Breast Cancer Research and Treatment 57: 183–192, 1999.
`© 1999 Kluwer Academic Publishers. Printed in the Netherlands.
`
`Report
`The effect of combining aromatase inhibitors with antiestrogens on tumor
`growth in a nude mouse model for breast cancer
`
`Qing Lu, Yang Liu, Brian J. Long, Dmitry Grigoryev, Mark Gimbel, and Angela Brodie
`Department of Pharmacology and Experimental Therapeutics, University of Maryland, School of Medicine,
`Baltimore, MD, USA
`
`Key words: breast cancer, nude mice, aromatase inhibitors, antiestrogens
`
`Summary
`
`We have previously established a model for postmenopausal, hormone-dependent breast cancer in nude mice which
`is responsive to both antiestrogens and aromatase inhibitors. In this model, MCF-7 human breast carcinoma cells
`transfected with the aromatase gene (MCF-7CA) synthesize sufficient estrogen to form tumors in ovariectomized
`nude mice. In the present study we used this intratumoral aromatase model to investigate the effects on tumor
`growth of the new nonsteroidal aromatase inhibitors letrozole (CGS 20,267) and anastrozole (ZD 1033) and
`the antiestrogens tamoxifen (ICI 47,474) and faslodex (ICI 182,780). Furthermore, we determined whether the
`inhibition of estrogen synthesis together with inhibition of estrogen action would be more effective in controlling
`breast tumor growth. The results of our studies indicate that the aromatase inhibitors anastrozole and letrozole, as
`well as the new pure antiestrogen faslodex, have potent antitumor effects in the mouse model. In the treatment
`of mice with mammary tumors, letrozole was more effective in suppressing tumor growth than anastrozole. This
`was consistent with the Ki values of these inhibitors against placental aromatase and the IC50 values in cell culture
`(MCF-7CA), which indicated the greater potency of letrozole as an aromatase inhibitor. Letrozole also had greater
`antitumor effects than tamoxifen and faslodex. The antitumor effect of letrozole was substantial, making it difficult
`to detect any additional effect on the tumors when letrozole was combined with the antiestrogens. However, the
`combined treatment of anastrozoleC tamoxifen and anastrozoleC faslodex also did not increase efficacy compared
`to the aromatase inhibitor alone. In addition, combining the two antiestrogens did not suppress tumor growth more
`effectively than faslodex alone. Our results show that treatment with the combinations of aromatase inhibitors
`with either tamoxifen or faslodex are not more effective in blocking estrogen stimulation of tumor growth than the
`aromatase inhibitors alone.
`
`Abbreviations: Faslodex (ICI 182,780): 7a-[9-(4,4,5,5,5-pentafluoropentylsulfinyl) nonyl]estra-1,3,5(10)-triene-
`3,17b-diol; Letrozole (CGS 20,267): 4-[1-(cyanophenyl)-1-(1,2,4-triazolyl)methyl]benzonitrile; Anastrozole (ZD
`0
`1033): 2,2
`[5-(1H-1,2,4-triazol-1-yl-methyl)-1,3-phenylene] bis(2-methlypropiononitrile); HPC: hydroxypropyl
`cellulose.
`
`Introduction
`
`Estrogens are known to be important in the growth
`of breast cancers in both pre and postmenopausal
`women. Estrogen receptor concentrations in breast
`cancers increase with age even after menopause. This
`results in a higher fraction of postmenopausal patients
`with hormone sensitive cancers than among premen-
`opausal patients [1]. While the ovary is no longer the
`
`major source of estrogens in older women, production
`is increased in peripheral sites and contributes to the
`stimulation of breast cancers [2].
`Two strategies that are now used to ameliorate the
`growth effects of estrogens are inhibition of estrogen
`action by compounds interacting with estrogen recept-
`ors, and inhibition of estrogen synthesis. The first
`method used clinically was the inhibition of estrogen
`action. The antiestrogen tamoxifen (ICI 46,474), a
`
`MYLAN PHARMS. INC. EXHIBIT 1030 PAGE 1
`
`

`
`184
`
`Q Lu et al.
`
`nonsteroidal antiestrogen, was developed in 1967 and
`entered clinical trials for advanced breast cancer in
`1971 [3]. Tamoxifen is effective as adjuvant therapy
`for postmenopausal, hormone responsive breast can-
`cer [4]. However, it has been reported that long-term
`tamoxifen treatment increases the risk of developing
`endometrial carcinoma and the incidence is correl-
`ated with the duration of treatment [5]. This effect
`is thought to be due to the partial agonistic action of
`tamoxifen. The steroidal antiestrogen faslodex (ICI
`182,780) was developed in the late 1980’s. This com-
`pound is a more potent antiestrogen than tamoxifen
`and is without agonistic effects [6, 7]. Clinical trials
`with faslodex in breast cancer patients are in progress.
