Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Hormone therapy timing hypothesis and atherosclerosis
(USC Thesis Other)
Hormone therapy timing hypothesis and atherosclerosis
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
HORMONE THERAPY TIMING HYPOTHESIS AND ATHEROSCLEROSIS
by
Intira Sriprasert
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(EPIDEMIOLOGY)
-- AUGUST 2020 --
Copyright 2020 Intira Sriprasert
ii
ACKNOWLEDGEMENTS
I would like to express my deepest appreciation to Dr. Wendy Mack and Dr. Howard Hodis
who have given me the opportunities to pursue my graduate studies. I have learned so much
from their invaluable knowledge and experience. They always supported and guided me through
any difficulties throughout the program. To me, they are not only mentors but also inspirational
role models in the academic professions.
I would like to acknowledge the other committee members of this dissertation who have
provided me with their guidance, feedback, insight, and support including Dr. Roksana Karim, Dr.
Frank Stanczyk and Dr. Roberta McKean-Cowdin. I would like to give special thanks to all staff
members from the Atherosclerosis Research Unit for providing great support for my analysis,
especially Ms. Naoko Kono. In addition, I would like to acknowledge Dr. Sebastian Mirkin and Dr.
Brian Bernick from TherapeuticsMD for their support and the great opportunity to use their data
in my studies.
Thank you to all my beloved friends in the program: Ugonna Ihenacho, Zhi Yang, Ashley
Song, Zhaohui Du, Charlotte Du and Malcolm Barrett. Thank you so much for being great
companions through the program and always being there for me.
Last but not least, I would like to sincerely thank my family for their unconditional and
unlimited understanding. I am very grateful for your support. You mean the world to me and I
could not be who I am today without you.
iii
TABLE OF CONTENTS
CONTENT PAGE
ACKNOWLEDGEMENTS ii
LIST OF TABLES vii
LIST OF FIGURES ix
ABBREVIATIONS xi
ABSTRACT xiii
CHAPTER 1: INTRODUCTION AND BACKGROUND OF HORMONE THERAPY AND 1
CORONARY HEART DISEASE
1.1 CORONARY HEART DISEASE AMONG POSTMENOPAUSAL WOMEN 2
1.1.1 Incidence of coronary heart disease among postmenopausal 2
women
1.1.2 Estrogen deficiency as a risk factor for atherosclerosis and 3
coronary heart disease
1.2 HORMONE THERAPY AND CORONARY HEART DISEASE 5
1.2.1 Observational studies of primary prevention 5
1.2.2 Randomized Clinical trials of secondary prevention 6
1.2.3 Randomized Clinical trials of primary prevention 9
1.2.4 Discrepancy of results from observational studies and 10
clinical trials
1.2.5 Women’s Health Initiative trial stratified results 13
1.2.6 Randomized Clinical trials of primary prevention among 15
younger women
1.3 HORMONE THERAPY TIMING HYPOTHESIS 16
1.3.1 Pathology of vascular aging 17
1.3.2 Mouse studies 18
1.3.3 Primate studies 19
1.3.4 Human data 21
1.4 STUDY OBJECTIVES 23
1.5 TABLES AND FIGURES 26
1.6 REFERENCES 31
CHAPTER 2: DATASET AND BACKGROUND OF STUDY VARIABLES RELATED 41
TO HORMONE THERAPY EFFECT ON CORONARY HEART DISEASE
2.1 DATESET FOR ANALYSIS 42
2.1.1 The Early versus Late Intervention Trial of Estradiol (ELITE) 42
2.1.2 REPLENISH trial 44
2.2 EXPOSURE: HORMONE THERAPY 49
2.2.1 Estrogen and progesterone as hormone therapy 49
2.2.2 Estrogen structure and receptors 49
iv
2.2.3 Estrogen effects on coronary heart disease 52
2.2.4 Progesterone structure and receptors 54
2.2.5 Progesterone effects on coronary heart disease 55
2.2.6 Timing as a possible modifying factor of HT effect on 56
coronary heart disease
2.2.7 Hormone regimen used in ELITE and REPLENISH 58
2.2.8 Measurement of estrogens and progesterone 58
in ELITE and REPLENISH
2.2 OVERVIEW OF MEDIATORS OF ESTROGEN EFFECT ON 61
CORONARY HEART DISEASE
2.3.1 Estrogen effect on lipid metabolism and coagulation factors 61
2.3.2 Estrogen effect on metabolic inflammation 61
2.3.3 Estrogen effect on vascular tissue (endothelium and 62
smooth muscle cell)
2.4 MEDIATOR: LIPIDS 63
2.4.1 Pathway of lipid metabolism 63
2.4.2 Lipids and coronary heart disease 63
2.4.3 Effect of hormone therapy through lipids in coronary 64
heart disease
2.4.4 Effect of estrogens on lipoprotein oxidation 66
2.4.5 Estrogen effect on atherosclerosis through lipids 69
2.4.6 Progesterone effect on atherosclerosis through lipids 69
2.4.7 Measurement of lipids in ELITE and REPLENISH 70
2.5 MEDIATOR: COAGULATION FACTORS 71
2.5.1 Thrombosis 71
2.5.2 Effect of hormone therapy on venous thrombosis 72
2.5.3 Effect of hormone therapy on arterial thrombosis 74
2.5.4 Coagulation pathway, coagulation factors, anti-coagulation 74
factors and interpretation
2.5.5 Effect of hormone therapy on hemostatic system 78
2.5.6 Measurement of coagulation factors in REPLENISH 82
2.6 OUTCOME: ATHEROSCLEROSIS 83
2.6.1 Imaging technique for atherosclerosis 83
2.6.2 Carotid intima media thickness and coronary heart disease 85
2.6.3 Estrogen effect on carotid intima media thickness 97
2.6.4 Measurement of carotid intima media thickness in ELITE 90
2.7 TABLES AND FIGURES 93
2.8 REFERENCES 118
v
CHAPTER 3: DIFFERENTIAL EFFECT OF PLASMA ESTRADIOL LEVELS ACHIEVED 138
WITH HORMONE THERAPY ON THE PROGRESSION OF SUBCLINICAL
ATHEROSCLEROSIS IN EARLY AND LATE POSTMENOPAUSAL WOMEN
3.1 ABSTRACT 139
3.2 INTRODUCTION 141
3.3 MATERIALS AND METHODS 141
3.4 RESULTS 145
3.5 DISCUSSION 147
3.6 CONCLUSION 175
3.7 TABLES AND FIGURES 1
3.8 REFERENCES 159
CHAPTER 4: EFFECT OF ESTRADIOL DOSE AND ESTRADIOL LEVEL ON THE 162
METABOLIC MEASURES IN EARLY AND LATE POSTMENOPAUSAL WOMEN
4.1 ABSTRACT 163
4.2 INTRODUCTION 165
4.3 MATERIALS AND METHODS 166
4.4 RESULTS 169
4.5 DISCUSSION 171
4.6 CONCLUSION 175
4.7 TABLES AND FIGURES 176
4.8 REFERENCES 181
CHAPTER 5: EFFECT OF ESTRADIOL DOSE AND ESTRADIOL LEVEL ON THE 186
COAGULATION MEASURES IN EARLY AND LATE POSTMENOPAUSAL WOMEN
5.1 ABSTRACT 187
5.2 INTRODUCTION 189
5.3 MATERIALS AND METHODS 190
5.4 RESULTS 193
5.5 DISCUSSION 195
5.6 CONCLUSION 198
5.7 TABLES AND FIGURES 199
5.8 REFERENCES 205
CHAPTER 6: FACTORS ASSOCIATED WITH SERUM ESTRADIOL LEVELS AMONG 208
POSTMENOPAUSAL WOMEN USING HORMONE THERAPY
6.1 ABSTRACT 209
6.2 INTRODUCTION 211
6.3 MATERIALS AND METHODS 212
6.4 RESULTS 215
6.5 DISCUSSION 217
6.6 CONCLUSION 221
6.7 TABLES 223
6.8 REFERENCES 235
vi
CHAPTER 7: SUMMARY AND FUTURE WORK 240
7.1 SUMMARY 241
7.2 RELATED WORK 242
7.2.1 Modifying effect of APOE4 genotype on the association 242
between metabolic phenotype and subclinical atherosclerosis
in postmenopausal women
7.2.2 Determinants of attained estradiol levels in response to 243
oral estradiol plus progesterone therapy
7.2.3 Modifying effect of time-since-menopause and age on 244
the association of estradiol dose and estradiol level on
coagulation measures
7.3 FUTURE DIRECTIONS 245
7.3.1 Effect of hormone therapy on atherosclerosis by extent of 245
underlying atherosclerosis at time of hormone therapy initiation
7.3.2 Possible mechanisms of hormone therapy effect on 246
atherosclerosis progression
7.3.3 A new atherosclerosis measure 247
7.3.4 Future studies 247
7.4 REFERENCES 249
vii
LIST OF TABLES
TABLE PAGE
Table 1.1 List of studies exploring the effect of hormone therapy on 27
coronary heart disease (CHD) and atherosclerosis
among healthy postmenopausal women and women with
preexisting coronary heart disease
Table 1.2 Women’s Health Initiative trial results on risk ratio for 29
cardiovascular disease stratified by age and
time-since-menopause
Table 1.3 Summary of primate studies on effect of hormone therapy 30
on coronary atherosclerosis
Table 2.1 Risk factors for thrombosis stratified by Virchow’s triads 94
Table 2.2 Studies evaluating association between hormone therapy 95
and venous thromboembolism (VTE)
Table 2.3 Women’s Health Initiative trial results on risk ratio for 96
venous thrombosis stratified by age and time-since-menopause
Table 2.4 Summary of coagulation proteins/clotting factors 98
Table 2.5 List of studies of effect of hormone therapy on coagulation, 99
anti-coagulation and fibrinolytic factors
Table 3.1 Baseline characteristics of women by 153
time-since-menopause strata
Table 3.2 Mean estradiol level during the trial and change of 154
estradiol level from baseline among total sample and
participants in hormone therapy group by
time-since-menopause strata
Table 3.3 Mixed model linear regression analysis of the association of 155
mean estradiol level during the trial (pg/ml) with carotid intima
media thickness (CIMT) progression rate (µm/year):
by time-since-menopause strata
Table 3.4 Evaluation of the differential estradiol (pg/ml) association with 156
carotid intima media thickness (CIMT) progression rate
(µm/year) by time-since-menopause strata from mixed model
linear regression analysis among total cohort and participants
in hormone therapy group
viii
Table 4.1 Baseline characteristics by postmenopausal strata; 177
early (<6 years-since-menopause) and
late (≥10 years-since-menopause) postmenopause
Table 4.2 Estimated change from baseline of metabolic measures 172
per 0.25 mg increase of E2 dose and 1 pg/ml serum increase of
E2 levels by postmenopausal strata
Table 5.1 Baseline characteristics by postmenopausal strata; 200
early (<6 years-since-menopause) and
late (≥10 years-since-menopause) postmenopause.
Table 5.2 On-trial levels of hormone, coagulation and 201
anti-coagulation measures by postmenopausal strata;
early (<6 years-since-menopause) and
late (≥10 years-since-menopause) postmenopause.
Table 5.3 Estimated change from baseline of coagulation and 202
anti-coagulation measures per 0.25 mg increase of E2 dose
and 1 pg/mL serum increase of E2 levels by postmenopausal strata
Table 6.1 Baseline demographic and clinical characteristics 224
Table 6.2 Association of serum estradiol levels while taking 226
hormone therapy with demographic and clinical
characteristics and median estradiol level by categorical variable
Table 6.3 Association of log serum estradiol levels while taking 229
hormone therapy with demographic and clinical characteristics
Table 6.4 Multivariable association of estradiol levels while taking 232
hormone therapy with demographic and clinical characteristics
among total analysis sample and by postmenopausal strata
Table 6.5 Multivariable association of log-transformed estradiol 233
levels while taking hormone therapy with
demographic and clinical characteristics among
total sample and by postmenopausal strata
Table 6.6 Estimated serum estradiol levels for each multivariable 234
model determinant
ix
LIST OF FIGURES
FIGURE PAGE
Figure 2.1 Structure of equine estrogens 103
Figure 2.2 Estrogen receptors and their pathways 104
Figure 2.3 Estrogen effect on coronary heart disease (CHD) 105
in early atherosclerogenesis versus established atherosclerosis
Figure 2.4 Lipoprotein metabolism pathways 106
Figure 2.5 Role of estrogen in lipid peroxidation in the etiology 107
of atherosclerosis in postmenopausal women
Figure 2.6 Virchow’s triad 108
Figure 2.7 Coagulation pathway tested by prothrombin time (PT) 109
Figure 2.8 Coagulation pathway tested by activated partial 110
thrombin time (APTT)
Figure 2.9 Coagulation pathway tested by fibrinogen (FIB) 111
Figure 2.10 Anticoagulation pathway involving antithrombin 112
Figure 2.11 Anticoagulation pathway involving protein C (PROTC) 113
and protein S (PROTS)
Figure 2.12 Ultrasound image of longitudinal axis of common carotid 114
artery (CCA) demonstrates the carotid intima media thickness
(CIMT) which lies between the lumen-intima and
the media-adventitia interface
Figure 2.13 Forest plot of hazard ratios for coronary heart disease (CHD) 115
A: per 1 SD increment in common carotid artery intima-media
thickness (CIMT), B: per 1 mm increment in CIMT
Figure 2.14 Anatomy of common carotid artery and carotid artery bulb 116
(transverse and longitudinal view)
Figure 2.15 Proper position for ultrasound image for common carotid 117
artery intima-media thickness (CIMT) measurement, which
includes common carotid artery and jugular vein
Figure 3.1 1A Model-estimated carotid artery intima-media thickness (CIMT) 157
progression rates at different quartiles of estradiol level
according to time since menopause strata among total cohort
Figure 3.1 1B Model-estimated carotid artery intima-media thickness (CIMT) 157
progression rate at different quartiles of estradiol level
according to time since menopause strata
among participants in hormone therapy group
x
Figure 4.1 Estimated change from baseline of metabolic measures 179
per 0.25 mg increase in estradiol dose by postmenopausal strata
Figure 4.2 Estimated change from baseline of metabolic measures 180
per 1 pg/ml increase in estradiol levels by postmenopausal strata
Figure 5.1 Estimated change from baseline of coagulation and 203
anti-coagulation measures per 0.25 mg increase in estradiol dose
by postmenopausal strata
Figure 5.2 Estimated change from baseline of coagulation and 204
anti-coagulation measures per 1 pg/mL increase in estradiol level
by postmenopausal strata
xi
ABBREVIATIONS
APC Activated protein C
APOE Apolipoprotein E
APTT Activated partial thromboplastin time
AT Antithrombin
BMI Body mass index
CAC Coronary Artery Calcium
CAS Coronary artery stenosis
CCA Common carotid artery
CEE Conjugated equine estrogen
CHD Coronary heart disease
CI Confidence interval
CIMT Carotid intima media thickness
CT Cardiac computed tomography
CVD Cardiovascular disease
DHEAS Dehydroepiandrosterone sulfate
DOPS Danish Osteoporosis Prevention Study
DVT Deep vein thrombosis
E1 Estrone
E2 Estradiol
E3 Estriol
ECG Electrocardiogram
ELITE Early versus Late Intervention Trial with Estradiol
eNOS Endothelial NO synthase
EPAT Estrogen in Prevention of Atherosclerosis Trial
EPT Combined estrogen-progesterone therapy
ER Estrogen receptors
ERA Estrogen Replacement and Atherosclerosis
ESPRIT oEStrogen in the Prevention of Reinfarction Trial
ET Estrogen therapy
FDP Fibrin degradation products
FIB Fibrinogen
FSH Follicular stimulating hormone
GSM Grey Scale Median
HER Heart and Estrogen-progestin Replacement Study
HDL-C High-density lipoprotein cholesterol
HR Hazard ratio
HT Hormone therapy
IDL-C Intermediate density lipoprotein cholesterol
KEEPS Kronos Early Estrogen Prevention Study
LDL-C Low-density lipoprotein cholesterol
MENQOL Menopause specific quality of life questionnaire
xii
MOS Medical outcomes study
MP Micronized progesterone
MPA Medroxyprogesterone acetate
NETA Norethindrone acetate
NHS Nurses’ Health Study
NO Nitric oxide
P4 Progesterone
PAI-1 Plasminogen activator
PE Pulmonary embolism
PEPI Postmenopausal Estrogen/Progestin Interventions
PHASE Papworth HRT Atherosclerosis Study Enquiry
PHOREA Postmenopausal Hormone Replacement against Atherosclerosis
PR Progesterone receptor
PROTC Protein C
PROTS Protein S
PT Prothrombin time
QCA Quantitative coronary angiography
RCT Randomized controlled trial
RR Relative risk
SCE Synthetic conjugated estrogens
SHBG Sex hormone binding globulin
SPE Soy phytoestrogen
SWAN Study of Women’s Health Across the Nation
TC Total cholesterol
TF Tissue factors
TFPI Tissue factor pathway inhibitor
TG Triglycerides
tPA Tissue plasminogen activator
VLDL-C Very low-density lipoprotein cholesterol
VMS Vasomotor symptom
VTE Venous thromboembolism
WAVE Women’s Angiographic Vitamin and Estrogen
WELL-HART Women’s Estrogen-progestin Lipid-Lowering Hormone Atherosclerosis
Regression Trial
WHI Women’s Health Initiative
WISH Women’s Isoflavone Soy Health
xiii
ABSTRACT
Cardiovascular disease, mainly coronary heart disease (CHD) is the leading cause of death
among women. CHD incidence and CHD-related mortality increase with older age, and
menopausal status is also associated with increased CHD risk, especially among younger
menopausal women with estradiol (E2) deficiency.
Epidemiologic and clinical trial data suggest that hormone therapy (HT), specifically E2
therapy may reduce CHD risk as a primary prevention among healthy postmenopausal women
but not as secondary prevention among women with preexisting CHD. The important discrepancy
of the HT effect on CHD noted between observational studies and the largest randomized clinical
trial, the Women’s Health Initiative trials motivated additional research regarding possible
reasons for this discrepancy. Cumulative evidence from vascular pathology, animal studies, and
human studies suggest that HT has a different effect on CHD, with modification by time of HT
initiation according to age and time-since-menopause. The HT timing hypothesis postulates that
postmenopausal women who initiate HT at younger ages or sooner after menopause have
reduced risk of CHD, CHD-related mortality and all-cause mortality compared with placebo.
This dissertation aims to examine and estimate the differential effect of HT on CHD by
time-since-menopause. We propose that the E2 effect on mediators such as lipid and coagulation
measures differs by time-since-menopause, and hence may explain the different effect of HT on
CHD. Dataset from two completed clinical trials, the Early versus Late Intervention Trial with
Estradiol and the REPLENISH trial, will be used to test the hypotheses. Furthermore, we will
evaluate the factors associated with serum E2 level among postmenopausal women using HT.
This study aims to identify the related factors to E2 level, leading to a practice to achieve the
desirable E2 levels for maximal benefits and minimal risk.
CHAPTER 1
INTRODUCTION AND BACKGROUND
OF
HORMONE THERAPY AND CORONARY HEART DISEASE
Page 2
1.1 CORONARY HEART DISEASE AMONG POSTMENOPAUSAL WOMEN
1.1.1 Incidence of coronary heart disease among postmenopausal women
Cardiovascular disease (CVD) is the leading cause of death, causing an overall of 17.9
million deaths globally in 2015(1). In the United States, national vital statistics based on
information from all death certificates filed in the 50 states and the District of Columbia in 2015
reported that heart disease was the leading cause of death for both males and females of all
ages. CVD mortality was higher in males than females until age 65 years when the CVD mortality
was similar for both genders. Among females, 298,840 deaths from CVD accounted for 22.3% of
all female deaths in the United States(2).
Coronary heart disease (CHD) is the primary cause of death attributable to CVD in the
United States. Atherosclerosis of coronary arteries, narrowing of arteries over time, is an
underlying etiology of CHD. Formation of blood clots after an atherosclerotic plaque rupture may
cause coronary artery occlusion. The initial presentation of CHD is acute coronary syndrome, a
sudden imbalance between myocardial oxygen consumption and demand, which is usually the
result of coronary artery obstruction. Signs and symptoms of acute coronary syndrome include
angina pectoris, ST segment elevation from the electrocardiogram or elevation of cardiac
biomarkers(3). In 2015, more than 360,000 people died of CHD, which accounted for 43.8% of
deaths from CVD in the United States(4). A recent report from the American Heart Association
showed that among females, CHD has the highest mortality rate when compared to other
common causes of death such as stroke, lung cancer and breast cancer; the data were consistent
for all races, with an age-adjusted mortality rate for CHD (per 100,000 population) for Non-
Page 3
Hispanic White females of 71.2, for Non-Hispanic Black females of 86.7, and for Hispanic females
of 56.0(4).
In addition to the fact that CHD incidence increases with age, menopausal status also
increases risk of CHD. Among women within the same age group, postmenopausal women have
higher incidence of CHD than premenopausal women(5). Data from the Framingham Heart Study,
a long-term, population-based cardiovascular cohort, reported incidence of CHD (per 1000 per
year) among women in pre- vs. post-menopause as: 0.2 versus 3.6 at age 40-44 years; 1.1 versus
2.9 at age 45-49 years; and 3.6 versus 4.3 at age 50-54 years(5). In a subsequent report at over
30 years of follow-up of the same Framingham cohort, the age-adjusted risk of CHD in
postmenopausal women was 2-3 times higher than premenopausal women(6). Among
premenopausal women, lower estradiol (E2) level was significantly associated with higher
occurrence of CHD defined by at least 70% stenosis of at least one epicardial coronary artery
even after controlling for age(7).
These data suggest that the elevated risk of CHD with postmenopausal status especially
among younger women could be due to the decline in reproductive sex hormones especially E2
levels after menopause.
1.1.2 Estrogen deficiency as a risk factor for atherosclerosis and CHD
As postmenopausal women have higher risk of CHD than premenopausal women,
elevated CHD risk may in part be attributable to changes in sex hormone levels during
menopause. At menopausal transition, as ovarian function declines, sex hormone levels are
drastically altered. In particular, decreased E2 levels yield an increase in follicular stimulating
hormone (FSH), whereas androgen levels such as total testosterone and
Page 4
dehydroepiandrosterone sulfate (DHEAS) are unchanged due to continuing production from the
adrenal gland(8, 9). A 9-year prospective study of women aged 45-55 years at baseline reported
mean E2 levels of 78 (39-158) pg/ml at 4 years prior to menopause, falling by 60% to 30 (23-42)
pg/ml at the time of menopause and to 10 (8-11) pg/ml at 18 months after menopause(9). At 2-
5 years after menopause, E2 levels were reduced to 5 pg/ml or less(8). The analysis of CHD risk
at 8 years of follow-up in this study showed a higher CHD risk in association with lower E2 level(9).
A meta-analysis including 32 studies showed significant higher CHD risk among women
with premature or early menopause. Compared with women who experienced menopause at 45
years or older, women who experienced menopause at younger than 45 years were 1.50 (95%CI
1.28-1.76) times as likely to develop CHD(10). Another meta-analysis including 10 studies
reported increased risk of developing and dying from CHD among women with primary ovarian
insufficiency, defined as age of menopause less than 40 years with hazard ratio of 1.69 (95%CI
1.29-2.21, p=0.0001)(11). As premature or early menopause and primary ovarian insufficiency
indicate cessation of ovarian function, the increased risk of CHD in these women could be due to
decreased E2 level. Nurses’ Health Study, a prospective cohort study of 73,814 women showed
that a shorter duration of reproductive life span is associated with a higher CVD risk(12). Similar
results of age at menopause and CVD risk in women with surgical and natural menopause in this
study support the hypothesis that alteration in sex hormones, particularly E2, with menopause
might be responsible for the association between reproductive life span and cardiovascular risk.
Atherosclerosis has been used as an indicator for the presence of CHD. Coronary
angiography and carotid ultrasonography (e.g., measuring carotid intima media thickness (CIMT))
have been used to measure atherosclerosis. In the Estrogen in Prevention of Atherosclerosis Trial
Page 5
(EPAT), lower E2 levels were significantly associated with progression of subclinical
atherosclerosis measured by CIMT over 2 years. After controlling for age and body mass index
(BMI), CIMT progression was significantly higher among women with lower total E2, free E2,
estrone (E1) and sex hormone binding globulin (SHBG)(13).
1.2 HORMONE THERAPY AND CORONARY HEART DISEASE
1.2.1 Observational studies of primary prevention
Hormone therapy (HT) is commonly used to treat menopausal symptoms (such as
vasomotor symptoms and the genitourinary syndrome of menopause) as well as to prevent
menopausal bone loss and osteoporosis fracture(14). As HT was postulated to prevent CHD,
many observational studies and clinical trials were conducted to explore this association. Since
the 1980s, approximately 40 observational studies showed that HT is associated with a 30-50%
reduction in CHD and overall mortality in postmenopausal women(15). A meta-analysis of 16
prospective studies comparing women who had ever used HT versus those who had never used
HT showed a reduction in CHD with a pooled relative risk (RR) of 0.70 (95% confidence interval
(CI), 0.63-0.77). When limited to women who ever used HT, those with current use showed a
reduction in CHD with pooled RR of 0.50 (95%CI, 0.45-0.59)(16). These results are consistent
across case-control studies, cross sectional studies and cohort studies.
One of the largest observational studies exploring the association of long-term HT
exposure on CHD was the Nurses’ Health Study (NHS), a prospective cohort study initiated in
1976 to investigate risk factors for chronic disease among 121,701 female registered nurses from
11 US states. Results from 48,470 postmenopausal women age 30-63 years in NHS with up to 10
years of follow-up (337,854 person-years) showed a decreased risk of CHD among women
Page 6
currently taking estrogen, with RR of 0.56 (95%CI, 0.40-0.80)(17). A follow-up reported from
among 59,337 women age 30-55 years of age at baseline with up to 16 years of follow-up showed
a decrease CHD risk among women who took estrogen with progestin (RR=0.39; 95 %CI, 0.19-
0.78) or estrogen alone (RR=0.60; 95%CI, 0.43-0.83), compared with women who did not use
HT(18). At a 20-year follow-up of NHS, current HT users showed a decreased risk of CHD
compared with never users (RR=0.61; 95% CI, 0.52-0.71). The reduction in CHD risk was similar
among women taking different doses of conjugated equine estrogen (CEE) with RR of 0.54
(95%CI; 0.44-0.67) for women taking CEE 0.625 mg daily and RR of 0.58 (95%CI; 0.37-0.92) among
women taking CEE 0.3 mg daily compared with never users(19). In summary, NHS showed
reduced CHD risk among women taking HT either with estrogen alone, estrogen plus progestin
and different doses of estrogen.
1.2.2 Randomized Clinical trials of secondary prevention
Motivated by these strong and consistent findings in observational studies of reduced
CHD associated with postmenopausal HT, in the late 1990s and early 2000s several randomized
clinical trials evaluated the effect of HT on CHD among postmenopausal women with and without
preexisting CHD. These trials used either clinical outcomes of CHD or imaging measures of
atherosclerosis as trial endpoints. Table 1.1 summarizes results from studies on HT as primary
and secondary prevention on CHD.
Trials with CHD clinical outcomes
Regarding the effect of HT on recurrent CHD or progression of coronary atherosclerosis,
all studies reported consistent results that HT does not have a beneficial effect on CHD compared
with placebo (Table 1.1). Studies using recurrence of clinical CHD as an outcome include the Heart
Page 7
and Estrogen-progestin Replacement Study (HERS), the Women’s Estrogen for Stroke Trial
(WEST), the Papworth HRT Atherosclerosis Study Enquiry (PHASE) and the EStrogen in the
Prevention of Reinfarction Trial (ESPRIT).
The HERS trial, aiming to test the effect of HT in secondary prevention of CHD, was a
placebo-controlled randomized controlled trial (RCT) of 2,763 postmenopausal women at a mean
age of 66.7 years with a previous CHD event. Women were randomized to use of CEE 0.625 mg
per day plus medroxyprogesterone acetate (MPA) 2.5 mg per day or placebo. After 4.1 years of
follow-up, CEE plus MPA was not significantly associated with a new CHD event with hazard ratio
(HR) of 0.99 (95%CI, 0.8-1.22)(20).
Similarly, the WEST trial, an RCT of 1 mg 17β-E2 or placebo conducted among 664
postmenopausal women with recent ischemic stroke or transient ischemic attack showed no
effect on CHD RR=1.2 (95%CI, 0.5-2.5) at the mean follow up of 2.8 years(21). This study was
primarily designed to study the role of estradiol therapy as a secondary prevention for
cerebrovascular disease with the hypothesized treatment effect size of 0.60(22).
