The new england journal of medicine

review article

drug therapy
Alastair J.J. Wood, M.D., Editor

Selective Estrogen-Receptor Modulators —

Mechanisms of Action and Application
to Clinical Practice
B. Lawrence Riggs, M.D., and Lynn C. Hartmann, M.D.

he selective estrogen-receptor modulators (serms) represent

From the Division of Endocrinology and
Metabolism, Department of Internal Medi-
cine (B.L.R.), and the Department of Oncol-
ogy (L.C.H.), Mayo Clinic and Mayo Foun-
dation, Rochester, Minn. Address reprint
requests to Dr. Riggs at the Mayo Clinic,
t a major therapeutic advance for clinical practice. Unlike estrogens, which are
uniformly agonists, and antiestrogens, which are uniformly antagonists, the
SERMs exert selective agonist or antagonist effects on various estrogen target tissues.
The SERMs are chemically diverse compounds that lack the steroid structure of estro-
200 First St. SW, N. 6 Plummer, Rochester,
MN 55905. gens (Fig. 1) but possess a tertiary structure that allows them to bind to the estrogen re-
ceptor. Although some members of this class of drugs have been available for decades,
N Engl J Med 2003;348:618-29. their tissue-specificity in humans has only recently been recognized. Certain phytoestro-
Copyright © 2003 Massachusetts Medical Society.
gens, such as genistein, also appear to have SERM-like properties.
In this article we review the emerging understanding of the molecular basis of action
of SERMs, summarize their tissue-selective agonist–antagonist effects, and place in per-
spective their therapeutic uses as compared with estrogen or nonestrogen alternatives.

mechanisms of action
In the classic model of estrogen action, the unoccupied nuclear estrogen receptor re-
sides in the nuclei of target cells in an inactive form. Binding to an agonist, such as
estradiol, alters the physicochemical properties of the estrogen receptor, allowing the
receptor dimer to interact with specific DNA sequences (estrogen response elements)
within the promoters of responsive genes.1 The DNA-bound estrogen receptor then reg-
ulates target-gene transcription, either positively or negatively (Fig. 2). However, the
recognition that tamoxifen and other SERMs have tissue-specific agonist–antagonist
activity led to the realization that the classic model was incomplete and that estrogen
action was more complex than had been thought.2,3 The mechanisms of the tissue-
selective, mixed agonist–antagonist action of SERMs, although still only partly under-
stood, are gradually becoming clearer.
Most of the unique pharmacology of SERMs can be explained by three interactive
mechanisms: differential estrogen-receptor expression in a given target tissue, differ-
ential estrogen-receptor conformation on ligand binding, and differential expression
and binding to the estrogen receptor of coregulator proteins (Fig. 2).
First, target cells for estrogen action contain varying concentrations of homodimers
of one or both of two species of estrogen receptors — estrogen receptor a and estrogen
receptor b — as well as estrogen receptor a–estrogen receptor b heterodimers. Mice
with genetic disruptions of estrogen receptor a and estrogen receptor b display differ-
ent phenotypes, demonstrating that each receptor has a distinct action.4,5 Estrogen
receptor a is almost always an activator, whereas estrogen receptor b can inhibit the
action of estrogen receptor a by forming a heterodimer with it. Moreover, microarray
analysis in mice with deletions of estrogen receptor a or estrogen receptor b showed

618 n engl j med 348;7 www.nejm.org february 13, 2003

 drug therapy

N
O

OH
CH3

HO

17b-Estradiol Tamoxifen

N N

O O

OH
S
HO HO

Cl

Toremifene Raloxifene

Figure 1. Chemical Structure of 17b-Estradiol, the Main Physiologically Relevant Estrogen, and the Three SERMs Ap-
proved by the Food and Drug Administration.
17b-Estradiol has a cyclophenanthrene structure, whereas tamoxifen and toremifene have a triphenylethylene structure
and raloxifene has a benzothiophene structure. Although the primary structure of these SERMs differs strikingly from
that of estrogens, they have a conformation that allows them to bind to the ligand-binding domain of the estrogen receptor.

that estrogen receptor b inhibited the transcription
of 240 estrogen-responsive genes by 46 percent.6
Thus, the relative levels of expression of these two
receptor isoforms will affect the cellular respon-
siveness to estrogens.7,8 Since two SERMs, ralox-
ifene and tamoxifen, also bind to both isoforms,
these drugs will affect the cellular responsiveness
as well. Indeed, these SERMs function as pure an-
tagonists when acting through estrogen receptor
b on genes containing estrogen response ele-
ments but can function as partial agonists when
acting on them through estrogen receptor a.8
Second, protein crystallography and techniques
that evaluate surface changes have shown that bind-
ing by estradiol, tamoxifen, raloxifene, or the pure
estrogen antagonist ICI 164,384 results in a unique
estrogen-receptor conformation for each ligand.9-12
Thus, the ligand binding results in various estrogen-
receptor conformations that range from that as-
sumed when the receptor is bound to estrogens at
one extreme to that assumed when the receptor is
bound to antiestrogens at the other extreme. SERM-
bound estrogen receptors assume a continuum of
intermediate shapes.
Third, more than 20 coregulator proteins have
been discovered that bind to estrogen receptors and
modulate their function, each acting as either a pos-
itive or a negative transcriptional regulator (a coac-
tivator or a corepressor, respectively).13-16 Depend-
ing on the unique receptor conformation induced
by ligand binding, varying combinations of coreg-
ulator proteins interact with the estrogen receptor
and modulate its function in a variety of ways. The
relative and absolute levels of expression of coreg-

n engl j med 348;7 www.nejm.org february 13, 2003 619

 The new england journal of medicine

Corepressor
protein

X
Antagonist
ERE

Estrogen SERMs
receptor

ERE
Unidentified
Agonist coactivator

Coactivator
ERE

Figure 2. Estrogen-Receptor Action.

On binding an agonist or an antagonist, the estrogen receptor (the a or b isoform) undergoes a conformational change
that permits its spontaneous dimerization and facilitates the subsequent interaction of the dimer with estrogen re-
sponse elements (EREs) located within target genes. It has been determined that estrogen facilitates the interaction of
the estrogen receptor with coactivators. An antagonist-activated estrogen receptor, on the other hand, interacts prefer-
entially with a corepressor protein. The binding of different SERMs to the receptor permits it to adopt conformational
states distinct from that induced by classic agonists or antagonists. The weight of available evidence suggests that the
structure of some SERM–estrogen-receptor complexes favors corepressor recruitment and that of others favors some
affinity for known coactivators. Some SERMs may also facilitate the interaction of the estrogen receptor with yet-to-be-
identified coactivators with which it would not normally couple. The implication of this model is that SERM activity will
be influenced by the relative levels of expression of the cofactors (corepressors and coactivators) in target cells.

ulator proteins vary among estrogen target cells.13
In an important recent study, Shang and Brown17
found that tamoxifen and raloxifene, which are an-
tiestrogens for the breast, act on mammary cells by
recruiting corepressors to estrogen-receptor tar-
get promoters. In contrast, tamoxifen, which is
an estrogen agonist for the endometrium, acts in
endometrial cells by recruiting coactivators to the
estrogen-receptor target promoters, whereas this
recruitment does not occur with raloxifene, which
has a neutral effect on the endometrium. The inves-
tigators also found that the tamoxifen–estrogen
receptor a complex activates transcription by teth-
ering to promoters that do not contain estrogen
response elements through protein-to-protein con-
tacts with other DNA-bound transcription factors.
Moreover, the agonist effect of tamoxifen was de-
pendent on a higher concentration of a key coacti-
vator, steroid receptor coactivator-1 (SRC-1), in the
endometrial cells. Thus, variable local concentra-
tions of different coregulator proteins may contrib-
ute to the tissue-selective pharmacology of SERMs.
A model of the molecular action of estrogens and
SERMs is shown in Figure 2.
action on target tissues
In recent years estrogen has been the most frequent-
ly prescribed drug in the world. It was reported in
1999 to be used by 38 percent of postmenopausal
American women.18 However, estrogen use has de-
clined sharply in the wake of the early termination

