Articles
Register for The Editorial:
Is HRT still indicated for the primary prevention of osteoporosis?
by Martin H. Birkhäuser
Reprinted with permission of the International Society of Gynecological Endocrinology
1. Hormone Replacement Therapy (HRT) for Fracture Prevention
Hormone Replacement Therapy: In most European countries, the actual Guideline on
the Evaluation of New Medicinal Products in the Treatment of Primary
Osteoporosis is incomplete in that, unlike the previous Note for Guidance on
Postmenopausal Osteoporosis in Women (EMEA 2001 Guidance), it fails to address
the prevention of osteoporosis indication.
Early prevention of osteoporosis is essential because the majority of fractures
occur in the non-osteoporotic population. Furthermore, it is well known that
individuals who had one osteoporotic fracture approximately double their risk to
have another. Although treatment of osteoporosis can reduce the risk for a
subsequent fracture, it cannot eliminate the excess risk. It is therefore
essential to maintain bone architecture and quality as early as in the peri- and
early postmenopause. Gynaecological endocrinologists are therefore greatly
astonished that EMEA did not maintain the prevention of osteoporosis by
estrogens as a first line medical intervention in women with an increased risk
of osteoporosis and bone fractures in its latest recommendations.
It is well known that the first 5–10 years after menopause are associated with
an accelerated rate of bone loss. Primary prevention of osteoporosis is directed
at women identified as being at increased risk for the disease but without
established disease. HRT has been shown to reduce postmenopausal bone loss and
reduce fracture incidence in several RCT studies, including the PEPI trial
(Writing group for the PEPI trial, 1996) and the HOPE trial (Lindsay et al.,
2002) trial. More recently, the WHI trial, using a fixed combination of
conjugated equine estrogens 0.625 mg and medroxyprogesterone acetate 2.5 mg, has
shown that the resulting loss of bone architecture and quality this is
preventable with hormone therapy and confirmed the fracture reduction efficacy
of HRT, including hip and spine fractures (Cauley et al., 2003; Women’s Health
Initiative Steering Committee, 2004). The WHI trial was of women aged between 50
and 80 years, who were not known to have increased fracture risk and who were
supposed to be in general good health. In addition, the Women’s Health
Initiative (Cauley et al., 2003) has demonstrated fracture prevention with
hormone therapy in a non-osteoporotic population, too. This preventive effect is
maintained: the PERF study (Bagger et al., 2004) demonstrated that
administration of hormone therapy in early postmenopause offers a long-lasting
benefit for the prevention of postmenopausal bone loss and osteoporotic
fracture. Lower doses than previously thought necessary are now proving
effective (Lees and Stevenson, 2001; Lindsay et al., 2002; Ettinger et al.,
2004, Greenwald et al., 2005), and a very low dose estrogen product has recently
been licensed for osteoporosis prevention by the US Food and Drugs
Administration.
Hormone Replacement Therapy (HRT) is an effective treatment for prevention of
osteoporotic fractures in women without established disease. These women tend to
be younger, since older women have an increasing prevalence of osteoporosis as
defined by bone density criteria (Kanis et al., 2001). Are younger women with
osteopenia, rather than osteoporosis, at increased fracture risk? In the age
group 50–80 years, around 36% of classical osteoporotic fractures occur in those
below age 65 years (Singer et al., 1998; Kanis et al., 2004). In the UK, this
amounts to over 100 000 fractures per year in women aged 50–65 years, over 9000
of which are hip fractures (Singer et al., 1998; Kanis et al., 2004;
www.uk2u.net).
There is no doubting the efficacy of HRT for the primary prevention of
osteoporosis in post-menopausal women. Its current role in prevention of
fractures seems best suited to those younger post-menopausal women with
increased risk. In such a population, HRT use for osteoporosis prevention would
be cost-effective (Lamy et al., 2003).
2. Alternatives to HRT:
The currently licensed alternatives to HRT for the secondary prevention of
osteoporosis, preventing fractures in women with existing osteoporosis, include
bisphosphonates, raloxifene, teriparatide, strontium ranelate, calcitonin and
anabolic steroids. Calcium supplements and Vitamine D are licensed as an adjunct
to therapy.
Of these, only bisphosphonates such as alendronate, risedronate, ibandronate and
zoledronate have been shown to reduce the incidence of hip fractures, yet hip
fracture is the most important osteoporotic fracture. However, both alendronate
(Cummings et al., 1998) and risedronate (McClung et al., 2001) and other
bisphoshonates have only been shown to prevent hip fracture in elderly women,
mean age above 65 years, who already had osteoporosis, with bone density T-score
in either hip or spine below –2.5, and in many cases had already sustained a
fracture. Hip fractures occur most commonly in elderly women. However,
bisphosphonates have not been shown in prospective studies to prevent hip
fractures in younger women with increased risk rather than established disease.