In the PHASE trial, 255 postmenopausal women with angiographically-proven CHD were
randomized to receive 2 mg per day of transdermal 17β-E2 and 4 mg of cyclic norethindrone
acetate (NETA) or placebo for 4 years. This study estimated that the annual event rate of primary
endpoints would decrease from 18% among placebo group to 12% among treated group. In a
primary composite endpoint of hospitalization for unstable angina, myocardial infarction or
death, HT had no significant effect (RR=1.29, 95%CI, 0.84-1.95) compared with placebo(23).
The ESPRIT was conducted in England and Wales among 1017 postmenopausal women
aged 50-59 years and survived their first myocardial infarction. Women were randomly assigned
Page 8
to 2 mg estradiol valerate daily or placebo for 2 years. The frequency of reinfarction or cardiac
death was similar in women receiving HT compared to those with placebo (RR=0.99, 95%CI, 0.70-
1.41)(24).
Trials using atherosclerosis-imaging outcomes
Studies using progression of coronary atherosclerosis as an outcome include the Estrogen
Replacement and Atherosclerosis (ERA) trial, the Postmenopausal Hormone Replacement against
Atherosclerosis (PHOREA) trial, the Women’s Angiographic Vitamin and Estrogen (WAVE) trial,
and the Women’s Estrogen-progestin Lipid-Lowering Hormone Atherosclerosis Regression Trial
(WELL-HART). Imaging techniques commonly used to measure progression of atherosclerosis
were carotid ultrasound to measure CIMT and coronary angiography.
In ERA, 309 postmenopausal women with a mean age of 66 years who had at least one
epicardial coronary stenosis of at least 30% of the luminal diameter were randomized to one of
three groups of 0.625 mg/day CEE, 0.625 mg/day CEE plus 2.5 mg/day of MPA or placebo. HT did
not significantly alter the progression of coronary atherosclerosis measured by quantitative
angiography compared with placebo during 3.2 years of follow up(25).
The PHOREA trial included 321 healthy postmenopausal women with subclinical
atherosclerosis defined with CIMT of at least 1 mm in at least one of the predefined carotid artery
segments. Women were randomized to placebo or 1 mg 17β-E2 daily with continuous or
sequential 0.025 mg gestodene. The progression of atherosclerosis between baseline and at 48
weeks follow-up among both HT regimen and placebo were not statistically different (26).
The WAVE trial was an RCT of 423 postmenopausal women with at least one 15-75%
coronary stenosis at baseline. Women were randomized to receive 0.625 mg CEE with 2.5 mg
Page 9
MPA per day or placebo and vitamin E and vitamin C or placebo. The trial atherosclerosis outcome
evaluated the annualized mean change in minimal lumen diameter of coronary artery from
coronary angiograms obtained at baseline and at study exit. At a mean follow-up of 2.8 years,
women randomized to HT showed similar coronary progression as placebo (p=0.17)(27).
WELL-HART, a RCT of 226 postmenopausal women with a mean age of 63.5 years with
prior CHD defined by established coronary artery atherosclerosis of at least one coronary artery
lesion occluding at least 30% of the luminal diameter or women with lesion at least 20% of the
luminal diameter who underwent percutaneous transluminal coronary angioplasty or coronary
artery bypass grafting. The study showed that either 1 mg/day of 17β-estradiol (17β-E2) or 17β-
E2 plus 2.5 mg/day of MPA had no significant effect on the progression of atherosclerosis
measured by quantified coronary angiography during 3.3 years follow up(28). In summary, HT
was consistently shown in RCTs to have no benefit in the secondary prevention of CHD nor reduce
or reverse existing atherosclerosis.
1.2.3 Randomized Clinical trials of primary prevention
The Women’s Health Initiative (WHI) study is the largest RCT to test the effect of HT as a
primary prevention (although at least 10% of the cohort had history of previous CVD and were
included in all analyses)(29) of CHD among 16,608 older women distant from menopause in the
CEE plus MPA trial and 10,739 women in the CEE alone trial. In July 2002, there was an early
termination of WHI CEE plus MPA trial due to presumed increased risk of stroke and breast
cancer. The WHI study reporting on postmenopausal women with a mean age of 63.2 years and
average time-since-menopause of 12 years when randomized to CEE plus MPA showed A
significantly increased CHD risk (in women with and without history of CVD) with HR of 1.29
Page 10
(95%CI, 1.02-1.63) during 5.2 years of intervention, however the HR was not significant (HR 1.29;
95%CI, 0.85-1.97) when adjusted for baseline risk factors for cardiovascular disease
(race/ethnicity, education, physical activity, prior hormone use, body mass index, left ventricular
hypertrophy, current smoking, hypertension, treated diabetes and treated high serum
cholesterol level)(30). This report had a great impact on both scientific and clinical communities
regarding the use of HT; clinical use of HT declined dramatically. Subsequently, and contrary to
standards of trial conduct, the WHI CEE trial was terminated by the National Institute of Health
(not the independent the data and safety monitoring board), the funding source of the trial,
earlier than planned at a mean of 6.8 years of intervention. Whereas it was reported that CEE
was not significantly associated with CHD with HR of 0.91 (95%CI, 0.75-1.12) and adjusted HR of
0.91 (95%CI, 0.72-1.15)(31), the effect of CEE on breast cancer risk was reported as statistically
significantly reduced especially in those women who were compliant with CEE therapy(31).
However, an important subsequent report of the HT effect on CHD from the WHI study after
central adjudication of clinical outcomes no longer showed significant risk of HT on CHD with
either HT regimen. Women using CEE plus MPA trial had overall HR of 1.24 (95%CI, 1.00-1.54)(32)
and women using CEE had overall HR of 0.95 (95%CI, 0.79-1.16)(33) compared with placebo.
1.2.4 Discrepancy of results from observational studies and clinical trials
It is obvious that results from previous observational studies and the WHI trial regarding
the association between HT and CHD are discordant. While meta-analysis of observational
studies demonstrate that postmenopausal HT is associated with a decreased risk of CHD by 30-
50%, the WHI trial with CEE plus MPA first reported a significant increased risk of CHD by
approximately 30%.
Page 11
As RCTs are thought to be the gold standard of study types over any types of observational
studies, it does not necessary mean that result from WHI study is more valid than observational
studies just because it is a large RCT. While larger sample size could reduce random error, it does
not reduce systematic error such as selective drop out and misclassification. Therefore, the
validity of study results could not be inferred based purely on study design(34). Instead, to
compare and contrast results of studies on a similar research question, we need to carefully
examine the study methods, population, interpretation and inference of results. Regarding the
association between HT and CHD, the different results from observational studies and RCT, in
particular the WHI trial, may be related to a possible modifying effects of timing of HT initiation
relative to age and time since menopause and HT regimen of estrogen (alone or combined
estrogen and progestin).
In observational studies, HT was commonly initiated around or before menopause, to
treat menopause-related symptoms. As the mean age of menopause for women in the United
States is 51 years (35), it is obvious that HT initiation in observational studies was at a far younger
age and close to menopause than women in WHI trial (with a mean age of 63.2 years and 63.6
years for CEE plus MPA trial and CEE alone trial, respectively) who were on average 12 years-
since-menopause when started on HT. In fact, WHI was not designed to test the HT hypothesis
on CVD prevention in younger women who were recently postmenopausal. As reviewed by the
institute of medicine, the HT effect in preventing CVD in young women who were started on HT
close to menopause was conclusive and that the question remaining was whether the effect of
HT in preventing CVD would have similar effects in older women more distant from menopause.
As such, WHI was designed to test HT CVD prevention hypothesis in older postmenopausal
Page 12
women. In addition, WHI was not a primary prevention trial in that the study included at least
10% of women with pre-existing CVD. Further, WHI analyses have never been conducted with
censoring of the women with pre-existing CVD. As such, all data/analyses from WHI that relates
to CVD as presented in the light of primary prevention must be questioned.
A comparison of results on the association of HT use and CHD from the WHI RCTs and the
NHS, a large prospective cohort study, suggests that the discrepancies may be largely explained
by differences in the distribution of time since menopause and length of follow-up (36). In the
WHI trial, HT was assigned to women with a mean age of 63.2 years while in the NHS HT was
initiated at younger age and closer to time-since-menopause. In the NHS, women were at 30-55
years at enrollment and nearly 80% of HT users initiated HT within 2 years of menopause, which
occurred on average at 50 years. Although the overall results from NHS showed a decreased risk
of CHD(17-19) while WHI showed an increased risk of CHD(30), when restricted to women in
comparable age group at HT initiation and using similar HT formulation as women in the WHI
trial, the results were consistent(36). For use of CEE alone, both NHS and WHI showed a
decreased CHD risk among women age 50-59 years and within 10 years since menopause.
Although, use of CEE plus MPA among women within 10 years since menopause showing no
significant risk of CHD were consistently reported by both studies, NHS showed a decreased risk
while WHI showed an increased risk of CHD among women age 50-59 years(36). Using data from
NHS to imitate the WHI trial design and intention to treat analysis of HT on CHD, women within
10 years after menopause showed a reduced HR for CHD among HT users was 0.54 (95%CI, 0.19-
1.51), whereas in women with 10 years or more after menopause the HR among HT users was
1.20 (95%CI, 0.78-1.84) compared to non-users (p for interaction=0.01)(37). The comparison of
Page 13
results from NHS and WHI suggested that the discrepancies of results from observational studies
and RCT could be attributed to the differences in the age structure of study populations and age
at HT initiation relative to menopause.
1.2.5 Women’s Health Initiative trial stratified results
To evaluate the effect of HT on CHD based on HT initiation according to age and time-
since-menopause, WHI investigators reported trial results stratified by 10-year age group and by
10-year time-since-menopause category (Table 1.2). When HT was initiated in women less than
60 years or less than 10 years since menopause, CHD risk tend to be reduced with both HT
regimens. In a combined analysis of data from the CEE plus MPA and CE alone trials, CHD risk was
reduced compared to placebo when HT was initiated among women 50-59 years old (HR=0.93,
95%CI, 0.65-1.33) and within 10 years since menopause (HR=0.76, 95%CI, 0.50-1.16) with a
significant trend in decreasing risk with shorter time-since-menopause (p trend=0.02) (32).
For CEE plus MPA, centrally-adjudicated results of CHD risk with HT based on time-since-
menopause categories of 10 years, 10-19 years and 20 years or more estimated HRs of 0.89, 1.22
and 1.77, respectively(29). Although the 2 degree of freedom test of modification by time-since-
menopause was not significant (p interaction=0.33), the trend test of higher risk with longer time
since menopause was significant (p trend=0.036) without reporting the effect size(38). A
subsequent report showed HRs for CHD as 0.88, 1.23 and 1.66 with 10 years, 10-19 years and 20
years or more since menopause with marginally significant trend (p trend=0.05)(32). A recent
report of WHI data after a cumulative follow-up of 13.2 years showed that during the active
intervention phase of 5.6 years, HRs for CHD with 10 years, 10-19 years and 20 years or more
since menopause were 0.90, 1.19 and 1.52 with p trend=0.08(39). In summary, all WHI reports
Page 14
were consistent that there is a reduced risk of CHD when CEE plus MPA is initiated close to
menopause with an increasing risk with longer time-since-menopause when initiated.
When the analysis was stratified by 10-year age group, there was no significant risk of
CHD with use of CEE plus MPA among 50-59 years old women, while there was a significant risk
among 70-79 year-old women. CHD risk for women 50-59, 60-69 and 70-79 years old during 5.6
years of intervention phase were HR=1.34 (95%CI, 0.82-2.19), HR=1.01 (95%CI, 0.73-1.39), and
HR=1.31 (95%CI, 0.93-1.84) and risk during 13.2 years of cumulative follow-up were HR=1.25
(95%CI, 0.88-1.76), HR=0.99 (95%CI, 0.80-1.24), and HR=1.34 (95%CI, 1.05-1.72), respectively
with non-significant trend with older age(39). Combining all results from WHI CEE plus MPA trial,
this regimen of HT was not significantly associated with CHD risk.
For CEE alone, CHD risk during intervention based on time-since-menopause of 10 years,
10-19 years and 20 years or more were HR=0.48 (95%CI, 0.20-1.17), HR=0.96 (95%CI, 0.64-1.44),
and HR=1.12 (95%CI, 0.86-1.46), respectively(32). A recent report of WHI data after cumulative
follow-up of 13.2 years showed that during the intervention phase, HRs for CHD with 10 years,
10-19 years and 20 years or more since menopause were 0.50, 1.00 and 1.08 with a non-
significant trend (p trend=0.16)(39).
Stratifying the analysis by 10-year age group, there was no significant risk of CHD with use
of CEE alone among any age group. Centrally-adjudicated results of CHD risk for women in CEE
trial showed no significant association between CEE and CHD when compared with placebo with
HR=0.95 (95%CI, 0.79-1.16) during 7.1 years of follow-up(33). Stratified CEE effect on CHD
compared with placebo by age 50-59, 60-69 and 70-79 years were HR=0.63 (95%CI, 0.36-1.08),
HR=0.94 (95%CI, 0.71-1.24), and HR=1.11 (95%CI, 0.82-1.52)(33). At 10.7 years follow-up, CEE
Page 15
showed a significant decreased CHD risk among women 50-59 years old (HR=0.59, 95%CI, 0.38-
0.90) with marginal significant trend (p trend=0.05)(40). Recent report on CHD risk during 13
years of cumulative follow-up were HR=0.60 (95%CI, 0.39-0.91), HR=1.03 (95%CI, 0.82-1.31), and
HR=1.25 (95%CI, 0.95-1.65), respectively with significant trend (p trend=0.007)(39). In summary
from the WHI CEE alone trial, CEE was shown to reduce CHD risk if initiated among 50-59 years
old women, while had no significant effect among women with older age(33).
1.2.6 Randomized Clinical trials of primary prevention of coronary heart disease among
younger women
In addition to the evidence of a possible differential effect of HT in younger versus older
women when stratified by age and time since menopause from meta-analyses of the cumulated
data, including the WHI trials, several studies have reported a reduced or no significant risk of
CHD and atherosclerosis with HT use among healthy younger postmenopausal women (Table
1.1).
Trials with CHD clinical outcomes
The Danish Osteoporosis Prevention Study (DOPS) was an RCT specifically designed to test
the HT effect on clinical endpoints in 1006 healthy recently postmenopausal women with a mean
age of 49.7 years and mean time since menopause of 7 months. The monthly HT regimen for
women with an intact uterus was 2 mg synthetic 17β-E2 for 12 days, 2 mg 17β-E2 plus NETA for
10 days and 1 mg 17β-E2 for 6 days. The monthly HT regimen for women without a uterus was 2
mg synthetic 17β-E2 daily. The mean duration for the intervention phase was 10.1 years and the
mean cumulative follow-up was 15.8 years. In DOPS, HT significantly reduced CHD (defined as
hospitalization for myocardial infarction or heart failure) compared to placebo, with HR=0.48
Page 16
(95%CI, 0.27-0.89) during the intervention phase and HR=0.61 (95%CI, 0.39-0.94) over the
cumulative follow-up(41).
Trials using atherosclerosis-imaging outcomes
The Estrogen in the Prevention of Atherosclerosis Trial (EPAT) tested the effect of HT as
primary prevention of atherosclerosis progression among 222 healthy postmenopausal women
(who did not have prior CHD) with a mean age of 62.2 years. Women were randomized to 1
mg/day of 17β-E2 compared or placebo. With CIMT measured every 6 months for 2 years,
women randomized to HT showed significantly lower atherosclerosis progression compared to
placebo (p=0.046)(42).
The Kronos Early Estrogen Prevention Study (KEEPS) placebo-controlled RCT was designed
to explore the effect of HT on atherosclerosis progression among women in early
postmenopause. KEEPS was conducted among healthy postmenopausal women 42-58 years old
who were between 6 and 36 months from their last menses with an intervention of low dose oral
CEE (0.45 mg/day of CEE) with placebo patch, 17β-E2 patch (50 mcg/day) with placebo tablet, or
placebo patch and placebo tablet during 4 years follow-up. This trial reported no significant effect
of any HT regimen on CIMT progression compared to placebo(43).
According to this evidence from observational studies and RCTs with long follow up
period, the effect of HT on CHD seem to be modified by timing of HT initiation according to time
since menopause. In younger postmenopausal women, HT tends to reduce CHD risk, while in
older postmenopausal women; HT has no effect or may have increased CHD risk.
1.3 HORMONE THERAPY TIMING HYPOTHESIS
From existing evidence of a possible differential effect of HT according to time since
Page 17
menopause, animal and human studies have explored the HT effect on CHD, in different
experimental groups defined by age and menopausal status.
1.3.1 Pathology of vascular aging
The pathology of vascular aging shows that coronary artery atherosclerosis progresses
with both age and menopausal status. The most active progression of atherosclerosis lesions
occurs after menopause and by the time of 60-65 years of age when around 60% of women had
complicated atherosclerosis, defined as the extent of coronary calcium score of 90 percentile or
higher while this occurs in 20% among women less than 60 years old(44). Although HT has been
shown to have a beneficial effect on a healthy vascular wall, the beneficial HT effect may be lost
or even become adverse in vessels with existing atherosclerosis lesions(45).
While in vitro experiments showed that estrogen can induce vasodilatation in healthy
endothelium, HT also has been shown to stimulate vasoconstriction in endothelium with
atherosclerotic plaque(46). Moreover, estrogen can potentially cause destabilization and rupture
of complicated atherosclerotic lesion(47) as estrogen was reported to increase inflammatory
marker levels such as matrix metalloproteinase(48) and C-reactive protein among
postmenopausal women(49). An in vitro study on role of annexin II in E2-induced matrix
metalloproteinase-9 activity(50). Annexin II is a receptor for tissue plasminogen activator and
plasminogen for the conversion to plasminogen then matrix metalloproteinase-9. Finding from
the study revealed that annexin II plays a central role in modulating matrix metalloproteinase-9
in response to E2. These E2 effect can be inhibited by simvastatin leading to increased matrix
metalloproteinase-9 activity hence relates to plaque destabilization(50). The study suggested
that E2 therapy could be both atheroprotective or atherogenic depending on the absence or
Page 18
presence of atherosclerosis in the individual women. Despite the increased C-reactive protein
among postmenopausal women using estrogen, this activation may be associated with CHD risk
independently of a systemic proinflammatory effects(51). This speculation was confirmed by the
finding form EPAT showing no association between C-reactive protein with atherosclerosis
progression despite the increased C-reactive protein levels among women with E2 therapy
compared with placebo(52).
It has been hypothesized that the underlying vascular heath may have an impact on the
effect of HT, with a beneficial effect on healthy arterial wall and a null or even adverse effect in
atherosclerotic arteries(45). Healthy endothelium hypothesis in relation to human
atherosclerosis was first proposed to explain divergent outcomes of E2 on atherosclerosis in
women with (WELL-HART) and without (EPAT) preexisting cardiovascular disease. Timing of HT
intervention relative to the stage of atherosclerosis may indicate these different effect of E2 on
atherosclerosis(28).
The timing, critical window or window of opportunity hypothesis of HT posits that HT
benefits and risks depend on the temporal initiation of HT relative to time since menopause, age,
or both, which are in turn related to the health of the underlying vascular tissue or to some other
factors such as age-related reduction in estrogen receptors(53).
1.3.2 Mouse studies
Experimental studies in mice showed that E2 inhibited the initiation of atherosclerosis but
not the progression of established atherosclerotic lesions. In a study testing the effect on 17β-E2
versus placebo on established atherosclerosis lesion among apolipoprotein E deficient mice,
which reproducibly develop advanced atherosclerotic lesions at several sites. The E2 did not
Page 19
cause the regression of established atherosclerotic lesions in carotid artery, aortic arch and
thoracic aorta but prevented the initiation of new lesions in the abdominal aorta, iliac, femoral
and popliteal arteries. E2 did not alter the progression of established lesions through stages of
instability and healing(54).
1.3.3 Primate studies
Primate studies showed results consistent with mouse studies that underlying vascular
health and timing of estrogen administration modify its association with atherosclerosis. Female
cynomolgus monkeys were induced into menopause by surgical ovariectomy (removal of ovaries
to induce menopause status) and HT or placebo was administered at different stages of
menopause. Results from this study series showed that when E2 was administered at the time of
menopause, size of atherosclerotic lesion was reduced compared to baseline or control group,
whereas, E2 had no effect when administered in 2 years after menopause (equivalent to 6 years
in human) when there were atherosclerotic lesions(55-58) (Table 1.3).
Among surgically postmenopausal monkeys previously fed a healthy diet and with little
or no atherosclerosis at the time of menopause, estrogen therapy significantly reduced vascular
lesion size by 70% compared with untreated monkeys after 2-3 years follow up(56, 57). A study
randomized 91 postmenopausal cynomolgus monkeys (previously fed with atherogenic diet for
4 months before menopause) into untreated, CEE (equivalent to 0.625 mg dose in human), MPA
(equivalent to 2.5 mg dose in human), or CEE plus MPA groups. After 30 months of treatment
along with atherogenic diet, average coronary artery plaque size was significantly reduced among
monkeys in the CEE group compared with those in the untreated group (72% reduction, p<0.004).
The plaque size of monkeys in the MPA and CEE plus MPA groups did not differ from those in the
Page 20
untreated group(56). Another study on 103 postmenopausal monkeys randomized to placebo,
CEE, low dose raloxifene (equivalent to 1 mg/kg dose in human), or high dose raloxifene
(equivalent to 5 mg/kg dose in human) showed that CEE significantly reduced percent difference
in mean coronary artery atherosclerosis from placebo group by 70% (p=0.007). Neither low dose
or high dose raloxifene differed from placebo on coronary atherosclerosis when compared(57).
These two studies suggested that CEE significantly reduced coronary atherosclerosis when
started at the time of menopause despite the consumption of an atherogenic diet, while MPA
and raloxifene did not. Moreover, the benefit of CEE in atherosclerosis reduction may be lost with
the addition of MPA.
A study among postmenopausal monkeys with existing atherosclerosis showed that the
benefit of CEE was not as strong compared with monkeys who were free of atherosclerosis. This
study was conducted in 189 monkeys previously fed an atherogenic diet for 26 months before
menopause, when they were randomly assigned to CEE, soy phytoestrogen (SPE) or control,
which received isolated soy protein that had been alcohol washed to remove the SPEs for 36
months starting at the time of menopause. The mean coronary plaque size was significantly
reduced among monkeys in the CEE group by approximately 50% compared to control
(p=0.0002), while the plaque size among those in the SPE and control groups were similar(58).
Another study showed that CEE had no effect on coronary atherosclerosis when the
administration of CEE was delayed after the time of menopause. This study was conducted
among 88 postmenopausal monkeys fed an atherogenic diet for 2 years after menopause
(equivalent to 6 years in humans) before they were randomly assigned to a lipid lowering diet,
lipid lowering diet plus CEE or lipid lowering diet plus CEE and MPA. At 30 months of intervention,
Page 21
the mean coronary artery plaque sizes showed no significant differences among groups
(p=0.55)(55).
In summary, the results from these studies showed that CEE significantly reduced
coronary atherosclerosis among postmenopausal monkeys when started at the time of
menopause, however, this beneficial effect of CEE was not as strong among postmenopausal
monkeys with existing atherosclerotic lesions. In addition, there was no significant effect of CEE
on atherosclerosis when the treatment was delayed until later years after menopause.
1.3.4 Human data
Consistent with animal studies, different time of HT initiation was reported to have
different effect on CHD as shown in EPAT and WELL-HART trials(28, 42). A meta-analysis of 23
RCTs including WHI trials with 39,049 participants followed for 191,340 person-years reported
associations between HT and CHD stratified by younger and older postmenopausal women
defined with a cut point of time-since-menopause of 10 years or age of 60 years(59). The results
showed that HT significantly reduced CHD events in younger women with odds ratio (OR) of 0.68
(95%CI, 0.48-0.96). Among older women, HT increased CHD events in the first year then reduced
CHD events after 2 years. Overall, HT had no significant effect on CHD in older women with OR
of 1.03 (95%CI, 0.91-1.16) when compared with placebo(59).
A Cochrane review including data from 19 RCTs including WHI trials with a total of 40,410
postmenopausal women reported a reduced CHD risk among younger postmenopausal women
(less than 10 years after menopause or less than 60 years old). The risk ratio (RR) for CHD among
younger women was 0.52 (95%CI, 0.29-0.96) and among older women (10 years or more after
menopause or 60 years old or more) was 1.07 (95%CI, 0.96-1.20)(60). This study also reported a
Page 22
significant reduction in all-cause mortality among younger women who are taking HT (RR=0.70,
95%CI 0.52-0.95) with no significant effect among older women (RR=1.06, 95%CI, 0.95-1.18)(60).
The reduction in all-cause mortality from HT among young postmenopausal women was also
shown in another meta-analysis of 30 RCTs with 26,708 women. HT reduced mortality in younger
age group (OR=0.61, 95%CI, 0.39-0.95) but not in the older age group (OR=1.03, 95%CI, 0.90-
1.18)(61).
Early versus Late Intervention Trial with Estradiol (ELITE) study(62) was specifically
designed to test that the cardiovascular effects of postmenopausal HT vary with the timing of HT
initiation showed that the effect of E2 therapy on subclinical atherosclerosis statistically differed
between women with early (within 6 years of menopause) and late postmenopause (with 10
years or more after menopause) (interaction p=0.007). This study is a RCT conducted among 643
healthy postmenopausal women to identify the effect of daily 1 mg oral 17β-E2 with vaginal P4
gel 45 mg/day 10 days a month (among women with intact uterus) on progression of subclinical
atherosclerosis measured by CIMT when compared with placebo at median follow up of 4.8
years. Among early postmenopausal women, the mean CIMT increased by 0.0044 mm per year
versus 0.0078 mm per year in the estradiol versus placebo group(p=0.008). Among late
postmenopausal women, the mean CIMT increased by 0.0100 mm per year versus 0.0088 mm
per year in the estradiol versus placebo group(p=0.29). In summary, ELITE showed that E2
therapy was significantly associated with less progression of subclinical atherosclerosis than
placebo when it was initiated within 6 years of menopause but had no effect when it was initiated
10 years or more after menopause.
Page 23
An analysis from WHI CEE trial studied HT effect on coronary artery calcium (CAC) as a
marker for atheromatous-plaque burden known as a late manifestation of atherosclerosis.
Among postmenopausal women who were 50-59 years old at randomization, mean CAC was
significantly lower in women assigned to CEE compared with placebo. This study showed that
CEE reduced CAC while no other therapy including lipid lowering medication showed no effect
on CAC in human or in non-human primate studies.
All relevant data including animal studies, observational studies and clinical trials
consistently support the HT timing hypothesis, which postulates that women who initiate HT at
younger age or sooner after menopause have reduced risk of CHD and related mortality
compared with placebo. Although there is no study to identify specific cut point for the timing of
HT initiation, which will result in beneficial, null or harmful effect, cumulative data from primate
studies and meta-analysis of trial among postmenopausal women suggest that a window of
opportunity for HT in reduction of CHD is when it is initiated before 60 years and/or within 10
years since menopause(15).
1.4 STUDIES OBJECTIVES
The Overall theme of this dissertation are to explore the effect of HT, particularly the E2
effect on CHD risk by exploration of the association of E2 with atherosclerosis through changes
in metabolic and coagulation measures. The analyses were conducted by using data from ELITE
(62) and REPLENISH trial(63, 64).
All three studies will focus on whether the associations between E2 with atherosclerosis,
metabolic and coagulation measures are different by time of HT initiation. Early postmenopausal
women are defined by women within 6 years since menopause. Late postmenopausal women
Page 24
are defined by women with 10 years or more since menopause. These cut points were based on
animal studies data(55) and previous stratified analysis(59).
CHD commonly occurs through a process involving formation of plaque in a coronary
vessel causing stenosis in combination with thrombus formation leading to coronary vessel
occlusion. Overall, we aim to study the effect of HT, particularly the E2 effect on CHD risk by
exploration of association of E2 with atherosclerosis through metabolic and coagulation
measures.
The first study, “Differential Effect of Plasma Estradiol Levels Achieved with Hormone
Therapy on the Progression of Subclinical Atherosclerosis in Early and Late Postmenopausal
Women” aims to evaluate whether there is a differential association between plasma E2 levels
and progression of subclinical atherosclerosis in early versus late postmenopausal women using
ELITE data.
The second study is “Effect of Estradiol Dose and Estradiol Level on the Metabolic and
Coagulation Measurements in Early and Late Postmenopausal Women” aims to evaluate whether
there is a differential association between E2 dose and the change in metabolic/coagulation
measures and to evaluate whether there is a differential association between E2 levels and the
change in metabolic/coagulation measures in early and late postmenopausal women using
REPLENISH trial data.
The third study, Determinants of Serum Estradiol Levels Among Postmenopausal Women
Using Hormone Therapy” aims to identify factors associated with serum E2 levels among healthy
postmenopausal women using HT in ELITE data.