620 n engl j med 348;7 www.nejm.org february 13 , 2003

 drug therapy
last summer of the Women’s Health Initiative after
the data and safety monitoring board decided that
risks exceeded benefits.19 The Women’s Health Ini-
tiative was a prospective trial, sponsored by the
National Institutes of Health, involving 16,608 post-
menopausal women randomly assigned to treat-
ment daily with 0.625 mg of conjugated estrogen
and 2.5 mg of medroxyprogesterone acetate, a com-
monly used regimen of hormone-replacement ther-
apy, or with placebo. The trial was originally sched-
uled to last 8.5 years but was stopped after 5.2 years,
although the subgroup of women who had under-
gone hysterectomy and were taking estrogen with-
out medroxyprogesterone was allowed to contin-
ue. Although other estrogen regimens or dosages
would not necessarily have provided similar re-
sults, the trial results bring into sharper focus the
risks of hormone-replacement therapy in its asso-
ciation with an increase in cardiovascular disease,
stroke, pulmonary embolism, and breast cancer.19
But the Women’s Health Initiative also provides the
first controlled data demonstrating that hormone-
replacement therapy protects against osteoporotic
fractures and may decrease the occurrence of co-
lorectal cancer.19
One consequence of the Women’s Health Initia-
tive findings has been an increased interest in thera-
py with SERMs, because of their potential to retain
most of the beneficial effects of estrogen while
avoiding most of its adverse effects. In the sections
that follow, we review the corresponding and con-
trasting actions of estrogen and SERMs on various
target tissues (Table 1).
bone
Estrogen acts on bone cells, which contain both
isoforms of estrogen receptors.21,22 However, con-
centrations of estrogen receptor b are higher in de-
veloping cancellous bone, whereas concentrations
of estrogen receptor a are higher in developing cor-
tical bone.23 Estrogen deficiency is the main cause
of postmenopausal osteoporosis.21 When estrogen
is deficient, bone turnover increases, and bone re-
sorption increases more than bone formation, lead-
ing to bone loss. Hormone-replacement therapy re-
verses these changes in women in both the early and
the late phases of the postmenopausal period.21 Al-
though it had been demonstrated by many observa-
tional studies, the Women’s Health Initiative was
the first large, randomized clinical trial to show that
hormone-replacement therapy reduces osteoporot-
ic fractures, including a 34 percent reduction in both
n engl j med 348;7 www.nejm.org
Table 1. Comparison of Selected Actions and Side Effects of Estrogen
and Clinically Available SERMs.*
Side Effect Estrogen Tamoxifen Toremifene Raloxifene
Hot flashes ↓↓↓ ↑† ↑† ↑†
Uterine bleeding ↑↑↑ ↑ ↑ ←
→
Risk of endometrial cancer ↑↑‡ ↑ ? ←
→
Prevention of postmenopausal
bone loss ↑↑↑ ↑ ←
→ ↑↑
Risk of breast cancer ↑↑ ↓↓ ↓↓§ ↓↓
Favorable pattern of serum lipids ↑↑↑¶ ↑ ↑↑ ↑
Venous thrombosis ↑↑ ↑↑ ? ↑↑
* Ascending arrows indicate that the drug increases the effect, and descending ar-
rows that it decreases the effect. Horizontal arrows indicate no change. The num-
ber of arrows indicates the size of the change.
† In perimenopausal women the action would be ↑↑.
‡ This effect can be prevented by concurrent treatment with a progestin.
§ The only available data are for inhibition of breast-cancer growth.
¶ This effect may be attenuated by concurrent treatment with androgen-derived
progestins.20
vertebral and hip fractures.19 This reduction oc-
curred even though the study subjects were at low
risk for fractures.
Although early work established that tamoxi-
fen was an estrogen antagonist for the breast, sub-
sequent studies in animals24,25 and clinical stud-
ies26,27 showed that it was a weak agonist for bone.
In postmenopausal women, however, the gain in
bone density after two years of tamoxifen therapy
was small (Table 2), and half of this short-term in-
crease was lost after five years of continued treat-
ment.34 Tamoxifen has been reported both to
increase35 and to decrease36 the risk of hip frac-
ture. From the limited available data, toremifene ap-
pears to be a weaker bone agonist than tamoxifen.37
The effects of raloxifene on bone are well estab-
lished. In postmenopausal women with osteoporo-
sis, treatment with raloxifene decreased markers of
bone turnover by 30 to 40 percent after one year and
increased bone density at several scanning sites by
2 to 3 percent after three years30-33 (Table 2). Ral-
oxifene also decreased the incidence of vertebral
fractures by 30 to 50 percent, depending on dosage,
but did not decrease the incidence of hip fracture or
other nonvertebral fractures.32 Although bispho-
sphonate (alendronate or risedronate) therapy does
decrease nonvertebral fractures, the decrease in ver-
tebral fractures of approximately 50 percent is only
slightly greater than with raloxifene therapy,38,39
february 13, 2003 621
 The new england journal of medicine

Table 2. Results of Major Randomized Clinical Trials of SERMs with Regard to Bone Mineral Density.

Change in Bone Mineral Density

Trial Study Subjects Duration as Compared with Placebo Group

Total-Body Lumbar Proximal

Bone Mineral Spine Femur
months percent
Tamoxifen (20–30 mg/day)
Love et al.27 140 postmenopausal women with breast cancer 24 — 1.6* —
Grey et al.28 57 normal late-postmenopausal women 24 0.5† 2.1† 0.6
Powles et al.29 179 healthy women in chemoprevention trial for
breast cancer
54 postmenopausal women 36 — 4.7* 3.6†
125 premenopausal women 36 — ¡2.6* ¡4.3†
Raloxifene (60 mg/day)
Delmas et al.30 302 normal postmenopausal women 24 2.0* 2.4* 2.4*
Lufkin et al.31 143 postmenopausal women with osteoporosis 12 ¡0.1 1.8† 1.0†
and vertebral fractures
Ettinger et al.32 5140 postmenopausal women with osteoporosis 36 — 2.6* 2.1*
Johnston et al.33 576 healthy early-postmenopausal women 36 1.7* 2.6* 2.5*

* P<0.05 for the comparison with placebo.

