Antiestrogens: Future Prospects
Antiestrogens: Future Prospects
Triphenylethylene structures similar to tamoxifen (Nolvadex) were first
used in the 1940s, but in comparison to synthetic and natural estrogens
used at that time, they were not particularly potent antitumor agents.
[1,2] Tamoxifen was introduced into the clinic in 1969. Because of its
much improved toxicity profile over the estrogens and androgens in use
at that time, it rapidly gained acceptance as the endocrine treatment of
choice for advanced breast cancer and subsequently as adjuvant therapy.[3-5]
Three different strategies have been used to develop new antiestrogens:
(1) chemical alteration of the triphenylethylene ring of tamoxifen, (2)
production of new nonsteroidal ring structures (eg, the benzothiophenes),
and (3) production of steroidal estradiol analogs with pure antiestrogenic
activity (Figure 1).
All of the nonsteroidal antiestrogens discussed in this article are
partial estrogen agonists to differing degrees. In contrast, the steroidal
pure antiestrogens lack intrinsic estrogenic activity and appear to have
a somewhat different mechanism of action. Although both types of compounds
bind to the estrogen receptor (ER), the tamoxifen analogs allow partial
activation of estrogen receptors, permitting transcription of some estrogen-regulated
genes, whereas the steroidal antiestrogens appear to completely inhibit
transcriptional activity via the estrogen receptor.
The six nonsteroidal antiestrogens shown in Figure
1 are either marketed or in clinical trials. All but one, raloxifene,
are based on the triphenylethylene molecule. In addition, pure steroidal
antiestrogens (ICI 164,384 and ICI 182,780) have been developed in order
to produce a state of complete estrogen deprivation. The overall goal for
the development of new antiestrogens is to increase their efficacy and
safety (Table 1).
This last characteristic is important because many women taking these
agents are treated for long periods with adjuvant therapy. Since most breast
cancer patients are unlikely to receive hormone replacement therapy due
to fear of disease recurrence, new endocrine breast cancer agents with
an improved side-effect profile and favorable effects on bone mineral density
and the cardiovascular system would be highly valuable.
The purpose of this review is: (1) to describe the current status and
ultimate potential of the nonsteroidal antiestrogen agents and (2) to introduce
a new class of pure steroidal antiestrogen agents.
Tamoxifen has gained wide clinical acceptance, first for metastatic
breast cancer and then for adjuvant therapy. Currently, tamoxifen is being
studied for the chemoprevention of breast cancer in healthy women at high
risk for the disease. The use of tamoxifen has expanded over the years
because of its efficacy in prolonging overall and disease-free survival
as well as reducing the incidence of contralateral breast cancer. Because
tamoxifen is a mixed estrogen agonist and antagonist, it has been shown
to have positive, estrogenic effects on bone and the cardiovascular system
while maintaining antitumor activity via its antiestrogenic effects. The
estrogenic activity of tamoxifen has been associated with a proliferative
effect on the endometrium and a two- to threefold increase in the risk
of endometrial carcinoma. This association has stimulated interest in the
discovery of new antiestrogens with no uterotropic activity. The methods
for evaluating these compounds are shown in Table
Toremifene (Fareston) differs from tamoxifen by only a single chlorine
atom (see Figure 1). Although it produces
different metabolites due to the stability of the chlorine atom, toremifene
has a pharmacologic profile similar to tamoxifen.
Toremifene displaced [3H]estradiol binding to estrogen receptors
by 50% at a concentration of 0.5 mmol/L. In the ER-positive human breast
carcinoma cell line MCF-7, toremifene showed a pattern of activity similar
to tamoxifen: growth inhibition at low concentrations and oncolytic activity
at high concentrations. A concentration of 5 × 10-6
M toremifene killed all MCF-7 cells within two days. The activity of
toremifene appears to be estrogen-dependent; growth inhibition of MCF-7
cells at concentrations less than 10-6 M can be reversed with
Against dimethylbenzanthracene (DMBA)-induced rat mammary cancer, toremifene
and tamoxifen showed similar antitumor activity; the one difference was
that 45 mg/kg tamoxifen was toxic to the rat, whereas 45 mg/kg toremifene
showed an antitumor effect.
In the rat uterus, the estrogenic effects of toremifene appear to be
lower than tamoxifen at low and moderate concentrations; however, the maximum
estrogenic response was similar. Toremifene has also demonstrated estrogenic
effects on human endometrial tissue in postmenopausal breast cancer patients,
including increasing endometrial thickening and proliferation.
Unlike tamoxifen, toremifene did not produce liver tumors in rats at
doses up to 48 mg/kg for one year.
Toremifene has been studied in a comparative trial with tamoxifen in
648 perimenopausal or postmenopausal women with metastatic breast cancer
(hormone receptor positive or unknown receptor status). The three treatment
arms consisted of tamoxifen (20 mg/d), toremifene (60 mg/d), or toremifene
(200 mg/d). In the intent-to-treat analysis, the frequency of objective
response (complete or partial) was 19% for tamoxifen, 21% for 60-mg toremifene,
and 23% for 200-mg toremifene (P not significant between treatments). The
median response durations were 19.1 months, 16.9 months, and 18.4 months,
respectively (P not significant between treatments). Likewise, there were
no significant differences among the three treatments in the median time
to progression or median overall survival.