`Estrogen synthesis is mediated by the enzyme
`aromatase and is the last step in the steroidogenic
`pathway. Inhibition of aromatase should effectively re-
`duce estrogen production from all sources in the body
`without affecting the biosynthesis of other steroid
`hormones. Selective aromatase inhibitors were first
`reported by our laboratory in 1973 [8]. One of these
`compounds, 4-hydroxyandrostenedione [9], has now
`been developed for breast cancer treatment [10, 11].
`Recently, several non-steroidal aromatase inhibitor
`compounds have been reported [12]. These include
`triazole derivatives based on antifungal agents that
`inhibit P-450 enzymes. Triazoles are noted for their
`favorable pharmacology. Two triazole compounds, let-
`rozole (CGS 20,267) [13] and anastrozole (ZD 1033)
`[14], are now approved as second line agents in the
`treatment of advanced breast cancer in postmeno-
`pausal patients. Both are selective aromatase inhibitors
`and are well tolerated by patients [15–18].
`Comparison of these new agents is difficult in the
`clinic since patients receive approved drugs until they
`relapse. Therefore, most studies to date with aro-
`matase inhibitors and new antiestrogens have been
`carried out in patients with advanced disease who
`have been previously treated with tamoxifen or other
`agents. To address this problem, we developed a
`model for postmenopausal breast cancer which is re-
`sponsive to both antiestrogens and aromatase inhibit-
`ors [19, 20]. After menopause, estrogens are produced
`by the aromatization of androgens in non-ovarian tis-
`sue [1], including the normal breast [21] and breast
`tumors themselves [21–23]. In our model, hormone
`responsive MCF-7 human breast cancer cells are trans-
`fected with the human aromatase gene and serve as
`the source of estrogen in the ovariectomized mouse. In
`the present study, we used this intratumoral aromatase
`model to determine the effects on tumor growth of
`
`the new nonsteroidal aromatase inhibitors letrozole
`and anastrozole, and the antiestrogens tamoxifen and
`faslodex. Furthermore, we addressed the question of
`whether the inhibition of estrogen synthesis together
`with inhibition of estrogen action would be more ef-
`fective in controlling breast tumor growth than either
`one alone. Therefore, the effect of combined treatment
`with aromatase inhibitors and antiestrogens was also
`investigated.
`
`Methods
`
`Radiometric (3H2O) assay of aromatase in human
`placental microsomes
`
`Preparation of human placental microsomes and aro-
`matase activity assays were performed as previ-
`ously described [9,19]. Briefly, 200 mg of placental
`microsomes were mixed with 0.3 mCi of [1b3H]-
`androstenedione, androstenedione (0–2000 nM), with/
`without 5 mM of the aromatase inhibitors letrozole and
`anastrozole, and 1.25 IU/ml of the NADPH generating
`system (NADP, glucose-6-phosphate, and glucose-6-
`phosphate dehydrogenase) in 1 ml 0.1 M phosphate
`(cid:14)
`C,
`buffer, pH 7.4. Following a 30 min incubation at 37
`the assay was terminated by adding 2 ml of chloroform
`to extract unconverted substrate and other steroids. An
`aliquot of 0.7 ml of the aqueous phase was treated with
`2.5% activated charcoal suspension (0.7 ml) to remove
`any residual steroids. Tritiated water (3H2O) formed
`during the aromatization of [1b3H]-androstenedione
`to estrogen was measured by counting the radioactiv-
`ity in the aqueous supernatant. The Ki values were
`determined by the Lineweaver and Burke method.
`
`Radiometric (3H2O) assay of aromatase in MCF-7CA
`cells
`
`Cell Culture
`Human breast carcinoma cells (MCF-7) stably trans-
`fected with the human placental aromatase gene
`(MCF-7CA) were used as previously described [19].
`The cells were cultured in Eagle’s minimum essential
`medium containing 5% fetal bovine serum and neo-
`mycin (600 mg/ml). The culture medium was changed
`twice weekly.