Page 25
In summary, the combined results from all three studies will potentially explain the
association between HT and CHD through E2 effects on metabolic and coagulation measures,
which in turn affect atherosclerosis. The results of these studies will be investigated as to whether
these effects differ between early and late postmenopausal women.
Page 26
1.5 TABLES AND FIGURES
Page 27
Table 1.1 List of studies exploring the effect of hormone therapy on coronary heart disease (CHD) and atherosclerosis among healthy
postmenopausal women and women with preexisting coronary heart disease
*CDH=coronary heart disease, TSM=Time-since-menopause, CEE=conjugated equine estrogen, MPA=medroxyprogesterone acetate,
E2=estradiol, P4=progesterone, NETA=, MI=myocardial infarction, HF=heart failure, TIA=transient ischemic attack, LDL=low density
Page 28
lipoprotein cholesterol, CIMT=carotid intima media thickness, NHS=Nurses’ Health Study, WHI= Women’s Health Initiative study,
DOPS=Danish Osteoporosis Prevention Study, EPAT=Estrogen in the Prevention of Atherosclerosis Trial, KEEPS=Kronos Early Estrogen Prevention
Study, ELITE=Early versus Late Intervention Trial with Estradiol, HERS=Heart and Estrogen-progestin Replacement Study, WEST=Women’s
Estrogen for Stroke Trial, PHASE=Papworth HRT Atherosclerosis Study Enquiry, ESPRIT=EStrogen in the Prevention of Reinfarction Trial,
ERA=Estrogen Replacement and Atherosclerosis, PHOREA=Postmenopausal Hormone Replacement against Atherosclerosis, WAVE=Women’s
Angiographic Vitamin and Estrogen, WELL-HART=Women’s Estrogen-progestin Lipid-Lowering Hormone Atherosclerosis Regression Trial
Page 29
Table 1.2 Women’s Health Initiative trial results on risk ratio for cardiovascular disease stratified by age and time-since-menopause
*CEE=conjugated equine estrogen, MPA=medroxyprogesterone acetate, TX=treatment period, FU=follow-up period, HR=hazard
ratio, yellow hi-lights represent statistically significant
Page 30
Table 1.3 Summary of primate studies on effect of hormone therapy on coronary atherosclerosis
*CEE=conjugated equine estrogen, MPA=medroxyprogesterone acetate
Page 31
1.6 REFERENCES
1. Global Burden of Disease Study 2015 Results. Seattle, WA: Institute for Health Metrics
and Evaluation (IHME), University of Washington; 2016. ; 2016.
(http://ghdx.healthdata.org/gbd-results-tool.). (Accessed).
2. Heron M. Deaths: Leading Causes for 2015. Natl Vital Stat Rep 2017;66(5):1-76.
3. Amsterdam EA, Wenger NK, Brindis RG, et al. 2014 AHA/ACC Guideline for the
Management of Patients with Non-ST-Elevation Acute Coronary Syndromes: a report of
the American College of Cardiology/American Heart Association Task Force on Practice
Guidelines. J Am Coll Cardiol 2014;64(24):e139-e228.
4. Benjamin EJ, Virani SS, Callaway CW, et al. Heart Disease and Stroke Statistics-2018
Update: A Report From the American Heart Association. Circulation 2018;137(12):e67-
e492.
5. Kannel WB, Hjortland MC, McNamara PM, et al. Menopause and risk of cardiovascular
disease: the Framingham study. Annals of internal medicine 1976;85(4):447-52.
6. Kannel WB. Metabolic risk factors for coronary heart disease in women: perspective
from the Framingham Study. American heart journal 1987;114(2):413-9.
7. Bairey Merz CN, Johnson BD, Sharaf BL, et al. Hypoestrogenemia of hypothalamic origin
and coronary artery disease in premenopausal women: a report from the NHLBI-
sponsored WISE study. J Am Coll Cardiol 2003;41(3):413-9.
Page 32
8. Burger HG, Dudley EC, Hopper JL, et al. Prospectively measured levels of serum follicle-
stimulating hormone, estradiol, and the dimeric inhibins during the menopausal
transition in a population-based cohort of women. The Journal of clinical endocrinology
and metabolism 1999;84(11):4025-30.
9. Guthrie JR, Dennerstein L, Taffe JR, et al. The menopausal transition: a 9-year
prospective population-based study. The Melbourne Women's Midlife Health Project.
Climacteric : the journal of the International Menopause Society 2004;7(4):375-89.
10. Muka T, Oliver-Williams C, Kunutsor S, et al. Association of Age at Onset of Menopause
and Time Since Onset of Menopause With Cardiovascular Outcomes, Intermediate
Vascular Traits, and All-Cause Mortality: A Systematic Review and Meta-analysis. JAMA
Cardiol 2016;1(7):767-76.
11. Roeters van Lennep JE, Heida KY, Bots ML, et al. Cardiovascular disease risk in women
with premature ovarian insufficiency: A systematic review and meta-analysis. European
journal of preventive cardiology 2016;23(2):178-86.
12. Ley SH, Li Y, Tobias DK, et al. Duration of Reproductive Life Span, Age at Menarche, and
Age at Menopause Are Associated With Risk of Cardiovascular Disease in Women.
Journal of the American Heart Association 2017;6(11).
13. Karim R, Hodis HN, Stanczyk FZ, et al. Relationship between serum levels of sex
hormones and progression of subclinical atherosclerosis in postmenopausal women. The
Journal of clinical endocrinology and metabolism 2008;93(1):131-8.
14. The NHTPSAP. The 2017 hormone therapy position statement of The North American
Menopause Society. Menopause 2017;24(7):728-53.
Page 33
15. Hodis HN, Collins P, Mack WJ, et al. The timing hypothesis for coronary heart disease
prevention with hormone therapy: past, present and future in perspective. Climacteric :
the journal of the International Menopause Society 2012;15(3):217-28.
16. Grodstein F, Stampfer M. The Epidemiology of Coronary Heart-Disease and Estrogen
Replacement in Postmenopausal Women. Progress in Cardiovascular Diseases
1995;38(3):199-210.
17. Stampfer MJ, Colditz GA, Willett WC, et al. Postmenopausal estrogen therapy and
cardiovascular disease. Ten-year follow-up from the nurses' health study. The New
England journal of medicine 1991;325(11):756-62.
18. Grodstein F, Stampfer MJ, Manson JE, et al. Postmenopausal estrogen and progestin use
and the risk of cardiovascular disease. The New England journal of medicine
1996;335(7):453-61.
19. Grodstein F, Manson JE, Colditz GA, et al. A prospective, observational study of
postmenopausal hormone therapy and primary prevention of cardiovascular disease.
Annals of internal medicine 2000;133(12):933-41.
20. Hulley S, Grady D, Bush T, et al. Randomized trial of estrogen plus progestin for
secondary prevention of coronary heart disease in postmenopausal women. Heart and
Estrogen/progestin Replacement Study (HERS) Research Group. Jama 1998;280(7):605-
13.
21. Viscoli CM, Brass LM, Kernan WN, et al. A clinical trial of estrogen-replacement therapy
after ischemic stroke. New England Journal of Medicine 2001;345(17):1243-9.
Page 34
22. Kernan WN, Brass LM, Viscoli CM, et al. Estrogen after ischemic stroke: clinical basis and
design of the Women's Estrogen for Stroke Trial. J Stroke Cerebrovasc Dis 1998;7(1):85-
95.
23. Clarke SC, Kelleher J, Lloyd-Jones H, et al. A study of hormone replacement therapy in
postmenopausal women with ischaemic heart disease: the Papworth HRT
Atherosclerosis Study. Bjog-an International Journal of Obstetrics and Gynaecology
2002;109(9):1056-62.
24. Cherry N, Gilmour K, Hannaford P, et al. Oestrogen therapy for prevention of
reinfarction in postmenopausal women: a randomised placebo controlled trial. Lancet
2002;360(9350):2001-8.
25. Herrington DM, Reboussin DM, Brosnihan KB, et al. Effects of estrogen replacement on
the progression of coronary-artery atherosclerosis. New England Journal of Medicine
2000;343(8):522-9.
26. Angerer P, Stork S, Kothny W, et al. Effect of oral postmenopausal hormone
replacement on progression of atherosclerosis : a randomized, controlled trial.
Arteriosclerosis, thrombosis, and vascular biology 2001;21(2):262-8.
27. Waters DD, Alderman EL, Hsia J, et al. Effects of hormone replacement therapy and
antioxidant vitamin supplements on coronary atherosclerosis in postmenopausal
women - A randomized controlled trial. Jama-Journal of the American Medical
Association 2002;288(19):2432-40.
Page 35
28. Hodis HN, Mack WJ, Azen SP, et al. Hormone therapy and the progression of coronary-
artery atherosclerosis in postmenopausal women. New England Journal of Medicine
2003;349(6):535-45.
29. Manson JE, Hsia J, Johnson KC, et al. Estrogen plus progestin and the risk of coronary
heart disease. The New England journal of medicine 2003;349(6):523-34.
30. Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus
progestin in healthy postmenopausal women - Principal results from the Women's
Health Initiative randomized controlled trial. Jama 2002;288(3):321-33.
31. Anderson GL, Limacher M, Assaf AR, et al. Effects of conjugated equine estrogen in
postmenopausal women with hysterectomy: the Women's Health Initiative randomized
controlled trial. Jama 2004;291(14):1701-12.
32. Rossouw JE, Prentice RL, Manson JE, et al. Postmenopausal hormone therapy and risk of
cardiovascular disease by age and years since menopause. Jama 2007;297(13):1465-77.
33. Hsia J, Langer RD, Manson JE, et al. Conjugated equine estrogens and coronary heart
disease: the Women's Health Initiative. Arch Intern Med 2006;166(3):357-65.
34. Rothman KJ. Six persistent research misconceptions. J Gen Intern Med 2014;29(7):1060-
4.
35. Nichols HB, Trentham-Dietz A, Hampton JM, et al. From menarche to menopause:
trends among US Women born from 1912 to 1969. Am J Epidemiol 2006;164(10):1003-
11.
Page 36
36. Bhupathiraju SN, Grodstein F, Rosner BA, et al. Hormone Therapy Use and Risk of
Chronic Disease in the Nurses' Health Study: A Comparative Analysis With the Women's
Health Initiative. Am J Epidemiol 2017;186(6):696-708.
37. Hernan MA, Alonso A, Logan R, et al. Observational studies analyzed like randomized
experiments: an application to postmenopausal hormone therapy and coronary heart
disease. Epidemiology 2008;19(6):766-79.
38. Grodstein F, Manson JE, Stampfer MJ. Hormone therapy and coronary heart disease: the
role of time since menopause and age at hormone initiation. Journal of women's health
2006;15(1):35-44.
39. Manson JE, Chlebowski RT, Stefanick ML, et al. Menopausal Hormone Therapy and
Health Outcomes During the Intervention and Extended Poststopping Phases of the
Women's Health Initiative Randomized Trials. Jama-Journal of the American Medical
Association 2013;310(13):1353-68.
40. LaCroix AZ, Chlebowski RT, Manson JE, et al. Health outcomes after stopping conjugated
equine estrogens among postmenopausal women with prior hysterectomy: a
randomized controlled trial. Jama 2011;305(13):1305-14.
41. Schierbeck LL, Rejnmark L, Tofteng CL, et al. Effect of hormone replacement therapy on
cardiovascular events in recently postmenopausal women: randomised trial. BMJ
2012;345:e6409.
42. Hodis HN, Mack WJ, Lobo RA, et al. Estrogen in the prevention of atherosclerosis - A
randomized, double-blind, placebo-controlled trial. Annals of internal medicine
2001;135(11):939-53.
Page 37
43. Harman SM, Black DM, Naftolin F, et al. Arterial Imaging Outcomes and Cardiovascular
Risk Factors in Recently Menopausal Women A Randomized Trial. Annals of internal
medicine 2014;161(4):249-60.
44. Raggi P, Callister TQ, Cooil B, et al. Identification of patients at increased risk of first
unheralded acute myocardial infarction by electron-beam computed tomography.
Circulation 2000;101(8):850-5.
45. Mikkola TS, Clarkson TB, Notelovitz M. Postmenopausal hormone therapy before and
after the women's health initiative study: what consequences? Ann Med
2004;36(6):402-13.
46. Holm P, Andersen HL, Andersen MR, et al. The direct antiatherogenic effect of estrogen
is present, absent, or reversed, depending on the state of the arterial endothelium. A
time course study in cholesterol-clamped rabbits. Circulation 1999;100(16):1727-33.
47. Koh KK, Ahn JY, Kang MH, et al. Effects of hormone replacement therapy on plaque
stability, inflammation, and fibrinolysis in hypertensive or overweight postmenopausal
women. The American journal of cardiology 2001;88(12):1423-6, A8.
48. Zanger D, Yang BK, Ardans J, et al. Divergent effects of hormone therapy on serum
markers of inflammation in postmenopausal women with coronary artery disease on
appropriate medical management. Journal of the American College of Cardiology
2000;36(6):1797-802.
49. Pradhan AD, Manson JE, Rossouw JE, et al. Inflammatory biomarkers, hormone
replacement therapy, and incident coronary heart disease: prospective analysis from
the Women's Health Initiative observational study. Jama 2002;288(8):980-7.
Page 38
50. Hwang J, Hodis HN, Hsiai TK, et al. Role of annexin II in estrogen-induced macrophage
matrix metalloproteinase-9 activity: the modulating effect of statins. Atherosclerosis
2006;189(1):76-82.
51. Herrington DM, Klein KP. Randomized clinical trials of hormone replacement therapy for
treatment or prevention of cardiovascular disease: a review of the findings.
Atherosclerosis 2003;166(2):203-12.
52. Hodis HN, St John JA, Xiang M, et al. Inflammatory markers and progression of
subclinical atherosclerosis in healthy postmenopausal women (from the Estrogen in the
Prevention of Atherosclerosis Trial). The American journal of cardiology
2008;101(8):1131-3.
53. Hodis HN, Mack WJ, Shoupe D, et al. Methods and baseline cardiovascular data from the
Early versus Late Intervention Trial with Estradiol testing the menopausal hormone
timing hypothesis. Menopause-the Journal of the North American Menopause Society
2015;22(4):391-401.
54. Rosenfeld ME, Kauser K, Martin-McNulty B, et al. Estrogen inhibits the initiation of fatty
streaks throughout the vasculature but does not inhibit intra-plaque hemorrhage and
the progression of established lesions in apolipoprotein E deficient mice. Atherosclerosis
2002;164(2):251-9.
55. Williams JK, Anthony MS, Honore EK, et al. Regression of Atherosclerosis in Female
Monkeys. Arteriosclerosis Thrombosis and Vascular Biology 1995;15(7):827-36.
Page 39
56. Adams MR, Register TC, Golden DL, et al. Medroxyprogesterone acetate antagonizes
inhibitory effects of conjugated equine estrogens on coronary artery atherosclerosis.
Arteriosclerosis Thrombosis and Vascular Biology 1997;17(1):217-21.
57. Clarkson TB, Anthony MS, Jerome CP. Lack of effect of raloxifene on coronary artery
atherosclerosis of postmenopausal monkeys. Journal of Clinical Endocrinology &
Metabolism 1998;83(3):721-6.
58. Clarkson TB, Anthony MS, Morgan TM. Inhibition of postmenopausal atherosclerosis
progression: A comparison of the effects of conjugated equine estrogens and soy
phytoestrogens. Journal of Clinical Endocrinology & Metabolism 2001;86(1):41-7.
59. Salpeter SR, Walsh JME, Greyber E, et al. Brief report: Coronary heart disease events
associated with hormone therapy in younger and older women. Journal of General
Internal Medicine 2006;21(4):363-6.
60. Boardman HMP, Hartley L, Eisinga A, et al. Hormone therapy for preventing
cardiovascular disease in post-menopausal women. Cochrane Database of Systematic
Reviews 2015(3).
61. Salpeter SR, Walsh JME, Greyber E, et al. Mortality associated with hormone
replacement therapy in younger and older women - A meta-analysis. Journal of General
Internal Medicine 2004;19(7):791-804.
62. Hodis HN, Mack WJ, Henderson VW, et al. Vascular Effects of Early versus Late
Postmenopausal Treatment with Estradiol. New England Journal of Medicine
2016;374(13):1221-31.
Page 40
63. Mirkin S, Amadio JM, Bernick BA, et al. 17beta-Estradiol and natural progesterone for
menopausal hormone therapy: REPLENISH phase 3 study design of a combination
capsule and evidence review. Maturitas 2015;81(1):28-35.
64. Lobo RA, Archer DF, Kagan R, et al. A 17beta-Estradiol-Progesterone Oral Capsule for
Vasomotor Symptoms in Postmenopausal Women: A Randomized Controlled Trial.
Obstetrics and gynecology 2018;132(1):161-70.
CHAPTER 2
DATASET AND BACKGROUND OF STUDY VARIABLES
RELATED TO HORMONE THERAPY EFFECT ON CORONARY HEART DISEASE
Page 42
2.1 DATESET FOR ANALYSIS
2.1.1 The Early versus Late Intervention Trial of Estradiol (ELITE)
The Early versus Late Intervention Trial with Estradiol (ELITE) was a single-center
randomized, double-blind, placebo-controlled trial specifically designed to evaluate this timing
hypothesis by testing whether the effect of hormone therapy (HT) on subclinical atherosclerosis
progression relative to hormone initiation differed according to time since menopause. The
primary hypothesis was that postmenopausal HT reduces the progression of subclinical
atherosclerosis, coronary artery disease and cognitive changes in postmenopausal women
without clinical evidence of cardiovascular disease (CVD) when initiated soon after menopause
(<6 years), but HT has no effect on these outcomes when initiated distant from menopause (≥10
years). The primary endpoint of subclinical atherosclerosis was represented with common
carotid intima media thickness (CIMT). The trial was funded by the National Institute on Aging,
National Institutes of Health (R01AG-024154) and registered with ClinicalTrials.gov as
NCT00114517.
ELITE included healthy postmenopausal women without clinical evidence of CVD, with a
serum estradiol (E2) level lower than 25 pg/mL and with cessation of regular menses for a
minimum of 6 months. Eligible women were postmenopausal for less than 6 years or for 10 years
or more at the time of randomization. Exclusion criteria were women in whom time since
menopause could not be determined; fasting plasma triglycerides higher than 500 mg/dL;
diabetes mellitus or fasting serum blood glucose level higher than 140 mg/dL; serum creatinine
level higher than 2.0 mg/dL; uncontrolled hypertension (systolic blood pressure/diastolic blood
pressure > 160/110 mmHg); untreated thyroid disease; life-threatening disease with prognosis
Page 43
of less than 5 years; history of deep vein thrombosis, pulmonary embolism, or breast cancer; and
current postmenopausal HT (within 1 month of screening).
Participants were screened for eligibility at a screening visit and baseline visit. Eligible
participants were randomized to active HT or placebo using stratified block randomization based
on time since menopause (<6 years or ≥10 years) with additional randomization stratification
factors of hysterectomy (yes or no) and baseline CIMT category (<0.75 or ≥0.75 mm)(1). The HT
used in the study was oral micronized 17β-estradiol (17β-E2) 1 mg/day with or without 4% vaginal
micronized progesterone (P4) gel 45 mg/day for 10 days each month. Participants, investigators,
staff, imaging specialists and data monitors were blinded to treatment assignment.
Participant evaluations were conducted at the Atherosclerosis Research clinic at USC.
Clinic research evaluations were scheduled every month for the first 6 months of the trial and
then every other month until the trial completion. Recruitment was initially based on a 5-year
trial (3-year recruitment with 2-5 years of randomly assigned treatment or placebo). In the fifth
year of follow up, additional funding was obtained to extend the trial for 2.5 years. The extended
funding and follow-up allowed for additional data collection including cardiac computed
tomography (CT) at the time participants completed their course of assigned intervention. In
total, the trial was conducted from 2005-2013 with median of 4.8 years follow up per participant.
The trial was approved by the Institutional Review Board of the University of Southern California.
All participants provided a written informed consent form. The trial conduct and its safety was
assessed by an External Data Safety Monitoring Board (appointed by the National Institute on
Aging, National Institutes of Health) whose members had expertise in women’s health,
menopausal health, HT, CVD, clinical trials, and biostatistics.
Page 44
The primary results from the ELITE study showed that the effect of E2, with or without
P4, on CIMT progression differed between the early and late postmenopause strata (P=0.007 for
the interaction), as hypothesized. Among early postmenopausal women (<6 years since
menopause), the mean CIMT increased by 0.0044 mm per year in the E2 group versus 0.0078
mm per year in the placebo group (P=0.008). Among late postmenopausal women (≥10 years
since menopause), the rate of CIMT progression in the E2 group (0.0100 mm per year) was similar
to the placebo group (0.0088 mm per year) (P=0.29). CT measures of coronary-artery calcium,
total stenosis, and plaque, measured only at the end of trial follow-up (not at baseline), did not
differ significantly between the E2 and placebo groups in any postmenopause stratum. In
summary, the ELITE study concluded that HT significantly reduced the progression of subclinical
atherosclerosis if initiated in early postmenopause, however, but had no effect in late
postmenopause(2).
2.1.2 REPLENISH trial
The REPLENISH trial is a phase 3 randomized, placebo-controlled, multicenter trial (80
sites in the United States) testing a new formulation of HT, TX-001HR (TherapeuticsMD, Inc, Boca
Raton, FL), a combined 17β-E2 and natural P4 in a single gelatin capsule. Preliminary reports
showed that the medication is molecularly identical to endogenous E2 and P4 and has
bioavailability similar to their respective 17β-E2 tablets and micronized P4 capsules administered
together(3). The new formulation was intended to provide an option for postmenopausal women
who prefer bioidentical or natural HT. The purpose of this trial was to determine whether
different doses of TX-001HR are effective in reducing the frequency and severity of moderate to
severe vasomotor symptoms (VMS) compared to placebo. In addition, the endometrial safety of
Page 45
the medication after 12 months of continuous use was evaluated(4).
The primary trial endpoints were: (1) mean change in frequency of moderate to severe
VMS from baseline to week 4 and to week 12 for each active HT formulation compared with
placebo, and (2) mean change in severity of moderate to severe VMS from baseline to mild,
moderate and severe VMS to week 4 and to week 12 for each active HT compared with placebo.
The primary safety endpoint was the incidence of endometrial hyperplasia evaluated at 12
months. In addition, the Menopause-specific Quality of Life questionnaire (MENQOL) and
Medical Outcomes Study (MOS)-sleep questionnaire were collected at baseline, week 12, month
6 and month 12. The trial was registered with ClinicalTrials.gov as NCT01942668.
Eligible participants were: (1) healthy postmenopausal women; (2) with a uterus; (3)
seeking treatment for VMS; (4) aged 40-65 years; (5) serum E2 ≤ 50 pg/mL; (6) body mass index
(BMI) ≤ 34 kg/m
2
; (7) using ≤ 2 antihypertensive drugs; (8) had a negative screening
mammography, normal breast exam and endometrial biopsy. Exclusion criteria were
contraindications to HT, heavy smokers (≥15 cigarettes per day), history of endometrial
hyperplasia or undiagnosed vaginal bleeding, history of cancer or clinically significant relevant
physical or mental illness.
Approximately 4000 women were screened for study eligibility to recruit 1750 women
meeting the study inclusion and exclusion criteria. At baseline eligible participants were
randomly assigned to one of five intervention groups (17β-E2 0.25 mg plus P4 50 mg, 17β-E2 0.5
mg plus P4 50 mg, 17β-E2 0.5 mg plus P4 100 mg, 17β-E2 1 mg plus P4 100 mg) or placebo.
Women with hot flushes (≥7/day or ≥50/week) were included in the VMS sub-study and were
randomized 1:1:1:1:1 to daily 17β-E2/P4 of 1.0mg/ 100 mg (n=141), 0.5 mg/100 mg (n=149), 0.5
Page 46
mg/50 mg (n=147), 0.25mg/50mg (n=154), or placebo (n=135). Women who did not meet criteria
for the VMS sub-study were randomized 1:1:1:1 to active17β-E2/P4 doses as part of the primary
safety endpoint analysis of endometrial hyperplasia (there was no placebo comparator).
Randomization at each study site was achieved using a reproducible, computer-generated block
randomization schedule. Study participants and staff were blinded to the assigned intervention
throughout the study.
Participants were evaluated at 6 follow-up visits at week 4, week 8, week 12, month 6,
month 9 and month 12. The primary outcome assessment was rate of improvement in VMS
frequency and severity from baseline using a 7-level response scale ranging from ‘‘very much
improved’’ to ‘‘very much worse’’; the scores were then categorized into a clinically meaningful
response (CGI ratings of ‘‘much improved’’ or ‘‘very much improved’’), minimally improved
response (rating of ‘‘minimally improved’’), and no change or worse (ratings of ‘‘no change’’ to
‘‘very much worse’’). The VMS outcome was assessed from each subject in VMS sub-study
through the questions "Rate the total improvement, whether or not in your judgment it is due
entirely to drug treatment compared to your condition at admission to the study, and how much
has it changed?". The HT effect on the endometrium was evaluated with incidence of
endometrial hyperplasia at 12 months from endometrial biopsy. In addition, diary for bleeding
(requiring sanitary protection) and spotting (not requiring sanitary protection) were completed
on daily basis up to month 12.
In addition to these main trial outcomes, at each visit, vital signs, adverse events,
concomitant drug use were recorded. In addition, blood sample was collected to test for
hormonal levels such as estrone (E1), E2, P4 and blood chemistry including glucose, lipids, and
Page 47
coagulation factors levels. The REPLENISH trial was conducted from August 2013 to October
2016.
Recent reports from the REPLENISH trial showed that that the study intervention, TX-
001HR, was effective in reducing both frequency and severity of VMS and had no incidence of
endometrial hyperplasia (5). The active intervention demonstrated clinically meaningful
improvements in VMS frequency in postmenopausal women compared to placebo. As assessed
by the CGI, women assigned to TX-001HR showed higher proportions of clinically-meaningful
response compared with placebo (50-63% versus 33%, P < 0.01) at week 4 and at week 12 (73-
82% vs 53%, P<0.01). A beneficial effect of TX-001HR across all doses compared to placebo was
observed, with statistically significant improvements in the frequency of VMS, as well as in the
vasomotor domain of the MENQOL questionnaire when compared with placebo(6). All doses of
TX-001HR significantly reduced the frequency of VMS at week 4 and week 12 follow up compared
to placebo, with the mean daily number of moderate to severe VMS decreasing from 10-11/day
at baseline to 2-4/day with TX-001HR (5/day for placebo). In terms of VMS severity, TX-001HR in
dose of 17β-E2/P4 as 1.0 mg/100 mg, 0.5 mg/100 mg and 0.5 mg/50 mg significantly improved
moderate to severe VMS (p<0.05); VMS severity did not improve with TX-001HR in dose of 17β-
E2/P4 as 0.25 mg/50 mg compared to placebo. These response rates for 50% and 75% reduction
in their moderate to severe VMS frequency achieved with TX-001HR were statistically significant
different from those achieved with placebo(7).
Sleep parameters were evaluated with MOS-sleep, a 12- item questionnaire measuring 6
sleep dimensions: initiation (time to fall asleep), quantity (hours of sleep each night),
maintenance, respiratory problems, perceived adequacy and somnolence over the past 4 weeks.
Page 48
The subscales are reported as Sleep Problem Index, sleep disturbance, sleep somnolence, snoring
and sleep shortness of breath or headache. At week 12, there were significant improvements in
sleep parameters from baseline among women taking TX-001HR and the effect was maintained
up to 12 months when compared with placebo (p<0.05). The treatment had no effects on the
snoring subscale, or the sleep shortness of breath or headache subscale(8).
For endometrial safety, TX-001HR was associated with high amenorrhea rates
(cumulative amenorrhea from cycle 1-13 56-73%; no bleeding 7-90%). Incidence of vaginal
bleeding from treatment with TX-001HR and placebo was 1.0-4.6% and 0.7%, respectively.
Uterine bleeding and spotting decreased over time. No cases of endometrial hyperplasia or
cancer were observed at the 12 month trial endpoint(9).
In summary, the REPLENISH trial showed that over 12 months of treatment, TX-001HR, a
combined 17β-E2 and natural P4, reduced the frequency and severity of VMS, and improved
sleep parameters. The potential for uterine bleeding and abnormal pathology were avoided with
the P4 doses of 50mg and 100mg used in this formulation.
