`
`71
`
`Selective Estrogen Receptor Modulators (SERMS)
`
`T. A. Grese* and J. A. Dodge
`
`Lilly Research Laboratories, Eli Lilly and Company, lrrrlia/rapo/is, Indiana 46285, USA
`
`
`
`|7B-estradiol and estrone, have
`Abstract: Naturally occurring estrogens, such as
`traditionally been thought to play a central role in the development and maintenance ol
`the female reproductive system and secondary sexual characteristics. In recent years, their
`beneficial effects on the skeleton,
`the cardiovascular system, and the central nervous
`system. as well as the cancer risks associated with long term exposure have also been
`recognized. The widespread use of “ anticstrogcns “ such as tamoxifen for the prevention
`and treatment of breast cancer has revealed that such compounds, whilc functioning as estrogen aritagorrrsts in
`mammary tissue, actually mimic the effects of estrogen in other tissues. The search for more selective agents
`has led to the development of raloxifene. a Selectivc Estrogen Receptor Modulator, which functions as an
`estrogen antagonist
`in the breast and uterus and as an estrogen agonist
`in the skeleton and cardiovascular
`system. Recent progress in the development of SERr\rls is the subject of this review, with an emphasis on
`structure activity relationships and on their effects in non—traditional target tissues.
`
`Introduction
`
`The central role played by endogenous estrogens. such as
`l7|3~estradiol, 1, and estrone, 2,
`in the development and
`maintenance of the female sex organs, mammary glands, and
`other sexual characteristics has long been recognized [L2].
`Recently,
`their involvement
`in the growth and function of a
`number of other tissues, such as the skeleton, the cardiovascular
`system, and the central nervous system,
`in both males and
`females has been recogrrized [3,4].
`
`The primary site of estrogen biosynthesis in the adult female
`is the ovary. After the menopause,
`the ovarian prodrrctiori of
`estrogens declines dramatically producing a wide range of
`primary and secondary physiological effects [5,6]. The decline
`in levels of circulating estrogens has also been linked to a
`number of pathological conditions
`incltrding osteoporosis
`[7,8], coronary artery disease [83,10]. depression [I l,l2]. and
`Alzheimer‘s disease [12,13]. Estrogen replacement
`therapy
`(ERT) has proven effective in reducing the freqrrency and severity
`of these pathologies, but
`the increased risk of endometrial
`cancer observed with ERT has necessitated the development of
`therapeutic regimens in which the uterine effects of estrogen are
`opposed by progestirr treatment (hormone replacement
`therapy
`or HRT)
`[14,15]. Side—effects of progestin treatment. such as
`resumption of menses, central nervous system disturbances, and
`the possibility of attenuated cardiovascular benefits. have
`unfortunately resulted in decreased patient compliance [l6,l7].
`
`Table 1.
`
`Classification of Estrogen Receptor Modulators
`
`Furthermore, recent studies which confirm the increased risk of
`breast and endometrial cancer associated with long term ERT or
`HRT have led to the search for treatment alternatives [l8,l9|.
`
`importance of estrogen in the development and
`The
`maintenance of the female reproductive system has led to the
`pharmaceutical development of a variety of steroidal and non-
`steroidal eomporrnds which interact with the estrogen receptor
`(ER) as contraceptives and for the treatment of breast cancer,
`uterine dysfunction, and other reproductive disorders. Several
`reviews of ER—modu|ators, with a particular‘ emphasis on their
`rrtrlity in the treatment of breast cancer, have been recently
`published
`[20,2l]. Early synthetic
`cstrogens
`such
`as
`dietlrylstilbestrol (DES), 3, and hexestrol. 4, were once widely
`utilized as estrogen substitutes, but due to concerns similar to
`those encountered with the natural hormones and other side
`
`their utility has diminished. The discovery that
`effects
`compounds such as
`\’lER—25, 5, antagonize the action of
`estrogen in breast
`tissue led to intensive pharmaceutical
`research, culminating in the development of tamoxifen, 6.
`which has found great rrtility in the treatment of breast cancer
`[22]. Early concerns
`that
`the
`long—term rise of
`these
`"antiestrogens" would lead to increased risks of osteoporosis
`and cardiovascular disease have been dispelled by
`the
`paradoxical
`finding that some compounds (Vi.e.
`tamoxifen and
`raloxifene, 7) actually mimic the effects of estrogen in skeletal
`and cardiovasctrlar tissues. although others (’i.e.