Overall, the type and frequency of adverse effects were similar among
the three groups, including the incidence of tumor flare and thromboembolic
and cardiac events; however, the 200-mg toremifene dose was associated
with a greater frequency of elevated aspartate aminotransferase (AST) levels
At this time, there does not appear to be any benefit of toremifene
over tamoxifen. The clinical experience with toremifene will need to be
expanded substantially to determine if there is any benefit of the drug
with respect to rare adverse events such as endometrial cancer.
Droloxifene (3-hydroxytamoxifen) is currently undergoing clinical trials
in advanced breast cancer. Unlike tamoxifen, droloxifene is itself the
active moiety and thus does not require metabolism for activation.
Droloxifene shows a higher binding affinity for ER compared with tamoxifen.
The IC50 for the displacement of 17-beta-estradiol from ER is
1 × 10-8 M. In rats, droloxifene exhibits higher antiestrogenic
activity and lower estrogenic activity than tamoxifen.
Droloxifene is more effective than tamoxifen in inhibiting the growth
of ER-positive breast cancer cells, even at therapeutic concentrations
(0.1 to 0.4 mM), and its activity is related to ER content. Interestingly,
short-term (1-hour) exposure to droloxifene in vitro produced maximum growth
inhibition, leading investigators to conclude that the agent may be suitable
for intermittent therapy. Moreover, droloxifene produced a greater antitumor
effect than tamoxifen against DMBA-induced breast tumors in rats. Droloxifene
(at doses up to 200 mg/kg/d for 6 months or doses up to 90 mg/kg/d for
24 months) is not hepatocarcinogenic in the rat.
A large phase II trial of droloxifene was conducted in centers located
in Europe, Canada, and Brazil. The purpose of the study was to compare
three daily doses (20 mg, 40 mg, and 100 mg) in postmenopausal women with
advanced breast cancer. Eligible patients had no prior exposure to systemic
hormonal therapy and had positive or unknown hormone-receptor status; of
369 randomized patients, 268 were evaluable for response.
Complete or partial responses were seen in 30% of the 20-mg group, 47%
of the 40-mg group, and 44% of the 100-mg group. The differences were significant
between the 20-mg group and the 40-mg group (P = .02) and between the 20-mg
group and the 100-mg group (P = .04). Half of the responses were seen within
the first two months of starting treatment.
The median duration of response was 12 months for 20 mg, 15 months for
40 mg, and 18 months for 100 mg. Again, the 40-mg (P = .02) and the 100-mg
(P = .01) groups had significantly better results than the 20-mg group.
Side effects reported by more than 20% of patients included hot flashes,
lassitude, and nausea; the frequency of these effects did not appear to
be dose related. Thromboembolic events did not occur more frequently than
with other antiestrogens.
Issues that remain to be resolved for droloxifene include its effects
on the uterus, bone, and the cardiovascular system as well as its potential
for cross-resistance with tamoxifen. Droloxifene has been shown to have
estrogenic effects on bone in the rat, and clinically, has produced
decreases in plasma cholesterol without affecting cholesterol synthesis.
Although these results look promising, as with toremifene, droloxifene's
ultimate clinical profile will only be determined in trials comparing it
Raloxifene is a benzothiophene derivative with a high binding affinity
for ER, reportedly 2.9-fold greater than that of estradiol. In the
rat, raloxifene has exhibited antiestrogenic activity in the breast and
uterus and estrogen agonist activity on bone and lipids.
In contrast to tamoxifen, raloxifene showed a lack of uterotropic effect
after four days of treatment in ovariectomized rats. Near complete
antagonism of estrogen-induced uterotropic activity was observed at 1 mg/d.
Moreover, three days of raloxifene was able to block the uterotropic action
of estradiol for 10 subsequent days. Raloxifene was unable to reverse the
uterotropic response to tamoxifen. Raloxifene dose-dependently inhibited
estrogen-stimulated proliferation of MCF-7 cells.
Sato et al compared the effects of raloxifene and tamoxifen on bone,
cholesterol, and the uterus in six-month-old, ovariectomized rats. The
effect of raloxifene on bone mineral density was dose-dependent, with an
ED50 of 0.3 mg/kg/d (35 days of treatment), while the ED50 for tamoxifen
was 0.1 mg/kg/d. At doses of 0.1-10 mg/kg, raloxifene dose-dependently
reduced cholesterol levels to 51% to 62% of ovariectomized controls. The
results were similar for tamoxifen. In contrast, uterine epithelial thickness
increased by 250% with tamoxifen therapy, compared with only 60% seen with
Raloxifene (100 mg orally, two times a day) was studied in 14 patients
with disseminated breast cancer refractory to tamoxifen. There was
one minor response in a patient with soft-tissue disease and in five patients
with stable disease. Side effects included hot flashes (n = 4), fatigue
(n = 3), leg cramps (n = 1), and mild nausea (n = 3).
Although the antitumor activity of raloxifene needs further evaluation,
the drug may have a role in the prevention of osteoporosis in postmenopausal
women. In this population, raloxifene doses 50 or more mg/d have significantly
reduced total serum and low-density lipoprotein (LDL) cholesterol as well
as serum markers of bone turnover (eg, osteocalcin, alkaline phosphatase).