Page 49
2.2 EXPOSURE: HORMONE THERAPY
2.2.1 Estrogen and progesterone as hormone therapy
Common HT regimens used in postmenopausal women are estrogen therapy (ET) and
combined estrogen-progestogen therapy (EPT). For women who have had a hysterectomy, ET, or
unopposed estrogen is used, while among women with a uterus, EPT is suggested. Although the
main benefit of EPT is derived from the estrogen component, the progestogen component
(progesterone, progestin or synthetic forms of progesterone) is used to reduce the risk of
endometrial carcinoma, a risk that significantly increased with use of ET(10).
2.2.2 Estrogen structure and receptors
Estrogen is defined as a variety of chemical compounds that have an affinity for estrogen
receptors (ERs), which mainly are ER-α and are ER-β receptors. Estrogen can be divided into 6
main types; human estrogens, nonhuman estrogens, synthetic estrogens, synthetic estrogen
analogs with a steroid molecular structure, synthetic estrogen analogs without a steroid skeleton,
and plant-based estrogens without a steroid skeleton.
Human estrogens are naturally produced in the human body. The endogenous human
estrogens are E1, E2 and estriol (E3). Among these 3 types of human estrogens, E2 is the most
biologically active, while E1 is 50-70% less active and E3 is 10% as active compared to E2. The
principal estrogen secreted by the ovaries is E2, which is metabolized to E1; both E1 and E2 can
be metabolized to E3. There is human fetal steroid estetrol (E4) which shows selective estrogen
receptor modulator effect in rat(11). While E4 acts as an estrogen on all tissue investigated, it
has antagonistic properties in breast tumor tissue in presence of E2. The currently available
exogenous human estrogen is 17-β E2. The exogenous E4 has been used as contraceptive
Page 50
hormone and is under investigation by Mithra company as HT in postmenopausal women(12-15).
Nonhuman estrogens generally refer to conjugated equine estrogen (CEE), a mixture of
at least 10 estrogens (Figure 2.1) extracted from pregnant mares’ urine. The CEE consists of the
sulfate esters of ring B saturated estrogens (classical estrogens): estrone (E
1
; 3-hydroxy-1,3,5(10)
estratrien-17-one), 17-β estradiol (17-E
2
; 1,3,5(10) estratrien-3,17 β -diol) and 17-α estradiol (17-
α E
2
; 1,3,5(10) estratriene-3,17 α -diol); and the unique ring B unsaturated estrogens: equilin (Eq;
3-hydroxy-1,3,5(10),7- estratetraen-17-one), equilenin (Eqn; 3-hydroxy-1,3,5(10) 6, 8-
estrapentaen-17-one), 17-α dihydroequilin (17-α Eq; 1,3,5(10) 7-estratetraen-3,17-diol), 17-β
dihydroequilin (17-β Eq; 1,3,5(10) 7-estratetraen-3, 17-β diol), 17-α dihydroequilenin (17 α -Eqn;
1,3,5(10) 6,8-estrapentaen-3,17-α diol), 17-β dihydroequilenin (17-β Eqn; 1,3,5(10)6,8-
estrapentaen-3,17-β diol), and delta-8-estrone (Δ
8
-E
1
; 3-hydroxy-1,5(10)8-estratetraen-17-one).
There is also an additional estrogen, Δ
8
-17 β estradiol (1,5(10)8- estratetraen-3,17 β -diol), a ring
B unsaturated estrogen that is formed in humans from Δ
8
-E
1. The pharmacologic effects of CEE
are a result of the sum of activities of these estrogens.
There are 2 types of synthetic estrogens; esterified estrogens (contains 75-85% sodium
sulfate) and synthetic conjugated estrogens (SCE), which are SCE-A (a mixture of 9 estrogens
found in CEE) and SCE-B (a mixture of all 10 estrogens found in CEE). The synthetic estrogen
analogs with a steroid molecular structure are ethinyl estradiol (commonly used for
contraception) and estropipate. The synthetic estrogen analogs without a steroid skeleton are
stilbesterol derivatives, which are not used for menopausal related therapy. The plant-based
Page 51
estrogens without a steroid skeleton are known as phytoestrogens, which have weak estrogenic
and antiestrogenic properties depending on the target tissue.
Estrogens can induce cellular change through several mechanisms, primarily acting
through ER-α or ER-β receptors. In women, ERs have been identified in many tissues with
differing distribution of ERs-α and ERs-β in different tissues, states of health and age. ERs-α are
primarily expressed in the reproductive tract and breast but also are found in liver, bone, adipose
tissue and brain. ERs-β are commonly found in colon, vascular endothelium, lung, bladder and
brain. Both ER-α and ER-β are present in the ovary, the central nervous system and cardiovascular
system(16).
Estrogens affect cellular change though several pathways; ER-dependent nuclear initiated
estrogen signaling, ER-dependent membrane initiated estrogen signaling, the ER-independent
pathway and ligand-independent activation of ER. (Figure 2.2)
The classical pathway is ER-dependent, nuclear initiated estrogen signaling, where
estrogen enters the cytoplasm through the cell membrane and binds to an ER at the nucleus. The
ER bound estrogen forms a ligand-activated transcription factor that regulates transcription of
target genes in the nucleus. The transcription occurs after when a ligand-activated transcription
factor binds to estrogen response element regulatory sequences in target genes then recruits
coregulatory proteins such as coactivators. The DNA receptor-ligand complex will either activate
or turn off gene transcription. After transcription, the ER or coactivator can be degraded by the
26S proteasome, and subsequently another cycle of assembly, activation and destruction of
ligand-activated transcription factor occurs(16).
Page 52
For ER-dependent, membrane initiated estrogen signaling, estrogen binds to the ER
located at the plasma membrane and works with the adaptor proteins to result in a rapid non-
genomic effect. This pathway occurs in the nervous system, skeleton, liver and other tissues. The
signaling from this pathway involves modification of growth factors, neurotransmitter receptors,
tyrosine kinase receptors and insulin-like growth hormone receptors(16).
In the ER-independent pathway, estrogen triggers a cellular response through the non-
ER membrane-associated estrogen-binding proteins. Demonstrated estrogen effects through this
pathway include antioxidant effects and suppression of oxidative stress(16).
Apart from estrogens, ERs can be activated by a variety of factors though ligand-
independent activation of the ER. Activators include neurotransmitters such as dopamine,
growth factors such as epidermal growth factor and insulin-like growth factor and activators of
intracellular signaling pathway such as protein kinase(16).
2.2.3 Estrogen effects on coronary heart disease
Coronary heart disease (CHD)-related effects of estrogen effects can exert through both
ER-dependent and ER-independent pathways.
The estrogen effect on CHD through ER-dependent pathways has been shown to occur in
both ER-α and ER-β. A study of ER expression in postmortem coronary artery samples from
women indicated that ER-α was expressed in significantly higher proportion of normal (non-
atherogenic) arteries (15 of 21) than those of atherogenic arteries (6 of out 19) (p=0.01). The
result suggested that atherosclerosis in women is associated with reduced levels of ER-α in
arteries(17).
Page 53
Cardiovascular effects of estrogen functioning through ER-dependent, nuclear initiated
estrogen signaling includes the inhibition of response to vascular injury and protection against
atherosclerosis. A study measuring function of the gene encoding ER-α demonstrated increased
in methylation and decreased gene expression in coronary atherosclerotic plaques when
compared with healthy proximal aorta (ER-methylation percent of 10% vs. 4%, p<0.01,
respectively). There was also a significant age-related increase in ER-α gene methylation in the
right atrium (6-19%, R=0.36, p<0.05)(18). Another study in vascular smooth muscle cells and
blood vessels from ER-β deficient mice showed multiple cardiovascular abnormalities suggesting
that ER-β has an important role in regulation of vascular function. Compared to wild-type mouse,
these mice developed hypertension upon aging and exhibited abnormal vascular function
through abnormal ion channel function of vascular smooth muscle cells(19). Polymorphisms in
both ER-α and ER-β have been associated with CVD. Among 3791 postmenopausal women in the
Rotterdam Study, a prospective cohort study, the haplotype encompassing ER-α polymorphism
was associated with an increased risk of myocardial infarction and ischemic heart disease after
adjustment for known cardiovascular risk factors(20). An analysis of ER-β polymorphisms in 702
women from the Framingham Heart Study showed that ER-β polymorphisms were associated
with hypertension (p=0.0006-0.01), left ventricular mass and left ventricular wall thickness
(p=0.007-0.03), which all are significant risk factors for CVD(21). In addition to direct effects on
vascular and cardiac function, estrogen also effects clotting factors through ER-dependent,
nuclear initiated estrogen signaling; this will be discussed in the following section.
In the ER-dependent, membrane initiated estrogen signaling pathway, estrogen has
non-genomic effects on vascular smooth muscle and endothelial cells as it induces nitric oxide
Page 54
(NO) synthesis and results in relaxation of vascular smooth muscle and inhibits platelet activation.
Both ER-α and ER-β are expressed in human vascular endothelial cells, vascular smooth muscle
cells and cardiomyocytes. Estrogen can increase the endothelial NO synthase (eNOS) by
activating transcription of the eNOS gene through an ER-dependent mechanism(22).
In studies of intact bovine endothelial cells, estrogen activated eNOS, which was fully
inhibited with ER antagonists. In addition, overexpression of the ER-α transcription factor led to
marked enhancement of the acute response to estrogen, and ER antagonists again blocked this
effect. These results suggest that estrogen causes rapid blood vessel dilatation through activation
of eNOS via ER-α(23). This process also occurs through ER-β, as a study of mouse blood vessel
showed that estrogen attenuates vasoconstriction among normal mice but augments
vasoconstriction among ER-β deficient mice. This study showed that the effect of estrogen on
vascular smooth muscle occurs through ER-β to induce an increase in nitric oxide synthase
expression(19).
The estrogen effect on CVD through ER-independent pathways is demonstrated on the
oxidation of lipoproteins. Although there is no proven mechanism of the known antioxidant
effects of estrogens, estrogens with higher affinity for ERs have weaker antioxidant properties
than those with lower affinity for ERs. Therefore, the oxidative effect of estrogen is thought to
be more likely non-genomic and independent of ER(24). The effect of estrogens on lipoprotein
oxidation will be addressed in the following section (section 2.4.4).
2.2.4 Progesterone structure and receptors
Progestrogen includes progesterone, progestin and synthetic forms of progesterone, and
acts through progesterone receptors (PRs), PR-α and PR-β. The PR-α predominates in the uterus
Page 55
and ovary and the PR-β is mainly found in the breast. Both PR-α and PR-β are expressed in blood
vessels; however, little is known about their functions in cardiovascular physiology and
pathology(25).
2.2.5 Progesterone effects on coronary heart disease
In light of the different effect between estrogen alone and estrogen plus progestogen on
CHD (reviewed in section 1.2.5), the effect of progestrogen on cardiovascular system was
identified through effects on vascular and cardiac function. Effect on lipids and clotting factors
will be discussed in following sections (section 2.4.6).
As estrogen has been observed to improve vasodilatation, some progestogens have been
found to reduce this effect. In a randomized double blinded cross-over trial of HT effect among
postmenopausal women, cyclical addition of dydrogesterone and norethisterone to estrogen-
only therapy significantly increased the carotid artery pulsatility index (PI) (p=0.02)(26). In
cynomolgus monkeys, a CEE effect on coronary artery dilatation was reduced by 50% when
medroxyprogesterone acetate (MPA) was administered concomitantly either in a cyclic (10 days
per month) or continuous regimen(27). In rhesus monkeys, administration of MPA alone had an
adverse effect on coronary artery hyperactivity; this effect did not occur with administration of
progesterone (P4) alone(28). However, MPA did not reduce a CEE benefit on vascular reactivity
measured by brachial artery flow mediated dilation and did not alter the markers of inflammation
such as proinflammatory cytokines, cell adhesion molecules and C-reactive protein among
postmenopausal women(29). Another study comparing the effect of 0.625 mg CEE alone with
the CEE plus 2.5 mg MPA on PI in the internal carotid artery among 75 postmenopausal women
Page 56
showed a significant reduction in PI from baseline in both groups and with no difference between
groups(30).
In a study investigating the effect of MPA on estrogen-induced increase in endothelium-
dependent vasodilation in postmenopausal women showed that MPA administration may offset
favorable effects of estrogen on endothelial function. MPA was shown to inhibit the vasodilation
effect from estrogen; however, it did not reduce LDL-lowering and
antioxidant effects of estrogen. This suggests that MPA-induced inhibition of endothelium-
dependent vasodilation is possibly independent of changes in oxidative susceptibility and plasma
concentration of low density lipoprotein cholesterol (LDL-C)(31).
2.2.6 Timing as possible factors of estrogen effect on coronary heart disease
According to the prior evidence that HT in particular E2 was associated with less
subclinical atherosclerosis, CHD incidence and mortality when the therapy was initiated soon
after menopause but had no effect or increased risk when the therapy was initiated later after
menopause. Use of 17β-E2 in combination with vaginal P4 was associated with less progression
of subclinical atherosclerosis among women with less than 6 years since menopause but had no
effect among women with 10 years or more since menopause(2). With combined regimen of 17β-
E2 plus norethisterone acetate (NETA) significantly reduced all-cause mortality, hospitalization
for myocardial infarction or heart failure among recently postmenopause women(32). The
combined regimen of CEE plus MPA showed a reduced risk of CHD (33) and mortality (34) among
young postmenopausal women (50-59 years old) and women initiated HT within 10 years since
menopause. Different age effect of E2 on atherosclerosis, CHD and CVD is possibly due to
different dose, estrogen receptors level and function as well as underlying vascular health tissue.
Page 57
A study evaluating the effect of different doses of E2 on coagulation measured with fibrin
production and degradation (D-dimer), prothrombin fragment 1 and 2 and thrombin-
antithrombin complex showed that higher dose of E2 had more coagulation activation than lower
dose of E2(35). Another study reported more decreases in factor VII, less reduction of
antithrombin and protein C and increased levels in protein S and tissue factor pathway inhibitor
among higher dose of E2 than lower dose of E2(36). Moreover, higher dose of E2 was associated
with significant increase in matrix metalloproteinases, which reflects vascular remodeling(37).
Therefore, higher dose of E2 could increase risk of thrombosis and atherosclerosis compared with
lower dose of E2.
In older postmenopausal women, with decreased amount of estrogen receptor along with
its function could be accounted for lack of E2 effect on atherosclerosis as there was an age related
increase in methylation associated with inactivation of the estrogen receptor alpha gene in
vascular tissue(18). The effect of E2 differs according to stages of atherosclerotic lesion in
vascular tissue. As E2 could improve or reverse the endothelial dysfunction in vessels with early
atherosclerotic lesion, in advanced atherosclerotic lesion, vessels are more susceptible to
inflammatory and hemostasis abnormality when E2 was administered(25) (Figure 2.3). In
addition, a study in mice showed that higher levels of 27-hydroxycholesterol presented in
atherosclerotic lesion could be a contributing factor in the loss of E2 effect on vascular
disease(38). Moreover, there could be an increases risk of CHD among older women as E2
increases matrix metalloproteinases level and could lead to the breakdown of the fibrous cap in
atheromatous arterial plaque therefore results in plaque instability and possible rupture and
thrombosis(39).
Page 58
2.2.7 Hormone regimen used in ELITE and REPLENISH
In ELITE, the HT regimen among women with uterus was oral micronized 17β-E2 1 mg/day
with 4% vaginal micronized P4 gel 45 mg/day for 10 days each month and the HT regimen among
women without uterus was 17β-E2 alone. Although vaginal P4 gel is not a standard treatment for
endometrial protection, it was used in this study with an intention to minimize possible interfere
effect from P4 on the association of 17β-E2 and CIMT. In addition, endometrial safety has been
closely monitored throughout the trial and there was no case of endometrial hyperplasia or
cancer during the whole follow up period.
Regimen of HT used in REPLENISH was TX-001HR, a combined 17β-E2/P4 with dose of
1.0mg/ 100 mg, 0.5 mg/100 mg, 0.5 mg/50 mg, and 0.25mg/50mg(3). This formulation was
reported to have bioavailability similar to the respective reference products of 17β-E2 and P4
given separately. The pharmacokinetic endpoints of all AUC and C max met bioequivalent criteria
for all analytes of E2, E1 and P4 except C max for E1. The absorption of E2 and P4 were similar
between TX-001HR and 17β-E2 and P4. A review of published data suggested that the use of this
natural formulation of 17β-E2 combined with P4 might be a more preferable option over other
combinations of estrogens and synthetic progestins in terms of tolerability of progesterone
formulation, progesterone effect on breast, cardiovascular system, diabetes, and estrogen effect
on cardiovascular system(4).
2.2.8 Measurement of estrogens and progesterone (ELITE and REPLENISH)
ELITE
In ELITE study, plasma E2 level is measured at baseline and every 6 months follow up visit.
The assay sensitivity is 2 pg/ml and the interassay coefficients of variation are 11%, 13% and 12%
Page 59
at 15, 36 and 101 pg/ml, respectively.
The measurement method was radioimmunoassay (RIA) with preceding organic solvent
extract and Celite column partition chromatography steps(40) in the University of Southern
California Reproductive Endocrine Laboratory. The steroid was first extracted with hexane:ethyl
acetate (9:1) and then with ethyl acetate:hexane (3:2). After pooling and evaporating the organic
solvents, the residue was reconstituted in isooctane and applied on a Celite partition
chromatography column to separate E2 from other steroids. E2 is eluted off the column with 40%
(v/v) ethyl acetate in isooctane. After each fraction was dried down, the residue was
reconstituted in assay buffer (0.1 M phosphate buffer, pH 7.4, to which sodium chloride and
gelatin are added). E2 was then measured by separate RIAs along with the appropriate standards
and quality control samples. Each RIA utilizes a specific antiserum in conjunction with an
iodinated radioligand. A second antibody is used to separate antibody-bound from unbound E2.
The intraassay and interassay coefficients of variation are, on the average, <7% and <13%,
respectively, when low, medium and high controls are used.
REPLENISH
In REPLENISH study, serum E2 and P4 levels were analyzed at central lab in Ventiv Health
Clinical Lab Inc in Princeton, NJ. Levels of E2 and E1 were measured at screening visit, week 4,
week 12, month 6, month 9, and month 12. For E2 and E1, there was a single GC-MS/MS method
that measured both analytes simultaneously. The method was fully validated for human serum
in the range of 2 to 500 pg/mL for E2 and 5 to 1000 pg/mL for E1. The interassay coefficients of
variation was <8.5% for E2 and <7.0% for E1. Level of P4 was measured at screening visit, week12
and month 12. For P4, there was an LC-MS/MS assay that was fully validated for human serum.
Page 60
The range of the assay was 0.05 to 50 ng/mL and the interassay coefficients of variation was
<5.5%.
Page 61
2.3 OVERVIEW OF MEDIATORS OF ESTROGEN EFFECT ON CORONARY HEART DISEASE
As epidemiologic and clinical studies suggested that estrogen has protective effect on
CHD, basic science studies have explored the biological mechanism of estrogen action on CHD.
Traditional CHD risk factors among women are dyslipidemia, obesity, diabetes, hypertension,
smoking, physical inactivity(41). Several plausible mechanisms have been proposed to explain
the estrogen-mediated effect on these CHD risk factors through both ER-dependent and ER-
independent pathway as mentioned in detail in the previous section (section 2.2.2). The
mediators of estrogen effect on CHD include its effect on lipid metabolism, coagulation factors,
metabolic inflammation, and vascular tissue.
The main vascular dysfunctions leading to cardiovascular disease are atherosclerosis and
blood clotting(42). The mediation effect of estrogen on CHD through lipid metabolism described
as dyslipidemia is an important risk factor of atherosclerosis. In addition, effect of estrogen on
coagulation factors would represent possibility of vascular blood clot leading to CHD.
2.3.1 Estrogen effect on lipid metabolism and coagulation factors
The effect of estrogen on lipid metabolism and coagulation factors will be discussed in
the following section (section 2.4 and section 2.5)
2.3.2 Estrogen effect on metabolic inflammation
The association between estrogen and decreased metabolic inflammation could explain
the preventive effect of estrogen on CHD. Estrogen has been shown to regulate adipose tissue
and insulin metabolism to prevent obesity and diabetes, which are important risk factors for CHD.
By keeping adequate energy balance, estrogen decreases food intake and favors energy
expenditure. The mechanism may be explained by regulation of leptin signaling by estrogen in
Page 62
animal model, as the data in human are still controversial(43). Influence of estrogen on
glucocorticoid is a possible mechanism as excess glucocorticoid signaling related to the
development of visceral obesity and insulin resistance. Estrogen regulates the amount of body
fat accumulation and its distribution to prevent visceral or central obesity. In pancreatic cells,
estrogen regulate insulin secretion and nutrient homeostasis. The action of estrogen in skeletal
muscle, liver, adipose tissue and immune cells are involved insulin sensitivity and prevention of
lipid accumulation and inflammation(44, 45).
Overall, estrogen decreases metabolic inflammation by maintaining energy balance,
prevent obesity and diabetes thus reduces CHD risk(46).
2.3.3 Estrogen effect on vascular tissue (endothelium and smooth muscle cell)
With antioxidant properties, estrogen increases nitric oxide and regulate superoxide in
endothelial and smooth muscle cell. Estrogen induces endothelium-dependent vasodilation by
increasing nitric oxide and prostacyclin. In addition, estrogen inhibits vasoconstriction by
decreasing endothelin and angiotensin II.
Estrogen also causes relaxation of vascular smooth muscle cell through direct effect on
the contraction mechanism(47). Estrogen decreases development of fatty streak deposit and
atherosclerosis plaque by regulation of immune-inflammatory process in vessels(48). The
possible mechanisms at vascular tissue related to decreased CHD are reduced fibrosis,
stimulation of angiogenesis, vasodilation, improved mitochondrial function and reduce oxidative
stress(49).
Page 63
2.4 MEDIATOR: LIPID
2.4.1 Pathway of lipid metabolism
Major lipid particles in human body are cholesterol and triglyceride (TG), which both
transports in blood circulation as lipoproteins. Plasma lipoproteins are composed of a TG core
and cholesterol ester, coated by phospholipid, unesterified cholesterol and apolipoproteins.
There are five main lipoproteins; chylomicrons, very low-density lipoprotein cholesterol (VLDL-
C), intermediate density lipoprotein cholesterol (IDL-C), low-density lipoprotein cholesterol (LDL-
C) and high-density lipoprotein cholesterol (HDL-C). In laboratory, total cholesterol (TC), TG and
HDL are measured directly while VLDL-C and LDL-C are calculated estimation values.
The generation and transport of lipids in the body includes 3 main pathways (Figure 2.4);
exogenous pathway, endogenous pathway and pathway of reverse cholesterol transport(50).
Exogenous pathway begins with dietary fat digestion and absorption, followed by chylomicrons
formation in the intestine from TG and cholesterol. The chylomicrons are hydrolyzed into TG and
free fatty acids in the adipose tissue and muscle cells then some remnants are transformed into
other lipoproteins. Lipoprotein synthesis in liver is considered to be the endogenous pathway.
Liver cells generate TG and cholesterol then releases them as VLDL-C particles into blood
circulation. The VLDL-C is hydrolyzed in tissue then the remnants form IDL-C and LDL-C, which
are retaken back to the liver through their receptors. Pathway of reverse cholesterol transport
is the process of reabsorption of cholesterol from the tissues back to the liver. HDL-C is the key
protein involving with this process. It is formed by precursors from liver and intestine through a
maturation process.
2.4.2 Lipids and coronary heart disease
Page 64
Elevated LDL-C has been identified as the key lipid related to elevated CHD risk in both
genders, whereas lower levels of HDLC and elevated TG have been more closely associated
with cardiovascular risk in women(51). The Framingham Heart Study was one of the first cohort
studies to report that HDL-C level was inversely correlated with the incidence and mortality
of CHD among women, with adjustment for other cardiovascular risk factors. The relative risks
comparing the bottom HDL-C quintile (less than 45 mg/dl) to the top quintile (greater than 69
mg/dl), was 3.1(52). It has been estimated that each 1 mg/dl increase in HDL-C is associated with
a 3% lower risk of CHD and 4.7% lower mortality rate in women(53).
Menopause has been associated with an increase in total and LDL-C and a decrease
in HDL-C. Estrogen therapy has been reported to increase HDL-C and decrease LDL-C levels, which
could explain the reduced risk of CHD. Two randomized, double-blind, placebo-controlled
crossover studies in 1,031 healthy postmenopausal women with normal lipid values at baseline
reported lipid levels after three months of treatment with CEE (0.625 mg and 1.25 mg per day).
These CEE doses significantly decreased the mean LDL-C level by 15% and 19% (p<0.0001) and
significantly increased the HDL-C level by 16%and 18% (p<0.0001), respectively(54).
2.4.3 Effect of hormone therapy through lipids in coronary heart disease
Change in lipoprotein metabolism after menopause was related to increased CHD risk.
Therefore it has been hypothesized that estrogen and progestin effect on CHD could be mediated
by lipoprotein metabolism(55). The Postmenopausal Estrogen/Progestin Interventions (PEPI)
trial was designed to determine effects on CHD risk factors, in particular lipoprotein levels, in
postmenopausal women treated with placebo, estrogen alone and combined estrogen plus
progestin (56). The PEPI trial randomized healthy postmenopausal women ages 45-64 years old
Page 65
(mean age of 56.1 years) into one of five treatment arms in 28 day cycles; placebo, CEE 0.625
mg/day, CEE 0.625 mg/day plus MPA 10mg/day for the first 12 days, CEE 0.625 mg/day plus MPA
2.5 mg/day, CEE 0.625 mg/day plus micronized progesterone (MP) 200 mg/day for the first 12
days. The lipoprotein profile, blood pressure, insulin and fibrinogen were measured at 3, 6 and
12 months during the first year then every 6 months for a total 3 years of follow-up. Overall,
estrogen alone or in combination with progestin improved lipoproteins and lowered fibrinogen
levels. There were no significant effects on post-challenge insulin level or blood pressure. All HT
regimens significantly increased HDL-C from baseline with an average of 2.4 mg/dl of HDL-C
higher than placebo. Among active treatments, the CEE-alone regimen had a higher mean change
in HDL-C (5.6 mg/dl) than CEE plus MPA (1.2-1.6 mg/dl) and CEE plus MP (4.1 mg/dl). All HT
regimens significantly decreased mean LDL-C with an average of 10.1 mg/dl relative to placebo
(p<0.001). However, all HT regimens increased mean TG compared with placebo (p<0.001). From
there results of PEPI trial, it has been proposed that the beneficial effect of estrogen therapy on
the progression of subclinical atherosclerosis in postmenopausal women could be due to these
beneficial changes in lipid metabolism noted.
A systematic review and meta-analysis of 28 randomized clinical trials evaluated the
effect of low-dose and conventional-dose HT on lipid profile(57). Compared with placebo, low-
dose HT was significantly associated with lower LDL-C and total cholesterol but was not
associated with HDL-C and TG. Compared with conventional-dose HT, low-dose HT was
significantly associated with higher total cholesterol and LDL-C. The data suggested that effect of
HT on lipid levels varies with different dose of HT, more specifically estrogen.
Page 66
The effect of E2 on atherosclerosis through lipids were studied in a post-trial analysis of
the Estrogen in the Prevention of Atherosclerosis Trial (EPAT), a randomized trial testing the
effect of unopposed 17β-E2 on CIMT among healthy postmenopausal women. As previous
evidence supported the role of decreased HDL-C and increased LDL-C in the causal pathway of
atherosclerosis, the analysis determined the extent to which the estrogen-induced changes in
these lipids metabolism over 2 years follow-up(58). CIMT progression was inversely associated
with on-trial levels of HDL-C (p=0.04) and positively associated with on-trial levels of LDL-C
(p=0.005). Women treated with 17β-E2 had higher on-trial HDL-C (percent change from baseline
of 12.6% vs. 7.2%, p=0.01) and lower LDL-C (percent change from baseline of -9.7% vs. -2.2%,
p=0.001) compared with women who received placebo. The univariate analysis of treatment
mediation showed that on-trial HDL-C (p=0.04) and LDL-C level (p=0.005) were significantly
independent determinants of CIMT progression. Both HDL-C and LDL-C levels jointly explained
30% of the effect of 17β-E2 on CIMT progression. For carbohydrate metabolism, women treated
with 17β-E2 showed reductions in fasting glucose, insulin and hemoglobin A1C; however, these
factors were not significantly related to CIMT progression (58). Another post-trial report from
EPAT a showed that estrogen and sex hormone binding globulin (SHBG) levels were associated
with reduced CIMT progression and these associations were partially mediated by the effects on
lipids.(59) Serum E1 , total E2, free E2 and SHBG levels were significantly inversely associated
with LDL-C and positively associated with HDL-C (p<0.0001)(59). These results suggest that the
beneficial effect of estrogen therapy on CHD is partly mediated by lipid metabolism.