`ICI 182780,
`8b) do not
`[23]. Findings such as
`these have led to a
`
`Classification
`
`Uterine Stimulation
`
`l3one/Cardiovascular
`
`Example
`
`Pure Antiestrogerrs
`
`Estrogen Agonists
`
`Partial Agonists (|st Generation SERMs)
`
`2nd Generation SERMS
`
`agonisi
`
`agonist
`
`agonrsr
`
`l7[3—estradiol, 1
`
`tamoxifen. 6
`
`i'aloxrl'crrc, 7
`
`antagonist/neutral
`
`lC1 1827808!)
`
`r3sr—er2s/9s $|5 f)()+.00
`
`© 1998 Bentham Science Publishers BV.
`
`AstraZeneca Ex. 2023 p. 1
`Mylan Pharrns. Inc. V. AstraZeneca AB IPR2016-01325
`
`
`
`72 Current Phamuiceutical Design, I 998, Vol. 4, No. I
`
`Grese and Dodge
`
`HO
`
`pas, 3
`
`
`
`
`
`Me OH
`
`
`
`17B-estradiol, 1
`
`N Et
`
`O/$/
`
`2
`
`MER—25, 5
`
`hexestrol, 4
`
`tamoxifen, 6
`
`
`
`raloxifene, 7
`
`Fig. (1). Representative estrogen receptor modulators.
`
`reclassification of estrogen receptor ligands (Table 1) [24] on
`the basis of
`their effects in various traditional and non-
`
`traditional target tissues. An explosion of research to understand
`the molecular basis for this specificity [25] and a race to develop
`these “designer estrogens” or Selective Estrogen Receptor
`Modulators (SERMs) as pharmaceutical products has also taken
`place [26]. The prototypical 2nd generation SERM, raloxifene,
`7,
`is currently undergoing clinical evaluation for the prevention
`and
`treatment of postmenopausal
`osteoporosis
`[27].
`Nevertheless,
`it should be noted these distinctions may be
`somewhat arbitrary, since there is likely to be a continuum of
`activities from full agonist
`to full antagonist and the relative
`activity of an individual compound may be different for each
`tissue or animal species examined.
`
`In this review, we will discuss the known pharmacology of
`various structural classes of estrogen receptor modulators,
`particularly with respect
`to their effects in non—traditional
`tissues. We will describe the structure-activity relationships of
`these compounds, where such data is available, concentrating
`upon how elements of structure contribute to their tissue—specific
`actions. Finally, we will provide a brief overview of the current
`theories which have been developed to account
`for
`tissue
`specificity of ER-modulators.
`
`on
`
`HO
`
`" (Cl-l2)XR
`
`ICI 164384, 8a x = IO, R = CON(Me)rz—Bu
`lCl l82780, Sb x = 9, R = SO(CH2)3CF2CF3
`
`Steroidal ER Modulators and the Estrogen
`Pharmacophore
`
`Natural and synthetic steroidal estrogens have shown great
`utility and significant therapeutic benefits in the replacement of
`endogenous hormones in postmenopausal women [7—l0,l4-18].
`Although most studies have focused on the efficacy of ERT or
`HRT in the prevention of osteoporosis, cardiovascular disease,
`and disorders of the urogenital
`tract [28], recent reports have
`also described benefits in the central nervous system, including
`improvements
`in cognitive
`function,
`and palliation of
`Alzheimer's disease and postmenopausal depression [1 1-
`13,2930].
`
`ERT and/or HRT have been demonstrated to provide a variety
`of cardiovascular benefits, resulting in a 40-50% reduction in the
`relative risk of coronary disease and atherosclerosis [3l,32].
`The effects of estrogens on cardiovascular risk factors include
`raising serum levels of high—density lipoprotein (HDL)
`cholesterol and apolipoprotein A-l, and lowering levels of low-
`density lipoprotein (LDL)
`cholesterol,
`lipoprotein (a),
`endothelin-l, and apolipoprotein B [33,34]. Estrogen has also
`been demonstrated to have direct and indirect effects on blood
`
`vessel walls
`
`including increased nitric oxide synthesis,
`
`AstraZeneca Ex. 2023 p. 2
`
`
`
`Estrogen Receptor Modulators
`
`Current Phtznmzceutical Design, I 998, Vol. 4, No. I 73
`
`inhibition of vascular smooth muscle cell proliferation, and
`increased vasodilation [35]. Currently it
`is
`felt
`that
`the
`combination of these effects on serum lipids and on vascular
`tone are responsible for the overall cardioprotective effects of
`estrogen therapy.
`
`function
`In the prevention of osteoporosis, estrogens
`primarily as antiresorptive agents, leading to decreased turnover
`of both cortical and cancellous bone [36,37]. As with other
`antiresorptive agents,
`this benefit
`is partially offset by a
`subsequent decrease in bone formation, however the overall
`result of ERT or HRT is a substantial increase in bone mineral
`
`density and a decrease in fracture incidence [38,39]. Although
`ERs have been detected in both osteoblasts and osteoclasts, it is
`currently unclear if the effects of estrogens on bone metabolism
`are direct or indirect [40].