† P<0.005 for the comparison with placebo.

despite the fact that bone density increases to a
much greater extent (4 to 9 percent more at the same
bone-density scanning sites).40,41 Although the ex-
planation for this paradox is unclear, Riggs and Mel-
ton42 have suggested that much of the antifracture
effect of antiresorptive drugs such as raloxifene at
cancellous bone sites results from normalization of
the high level of bone turnover and thus the preven-
tion of further microarchitectural disruption. They
have also suggested that it is necessary to reach only
a low therapeutic threshold to prevent osteoclasts
from perforating the trabecular plates and thus to
reduce fractures at sites of cancellous bone, such as
the vertebrae. This low threshold can be achieved
with less potent antiresorptive drugs, such as ralox-
ifene. However, to reduce fractures at sites of corti-
cal bone, such as the hip, it is necessary to increase
bone density more substantially by using potent an-
tiresorptive or formation-stimulating agents. Ta-
ble 2 summarizes the effect of two SERMs on bone
mineral density according to the results of major
randomized, controlled trials.
breast
Estrogen stimulates the proliferation of breast epi-
thelial cells, and both endogenous and exogenous
estrogens have been implicated in the pathogenesis
of breast cancer.43 The Women’s Health Initiative
showed that hormone-replacement therapy was as-
sociated with a 27 percent relative increase in inva-
sive breast cancer (38 vs. 30 cases per 10,000 pa-
tient years),19 a statistic similar to the 35 percent
increase found in a meta-analysis of 51 observa-
tional studies.44
Tamoxifen has demonstrated efficacy for the
treatment and prevention of estrogen-receptor–pos-
itive breast cancer. Adjuvant tamoxifen therapy (ad-
ministered after the initial surgery) significantly re-
duced the risk of recurrence and death from breast
cancer in all age groups studied.45 In an overview of
37,000 women with breast cancer from 55 trials of
adjuvant therapy,45 the proportional reduction in re-
currence was 47 percent after 5 years of treatment
with tamoxifen and the proportional reduction in
mortality was 26 percent after 10 years. The absolute
improvements in 10-year survival were 10.9 percent
in node-positive and 5.6 percent in node-negative
breast cancer. Women with estrogen-receptor–neg-
ative disease had little, if any, benefit.45,46
Best results appear to be achieved after five years
of treatment; thereafter, beneficial effects decrease
and toxicity increases,47-49 although the optimal du-
ration of administration is still under investigation.
Tamoxifen is mainly cytostatic and slows the prolif-
eration of breast-cancer cells by inhibiting their pro-
gression from the G1 phase of the cell cycle,48 but it

622 n engl j med 348;7 www.nejm.org february 13 , 2003

 drug therapy

also induces apoptosis in vitro and thus may pos-
sess cytocidal properties in vivo.50 About half of
women with advanced estrogen-receptor–positive
breast cancer will have a response to tamoxifen
therapy, whereas only 5 percent of those whose
cancer is estrogen-receptor–negative will have a
response.48,51
The reduction in the risk of contralateral breast
cancer in adjuvant trials of tamoxifen led to its in-
clusion in randomized primary-prevention trials.
Among the 13,388 participants in the Breast Can-
cer Prevention Trial, there was a 49 percent reduc-
tion in the risk of invasive breast cancer, but benefit
was limited to estrogen-receptor–positive tumors.36
Two smaller primary-prevention trials from the Roy-
al Marsden Hospital in London52 and from Italy53
failed to demonstrate that tamoxifen reduced the
risk of breast cancer (Table 3). The recently report-
ed International Breast Cancer Intervention Study I
(IBIS-I) trial showed a 25 percent reduction in inva-
sive breast cancer with tamoxifen.54 However, no
trial has shown improvement in survival with ta-
moxifen; in fact, there were slightly more deaths in
the tamoxifen group in the IBIS-I trial owing to an
excess of thromboembolic events.54 The recent
technology assessment by the American Society of
Clinical Oncology concluded that tamoxifen’s fa-
vorable effect on the risk of breast cancer must be
weighed against its potential side effects in individ-
ual women.55
Toremifene is approved by the Food and Drug
Administration for the treatment of advanced breast

Table 3. Results of Major Trials of SERMs for the Primary Prevention of Breast Cancer.*

Royal Marsden
Variable NSABP P-1 Study36 Hospital Trial52 Italian Trial53 IBIS-I54

Primary outcome Breast cancer Breast cancer Breast cancer Breast cancer

Secondary outcome Bone, cardiovascular — Cardiovascular, Thromboembolic, car-

psychometrics diovascular, en-
dometrial cancer
Eligibility Age ≥60 yr, or 35–59 yr Age 30–70 yr with a Age 35–70 yr after Age 35–70 yr with a
with a 5-yr predicted family history hysterectomy family history, lob-
risk of ≥1.66%, ular carcinoma in
lobular carcinoma situ or atypia
in situ
No. of women 13,388 2471 5408 7152
Study drugs Tamoxifen vs. placebo Tamoxifen vs. placebo Tamoxifen vs. placebo Tamoxifen vs. placebo
Age distribution (%)
<50 yr 39 61 38 —†
50–60 yr 31 39 50
>60 yr 30 12
Family history of breast cancer 77 96 18 97
(%)
Mean follow-up (mo) 55 70 46 50

Effect on invasive breast cancer

No. taking placebo in 175 64 Not reported 85
whom breast cancer 89 54 — 64
developed ¡49‡ —§ —§ ¡25¶
No. taking tamoxifen in
whom breast cancer
developed
Relative difference (%)

* IBIS-I denotes International Breast Cancer Intervention Study I, and NSABP National Surgical Adjuvant Breast and Bowel Project.
† The median age in the trial was 51 years.
‡ P<0.001.
§ The difference was not significant.
¶ The confidence interval was 54 to 104.