Furthermore, there are no data are available in precocious and in early
menopause or in longstanding amenorrhea. However, today, there is an agreement
that bisphosphonate use is more appropriate in asymptomatic elderly women being
10 or more years after their menopause than HRT regimens.
SERMs (selective estrogen receptor modulators) SERMs are licensed for the
prevention and treatment of spinal osteoporosis in postmenopausal women. They
have not been shown to reduce hip fracture risk. There is good evidence (level
1) for a reduction in estrogen receptor-positive breast cancers in users of
SERMs. Therefore, there is an indication for SERMs in women with an increased
risk for breast cancer. Currently available SERMs are not able to relieve
climacteric symptoms in early postmenopausal women and may make them worse.
SERMs have the same thromboembolic risk as per-oral HRT.
3. Main non-skeletal benefits and riks for HRT
Climacteric syndrome: The uncontested main indication for HRT remains the
relief of post-menopausal symptoms, which brings a major improvement in quality
of life (Utian et al., 2001; Birkhäuser et al., 2008; Pines et al, 2007 and
2008). No other therapy has proved to be more effective than HRT in this
respect.
Cardiovascular risk: EMEA expressed concern because of a possible
increase of the cardiovascular (CVD ) risk of HRT in December 2003, mainly based
on the publication of the first analysis of the WHI data in the combined in the
estrogen-only arm (Manson et al., 2003; The Women’ Health Initiative Steering
Committee, 2004). In fact, the WHI study was not designed, and therefore was not
powered, to investigate the consequences of hormone replacement therapy (HRT) in
women below 60 years of age. Therefore, any attempt to present the results of
the study as indicating that HRT may inflict damage to the heart in general is
profoundly wrong and must be amended.
It should be noted that the recent final analysis of WHI data (Hsia et al.,
2006), as well as the latest update of the results of the Nurses’ Health Study (Grodstein
et al., 2006 and 2000) have uniformly shown that HRT, when started soon after
the start of menopause, does not increase the risk of CVD but may even protect
against myocardial infarction. The outcomes in the youngest cohort of the WHI
study (age 50–59 years or within 10 years of menopause) confirms previous
evidence regarding a favourable effects on cardiovascular risk factors, vascular
structure and function in nonatherosclerotic vessels, a reduced insulin
resistance and reduction in the risk of diabetes and the preventative benefit of
HRT for CHD in the majority of pre-clinical and observational studies. Young
healthy postmenopausal women can be started on HRT when clinically warranted
without fear of increased cardiovascular disease risk.
Thromboembolism and Stroke: The risks of ET for venous thromboembolism
have been overstated in the WHI (Cushman, 2004) for the younger study
participants. The VTE risk was age-dependent and significantly higher in the
older and in obese women (BMI>25). The same has been found for stroke (Lobo et
al., 2007). As Canonico et al. (2007) have shown, there is no significantly
increased venous thromboembolic risk in women with parenteral estrogen
application. Women seeking HRT who have potential or confirmed risk factors for
venous thromboembolism and stroke need individualized counselling; in these
situations, transdermal HRT might be preferable to oral formulations.
Breast cancer risk: The intention to eliminate the use of HRT from the
prevention of osteoporsis / osteoporotic fractures, as expressed e.g. by EMEA in
December 2003, is largely the consequence of the believe that HRT should
increase the risk of breast cancer, mainly deduced again from the first
publications of the WHI data (Writing Group for the Women’s Health Initiative
Investigators, 2002; Chlebowski et al., 2003). However, newer publications from
the WHI are showing a quite different picture. Because of its inherent
weaknesses, we believe that the MWS (Million Women Study Collaborators, 2003; an
observational study and not an RCT) should not be used as part of an evidence
base for HRT and should be discounted by regulatory authorities.
The new sub-analysis of the unopposed estrogen arm of the WHI study has modified
further the original alarmist messages (Stefanic et al., 2006). The WHI study
investigators have released the details of breast cancer and mammography
screening data for 10,739 women with prior hysterectomy, who received either
estrogen replacement treatment (ERT) or placebo for a mean follow-up of 7.1
years (Stefanic et al., 2006). Thirty per cent of the subjects were aged 50–59
years and 24% were 70–79 years old at study entry; over half had never taken ET
prior to the study. The initial report on the estrogen-only arm (The Women’
Health Initiative Steering Committee, 2004) concluded that there was a
non-significant (p < 0.06) decreased risk for breast cancer in ERT users, but
subgroup analyses now reveal that first lifetime exposure to ET at the trial was
associated with fewer breast cancer cases as compared to placebo (hazard ratio
(HR), 0.76; 95% confidence interval (CI), 0.58–0.99; p < 0.05). On the other
hand, mammographic breast density increased in ERT users, with 9.2% having
abnormalities in the ET group vs. 5.5% in the placebo group at 1 year (p <
0.001) and a cumulative percentage of 36.2% and 28.1%, respectively, leading to
more breast biopsies.