`
`Steroidal Estrogens
`
`Early efforts to identify selective estrogens focused on
`changes in the parent steroid to elicit tissue specific biological
`responses. For example,
`the estrogen metabolites estriol, 9, and
`l7ot-estradiol, 10, were found to be time-dependent mixed
`agonist—antagonists of estrogen, which stimulate
`early
`uterotrophic responses but have little effect on true uterine
`
`hypertrophy and hyperplasia unless administered chronically at
`high doses [4l,42]. Estriol causes significantly less uterine
`hyperplasia than l7B—estradiol and inhibits the development of
`breast cancer in rodents [43]. In addition,
`l7ot-estradiol has been
`shown to exert
`a neuroprotective
`effect
`in
`a human
`neuroblastoma cell
`line
`(SK—N-SH)
`[44]. The estrogen
`metabolite, 2-methoxyestradiol, 11, has been implicated in the
`angiogenesis of vascular tissue and a number of analogs have
`been reported which potently inhibit
`tubulin polymerization
`[45,46]. The estrogen analog l7ot-ethynylestradiol
`(E32), 12,
`has been extensively studied for its bone, uterine, and lipid
`effects due,
`in large part,
`to an enhanced oral activity profile
`relative to 17B-estradiol.
`
`Improvements in tissue selectivity have been observed with
`a family of D~ring halogenated estrones (such as 13) which have
`demonstrated potent
`lipid lowering yet diminished uterine
`hypertrophy relative to estrone [47]. Other attempts to attenuate
`the estrogenic activity of steroids via opening of the steroid
`nucleus, such as 9,11-seco steroids, 14, have met with limited
`success [48,49].
`
`Recently, the components of Premarin® (the most prescribed
`form of ERT) have been evaluated for their lipid lowering effects.
`These conjugated equine estrogens contain sulfate esters of two
`distinct estrogen structural classes; (1) ring B saturated steroids
`
`
` MCO
`
`14
`
` HO
`
`'l7[3—dihydroquiIinen, 17
`
`
`
`l7oL—estradiol, 10
`
`
`
`
`
`HO
`
`cquilin,15
`
`equilinen, 16
`
`
`
`l70t—dihydr0equilinen, 18
`Fig. (2). Steroidal estrogen receptor modulators.
`
`
`
`tibolone, 19
`
`AstraZeneca Ex. 2023 p. 3
`
`
`
`74 Current Pharmaceutical Design. 1998, V01. 4, No. 1
`
`Grese and Dodge
`
`including traditional sex steroid hormones such as estrone, 17B-
`estradiol, and l70t—estradiol, and (2) ring B unsaturated estrogens
`such as equilin (Eq), 15, equilenin (Eqn), 16,
`17B-
`dihydroequilenin (l7B-DHEqn), 17,
`l7]3—dihydroequilin (17[3-
`DHEq),
`l7ot—dihydroequilenin (170t—DHEqn), 18, and l7ot—
`dihydroequilin (170t—DHEq). In 1991, Bhavnani and co—workers
`examined these individual steroids,
`in their unconjugated form.
`to determine their relative binding affinities for the estrogen
`receptor and their in vivo effects on uterine hypertrophy in the
`immature rat
`[50].
`in this
`study,
`the majority of equine
`components mimicked estrogen in their ability to increase
`uterine weight relative to vehicle treated animals. The notable
`exception to this uterotrophic response was l7ot—DHEqn which
`did not cause a significant effect at the dose examined (2 mg/kg).
`More recently,
`the sulfate ester conjugate of l7ot-DHEqn has
`been shown to
`lower
`serum cholesterol
`and increase
`
`hippocampal dendritic spine density in rats, and improve arterial
`vasomotor function in macaques [51].
`Work from our own laboratories on the relative effects of
`
`conjugated equine estrogens on bone versus uterus has shown
`that
`l7ot-DHEqn is a partial estrogen agonist [52]. in this study,
`uterotrophic effects were observed after 4 days of oral dosing for
`Eq. 15, Eqn. 16.
`l7B—DHEqn, 17, and l7ot-DHEqn. 18.
`Increases in uterine wet weight relative to ovariectomized (OVX)
`controls ranged from 263% for Eq to 100% for
`l70t-DHEqii.
`Serum cholesterol levels were lowered with similar potencies for
`all equine estrogens [52]. Bone mineral density measurements
`indicated that 17ot-DHEqn effectively prevented osteopenia in a
`dose-dependent
`fashion after 5—weeks of oral administration
`(59.9% of ovariectomy-induced bone loss was prevented at
`l
`nig/kg,
`ll9 % at
`l0 mg/kg).
