`
`—w*— RU 58668
`
`-——O— ICI 182,780
`—O— 4-OH-tamoxifen
`
`
`
`0
`10712
`
`19-12
`
`19-10
`
`19-9
`
`10 °8
`
`10°7
`
`19-6
`
`concentration (M)
`
`Fig. 1. In vitro activity of RU 58668, ICI 182,780 and 4-OQH-
`tamoxifen on the MCF-7vhtcells. The ras-transfected MCF-
`7vht cells were cultured in DMEM without phenol red in the
`presence of 5% charcoal-stripped serum and the indicated
`concentration of test compounds. Medium was changed
`every two to three days, and the DNA wasassayedafter eight
`days. Results are mean + SEM ofthree or four experiments.
`
`0.035 + 0.010 nM, 0.09 + 0.03 nM and 1.4+0.6nM
`(mean +SEM of three or four experiments). More-
`over,
`the two former compounds induced a nearly
`complete inhibition of cell growth (©90%), whereas
`4-OH-tamoxifen induced only a limited growth inhi-
`bition (50-60%) of the MCF~7vhtcell line.
`Effect on invasiveness of MCF-7vht in chick embryonic
`heart. Striking differences between 4-OH-tamoxifen
`and RU 58668 werealso observed on the in vitro inhi-
`bition of MCF-—7vht
`invasiveness in chick embryo
`heart. Fig. 2b shows chick embryonic heart after a
`three week contact with a nodule of MCF-—7vhtin the
`presence of RU 58668 at 10~° M. Noepithelial cells,
`stained with the specific epithelial cell antibody KL1
`cytokeratin, had invaded the heart fragment. At the
`same concentration, 4-OH-tamoxifen (Fig. 2c) was
`unable to prevent invasiveness as shown by the pre-
`sence of several stained cells in the embryonic tissue
`(see arrows). In the untreated-control group, embryo-
`nic tissue cannot be distinguish from MCF—7vhtcells
`(Fig. 2a).
`
`In vivo studies
`
`Dose activity relationship of MCF-7 tumours implanted
`in nude mice. A dose activity relationship study was
`undertaken in order
`to determine
`the maximal
`tumourinhibitory effect of RU 58668 in this model.
`As shown in Fig. 3, the maximal antitumoral activity
`of RU 58668 and ICI 182,780 was reached at the
`dose of 50 mg/kg weekly: RU 58668 induced a re-
`gression of the tumours, whereas ICI 182,780 was
`only able to slow down the tumour growth (respective
`ratio
`of
`tumour
`volumes:
`0.35+0.05
`and
`1,72 + 0.39, P< 0.01). A higher dose of the two com-
`pounds (250 mg/kg/week) did not inducesignificantly
`stronger effects. The regression of MCF—7 tumours
`induced by RU 58668 was observed at dosesstarting
`from 10 mg/kg/week. On the contrary, ICI 182,780
`was not able to induce any regression of the tumours
`on that model, whatever the doses used. Nevertheless,
`in these animals, the two compounds displayed simi-
`lar antiuterotrophic effects at equivalent doses (Fig. 4).
`In order to test
`the sensitivity of these MCF-7
`tumours to chemotherapeutic agents used in the
`clinic, we chose the established drug adriamycin. The
`dose used (two injections of 10 mg/kg) was the high-
`est one compatible with the toxicity of the compound.
`In this model, adriamycin only led to a decrease in
`tumoral growth, without regression, an effect signifi-
`cantly smaller than that of 10 mg/kg/week RU 58668
`(respective ratios of tumour volumeat the end of the
`experiment: 1.56+0.18 and 0.61+40.07, P<0.01).
`Therelatively poor effect of adriamycin in this model
`shows that these tumours have a strong growing abil-
`ity when implanted in nude mice.
`Effect of RU 58668 on tumour regrowth. We havepre-
`viously shown that RU 58668 induced the regression
`of E2-stimulated MCF-7 tumours implanted in nude
`
`