n engl j med 348;7 www.nejm.org february 13, 2003 623

 The new england journal
cancer. However, in head-to-head randomized tri-
als, tamoxifen and toremifene were found to have
very similar efficacy and side-effects profiles56,57
and exhibited cross-resistance with each other.58
In the Multiple Outcomes of Raloxifene Evalua-
tion (MORE) trial, which was designed to assess
antifracture efficacy, raloxifene reduced the risk of
breast cancer by 76 percent.59 Because the MORE
trial was conducted in older women with osteoporo-
sis who were not at increased risk for breast cancer,
these favorable results cannot necessarily be extrap-
olated to younger women who are at high risk for
breast cancer. This issue should be resolved by the
ongoing Study of Tamoxifen and Raloxifene (STAR)
trial, which is a head-to-head comparison of the ef-
ficacy of raloxifene and tamoxifen in women at in-
creased risk for breast cancer60 (Table 4). A phase 1
trial with arzoxifene (LY353381.HCl), a new third-
generation SERM, demonstrated stable disease in
6 of 32 tamoxifen-resistant patients with metastatic
Table 4. Major Large-Scale Trials Involving SERMs.
ATLAS (Adjuvant Tamoxifen Longer against Shorter) trial61*
Assessment of optimal duration of tamoxifen adjuvant therapy
Accrual goal is 20,000 pre- and postmenopausal patients with breast cancer
who are receiving adjuvant tamoxifen
Therapy: tamoxifen for 5 yr vs. 10 yr (or longer)
ATTOM (Adjuvant Tamoxifen Treatment Offers More) trial62*
Assessment of optimal duration of tamoxifen adjuvant therapy
Accrual goal is 8000–20,000 pre- and postmenopausal patients with breast
cancer who are receiving adjuvant tamoxifen
Therapy: tamoxifen for 2 yr (group 1) vs. 7 yr (group 2)
IBIS2 (International Breast Cancer Intervention Study2)63*
Primary prevention of breast cancer
Accrual goal is 16,000 women at high risk for breast cancer (age, 35–70 yr)
Therapy: anastrozole vs. placebo
STAR (Study of Tamoxifen and Raloxifene)60
Primary prevention of breast cancer
Accrual goal is 22,000 postmenopausal women at high risk for breast cancer
Therapy: 20 mg tamoxifen per day vs. 60 mg of raloxifene per day for 5 yr
CORE (Continuing Outcomes Relevant to Evista)
4000 postmenopausal women who were previous participants in the MORE
(Multiple Outcomes of Raloxifene Evaluation) trial
Receiving raloxifene or placebo for an additional 4 yr, completion by 2003
Primary end point is breast-cancer prevention; secondary end points are non-
vertebral fractures and uterine safety
RUTH (Raloxifene Use in the Heart)64†
Effect of raloxifene vs. that of placebo in prevention of coronary events and
death from coronary causes
10,000 postmenopausal women at risk for coronary disease
Duration of 7.5 yr, completion by 2005
* The trial is not open in the United States.
† The trial has been completed, and the data are under analysis.
624 n engl j med 348;7 www.nejm.org
of medicine
breast cancer,65 and phase 2 studies are under way.
Several large trials addressing unresolved questions
regarding the role of SERMs in the prevention and
treatment of breast cancer are ongoing (Table 4).
genitourinary tract
Estrogen deficiency induces atrophic changes in the
genitourinary tract and symptoms of dyspareunia.
In postmenopausal women with an intact uterus,
administration of unopposed estrogen increases
the incidence of endometrial carcinoma.66 Com-
bination treatment with progestin prevents this
increase,66 but the progestin component of hor-
mone-replacement therapy is the main cause of the
common treatment-related problems of edema and
premenstrual syndrome–like symptoms. Yet anoth-
er major problem with hormone-replacement ther-
apy is breakthrough and withdrawal vaginal bleed-
ing. The use of raloxifene obviates both problems,
because it does not stimulate the endometrium and
does not require progestin treatment. Over a period
of three years, endometrial thickness as assessed by
ultrasonography and vaginal bleeding were similar
in raloxifene-treated subjects and those given pla-
cebo.59 Moreover, hormone-replacement therapy,
but not raloxifene or tamoxifen therapy, significant-
ly increases urinary incontinence.67 However, like
estrogen, but unlike raloxifene, tamoxifen is asso-
ciated with a 2.5-fold increase in endometrial carci-
noma,36 so vaginal bleeding in women receiving ta-
moxifen should be promptly investigated.
cardiovascular system
Coronary artery disease accounts for one third of all
deaths in postmenopausal women.68 Both estrogen
and SERM therapy induce a beneficial serum lipid
profile, although the patterns of these changes dif-
fer (Table 5). The main serum lipid changes with
oral estrogen are increases in high-density lipopro-
tein (HDL) cholesterol and triglycerides and de-
creases in low-density lipoprotein (LDL) choles-
terol. Tamoxifen, toremifene, and raloxifene also
decrease LDL cholesterol but, unlike estrogen, do
not increase triglycerides. Toremifene, unlike oth-
er SERMs, increases HDL cholesterol.71 Treatment
with estrogen or SERMs changes the blood-coagu-
lation indexes in the direction of enhanced clotting,
and estrogen, but not raloxifene, increases indexes
of inflammation (Table 5). Also, estrogen, but not
raloxifene, retarded experimentally induced athero-
sclerosis in a monkey model,74 whereas raloxifene
was 75 percent as effective as estrogen in a rabbit
february 13 , 2003
 drug therapy

model.75 Moreover, in ovariectomized rats, ralox-

ifene was as effective as estrogen in enhancing nitric Table 5. Comparative Effects of Oral Hormone-Replacement Therapy and
SERMs on Serum Lipids, Indexes of Inflammation, and Blood Coagulation.*
oxide–induced coronary-artery dilatation76 and in
retarding injury-induced intimal thickening of the Hormone-
carotid artery.77 Replacement
Variable Therapy Tamoxifene Toremifene Raloxifene
Observational studies have found that women
receiving hormone-replacement therapy have a 30 percent difference from change with placebo
to 35 percent lower risk of coronary disease.78 Thus, Low-density lipoprotein ¡12† ¡19† ¡21† ¡12†
the reports of two randomized clinical trials of hor- cholesterol
mone-replacement therapy that failed to show ben- High-density lipoprotein 7† ¡2 14† 0
efit in postmenopausal women with established cholesterol
coronary artery disease79,80 were unexpected, al- Triglycerides 18† 31† ¡4 ¡4
though another trial demonstrated that estrogen
Apolipoprotein A-I 13† 5 13† 3
therapy retarded the development of subclinical
atherosclerosis.81 However, the results of the Wom- Apolipoprotein B ¡4 ¡9† ¡10† ¡9†
en’s Health Initiative in postmenopausal women, Lp(a) lipoprotein ¡19† ¡14† ¡53† ¡7†
the large majority of whom were not at high risk for Fibrinogen ¡1 — — ¡10†
coronary heart disease, are of particular concern:
Plasminogen activator ¡19† — — 8
participants had a 23 percent increase in cardiovas- inhibitor type 1
cular disease (37 vs. 30 cases per 10,000 person-
Homocysteine ¡7† — — ¡8†
years) and a 38 percent increase in strokes (29 vs.
21 cases per 10,000 person-years).19 Thus, there is C-reactive protein 84† — — ¡7
no current justification for the use of hormone- Tumor necrosis factor a ¡11† — — ¡5†
replacement therapy to prevent or treat cardiovas- Interleukin-6 11 — — 1
cular disease.
How to reconcile these findings with previous * Data are from the Postmenopausal Estrogen/Progestin Interventions Trial Writing
data showing benefit is unclear. One possible in- Group,20 Love et al.,69 Walsh et al.,70 Saarto et al.,71 Walsh et al.,72 and Cox et al.73
terpretation is that cardiovascular outcomes asso- † P<0.05 for the comparison with placebo.

ciated with hormone-replacement therapy represent

the algebraic summation of its prothrombotic or
proinflammatory effects and its antiatherosclerotic
effects and that the balance between them will be
determined by such variables as oral as compared
with transdermal administration, the types and dos-
ages of estrogens and progestins, the age at onset of
therapy, and other factors. Consistent with this
speculation is the recent report that raloxifene de-
creased cardiovascular events by 40 percent in the
1035 women in the MORE trial who had cardiovas-
cular risk factors at base line.82 A prospective study
of 27,939 healthy U.S. women found that the serum
level of C-reactive protein was a stronger predictor
of cardiovascular events than the LDL cholesterol
level.83 Hormone-replacement therapy increases
C-reactive protein levels, whereas raloxifene does
not (Table 5). The large, yet to be reported, random-
ized trial of Raloxifene Use in the Heart (RUTH)64 is
specifically powered to determine whether ralox-
ifene is cardioprotective for postmenopausal wom-
en at risk for coronary artery disease (Table 4).
Estrogen,19,78,84 tamoxifen,36 and raloxifene32
therapy are associated with increases by a factor
of 1.5 to 3 in venous thromboembolic disease, al-
though the absolute risk is small.84
central nervous system
Estrogen has diverse effects on brain function.85 Es-
trogen receptor a is found mainly in the hypothala-
mus. Estrogen receptor b is more widely distributed
throughout the brain4,22 and is concentrated at loci
involved in cognition and memory. Many studies
have shown that estrogen treatment enhances both
of these functions.85 However, data from objective
testing carried out during the Heart and Estrogen/
Progestin Replacement Study (HERS)86 failed to
show that hormone-replacement therapy increased
energy levels or improved mental health and depres-
sive symptoms any more than placebo. A recent
3-year prospective observational study in elderly
women in Cache County, Utah, found that estrogen
use for 10 or more years reduced the risk of Alzhei-
mer’s disease by 67 percent.87 However, this protec-
tion has not as yet been demonstrated in random-
ized clinical trials.88 No studies of the effect of