`in addition, an average uterine
`weight gain of 100.4% relative to OVX controls was observed at
`the l nig/kg dose [52]. These data demonstrate that
`l7ot~DHEqn
`is a ftill estrogen agonist on bone, but a partial agonist on the
`uterus in the OVX rat and further highlight
`the structural
`significance of both the stereochemistry at
`the l7—position and
`unsaturation in the B»ring.
`
`is a unique steroid that possesses
`Tibolone (OD-14). 19,
`estrogenic, progestenic and androgenic properties. At doses of
`less
`than 2.5 mg/day, OD—l4 appears
`to reduce skeletal
`remodeling without producing concomitant
`endometrial
`stimulation [53]. However, because of its estrogenic activity,
`eiidometrial hypertrophy over
`the
`long term remains
`a
`possibility.
`
`Pure Antiestrogens
`
`While estrogen agoiiists, partial agonists, and SERMs can
`mimic the pharmacology of
`the natural hormone, pure
`antiestrogens (e. g., 8a,b) represent a class of therapeutic
`agents which are devoid of estrogen agonism regardless of the
`target
`tissue.
`initially introduced by Wakeling in 1988,
`these
`compounds demonstrate an absence of estrogeiiic activity on the
`rat uterus, vagina, and hypothalainic—pituitary axis as well as
`effectively antagonizing the stimulatory effects of estrogen
`[54]. In non-reproductive tract tissue, pure antiestrogens behave
`like estrogen antagonists as well. For example,
`lCl 164,384.
`821, and lCl 182,790, 8b, exhibited no capacity for lowering
`serum cholesterol or sparing bone loss in the OVX rat model
`[55]. Recent data suggests that
`[Cl 182,780 has significantly
`complex effects on rat skeletal tissue [56]. For example,
`loss of
`
`cancellous bone is observed in intact rats after administration of
`
`the compound whereas no bone loss is observed in OVX rats.
`
`Estradiol Pharmacophore
`
`Recently, Katzenellenbogen, et. al. have combined literature
`ER binding affinity data for a large number of steroidal estrogen
`analogs with molecular modeling and receptor sequence analysis
`to develop a detailed picture of the estradiol pharmacophore
`[57]. Their study recognizes the important contributions of the
`two hydroxy groups of estradiol to receptor binding, with the 3~
`hydroxy acting primarily as
`a hydrogen bond donor and
`contributing approximately 1.9 kcal/mol
`to the binding free
`energy. while the
`l7B~hydroxy functions primarily as
`a
`hydrogen bond acceptor and contributes approximately 0.6
`kcal/mol
`[58]. The preferred distance between the hydroxy
`functionalities appears to be somewhat flexible, possibly due to
`the inclusion of water molecules in the binding cavity [57].
`Large, preformed, hydrophobic pockets apparently exist within
`the ligand binding domain which are able to accommodate large
`substituents at the 1115- and 70t—p0siti0ns [59]. Smaller pockets
`appear to exist at the 16B~ and l7B—positions, while the 16a-
`position and the aromatic A~ring are relatively intolerant of
`substitution [57]. These properties of the ER ligand binding
`cavity, which were primarily determined empirically, appear to
`be supported by the recently reported X~ray crystal structure of
`the ER ligand binding domain complexed with estradiol [60].
`
`Triphenylethylenes
`
`The most thoroughly investigated class of non—steroidal ER
`modulators
`are
`the
`triphenylethylenes
`(TPE’s),
`such as
`tamoxifen, 6. and clomiphene, 20, A common structural motif
`which is
`incorporated in many classes of molecules with
`estrogen antagonist activity involves the attachment of
`a
`sidechain containing a hydrogen bond acceptor to an ER binding
`core unit. This theme is illustrated for the triphenylethylenes via
`the progression from DES to MER—25 and tamoxifen. Originally
`investigated for contraceptive activity,
`the strong estrogen
`antagonist activity observed with many of these compounds in
`mammary tissue, has led to their development for treatment of
`breast cancer [20d]. The success of tamoxifen in this arena, has
`led to the investigation and development of a wide variety of
`analogs. To date, SAR work in this series has been confined
`primarily to the investigation of antagonist effects in mammary
`and uterine tissue [20]. Recently, reports of estrogen agonist
`effects of some of
`these compounds in the skeletal and
`cardiovascular system have begun to appear [26].
`In general,
`although they partially antagonize the effects of estrogen on the
`uterus, the members of this structural class tend to induce some
`level of uterine stimulation in the absence of endogenous
`estrogen, therefore they have been classified as partial agonists
`or first generation SERMS [61].