`

`
`
`
`
`AstraZeneca Exhibit 2033 p. 5
`
`453
`
`5
`
`Volumeweek5/volumeweek0
`
`BBR &2 + Rv ssece
`V//, £2 + ICI 182,780
`
` Oe
`
`Dose (mg/kg/week)
`
`Fig. 3. Dose effect relationship of RU 58668 and ICI 182,780
`on in vivo MCF-7 tumour evolution. After MCF-7 tumour
`settlement under estradiol, the animals received a weekly s.c.
`administration of the indicated doses of RU 58668 or ICI
`182,780 along with 250 ugikg estradiol for five weeks. The
`effect of these treatments was comparedto that ofi.p. injec-
`tions of 10 mg/kg adriamycin on weeks 0 and three of the ex-
`periment. *P < 0.05; **P < 0.01 vs E2 (Mann-Whitneytest).
`
`Further Pharmacological Studies of RU 58668
`
`mice [15, 23]. Similar results (Fig. 5) were obtained
`during the 10-week experiment described here; ani-
`mals treated with RU 58668 plus estradiol showed a
`ratio: volume week 10/volume week 0 of 0.48 + 0.11.
`Under the same conditions,
`the groups of animals
`treated with ICI 182,780 and tamoxifen led to re-
`spective ratios of 1.43+0.28 and 2.66 + 0.63. Con-
`trol mice treated with E2 alone showed a ratio of
`
`6.05 + 1.09. In order to check if the RU 58668-trea-
`ted tumours retain some growingability after this 10-
`week treatment, 1-2 mm pieces of tumour from each
`of these groups were reimplanted into naive nude
`mice and stimulated for six weeks with 5 mg/kg E2
`weekly, without any other treatment. The tumours
`from the different pretreated groups displayed differ-
`ent growth rates (Fig. 6), and significantly different
`weights at
`the end of the six weeks
`(Table 2).
`Tumours excised from animals
`treated with RU
`58668 displayed the lowest regrowing ability and the
`
`Fig. 2. Decreased invasiveness of MCF-7 transfected cells in
`embryonic chick heart. Embryonic chick hearts were main-
`tained in culture closely with MCF-7vht nodules in the
`absence of estradiol, and treated with vehicle (2a), RU 58668
`10° M (2b) or OH-tamoxifen 10°°M (2c). After three-week
`
`fragment complexes were fixed in
`culture, nodule-heart
`Bouin’s solution and included in paraffin. Epithelial cells
`were revealed by the three-step indirect method using mur-
`ine monoclonal antibodies to cytokeratin gp 56Kd (KL1,
`Immunotech, France) and appeared in brown (arrows) in
`photomicrographs of paraffin sections from embryonic heart
`(scale bar = 50 ym).
`
`