`
`Assay
`MCF-7CA cells were plated into 6-well plates
`(100,000 cells/well) and left overnight to attach. For
`
`MYLAN PHARMS. INC. EXHIBIT 1030 PAGE 2
`
`

`
`Aromatase inhibitors and antiestrogens in a mouse model
`
`185
`
`determination of IC50 values for the aromatase inhib-
`itors, the cells were washed and incubated for 2 h with
`0.5 mCi of [1b3H]-androstenedione in 1 ml of medium
`in the presence of letrozole or anastrozole at concen-
`trations of 0.001–100 nM. After incubation, medium
`was transferred to a glass test-tube and 300 ml of tri-
`chloracetic acid was added to precipitate the proteins.
`After centrifugation, 1 ml of medium was mixed with
`2 ml of chloroform to extract unconverted substrate
`and other steroids. The aromatization assay was then
`performed as above [19].
`
`Athymic mice
`
`Female BALB/c athymic mice 4–6 weeks of age were
`obtained from Charles River Laboratories (Boston,
`MA). The animals were housed in a pathogen-free
`environment under controlled conditions of light and
`humidity and received food and water ad libitum.
`Ovariectomy was carried out under metathane anes-
`thesia 1–3 days before cell inoculation.
`
`Inoculation of MCF-7CA cells to athymic mice
`
`Subconfluent MCF-7CA cells were scraped into
`Hank’s solution and centrifuged at 1000(cid:2)g for 10 min
`(cid:14)
`C. The cells were then resuspended in Matrigel
`at 4
`(10 mg/ml, kindly provided by Dr Hynda Kleinman,
`NIH) to make a cell suspension of 3 (cid:2) 107 cells/ml.
`Each mouse was inoculated sc in four sites with
`0.1 ml of the cell suspension. Beginning on the day
`of inoculation and for the duration of the experiment,
`animals were injected with 0.1 mg per mouse per day
`sc androstenedione, substrate for aromatization to es-
`trogens. Animals were weighed and tumor growth
`rates were determined by measuring the tumors with
`calipers every week. Tumor volumes were calculated
`according to the formula 4=3 (cid:2) (cid:25) (cid:2) r2
`(cid:2) r2.r1 < r2/
`[19, 20].
`
`1
`
`Treatments
`
`Treatments began when all tumors had reached a
`measurable size ((cid:24)500 mm3). This occurred 28–35
`days after androstenedione injections began. Mice
`were then treated daily with sc injections of the aro-
`matase inhibitors, letrozole (CGS 20,267; MW 285.1)
`(kindly provided by Dr Ajay Bhatnagar, Novartis,
`Basel Switzerland) and anastrozole (ZD 1033; MW
`293.4) (kindly provided by Dr Michael Dukes, Zeneca
`Pharmaceuticals, Macclesfield, UK), and with tamox-
`ifen (ICI 47,474), or a combination of letrozole plus
`
`tamoxifen, or anastrozole plus tamoxifen in 0.3% hy-
`droxypropylcellulose (HPC). The antiestrogen faslo-
`dex (ICI 182,780) was injected once per week in
`oil (as kindly provided by Dr. A Wakeling, Zeneca
`Pharmaceuticals, Macclesfield, UK). Control anim-
`als were given injections of the vehicle (0.3% HPC,
`0.1 ml/mouse/day) sc daily. The treatments lasted 5–6
`weeks as indicated in the figures. Animals were autop-
`sied 4–6 h after the last injection. Tumors and uteri
`were removed from the mice, cleaned, and weighed.
`
`Tumor estradiol and letrozole concentrations
`
`Tumors from mice in each treatment group were
`(cid:14)
`pooled and homogenized in PBS buffer at 4
`C. The
`steroids were extracted with diethyl ether and estradiol
`isolated by celite chromatography. Estradiol concen-
`trations were measured in triplicate aliquots of the
`homogenates by RIA. The assay was performed us-
`ing an estradiol antibody and iodinated estradiol (ICN,
`Boston). The sensitivity of the assay was 1 pg/ml [25,
`26]. Serum concentrations of letrozole were measured
`using HPLC/UV and kindly performed by G. Leferre,
`Novartis Pharma, DMPK, France.
`
`Statistics
`
`The effects of treatment on tumor weights were com-
`pared using one way ANOVA followed by Newman
`Keuls multiple range test when required.
`
`Results
`
`Ki values for the aromatase inhibitors letrozole and
`anastrozole were obtained by incubating placental mi-
`crosomes with 0.3 mCi [1b3H]-androstenedione for
`30 min with a range of inhibitor concentrations. The
`−9 M for letrozole and 5:35 (cid:2)
`Ki value was 1:02 (cid:2) 10
`−9 M for anastrozole. Thus,
`letrozole is approx-
`10
`imately five times more potent than anastrozole in
`inhibiting aromatase activity of placental microsomes.