2.4.4 Effect of estrogens on lipoprotein oxidation
Page 67
The effect of estrogens on CHD is also reported to be mediated by lipoprotein oxidation
through both ER-dependent and ER-independent pathways. As noted above, circulating lipid
levels play an important role in the etiology of CHD as lipid abnormalities contribute to
atherosclerosis development and progression(60, 61). ER-dependent lipoprotein oxidation is
demonstrated as oral estrogen therapy undergoes extensive first pass metabolism through ER-
α(25). As a result, oral estrogen decreases LDL-C, increase HDL-C and increase TG levels. In
addition, estrogen induced lipoprotein oxidation can also be ER-independent as in general
estrogens with higher affinity for ERs are weaker antioxidants(61, 62).
The oxidative hypothesis of atherosclerosis has been attributed to the process of LDL-C
and HDL-C oxidation. Oxidatively-modified LDL-C can be formed in vivo by free radical-based
mechanisms. Compared to native LDL-C, oxidatively-modifed LDL-C is more strongly related to
atherogenesis and CHD as it modulates platelet aggregability in blood vessel(63). In addition,
oxidatively-modified HDL-C had impaired reverse cholesterol transport, which leads to
accumulation of LDL-C.
Estrogens has been shown to significantly inhibit LDL-C oxidation, however different types
of estrogen has different potency(64) in lipoprotein oxidation evaluated by products of oxidation
and lag time. Increased lag time reflects the inhibition of lipoprotein oxidation activity. Both E2
and CEE, most commonly used estrogen in HT, have been shown to inhibit LDL-C oxidation. A
study in 18 postmenopausal women with intra-arterial administration of and 17β-E2 showed a
significant increase of lag time from 134 minutes at baseline to 167 minutes (p=0.01). Among 12
postmenopausal women receiving transdermal 17β-E2 for 1 month, a significant increase of lag
time from 132 minutes at baseline to 178 minutes (p=0.009) was observed(65). A randomized
Page 68
cross-over study among 8 postmenopausal women showed that the 3 most prevalent
constituents in CEE (estrone sulfate, equilin sulfate and 17-α dihydroequilin sulfate)
demonstrated antioxidation activity on LDL-C oxidation measures of negatively charged LDL-C
and lag phase duration(66).
The estrogen effect on oxidation of LDL-C seems to differ by duration of use. While long-
term estrogen use significantly inhibits LDL-C oxidation, short-term estrogen use dose not. An
RCT among 71 postmenopausal women evaluating the effect of 0.625 mg CEE, 0.625 mg CEE plus
5 mg MPA for 10 days per cycle and placebo showed that at 1 year follow up, women in both HT
regimens had a significant increase in the lag time of LDL oxidation compared to placebo,
p<0.01)(67). In contrast, short-term administration of estrogen (4-6 weeks) may not reduce LDL-
C oxidation. A study reported that there was no significant change in the lag time to oxidation,
or the maximum rate of propagation of the reaction observed with various estrogen-containing
regimens(68).
HDL-C oxidation also contributed to etiology of CHD when oxidatively-modified HDL-C is
related to a breakdown of reverse cholesterol transport. Benefit of higher HDL-C levels on CVD
could be in part from its ability to prevent LDL-C oxidation(69, 70). The HDL-C associated enzymes
may play a role in protective effect on LDL-C oxidation as paraoxonase and PAF-acetylhydrolase,
were showed to detoxify oxidized phospholipids produced by lipid peroxidation(69). Estrogen
was shown to protect both LDL-C and HDL-C from oxidation. The HDL-C delayed LDL-C oxidation
and this inhibitory effect of HDL-C was further enhanced with addition of equilenin or 17β-E2
(71).
Page 69
The precise mechanism of an estrogen effect on LDL-C and HDL-C oxidation remains to be
identified. So far, existing evidence indicates that estrogens have antioxidant properties that may
inhibit or delay both LDL-C and HDL-C oxidation especially with long-term administration.
2.4.5 Estrogen effect on atherosclerosis through lipids
In summary, with low estrogen levels, there are increased plasma LDL-C and decreased
plasma HDL-C levels. These lipids undergo the lipid peroxidation process that results in increased
oxidized LDL-C and oxidized HDL-C. The oxidized LDL-C induces the development of
atherosclerosis through formation of foam cells, fatty streaks in the vessels causing endothelial
injury and eventually resulting in advanced atherosclerosis lesions. In addition, the increased
oxidization of HDL-C inhibits the process of reverse cholesterol transport, which results in
accumulation of LDL-C as well as oxidized LDL-C, further contributing to the atherosclerosis
process. Administration of estrogen reduces plasma LDL-C and increased plasma HDL-C, as well
as reducing lipid peroxidation to reduce both oxidized LDL-C and oxidized HDL-C. The decreased
level of oxidized HDL-C leads to decreased levels of LDL-C and oxidized LDL-C, which helps protect
or reduce the risk against atherosclerosis process (Figure 2.5).
2.4.6 Progesterone effect on atherosclerosis through lipids
A review and pooled analysis of 248 prospective studies reported that progestin had
little effect on estrogen-induced reductions in LDL-C and total cholesterol. However, estrogenic
effect of increased HDL-C and TG were attenuated by progestins(72). This opposing effect varies
with types of progestin. The greatest effect was found in norethindrone acetate followed by
MPA, cyproterone acetate, progesterone, medrogestone and dydrogesterone. A study of 33
surgically postmenopausal women compared the effects of daily 0.6 mg CEE versus 10 mg MPA
Page 70
on lipid profiles at 12 months after menopause. Women in the CEE group had greater increases
in TG and HDL-C compared with those in the MPA group(73). Although progestin was reported
to reduce the beneficial effect of estrogen on HDL-C and TG, there is no clear link between altered
levels as a result of progestin treatment and CVD.
2.4.7 Measurement of lipids (ELITE and REPLENISH)
In ELITE, plasma lipids and lipoproteins were measured at baseline and every 6 month by
preparative ultracentrifugation and enzymatic assays standardized to the Centers for Disease
Control using Lipid Research Clinic protocol(74) including total cholesterol, total TG, LDL-C, LDL-
TG, VLDL-C, VLDL-TG, HDL-C, and HDL-TG. Plasma hemoglobin A1c is determined by high-
pressure liquid chromatography Yion exchange chromatography (Diabetes Control and
Complications Trial approved method)(75).
In REPLENISH, fasting (at a minimum of 8 hours) lipid levels including total cholesterol,
LDL-C and HLD-C were measured at screening visit, month 3, 6, 9, and 12 at central lab with
standard laboratory method.
Page 71
2.5 MEDIATOR: COAGULATION FACTORS
2.5.1 Thrombosis
Thrombosis, described as clotting in blood vessels, is a serious vascular disorder occurring
in both veins and arteries(76). Venous thrombosis or venous thromboembolism (VTE) includes
pulmonary embolism (PE) and deep vein thrombosis (DVT). Arterial thrombosis occurring in a
coronary artery can lead to myocardial infarction (MI), and the thrombosis in a cerebral artery
can lead to stroke.
Risk factors for thrombosis can be divided into 3 causes according to Virchow’s triad
theory; circulation stasis, hypercoagulable state and vascular injury(77). Circulation stasis is a
result of: (1) blood vessel abnormalities causing turbulent flow; (2) irregular stenosis in the
vascular lumen; or (3) immobilization. A hypercoagulable state is described as abnormalities in
platelets as well as coagulation and fibrinolytic pathways. Vascular injury may arise from surgery,
a vessel catheter, trauma, chemotherapy, vasculitis, sepsis or endothelial cell injury(78).
Circulation stasis and hypercoagulable state are important in the development of venous
thrombosis; previous VTE is the strongest predictor of subsequent VTE due to damage to the
venous valve, which leads to venous stasis (i.e., circulation stasis). In contrast, vascular injury is
more related to arterial thrombosis as the occurrence of arterial thrombosis is mainly determined
by the development of atherosclerotic changes in the arterial wall. Age is related to both venous
and arterial thrombosis as older age may be accompanied by greater circulation stasis, increased
hypercoagulable stage and vessel wall injury(79). The common causes of Virchow’s triads are
listed in Figure 2.6. Risk factors for thrombosis stratified by Virchow’s triads are listed in Table
2.1.
Page 72
2.5.1 Effect of hormone therapy on venous thrombosis
Several observational studies conducted primarily in the late 1990’s first reported an
increased risk of VTE (pooled relative risk confidence interval of 1.6-3.2) with use of estrogen
therapy among postmenopausal women(79). A recent review showed that despite the
heterogeneity in study methodology, the increased risks of VTE from HT use was consistently
reported and highest in the first year of HT use(80, 81). (Table 2.2)
A meta-analysis of observational studies and randomized clinical trials explored the
association between postmenopausal estrogen therapy and VTE (81). Data from 12 included
observational studies showed that oral estrogen increased risk of VTE with an odds ratio of 2.5
(95%CI, 1.9-3.4) while transdermal estrogen was not significantly associated with VTE (OR=1.2,
95%CI, 0.9-1.7). Data from 9 included clinical trials also supported the finding that oral estradiol
increased risk of VTE with similar effect estimate (OR=2.1, 95%CI, 1.4-3.1). The risk of VTE is
higher within the first year of estrogen treatment (OR=4.0, 95%CI, 2.9-5.7) compared to VTE risk
beyond one year (OR=2.1, 95%CI, 1.3-3.8). This meta-analysis reported that unopposed estrogen
(OR=2.2, 95%CI, 1.6-3.0) and estrogen plus progestin (OR=2.6, 95%CI, 2.0-3.2) were associated
with a similar risk of VTE. However, another meta-analysis of 6 observational studies showed that
risk of VTE among women using estrogen plus progestin (OR=2.3, 95%CI, 1.7-3.2) was similar to
those using unopposed estrogen (OR=1.7, 95%CI, 1.3-2.2) (p=0.15)(82).
Another meta-analysis compared the risk of VTE between oral and transdermal estrogen
therapy from 15 observational studies among postmenopausal women who had a mean ± SD age
of 57.7 ± 6.3 years with 3-20 years of follow-up(83). When compared to transdermal estrogen,
oral estrogen significantly increased risk of VTE (relative risk (RR)=1.63, 95%CI, 1.4-1.9) and DVT
Page 73
(RR=2.09, 95%CI, 1.35-3.23). When compared with transdermal route, oral estrogen alone
(OR=1.37, 95%CI, 1.21-1.56) had a lower risk of VTE than oral estrogen plus progestin (OR=1.83,
95%CI, 1.17-2.83).
A Cochrane review of 19 clinical trials with a total of 40,410 postmenopausal women
reported an overall risk of VTE among HT users as 1.92 times (95%CI 1.24-2.9) that of non-
users(84). When the trials were stratified by time-since-menopause at the start of HT use, VTE
risk among women within 10 years of menopause (VTE OR=1.74, 95%CI, 1.11-2.73) seem to be a
little lower than risk among those who were 10 years or more since menopause (OR=1.96, 95%CI,
1.37-2.80). The result suggested that as longer years since menopause usually reflects older age,
higher VTE relative risks may be in part due to older age. The sub-analysis of Heart and Estrogen-
progestin Replacement Study (HERS) II study reported a higher risk of VTE with HT use in women
older than 65 years compared to HT use in women younger than 65 years (hazard ratio (HR)=1.9,
95%CI, 1.0-3.6)(85). The Women’s Health Initiative (WHI) estrogen plus progesterone trial
reported a higher risk of VTE with estrogen plus progesterone use in advance age. The VTE risk
of estrogen plus progesterone use in sixth, seven and eight decades of life were HR=2.27
(95%CI1.19-4.33), HR=4.28 (95%CI, 2.38-7.22) and HR=7.46 (95%CI, 4.32-14.38), respectively(86).
The WHI estrogen alone trial also reported higher VTE risk with estrogen use in older age(87).
However, the recent WHI report did not show different CEE and MPA effect on PE and DVT risks
between age or year-since-menopause stratification (p trend >0.05). In addition, CEE did not
increase PE and DVT risks in any age or year-since-menopause stratification(33) (Table 2.3). When
limited to women 50-59 years of age in WHI study, CEE had no significant effect on any VTE and
Page 74
CEE and MPA significantly increased DVT but not PE and the absolute number to treat is rare (10
events per 10,000 women per year)(88).
In summary, results of the individual studies and meta-analysis consistently show that
postmenopausal HT is associated with increased risk of VTE especially during the first year of use.
While HT-associated elevations in VTE risk between estrogen alone versus estrogen plus
progestin is inconclusive, the risk varies with route of administration as oral estrogen had higher
risk than transdermal estrogen. In addition, HT-associated VTE risk tends to be higher with longer
years since menopause, or older age.
2.5.3 Effect of hormone therapy on arterial thrombosis
Arterial thrombosis commonly manifests as to coronary artery thrombosis, which lead to
coronary artery occlusion leading to CHD. The effect of HT on arterial thrombosis, specifically,
CHD is reviewed in previous section (section 2.2).
2.5.4 Coagulation pathway, coagulation factors, anti-coagulation factors and interpretation
Hemostasis is defined as an arrest of bleeding, and is controlled by the blood coagulation
system, a dynamic process representing a balance between coagulation, anti-coagulation, and
fibrinolysis pathways. Imbalance of the coagulation system leads to a tendency to either
excessive bleeding or blood clotting known as thrombosis. Coagulation proteins, also called
clotting factors, are core components of the coagulation system. In the presence of bleeding,
these clotting factors interact with one another and result in the conversion of soluble fibrinogen
to an insoluble fibrin strand to form a blood clot. The majority of clotting factors are produced
by the liver (except for factor III, IV and VIII) as precursors of proteolytic enzymes that circulate
in an inactive form and are activated in the coagulation pathway to exert their actions. The
Page 75
clotting factors involved in the coagulation pathway are summarized in Table 2.4.
Coagulation pathway
The coagulation pathway is classified into the intrinsic and extrinsic pathways to explain
the in vitro coagulation tests. When clotting times were measured after the clotting was initiated
by surface contact of glass, it is called contact activation or intrinsic pathway and when the
clotting was initiated by thromboplastin, it is called tissue factor or extrinsic pathway. Then both
intrinsic and extrinsic pathways activate the final common pathway of factor X, thrombin and
fibrin(89). However, this intrinsic/extrinsic classification may not fully explain the in vivo
coagulation process(90). The extrinsic pathway is considered the first step in plasma-mediated
hemostasis. It is activated by tissue factors (TF) expressed in subendothelial tissue. In normal
conditions, the vascular endothelium minimizes contact between TF and plasma procoagulants.
However vascular injury exposes TF to factor VII to form the TF-VIIa complex that promotes the
conversion of factor X to Xa. Concurrently, the intrinsic pathway is activated after vascular injury
as a parallel pathway. It begins with activation of prekallekerin, kininogen, HMW, and factor XII.
Factor XIIa further activates factor IX, which then acts with its cofactor (factor VIIIa) to form a
complex on the phospholipid surface to activate factor X. In the common pathway, factors Xa
along with its cofactor (factor Va), tissue phospholipids, platelet phospholipids and calcium forms
the prothrombinase complex to convert prothrombin to thrombin. Thrombin further converts
circulating fibrinogen to insoluble fibrin monomer and polymer. With factor XIIIa, fibrin polymer
incorporates itself into the platelet plug to create a fibrin network to stabilize a blood clot.
Common laboratory measures of the coagulation pathway
Page 76
Common laboratory measures that reflect activity of coagulation pathway are
prothrombin time (PT), activated partial thromboplastin time (APTT), fibrinogen (FIB),
prothrombin fragment 1+2, procoagulation factors (factor VII, VIII), and thrombin-anti-thrombin
complex
PT measures the extrinsic and common coagulation pathways. PT reflects activity of
procoagulation factors (factor VII, X, V, II) and fibrinogen. Increased or prolonged PT is related to
factor VII deficiency, a bleeding disorder with prolong and excessive bleeding, while decreased
or shortened PT is found in VTE. (Figure 2.7)
APTT measures intrinsic and common coagulation pathways, and reflects activity of
procoagulation factors (factor XII, XI, IX, VIII). (Figure 2.8)
Increased or prolonged APTT is related to factor XII, XI, IX, and VIII deficiencies, while
decreased or shortened APPT is found in VTE.
FIB is a clotting factor in the common coagulation pathway. Increased FIB is associated
with both venous thrombosis and arterial thrombosis, especially CHD. (Figure 2.9)
Anti-coagulation pathway
The anti-coagulation pathway inhibits the coagulation pathway to maintain balance of
the blood coagulation system. The key elements related to the anti-coagulation process are
tissue factor pathway inhibitor (TFPI), protein C (PROTC), protein S (PROTS), antithrombin (AT),
and thrombomodulin(90). TFPI is an anticoagulant protein that inhibits early phases of the
procoagulant response of TF-VIIa complex and factor Xa. It is a serine protease with properties
of anti-coagulation, profibrinolytic and anti-inflammation. TPFI is activated by
thrombomodulin-thrombin complex to form activated protein C (APC) and works with PROTS
Page 77
to inhibit factor Va and VIIIa. In addition to working with PROTC, PROTS degrades procoagulant
factor Xa and VIII. AT is a serine protease acting as the main inhibitor of thrombin. It inactivates
thrombin, factor IXa, Xa, Xia and XIIa. AT is activated when binding with heparin on the
endothelial cell surface. Thrombomodulin is a transmembrane receptor on the endothelial
cells. It binds to thrombin and forms thrombomodulin-thrombin complex to prevent the
formation of blood blot.
Common laboratory measures of the anti-coagulation pathway
Common laboratory measures that reflect activity of anti-coagulation pathway are AT,
PROTC and PROTS.
As AT is an anticoagulant, decreased levels of AT level are found in VTE. (Figure 2.10)
Decreased levels of PROTC and PROTS are found with VTE. (Figure 2.11)
Fibrinolysis pathway
The fibrinolytic pathway is a parallel system along with the coagulation and anti-
coagulation pathways, and acts to limit the size of a blood clot. The fibrinolytic pathway is an
enzymatic process that dissolves the fibrin clot into fibrin degradation products (FDPs) and D-
dimers by plasmin. Plasmin is converted from plasminogen with catalyzation by tissue
plasminogen activator (tPA) and urokinase plasminogen activator (uPA), which are released from
the vascular endothelium and are inhibited irreversibly by plasminogen activator inhibitor (PAI-
1). Plasmin activity is also regulated by its inhibitor, -2 antiplasmin.
Common laboratory measures of the fibrinolytic pathway
Page 78
Common laboratory measures that reflect activity of fibrinolytic pathway are fibrin
degradation products (FDPs), D-dimers, fibrinolytic inhibitor: Plasminogen activator (PAI-1),
plasminogen activity, tissue plasminogen activator (tPA) and plasmin-antiplasmin complex
FDPs assess the activity of fibrinolytic system. Increased level of FDPs reflects fibrinolysis.
D-dimers are produced by digestion of cross-linked fibrin. They are specific indicators of
fibrinolysis, and are elevated in VTE
PAI-1 is a profibrinolytic inhibitor, which increases both venous and arterial thrombosis.
2.5.5 Effect of hormone therapy on the hemostatic system
It is difficult to quantify the thrombosis effect of HT by epidemiologic methods due to a
relatively low incidence of clinical thrombotic events and variable presentation of arterial and
venous thrombosis. Therefore, studies on the effect of HT on the hemostatic system were used
to comment on the thromboembolic potential of HT. Changes in several coagulation factors are
reported to be associated with venous and arterial thrombosis. Specifically, elevated
fibrinogen(91-95) and factor VII activity(92, 95) have been reported to be risk factors for CHD and
stroke. Increased PAI-1 activity and tPA level are also related to increased CHD risk(96-98).
Decreased anti-coagulation levels including AT, PROTC, and PROTS are associated with elevated
risk of venous thrombosis(99).
Among all coagulation factors, fibrinogen has been most commonly reported as a strong
risk factor for CHD, as fibrinogen is a major coagulation protein in blood by mass, the precursor
of fibrin and an important determinant of blood viscosity and platelet aggregation. A meta-
analysis of 18 studies involving 4018 CHD cases estimated a CHD relative risk of 1.8 (95%CI, 1.6-
2.0) per 1 g/L (2.0 pmol/L) increase in plasma fibrinogen(100).
Page 79
Another meta-analysis from 31 prospective studies among 154,211 participants with 1.38
million person-years of follow up further explored the association between fibrinogen and CHD
with adjustment for known cardiovascular risk factors. The study reported moderately strong
associations between plasma fibrinogen and CHD risk with adjustment for age and sex (HR=2.42,
95%CI, 2.24-2.60 per 1 g/L increase in plasma fibrinogen)(101). The association was strongest
among younger participants as age-specific associations for participant aged 40-59 years, 60-69
years and 70 years and above were 2.93 (95%CI, 2.59-3.31), 2.18 (95%CI, 1.92-2.48) and 2.05
(95%CI, 1.74-2.41), respectively. This association did not differ substantially according to sex,
smoking, blood pressure, lipid levels or study design.
Although the clinical relevance of changes of certain coagulation factors and thrombosis
is inconclusive, the effect of HT on coagulation factors may in part explain the effect of HT on VTE
and CHD. HT is associated with multiple changes in the hemostatic system including coagulation,
anti-coagulation and fibrinolysis. However, individual changes in coagulation factors associated
with HT are minor and the levels of coagulation factors are usually maintained within the normal
range in the presence of HT(102). Although many studies have explored the effect of HT on
coagulation factors, the results are not consistent, likely due to differences in study design,
method of what, and measurement of different coagulation factors. Table 2.5 summarized
results from the clinical trials of HT effect on various coagulation factors.
Effect of hormone therapy on the coagulation pathway
HT has been consistently shown to increase prothrombin fragment 1+2(35, 103-108),
although the increase was not significant in some studies(109-112); the increase is estrogen dose-
dependent(103).
Page 80
Many studies reported that HT significantly decreased fibrinogen levels compared with
baseline levels or placebo(36, 56, 110, 111, 113). Some studies showed non-significantly
decreased fibrinogen levels with use of HT(36, 107, 108, 110-112, 114). Only one study showed
a significant increase in fibrinogen with HT(115).
While some studies showed that HT increased procoagulation factors (factor VII, IX, X, XII
and XIII), other studies including a meta-analysis of clinical trials showed that HT significantly
decreased some factors (factor VII). From our review, studies reported either increased(114, 116)
or decreased(36, 110, 111, 113) factor VII level from HT. Several studies reported no association
between factor VIII level and HT use(36, 117).
Studies evaluating the effect of HT on PT and APTT are scarce and the results are
inconclusive as some studies reported no association of these measures with HT use while some
reported increase or decrease PT, APTT with HT use.
Effect of hormone therapy on the anti-coagulation pathway
The effect of HT on anti-coagulation factors are consistent across the majority of studies.
HT significantly decreased AT levels(36, 107, 108, 112, 113, 116, 117) as well as decreased
thrombin-antithrombin complexes(35, 104, 110, 111, 117). PROC and PROTS levels were also
significantly reduced with any HT regimen across studies(36, 110, 111, 113, 117).
Effect of hormone therapy on the fibrinolysis pathway
The activity of the fibrinolytic pathway is stimulated with HT as there were reported
increases in fibrinolytic products (D-dimer and FDPs) and decreases in profibrinolytic factors (tPA)
and fibrinolytic inhibitors (PAI-1). Several studies reported a significant increase in D-dimer
level(35, 106, 110, 111, 113) with HT, while some studies reported no significant association(35,
Page 81
110, 111, 118). One study showed that HT significantly increased level of FDPs(113). Overall,
studies showed that HT was associated with increased plasminogen activity(113, 114, 116, 117),
decreased tPA level(36, 110, 111, 113, 114, 119) and increased plasmin-antiplasmin complexes
level(113). In addition, the fibrinolytic inhibition process was limited with HT use as evaluated by
decreased PAI-1 level(36, 106-111, 113, 114, 118, 120).
A meta-analysis of 48 studies including 40-68 years old postmenopausal women (6,229
HT users and 24,974 non-users) explored the association between HT use on coagulation factors.
The study concluded that HT was associated with significantly decreased fibrinogen, factors VII,
AT, PROC, PROS but significantly increased plasminogen levels(117). The effect of HT differs with
different regimen (estrogen alone or in combination with progestins), route of administration
(oral or transdermal) as well as different estrogen preparation and types of progestins. The
results from this meta-analysis suggest that the decrease in procoagulation factors (fibrinogen,
factor VII) might explain the HT effect on reduced risk for arterial thrombosis such as CHD and
stroke. In addition, the increase in anti-coagulation factors (AT, PROTC, PROTS) from HT might
explain the higher incidence of venous thrombosis associated with HT(117).
In summary, HT may affect the hemostatic system through all three main pathways
(coagulation, anti-coagulation and fibrinolytic pathways). These changes might explain the effect
of HT on CHD and VTE event risk. Although many studies have evaluated the effect of HT on
coagulation factors, however, no study compared the effect of HT between early and late
postmenopausal women. In addition, all studies calculated the difference in these coagulation
factors either between pre- and post-treatment periods or compared with placebo. Our
proposed study, the effect of HT on coagulation factors among early and late postmenopausal
Page 82
women will explore if there are different HT effects on coagulation factors using the rate of
change in repeated measurements over a one-year period in the REPLENISH. The repeated
measures should reduce measurement error of HT-associated changes compared to simple
pre/post treatment period comparisons.
2.5.6 Measurement of coagulation and anti-coagulation factors (REPLENISH)
In REPLENISH, coagulation and anti-coagulation factors including prothrombin time (PT),
activated partial thromboplastin time (APTT), antithrombin (AT), fibrinogen (FIB), protein C
(PROTC), and protein S (PROTS) were measured at screening visit, week 12, month 6, month 9,
and month 12.
Page 83
2.6 OUTCOME: ATHEROSCLEROSIS
2.6.1 Imaging technique for atherosclerosis
Atherosclerosis is a generalized inflammatory disorder characterized by an accumulation
of lipids, inflammatory cells and scar tissue covered by a fibrous cap within the wall of
arteries(121). Atherosclerosis in larger- and medium-sized elastic and muscular arteries can lead
to arterial obstruction and ischemia of the heart resulting in infarction or CHD.
As a progressing process through several stages, atherosclerosis can be measured by
different imaging techniques. At early stage, it is characterized by endothelial injury and
inflammatory response, which may be manifested as a change in vascular reactivity or
stiffness(121). Among patients with clinical symptom of CHD with advanced atherosclerosis
lesion, coronary angiography is the method of choice in clinical practice to assess the
atherosclerosis lesion. The quantitative coronary angiography (QCA) indirectly measures the
change in the arterial wall by visualizing the vascular lumen. The QCA is well correlated with
clinical CHD and used as a direct measure for CHD progression(122), however, this method is
limited to symptomatic patients according to exposure to radiation and high cost.
Several techniques have been used to assess the early stages of atherosclerosis at the
time when the lumen is not obstructed. Intravascular ultrasound could visualize the coronary
arterial wall but is too invasive to perform in asymptomatic patients. Magnetic resonance
imaging is non-invasive but expensive and does not have the speed and resolution to visualize
coronary artery. Extra-thorax ultrasonography is limited to detect the constant motion of
coronary arteries and the visual is interfered with bony structures. Fast computerized
Page 84
tomography, although can be used to measure vascular calcium but does not have the resolution
to directly measure the arterial wall.
Due to difficulties in imaging coronary arteries, the peripheral blood vessels is used as
vascular end points for assessing generalized atherosclerosis as the lesions across major arterial
beds have been shown to parallel each other. Carotid artery is used as a surrogate point as it lies
at a shallow tissue path yielding easy approach and the degree of atherosclerosis in the carotid
artery correlates with that in the coronary arteries and the abdominal aorta(123).
Carotid intima media thickness (CIMT) measured with B-mode ultrasound has been used
as a primary indicator of atherosclerosis in the common carotid artery (CCA) since 1986(124).
CIMT has been associated with traditional cardiovascular risk factors such as age, sex, race,
smoking, alcohol consumption, habitual endurance exercise, blood pressure, dyslipidemia,
dietary patterns, risk-lowering therapy, glycemia, hyperuricemia, obesity-related anthropometric
parameters, obesity and obesity-related disease. In addition, CIMT has been associated with
novel risk factors for cardiovascular disease including heredity, some genotypic indices of CHD
(telomere shortening, haptoglobin genotype), anthropometric cardiovascular parameters (level
of coronary artery calcification, the coronary SYNTAX score), rheumatoid arthritis, immunological
diseases, inflammatory cytokines, lipid peroxidation, anthropometric hemocyte parameters
(neutrophil to lymphocyte ratio), infectious diseases, serum vitamin D level, and matrix
metalloproteinases(125).