`

`P. Van de Veldeez ai.
`
`454
`
`200
`
`180
`
`160
`
`140
`
`120
`
`100
`
`80
`
`60
`
`40
`
`20
`
`
`
`Uteriweight(mg)
`
`
`
`
`
`Oe
`BBB £2 + Rv sa6es
`E2 + ICI 182,780
`
`
`
`lowest weights, suggesting that RU 58668 is able to
`induce qualitative modifications in the tumours.
`Histological studies. After a separate 14-week exper-
`iment, designed with the classical protocol comparing
`RU 58668,
`ICI 182,780 and tamoxifen,
`tumours
`were submitted to histological examination. Those
`from RU 58668-treated animals (Fig. 7D) displayed a
`high level of collagen (which appears in green) and a
`lower density of epithelial cells (which appear in red)
`than controls (Fig. 7A). Tumours from RU 58668-
`treated animals displayed a fibroadenomateousaspect.
`Tumours excised from ICI 182,780 (Fig. 7C) or 4-
`OH-tamoxifen-treated (Fig. 7B) animals displayed in-
`termediate histological profiles. In addition, necrotic
`structures, recognized by the loss of nuclear details
`with karyorrhexis or extreme pyknosis and the loss of
`nuclear staining characteristics, were seen consistently
`in control and tamoxifen-treated tumours, in contrast
`to ICI 182,780 and RU 58668-treated tumours (data
`not shown).
`
`DISCUSSION
`
`This paper describes in vitro and in vivo exper-
`iments carried out in order to complete the pharma-
`cological profile of RU 58668 and to gain more
`
`0
`
`2
`
`20
`
`50
`
`250
`
`Dose (mg/kg/week)
`
`Fig. 4. Dose effect relationship of RU 58668 and ICI 182,780
`on mice uteri weight. At the end of the five-week experiment
`described in Fig. 3, the animals were killed and the uteri
`were removed and weighed. On this parameter, RU 58668
`and ICI 182,780 displayed the same activity. *P< 0.05;
`**P< 0.01 vs E2 (Mann-Whitneytest).
`
`volumeweekn/volumeweek0
`
`As
`
`
`
`
`E2
`
`E2 + tamoxifen
`
`E2 + ICI 182,780
`
`E2 + RU 58668
`
`
`
`weeks
`
`Fig. 5. MCF-7 tumourevolution over 10 weeks. After five weeks of tumour growth stimulation before week 0,
`mice were randomly allocated to groups of five animals and received, for 10 weeks, a weekly p.c. adminis-
`tration of 250 pg/kg E2 with or without a single s.c. injection of 250 mg/kg test compounds on weeks 0 andfive.
`The animals were killed on week 10, and small pieces of each tumour were reimplanted into nude mice as
`described in Fig.6.
`
`
`AstraZeneca Exhibit 2033 p. 6
`
`

`

`
`
`
`AstraZeneca Exhibit 2033 p. 7
`
`200
`
`175
`
`150
`
`125
`
`100
`
`75
`
`50
`
`25
`
`
`
`
`
`tumorvolume(mm?)
`
`Further Pharmacological Studies of RU 58668
`
`455
`
`Previous treatments :
`
`
`E2
`
`
`E2 + tamoxifen
`
`E2 + ICI 182,780
`
`E2 + RU 58668
`
`
`
`Fig. 6. Growth of E2-stimulated tumour pieces coming from the experiment described in Fig. 5. Tumour
`pieces from the experiment described in Fig. 5 were implanted into nude mice and their growth was stimu-
`lated by a weekly p.c. administration of 5 mg/kg estradiol without any other treatment.
`
`weeks
`
`insight into its improved efficacy over ICI 182,780, in
`the model of MCF-7 tumours implanted in mice that
`had previously been shown.
`The effect of RU 58668 on the MCF-7 cell cycle
`(Table 1) giving rise to an accumulation of cells in
`the GO/G1 phases wassignificant after 24h at 10° M
`and more pronouncedafter 48 h. ICI 182,780 had no
`effect at 24h but at 48h the effect was close to that
`caused by RU 58668. The issue of this timing differ-
`ence between the two compounds observed in this im
`vitro test is still difficult to correlate with what occurs
`
`in animals. Other in vitro experiments with MCF-7
`cells, whose growth is endogenously stimulated by a
`stable transfection with the ras oncogene (MCF-
`
`Table 2. Weight of tumours and uteri of the experiment described
`in Fig. 6
`
`Original source of
`tumours
`
`Estradiol
`E2 + tamoxifen
`E2 + ICI 182,780
`E2 + RU 58668
`
`Tumour
`weight(mg)
`
`91.1415.7
`30.2 + 12.4
`46.0 + 9.8*
`25.444.3**
`
`(n = 25)
`(n = 24)
`(n = 20)
`(n= 24)
`
`Uterus
`weight(mg)
`
`123.9 + 8.3
`129.8 + 6.0
`128.24 5.2
`126.64 4.9
`
`After the previous treatment stated, small pieces of tumours were
`reimplanted into naive nude mice and their growth was stimu-
`lated for six weeks with 5 mg/kg estradiol; the animals were then
`killed and the tumours and uteri removed and weighed.
`*P< 0.05 vs E2.
`**P<0.01 vs E2 (Mann-Whitneytest).
`
`7vht), also do not explain the differences seen in vivo
`between the two compounds(Fig. 1).
`One could have argued that higher doses of ICI
`182,780 could have induced a decrease in the tumour
`volume, not found so far with a 250 mg/kg/month
`treatment. Our results (Fig. 3), however, show that
`such an assumption has to be excluded because,
`if
`both compounds displayed a similar quantitative ac-
`tivity (maximal effect at the weekly dose of 50 mg/kg),
`they showed a very different qualitative pattern
`(regression, compared to week 0, of the tumour
`volume with RU 58668, and a slowing down of the
`tumour growth with ICI 182,780). With respect to
`tumour volume, RU 58668 produces the same effect
`as ICI 182,780 but at doses 25-125 times lower.
`Small
`tumour pieces,
`removed from RU 58668-
`treated mice, displayed a lower growth ability, when
`teimplanted into naive animals,
`than those coming
`from E2,
`tamoxifen or ICI 182,780-treated mice.
`This suggests that pretreatment with RU 58668, and
`to a less extent ICI 182,780, caused a sustained
`growth inhibition of the reimplanted tumours.
`Histological examination of tumours treated with
`estradiol plus RU 58668 for 14 weeks showed a
`decrease in the ratio of epithelial cells to stroma com-
`pared to that treated with estradiol alone; this could
`be one explanation for
`the above-described slow-
`growing aspect of
`these tumours, but does not
`exclude the possibility of a retentive effect of RU
`58668 onthe dividing ability of MCF-—7 cells.
`
`

