© 1997 Cancer Research Campaign
`
`The novel anti-oestrogen idoxifene inhibits the growth
`of human MCF-7 breast cancer xenografts and reduces
`the frequency of acquired anti-oestrogen resistance
`
`SRD Johnston’2, S Riddler’, BP Haynes’, R A’Hern?, IE Smith?, M Jarman‘ and M Dowsett'
`
`Departments of ‘Academic Biochemistry and 2Medicine, Royal Marsden Hospital, Fulham Road, London SW36JJ; *Biological Services Unit
`and ‘Cancer Research Campaign Centre for Cancer Therapeutics,Institute of Cancer Research, Sutton, Surrey SM2 5NG, UK
`
`Summary Theeffect of idoxifene, a novel anti-oestrogen with less agonist activity than tamoxifen, was compared with that of tamoxifen on
`the growth of hormone-dependent MCF-7 breast cancer xenografts. Forty tumours were established with oestradiol support in ovariectomized
`athymic mice, allowed to grow to a median volume of 420 mm? and then continued with oestradiol, no support, tamoxifen or idoxifene
`delivered by 1.5-cm silastic capsule. Tumour regression occurred with both anti-oestrogens, although maximum regression was observed
`following oestradiol withdrawal alone. While prolonged anti-oestrogen therapy was associated with static growth, tumour volumes were
`significantly lower with idoxifene (P=0.01). After 6 months, 0/10 idoxifene-treated tumours developed acquired resistance compared with 3/10
`tumours treated with tamoxifen. In separate experiments, 94 animals weretreated initially with oestradiol, tamoxifen, idoxifene or placebo
`following implantation with 1-mm%piecesof either wild-type (WT) or tamoxifen-resistant (TR) MCF-7 tumour. After 4 months, only 1/11 WT
`tumours becameestablished with idoxifene compared with 4/11 with tamoxifen, 8/12 with oestradiol and 0/12 with placebo. Likewise, fewer
`TR tumours were supported by idoxifene (3/12) than by tamoxifen (8/12) or oestrogen (11/12). These data indicate that, compared with
`tamoxifen, idoxifene shows reduced growth support of MCF-7 xenografts and may share only partial cross-resistance. Furthermore, the
`developmentof acquired anti-oestrogen resistance may be reduced during long-term idoxifene therapy. The drug’s reduced agonist activity
`may,in part, explain these observations andindicate a preferable biochemicalprofile for breast cancer treatment.
`
`Keywords: breast cancer; idoxifene; tamoxifen; acquired resistance
`
`The anti-oestrogen tamoxifen is established as first-line endocrine
`therapy for women with breast cancer. In advancedbreast cancer,it
`is most effective in oestrogen receptor (ER)-positive tumours
`(McGuire, 1978). However, most tumours that respond eventually
`develop acquired resistance andstart to regrow. MCF-7cells are an
`ER-positive hormone-dependent humanbreast cancercell line, and
`an animal model using MCF-7 xenografts in athymic mice has been
`developed by several groups to investigate the phenomenon of
`acquired anti-oestrogen resistance (Osborne et al, 1985; Gottardis
`et al, 1988). It has been demonstrated that resistant tumours often
`become growth dependent on tamoxifen and can be stimulated by
`the drug in a dose-dependent manner(Gottardis and Jordan, 1988).
`This growth can be reversed by tamoxifen withdrawalor inhibited
`by the ‘pure’ non-steroidal anti-oestrogen ICI 164,384 (Gottardis et
`al, 1989). It has been suggested that the partial agonist activity of
`tamoxifen or its metabolites may be responsible for the acquisition
`of tamoxifen-stimulated growth.
`Idoxifene is a novel anti-oestrogen that is structurally related to
`tamoxifen (Figure 1) (McCague, 1986). Analogues of tamoxifen that
`include an iodine atom at position 4 have been found to have
`increased affinity for ER (McCagueet al, 1989). Such compounds
`cannot undergo glucuronidation via 4-hydroxylation, which probably
`
`Received 19 August 1996
`Revised 18 October 1996
`Accepted 21 October 1996
`
`Correspondence to: SRD Johnston, Department of Academic Biochemistry,
`Royal Marsden Hospital, Fulham Road, London SW3 6JJ, UK
`
`804
`
`aids the excretion of tamoxifen (McCagueet al, 1990a), and, unlike
`trans-4-hydroxytamoxifen, they cannot isomerize to the cis isomer,
`which has much weakeranti-oestrogenic properties for tamoxifen,
`while retaining partial agonist activity (Murphyetal, 1990). In addi-
`tion, substitution of the dimethylamino group on the side-chain of
`tamoxifen by the pyrrolidine ring prevents conversionbytheliver to
`desmethy] and didesmethyl metabolites, which are the predominant
`circulating metabolites of tamoxifen found in humans (Danielet al,
`1981; Jordan et al, 1983). Studies have confirmed that these struc-
`tural modifications result in a compoundthat is metabolically more
`stable than tamoxifen (McCague et al, 1990b; Hayneset al, 1991).