The right CCA originates from the truncus brachiocephalicus and the left CCA originates
from the aortic arch. The CCA has a superficial course and bifurcates into internal and external
carotid arteries at the level of the forth cervical vertebra. Histologically, the CCA is composed of
Page 85
3 layers; tunica intima, tunica media and tunica externa or adventitia. The CIMT measured by B-
mode ultrasound is defined as the viewable distance between the lumen-intima and the media-
adventitia interface(126) (Figure 2.12).
2.6.2 Carotid intima media thickness and coronary heart disease
Several studies have reported associations in the expected direction between CIMT and
cardiovascular risk factors and prospectively-assessed cardiovascular events. Systematic review
and meta-analysis of the cohort studies on the association of baseline CIMT and first-time
cardiovascular events during follow-up was conducted. Results from a meta-analysis of 15 studies
from USA, Europe and China among 28,177 subjects also showed a positive association between
baseline CIMT and CHD with similar effect size. The included cohort studies were conducted
among participants age 19-90 years with mean age of 55 years. After adjusting for age and sex,
the summarized estimates showed the association of CIMT and future myocardial (MI) and 1
standard deviation (SD) (0.17 mm) increase in CIMT (HR=1.26, 95%CI, 1.20-1.31) and 0.1 mm
increase in CIMT (HR=1.15, 95%CI, 1.12-1.18)(127)(Figure 2.13). Another meta-analysis on 14
population-base cohort studies among 45,828 subjects with a median follow-up of 11 years
explored the association between baseline CIMT and a first-time MI. The study reported
significant association between baseline CIMT and first time MI risk after adjustment for the
cardiovascular risk factors of the Framingham Risk Score (HR=1.08, 95%CI, 1.05-1.11 with 0.1 mm
increase in CIMT)(128). The adjustment for cardiovascular risk factors may assess the added value
of CIMT in prediction of future MI. However, as CIMT is associated with several cardiovascular
risk factors, it leads to collinearity in the model, which results in model overfitting and inaccurate
effect estimates.
Page 86
Atherosclerosis is the main trigger of overall cardiovascular disease. Subclinical or early
stage of atherosclerosis is when there is no present clinical symptom. It could be measured by
endothelial dysfunction, carotid plaque or CIMT. As CIMT is a non-invasive procedure to assess
subclinical atherosclerosis and a useful measure as a surrogate end point for CHD(129). As an
absolute value of CIMT, a case-control study among patients with chest pain who underwent
coronary angiography reported that CIMT equal to or greater than 0.85 mm had a positive
predictive value of 83% for CHD(130).
CIMT progression was also reported to significantly predict recurrent coronary events
among patients who previously had coronary artery bypass graft surgery (p<0.02)(131). For each
0.03 mm increase per year in CIMT (that was measured longitudinally, every 6 months over 2
years), the relative risk for non-fatal CHD or death was 2.2 (95%CI, 1.4-3.6) and the relative risk
for any coronary events was 3.1 (95%CI, 2.1-4.5). In addition, a study showed that there was a
significant correlation between change in CIMT and change in coronary artery atherosclerosis
measured by quantitative coronary angiogram(132). This correlation analysis was conducted
within the Cholesterol Lowering Atherosclerosis Study, and RCT including 188 male subjects with
prior CHD. The correlation in longitudinally-assessed progression of CIMT and coronary
atherosclerosis was significant among all coronary lesions and mild/moderate-sized coronary
lesions (<50% diameter stenosis) but not severe coronary lesion ( 50% diameter stenosis). In
mild/moderate coronary lesions, the correlations between change in CIMT and percent diameter
stenosis (r=0.28, p=0.002), minimum lumen diameter (r=-0.28, p-=0.002) and vessel edge
roughness (r=0.25, p=0.003) were statistically significant. The result from this study suggested
that CIMT may be a useful non-invasive procedure to assess atherosclerosis in clinical trials(132).
Page 87
2.6.3 Estrogen effect on carotid intima media thickness
The known association of menopause with elevated risk of CHD may be explained by
progressive atherosclerosis following the menopause-related decline of estrogen from levels in
the premenopausal period. Subclinical atherosclerosis progression after menopause was studied
among women with and without oophorectomy from the Women’s Isoflavone Soy Health (WISH)
trial(133). Compared to baseline, the CIMT rate increased during the follow-up period of 3 years
among all menopausal women. As time from menopause transition increased, retained ovaries
are associated with a slower rate of CIMT compared with menopausal women with
oophorectomy. The result suggested that retained ovaries has a beneficial effect in slowing the
rate of atherosclerosis measured by CIMT.
In addition, an analysis from the Study of Women’s Health Across the Nation (SWAN)
reported progression in CIMT during menopause transition(134). The SWAN study is an ongoing
longitudinal multi-ethnic study of changes during the menopause transition. The study recruited
3320 women age 42-52 years who did not use HT from 7 study sites across US in the year 1996.
An analysis on effect of menopause on subclinical atherosclerosis was conducted among 257
women from SWAN ancillary study who had up to 5 carotid scan over a maximum of 9 years of
follow-up (median 3.7years). Menopause stages were defined as premenopause (monthly
bleeding with no change in cycle interval in the past year), early peri-menopause (monthly
bleeding with change in cycle interval but at least one menstrual period within the past 3 months,
late peri-menopause (3 consecutive months of amenorrhea), postmenopause (12 consecutive
months of amenorrhea). The overall rate of change in CIMT was 0.007 mm/year. The CIMT
progression increased substantially in late peri-menopause stage (0.17 mm/year) compared to
Page 88
both premenopausal (0.007 mm/year) and early peri-menopausal stage (0.005 mm/year)
(p<0.05) after adjusted for age and race. Compared to premenopausal stage, postmenopausal
stage was independently associated with higher level of CIMT (p<0.05).
Several studies reported the effect of estrogen therapy on CIMT progression and the
significant effect appears to be limited to postmenopausal women who initiated treatment at
the time of menopause or soon after menopause.
The PHOREA trial, a randomized single-blinded, placebo-controlled study of estradiol
effect on CIMT progression reported no difference in CIMT progression among 3 treatment
groups of 17 -E2 with gestodene 0.025 mg for 12 days per cycle, 17 -E2 with daily gestodene
0.025 mg or placebo (p>0.2)(135). This study included 321 postmenopausal women 40-70 years
who had CIMT > 1 mm, which reflected baseline advanced atherosclerosis lesion, hence cold lead
to null result. In addition, the trial followed the effect in 48 weeks, which is not likely to see the
treatment effect.
In the EPAT trial that randomized 222 healthy postmenopausal women with a LDL
cholesterol of 130 mg/dl or greater and a mean age of 62.2 years , oral 17 -E2 1 mg per day
resulted in lower CIMT rate (-0.0017 mm/year) compared to placebo (0.0036 mm/year)(74). The
mean difference of CIMT rate between estradiol and placebo was 0.0053 mm/year (95%CI,
0.0001-0.0105 mm/year) (p=0.046). Moreover, the stronger result was found among participants
who did not take lipid-lowering medication with the mean difference of CIMT rate between
estradiol and placebo of 0.0147 mm/year (95%CI, 0.0055-0.0240 mm/year) (p=0.002).
A cross-sectional study evaluated the thickness of carotid intima and media among 17
long-term estrogen users (initiated therapy at time of menopause with mean duration of use 20
Page 89
years), 17 postmenopausal non-users and 20 premenopausal women. It was reported that the
long-term estrogen users had significantly thinner carotid intima layer (-25%, p=0.0002), thicker
media layer (74%, p=0.0002) and lower intima/media thickness (I/M) ratio (-54%, p<0.0001)
compare with age-matched non-users. The values among estrogen users were similar to those of
premenopausal women. These data indicated that long-term use of estrogen therapy initiated at
the time of menopause preserves a thin intima and maintains the media thickness and I/M ratio
at values similar to those in premenopausal women(136).
The Kronos Early Estrogen Prevention Study (KEEPS) trial did not find a significant effect
of estrogen on CIMT rate (measured annually during 4 years of follow-up) compared to placebo.
During 48 months of treatment, mean CIMT increased at a rate of 0.007 mm/year similarly among
all treatment groups (oral CEE 0.45 mg/day, transdermal 17- E2 50 mcg/day, each with 200 mg
oral P4 for 12 days per month or placebo)(137). A recent report comparing the CIMT rate among
KEEPS trial participants after cessation of therapy showed that the CIMT rate (measured annually
during 4 years of study period and at 3 years after cessation of study treatment) was significantly
higher than those on-treatment period of oral CEE (0.010, 95%CI, 0.002-0.017) but not with
transdermal 17- E2 or placebo(138). The data suggested that cessation of higher dose estrogen
therapy increase CIMT rate while had no effect in lower estrogen formulation.
The ELITE trial, an RCT specifically designed to test HT timing hypothesis on subclinical
atherosclerosis measured by CIMT rate during a median of 5 years of follow-up reported a
different estrogen effect on CIMT rate between early and late postmenopausal women. Among
women who were less than 6 years since menopause, mean CIMT rate in estradiol group (0.0044
mm/year) was significantly lower than the placebo group rate (0.0078 mm/year) (p=0.008).
Page 90
Among women who were 10 years or more since menopause, the mean CIMT rate in the estradiol
group (0.0100 mm/year) was similar to the placebo group (0.0088 mm/year) (p=0.29)(2).
The null results of 0.5 mg E2 on CIMT in KEEPS in relation to significant 1.0 mg E2 effect
on significant reduction of CIMT in both EPAT and ELITE could be explained by different doses of
E2 being studied. The dose response analysis of estrogen on atherosclerosis showed that
increasing doses of estrogen significantly reduced CIMT among hypogonadal women (p
trend=0.001) (139).
2.6.4 Measurement of carotid intima media thickness (ELITE)
In the ELITE trial, CIMT was assessed by means of computer image processing of B-mode
ultrasonograms that were obtained at two baseline assessments and every 6 months during the
trial follow-up. High-resolution B-mode ultrasonographic imaging and CIMT measurement were
performed with the use of standardized procedures and with technology that was specifically
developed in house for longitudinal measurements of changes in atherosclerosis(74, 140-142).
Using the two CIMT baseline CIMT measurements, the short-term coefficient of variation for
baseline CIMT measurement was 0.69%; the coefficient of variation for repeated CIMT
measurement is typically<3% and often approaches 1%(1, 2).
Development of carotid intima media thickness measurement method
In the past the measurement of CIMT relied on human visual judgement to manually
identify echo coordinates with a computer pointing device. This manual method is subject to
measurement error as limited by human variability in operation of the pointing devices and by
the resolution of the displayed ultrasound image. Therefore, an automated measurement
method using a computerize edge detection algorithm with internal standardization of
Page 91
anatomical landmarks was developed to improve accuracy and precision of CIMT
measurement(122, 140, 141). This automated method was tested and found to be equally
precise over short (60 days) and long intervals (48 months). In addition, the automated
computerized edge tracking method had 2-4 times lower measurement variability compared with
a manual method of edge detection (140). Subsequently, an improved automated CIMT
measurement method was developed to further improve reproducibility of CIMT measurement.
This new method utilized computerized edge tracking-multi-frame image processing that
automatically measured arterial diameter and CIMT in multiple sequential frames spanning
several cardiac cycles.
Carotid intima media thickness measurement method used in ELITE
The primary end point in ELITE was the rate of change in the right distal CIMT on computer
image-processed B-mode ultrasonograms. The images were acquired via a linear array 7.5 MHz
transducer attached to a Siemen Acuson CV70 (Mountain View, California). The ultrasound was
performed along with a single lead electrocardiogram (ECG). The carotid artery was imaged
transversely and then longitudinally with the jugular vein stacked above the carotid artery (Figure
2.14 and Figure 2.15); this imaging technique is intended to standardize positioning of the
ultrasound probe for longitudinal image collections. For each individual, probe angulation
(identified by internal anatomical landmarks) and instrumental setting (depth of field, gain, input
power, dynamic range, monitor intensity) were maintained for all follow-up measurements. All
images were recorded from a personal computer with a video digitizer board interfaced to an
SVHS video tape recorder for automated multi-frame CIMT measurement. CIMT was determined
as the average of 70-100 CIMT measurements between the intima-lumen and media-adventitia
Page 92
echo interface along a 1 cm length just proximal to the carotid artery bulb at the same point in
the cardiac cycle standardized to the ECG signal. This procedure standardizes the timing, location
and distance over which CIMT is measured to ensure comparability within and across
participants(140, 141)
Page 93
2.7 TABLES AND FIGURES
Page 94
Table 2.1 Risk factors for thrombosis stratified by Virchow’s triads(79)
Page 95
Table 2.2 Studies evaluating association between hormone therapy and venous thromboembolism (VTE)(80)
Page 96
Table 2.3 Women’s Health Initiative trial results on risk ratio for venous thrombosis stratified by age and time-since-
menopause (33, 86, 87, 143, 144)
Page 97
Table 2.3 (Continue)
*CEE=conjugated equine estrogen, MPA=medroxyprogesterone acetate, VTE=venous thromboembolism, PE=pulmonary embolism,
DVT=deep vein thrombosis
Page 98
Table 2.4 Summary of coagulation proteins/clotting factors(90)
Page 99
Table 2.5 List of studies of effect of hormone therapy on coagulation, anti-coagulation and fibrinolytic factors
Page 100
Table 2.5 (continue)
Page 101
Table 2.5 (continue)
Page 102
*RCT=randomized clinical trial, E2=estradiol, CEE=conjugated equine estrogen, MPA=medroxyprogesterone acetate, MP=micronized
progesterone, green hi-lights indicate significantly increased levels, red hi-lights indicate significantly decreased levels, blue hi-lights
indicate no change in levels
Page 103
Figure 2.1 Structure of equine estrogens(24)
Page 104
Figure 2.2 Estrogen receptors and their pathways(16)
*(i) ER-dependent, nuclear initiated estrogen signaling, (ii) ER-dependent, membrane initiated estrogen signaling, (iii) the ER-
independent pathway; SERMs=selective estrogen receptor modulators, ER=estrogen receptor, ERE=estrogen response element,
CoRegs=coregulatory proteins, EBPs=estrogen-binding proteins
Page 105
Figure 2.3 Estrogen effect on coronary heart disease (CHD) in early atherosclerogenesis versus established
atherosclerosis(25)
*LDL=low density lipoprotein, CAMs=cell adhesion molecules, MCP-1=monocyte chemoattractant protein 1, TNF-α=tumor
necrosis factor-α, VSMC=vascular smooth muscle cell, MMP=matrix metalloproteinase, COX-2=cyclooxygenase 2
Page 106
Figure 2.4 Lipoprotein metabolism pathways (145)
Page 107
Figure 2.5 Role of estrogen in lipid peroxidation in the etiology of atherosclerosis in postmenopausal women (24)
Page 108
Figure 2.6 Virchow’s triad (146)
Page 109
Figure 2.7 Coagulation pathway tested by prothrombin time (PT)(147)
Page 110
Figure 2.8 Coagulation pathway tested by activated partial thrombin time (APTT)(147)
Page 111
Figure 2.9 Coagulation pathway tested by fibrinogen(147)
Page 112
Figure 2.10 Anti-coagulation pathway involving antithrombin(147)
Page 113
Figure 2.11 Anti-coagulation pathway involving protein C (PROTC) and protein S (PROTS)(147)
Page 114
Figure 2.12 Ultrasound image of longitudinal axis of common carotid artery (CCA) demonstrates the carotid intima media
thickness (CIMT) which lies between the lumen-intima and the media-adventitia interface(126)
Page 115
Figure 2.13 Forest plot of hazard ratios for coronary heart disease (CHD) A: per 1 SD increment in common carotid artery
intima-media thickness (CIMT), B: per 1 mm increment in CIMT(127)
Page 116
Figure 2.14 Anatomy of common carotid artery and carotid artery bulb (transverse and longitudinal view)
Page 117
Figure 2.15 Proper position for ultrasound image for common carotid artery intima-media thickness (CIMT) measurement,
which includes common carotid artery and jugular vein
Page 118
2.8 REFERENCES
1. Hodis HN, Mack WJ, Shoupe D, Azen SP, Stanczyk FZ, Hwang-Levine J, et al. Methods and
baseline cardiovascular data from the Early versus Late Intervention Trial with Estradiol testing
the menopausal hormone timing hypothesis. Menopause-the Journal of the North American
Menopause Society 2015 Apr;22(4):391-401.
2. Hodis HN, Mack WJ, Henderson VW, Shoupe D, Budoff MJ, Hwang-Levine J, et al. Vascular
Effects of Early versus Late Postmenopausal Treatment with Estradiol. New England Journal of
Medicine 2016;374(13):1221-31.
3. Pickar JH, Bon C, Amadio JM, Mirkin S, Bernick B. Pharmacokinetics of the first combination
17beta-estradiol/progesterone capsule in clinical development for menopausal hormone
therapy. Menopause 2015 Dec;22(12):1308-16.
4. Mirkin S, Amadio JM, Bernick BA, Pickar JH, Archer DF. 17beta-Estradiol and natural
progesterone for menopausal hormone therapy: REPLENISH phase 3 study design of a
combination capsule and evidence review. Maturitas 2015 May;81(1):28-35.
5. 28th Annual Meeting of The North American Menopause Society October 11-14, 2017,
Philadelphia, PA. Menopause 2017 Dec;24(12):1418-61.
6. Constantine G, Revicki DA, Kagan R, Simon J, Graham S, Mirkin S. TX-001HR is Associated with
a Clinically Meaningful Effect on Vasomotor Symptoms (28th Annual Meeting of The North
American Menopause Society October 11-14, 2017, Philadelphia, PA.). Menopause 2017
Dec;24(12):1428.
Page 119
7. Lobo RA, Archer DF, Constantine G, Pickar JH, Kaunitz A, Graham S, et al. 17b-
Estradiol/Progesterone in a Single Oral Softgel Capsule (TX-001HR) Significantly Reduced
Moderate-to-Severe Vasomotor Symptoms without Endometrial Hyperplasia (28th Annual
Meeting of The North American Menopause Society October 11-14, 2017, Philadelphia, PA.).
Menopause 2017 Dec;24(12):1430.
8. Kagan R, Constantine G, Kaunitz A, Gasper G, Bernick B, Mirkin S. TX-001HR Improved the
Medical Outcomes Study-Sleep (MOS-Sleep) questionnaire in Menopausal Women with
Vasomotor Symptoms (28th Annual Meeting of The North American Menopause Society
October 11-14, 2017, Philadelphia, PA.). Menopause 2017 Dec;24(12):1431.
9. Goldstein SR, Constantine G, Archer DF, Pickar JH, Graham S, Bernick B, et al. Effects of TX-
001HR on Uterine Bleeding Rates in Menopausal Women with Vasomotor Symptoms (28th
Annual Meeting of The North American Menopause Society October 11-14, 2017, Philadelphia,
PA.). Menopause 2017 Dec;24(12):1431.
10. Grady D, Rubin SM, Petitti DB, Fox CS, Black D, Ettinger B, et al. Hormone therapy to prevent
disease and prolong life in postmenopausal women. Annals of internal medicine 1992 Dec
15;117(12):1016-37.
11. Coelingh Bennink HJ, Holinka CF, Diczfalusy E. Estetrol review: profile and potential clinical
applications. Climacteric : the journal of the International Menopause Society 2008;11 Suppl
1:47-58.
12. Coelingh Bennink HJ, Verhoeven C, Zimmerman Y, Visser M, Foidart JM, Gemzell-Danielsson
K. Clinical effects of the fetal estrogen estetrol in a multiple-rising-dose study in
postmenopausal women. Maturitas 2016 Sep;91:93-100.
Page 120
13. Coelingh Bennink HJT, Verhoeven C, Zimmerman Y, Visser M, Foidart JM, Gemzell-
Danielsson K. Pharmacodynamic effects of the fetal estrogen estetrol in postmenopausal
women: results from a multiple-rising-dose study. Menopause 2017 Jun;24(6):677-85.
14. Mattsson LA, Skouby S, Rees M, Heikkinen J, Kudela M, Stadnicki-Kolendo A, et al. Efficacy
and tolerability of continuous combined hormone replacement therapy in early
postmenopausal women. Menopause Int 2007 Sep;13(3):124-31.
15. von Schoultz B. Clinical efficacy and safety of combined estradiol valerate and dienogest: a
new no-bleed treatment. Climacteric : the journal of the International Menopause Society 2003
Aug;6 Suppl 2:24-32.
16. Deroo BJ, Korach KS. Estrogen receptors and human disease. J Clin Invest 2006
Mar;116(3):561-70.
17. Losordo DW, Kearney M, Kim EZ, Jekanowski J, Isner JM. Variable Expression of the
Estrogen-Receptor in Normal and Atherosclerotic Coronary-Arteries of Premenopausal Women.
Circulation 1994 Apr;89(4):1501-10.
18. Post WS, Goldschmidt-Clermont PJ, Wilhide CC, Heldman AW, Sussman MS, Ouyang P, et al.
Methylation of the estrogen receptor gene is associated with aging and atherosclerosis in the
cardiovascular system. Cardiovasc Res 1999 Sep;43(4):985-91.
19. Zhu Y, Bian Z, Lu P, Karas RH, Bao L, Cox D, et al. Abnormal vascular function and
hypertension in mice deficient in estrogen receptor beta. Science 2002 Jan 18;295(5554):505-8.
20. Schuit SC, Oei HH, Witteman JC, Geurts van Kessel CH, van Meurs JB, Nijhuis RL, et al.
Estrogen receptor alpha gene polymorphisms and risk of myocardial infarction. Jama 2004 Jun
23;291(24):2969-77.
Page 121
21. Peter I, Shearman AM, Vasan RS, Zucker DR, Schmid CH, Demissie S, et al. Association of
estrogen receptor beta gene polymorphisms with left ventricular mass and wall thickness in
women. American journal of hypertension 2005 Nov;18(11):1388-95.
22. Chambliss KL, Shaul PW. Estrogen modulation of endothelial nitric oxide synthase. Endocr
Rev 2002 Oct;23(5):665-86.
23. Chen Z, Yuhanna IS, Galcheva-Gargova Z, Karas RH, Mendelsohn ME, Shaul PW. Estrogen
receptor alpha mediates the nongenomic activation of endothelial nitric oxide synthase by
estrogen. J Clin Invest 1999 Feb;103(3):401-6.
24. Bhavnani BR, Stanczyk FZ. Pharmacology of conjugated equine estrogens: Efficacy, safety
and mechanism of action. The Journal of steroid biochemistry and molecular biology 2014
2014/07/01/;142:16-29.
25. Mendelsohn ME, Karas RH. Molecular and cellular basis of cardiovascular gender
differences. Science 2005 Jun 10;308(5728):1583-7.
26. Luckas MJ, Gleeve T, Biljan MM, Buckett WM, Aird IA, Drakeley A, et al. The effect of
progestagens on the carotid artery pulsatility index in postmenopausal women on oestrogen
replacement therapy. Eur J Obstet Gynecol Reprod Biol 1998 Feb;76(2):221-4.
27. Williams JK, Honore EK, Washburn SA, Clarkson TB. Effects of hormone replacement therapy
on reactivity of atherosclerotic coronary arteries in cynomolgus monkeys. J Am Coll Cardiol
1994 Dec;24(7):1757-61.
Page 122
28. Mishra RG, Hermsmeyer RK, Miyagawa K, Sarrel P, Uchida B, Stanczyk FZ, et al.
Medroxyprogesterone acetate and dihydrotestosterone induce coronary hyperreactivity in
intact male rhesus monkeys. The Journal of clinical endocrinology and metabolism 2005
Jun;90(6):3706-14.
29. Clarkson TB, Appt SE. MPA and postmenopausal coronary artery atherosclerosis revisited.
Steroids 2003 Nov;68(10-13):941-51.
30. Wender MC, Vigo F, Oliveira PP, Bittellbrun F, Furtado AP, Schvartzman L, et al. Effect of
conjugated estrogen versus conjugated estrogen associated with medroxyprogesterone acetate
in postmenopausal women on internal carotid artery pulsatility index: a randomized pilot study.
J Obstet Gynaecol Res 2011 Jul;37(7):815-8.
31. Wakatsuki A, Okatani Y, Ikenoue N, Fukaya T. Effect of medroxyprogesterone acetate on
endothelium-dependent vasodilation in postmenopausal women receiving estrogen. Circulation
2001 Oct 9;104(15):1773-8.
32. Schierbeck LL, Rejnmark L, Tofteng CL, Stilgren L, Eiken P, Mosekilde L, et al. Effect of
hormone replacement therapy on cardiovascular events in recently postmenopausal women:
randomised trial. BMJ 2012 Oct 09;345:e6409.
33. Manson JE, Chlebowski RT, Stefanick ML, Aragaki AK, Rossouw JE, Prentice RL, et al.
Menopausal Hormone Therapy and Health Outcomes During the Intervention and Extended
Poststopping Phases of the Women's Health Initiative Randomized Trials. Jama-Journal of the
American Medical Association 2013 Oct 2;310(13):1353-68.
Page 123
34. Manson JE, Aragaki AK, Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, et al. Menopausal
Hormone Therapy and Long-term All-Cause and Cause-Specific Mortality: The Women's Health
Initiative Randomized Trials. Jama 2017 Sep 12;318(10):927-38.
35. Eilertsen AL, Qvigstad E, Andersen TO, Sandvik L, Sandset PM. Conventional-dose hormone
therapy (HT) and tibolone, but not low-dose HT and raloxifene, increase markers of activated
coagulation. Maturitas 2006 Oct 20;55(3):278-87.
36. Eilertsen AL, Sandvik L, Mowinckel MC, Andersen TO, Qvigstad E, Sandset PM. Differential
effects of conventional and low dose oral hormone therapy (HT), tibolone, and raloxifene on
coagulation and fibrinolysis. Thromb Res 2007;120(3):371-9.
37. Wingrove CS, Garr E, Godsland IF, Stevenson JC. 17beta-oestradiol enhances release of
matrix metalloproteinase-2 from human vascular smooth muscle cells. Biochim Biophys Acta
1998 Mar 5;1406(2):169-74.
38. Umetani M, Domoto H, Gormley AK, Yuhanna IS, Cummins CL, Javitt NB, et al. 27-
Hydroxycholesterol is an endogenous SERM that inhibits the cardiovascular effects of estrogen.
Nat Med 2007 Oct;13(10):1185-92.
39. Hwang J, Hodis HN, Hsiai TK, Asatryan L, Sevanian A. Role of annexin II in estrogen-induced
macrophage matrix metalloproteinase-9 activity: the modulating effect of statins.
Atherosclerosis 2006 Nov;189(1):76-82.
40. Probst-Hensch NM, Ingles SA, Diep AT, Haile RW, Stanczyk FZ, Kolonel LN, et al. Aromatase
and breast cancer susceptibility. Endocr Relat Cancer 1999 Jun;6(2):165-73.
41. Garcia M, Mulvagh SL, Merz CN, Buring JE, Manson JE. Cardiovascular Disease in Women:
Clinical Perspectives. Circ Res 2016 Apr 15;118(8):1273-93.
Page 124
42. Ritzel RM, Capozzi LA, McCullough LD. Sex, stroke, and inflammation: the potential for
estrogen-mediated immunoprotection in stroke. Horm Behav 2013 Feb;63(2):238-53.
43. Springer AM, Foster-Schubert K, Morton GJ, Schur EA. Is there evidence that estrogen
therapy promotes weight maintenance via effects on leptin? Menopause 2014 Apr;21(4):424-
32.
44. Mauvais-Jarvis F, Clegg DJ, Hevener AL. The role of estrogens in control of energy balance
and glucose homeostasis. Endocr Rev 2013 Jun;34(3):309-38.
45. Faulds MH, Zhao C, Dahlman-Wright K, Gustafsson JA. The diversity of sex steroid action:
regulation of metabolism by estrogen signaling. J Endocrinol 2012 Jan;212(1):3-12.
46. Monteiro R, Teixeira D, Calhau C. Estrogen signaling in metabolic inflammation. Mediators
Inflamm 2014;2014:615917.
47. Masood DE, Roach EC, Beauregard KG, Khalil RA. Impact of sex hormone metabolism on the
vascular effects of menopausal hormone therapy in cardiovascular disease. Curr Drug Metab
2010 Oct;11(8):693-714.
48. Reslan OM, Khalil RA. Vascular effects of estrogenic menopausal hormone therapy. Rev
Recent Clin Trials 2012 Feb;7(1):47-70.
49. Iorga A, Cunningham CM, Moazeni S, Ruffenach G, Umar S, Eghbali M. The protective role of
estrogen and estrogen receptors in cardiovascular disease and the controversial use of estrogen
therapy. Biol Sex Differ 2017 Oct 24;8(1):33.
50. Kingsbury KJ, Bondy G. Understanding the essentials of blood lipid metabolism. Prog
Cardiovasc Nurs 2003 Winter;18(1):13-8.
Page 125
51. Cifkova R, Krajcoviechova A. Dyslipidemia and cardiovascular disease in women. Curr
Cardiol Rep 2015 Jul;17(7):609.