`

`456
`
`P. Van de Velde et ail.
`
`Fig. 7. Histological modifications of MCF-7 xenografted tumours in nude mice. Tumourpieces from an exper-
`imentsimilar to that described in Fig. 5, but lasting 14 weeks, were examined by histological processing. On
`week 14, mice were killed and tumours were removed and fixed in Bouin’s solution, included in paraffin and
`processed for histological examination. Preparations were stained with Masson’s trichrome technique in
`in crimson red and cytoplasm in reddish brown.
`cell nuclei
`which collagen appears
`in green,
`Photomicrographs were paraffin sections from MCF-7 tumourcells xenografted in nude mice treated with
`estradiol (a), OH-tamoxifen (b), ICI 182,780 (c), and RU 58668 (d) (scale bar = 50 ym).
`
`the density of
`In human infiltrating carcinomas,
`fibrous tissue of enveloping stroma varies greatly, and
`the degree of fibrosis in breast cancer has no clinical
`relevance. Nevertheless, metastases
`are
`essentially
`constituted of pure epithelial cells, so a decrease in
`the ratio of epithelial cells to stroma could be con-
`sidered as a reduction of the metastatic potential.
`These data, along with the inhibition of the invasive
`properties of the highly tumorigenic ras-transfected
`MCE-7cells into embryonic chicken heart (Fig. 2),
`suggest that the potential usefulness of RU 58668 in
`reducing the tumour evolution could be associated
`with
`the
`prevention
`of
`a metastatic
`process.
`Moreover, carcinomas evidencing necrosis are usually
`of a high grade. Therefore, a reduction of necrotic
`
`in RU 58668-treated tumours could be inter-
`foci
`preted as a good evolution prognosis.
`In conclusion, RU 58668 displayed a high and
`long-term antitumoral activity. It also decreased,
`in
`vitro,
`the invasiveness of ras transfected MCF-7vht
`cells and reduced,
`in vivo, the ability of regrowth of
`MCF-7 tumours. The results described herearestill
`unable
`to explain the discrepancy between the
`potency difference between RU 58668 and ICI
`182,780 in vivo and in vitro, where the former com-
`poundis surprisingly more effective than ICI 182,780
`in vivo. A higher bioavailability could be involved, but
`both compounds displayed the same in vivo effect on
`uteri (Fig. 4, Table 2 and Van de Velde ez al.
`[15}).
`Effects on angiogenesis, or differential effects on the
`
`
`
`
`
`AstraZeneca Exhibit 2033 p. 8
`
`