`Idoxifene was more effective than tamoxifen at inhibiting MCF-7
`cell growth and rat mammary tumour growth (Chanderetal, 1991).
`Furthermore, observations that idoxifene has reduced agonistactivity
`in the immature rat uterotrophic assay compared with tamoxifen
`(Chanderet al, 1991) suggest that the drug may have a preferable
`biochemical profile for clinical use, and could be an effective anti-
`oestrogen in circumstances in which tamoxifen’s agonist activity is
`predominant.
`We established a xenograft model to investigate the growth-
`suppressive activity of idoxifene in acquired tamoxifen-resistant
`human breast cancer (MCF-7 cells). In particular, we wished to
`compare in established hormone-dependent xenografts the growth
`inhibition achieved with idoxifene with that observed with tamox-
`
`ifen treatment or oestradiol withdrawal. In view of its reported
`lower agonist activity, we wished to determine whether long-term
`administration of idoxifene would reduce or delay the emergence of
`acquired resistance, and whether
`tamoxifen-resistant
`tumours
`would remain sensitive to idoxifene in cross-resistance experiments.
`
`AstraZeneca Exhibit 2013 p. |
`InnoPharma Licensing LLC v. AstraZeneca AB IPR2017-00900
`
`

`

`> NCH,CH,O
`
`" Oo c=—=c
`CY \CH,CH,
`
`Tamoxifen
`
`NCH,CH,O
`
`c=—c
`
`1ot ~
`
`Idoxifene
`
`Figure 1 Chemical structures of the anti-oestrogens tamoxifen and idoxifene ,
`
`MATERIALS AND METHODS
`
`Chemicals
`
`The twoanti-oestrogens used in this study were tamoxifen {{Z-
`trans-\-{4-[2-(dimethylamino)ethoxy ]pheny]}-1,2-diphenyl-1-
`butene}} from Sigma Chemical Co., Poole, UK, and idoxifene
`{E-trans-1-(4-iodophenyl)-1-[4-(2-pyrrolidinoethoxy) phenyl]-2-
`phenyl-1-butene} (McCague, 1986) synthesized at The Cancer
`Research Campaign Centre for Cancer Therapeutics, Institute of
`Cancer Research, Sutton, UK.
`
`Drug delivery system
`
`Slow-release silastic capsules of tamoxifen and idoxifene were
`madeaccording to previously published methods (Gottardis et al,
`1989). Briefly, these were formed by plugging one end of a 1.5-cm
`length of medical-grade silastic tubing (0.078 inch internal diam-
`eter by 0.125 inch outside diameter; Dow Corning, Midland, MI,
`USA) with silastic 382 medical-grade adhesive. After drying,
`these werefilled with either tamoxifen free base or the crystalline
`form of idoxifene. Based on the relative molecular weight of each
`drug (tamoxifen, 371.3;
`idoxifene, 497.4), an estimated molar
`equivalent amount of each drug wasput into each capsule (20 mg
`of tamoxifen and 26 mg of idoxifene per capsule), and the capsule
`was sealed by plugging the open end with adhesive. All capsules
`
`were sterilized by y-irradiation (200 Gy) before subcutaneous
`implantation under general
`anaesthetic on the
`left dorsal
`paraspinalarea.
`
`Serum drug levels
`
`In preliminary experimentsto establish the serum levels and phar-
`macological profile of each drug, a total of 24 mice were
`implanted along theleft flank (parallel to the spine) with 1.5-cm
`silastic capsules containing either tamoxifen or idoxifene. Four
`mice from each group weresacrificed and bledat 2, 4 and between
`6 and 8 weeks. Thetotal serum level of each drug was measuredat
`these time points by high-performance liquid chromatography
`(HPLC)
`according
`to
`previously
`published methodology
`(Johnston et al, 1993). The recovery from the mouse serum for
`each drug was 97%, and the lower detection limit for the assay was
`0.1 ng mt' for tamoxifen and 0.2 ng ml! for idoxifene.