n engl j med 348;7 www.nejm.org february 13, 2003 625

 The new england journal
SERMs in reducing the risk of Alzheimer’s disease
have been undertaken.
Although estrogen is highly effective in reducing
hot flashes, all SERMs studied thus far act as anti-
estrogens for the hypothalamic centers regulating
gonadotrophin secretion. During the early post-
menopausal period, tamoxifen increased the inci-
dence of hot flashes by 17 percent as compared with
placebo,37 and raloxifene increased them by 7 per-
cent.33 This is less of a problem in women who
are in the late postmenopausal period; hot flashes
caused only 0.7 percent of the patients to discon-
tinue therapy.32 In studies in ovariectomized rats,
raloxifene was able to reproduce the ability of es-
trogen to restore the reduction in choline acetyl-
transferase activity in the hippocampus but not in
the hypothalamus.89 It is important to note, howev-
er, that raloxifene treatment does not impair cogni-
tion in postmenopausal women.90
therapeutic uses
general
Because of their selective estrogen-agonist proper-
ties, SERMs can be used to prevent or treat diseases
caused by estrogen deficiency, such as osteoporosis,
without most of the undesirable actions of estro-
gen. Conversely, because of their selective estrogen-
antagonist properties, they can be used to prevent or
treat diseases, such as breast cancer, in which estro-
gen-agonist activity is undesirable for a given tissue.
Currently available SERMs have two major limita-
tions: they are only weak estrogen agonists and they
aggravate hot flashes, the most common indication
for estrogen therapy. Because tamoxifen stimulates
the endometrium, raloxifene is the preferred SERM
for uses other than the prevention or treatment of
breast cancer. The Food and Drug Administration
has approved three SERMs for clinical use: tamoxi-
fen for the prevention and treatment of breast can-
cer, toremifene for the treatment of advanced breast
cancer, and raloxifene for the prevention and treat-
ment of postmenopausal osteoporosis.
prevention and treatment of osteoporosis
Raloxifene at a dose of 60 mg per day is effective in
preventing and treating postmenopausal osteoporo-
sis. It has a very low incidence of side effects, may
protect against the development of breast cancer,
and retains at least some of the putative extraskele-
tal benefits of estrogen. As with other medications
used as antiresorptive therapy, combined treatment
with 1000 mg or more of elemental calcium and 400
626 n engl j med 348;7 www.nejm.org
of medicine
units of vitamin D daily is advisable. Tamoxifen and
toremifene are not indicated for the treatment of
osteoporosis.
There are now several effective treatment options
for osteoporosis, including the bisphosphonates,
nasal-spray calcitonin, and raloxifene.91 Estrogen is
also effective, although its risk–benefit ratio must
now be reevaluated.19 Treatment should be tailored
to the individual patient. Because raloxifene, like
nasal-spray calcitonin, is a less potent antiresorp-
tive drug than alendronate, risedronate, or estro-
gen,19,37,38,41,91 it is best suited for the prevention
of bone loss or the treatment of mild osteoporosis.
Because clinical trials have demonstrated that ral-
oxifene decreases the risk of vertebral fracture but
have failed to demonstrate that it decreases the risk
of extravertebral fracture, it should be used mainly in
postmenopausal women with predominantly spinal
osteoporosis.
treatment and prevention of breast cancer
The treatment of breast cancer is complex and in-
volves surgery, radiation, chemotherapy, and hor-
monal therapy, used singly or in combination, de-
pending on the stage and estrogen-receptor status
of the disease in the individual patient.53 Tamoxi-
fen remains a mainstay of the hormonal treatment
of all phases of estrogen-receptor–positive breast
cancer. The aromatase inhibitors, which block es-
trogen synthesis, are an alternative to tamoxifen for
hormonal therapy and, unlike tamoxifen, are not as-
sociated with an increased risk of thromboembolic
complications or endometrial cancer.92-95 Howev-
er, in postmenopausal women, aromatase inhibi-
tors elevate the rate of bone turnover96 and increase
the risk of fracture.95 In postmenopausal women
with advanced estrogen-responsive breast cancer,
tamoxifen and aromatase inhibitors have similar
antitumor efficacy.92-94
For adjuvant therapy in postmenopausal women,
however, the recently published results of the Arim-
idex, Tamoxifen Alone or in Combination (ATAC)
trial found that the aromatase inhibitor anastrozole
alone was equivalent or superior to both tamoxifen
and a combination of anastrozole and tamoxifen
with regard to various response indexes when stud-
ied over a period of 33 months.95 Nonetheless,
pending further studies, the American Society of
Clinical Oncology Technology Assessment state-
ment continues to recommend tamoxifen as the
preferred hormonal adjuvant therapy for estrogen-
receptor–positive breast cancer.97 Also, because
there is insufficient information on safety and effi-
february 13, 2003
 drug therapy

cacy, aromatase inhibitors are not recommended
for premenopausal women with breast cancer.
Tamoxifen is the only approved hormonal agent
for primary prevention of breast cancer. Debate con-
tinues about who should receive it for this indica-
tion, although reasonable candidates include wom-
en with a strong family history of breast cancer and
those with atypical hyperplasia or lobular carcinoma
in situ. Women with mutant BRCA1 or BRCA2 genes
are also potential candidates, especially since pro-
phylactic ovariectomy has recently been shown to
reduce their risk of breast cancer by about 50 per-
cent.98,99 However, because 80 percent of the breast
cancers that develop in carriers of BRCA1 mutations
are estrogen-receptor–negative, the efficacy of ta-
moxifen in this group has been questioned.100
Although the substantial reduction in the risk of
breast cancer in postmenopausal women reported
from the MORE trial was encouraging, the use of
raloxifene in primary prevention should await the
results of the STAR trial.60 A cautionary note on the
use of tamoxifen for prevention in premenopausal
women is that it acts as a bone antagonist that re-
sults in bone loss in this age group29 (Table 2). The
presumed mechanism is competition with the more
potent bone agonist 17b-estradiol for the estrogen
receptor.
future directions
As the molecular mechanisms of the action of
SERMs become more completely understood, ra-
tional drug design will replace the current empiri-
cal method for the discovery of new SERMs. It is
probable that the ultimate goal of SERM research
will be achieved — the discovery of a tissue-selective
drug that has all the beneficial effects of estrogen,
has none of its adverse effects, and offers protection
against breast cancer. Development of a SERM with
superagonist protective actions on the cardiovascu-
lar and skeletal systems may also become feasible.
Finally, recent studies suggest the even more excit-
ing possibility that the plasticity of estrogen action
exhibited by SERMs will be found to be a general
feature of the steroid nuclear-receptor family of
molecules. If so, corticosteroids may soon become
available that suppress immune function without
causing bone loss, as may androgens that are ana-
bolic without producing masculinizing effects or
adverse lipid patterns, progestins that can induce
endometrial atrophy without producing fluid reten-
tion, and vitamin D analogues that increase bone
density without inducing hypercalcemia or hyper-
calciuria. Should these events come to pass, the ini-
tial observation that tamoxifen has tissue-specific
effects will have proved to be the golden thread of
Ariadne. It will have led to the development of re-
markable new drugs that will selectively express the
desirable actions and selectively suppress the un-
desirable actions of the various steroid hormones.
Supported by grants (AR27065, AG04775, CA80181, and
CA15083) from the National Institutes of Health and by the Komen
Foundation.
We are indebted to Dr. Donald P. McDonnell for his assistance
with the section on the molecular actions of SERMs and Figure 2.