`
`to be utilized clinically was
`the first TPE’s
`One of
`clomiphene, 20. Although it was originally developed as a
`contraceptive, clomiphene has been mainly utilized for
`the
`induction of ovulation in anovulatory women [62]. Its effects on
`uterine tissue are complex. and are at least partly complicated by
`its availability as
`a mixture of double—bond
`isomers
`(zuclomiphene and enelomiphene), but
`it appears to cause
`significant stimulation of uterine epithelia in the rat [63,64].
`
`AstraZeneca Ex. 2023 p. 4
`
`
`
`Estrogen Receptor Modulators
`
`Current Phannaceutical Design, I 998, Vol. 4, No. I 75
`
`in
`Clomiphene has been reported to reduce serum cholesterol
`rats, similar to estrogen, however this may not be an ER-
`mediated effect
`[63]. Clomiphene has also been reported to
`inhibit bone resorption in vitro [65] and to protect against bone
`loss in both OVX rats [66] and in postmenopausal women [67].
`Interestingly,
`the individual
`isomers of clomiphene have been
`reported to have similar effects on bone metabolism, while the
`uterine effects are primarily induced by zuclomiphene [68].
`Recently a clomiphene analog, MDL—lO3,323, 2], with
`antiproliferative activity in breast cancer assays has been
`reported to protect against bone loss in OVX rats with minimal
`uterine stimulation [69].
`
`Consistent with the importance of the hydroxyl moieties of
`estradiol
`for receptor binding, tamoxifen, 6, binds only weakly
`to the ER, however evidence suggests
`that
`the primary
`biologically active species may be its 4—hydroxy metabolite
`[70,71]. Estrogen antagonist effects in mammary tissue have
`been demonstrated in a variety of cell
`lines and animal models
`[20d]. Tissue specific estrogen agonist effects have been
`demonstrated in the OVX rat model of estrogen deficiency, where
`
`tamoxifen reduced serum cholesterol by 50% at doses of (11-10
`mg/kg and protected against bone loss with an ED50 of 0.1
`mg/kg [6l,72,73].
`In vitro effects on bone resorption and
`osteoclast viability have also been demonstrated [64,74].
`In a
`primate model,
`tamoxifen was shown to significantly inhibit
`the progression of coronary artery atherosclerosis [75].
`
`Due to the widespread use of tamoxifen in the treatment of
`breast cancer, a large body of clinical evidence with respect to
`its effects in other tissues has also accumulated [22].
`ln the
`cardiovascular
`system,
`tamoxifen
`has been shown to
`significantly reduce risk factors of disease including LDL
`cholesterol,
`lipoprotein (a), and fibrinogen in postmenopausal
`women with little or no effect on triglycerides or HDL
`cholesterol [76,77]. A corresponding decrease in mortality due
`to cardiovascular disease has also been reported [78]. Clinical
`effects on the skeleton have included the preservation of bone
`mineral density at the lumbar spine, femoral neck, and forearm
`in postmenopausal women |77,79] as well as an estrogenalike
`reduction in serum markers of bone turnover
`[79b-d_,80].
`Interestingly,
`in premenopausal women decreases in bone
`
`O/\/
`
`NM 7
`C-
`
`O/\/
`
`NE
`
`2
`
`/K/Q NEI
`
`O
`
`2
`
`\
`
`4
`
`\
`
`Cl
`
`tatnoxifen, 6
`
`clomifene, 20
`
`M[)L.]()3;323, 21
`
`O/\/ NMB2
`
`O
`
`/\/ N
`
`O/\/ NMC2
`
`\
`
`OH
`droloxifene, 22
`
`\
`
`1
`
`idoxifene. 23
`
`\
`
`Cl
`
`toremifene. 24
`
`O/\/ NMc3 I’
`
`(l-lO)3P(())O
`
`Fig. (3). Triphenylcthylene estrogen receptor modulators.
`
`TAT/59, 25
`
`TMl, 26
`
`AstraZeneea Ex. 2023 p. 5
`
`
`
`76 Current Phannaceutical Design, 1998, Vol. 4, No. I
`
`Grese and Dodge
`
`mineral density have been observed with tamoxifen treatment
`[79a,e].
`
`the
`recently been approved for
`Toremifene, 24. has
`treatment of breast cancer and has demonstrated clinical effects
`
`Notwithstanding the positive effects described above for
`tamoxifen, there continues to be considerable concern about the
`increased risk of endometrial cancer which has been associated
`
`with tamoxifen use [81,82]. Estimations of the magnitude of
`this risk vary, however the average value of about a five-fold
`increase is similar to that observed with ERT [83]. Significant
`stimulation of uterine endometrial tissue is also observed in the
`
`[61,84] and DNA adduct formation has been
`OVX rat model
`observed in both rats and humans [85,86]. Tamoxifen has also
`been shown to induce liver cancer in rats [87].