`
`Animals and tumours
`
`MCF-7 xenografts were established from cells that had been
`growing in culture in RPMI-1640 medium (Life Technologies,
`Paisley, Strathclyde, UK) supplemented with 10% fetal calf serum
`(Life Technologies.), 2 mm L-glutamine, 5 U ml"! penicillin, 5 mg
`ml-! streptomycin and 12.5 ng ml-! amphotericin (Sigma Chemical
`Co.). Cells were recovered from six 80% confluent 175-cm?flasks
`by scraping and were resuspended immediately in 2.5 ml offresh
`medium. Approximately 107 cells (0.2 ml) were injected in
`suspension into the right flank of each of ten ovariectomized
`athymic nude mice (ICRF nunw mice; Harlan, Oxford, UK). At
`the same time, each mousereceived a 1.7-mg 60-day release 17B-
`oestrodiol (E,) pellet (Innovative Research of America, Toledo,
`OH, USA) implanted under the neck skin under general anaes-
`thesia. After 8-12 weeks, oestradiol-dependent wild-type (WT)
`tumours were established for passage in subsequent experiments.
`All procedures were approved by the Institute of Cancer Research
`ethics committee.
`
`Growthinhibition of MCF-7 xenografts
`
`Forty mice were implanted with 1-mm} pieces of WT tumourin
`the right flank and supported with E2 pellets implanted at the same
`time under general anaesthesia. Bidimensional tumour diameters
`were measured by caliper at weekly intervals, and tumour volume
`in cubic mm wasestimated using the formula:
`
`Tumour volume = (width)? x length
`2
`
`At week 9, animals were randomly sorted into four groups of
`ten, which were treated differently: (1) E2 support with 1.7-mg 60-
`day release pellet continued; (2) E, support withdrawn by removal
`of pellet; (3) tamoxifen; or (4) idoxifene delivered by silastic
`capsule as described above. In the last two groups, E, support was
`withdrawn at the same time as the anti-oestrogen capsule was
`implanted. Tumour measurements were made weekly and change
`in tumour size for each animal recorded as the percentage of the
`baseline size reached at week 9. During prolonged treatment, drug
`capsules were changed every 60 days under general anaesthesia.
`The experiment was continued for 6 months to determine whether
`any tumours would develop acquired resistance and start
`to
`regrow.
`
`© Cancer Research Campaign 1997
`
`British Journal of Cancer (1997) 75(6), 804-809
`
`AstraZeneca Exhibit 2013 p. 2
`
`

`

`Table 1 Serum concentrations of tamoxifen and idoxifene achieved 2, 4 and
`6 weeksafter implantation of 1.5-cm silastic capsulesfilled with equimolar
`amounts of tamoxifen and idoxifene
`
`Duration of
`n
`Tamoxifen
`n
`Idoxifene
`therapy (weeks)
`(ng mi“)
`(ng mi-")
`
`17.341.2
`4
`34.34 6.5
`4
`2
`23.8+4.2
`4
`36.7 + 6.5
`3
`4
`25.3 42.2
`3
`29.5+5.4
`4
`6-8
`
`
`Values are expressed in ng mi-' as means + s.e. Detection limits were
`tamoxifen (0.1 ng mi-') and idoxifene (0.2 ng mi’).
`
`250:
`
`200
`
`150
`
`100
`
`ou
`
`200
`
`780
`
`£00
`
`chen EZ (Pe 20)
`ee Tamoxten (a= 10)
`we Idoxiiene (4= 10)
`te Placebo (= 70)
`
`0
`
`5
`
`10
`
`20
`
`25
`
`30
`
`15
`Weeks
`
`Figure 3 Growth rate of three MCF-7 xenografts that regrew during
`prolonged tamoxifen therapy and developed acquired resistance. Tamoxifen
`capsules were changed every 8 weeksasindicated by the arrows
`
`Gapsile
`change
`
`similar design to the experiments with WT tumour, four groups of
`six mice (repeated in duplicate) were supported with E,, placebo,
`tamoxifen or idoxifene for 16 weeks, and the TR tumourtake rate
`and growth rate were recorded.
`
`ie. ES
`Suopod
`
`Statistics
`
`50
`
`
`
`Percentagebaselinetumourvolume
`Percentagebaselinelurmaurvolume
`
`
`oD Decco Bcc Bcc cogccfat
`
`Growth rates for individual tumours were calculated assuming an
`exponential model within individual time segments. The growth
`rate was calculated as the slope of the line of log (volume) plotted
`against time. Comparisons in growth rate were performed using
`the Kruskal-Wallis one-way analysis of variance for the three
`
`treatment groups (tamoxifen,
`idoxifene and placebo), with the
`Mann-Whitney test for two samples using a multiple comparison
`corrected P-value of 0.017 (i.e. 0.05/3).