references
1. Gruber CJ, Tschugguel W, Schneeberger
C, Huber JC. Production and actions of estro-
gens. N Engl J Med 2002;346:340-52.
2. McDonnell DP. The molecular pharma-
cology of SERMs. Trends Endocrinol Metab
1999;10:301-11.
3. Jordan VC. Selective estrogen receptor
modulation: a personal perspective. Cancer
Res 2001;61:5683-7.
4. Kuiper GGJM, Carlsson B, Grandien K,
et al. Comparison of the ligand binding
specificity and transcript tissue distribution
of estrogen receptors a and b. Endocrinol-
ogy 1997;138:863-70.
5. Couse JF, Korach KS. Estrogen receptor
null mice: what have we learned and where
will they lead us? Endocr Rev 1999;20:358-
417. [Erratum, Endocr Rev 1999;20:459.]
6. Lindberg MK, Movérare S, Skrtic S, et al.
Estrogen receptor-b reduces estrogen recep-
tor-a regulated gene transcription, support-
ing a “Ying Yang” relationship between
estrogen receptor a and b in mice. Mol
Endocrinol (in press).
7. Pettersson K, Delaunay F, Gustafsson
J-Å. Estrogen receptor b acts as a dominant
regulator of estrogen signaling. Oncogene
2000;19:4970-8.
8. Hall JM, McDonnell DP. The estrogen
receptor b-isoform (ERb) of the human
estrogen receptor modulates ERa transcrip-
tional activity and is a key regulator of the
cellular response to estrogens and antiestro-
gens. Endocrinology 1999;140:5566-78.
9. Shiau AK, Barstad D, Loria PM, et al.
The structural basis of estrogen receptor/
coactivator recognition and the antagonism
of this interaction by tamoxifen. Cell 1998;
95:927-37.
10. Brzozowski AM, Pike AC, Dauter Z, et
al. Molecular basis of agonism and antago-
nism in the oestrogen receptor. Nature
1997;389:753-8.
11. Paige LA, Christensen DJ, Grøn H, et al.
Estrogen receptor (ER) modulators each
induce distinct conformational changes in
ERa and ERb. Proc Natl Acad Sci U S A
1999;96:3999-4004.
12. Pike ACW, Brzozowski AM, Walton J, et
al. Structural insights into the mode of action
of a pure antiestrogen. Structure (Camb)
2001;9:145-53.
13. McKenna NJ, Lanz RB, O’Malley BW.
Nuclear receptor coregulators: cellular and
molecular biology. Endocr Rev 1999;20:
321-44.
14. Smith CL, Nawaz Z, O’Malley BW. Coac-
tivator and corepressor regulation of the
agonist/antagonist activity of the mixed anti-
estrogen, 4-hydroxytamoxifen. Mol Endo-
crinol 1997;11:657-66.
15. Norris JD, Paige LA, Christensen DJ, et
al. Peptide antagonists of the human estro-
gen receptor. Science 1999;285:744-6.
16. Jepsen K, Hermanson O, Onami TM, et
al. Combinatorial roles of the nuclear recep-
tor corepressor in transcription and devel-
opment. Cell 2000;102:753-63.
17. Shang Y, Brown M. Molecular determi-
nants for the tissue specificity of SERMs.
Science 2002;295:2465-8.
18. Keating NL, Cleary PD, Rossi AS,