`Both compounds showed competitive inhibition of the
`enzymes, as previously reported [13, 14].
`Inhibition of aromatase was determined in semi-
`confluent cultures of MCF-7CA cells.
`IC50 values
`for the aromatase inhibitors letrozole and anastro-
`zole were obtained by incubating MCFCA cells with
`0.5 mCi [b3H]-androstenedione for 2 h in the presence
`of compounds. The experiment was repeated three
`times and the mean values were presented in Figure 1.
`The IC50 value was 0:35 (cid:6) 0:07 nM (SE) for letrozole
`
`MYLAN PHARMS. INC. EXHIBIT 1030 PAGE 3
`
`

`
`186
`
`Q Lu et al.
`
`Figure 2a. Comparison of the effect of doses of anastrozole (10 and
`60 mg per mouse per day), faslodex (ICI, 5 mg per mouse per week),
`and letrozole (10 mg per mouse per day) on the volume of MCF-7CA
`breast tumors in ovariectomized nude mice.
`
`Figure 1. The IC50 values for letrozole (CGS 20,267) and anastro-
`zole (ZD 1033) in human breast carcinoma cells transfected with
`the aromatase gene (MCF-7CA). Semi-confluent cultures of cells
`were incubated with 0.5 mCi [1b3H]-androstenedione for 2 h in the
`presence of the concentrations of drugs indicated.
`
`and 3:62 (cid:6) 0:90 nM (SE) for anastrozole. Letrozole is
`approximately 10-fold more potent than anastrozole in
`inhibiting aromatase activity in the MCFCA cells.
`Several doses of antiestrogens and aromatase in-
`hibitors were initially evaluated (Figures 2A & 2B).
`Twenty-eight days after inoculation, groups of five
`mice received letrozole (10 mg per mouse per day, sc),
`anastrozole (10 mg per mouse per day, sc), anastrozole
`(60 mg per mouse per day, sc), or faslodex (5 mg per
`mouse per week, sc). After treatment for 28 days, the
`total tumor volumes for control mice had increased by
`145.9% (Figure 2A.). The total tumor volumes in the
`anastrozole (10 mg), anastrozole (60 mg per mouse per
`day, sc ), and faslodex groups decreased and to about
`the same extent during treatment, to 95.6%, 78.2%,
`and 63.2%, respectively. Tumor volume was markedly
`decreased by letrozole treatment to 22.4%. The mean
`tumor weights for letrozole (32:1(cid:6) 21:0 mg), anastro-
`zole (10 mg per mouse per day, 133:2 (cid:6) 22:0 mg), and
`faslodex (114:9(cid:6)24:4 mg) were significantly less than
`for the controls (270:0 (cid:6) 27:7 mg) (p < 0:05) (Figure
`2B). However, tumor weights of the anastrozole (10
`and 60 mg) treated mice were not significantly differ-
`ent from each other or from faslodex treated animals.
`Although anastrozole (10 mg/day) showed an erratic
`
`Figure 2b. Comparison of the effect of treatment with anastrozole
`(10 and 60 mg per mouse per day), faslodex (ICI, 5 mg per mouse
`per week), and letrozole (10 mg per mouse per day) for 28 days on
`the mean tumor weights of ovariectomized nude mice.
`
`pattern of growth and regression during the few weeks
`of treatment, the other treatments showed more typical
`gradual reduction in tumor volume.
`Since the experiment in Figure 2A indicated that
`10 mg per mouse per day letrozole caused marked
`tumor regression, 5 mg per mouse per day was ini-
`tially evaluated along with tamoxifen (3 mg per mouse
`per day) (Table 1, # 1).