`
`egocBc ge igo ga
`
`Weeks
`
`Figure 2 Effect of tamoxifen, idoxifene or oestradiol withdrawal on growth of
`MCF-7 xenografts in nude mice. All tumours wereinitially growth supported
`with oestradiol. At week 9, capsules were changedto either continued
`oestradiol, tamoxifen, idoxifene or placebo. Tumour volumes were measured
`weekly and expressed as the percentage baseline volume at 9 weeks
`(mean + s.e.)
`
`Growth support of WT and TR tumours
`
`Another 48 mice were implanted with 1-mm: pieces of WT tumour
`in sequential duplicate experiments (24 mice per experiment). In
`each experiment, four groups of six mice were supported from the
`time of tumour implantation with either E2 pellet, placebo pellet,
`tamoxifen capsule or idoxifene capsule for a total of 16 weeks. The
`pellets or capsules were replaced under general anaesthesia after 8
`weeks. The WT tumourtake rates (i.e. number of tumours that
`becomeestablished by week 16) in each group and growthrate of
`any established tumours were recorded.
`In parallel experiments, a further 48 mice were implanted with
`1-mm?pieces from two tamoxifen-resistant (TR) tumoursthat had
`developed acquired tamoxifen resistance during long-term tamox-
`ifen treatment. Theoriginal E,-established WT tumours from these
`two mice had regressed following tamoxifen therapy, but after 18
`and 21 weeks,started to regrow despite continued tamoxifen. In a
`
`RESULTS
`
`Serum tamoxifen and idoxifene concentrations
`
`In preliminary experiments with 24 mice, the serum concentra-
`tions of tamoxifen and idoxifene obtained using the slow-release
`silastic capsules were measured after 2, 4 and 6 weeks. These data
`are shownin Table 1. By 4 and 6 weeks when steady state would
`be expected, the mean concentration of tamoxifen and idoxifene
`was 35.3 + 4.3 ng ml"! and 28.4 + 2.2 ng mi’, respectively, which
`is equivalent to 95 nm for tamoxifen and 57 nM for idoxifene.
`
`Growth inhibition of MCF-7 xenografts
`
`initial oestradiol support varied
`Tumour size obtained after
`markedly with the smallest tumour measuring 78 mm? and the
`largest 2790 mm3. However, the median tumoursize and ranges in
`eachofthe four groups of ten mice were similar (413, 342, 434 and
`366 mm?) such that valid comparisons in growth inhibition could
`be made between groups.
`Tumourstreated with further E2 support continued to grow at a
`steady rate. Tumours in which E2 support was withdrawn
`regressed in size such that by week 26 the mean tumoursize was
`
`British Journal of Cancer (1997) 75(6), 804-809
`
`© Cancer Research Campaign 1997
`
`AstraZeneca Exhibit 2013 p. 3
`
`

`

`
`
`
`E, (11/12 take)
`Tamoxifen (8/12 take)
`
`Idoxifene (3/12 take)
`
`Placebo (0/12 take)
`
`
`
`600
`
`400
`
`200
`
`
`
`
`
`Tumourvolume(mm*)
`
`4
`
`6
`
`8
`
`10
`Weeks
`
`12
`
`14
`
`16
`
`Figure 4 Effect of oestrogen, tamoxifen and idoxifene on the growth support
`of tamoxifen-resistant (TR) MCF-7 xenografts. Tumour implants from an
`acquired tamoxifen-resistant tumour (1 mm? size) were implanted on day 0
`into 48 mice that were then randomlyallocated to receive oestradiol,
`tamoxifen, idoxifene or placebo by silastic capsule (capsules replaced again
`after 8 weeks). The weekly mean tumour volumesare displayed (+ s.e.) for
`the tumours that becameestablished (take-rate indicated in brackets)
`
`12% of baseline (Figure 2) and 3/10 tumours had regressed
`completely.
`In mice treated with tamoxifen,
`tumour growth
`continued initially and peaked at 112% at week 10. Thereafter,
`tamoxifen induced tumourregression, although this was signifi-
`cantly less
`than that observed following E2 withdrawal
`(P=0.0003). Idoxifene induced immediate tumour regression that
`wassignificantly greater than that in the tamoxifen-treated group
`(P=0.01) (Figure 2).
`During prolonged anti-oestrogen therapy, tumour volumes in
`mice treated with tamoxifen remained static and 3/10 tumours
`started to regrow after an interval of 15, 18 and 21 weeks of
`therapy respectively. The growth of these individual tumoursthat
`developed acquired tamoxifen resistance is shown in Figure 3. In
`contrast, no tumours treated with idoxifene started to regrow
`during this same time period.