n engl j med 348;7 www.nejm.org february 13, 2003 627

 The
Zaslavsky AM, Ayanian JZ. Use of hormone
replacement therapy by postmenopausal
women in the United States. Ann Intern
Med 1999;130:545-53.
19. Writing Group for the Women’s Health
Initiative Investigators. Risks and benefits
of estrogen plus progestin in healthy post-
menopausal women: principal results from
the Women’s Health Initiative randomized
controlled trial. JAMA 2002;288:321-33.
20. The Writing Group for the PEPI Trial.
Effects of estrogen or estrogen/progestin
regimens on heart disease risk factors in
postmenopausal women: the Postmeno-
pausal Estrogen/Progestin Interventions
(PEPI) Trial. JAMA 1995;273:199-208. [Erra-
tum, JAMA 1995;274:1676.]
21. Riggs BL, Khosla S, Melton LJ III. Sex
steroids and the construction and conserva-
tion of the adult skeleton. Endocr Rev 2002;
23:279-302.
22. Gustafsson J-A. Estrogen receptor b —
a new dimension in estrogen mechanism of
action. J Endocrinol 1999;163:379-83.
23. Bord S, Horner A, Beavan S, Compston
J. Estrogen receptors a and b are differen-
tially expressed in developing human bone.
J Clin Endocrinol Metab 2001;86:2309-14.
24. Jordan VC, Phelps E, Lindgren JU. Effects
of anti-estrogens on bone in castrated and
intact female rats. Breast Cancer Res Treat
1987;10:31-5.
25. Turner RT, Wakley GK, Hannon KS, Bell
NH. Tamoxifen inhibits osteoclast-mediated
resorption of trabecular bone in ovarian hor-
mone-deficient rats. Endocrinology 1988;
122:1146-50.
26. Turken S, Siris E, Seldin D, Flaster E,
Hyman G, Lindsay R. Effects of tamoxifen on
spinal bone density in women with breast
cancer. J Natl Cancer Inst 1989;81:1086-8.
27. Love RR, Mazess RB, Barden HS, et al.
Effects of tamoxifen on bone mineral den-
sity in postmenopausal women with breast
cancer. N Engl J Med 1992;326:852-6.
28. Grey AB, Stapleton JP, Evans MC, Tat-
nell MA, Ames RW, Reid IR. The effect of the
antiestrogen tamoxifen on bone mineral
density in normal late postmenopausal
women. Am J Med 1995;99:636-41.
29. Powles TJ, Hickish T, Kanis JA, Tidy A,
Ashley S. Effect of tamoxifen on bone min-
eral density measured by dual-energy x-ray
absorptiometry in healthy premenopausal
and postmenopausal women. J Clin Oncol
1996;14:78-84.
30. Delmas PD, Bjarnason NH, Mitlak BH,
et al. Effects of raloxifene on bone mineral
density, serum cholesterol concentrations,
and uterine endometrium in postmeno-
pausal women. N Engl J Med 1997;337:
1641-7.
31. Lufkin EG, Whitaker MD, Nickelsen T,
et al. Treatment of established postmeno-
pausal osteoporosis with raloxifene: a ran-
domized trial. J Bone Miner Res 1998;13:
1747-54.
32. Ettinger B, Black DM, Mitlak BH, et al.
Reduction of vertebral fracture risk in post-
menopausal women with osteoporosis
628 n engl j med 348;7
new england journal of medicine
treated with raloxifene: results from a 3-year
randomized clinical trial. JAMA 1999;282:
637-45. [Erratum, JAMA 1999;282:2124.]
33. Johnston CC Jr, Bjarnason NH, Cohen
FJ, et al. Long-term effects of raloxifene on
bone mineral density, bone turnover, and
serum lipid levels in early postmenopausal
women: three-year data from 2 double-
blind, randomized, placebo-controlled tri-
als. Arch Intern Med 2000;160:3444-50.
34. Love RR, Barden HS, Mazess RB,
Epstein S, Chappell RJ. Effect of tamoxifen
on lumbar spine bone mineral density in
postmenopausal women after 5 years. Arch
Intern Med 1994;154:2585-8.
35. Kristensen B, Ejlertsen B, Mouridsen
HT, Andersen KW, Lauritzen JB. Femoral
fractures in postmenopausal breast cancer
patients treated with adjuvant tamoxifen.
Breast Cancer Res Treat 1996;39:321-6.
36. Fisher B, Costantino JP, Wickerham DL,
et al. Tamoxifen for prevention of breast
cancer: report of the National Surgical Adju-
vant Breast and Bowel Project P-1 Study.
J Natl Cancer Inst 1998;90:1371-88.
37. Marttunen MB, Hietanen P, Tiitinen A,
Ylikorkala O. Comparison of effects of
tamoxifen and toremifene on bone bio-
chemistry and bone mineral density in post-
menopausal breast cancer patients. J Clin
Endocrinol Metab 1998;83:1158-62.
38. Cummings SR, Black DM, Thompson
DE, et al. Effect of alendronate on risk of
fracture in women with low bone density
but without vertebral fractures: results from
the Fracture Intervention Trial. JAMA 1998;
280:2077-82.
39. Reginster J-Y, Minne HW, Sorensen OH,
et al. Randomized trial of the effects of ris-
edronate on vertebral fractures in women
with established postmenopausal osteo-
porosis. Osteoporos Int 2000;11:83-91.
40. Liberman UA, Weiss SR, Bröll J, et al.
Effect of oral alendronate on bone mineral
density and the incidence of fractures in
postmenopausal osteoporosis. N Engl J
Med 1995;333:1437-43.
41. Fogelman I, Ribot C, Smith R, Ethgen D,
Sod E, Reginster JY. Risedronate reverses
bone loss in postmenopausal women with
low bone mass: results from a multinational,
double-blind, placebo-controlled trial. J Clin
Endocrinol Metab 2000;85:1895-900.
42. Riggs BL, Melton LJ III. Bone turnover
matters: the raloxifene treatment paradox of
dramatic decreases in vertebral fractures
without commensurate increases in bone
density. J Bone Miner Res 2002;17:11-4.
43. Clemons M, Goss P. Estrogen and the
risk of breast cancer. N Engl J Med 2001;
344:276-85. [Erratum, N Engl J Med 2001;
344:1804.]
44. LaCroix AZ, Burke W. Breast cancer and
hormone replacement therapy. Lancet 1997;
350:1042-3.
45. Early Breast Cancer Trialists’ Collabora-
tive Group. Tamoxifen for early breast can-
cer: an overview of the randomised trials.
Lancet 1998;351:1451-67.
46. Adjuvant therapy for breast cancer. NIH
www.nejm.org february 13, 2003
consensus statement 2000. Vol. 17. No. 4.
Bethesda, Md.: National Institutes of Health,
November 2000. (Also available at http://
consensus.nih.gov/cons/114/
114_statement.pdf.)
47. Fisher B, Dignam J, Bryant J, Wolmark N.
Five versus more than five years of tamoxi-
fen for the lymph-node negative breast can-
cer: updated findings from the National Sur-
gical Adjuvant Breast and Bowel Project B-14
randomized trial. J Natl Cancer Inst 2001;
93:684-90.
48. Osborne CK. Tamoxifen in the treatment
of breast cancer. N Engl J Med 1998;339:
1609-18.
49. Abrams JS. Tamoxifen: five versus ten
years — is the end in sight? J Natl Cancer
Inst 2001;93:662-4.
50. Ellis PA, Saccani-Jotti G, Clarke R, et al.
Induction of apoptosis by tamoxifen and ICI
182780 in primary breast cancer. Int J Can-
cer 1997;72:608-13.
51. Hortobagyi GN. Treatment of breast
cancer. N Engl J Med 1998;339:974-84.
52. Powles T, Eeles R, Ashley S, et al.
Interim analysis of the incidence of breast
cancer in the Royal Marsden Hospital tamox-
ifen randomised chemoprevention trial. Lan-
cet 1998;352:98-101.
53. Veronesi U, Maisonneuve P, Costa A, et
al. Prevention of breast cancer with tamoxi-
fen: preliminary findings from the Italian
randomised trial among hysterectomised
women: Italian Tamoxifen Prevention Study.
Lancet 1998;352:93-7.
54. IBIS Investigators. First results from the
International Breast Cancer Intervention
study (IBIS-I): a randomised prevention trial.
Lancet 2002;360:817-24.
55. Chlebowski RT, Col N, Winer EP, et al.
American Society of Clinical Oncology tech-
nology assessment of pharmacologic inter-
ventions for breast cancer risk reduction
including tamoxifen, raloxifene, and aro-
matase inhibition. J Clin Oncol 2002;20:
3328-43.
56. Hayes DF, Van Zyl JA, Hacking A, et al.
Randomized comparison of tamoxifen and
two separate doses of toremifene in post-
menopausal patients with metastatic breast
cancer. J Clin Oncol 1995;13:2556-66.
57. Pyrhönen S, Valavaara R, Modig H, et al.
Comparison of toremifene and tamoxifen in
post-menopausal patients with advanced
breast cancer: a randomized double-blind,
the ‘nordic’ phase III study. Br J Cancer 1997;
76:270-7.
58. Buzdar AU, Hortobagyi G. Update on
endocrine therapy for breast cancer. Clin
Cancer Res 1998;4:527-34.
59. Cummings SR, Eckert S, Krueger KA,
et al. The effect of raloxifene on risk of
breast cancer in postmenopausal women:
results from the MORE Randomized Trial.
JAMA 1999;281:2189-97. [Erratum, JAMA
1999;282:2124.]
60. Wickerham DL, Tan-Chiu E. Breast
cancer chemoprevention: current status
and future directions. Semin Oncol 2001;
28:253-9.