`In this experiment,
`the
`mean tumor weight for mice treated for 42 days
`with letrozole (5 mg per mouse per day) was 30:7 (cid:6)
`5:2 mg, with tamoxifen (3 mg per mouse per day) was
`128:7 (cid:6) 20:1 mg, and with letrozole plus tamoxifen
`was 115:5 (cid:6) 19:4 mg. All values were significantly
`less than the controls (166:7(cid:6)32:1 mg) (p < 0:05). In
`addition, the mean tumor weight of the letrozole group
`
`MYLAN PHARMS. INC. EXHIBIT 1030 PAGE 4
`
`

`
`Table 1. The effect of aromatase inhibitors and antiestrogens on tumor and uterine weight
`
`Aromatase inhibitors and antiestrogens in a mouse model
`
`187
`
`n
`
`5
`
`t
`
`42
`
`5
`
`35
`
`4
`
`42
`
`5
`
`35
`
`Experiment
`
`Treatment
`
`Control
`Letrozole
`Tamoxifen
`Let C Tam
`
`Control
`Anastrozole
`Tamoxifen
`Anast C Tam
`
`Control
`Letrozole
`Tamoxifen
`Let C Tam
`Anastrozole
`Anast C Tam
`
`1
`
`2
`
`3
`
`4
`
`Uterine weight (mg)
`Tumor weight (mg)
`43:5 (cid:6) 10:1
`146:7 (cid:6) 32:1
`30:7 (cid:6) 5:2a;b
`26:2 (cid:6) 4:2a
`50:6 (cid:6) 7:6
`128:7 (cid:6) 20:1
`51:0 (cid:6) 3:9
`115:5 (cid:6) 19:4
`68:3 (cid:6) 14:0
`743:3 (cid:6) 157:4
`452:9 (cid:6) 78:9a
`28:0 (cid:6) 7:2a
`62:8 (cid:6) 4:5
`661:9 (cid:6) 58:3
`412:7 (cid:6) 90:2a
`48:0 (cid:6) 5:0
`45:6 (cid:6) 2:4
`297:6 (cid:6) 54:5
`22:9 (cid:6) 3:7a
`20:2 (cid:6) 1:1a;b
`60:7 (cid:6) 3:2
`178:9 (cid:6) 22:9a
`52:0 (cid:6) 8:9a;b
`41:2 (cid:6) 2:4a;b
`24:5 (cid:6) 1:2a
`104:3 (cid:6) 14:3a;b
`44:6 (cid:6) 2:2
`132:8 (cid:6) 26:5a;b
`46:7 (cid:6) 3:3
`506:8 (cid:6) 112:5
`Control
`42:5 (cid:6) 6:8a;b;c;d
`19:4 (cid:6) 0:9
`Letrozole
`182:0 (cid:6) 35:5a
`14:6 (cid:6) 2:1a
`Faslodex
`14:5 (cid:6) 1:2a
`66:2 (cid:6) 16:9a;b;c;d
`Let C Fas
`18:4 (cid:6) 2:7a
`137:4 (cid:6) 19:3a;b
`Anastrozole
`19:8 (cid:6) 4:5a
`220:7 (cid:6) 27:8a
`Anast C Fas
`32:5 (cid:6) 2:5
`255:0 (cid:6) 55:1
`Tamoxifen
`24:8 (cid:6) 4:9
`345:4 (cid:6) 56:6
`Tam C Fas
`Groups of ovariectomized nude mice each with four tumors of MCF-7 breast cancer cells
`transfected with the aromatase gene were injected sc daily with androstenedione (0.1 ml per
`mouse per day). When tumors had reached a measurable size, the mice were treated with
`vehicle (controls), letrozole (let: 5 mg per mouse per day sc), anastrozole (anast: 5 mg per
`mouse per day sc), tamoxifen (tam: 3 mg per mouse per day sc), or faslodex (fas: 70 mg per
`mouse per week sc). Animals were sacrificed on the last day of treatment and tumors and
`uteri removed and weighed.
`n D number of mice.
`t D duration of treatment.
`a p < 0:05 from control.
`b p < 0:05 from tamoxifen.
`c p < 0:05 from faslodex.
`d p < 0:05 from anastrozole.
`
`was significantly less than of the tamoxifen and the
`letrozole plus tamoxifen groups (p < 0:05). Tamox-
`ifen treatment resulted in a reduction in tumor growth,
`whereas with letrozole treatment, tumor weights actu-
`ally decreased from initial volumes. This experiment
`determined that letrozole was effective at 5 mg per
`mouse per day for 42 days. In this model, uterine
`weight is maintained by estrogen produced by the tu-
`mor [20, 26]. The mean uterine wet weight of the
`group treated with letrozole (26:2 (cid:6) 4:3 mg) was sig-
`nificantly less than the control group (43:5(cid:6) 10:1 mg)
`(p < 0:05) (Table 1). In contrast, the mean uter-
`
`ine weight of the animals treated with letrozole plus
`tamoxifen was not significantly different from those of
`mice treated with tamoxifen alone or vehicle. The con-
`centration of letrozole measured in pooled serum was
`90 nmol/l for the group treated with letrozole alone,
`and 74.3 nmol/l for the group treated with letrozole
`plus tamoxifen; none was detected in the serum of the
`control group.
`In the second experiment (Table 1, # 2), we com-
`pared the effects of anastrozole (5 mg per mouse per
`day) in combination with tamoxifen. Treatments were
`started in groups of five mice, 35 days after inocu-
`
`MYLAN PHARMS. INC. EXHIBIT 1030 PAGE 5
`
`

`
`188
`
`Q Lu et al.