`
`Growth support of wild-type (WT) and tamoxifen-
`resistant (TR) tumours
`
`Twoof the 96 animals (one tamoxifen WT and one idoxifene WT)
`died shortly after
`tumour passage and capsule implantation
`(necropsies were not performed). The remaining 94 animals were
`observed for 16 weeks. WT tumour growth was dependent on E2
`support, as 0/12 tumours grew with placebo compared with 8/12
`treated with E2, these latter tumours arising after a median of 3
`weeks. With tamoxifen, 4/11 tumours took after a median of 9
`weeks, whereas only 1/11 WT tumours treated with idoxifene
`becameestablished (and then only at week 15).
`The growth of TR tumours was also dependent on endocrine
`support to become established; 11/12 E2-supported tumours took
`
`and grew atrates similar to those observed with WT tumours,
`whereas 0/12 tumours took when treated with placebo. More TR
`tumours were supported by tamoxifen than by idoxifene (8/12 vs
`3/12), although this was notstatistically significant (P=0.0995,
`Fisher’s exacttest, two-tail). No difference was observed between
`the TR tumour growth rates in idoxifene and tamoxifen-treated
`animals, although the median time to tumour take was longer with
`idoxifene (9 weeks vs 5 weeks). Growth rates with both anti-
`oestrogens were slowerthan with oestradiol (Figure 4).
`
`DISCUSSION
`
`MCF-7 xenografts established in ovariectomized nude mice and
`treated with long-term tamoxifen provide a model for investi-
`gating acquired anti-oestrogen resistance
`in breast
`cancer.
`Previous studies have established that, while tamoxifen will cause
`partial regression of established xenografts, continued administra-
`tion is associated with static growth and stable tumour volumes
`(Osborne etal, 1985; Gottardis et al, 1988). Following prolonged
`therapy, resistance eventually develops in this model and tumours
`regrow despite continued tamoxifen (Gottardis and Jordan, 1988).
`Subsequentstudies have shownthese tumours to be growth-stimu-
`lated by tamoxifen in a dose-dependent manner and that growth
`can be reduced by withdrawal of tamoxifen (Gottardis et al, 1989).
`In cross-resistance experiments,
`the pure anti-oestrogen,
`ICI
`164,384, can inhibit tamoxifen-stimulated growth of these resis-
`tant tumours (Gottardis et al, 1989), which implies a mechanism
`for acquired resistance that is specific to tamoxifen. This mecha-
`nism would explainthe clinical observation that more than 50% of
`patients with advancedbreast cancer, who previously responded to
`tamoxifen before developing acquired resistance, will respond to
`further endocrine therapies with either pure anti-oestrogens
`(Howell et al, 1995) or aromatase inhibitors (Dowsettet al, 1995).
`Acquired tamoxifen resistance may be associated with the
`agonist properties of the drug and/or its metabolites (Howellet al,
`1990). Several of the known metabolites have more agonist than
`antagonist effects. These include compounds formed following
`metabolism of the dimethylamino side-chain (i.e. the monophenol
`metabolite E and the bisphenol)
`(Lyman and Jordan, 1985).
`Although the cis isomer of 4-hydroxytamoxifen is probably as
`oestrogenic as the trans isomer, it is a much weaker antagonist
`(McCague et al, 1990a). It has been suggested that a relative
`increase within the tumour of more oestrogenic metabolites of
`tamoxifen could stimulate growth (Osborne et al, 1991; Wiebe et
`al, 1992), Analogues of tamoxifen in which the formation of these
`metabolites is prevented or reduced may prove more effective
`anti-oestrogens, and theoretically could delay the onset of any
`acquired resistance that was caused by stimulation by agonist
`metabolites.
`
`Idoxifene is a structural analogue of tamoxifen that is metaboli-
`cally more stable than tamoxifen. In vitro studies, using isolated
`rat hepatocytes, demonstrated that
`idoxifene was metabolized
`approximately three times more slowly than tamoxifen (Hayneset
`al, 1991). In vivo idoxifene was shownto havea significantly
`longer terminal half-life than tamoxifen in the rat (Hayneset al,
`1991). In a recent phase I study in women with advancedbreast
`cancer (Coombeset al, 1995), the terminal half-life of idoxifene
`was 3 weeks compared with a known half-life for tamoxifen of 7
`days (DeVos et al, 1992).
`In addition,
`the antagonist/agonist
`profile for idoxifene appears favourable to that for tamoxifen.