`
`The growing concern over the potential cancer causing or
`cancer promoting effects of tamoxifen has recently led to the
`development of a number of tamoxifen analogs. The hypothesis
`that metabolic hydroxylation of tamoxifen at
`the 4—position is
`in some way responsible for
`these effects has led to the
`investigation of agents in which this metabolic pathway is
`blocked [20d]. Examples of this strategy include droloxifene,
`22, and idoxifene, 23, which have been reported to show
`decreased
`levels
`of DNA adduct
`formation
`and
`
`ln toremifene, 24, chlorination
`hepatocarcinogenicity [88,89].
`of the aliphatic substituent of tamoxifen also appears to reduce
`DNA-adduct formation [90]. Alternative strategies for modifying
`the metabolic fate and/or tissue distribution observed with
`
`tamoxifen are represented by TAT—59, 25, and tamoxifen
`methiodide or TMI, 26. To date these compounds have been
`most extensively evaluated for the treatment of breast cancer
`[90.9l], however recently reports on their effects in non-
`traditional target tissues have begun to appear.
`
`the 3—position of the
`ln droloxifene, 22, hydroxylation at
`TPE core leads to an altered metabolic profile and decreased
`estrogen agonist activity relative to tamoxifen [92,93].
`ln OVX
`rats, droloxifene has been reported to reduce serum cholesterol
`40-46% and to protect against
`loss of bone mineral density,
`similar to tamoxifen but with reduced uterine stimulation [94]. in
`a head-to-head comparison with tamoxifen, droloxifene was
`found to be at least 10-fold less potent in terms of its effects on
`serum cholesterol and bone density, even though it has a tenfold
`higher binding affinity to the ER [95]. Estrogenic effects in the
`skeleton have also been observed by histomorphometry at both
`cancellous and cortical bone sites [96]. Recently, droloxifene
`has been observed to induce apoptosis of both MCF-7 cells and
`osteoclasts in culture, while estrogen has similar effects on
`osteoclasts but
`is mitogenic to MCF—7 cells [97]. This tissue-
`specific difference has led to the hypothesis that a common
`mechanism may account
`for both the estrogen agonist and
`antagonist activities of droloxifene. Although droloxifene has
`been evaluated clinically for efficacy in breast cancer treatment
`[9la], its effects on other estrogen target tissues in humans have
`not yet been reported.
`
`ldoxifene, 23, was designed to reduce both metabolic
`oxidation and N—demethy1ation, via iodination of the 4‘-
`position and replacement of
`the dimethylamino group of
`tamoxifen with a pyrrolidine ring, respectively [98]. As with
`droloxifene.
`idoxifene has been evaluated clinically for breast
`cancer treatment [9lb], and a preliminary report describing its
`effects on serum cholesterol and bone density in the OVX rat has
`also appeared [99].
`It has also been reported to be less
`uterotrophie than tamoxifen [I00].
`
`on serum cholesterol and bone mineral density which are similar
`to those of tamoxifen in postmenopausal breast cancer patients
`[lOl,l02]. Although it has been reported to be less uterotrophic
`in the rat
`[lO3],
`its estrogenic effects on the uterus
`in
`postmenopausal women have been reported to be comparable to
`those of tamoxifen [lO4].
`
`Tamoxifen methiodide, 26, was designed to inhibit crossing
`of the blood—brain barrier,
`in order to avoid possible estrogen
`antagonist effects
`in the central nervous
`system |91d].
`Interestingly,
`this compound has recently been reported to
`selectively stimulate creatinine klnase activity in bone cells but
`not uterine cell
`lines, while tamoxifen and estrogen stimulate
`this activity in both [lO5]. Similar effects have been described
`in vivo, although correlation of these effects with bone density
`and uterine stimulation have not yet been reported [lO6].
`
`Two new TPE’s which contain carboxylic acid functionality
`in place of the amine side chain moiety have also been reported.
`GW5638, 27, has been described as a bone—selective estrogen
`agonist. and has demonstrated decreased uterine stimulation,
`relative to tamoxifen.
`in OVX rats [l07,lO8].
`Interestingly,
`amide analogs of 27 showed an increased tendency toward
`uterine stimulation both in vivo and in vitro [lO7].