52. Wilson PW, Abbott RD, Castelli WP. High density lipoprotein cholesterol and mortality. The
Framingham Heart Study. Arteriosclerosis 1988 Nov-Dec;8(6):737-41.
53. Gordon DJ, Probstfield JL, Garrison RJ, Neaton JD, Castelli WP, Knoke JD, et al. High-density
lipoprotein cholesterol and cardiovascular disease. Four prospective American studies.
Circulation 1989 Jan;79(1):8-15.
54. Walsh BW, Schiff I, Rosner B, Greenberg L, Ravnikar V, Sacks FM. Effects of postmenopausal
estrogen replacement on the concentrations and metabolism of plasma lipoproteins. The New
England journal of medicine 1991 Oct 24;325(17):1196-204.
55. Kannel WB, Hjortland MC, McNamara PM, Gordon T. Menopause and risk of cardiovascular
disease: the Framingham study. Annals of internal medicine 1976 Oct;85(4):447-52.
56. Miller VT, Larosa J, Barnabei V, Kessler C, Levin G, Smithroth A, et al. Effects of Estrogen or
Estrogen/Progestin Regimens on Heart-Disease Risk-Factors in Postmenopausal Women - the
Postmenopausal Estrogen/Progestin Interventions (Pepi) Trial. Jama-Journal of the American
Medical Association 1995 Jan 18;273(3):199-208.
57. Casanova G, Bossardi Ramos R, Ziegelmann P, Spritzer PM. Effects of low-dose versus
placebo or conventional-dose postmenopausal hormone therapy on variables related to
cardiovascular risk: a systematic review and meta-analyses of randomized clinical trials. The
Journal of clinical endocrinology and metabolism 2015 Mar;100(3):1028-37.
Page 126
58. Karim R, Mack WJ, Lobo RA, Hwang J, Liu CR, Liu CH, et al. Determinants of the effect of
estrogen on the progression of subclinical atherosclerosis: Estrogen in the Prevention of
Atherosclerosis Trial. Menopause 2005 Jul-Aug;12(4):366-73.
59. Karim R, Hodis HN, Stanczyk FZ, Lobo RA, Mack WJ. Relationship between serum levels of
sex hormones and progression of subclinical atherosclerosis in postmenopausal women. The
Journal of clinical endocrinology and metabolism 2008 Jan;93(1):131-8.
60. Kuller LH, Gutai JP, Meilahn E, Matthews KA, Plantinga P. Relationship of endogenous sex
steroid hormones to lipids and apoproteins in postmenopausal women. Arteriosclerosis 1990
Nov-Dec;10(6):1058-66.
61. LaRosa JC. Women, lipoproteins and cardiovascular disease risk. Can J Cardiol 1990 May;6
Suppl B:23B-9B.
62. Bhavnani BR, Tam SP, Lu X. Structure activity relationships and differential interactions and
functional activity of various equine estrogens mediated via estrogen receptors (ERs) ERalpha
and ERbeta. Endocrinology 2008 Oct;149(10):4857-70.
63. Koller E, Volf I, Gurvitz A, Koller F. Modified low-density lipoproteins and high-density
lipoproteins. From investigation tools to real in vivo players. Pathophysiol Haemost Thromb
2006;35(3-4):322-45.
64. Tang M, Abplanalp W, Ayres S, Subbiah MT. Superior and distinct antioxidant effects of
selected estrogen metabolites on lipid peroxidation. Metabolism 1996 Apr;45(4):411-4.
65. Sack MN, Rader DJ, Cannon RO, 3rd. Oestrogen and inhibition of oxidation of low-density
lipoproteins in postmenopausal women. Lancet 1994 Jan 29;343(8892):269-70.
Page 127
66. Wilcox JG, Hwang J, Hodis HN, Sevanian A, Stanczyk FZ, Lobo RA. Cardioprotective effects of
individual conjugated equine estrogens through their possible modulation of insulin resistance
and oxidation of low-density lipoprotein. Fertility and sterility 1997 Jan;67(1):57-62.
67. Bhavnani BR, Cecutti A, Gerulath A, Woolever AC, Berco M. Comparison of the antioxidant
effects of equine estrogens, red wine components, vitamin E, and probucol on low-density
lipoprotein oxidation in postmenopausal women. Menopause 2001 Nov-Dec;8(6):408-19.
68. Mc Manus J, Mc Eneny J, Thompson W, Young IS. The effect of hormone replacement
therapy on the oxidation of low density lipoprotein in postmenopausal women. Atherosclerosis
1997 Nov;135(1):73-81.
69. Bonnefont-Rousselot D, Therond P, Beaudeux JL, Peynet J, Legrand A, Delattre J. High
density lipoproteins (HDL) and the oxidative hypothesis of atherosclerosis. Clin Chem Lab Med
1999 Oct;37(10):939-48.
70. Parthasarathy S, Barnett J, Fong LG. High-density lipoprotein inhibits the oxidative
modification of low-density lipoprotein. Biochim Biophys Acta 1990 May 22;1044(2):275-83.
71. Perrella J, Berco M, Cecutti A, Gerulath A, Bhavnani BR. Potential role of the interaction
between equine estrogens, low-density lipoprotein (LDL) and high-density lipoprotein (HDL) in
the prevention of coronary heart and neurodegenerative diseases in postmenopausal women.
Lipids Health Dis 2003 Jun 20;2:4.
72. Godsland IF. Effects of postmenopausal hormone replacement therapy on lipid, lipoprotein,
and apolipoprotein (a) concentrations: analysis of studies published from 1974-2000. Fertility
and sterility 2001 May;75(5):898-915.
Page 128
73. Kalyan S, Hitchcock CL, Sirrs S, Pudek M, Prior JC. Cardiovascular and metabolic effects of
medroxyprogesterone acetate versus conjugated equine estrogen after premenopausal
hysterectomy with bilateral ovariectomy. Pharmacotherapy 2010 May;30(5):442-52.
74. Hodis HN, Mack WJ, Lobo RA, Shoupe D, Sevanian A, Mahrer PR, et al. Estrogen in the
prevention of atherosclerosis - A randomized, double-blind, placebo-controlled trial. Annals of
internal medicine 2001 Dec 4;135(11):939-53.
75. Little RR, Rohlfing CL, Sacks DB, National Glycohemoglobin Standardization Program
Steering C. Status of hemoglobin A1c measurement and goals for improvement: from chaos to
order for improving diabetes care. Clin Chem 2011 Feb;57(2):205-14.
76. Rosendaal FR, Helmerhorst FM, Vandenbroucke JP. Female hormones and thrombosis.
Arteriosclerosis, thrombosis, and vascular biology 2002 Feb 1;22(2):201-10.
77. Virchow R. Phlogose und Thrombose im Gefa ̈ßsystem. Gesammelte Abhandlungen zur
Wissenschaftlichen Medizin. Frankfurt: Staats- druckerei 1856.
78. Kumar DR, Hanlin E, Glurich I, Mazza JJ, Yale SH. Virchow's contribution to the
understanding of thrombosis and cellular biology. Clin Med Res 2010 Dec;8(3-4):168-72.
79. Rosendaal FR, Van Hylckama Vlieg A, Tanis BC, Helmerhorst FM. Estrogens, progestogens
and thrombosis. Journal of thrombosis and haemostasis : JTH 2003 Jul;1(7):1371-80.
80. Eisenberger A, Westhoff C. Hormone replacement therapy and venous thromboembolism.
The Journal of steroid biochemistry and molecular biology 2014 2014/07/01/;142:76-82.
81. Canonico M, Plu-Bureau G, Lowe GD, Scarabin PY. Hormone replacement therapy and risk
of venous thromboembolism in postmenopausal women: systematic review and meta-analysis.
BMJ 2008 May 31;336(7655):1227-31.
Page 129
82. Canonico M, Plu-Bureau G, Scarabin PY. Progestogens and venous thromboembolism
among postmenopausal women using hormone therapy. Maturitas 2011 Dec;70(4):354-60.
83. Mohammed K, Abu Dabrh AM, Benkhadra K, Al Nofal A, Carranza Leon BG, Prokop LJ, et al.
Oral vs Transdermal Estrogen Therapy and Vascular Events: A Systematic Review and Meta-
Analysis. The Journal of clinical endocrinology and metabolism 2015 Nov;100(11):4012-20.
84. Boardman HMP, Hartley L, Eisinga A, Main C, Roque i Figuls M, Bonfill Cosp X, et al.
Hormone therapy for preventing cardiovascular disease in post-menopausal women. Cochrane
Database of Systematic Reviews 2015 2015(3).
85. Grady D, Wenger NK, Herrington D, Khan S, Furberg C, Hunninghake D, et al.
Postmenopausal hormone therapy increases risk for venous thromboembolic disease. The
Heart and Estrogen/progestin Replacement Study. Annals of internal medicine 2000 May
2;132(9):689-96.
86. Cushman M, Kuller LH, Prentice R, Rodabough RJ, Psaty BM, Stafford RS, et al. Estrogen plus
progestin and risk of venous thrombosis. Jama 2004 Oct 6;292(13):1573-80.
87. Curb JD, Prentice RL, Bray PF, Langer RD, Van Horn L, Barnabei VM, et al. Venous
thrombosis and conjugated equine estrogen in women without a uterus. Arch Intern Med 2006
Apr 10;166(7):772-80.
88. Lobo RA, Pickar JH, Stevenson JC, Mack WJ, Hodis HN. Back to the future: Hormone
replacement therapy as part of a prevention strategy for women at the onset of menopause.
Atherosclerosis 2016 Nov;254:282-90.
89. Hoffbrand AV. Essential Haematology. Oxford: Blackwell Science; 2002. p. 241-3.
Page 130
90. Palta S, Saroa R, Palta A. Overview of the coagulation system. Indian J Anaesth 2014
Sep;58(5):515-23.
91. Wilhelmsen L, Svardsudd K, Korsan-Bengtsen K, Larsson B, Welin L, Tibblin G. Fibrinogen as
a risk factor for stroke and myocardial infarction. The New England journal of medicine 1984
Aug 23;311(8):501-5.
92. Folsom AR, Wu KK, Shahar E, Davis CE. Association of hemostatic variables with prevalent
cardiovascular disease and asymptomatic carotid artery atherosclerosis. The Atherosclerosis
Risk in Communities (ARIC) Study Investigators. Arteriosclerosis and thrombosis : a journal of
vascular biology / American Heart Association 1993 Dec;13(12):1829-36.
93. Meade TW, Ruddock V, Stirling Y, Chakrabarti R, Miller GJ. Fibrinolytic activity, clotting
factors, and long-term incidence of ischaemic heart disease in the Northwick Park Heart Study.
Lancet 1993 Oct 30;342(8879):1076-9.
94. Scarabin PY, Aillaud MF, Amouyel P, Evans A, Luc G, Ferrieres J, et al. Associations of
fibrinogen, factor VII and PAI-1 with baseline findings among 10,500 male participants in a
prospective study of myocardial infarction--the PRIME Study. Prospective Epidemiological Study
of Myocardial Infarction. Thromb Haemost 1998 Nov;80(5):749-56.
95. Ten Cate H, Meade T. The Northwick Park Heart Study: evidence from the laboratory.
Journal of thrombosis and haemostasis : JTH 2014 May;12(5):587-92.
96. Wiman B, Hamsten A. Correlations between fibrinolytic function and acute myocardial
infarction. The American journal of cardiology 1990 Nov 6;66(16):54G-6G.
97. Wiman B, Hamsten A. The fibrinolytic enzyme system and its role in the etiology of
thromboembolic disease. Semin Thromb Hemost 1990 Jul;16(3):207-16.
Page 131
98. Ridker PM, Vaughan DE, Stampfer MJ, Manson JE, Hennekens CH. Endogenous tissue-type
plasminogen activator and risk of myocardial infarction. Lancet 1993 May 8;341(8854):1165-8.
99. Spencer FA, Becker RC. Diagnosis and management of inherited and acquired
thrombophilias. J Thromb Thrombolysis 1999 Apr;7(2):91-104.
100. Danesh J, Collins R, Appleby P, Peto R. Association of fibrinogen, C-reactive protein,
albumin, or leukocyte count with coronary heart disease: meta-analyses of prospective studies.
Jama 1998 May 13;279(18):1477-82.
101. Fibrinogen Studies C, Danesh J, Lewington S, Thompson SG, Lowe GD, Collins R, et al.
Plasma fibrinogen level and the risk of major cardiovascular diseases and nonvascular mortality:
an individual participant meta-analysis. Jama 2005 Oct 12;294(14):1799-809.
102. Sandset PM. Mechanisms of hormonal therapy related thrombosis. Thromb Res 2013
Jan;131 Suppl 1:S4-7.
103. Caine YG, Bauer KA, Barzegar S, ten Cate H, Sacks FM, Walsh BW, et al. Coagulation
activation following estrogen administration to postmenopausal women. Thromb Haemost
1992 Oct 5;68(4):392-5.
104. Kroon UB, Silfverstolpe G, Tengborn L. The effects of transdermal estradiol and oral
conjugated estrogens on haemostasis variables. Thromb Haemost 1994 Apr;71(4):420-3.
105. Scarabin PY, Alhenc-Gelas M, Plu-Bureau G, Taisne P, Agher R, Aiach M. Effects of oral and
transdermal estrogen/progesterone regimens on blood coagulation and fibrinolysis in
postmenopausal women. A randomized controlled trial. Arteriosclerosis, thrombosis, and
vascular biology 1997 Nov;17(11):3071-8.
Page 132
106. Teede HJ, McGrath BP, Smolich JJ, Malan E, Kotsopoulos D, Liang YL, et al. Postmenopausal
hormone replacement therapy increases coagulation activity and fibrinolysis. Arteriosclerosis,
thrombosis, and vascular biology 2000 May;20(5):1404-9.
107. Koh KK, Ahn JY, Jin DK, Yoon BK, Kim HS, Kim DS, et al. Significant differential effects of
hormone therapy or tibolone on markers of cardiovascular disease in postmenopausal women:
a randomized, double-blind, placebo-controlled, crossover study. Arteriosclerosis, thrombosis,
and vascular biology 2003 Oct 1;23(10):1889-94.
108. Koh KK, Shin MS, Sakuma I, Ahn JY, Jin DK, Kim HS, et al. Effects of conventional or lower
doses of hormone replacement therapy in postmenopausal women. Arteriosclerosis,
thrombosis, and vascular biology 2004 Aug;24(8):1516-21.
109. Koh KK, Horne MK, 3rd, Csako G, Waclawiw MA, Cannon RO, 3rd. Relation of fibrinolytic
potentiation by estrogen to coagulation pathway activation in postmenopausal women. The
American journal of cardiology 1999 Feb 1;83(3):466-9, A10.
110. Post MS, Christella M, Thomassen LG, van der Mooren MJ, van Baal WM, Rosing J, et al.
Effect of oral and transdermal estrogen replacement therapy on hemostatic variables
associated with venous thrombosis: a randomized, placebo-controlled study in postmenopausal
women. Arteriosclerosis, thrombosis, and vascular biology 2003 Jun 1;23(6):1116-21.
111. Post MS, van der Mooren MJ, van Baal WM, Blankenstein MA, Merkus HM, Kroeks MV, et
al. Effects of low-dose oral and transdermal estrogen replacement therapy on hemostatic
factors in healthy postmenopausal women: a randomized placebo-controlled study. Am J
Obstet Gynecol 2003 Nov;189(5):1221-7.
Page 133
112. Koh KK, Han SH, Shin MS, Ahn JY, Lee Y, Shin EK. Significant differential effects of lower
doses of hormone therapy or tibolone on markers of cardiovascular disease in post-menopausal
women: a randomized, double-blind, crossover study. European heart journal 2005
Jul;26(14):1362-8.
113. Winkler UH, Altkemper R, Kwee B, Helmond FA, Coelingh Bennink HJ. Effects of tibolone
and continuous combined hormone replacement therapy on parameters in the clotting
cascade: a multicenter, double-blind, randomized study. Fertility and sterility 2000 Jul;74(1):10-
9.
114. Hellgren M, Conard J, Norris L, Kluft C. Cardiovascular risk markers during treatment with
estradiol and trimegestone or dydrogesterone. Maturitas 2009 Mar 20;62(3):287-93.
115. Borowiecka M, Polac I, Nowak P, Radwan P, Ponczek MB, Wachowicz B. Changes in
hemostatic parameters after oral hormone therapy in postmenopausal women. Journal of
women's health 2010 Dec;19(12):2267-70.
116. Lobo RA, Pickar JH, Wild RA, Walsh B, Hirvonen E. Metabolic impact of adding
medroxyprogesterone acetate to conjugated estrogen therapy in postmenopausal women. The
Menopause Study Group. Obstetrics and gynecology 1994 Dec;84(6):987-95.
117. Acs N, Vajo Z, Miklos Z, Siklosi G, Paulin F, Felicetta JV, et al. The effects of postmenopausal
hormone replacement therapy on hemostatic variables: a meta-analysis of 46 studies. Gynecol
Endocrinol 2002 Aug;16(4):335-46.
118. Koh KK, Mincemoyer R, Bui MN, Csako G, Pucino F, Guetta V, et al. Effects of hormone-
replacement therapy on fibrinolysis in postmenopausal women. The New England journal of
medicine 1997 Mar 6;336(10):683-90.
Page 134
119. Koh KK, Jin DK, Yang SH, Lee SK, Hwang HY, Kang MH, et al. Vascular effects of synthetic or
natural progestagen combined with conjugated equine estrogen in healthy postmenopausal
women. Circulation 2001 Apr 17;103(15):1961-6.
120. Eilertsen AL, Hoibraaten E, Os I, Andersen TO, Sandvik L, Sandset PM. The effects of oral
and transdermal hormone replacement therapy on C-reactive protein levels and other
inflammatory markers in women with high risk of thrombosis. Maturitas 2005 Oct 16;52(2):111-
8.
121. Ross R. Atherosclerosis--an inflammatory disease. The New England journal of medicine
1999 Jan 14;340(2):115-26.
122. Hodis HN, Mack WJ. Atherosclerosis imaging methods: assessing cardiovascular disease
and evaluating the role of estrogen in the prevention of atherosclerosis. The American journal
of cardiology 2002 Jun 20;89(12A):19E-27E; discussion E.
123. Blankenhorn DH, Hodis HN. Arterial Imaging and Atherosclerosis Reversal. Arteriosclerosis
and Thrombosis 1994 Feb;14(2):177-92.
124. Pignoli P, Longo T. Ultrasound evaluation of atherosclerosis. Methodological problems and
technological developments. Eur Surg Res 1986;18(3-4):238-53.
125. Qu B, Qu T. Causes of changes in carotid intima-media thickness: a literature review.
Cardiovascular ultrasound 2015 Dec 15;13:46.
126. Bauer M, Caviezel S, Teynor A, Erbel R, Mahabadi AA, Schmidt-Trucksass A. Carotid intima-
media thickness as a biomarker of subclinical atherosclerosis. Swiss medical weekly 2012 Oct
25;142:w13705.
Page 135
127. van den Oord SC, Sijbrands EJ, ten Kate GL, van Klaveren D, van Domburg RT, van der Steen
AF, et al. Carotid intima-media thickness for cardiovascular risk assessment: systematic review
and meta-analysis. Atherosclerosis 2013 May;228(1):1-11.
128. Den Ruijter HM, Peters SA, Anderson TJ, Britton AR, Dekker JM, Eijkemans MJ, et al.
Common carotid intima-media thickness measurements in cardiovascular risk prediction: a
meta-analysis. Jama 2012 Aug 22;308(8):796-803.
129. Blankenhorn DH, Hodis HN. George Lyman Duff Memorial Lecture. Arterial imaging and
atherosclerosis reversal. Arteriosclerosis and thrombosis : a journal of vascular biology /
American Heart Association 1994 Feb;14(2):177-92.
130. Geroulakos G, O'Gorman DJ, Kalodiki E, Sheridan DJ, Nicolaides AN. The carotid intima-
media thickness as a marker of the presence of severe symptomatic coronary artery disease.
European heart journal 1994 Jun;15(6):781-5.
131. Hodis HN, Mack WJ, LaBree L, Selzer RH, Liu CR, Liu CH, et al. The role of carotid arterial
intima-media thickness in predicting clinical coronary events. Annals of internal medicine 1998
Feb 15;128(4):262-+.
132. Mack WJ, LaBree L, Liu CR, Liu CH, Selzer RH, Hodis HN. Correlations between measures of
atherosclerosis change using carotid ultrasonography and coronary angiography.
Atherosclerosis 2000 Jun;150(2):371-9.
133. Mucowski SJ, Mack WJ, Shoupe D, Kono N, Paulson R, Hodis HN. Effect of prior
oophorectomy on changes in bone mineral density and carotid artery intima-media thickness in
postmenopausal women. Fertility and sterility 2014 Apr;101(4):1117-22.
Page 136
134. El Khoudary SR, Wildman RP, Matthews K, Thurston RC, Bromberger JT, Sutton-Tyrrell K.
Progression rates of carotid intima-media thickness and adventitial diameter during the
menopausal transition. Menopause 2013 Jan;20(1):8-14.
135. Angerer P, Stork S, Kothny W, Schmitt P, von Schacky C. Effect of oral postmenopausal
hormone replacement on progression of atherosclerosis : a randomized, controlled trial.
Arteriosclerosis, thrombosis, and vascular biology 2001 Feb;21(2):262-8.
136. Naessen T, Rodriguez-Macias K. Menopausal estrogen therapy counteracts normal aging
effects on intima thickness, media thickness and intima/media ratio in carotid and femoral
arteries. An investigation using noninvasive high-frequency ultrasound. Atherosclerosis 2006
Dec;189(2):387-92.
137. Harman SM, Black DM, Naftolin F, Brinton EA, Budoff MJ, Cedars MI, et al. Arterial Imaging
Outcomes and Cardiovascular Risk Factors in Recently Menopausal Women A Randomized Trial.
Annals of internal medicine 2014 Aug 19;161(4):249-60.
138. Miller VM, Hodis HN, Lahr BD, Bailey KR, Jayachandran M. Changes in carotid artery
intima-media thickness 3 years after cessation of menopausal hormone therapy: follow-up from
the Kronos Early Estrogen Prevention Study. Menopause 2018 Jul 16.
139. Ostberg JE, Storry C, Donald AE, Attar MJH, Halcox JPJ, Conway GS. A dose-response study
of hormone replacement in young hypogonadal women: effects on intima media thickness and
metabolism. Clinical Endocrinology 2007 Apr;66(4):557-64.
140. Selzer RH, Hodis HN, Kwongfu H, Mack WJ, Lee PL, Liu CR, et al. Evaluation of
Computerized Edge Tracking for Quantifying Intima-Media Thickness of the Common Carotid-
Artery from B-Mode Ultrasound Images. Atherosclerosis 1994 Nov;111(1):1-11.
Page 137
141. Selzer RH, Mack WJ, Lee PL, Kwong-Fu H, Hodis HN. Improved common carotid elasticity
and intima-media thickness measurements from computer analysis of sequential ultrasound
frames. Atherosclerosis 2001 Jan;154(1):185-93.
142. Hodis HN, Mack WJ, LaBree L, Mahrer PR, Sevanian A, Liu CR, et al. Alpha-tocopherol
supplementation in healthy individuals reduces low-density lipoprotein oxidation but not
atherosclerosis: the Vitamin E Atherosclerosis Prevention Study (VEAPS). Circulation 2002 Sep
17;106(12):1453-9.
143. Anderson GL, Limacher M, Assaf AR, Bassford T, Beresford SA, Black H, et al. Effects of
conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's
Health Initiative randomized controlled trial. Jama 2004 Apr 14;291(14):1701-12.
144. LaCroix AZ, Chlebowski RT, Manson JE, Aragaki AK, Johnson KC, Martin L, et al. Health
outcomes after stopping conjugated equine estrogens among postmenopausal women with
prior hysterectomy: a randomized controlled trial. Jama 2011 Apr 6;305(13):1305-14.
145. Npatchett. Lipoprotein metabolism. 2015.
146. Bruni-Fitzgerald KR. Venous thromboembolism: An overview. J Vasc Nurs 2015
Sep;33(3):95-9.
147. Perry D, Todd T. Practical-Haemostasis: A practical guide to laboratory haemostatis. 2013
[cited July 1, 2018]; Available from: http://www.practical-haemostasis.com/index.html
CHAPTER 3
DIFFERENTIAL EFFECT OF PLASMA ESTRADIOL
ON SUBCLINICAL ATHEROSCLEROSIS PROGRESSION
IN EARLY VERSUS LATE POSTMENOPAUSE
Page 139
3.1 ABSTRACT
Context: The Early versus Late Intervention Trial with Estradiol (ELITE) showed that hormone
therapy (HT) reduced progression of atherosclerosis when initiated in early but not in late
postmenopause.
Objective: This post-trial analysis determined the association between plasma estradiol (E2)
levels and atherosclerosis determined by rate of change in carotid artery intima-media thickness
(CIMT) and tested whether this association is equally evident in early (<6 years) compared with
late (≥10 years) postmenopause.
Design: Randomized controlled trial stratified by time since menopause (ClinicalTrials.gov
number NCT00114517). Mixed-effects linear models tested the association of E2 levels with CIMT
rate of change.
Setting: Los Angeles, California, USA
Participants: Healthy postmenopausal women
Intervention: Oral E2 with/without cyclic vaginal progesterone
Main outcome measures: Plasma E2 levels and CIMT assessed every 6 months over an average
4.8 years.
Results: Among 596 postmenopausal women, higher E2 was inversely associated with CIMT
progression in early-postmenopausal women (p=0.041) and positively associated with CIMT
progression in late-postmenopausal women (p=0.006) (p-for-interaction<0.001). CIMT
progression rate for the lowest versus highest quartiles of E2 levels among early-postmenopausal
women was 8.5 μm/yr and 7.2 μm/yr, while among late-postmenopausal women was 9.8 μm/yr
and 11.7 μm/yr, respectively.
Page 140
Conclusion: E2 levels are differentially associated with atherosclerosis progression according to
timing of hormone therapy initiation. With higher E2 levels, CIMT progression rate is decreased
among early-postmenopausal women, but increased among late-postmenopausal women. These
results support the timing hypothesis of HT initiation on cardiovascular benefit, with reduced
atherosclerosis progression for initiation during early postmenopause.
Page 141
3.2 INTRODUCTION
Meta-analyses of randomized controlled trials suggest that women who initiate hormone
therapy (HT) when less than 60 years of age and/or within 10 years menopause have reduced
risk of coronary heart disease (CHD) and all-cause mortality compared with placebo, whereas
women who initiate HT when older than 60 years of age and/or more than 10 years since
menopause have a null risk(1-3). The HT timing hypothesis posits that women respond to HT
differentially according to the timing of HT initiation relative to age and/or time since
menopause(4).
In a direct evaluation of the HT timing hypothesis in healthy postmenopausal women, the
Early versus Late Intervention Trial with Estradiol (ELITE) showed that when initiated within 6
years of menopause, HT significantly reduced the progression of subclinical atherosclerosis,
measured by carotid artery intima-media thickness (CIMT). In contrast, HT initiation in women
10 or more years since menopause had no effect on the progression of atherosclerosis(4).
Although the effect of plasma estradiol (E2) on the progression of subclinical atherosclerosis has
been well-described,(5) the effect of plasma E2 levels on atherosclerosis progression according
to time of HT initiation in relation to time since menopause is unknown. In this secondary analysis
of the ELITE study, we evaluated whether there is a differential association between plasma E2
levels and the progression of subclinical atherosclerosis based on when HT was initiated in
relation to time since menopause.
3.3 MATERIALS AND METHODS
ELITE study design
ELITE (ClinicalTrials.gov number NCT00114517) was a single-center, randomized, double-
Page 142
blinded, placebo-controlled trial of HT administered to women, stratified within 6 years of
menopause (early postmenopause) and 10 years or more after menopause (late
postmenopause)(6). The trial was conducted from July 2005 to February 2013. ELITE was
specifically designed to test the HT timing hypothesis, i.e., whether the effects of HT vary
according to the timing of initiation in relation to menopause. Healthy postmenopausal women
were randomized to receive either HT or placebo according to time since menopause strata using
a 1:1 ratio of stratified blocked randomization. The HT regimen for women with an intact uterus
was daily oral micronized 17-beta-estradiol 1 mg/day with 4% vaginal micronized progesterone
gel 45 mg/day for 10 days each month. Women without an intact uterus were given 1 mg/day
oral micronized 17-beta-estradiol alone. The trial was approved by the Institutional Review Board
of the University of Southern California. In the primary outcome analysis, HT was associated with
less progression of subclinical atherosclerosis (measured as rate of change in carotid artery
intima-media thickness, CIMT) compared with placebo when the therapy was initiated in early,
but not in late postmenopausal women(4). The ELITE study was funded by the National Institute
on Aging, National Institutes of Health (R01AG-024154).