`

`Further Pharmacological Studies of RU 58668
`
`457
`
`feedback production ofestradiol by the ovaries follow-
`ing the binding of the two pure antiestrogens to the
`hypothalamo-pituary estrogen receptors, could also be
`involved. However, with regard to this last point, pre-
`vious results showed nearly the same antitumoral
`potency difference between RU 58668 and ICI
`182,780 between Ovx and non-Ovx nude mice [21].
`These data suggest that RU 58668 maybean effec-
`tive drug for the treatment of metastatic mammary
`carcinoma in women.
`
`REFERENCES
`
`. Stahl M., Wilke H., Schmoll H. J., Schober C., Diedrich H.,
`Casper J., Freund M. and Polidova H.: A phase II study of
`high dose tamoxifen in progressive, metastatic renal cell carci-
`noma. Ann. Oncol. 3 (1992) 167-168.
`. Tang B. L., Teo C. C., Sim K. J., Ng M. L. and Kon O. N::
`Cytostatic effect of antiestrogens in lymphoid cells: relationship
`to high affinity antiestrogen-binding sites and cholesterol.
`Biochim. Biophys. Acta 1014 (1989) 162-172.
`. Rumke P., Kleeberg U. R., MacKie R. M., Lejeune F. J.,
`Planting A. S., Brocker E. B., Bierhorst J. F. and Lentz M. A.:
`Tamoxifen as a single agent for advanced melanoma in post-
`menopausal women. A phase II study of the EORTC malig-
`nant melanoma cooperative group. Melanoma Res. 2 (1992)
`153-156.
`. Vertosick F,t. Jr, Selker R. J., Pollack I. F. and Arena V.: The
`treatment of intracranial malignant gliomas using orally admi-
`nistered tamoxifen therapy: preliminary results in a series of
`“failed” patients. Neurosurgery 30 (1992) 897-903.
`. Farinati F., De Maria N., Fornesiaro A., Salvagnini M.,
`‘Fagiuoli S., Chiaramonte M. and Naccarato R.: Prospective
`controlled trial with antiestrogen drug tamoxifen in patients
`with unresectable hepatocellular carcinoma. Dig. Dis. Sct. 37
`(1992) 659-662.
`. Panettiere F. J., Groppe C. W., Athens J. W., Costanzi J. J.,
`BonnetJ. D., Saiki J. H. and Chen T. T.: Tamoxifen therapy
`and colorectal adenocarcinoma: a southwest oncology group
`study. Cancer Treat, Rep. 69 (1985) 113-114.
`. Wong A., Chan A. and Arthur K.: Tamoxifen therapy in unre-
`sectable adenocarcinoma of the pancreas. Cancer Treat. Rep. 71
`(1987) 749-750.
`. Jordan V. C.: The only true antiestrogen is no estrogen. Molec.
`Cell, Endocr. 74 (1990) C91-C95.
`. Fornander T., Cedermark B., Mattsson A., Skoog L., Theve
`T., Askergren J., Rutqvist L. E., Glas U., Silfversward C.,
`Somell A., Wilking N. and Hjarlmar M.-L.: Adjuvant tamoxi-
`fen in early breast cancer: occurrence of new primary cancers.
`Lancet 8630 (1989) 117-119.
`Gottardis M. M. and Jordan V. C.: Development of tamoxifen-
`stimulated growth of MCF~7 tumors in athymic mice after
`long-term antiestrogen administration. Cancer Res. 48 (1988)
`5183-5187.
`
`10.
`
`11.
`
`12.
`
`13.
`
`14.
`
`15.
`
`16.
`
`17.
`
`18.
`
`19.
`
`20.
`
`21.
`
`22.
`
`23.
`
`24.
`
`25.
`
`Legault-Poisson S., Jolivet J., Poisson R., Beretta-Piccoli M.
`Band P. R.: Tamoxifen-induced tumorstimulation and with-
`drawal response. Cancer Treat. Rep. 63 (1979) 1839-1841.
`Cullen K. J., Smith H. S., Hill S., Rosen N. and Lippman M.
`E.: Growth factor messenger RNA expression by human breast
`fibroblasts from benign and malignant lesions. Cancer Res. 51
`(1991) 4