`Idoxifene has a greater relative binding affinity (RBA) for the
`
`© Cancer Research Campaign 1997
`
`British Journal of Cancer (1997) 75(6), 804-809
`
`AstraZeneca Exhibit 2013 p. 4
`
`

`

`oestrogen receptor compared with tamoxifen (tamoxifen, 5; idox-
`ifene, 12.5; oestradiol, 100) and is 1.5-fold more effective than
`tamoxifen at
`inhibiting oestrogen-induced MCF-7 cell growth
`(Chander et al, 1991). Idoxifene caused a greater percentage of
`tumour regression in the hormone-dependent NMU-induced rat
`mammary tumour model
`than tamoxifen (92% vs 75%).
`In
`uterotrophic studies in immature rats and mice,
`idoxifene had
`reduced agonist activity compared with tamoxifen in doses up to
`10 mg kg"! (Chanderet al, 1991). Overall, these data suggestthat,
`compared with tamoxifen, idoxifene is metabolically morestable,
`has reduced agonist activity and inhibits hormone-dependent
`tumour growth moreeffectively.
`Our data demonstrate greater tumourinhibition by idoxifene of
`MCF-7 xenografts compared with tamoxifen in vivo. Equimolar
`amounts of each drug were delivered by slow-release silastic
`capsule andresultedin relatively similar serum concentrations that
`remained stable during the 8 weeks each capsule was implanted.
`The slightly lower mean levels (57 nm vs 95 nm) could have
`resulted from variation in the release characteristics through
`silastic as a consequence of the idoxifene’s different chemical
`structure. Unlike oestrogen withdrawal in this model in which
`tumoursize diminished to 12% of baseline after 6 months, both
`tamoxifen and idoxifene maintained static tumour volumes
`following an initial period of tumour regression (Figure 2). The
`biological basis
`for
`these stable volumes
`remains unclear.
`Classical anti-oestrogens are thought to be cytostatic in action
`inhibiting cell cycle progression through G, (Sutherland et al,
`1983), and initial reduction in tumour volume may represent
`increased cell
`loss and/or reduced cell proliferation owing to
`antagonism of the mitogenic signal. The advantage of this MCF-7
`animal model, namely anti-oestrogen therapy in oophorectomized
`mice after removal of the oestradiol pellet, is that it allows the
`inherent agonist activity of the two drugs to be evaluated in the
`absence of endogeneous or exogenousoestradiol. The significant
`difference in the baseline level at which tumour size was main-
`
`tained between idoxifene and tamoxifen during long-term anti-
`oestrogen therapy could reflect the different agonist profiles of the
`two agents.
`The acquisition of resistance following prolonged therapy was
`seen only in tamoxifen-treated tumours. None of the idoxifene-
`treated tumours regrew during the 6-month experiment, although
`it is possible that with more prolonged therapy idoxifene-resistant
`tumours would have developed. Similar experiments
`that
`comparedthe effects of the pure anti-oestrogen, ICI 182,780, with
`those of tamoxifen in the same MCF-7 xenograft model demon-
`strated that the pure anti-oestrogen suppressed tumour growth for
`twice as long as treatment with tamoxifen (Osborneet al, 1995).
`Eventually, most
`tumours became resistant
`to ICI 182,780,
`although these experiments were conducted for much longer (11
`months) compared with the idoxifene studies. However, both
`studies imply that more effective oestrogen antagonism using
`drugs with reduced agonist activity may provide not only greater
`inhibition of tumour growth than tamoxifen, but may delay the
`onset of acquired resistance. Clearly, such a property would be
`highly advantageous for a novel endocrine agent, if this were
`translated in the clinic into prolonged time to disease progression.
`Experiments in which tumour implants are growth supported
`from the outset with either oestrogen or anti-oestrogen allow a
`comparison of the tumorigenic potential of each drug to be made.
`Fewer wild-type MCF-7 tumours were growth supported by idox-
`
`ifene compared with tamoxifen, with a longer time to tumour
`formation and reduction in tumourtake-rate. Similar observations
`have been reported with wild-type MCF-7 xenografts treated with
`the pure anti-oestrogen, ICI 182,780, compared with tamoxifen
`(Osborne et al 1995). These data support those from experiments
`that study growth inhibition of established MCF-7 tumours and
`suggest that anti-oestrogens with less agonist activity are less
`likely to support tumour growth.
`Tamoxifen-resistant (TR) tumours were growth supported by
`tamoxifen, although this phenotype clearly remained hormone
`dependentasillustrated by the lack of tumours that developed in
`the absence of any exogenous hormone.