`ln OVX rats,
`27 was observed to maintain bone mineral density at both the
`lumbar spine and the proximal tibia with an efficacy similar to
`that of 17B-estradiol or
`tamoxifen at doses of 1-10 mg/kg
`[lO7,lO8]. It has also been shown to reduce serum cholesterol in
`OVX rats with a maximal efficacy of 20-30% [lO8,lO9]. The
`magnitude of this effect, although similar to that observed with
`17B-estradiol, appears to be somewhat muted in comparison to
`the more bioavailable l7ot—ethynyl estradiol and other TPE’s,
`implying perhaps that multiple mechanisms may be involved in
`the regulation of serum lipid concentrations by these compounds
`[1 10]. Hydroxytatnoxifen acid. 28. a tamoxifen metabolite, has
`also been reported to have bone—selective effects in the OVX rat
`[lll],
`
`coat
`
`
`
`o/\ co2H
`
`/
`
`\ \
`
`/
`
`\
`
`HO
`
`owsezs, 27
`
`23
`
`Fig. (4). Acidic triphenylethylenes.
`
`Several groups have recently reported the application of
`parallel synthesis techniques for the preparation of TPE libraries
`[ll2].
`lt
`is expected that
`the ready availability of more
`structurally diverse members of this class,
`together with the
`development of molecular biological assays predictive of in
`viva tissue selectivity, will
`lead to greater understanding of the
`SAR of these compounds in multiple tissues.
`
`AstraZeneea Ex. 2023 p. 6
`
`
`
`Estrogen Receptor Modulators
`
`Benzothiophenes
`
`In order to avoid the problems associated with double bond
`isomerization of the TPE‘s, a variety of cyclic frameworks have
`been investigated for their ER modulating properties. Out of
`these
`structure—activity
`studies,
`raloxifene,
`7,
`a
`benzothiophene—containing compound with a unique profile of
`biological activity emerged [26h,c].
`Raloxifene has been shown to bind the estrogen receptor
`with high affinity and to function as
`a potent estrogen
`antagonist
`in mammary tumor cells and in rat models of
`mammary cancer [ll3,ll4].
`In contrast,
`in the cardiovascular
`
`Current Phannaceutical Design, 1998, Vol. 4, N0. 1 77
`
`system, raloxifene functions primarily as an estrogen agonist.
`ln cell culture, raloxifene has demonstrated estrogen—like effects
`on vascular smooth muscle cells and on the inhibition of LDL
`oxidation [1 l5,l 16]. In the OVX rat model, raloxifene has been
`shown to reduce serum cholesterol by 50-75% after 1-5 weeks of
`daily dosing [1l7,ll8]. Most
`importantly,
`in postmenopausal
`women treated daily with‘ raloxifene, significant reductions in
`total serum cholesterol and LDL cholesterol have been observed
`after both eight Weeks and mo years Oftreatmem [27]'
`Similarly,
`the effects of raloxifene on the skeleton seem to
`parallel
`those observed with estrogen.
`In vitro studies have
`shown similar effects of
`raloxifene and l7[3—estradiol on
`
`Table 2.
`
`ER Binding and Inhibition of MCF-7 Cell Proliferation by 2-Aryl Raloxifene Analogs [127a]
`
` MCF-7 lnhib. [C50 (nM)‘M
`
`estradiol
`
`4—0H—t21rri€
`
`raloxifene, 7
`7a
`
`7b M
`7c
`7d
`7c
`7r
`
`;
`
`‘_
`M
`
`i
`
`M
`
`M
`
`6—OH
`none
`
`M6M-OMe
`M6—OH
`6-OMe
`none
`6—OH
`
`L
`
`7
`
`M
`
`
`
`
`
`
`
`7g
`7h
`7i
`
`M
`
`M
`
`M
`
`M MM6M-CM)MH
`6—Cl
`6—OH
`
`M
`
`4—Cl
`4-OH
`4—Me
`
`0.046
`0006
`0.07
`
`I
`i000
`M50
`
`
`
`M
`
`7;
`
`‘
`
`_M
`
`M
`
`MM
`
`6~lVle
`
`M
`
`MM
`
`.
`
`4—()H
`
`M
`
`Md
`
`300
`
`M
`
`
`
`
`
`
`
`
`
`
`4—OH
`none
`
`4—OMc
`4—OMe
`4—oHM
`4-OH
`none
`
`M
`
`<0.002
`
`<0.002
`0073
`0.008
`0.003
`OM.O62
`
`M
`
`300
`
`M300MM
`i0M0M0MM
`250
`35MMMM
`2.5
`
`
`
`
`
`
`
`
`M
`
`
`
`
`
`
`7k
`71
`7Mrn
`7n
`M 70 M
`
`M
`
`MM
`
`M
`
`M
`M
`
`f
`
`M
`
`MM6M-Ol-l
`MMMMMM3.oii MM
`MM4lMoMHM
`5»OMMH M
`M6-Moii
`
`>
`
`4«F
`4~OH
`4—OH
`4—OH
`2-OH
`
`M
`
`M
`
`0.19
`M 0.02
`<0.M002
`0.10
`0.057
`
`7
`0
`
`M
`
`M
`
`M
`
`.