Overall, the WHI study is in line with older observational data (Bush et al.,
2001; Colditz et al., 1995) and carries the message that ERT for postmenopausal
women does not increase the risk of breast cancer for at least up to 7 years.
The incidence of diagnosis of breast cancer is increased with prolonged use of
estrogen and, to a greater extent, with estrogen/progestogen treatment. Whether
all progestogens are associated with an increase in breast cancer risk is
currently unclear.
There is evidence that there may be some differences between the various
estrogens, progestogens and progesterone in terms of their effect on the risk of
breast cancer, but this requires more studies to confirm. Patients should be
reassured that the possible risk of breast cancer after long-term HRT use (> 7
years in the WHI) is small and is comparable to the risk of a 1-year delay in
menopause. Breast cancer risk is influenced by the total duration of exposure to
endogenous and exogenous estrogens and progestogens. Some progestogens may
increase the estrogen-dependent proliferation of mammary tissue. The WHI found
no increase in breast cancer risk for duration of 7 years in participants with
no prior use of HRT (Anderson et al, 2006).
In conclusion, young postmenopausal women starting on combined HRT for the first
time should be advised that breast cancer risks do not appear to increase in the
first 7 years of usage. Hysterectomized women on unopposed estrogen are not at
increased risk of breast cancer and some may even have a small reduction in
risk.
HRT and total mortality: There is no increase in total all-cause mortality in
women using HRT (Salpeter et al., 2004), as might be the case if there were a
significant increase in the risk of cardiovascular disease and breast cancer. In
contrast, in younger women (50-60 years, or starting HRT less than 10 years
after menopause), a recent metaanalysis suggests a significantly decreased risk
of all-cause mortality. This supports the ‘window of opportunity’ theory where
early use of HRT is associated with preventative cardiovascular benefits and is
in agreement with the latest data from both the WHI and the Nurses’ Health
Study.
4. Conclusions
Once again, it is apparent that the alarmist reports that spread worldwide when
the first results of the WHI study were published in 2002 were unjustified based
on the more recent further analyses, particularly in peri- and early
postmenopausal women. It is very regrettable that, as a result, so many women
and their medical advisors have lost confidence in the merits of HT, which will
now be difficult to redress. Nevertheless, they should be reassured that, once
recommended for an approved indication, with correct timing and after proper
individual considerations, ERT/HRT is an effective therapy which may even have
some extra benefits (cardiac, metabolic) in certain subgroups of women, for the
period of time investigated in the WHI study. The risks of ERT/HRT have been
overstated. Lower or ultra-low dosages of ERT/HRT may have an even better safety
profile.
Finally, HRT is less expensive than any of the alternatives, and thus its use
for primary prevention should be actively encouraged targeting women in their
early postmenopause at high risk for fracture, such as those with osteopenia,
with a family history of hip fracture, with a low body mass index or with a
history of corticosteroid use. Bisphosphonates may be the more cost-effective
intervention for elderly osteoporotic women. However, new data from a long-term
prospective study suggest that even limited HRT use in women during early
menopause may result in fracture reduction later (Bagger et al., 2004). A recent
analysis of the WHI study data also showed that overall fracture benefit was
maintained 16 months after discontinuation of estrogen plus progestogen (Jackson
et al., 2004). Both these findings need to be confirmed.
With regard to safety issues, some regulatory authorities may have
misinterpreted the new data from the recent studies such as the WHI.
HRT should be started with the lowest appropriate dose and increased if
necessary.
Although low and ultra-low doses have been shown to be effective for menopause
symptom relief data on reduction of fracture risk are missing. Women with
premature ovarian failure generally require higher doses of HRT than those
conventionally used in peri- and postmenopausal women. Arbitrary limitations on
duration of therapy should be avoided. Benefits, risks and patient preferences
should be discussed at least annually. Counselling should include discussion of
benefits and risks in terms of absolute numbers of events rather than percentage
changes.
In conclusion, HRT should be considered a first-line option for the primary
prevention of osteoporosis-related fracture in precocious menopause even if
asymptomatic. In the absence of data for the efficacy and safety of alternative
preparations, HRT should be recommended in postmenopausal women at risk of
osteoporosis-related fractures below the age of 60 as a first-line therapy. It
should also be available to those older women with increased risk who either
have persisting menopausal symptoms or who make an informed choice to use it.
Correspondence:
Prof. M. Birkhäuser
President of the Swiss Society against Osteoporosis
Founding Honorary President of EMAS
Gartenstrasse 67
CH-4052 Basel
Switzerland
martin.birkhaeuser@balcab.ch
Continue to
page 2
© International Society of Gynecological Endocrinology - n. 30/2
February 2009