`Idoxifene-supported
`tumours were less frequent and developed later after a median of
`15 weeks compared with 9 weeks for tamoxifen. Although there
`was nostatistical difference in the growth rates of idoxifene- and
`tamoxifen-supported TR tumours,
`the latter were still growing
`actively at 16 weeks when the experiment was terminated (Figure
`4). The emergerence of tumours later that grow more slowly may
`representgreater sensitivity of TR tumoursto idoxifene. This could
`be interpreted as a lack of partial cross-resistance between the two
`anti-oestrogens, although the hormone dependence of these idox-
`ifene-supported tumours was not examined in further serial trans-
`plant experiments. However, when ICI 182,780-resistant tumours
`developed after 11 months and were transplanted into new mice,
`they were noted to have become completely endocrine independent
`and grew in the absenceof oestradiol (Osborneetal, 1995).
`These data imply that novel anti-oestrogens, such as idoxifene
`or ICI 182,780, may inhibit the growth of tamoxifen-resistant
`tumours more effectively, a feature which could be related to the
`drug’s reduced agonist activity. If the formation of oestrogenic
`metabolites were a significant mechanism for tamoxifen relapse,
`then structural analogues in which their formation is prevented or
`reduced could be more effective anti-oestrogens. However, recent
`data from two separate groups have shownthat using fixed-ring
`derivatives of tamoxifen in which isomerization is inhibited (thus
`preventing formation of the cis isomer of 4-hydroxytamoxifen),
`growth stimulation of resistant
`tumours occurred to the same
`extent as with tamoxifen itself (Wolf et al, 1993; Osborneetal,
`1994). Furthermore, analogues, such as deoxytamoxifen (in which
`cleavage of the dimethylamino side-chain is
`impaired,
`thus
`reducing formation of the oestrogenic metabolite E or bisphenol),
`nafoxidine and toremifeneall stimulated tumour growth (Osborne
`et al, 1994). Thus,
`the mechansism for tamoxifen-stimulated
`growth in this model remains unclear. Nonetheless, these data
`imply that the contribution of oestrogenic metabolites of tamox-
`ifen in stimulating the growth of established acquired tamoxifen-
`resistant tumours is probably low. However, it remains possible
`that structural analogues of tamoxifen with substantially less
`agonist activity are more effective at inhibiting the growth of
`hormone-sensitive tumours, which, compared with tamoxifen,
`mayresult in the delayed onset of acquired resistance.If this trans-
`lated in the clinical setting into prolonged disease control in the
`primary or adjuvant setting, this would represent a significant
`advantage for idoxifene over tamoxifen, currently the first-line
`endocrine therapy for breast cancer.
`
`ACKNOWLEDGEMENTS
`
`This work was funded by British Technology Group, Ltd. SRDJis
`a Cancer Research Campaign Clinical Research Training Fellow.
`
`British Journal of Cancer (1997) 75(6), 804-809
`
`© Cancer Research Campaign 1997
`
`AstraZeneca Exhibit 2013 p. 5
`
`

`

`REFERENCES
`
`Chander SK, McCague R, Luqmani Y, Newton C, Dowsett M, Jarman M and
`Coombes RC (1991) Pyrrolidino-4-iodotamoxifen and 4-iodotamoxifen, new
`analoquesofthe antioestrogen tamoxifen for the treatment of breast cancer.
`Cancer Res 51: 5851-5858
`Coombes RC, Haynes BP, Dowsett M, Qugley M,English J, Judson IR, Griggs LJ,
`Potter GA, McCague R and Jarman M (1995) Idoxifene: report of a phaseI
`studyin patients with metastatic breast cancer. Cancer Res 58: 1070-1074
`Daniel P, Gaskell SJ, Bishop H, Campbell C and Robertson RI (1981) Determination
`of tamoxifen and biologically active metabolites in human breast tumours and
`plasma. Eur J Cancer Clin Oncol 17: 1183-1189
`DeVos D, Slee PHTJ, Stevenson D and Briggs LJ (1992) Serum eliminationhalf-life
`of tamoxifen and its metabolities in patients with advanced breast cancer.
`Cancer Chemother Pharamacol 31: 76-78
`Dowsett M, Johnston SRD,Iveson TJ and Smith IE (1995) Responseto pure
`antioestrogen (ICI 182, 780) in tamoxifen-resistant breast cancer. Lancet 345:
`525
`Gottardis M and Jordan VC (1988) Developmentoftamoxifen-stimulated growth of
`MCF-7 tumorsin athymic mice after long-term antioestrogen administration.