`
`M
`
`M
`
`2.3
`300MMMM
`i90M
`l0(M)MM
`1M0
`
`
`
`M M
`
`
`
`
`
`3.2
`3
`ND)‘
`
`500
`2MMM
`i
`
`MM
`
`M
`
`M
`M
`
`M
`
`.
`
`7p
`7q
`7r
`
`7s
`7:
`7Mu
`
`M
`
`i
`
`,
`
`M
`
`M
`
`6-OMH
`5—F,6»OH
`5M—Me,6—OH
`
`5,7Adi(Me),6»OH
`6—OHM
`6—OH
`
`‘
`
`3-OHM
`4»OH
`M 4—OH
`
`M
`
`M
`
`4-OH
`2—MeM.M4—MOMlM‘l
`M3—Me,4—OH
`
`MM
`
`M
`
`‘
`
`0
`
`0.16
`0093
`0.07
`
`0.005
`0.4l
`O.l3
`
`
`
`
`
`7v
`M6—OH
`3—Cl,4/OH
`0.12
`2.3MMM
`M 7w
`6»OH
`3-F,4—OH
`0.20
`MM
`03
`
`
`
`6—OH
` 3,5~di(Mc),4—OH
`
`
`(Dose iequired to give 50% inhibition
`hAvt-rage of at least 2 determinations. Values are i |()%.
`“RBA 2 relative binding affinity by competition with 3H—i7[3-€SlI'E1dlOi.
`of a riizixinially effective (|0'H) dose of 17B-estradiol. Average of atleasi 3 determinations. Values are i lO% ‘INA : not active at
`the doses tested,
`“4—
`Hydroxyiamoxifen, the primary biologically active metabolite of tamoxifen.
`-1 ND : not determined
`
`AstraZeneca Ex. 2023 p. 7
`
`
`
`78 Current Pharmaceutical Design, 1998, Vol. 4, No. 1
`
`Grese rind Dlldgr
`
`osteoclastogenesis and on creatinine kinase activity in human
`osteoblast cells
`[ll9,l20].
`In rats,
`several
`studies have
`demonstrated a protective effect against ovariectomy—induced
`osteopenia at doses as
`low as 0.1
`rng/kg [1l7,l2l,l22].
`Positive effects on bone mineral density at both cortical and
`cancellous bone sites have been reported. as have positive
`effects on bone strength [l21,l23].
`Interestingly, although
`raloxifene suppresses bone resorption in the rat with efficacy
`which is approximately equal
`to that of estrogen, bone
`formation appears to be suppressed to a lesser degree, resulting
`in a net gain in bone mass with raloxifene [121,l23b]. These
`positive results in animal studies, have now been confirmed with
`clinical studies in postmenopausal women [27]. Significant
`positive effects on histomorphometric parameters, bone
`markers, and on bone mineral density at both the lumbar spine
`and hip were observed after two years of raloxifene treatment
`[27,l24].
`
`in terms of its pharmacology, raloxifene is distinguished
`from the TPE‘s primarily on the basis of its effects on the uterus,
`where a qualitative difference has been observed [61,l25].
`In a
`direct comparison with tamoxifen, droloxifene, and idoxifene.
`raloxifene was a significantly more effective antagonist of
`estrogen action in the immature female rat uterus [61a]. In this
`assay the TPE’s functioned as partial agonists,
`inhibiting the
`effects of estrogen on uterine weight gain only to the level of
`their own intrinsic agonist activity,
`\\/hile raloxifene functioned
`essentially as a complete antagonist. Similarly.
`in OVX rats, the
`TPE‘s have been found to induce a larger maximal stimulation of
`uterine weight and to induce uterine eosinophilia while
`raloxifene did not
`[61]. Although raloxifene has also been
`reported to stimulate a modest
`increase in uterine wet weight,
`this increase is not dose related and is not coincident with
`
`increases in other measures of uterine hypertrophy, and has
`therefore been attributed to water
`retention [1 17].
`In
`postmenopausal women, raloxifene has been reported to show
`no stimulatory effects on the uterus, even after 2 years of
`treatment [27,l26].
`
`SAR studies in the raloxifene series have centered on
`
`modifications of the 2-arylbenzothiophene [127] the amine-
`containing side chain [1 13,128], and the carbonyl hinge [129]
`The 2~arylbenz.othiophene unit appears to be the primary site of
`ER binding, mimicking the interactions of 17B-estradiol with
`the receptor. On the basis of binding and in vitro activation data
`(Table 2) the 6-hydroxy of raloxifene is believed to imitate the
`3—hydr