`
`Figure 3. The effet of treatment with tamoxifen (3 mg per mouse per day), anastrozole (ZD 1033) (5 mg per mouse per day), letrozole (CGS
`20,267) (5 mg per mouse per day), and the aromatase inhibitors in combination with tamoxifen on the volume of MCF-7CAbreast tumors in
`ovariectomized nude mice.
`
`lation, and continued for 35 days. The mean tumor
`weights for mice treated with anastrozole (5 mg per
`mouse per day) and with anastrozole plus tamox-
`ifen were significantly less than for the controls
`(p < 0:05). Tamoxifen and tamoxifen plus ana-
`strozole treatment did not reduce the uterine weight
`significantly compared to the control. However, the
`mean uterine weight of the anastrozole group was
`significantly less than that of the controls (p < 0:01).
`The above treatments were then compared together
`in one experiment, as shown in Figure 3 and Table 1,
`# 3. Twenty-eight days after inoculation, groups of
`four mice received letrozole (5 mg per mouse per
`day, sc), anastrozole (5 mg per mouse per day, sc),
`tamoxifen (3 mg per mouse per day, sc),
`letrozole
`(5 mg per mouse per day, sc) plus tamoxifen (3 mg
`per mouse per day, sc), anastrozole (5 mg per mouse
`per day, sc) plus tamoxifen (3 mg per mouse per day,
`sc), or vehicle. Consistent with the previous experi-
`ments, tumor volumes in the letrozole and letrozole
`plus tamoxifen treatment groups were decreased after
`6 weeks of treatment by 64% and 27%, respectively
`(Figure 3). Tumor growth in the anastrozole and ana-
`strozole plus tamoxifen groups was reduced compared
`to that of the control mice. Thus,
`tumors changed
`during treatment by 160.0% and 186.0% from their
`original volumes, respectively (Figure 3). Mean tumor
`weights of all treatment groups were significantly less
`
`than for the control group (p < 0:05) (Table 1, #
`3). The concentration of estradiol in the tumors was
`markedly reduced by letrozole from 460 to 20 pg/mg
`tissue. There was no difference between the effect of
`anastrozole and anastrozole plus tamoxifen on tumor
`weight, although letrozole was more effective alone
`than when combined with tamoxifen. Thus, the com-
`bination of aromatase inhibitors and tamoxifen was
`no more effective than the aromatase inhibitor alone
`for blocking estrogen stimulated growth in this mouse
`model.
`To compare the effect of the pure antiestrogen,
`faslodex (ICI 182, 780) with tamoxifen and the aro-
`matase inhibitors, all compounds were evaluated in
`the same experiment. As seen in Table 1, growth rates
`of tumors in the control mice varied among exper-
`iments. This was most likely due to the aromatase
`activity of the MCF-7 cells, which declines as cultured
`cells are passaged. Although we limit the number of
`passages, there is variation between first and last pas-
`sage. However, the pattern of response to treatment
`with the different compounds was quite consistent
`from experiment to experiment. The animals received
`the aromatase inhibitors and tamoxifen in the same
`doses as stated above, whereas faslodex was injected
`sc at 70 mg per mouse per week, beginning 28 days
`after inoculation in groups of five mice (Figure 4).
`The treatment continued for 35-days. The percentage
`
`MYLAN PHARMS. INC. EXHIBIT 1030 PAGE 6
`
`

`
`Aromatase inhibitors and antiestrogens in a mouse model
`
`189
`
`Figure 4. The effect of treatment with tamoxifen (3 mg/day), faslodex (ICI 182,780) (70 mg per mouse per week), anastrozole (ZD 1033)
`(5 mg per mouse per day), letrozole (CGS20,267) (5 mg per mouse per day), and combinations of aromatase inhibitors and antiestrogens on the
`volume of MCF-7CA breast tumors in nude mice.
`
`change in tumor volumes measured weekly are shown
`in Figure 4. The total tumor volume for the control
`group increased by 322.1% during the course of the
`experiment. Although tamoxifen alone or the two anti-
`estrogens combined reduced tumor growth compared
`to the control group,
`tumor volumes increased by
`177.6% with tamoxifen alone and by 201.2% with the
`combination during the 35-day treatment. The mean
`tumor weights from these two groups of animals were
`not significantly different from the controls or one
`another. The pure antiestrogen was more effective in
`reducing tumor growth when administered alone or
`combined with an aromatase inhibitor. Faslodex, ana-
`strozole, and the combination of faslodex plus anastro-
`zole reduced tumor growth to about the same extent.