61. ATLAS trial: NIH clinical trial. Oxford,
England: ATLAS Trial Office, 2003.
(Accessed January 24, 2003, at http://www.
ctsu.ox.ac.uk/˜ atlas/.)
62. Earl H, Gray R, Kerr D, Lee M. The opti-
mal duration of adjuvant tamoxifen treat-
ment for breast cancer remains uncertain:
randomize into aTTom. Clin Oncol (R Coll
Radiol) 1997;9:141-3.
63. Cuzick J. Chemoprevention of breast
cancer with tamoxifen. In: Hakama M, Beral
V, Buiatti E, Faivre J, Parkin DM, eds.
Chemoprevention in cancer control. Lyons,
France: International Agency for Research
on Cancer, 1996:95-109. (IARC scientific
publications no. 136.)
64. Mosca L, Barrett-Connor E, Wenger NK,
et al. Design and methods of the Raloxifene
Use for the Heart (RUTH) study. Am J Car-
diol 2001;88:392-5.
65. Münster PN, Buzdar A, Dhingra K, et
al. Phase I study of a third-generation
selective estrogen receptor modulator,
LY353381.HCL, in metastatic breast cancer.
J Clin Oncol 2001;19:2002-9.
66. Grady D, Gebretsadik T, Kerlikowske K,
Ernster V, Petitti D. Hormone replacement
therapy and endometrial cancer risk: a
meta-analysis. Obstet Gynecol 1995;85:
304-13.
67. Grady D, Brown JS, Vittinghoff E,
Applegate W, Varner E, Snyder T. Postmeno-
pausal hormones and incontinence: the
Heart and Estrogen/Progestin Replacement
Study. Obstet Gynecol 2001;97:116-20.
68. Wenger NK. Coronary heart disease: an
older woman’s major health risk. Br Med J
1997;315:1085-90.
69. Love RR, Wiebe DA, Newcomb PA, et al.
Effects of tamoxifen on cardiovascular risk
factors in postmenopausal women. Ann
Intern Med 1991;115:860-4.
70. Walsh BW, Kuller LH, Wild RA, et al.
Effects of raloxifene on serum lipids and
coagulation factors in healthy postmeno-
pausal women. JAMA 1998;279:1445-51.
71. Saarto T, Blomqvist C, Ehnholm C,
Taskinen M-R, Elomaa I. Antiatherogenic
effects of adjuvant antiestrogens: a random-
ized trial comparing the effects of tamoxifen
and toremifene on plasma lipid levels in
postmenopausal women with node-positive
breast cancer. J Clin Oncol 1996;14:429-33.
72. Walsh BW, Paul S, Wild RA, et al. The
effects of hormone replacement therapy and
raloxifene on C-reactive protein and
homocysteine in healthy postmenopausal
women: a randomized, controlled trial.
J Clin Endocrinol Metab 2000;85:214-8.
73. Cox DA, Sashegyi A, Paul S, et al. Effects
of raloxifene and hormone replacement ther-
apy on markers of inflammation in healthy
postmenopausal women. Osteoporos Int
2000;11:Suppl 2:S207. abstract.
74. Clarkson TB, Anthony MS, Jerome CP.
Lack of effect of raloxifene on coronary artery
atherosclerosis of postmenopausal mon-
keys. J Clin Endocrinol Metab 1998;83:721-
6.
n engl j med 348;7
drug therapy
75. Bjarnason NH, Haarbo J, Byrjalsen I,
Alexandersen P, Kauffman RF, Christiansen
C. Raloxifene and estrogen reduce progres-
sion of advanced atherosclerosis — a study
in ovariectomized, cholesterol-fed rabbits.
Atherosclerosis 2001;154:97-102.
76. Pavo I, Laszlo F, Morschl E, et al. Ralox-
ifene, an oestrogen-receptor modulator, pre-
vents decreased constitutive nitric oxide and
vasoconstriction in ovariectomized rats. Eur
J Pharmacol 2000;410:101-4.
77. Kauffman RF, Bean JS, Fahey KJ, Culli-
nan GJ, Cox DA, Bensch WR. Raloxifene
and estrogen inhibit neointimal thickening
after balloon injury in the carotid artery of
male and ovariectomized female rats. J Car-
diovasc Pharmacol 2000;36:459-65.
78. Barrett-Connor E, Stuenkel C. Hor-
mones and heart disease in women: Heart
and Estrogen/Progestin Replacement Study
in perspective. J Clin Endocrinol Metab
1999;84:1848-53.
79. Hulley S, Grady D, Bush T, et al. Ran-
domized trial of estrogen plus progestin for
secondary prevention of coronary heart dis-
ease in postmenopausal women. JAMA 1998;
280:605-13.
80. Herrington DM, Reboussin DM, Bros-
nihan KB, et al. Effects of estrogen replace-
ment on the progression of coronary-artery
atherosclerosis. N Engl J Med 2000;343:
522-9.
81. Hodis HN, Mack WJ, Lobo RA, et al.
Estrogen in the prevention of atherosclero-
sis: a randomized, double-blind, placebo-
controlled trial. Ann Intern Med 2001;135:
939-53.
82. Barrett-Connor E, Grady D, Sashegyi A,
et al. Raloxifene and cardiovascular events
in osteoporotic postmenopausal women:
four-year results from the MORE (Multiple
Outcomes of Raloxifene Evaluation) ran-
domized trial. JAMA 2002;287:847-57.
83. Ridker PM, Rifai N, Rose L, Buring JE,
Cook NR. Comparison of C-reactive protein
and low-density lipoprotein cholesterol lev-
els in the prediction of first cardiovascular
events. N Engl J Med 2002;347:1557-65.
84. Daly E, Vessey MP, Hawkins MM, Car-
son JL, Gough P, Marsh S. Risk of venous
thromboembolism in users of hormone
replacement therapy. Lancet 1996;348:977-
80.
85. McEwen BS, Alves SE. Estrogen actions
in the central nervous system. Endocr Rev
1999;20:279-307.
86. Hlatky MA, Boothroyd D, Vittinghoff E,
Sharp P, Whooley MA. Quality-of-life and
depressive symptoms in postmenopausal
women after receiving hormone therapy:
results from the Heart and Estrogen/Proges-
tin Replacement Study (HERS) trial. JAMA
2002;287:591-7.
87. Zandi PP, Carlson MC, Plassman BL, et
al. Hormone replacement therapy and inci-
dence of Alzheimer disease in older women:
the Cache County study. JAMA 2002;288:
2123-9.
88. Mulnard RA, Cotman CW, Kawas C, et
www.nejm.org february 13, 2003
al. Estrogen replacement therapy for treat-
ment of mild to moderate Alzheimer dis-
ease: a randomized controlled trial: Alzhei-
mer’s Disease Cooperative Study. JAMA
2000;283:1007-15. [Erratum, JAMA 2000;
284:2597.]
89. Wu X, Glinn MA, Ostrowski NL, et al.
Raloxifene and estradiol benzoate both fully
restore hippocampal choline acetyltransfer-
ase activity in ovariectomized rats. Brain Res
1999;847:98-104.
90. Yaffe K, Krueger K, Sarkar S, et al. Cog-
nitive function in postmenopausal women
treated with raloxifene. N Engl J Med 2001;
344:1207-13.
91. Eastell R. Treatment of postmenopausal
osteoporosis. N Engl J Med 1998;338:736-
46.
92. Bonneterre J, Thurlimann B, Robertson
JFR, et al. Anastrozole versus tamoxifen as
first-line therapy for advanced breast cancer
in 668 postmenopausal women: results of
the Tamoxifen or Arimidex Randomized
Group Efficacy and Tolerability study. J Clin
Oncol 2000;18:3748-57.
93. Nabholtz JM, Buzdar A, Pollak M, et al.
Anastrozole is superior to tamoxifen as
first-line therapy for advanced breast can-
cer in postmenopausal women: results of a
North American multicenter randomized
trial. J Clin Oncol 2000;18:3758-67.
94. Goss PE, Strasser K. Aromatase inhibi-
tors in the treatment and prevention of
breast cancer. J Clin Oncol 2001;19:881-94.
95. The ATAC Trialists’ Group. Anastrozole
alone or in combination with tamoxifen ver-
sus tamoxifen alone for adjuvant treatment
of postmenopausal women with early breast
cancer: first results of the ATAC randomised
trial. Lancet 2002;359:2131-9.
96. Heshmati HM, Khosla S, Robins SP,
O’Fallon WM, Melton LJ III, Riggs BL. Role
of low levels of endogenous estrogen in reg-
ulation of bone resorption in late postmeno-
pausal women. J Bone Miner Res 2002;17:
172-8.
97. Winer EP, Hudis C, Burstein HJ, et al.
American Society of Clinical Oncology tech-
nology assessment on the use of aromatase
inhibitors as adjuvant therapy for women
with hormone receptor-positive breast can-
cer: status report 2002. J Clin Oncol 2002;
20:3317-27.
98. Rebbeck TR, Lynch HT, Neuhausen SL,
et al. Prophylactic oophorectomy in carriers
of BRCA1 or BRCA2 mutations. N Engl J Med
2002;346:1616-22.
99. Kauff ND, Satagopan JM, Robson ME,
et al. Risk-reducing salpingo-oophorec-
tomy in women with a BRCA1 or BRCA2
mutation. N Engl J Med 2002;346:1609-15.
100. King M-C, Wieand S, Hale K, et al.
Tamoxifen and breast cancer incidence
among women with inherited mutations in
BRCA1 and BRCA2: National Surgical Adju-
vant Breast and Bowel Project (NSABP-P1)
Breast Cancer Prevention Trial. JAMA 2001;
286:2251-6.
Copyright © 2003 Massachusetts Medical Society.
629