`Cancer Res 48: 5183-5187
`Gottardis MM,Robinson SP and Jordan VC (1988) Estradiol-stimulated growth of
`MCEF-7 tumors implanted in athymic mice: a modelto study the tumoristatic
`action of tamoxifen. J Steroid Biochem 30: 311-314
`Gottardis MM,Jiang S-Y, Jeng M-H and Jordan VC (1989) Inhibition of tamoxifen-
`stimulated growth of an MCF-7 tumorvariant in athymic mice by novel
`steroidal antiestrogens. Cancer Res 49; 4090-4093
`Haynes BP, Parr IB, Griggs LJ and Jarman M (1991) Metabolism and
`pharmacokinetics of pyrrolidino-4-iodotamoxifen in the rat. Breast Cancer Res
`Treat 19: 174 (A72)
`Howell A, De Friend D, Robertson J, Blamey R and Walton P (1995) Response to
`the pure antioestrogen ICI 182,780in tamoxifen resistant breast cancer, Lancet
`345: 29-30
`Howell A, Dodwell DJ, Laidlaw I, Anderson H and Anderson E (1990) Tamoxifen
`as an agonist for metastatic breast cancer. In Endocrine Therapy of Breast
`Cancer, Goldhirsch A (ed.), pp. 49-58 Springer: New York
`Johnston SRD, Haynes BP, Sacks NPM, McKinna JA, Griggs LJ, Jarman M, Baum
`M,Smith IE and Dowsett M (1993)Effect of oestrogen receptor status and
`time onthe intra-tumoural accumulation of tamoxifen and N-
`desmethyltamoxifen following short-term therapy in human breast cancer.
`Breast Cancer Res Treat 28: 241-250
`Jordan VC, Bain RR, Brown RR, Gosden B and Santos MA (1983) Determination
`and pharmacology ofa new hydroxylated metabolite of tamoxifen observed in
`
`patient sera during therapy for advanced breast cancer. Cancer Res 43:
`1446-1450
`Lyman SD and Jordan VC (1985) Metabolism of tamoxifen and its uterotrophic
`activity. Biochem Pharmacol 34: 2787-2794
`McCague R (1986) Halogenated tamoxifens. British Patent Application No 8621908
`McCagueR, Leclercq G, Legros N, GoodmanJ, Blackburn GM, JarmanM, etal.
`(1989) Derivatives of tamoxifen; dependenceof antioestrogenicity on the 4-
`substituent. J Med Chem 32: 2527-2533
`McCagueR,Parr IB, Leclercq G, Leung O-T and Jarman M (1990a) Metabolism of
`tamoxifen by isolated rat hepatocytes; identification of the glucuronide of 4-
`hydroxytamoxifen. Biochem Pharmacol 39: 1459-1465
`McCagueR,Parr IB and Haynes BP (19906) Metabolism of4-iododerivative of
`tamoxifen by isolated rat hepatocytes. Biochem Pharmacol 40: 2277-2283
`McGuire WL (1978) Hormonereceptors; their role in predicting prognosis and
`response to endocrine therapy. Semin Oncol 5: 428-433
`Murphy CS, Langan FS, McCague R and Jordan VC (1990) Structure—-function
`relationships of hydroxylated metabolites of tamoxifen that control the
`proliferation of estrogen-responsive T47D breastcancercells in vitro. Mol
`Pharmacol 38: 737-743
`Osborne CK, Hobbs K and Clark GM (1985) Effect of estrogens and antiestrogens
`on growth of humanbreast cancercells in athymic nude mice. Cancer Res 45:
`584-590
`Osborne CK, Coronado E, Allred DC, Wiebe V and DeGregorio M (1991) Acquired
`tamoxifen resistance: correlation with reduced breast tumorlevels of tamoxifen
`and isomerisation oftrans-4-hydroxytamoxifen. J Natl Cancer Inst 83:
`1477-1482
`Osborne CK, Jarman M, McCague R, CoronadoEB, Hilsenbeck SG and Wakeling
`AE(1994) The importance of tamoxifen metabolism in tamoxifen-stimulated
`breast tumor growth. Cancer Chemother Pharmacol 34; 89-95
`Osborne CK, Coronado-Heinsohn EB, Hilsenbeck SG, McCue BL, Wakeling AE,
`McClelland RA, Manning DL and Nicholson RI (1995) Comparison ofthe
`effects of a pure steroidal antiestrogen with those of tamoxifen in a model of
`humanbreast cancer. J Natl Cancer Inst 87: 746-750
`Sutherland RL, Green MD,Hall RD, Reddell RR and Taylor TW (1983) Tamoxifen
`induces accumulation of MCF-