`The tumor weights were significantly less (p < 0:05)
`than for the control group of 506:8 (cid:6) 112:5 mg and
`were 137:4 (cid:6) 19:3 mg for anastrozole, 182 (cid:6) 35:5 mg
`for faslodex, and 220:7 (cid:6) 27:8 mg for the combined
`treatment (Table 1, # 4). Letrozole alone or in combin-
`ation with faslodex showed marked tumor regression
`as indicated by reductions in tumor volumes of 56.4%
`and 69.0%, respectively (Figure 4). Tumor weights
`were significantly reduced to 42:5 (cid:6) 6:8 mg for let-
`rozole and 66:2 (cid:6) 16:9 mg for letrozole plus faslodex
`compared to 506:8 (cid:6) 112:5 mg for the control group
`(p < 0:01). The mean uterine weights of groups
`treated with an aromatase inhibitor and faslodex, alone
`and combined, were all significantly less than that of
`
`the control group (p < 0:01). However, mice treated
`with tamoxifen alone had uterine weights similar to
`the controls.
`
`Discussion
`
`The results of these studies indicate that the aromatase
`inhibitors anastrozole and letrozole, as well as the new
`pure antiestrogen faslodex, have potent antitumor ef-
`fects in the mouse tumor model. In previous studies,
`we have shown that these compounds cause more sup-
`pression of tumor growth in this mouse model when
`administered in higher doses [27]. In the first exper-
`iment (Figure 2) we compared doses of anastrozole
`(10 mg per mouse per day; 60 mg per mouse per day),
`letrozole (10 mg per mouse per day), and faslodex
`(5 mg per mouse per week). Both doses of anastrozole
`were effective in suppressing tumor growth although
`there was no significant difference between their ef-
`fects on tumor weight. However, a rather erratic
`pattern of growth and regression was seen with the
`10 mg/day dose, whereas there was a large variation
`in the tumor weights of the 60 mg/day treated mice.
`The dose of faslodex was formulated in castor oil and
`used as previously reported by Osborne et al. to be
`effective in the mouse model [28]. Faslodex and let-
`rozole not only reduced tumor growth but also caused
`regression of established tumors, although the effect
`
`MYLAN PHARMS. INC. EXHIBIT 1030 PAGE 7
`
`

`
`190
`
`Q Lu et al.
`
`with letrozole was more marked. In order to address
`the question of whether the inhibitors of estrogen ac-
`tion together with inhibitors of estrogen synthesis are
`more effective than either type of compound alone,
`we have compared these compounds with one another
`within the same experiments at suboptimal and equiva-
`lent doses. Because the 10 mg dose of letrozole caused
`almost complete tumor regression, 5 mg per mouse per
`day was selected for letrozole; anastrozole was com-
`pared at this same dose. Although our previous studies
`showed tamoxifen to be effective in suppressing tumor
`growth at higher doses [19], we used a low dose of
`tamoxifen (3 mg per mouse per day) which had been
`shown to reduce tumor growth to about 60% of the
`control [20]. This allowed us to determine whether
`there is further reduction in tumor growth when an-
`imals are treated with a combination of tamoxifen and
`an aromatase inhibitor or the pure antiestrogen, com-
`pared with either compound alone. However, as even
`5 mg letrozole alone produced a marked suppression in
`tumor growth, any further reduction would have been
`difficult to detect.
`Our in vitro kinetic studies in human placental mi-
`crosomes indicate that letrozole is approximately five
`times more potent than anastrozole based on the Ki
`values. However, in cell culture (MCFCA), there is
`a 10-fold difference in IC50 values between the two
`inhibitors in favor of letrozole. This suggests that
`there may be cellular processes which affect these
`compounds differently. Consistent with these findings,
`letrozole treatment was more potent than anastrozole
`in suppressing growth of tumors of MCF-7CA cells in
`the mice. Although no such difference in the response
`to treatment with anastrozole and letrozole has been
`observed to date in breast cancer patients, this may
`be because the doses administered to patients are the
`maximum effective doses of each drug.
`In our mouse experiments, when tamoxifen was
`added to letrozole and anastrozole treatment, the com-
`bination was only about as potent as each aromatase
`inhibitor alone. Although there was no apparent dif-
`ference in tumor volumes (Figure 4), the mean tumor
`weights of mice receiving the combined treatment
`were greater than those of tumors from the letrozole
`group (Table 1). The concentration of estrogen meas-
`ured in the tumor tissue at the end of the experiment,
`where mice were treated with the inhibitors in combin-
`ation with either tamoxifen or faslodex, indicated that
`letrozole caused a marked and significant reduction
`in estradiol concentrations to very low levels com-
`pared to the control value. Unfortunately, values for
`
`tumor samples from the tamoxifen or tamoxifen plus
`aromatase inhibitor treated animals were invalid as
`tamoxifen appeared to compete with estradiol for the
`antibody used in the assay. Further studies are needed
`to provide useful information on the reduction in estra-
`diol levels in mice treated with combinations of these
`agents. Nevertheless, as estrogen levels were reduced
`by letrozole, the results suggest that the partial agon-
`ist effect of tamoxifen on the tumors may be more
`apparent when the concentration of estrogen is low.
`Tumor volumes compared at the beginning and end
`of the experiment, indicated that letrozole caused tu-
`mors to regress by 62%. Many tumors had disappeared
`almost completely at the end of the experiments. We
`had noted this effect previously and reported a signi-
`ficant difference in tumor weights at the beginning and
`end of the experiment [27]. Letrozole was consistently
`the most potent agent evaluated in this mouse model.
`The effect of combining tamoxifen with anastro-
`zole was similar to that of combining letrozole and
`did not cause an increase in antitumor efficacy over
`anastrozole alone. Thus, tamoxifen did not appear to
`block the effects of the reduced levels of estrogen on
`tumor growth.
`The results of combining the pure antiestrogen
`faslodex (70 mg per mouse per week) with the aro-
`matase inhibitor (Figure 3) did not cause further re-
`duction in tumor growth than with these agents alone.
`The results of combining faslodex with tamoxifen also
`suggest that tamoxifen has a partial agonistic effect
`on tumor growth. Faslodex is known to block dimer-
`ization of the estrogen receptor, thereby preventing
`transcription [29]. The dose of 70 mg per mouse per
`week of faslodex caused a greater response than 3 mg
`per mouse per day of tamoxifen alone, as shown in
`Figure 4. The combination of tamoxifen and faslodex
`was also less effective than faslodex alone, but similar
`to tamoxifen alone. While it is possible that altered
`pharmacology of the drugs in combination might ex-
`plain the lack of greater antitumor efficacy, the results
`suggest that this dose of faslodex (70 mg per mouse
`per week) only partially blocked the action of estrogen
`on the tumors and uterus. The finding that combin-
`ing faslodex with the aromatase inhibitors produced
`a similar response to those of the inhibitors alone is
`consistent with this conclusion. However, in combin-
`ation with tamoxifen, the dose of faslodex may not be
`sufficient to counteract completely the agonist action
`of tamoxifen. It should be pointed out that patients are
`likely to be treated with a higher dose relative to the
`doses used in these mouse experiments.
`
`MYLAN PHARMS. INC. EXHIBIT 1030 PAGE 8
`
`

`
`Aromatase inhibitors and antiestrogens in a mouse model
`
`191
`
`In conclusion, our results show that the combin-
`ation of aromatase inhibitors with either antiestrogen,
`tamoxifen or faslodex, are not more potent in blocking
`estrogen stimulated tumor growth than the inhibit-
`ors alone. The data suggests that letrozole alone may
`provide the most effective estrogen blockade. Inhibi-
`tion of estrogen synthesis and action with these agents
`did not result in greater antitumor efficacy. The cur-
`rent practice of sequential use of these agents appears
`to be more likely to benefit breast cancer patients by
`extending treatment options.
`
`Acknowledgements
`
`This work was supported by NIH CA-62483. Thanks
`are due to Dr Laurence Demers, Hershey Medical
`Center, Hershey, PA, for measuring the concentration
`of estradiol in the tumor tissue.
`
`References
`
`1. Hemsell DL, Grodin J, Brenner PF, Siiteri PK, MacDonald
`PC: Plasma precursors of estrogen II. Correlation of the extent
`of conversion of plasma androstenedione to estrone with age.
`J Clin Endocrinol Metab 38: 476–479, 1974
`2. McGuire WL: An update on estrogen and progesterone re-
`ceptors in prognosis for primary and advanced breast cancer.
`In: Iacobelli, et al. (ed.) Hormones in Cancer, Vol. 15, Raven
`Press, NewYork, 1980, pp. 337–344
`3. Cole MP, Jones CTA, Todd IDH: A new antiestrogen agent in
`the late breast cancer. An early clinical appraisal of ICI 46,474.
`Br J Cancer 25: 270–275, 1971
`4. Early Breast Cancer Trials Collaborative Groups: Sy