Study population
Participants were recruited from the general population. Inclusion criteria were healthy
postmenopausal women without clinical evidence of cardiovascular disease with a serum E2 level
lower than 25 pg/ml and with cessation of menses for a minimum of 6 months. Exclusion criteria
included the following: women in whom time since menopause could not be determined, fasting
plasma triglyceride levels greater than 500 mg/dl, diabetes mellitus or fasting serum glucose
levels greater than 140 mg/dl, serum creatinine level greater than 2.0 mg/dl, uncontrolled
Page 143
hypertension, untreated thyroid disease, life-threatening disease with prognosis less than 5
years, a history of deep vein thrombosis, pulmonary embolism or breast cancer, or current use
of postmenopausal HT within 1 month of screening. Based on time since menopause at study
enrollment, women were characterized as early postmenopause (<6 years) and late
postmenopause (≥10 years).
Follow-up
After randomization, women attended study clinic visits every month for the first 6
months and every other month thereafter until trial completion. The median duration of follow-
up was 4.8 (range 0.5 to 6.7) years.
Carotid artery intima-media thickness assessment
The rate of change in intima-media thickness of the far wall of the right distal common
carotid artery was assessed by means of computer image processing of B-mode ultrasonograms
that were obtained at the baseline examination and at every 6 months during the follow-up
period. The serial imaging and measurement methodology were specifically developed for
longitudinal measurements of change in atherosclerosis; the coefficient of variation for baseline
CIMT measurement was 0.69%(7-9).
Estradiol assay
Fasting blood samples for measurement of plasma total E2 were obtained every 6 months
while the women were on the trial. The plasma level of total E2 was quantified by
radioimmunoassay with preceding organic solvent extraction and Celite column partition
chromatography (10) of the samples obtained at baseline and at every 6 months during the
Page 144
follow-up period(6). The assay sensitivity is 2 pg/ml and the interassay coefficients of variation
are 11%, 13% and 12% at 15, 36 and 101 pg/ml, respectively.
Statistical methods
Baseline characteristics were reported separately for each time since menopause
stratum: early and late postmenopause. Continuous variables, including age, baseline CIMT,
baseline E2 level, mean E2 level during the trial and change of E2 level from baseline are reported
as mean (standard deviation), and were compared between time since menopause strata using
two-sample t tests. Categorical variables, including race, education, randomized treatment and
hysterectomy status were reported as frequency (percent), and were compared between strata
with chi-square tests.
Per-participant rate of change in CIMT was analyzed with mixed effects linear models;
mean E2 level (per-participant average E2) was included as a main independent variable, along
with time (years) since randomization. Random effects were specified for participant-specific
intercept (baseline CIMT) and slope (rate of change in CIMT). The analysis was first performed
separately for women in early versus late postmenopause; an interaction of E2 level with time
since randomization tested whether the rate of change in CIMT (association of CIMT with time)
varied by E2 level. In a combined analysis using participants from the total cohort, time since
menopause stratum was added into the model. The difference in the association of rate of
change in CIMT with E2 by early/late postmenopause was tested by including a product term
between time since menopause stratum, E2 level, and time since randomization. This analysis
was repeated, including only women randomized to HT. Analyses were further adjusted for body
mass index, race, education and hysterectomy status. Model-based estimates of the annual CIMT
Page 145
progression rates and 95% confidence intervals were calculated for each quartile cut-point (at
25
th
, 50
th
and 75
th
percentiles) of E2 levels separately for early and late postmenopause in the
total cohort and among women randomized to HT.
3.4 RESULTS
A total of 643 eligible postmenopausal women were randomized (271 early
postmenopause and 372 late postmenopause) in the ELITE; 596 women (248 early
postmenopause and 348 late postmenopause) with follow-up data on E2 levels and CIMT were
included in this analysis. Among these women, 297 were randomly assigned to receive HT (125
early postmenopause and 172 late postmenopause). Table 3.1 shows baseline characteristics of
the study participants by time since menopause strata. As expected, the mean age (±SD) for
women in early postmenopause (54.7±4.2 years) was lower than those in late postmenopause
(63.6±6.1 years). Women in late postmenopause had thicker CIMT at baseline compared with
women in early postmenopause. Women in late postmenopause were more likely to be White,
to have lower education level, and to have a prior hysterectomy compared to those in early
postmenopause.
The mean E2 level at baseline was 8.3±5.3 pg/ml and did not differ by time since
menopause strata (Table 3.1). Mean E2 levels during the trial also did not differ significantly
between early and late postmenopause in the total cohort, combining HT- and placebo-treated
women (Table 3.2). Among women in the HT group, mean E2 levels during the trial and change
of E2 levels from baseline were significantly higher in early postmenopause than late
postmenopause group despite similar compliance (measured by percentage ideal pill count)
between the two groups. Among women in the placebo group, both mean E2 levels and change
Page 146
of E2 levels from baseline were equivalent in early and late postmenopause. The HT effect on
change in E2 levels differed between the two time since menopause strata (treatment*time since
menopause strata interaction, adjusted for baseline E2 level, ANCOVA p = 0.02).
Combining HT- and placebo-treated women, the mixed effects analysis of the CIMT
progression rate based on the mean E2 during the trial showed that a higher level of E2 was
inversely associated with the CIMT progression rate in early postmenopausal women (beta
coefficient = -0.04 (95% confidence interval (CI): -0.09, -0.001) μm CIMT per year per 1 pg/ml
estradiol, p=0.04), but was positively associated (beta coefficient = 0.063 (95% CI: 0.018, 0.107)
μm CIMT per year per 1 pg/ml estradiol, p=0.006) with CIMT progression rate in the late
postmenopausal women (Table 3.3).
The effect of E2 levels on the CIMT progression rate during the trial differed significantly
between the time since menopause strata (time since randomization*mean E2 level*time since
menopause strata interaction p<0.001, when analyzed among the total cohort and p=0.004 when
analyzed only among HT participants), adjusted for body mass index, race, education and
hysterectomy status (Table 3.4). The association of E2 levels on the CIMT progression rate was
not significantly modified by level of baseline CIMT.
In the total cohort combining HT- and placebo treated women, the model-estimated CIMT
progression rates, estimated at quartile cut-points of the on-trial E2 level; the 25
th
percentile (9
pg/ml), the 50
th
percentile (17 pg/ml) and the 75
th
percentile (38 pg/ml), are presented
graphically in Figure 3.1 1A.
Page 147
Confining the analysis to women in the HT group, the model-estimated CIMT progression
rate by quartiles of the on-trial E2 levels; the 25
th
percentile (25 pg/ml), the 50
th
percentile (37
pg/ml) and the 75
th
percentile (57 pg/ml) are shown in Figure 3.1 1B.
3.5 DISCUSSION
In this analysis from the ELITE randomized controlled trial, E2 levels were inversely
associated with CIMT progression in early postmenopausal women (who were within 6 years of
menopause) and were positively associated with CIMT progression in late postmenopausal
women (who were at least 10 years post-menopause). The primary results of ELITE showed that
oral E2 therapy reduced the progression of CIMT when compared with placebo if therapy was
initiated in early postmenopausal women, but had no effect on CIMT progression when it was
initiated in late postmenopausal women(4). The results of the current analysis showed that E2
levels achieved with oral E2 therapy were associated with CIMT progression in both early and
late postmenopause, however, in opposite directions. These opposite effects on CIMT
progression occurred despite the fact that women in early and late postmenopause had achieved
similar E2 levels from oral HT during the trial. These results not only support the HT timing
hypothesis tested by the main trial, but also add an explanatory mechanism consistent with the
timing hypothesis.
Our findings demonstrated that the effect of HT-induced E2 level on atherosclerosis
progression differs according to the timing of HT initiation relative to time since menopause. The
timing of initiation could serve as a possible chronological indicator of underlying vascular health
and hence, a determinant as to whether E2 will reduce or accelerate the progression of
atherosclerosis(4).
Page 148
Human and animal data suggest that the athero-protective effect of HT is limited to the
endothelium without complicated atherosclerosis(4, 6). Estrogen inhibits atherosclerotic lesion
formation in mice, but does not alter the progression of established atherosclerotic lesions(11).
In non-human primates, HT reduced coronary atherosclerosis when initiated immediately
following surgical menopause but had no effect after a 2 year (equivalent to 6 years in human)
delay in initiating HT after surgical menopause(12, 13). ELITE extends these animal studies by
suggesting that a healthy vasculature without complicated atherosclerosis (as inferred by
chronological age or time since menopause) is required to respond beneficially to E2 levels with
the reduction of atherosclerosis progression. Randomized controlled trials evaluating coronary
artery atherosclerosis with coronary angiography show that HT does not affect the progression
of coronary artery atherosclerosis in women with established coronary artery disease(14).
Favorable vascular effects of E2 appear to be limited to those women who are in early
postmenopause (and possibly late postmenopause) but have not yet developed atherosclerotic
vascular disease and resistance to the atheroprotective effects of estrogen(7, 15).
Although the mechanism(s) underlying the differential response of arteries to similar E2
levels is unknown, estrogens are known to regulate estrogen receptor expression levels, which
are reduced after menopause and in the presence of atherosclerotic lesions(16-18). One possible
mechanism may be related to the expression and/or signaling of estrogen receptors in healthy
versus atherosclerosis-related tissues as a function of aging and/or time since menopause.
Immunostaining of human postmortem coronary artery specimens has demonstrated that the
number of estrogen receptors is reduced in postmenopausal compared to premenopausal
women. Estrogen receptor differences are even more pronounced in premenopausal women
Page 149
with atherosclerosis compared to those without(16). In this regard, an age-related increase in
methylation of the CpG islands in the promoter region of the estrogen receptor has been shown
in vascular tissue as well as in coronary atherosclerotic plaques compared to normal vessels(17).
Proliferating aortic smooth muscle cells that are characteristically found in atherosclerotic lesions
selectively show increased estrogen receptor promoter methylation(18).
The strength of this study is the randomized design testing the HT timing hypothesis, the
prospective data, and specimen collection with simultaneous serial measurements of E2 and
CIMT levels over 5 years. Repeated measurements reduce the variability of the exposure and
outcome measurement, thus yielding more precise results. As ELITE is the only trial with a-priori
stratification of early and late postmenopause, we had sufficient statistical power to identify the
opposite directionality of the association between plasma E2 levels and atherosclerosis
progression between the two strata of time since menopause.
Certain study limitations should be noted. The analysis was limited to evaluation of total
E2 levels and did not account for the other possibly related levels of free E2, estrone and sex
hormone-binding globulin levels. This was a single site study, which may limit the generalizability
of the results. However, the uniform study conduct provided by a single site study reduced
variability in trial administration, participant follow-up, data collection and longitudinal
ultrasound imaging. This trial included predominantly healthy, well-educated women. While
perhaps limiting generalizability, education level and race were well balanced between HT and
placebo groups. Although approximately 30% of the ELITE sample included other non-White
women, the subgroup sample size of non-White women was not sufficient to evaluate possible
effect modification, in particular within specific race or ethnic groups. Stratified analysis by White
Page 150
and non-White women found similar different association of E2 level and CIMT progression rate
between early and late postmenopause.
As participants in the ELITE study were healthy women without clinical evidence of
cardiovascular disease or diabetes, future studies among populations with more heterogeneity
in underlying atherosclerosis status are needed to test the hypothesis of differential response to
estradiol level based on relative vascular health. Unlike ELITE, the Kronos Early Estrogen
Prevention Study (KEEPS) showed a null effect of HT on atherosclerosis progression in young
postmenopausal women who were within 3 years of menopause. The hormone regimen in KEEPS
(oral conjugated equine estrogens 0.45 mg/day or transdermal 17 estradiol 50 mcg/day each
with 200 mg of oral progesterone for 12 days per month) differed from ELITE and was low-dose.
The increasing dose of estrogen has been reported to reduce CIMT comparing to low dose(19).
Although both ELITE and KEEPS excluded women with history of cardiovascular disease, KEEPS
additionally excluded women who had coronary artery calcium assessed on CT imaging.
Compared with ELITE, participants in KEEPS had lower levels of cardiovascular risk factors such
as lower glucose, total cholesterol, triglyceride and LDL cholesterol and higher HDL cholesterol.
These differences may account for the null effect of HT on atherosclerosis progression in KEEPS.
Additionally, KEEPS did not include a late postmenopausal group of women and therefore, cannot
provide an analysis such as that reported herein for ELITE (20).
3.6 CONCLUSION
In conclusion, plasma E2 levels achieved through oral E2 therapy had opposite effects on
the progression of atherosclerosis among women in early and late postmenopause. With higher
plasma E2 levels, the CIMT progression is decreased among early postmenopausal women and is
Page 151
increased among late postmenopausal women. These effects were statistically demonstrated in
both the total cohort and among women receiving HT.
Page 152
3.7 TABLES AND FIGURES
Page 153
Table 3.1 Baseline characteristics of women by time since menopause strata
Variables Total
participants
Early post
menopause
(<6 years)
Late post
menopause
( ≥10 years)
Number of participants N=596 N=248 N=348
Age* (years) 59.9±6.9 54.7±4.2 63.6±6.1
Estradiol level at baseline (pg/ml) 8.3±5.3 7.9±4.8 8.5±5.7
Carotid artery intima-media thickness at
baseline* (μm)
770.3±105.5 747.1±95.5 786.9 ±109.2
Race* White
Non-Hispanic
415
(69.6)
161
(64.9)
254
(72.9)
Black
Non-Hispanic
52
(8.7)
21
(8.5)
31
(8.9)
Hispanic 79
(13.3)
36
(14.5)
43
(12.4)
Asian/Pacific Islander 50
(8.4)
30
(12.1)
20
(5.8)
Education* High school graduate
or less
22
(3.7)
6
(2.4)
16
(4.6)
Trade/business
school/some college
173
(29.0)
60
(24.2)
113
(32.5)
Bachelor’s degree/
Graduate/professional
401
(67.3)
182
(73.4)
219
(62.9)
Treatment Placebo 299
(50.2)
123
(49.6)
176
(50.6)
Active hormone 297
(49.8)
125
(50.4)
172
(49.4)
Hysterectomy* No 487
(81.7)
237
(95.6)
250
(71.8)
Yes 109
(18.3)
11
(4.4)
98
(28.2)
*Continuous variables: mean±standard deviation and p-value from t-test
Categorical variables: frequency (percent) and p-value from chi-square test or Fisher’s exact
test; p value <0.05 from the comparison between early and late postmenopause
Page 154
Table 3.2 Mean estradiol level during the trial and change of estradiol level from baseline among total
sample and participants in hormone therapy group by time since menopause strata
Variables Early post
menopause
(<6 years)
Late post
menopause
( ≥10 years)
p-value
Total cohort
N=248 N=348
Mean estradiol level during the trial
(pg/ml)
29.7±31.8 25.5±22.5 0.06
Change of estradiol level from baseline
(pg/ml)
21.7±31.6 17.0±22.7 0.03
Participants in hormone therapy group
N=125 N=172
Mean estradiol level during the trial
(pg/ml)
48.2±35.8 40.2±23.6 0.02
Change of estradiol level from baseline
(pg/ml)
40.4±35.4 31.6±24.0 0.01
Participants in placebo group
N=123 N=176
Mean estradiol level during the trial
(pg/ml)
10.9±5.9 11.0±6.0 0.85
Change of estradiol level from baseline
(pg/ml)
2.8±5.9 2.6±6.8 0.88
*Continuous variables: mean±standard deviation and p-value from t-test
Among total cohort, test for interaction of treatment*menopausal strata adjusting for baseline
estradiol level p-value=0.02
Page 155
Table 3.3 Mixed model linear regression analysis of the association of mean estradiol level
during the trial (pg/ml) with carotid artery intima-media thickness (CIMT) progression rate
(µm/year): by time since menopause strata
Variables Early post
menopause
(<6 years)
p-value Late post
menopause
( ≥10 years)
p-value
Number of participants N=248 N=348
Number of CIMT data points N=2435 N=3173
Intercept 751.0
(734.6, 767.4)
<0.001 792.0
(774.5, 809.5)
<0.001
Time since randomization (years) 7.28
(5.43, 9.13)
<0.001 7.67
(6.15, 9.19)
<0.001
Mean estradiol level (pg/ml) -0.10
(-0.47, 0.28)
0.62 -0.18
(-0.70, 0.33)
0.49
Time since randomization (years)
x Mean estradiol level (pg/ml)
-0.040
(-0.090, -0.001)
0.041 0.063
(0.018, 0.107)
0.006
*Numbers in table are beta coefficients (95% confidence interval); p-value from Wald’s test
Page 156
Table 3.4 Evaluation of the differential estradiol (pg/ml) association with carotid artery intima-
media thickness (CIMT) progression rate (µm/year) by menopause strata from mixed model
linear regression analysis: total cohort and participants in hormone therapy group
Variables Total cohort p-value Hormone
therapy group
p-value
Number of participants N=596 N=297
Number of data points N=5608 N=2784
Intercept 692.8
(624.1, 761.6)
<0.001 706.9
(596.6, 817.1)
<0.001
Time since randomization (years) 9.2
(4.9, 13.6)
<0.001 7.8
(1.4, 14.2)
0.017
Mean estradiol level (pg/ml) -0.28
(-0.77, 0.21)
0.27 -0.29
(-0.96, 0.37)
0.39
Time since randomization (years)
x Mean estradiol level (pg/ml)
0.06
(0.02, 0.11)
0.007 0.10
(0.04, 0.17)
0.002
Time since menopause strata -42.4
(-67.2, -17.6)
<0.001 -38.7
(-81.9, 4.5)
0.08
Time since randomization (years)
x Time since menopause strata
-0.39
(-2.75, 1.95)
0.74 -0.50
(-4.81, 3.80)
0.82
Mean estradiol level (pg/ml) x
Time since menopause strata
0.16
(-2.7, 2.0)
0.62 0.10
(-0.73, 0.92)
0.82
Time since randomization (years)
x Mean estradiol level (pg/ml) x
Time since menopause strata
-0.11
(-0.17, -0.05)
<0.001 -0.12
(-0.20, -0.04)
0.004
*Numbers in table are beta coefficients (95% confidence interval); p-values from Wald’s test.
The interaction term of the E2 association with CIMT rate by time since menopause strata
interaction terms represents the early menopause group (using late menopause as the referent
group).
Page 157
Figure 3.1 1A. Model-estimated CIMT progression rates at different quartiles of estradiol level
according to time since menopause strata among total cohort
1B. Model-estimated CIMT progression rate at different quartile of estradiol level
according to time since menopause strata among participants in hormone therapy group
0
2
4
6
8
10
12
14
16
9 pg/ml 17 pg/ml 38 pg/ml
CIMT rate ( μ m / y e ar)
Estradiol levels
Early postmenopause Late postmenopause
0
2
4
6
8
10
12
14
16
25 pg/ml 37 pg/ml 57 pg/ml
CIMT rate ( μ m / y e ar )
Estradiol levels
Early postmenopause Late postmenopause
Page 158
*The lines represent standard error. Number of participants (%) in each quartile of estradiol:
1
st
quartile hormone therapy group N=10 (6.5%), placebo group N=144 (93.5%); 2
nd
quartile
hormone therapy group N=21 (14.9%), placebo group N=120 (85.1%); 3
rd
quartile hormone
therapy group N=122 (78.7%), placebo group N=33 (21.39%); 4
th
quartile hormone therapy group
N=144 (98.6%); placebo group N=2 (1.4%). Estimates of CIMT rate (with 95% CI) by estradiol
level: 1A. 25
th
percentile at 9 pg/ml: early postmenopause 8.5 (4.1, 12.8) μm/year, late
postmenopause 9.8 (5.5, 14.1) μm/year (p=0.18); 50
th
percentile at 17 pg/ml: early
postmenopause 8.1 (3.8, 12.4) μm/year, late postmenopause 10.3 (6.1, 14.6) μm/year (p=0.014);
75
th
percentile at 38 pg/ml: early postmenopause 7.2 (2.9, 11.5) μm/year, late postmenopause
11.7 (7.3, 16) μm/year, (p<0.0001). 1B. 25
th
percentile at 25 pg/ml: early postmenopause 6.8 (0.7,
12.9) μm/year, late postmenopause 10.4 (4.3, 16.4) μm/year (p=0.0144), 50
th
percentile at 37
pg/ml: early postmenopause 6.6 (0.6, 12.6) μm/year, late postmenopause 11.6 (5.6, 17.6)
μm/year (p<0.0001); 75
th
percentile at 57 pg/ml: early postmenopause 6.2 (0.2, 12.2) μm/year,
late postmenopause 13.6 (7.4, 20.0) μm/year (p<0.0001).
Page 159
3.8 REFERENCES
1. Boardman HMP, Hartley L, Eisinga A, et al. Hormone therapy for preventing
cardiovascular disease in post-menopausal women. Cochrane Database of Systematic
Reviews 2015(3).
2. Salpeter SR, Walsh JM, Greyber E, et al. Brief report: Coronary heart disease events
associated with hormone therapy in younger and older women. A meta-analysis. J Gen
Intern Med 2006;21(4):363-6.
3. Salpeter SR, Walsh JME, Greyber E, et al. Mortality associated with hormone replacement
therapy in younger and older women - A meta-analysis. Journal of General Internal
Medicine 2004;19(7):791-804.
4. Hodis HN, Mack WJ, Henderson VW, et al. Vascular Effects of Early versus Late
Postmenopausal Treatment with Estradiol. New England Journal of Medicine
2016;374(13):1221-31.
5. Karim R, Hodis HN, Stanczyk FZ, et al. Relationship between serum levels of sex hormones
and progression of subclinical atherosclerosis in postmenopausal women. The Journal of
clinical endocrinology and metabolism 2008;93(1):131-8.
6. Hodis HN, Mack WJ, Shoupe D, et al. Methods and baseline cardiovascular data from the
Early versus Late Intervention Trial with Estradiol testing the menopausal hormone timing
hypothesis. Menopause-the Journal of the North American Menopause Society
2015;22(4):391-401.
Page 160
7. Hodis HN, Mack WJ, Lobo RA, et al. Estrogen in the prevention of atherosclerosis. A
randomized, double-blind, placebo-controlled trial. Annals of internal medicine
2001;135(11):939-53.
8. Selzer RH, Hodis HN, Kwong-Fu H, et al. Evaluation of computerized edge tracking for
quantifying intima-media thickness of the common carotid artery from B-mode
ultrasound images. Atherosclerosis 1994;111(1):1-11.
9. Selzer RH, Mack WJ, Lee PL, et al. Improved common carotid elasticity and intima-media
thickness measurements from computer analysis of sequential ultrasound frames.
Atherosclerosis 2001;154(1):185-93.
10. Probst-Hensch NM, Ingles SA, Diep AT, et al. Aromatase and breast cancer susceptibility.
Endocr Relat Cancer 1999;6(2):165-73.
11. Rosenfeld ME, Kauser K, Martin-McNulty B, et al. Estrogen inhibits the initiation of fatty
streaks throughout the vasculature but does not inhibit intra-plaque hemorrhage and the
progression of established lesions in apolipoprotein E deficient mice. Atherosclerosis
2002;164(2):251-9.
12. Clarkson TB. Estrogen effects on arteries vary with stage of reproductive life and extent
of subclinical atherosclerosis progression. Menopause-the Journal of the North American
Menopause Society 2007;14(3):373-84.
13. Mikkola TS, Clarkson TB, Notelovitz M. Postmenopausal hormone therapy before and
after the women's health initiative study: what consequences? Ann Med 2004;36(6):402-
13.
Page 161
14. Herrington DM, Espeland MA, Crouse JR, et al. Estrogen replacement and brachial artery
flow-mediated vasodilation in older women. Arteriosclerosis Thrombosis and Vascular
Biology 2001;21(12):1955-61.
15. Hodis HN, Mack WJ, Azen SP, et al. Hormone therapy and the progression of coronary-
artery atherosclerosis in postmenopausal women. New England Journal of Medicine
2003;349(6):535-45.
16. Losordo DW, Kearney M, Kim EZ, et al. Variable Expression of the Estrogen-Receptor in
Normal and Atherosclerotic Coronary-Arteries of Premenopausal Women. Circulation
1994;89(4):1501-10.
17. Post WS, Goldschmidt-Clermont PJ, Wilhide CC, et al. Methylation of the estrogen
receptor gene is associated with aging and atherosclerosis in the cardiovascular system.
Cardiovasc Res 1999;43(4):985-91.
18. Ying AK, Hassanain HH, Roos CM, et al. Methylation of the estrogen receptor-alpha gene
promoter is selectively increased in proliferating human aortic smooth muscle cells.
Cardiovasc Res 2000;46(1):172-9.
19. Ostberg JE, Storry C, Donald AE, et al. A dose-response study of hormone replacement in
young hypogonadal women: effects on intima media thickness and metabolism. Clinical
Endocrinology 2007;66(4):557-64.
20. Harman SM, Black DM, Naftolin F, et al. Arterial Imaging Outcomes and Cardiovascular
Risk Factors in Recently Menopausal Women A Randomized Trial. Annals of internal
medicine 2014;161(4):249-60.
CHAPTER 4
EFFECT OF ESTRADIOL DOSE AND ESTRADIOL LEVEL
ON THE METABOLIC MEASURES
IN EARLY AND LATE POSTMENOPAUSAL WOMEN
Page 163
4.1 ABSTRACT
Objective
To identify the association of estradiol (E2) dose and serum E2 levels on metabolic
measures in early (<6 years) compared to late (≥10 years) postmenopausal women from the
REPLENISH trial.
Design
Post-hoc analysis of a randomized clinical trial.
Setting
Multicenter trial in the US.
Patients
1216 early and 297 late postmenopausal women.
Intervention
Four doses of TX-001HR, an oral combined E2 and progesterone (P4) or placebo were
tested.
Main outcome measures
Mixed effects linear models tested the association of E2 dose and serum E2 levels with
the changes in metabolic parameters; total cholesterol (CHOL), high-density lipoprotein
cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), triglyceride (TRIG) and glucose
(GLUC) from 6 visits over 12 months adjusted for serum P4 level.
Results
Higher E2 dose was significantly associated with lower CHOL (p=0.02) and LDL-C (p=0.002)
and higher HDL-C (p=0.04) in early but not late postmenopause. With longer time-since-
Page 164
menopause, the inverse association of E2 dose with CHOL and LDL-C and positive association
with HDL-C were attenuated (interaction p<0.05).
Higher serum E2 levels were significantly associated with lower CHOL (p=0.004), LDL-C
(p=0.0001), and fasting blood GLUC (p=0.003) and higher TRIG (p=0.002) in early postmenopause.
Conclusion
E2 dose differentially affects metabolic measures among early compared with late
postmenopausal women. No significant main effect of serum P4 level was found. As the
metabolic parameters studied are risk factors for cardiovascular events, these results support the
timing hypothesis of E2 therapy and cardiovascular benefit.
Keywords
Estradiol, Metabolic Measure, Menopause, Lipid, Glucose
ClinicalTrail.gov registration number
NCT01942668
Page 165
4.2 INTRODUCTION
The leading cause of death among postmenopausal women is cardiovascular disease
(CVD) (1, 2). Hormone therapy (HT) has been shown to reduce CVD and mortality among
postmenopausal women without prior CVD when initiated at a younger age or closer to
menopause(3, 4). In addition to the direct effect of estrogen on vascular endothelium and
vascular smooth muscle, the protective effect of estrogen on CVD is mediated through metabolic,
inflammatory and thrombotic processes (5). Pooled data from 107 trials, with 33,315 participants
showed an HT effect on several cardiovascular risk factors; HT increased lean body mass (fat-free
body mass) and high-density lipoprotein cholesterol (HDL-C) and decreased abdominal obesity,
insulin resistance, fasting glucose (GLUC), new onset diabetes, cholesterol (CHOL), low density
lipoprotein cholesterol (LDL-C), blood pressure and procoagulant factors(6). The aforementioned
favorable effects may explain the protective effect of HT on atherosclerosis and CVD.
Estrogen alone and estrogen plus progestin effect changes in lipid and carbohydrate
metabolism. Conjugated equine estrogen (CEE) therapy significantly reduced LDL-C and
increased HDL-C compared with placebo over a 3 month follow-up, and the effect was dose-
dependent(7). The Postmenopausal Estrogen/Progestin Interventions trial tested the effect of 4
different HT regimens on CVD risk factors during 3 years. CEE alone, CEE plus sequential
medroxyprogesterone acetate (MPA), CEE plus continuous MPA and CEE plus progesterone (P4)
significantly reduced LDL-C, GLUC and insulin, and increased HDL-C and triglyceride (TRIG) levels
(8). The HT effect on metabolic measures is dose-dependent, as shown in a meta-analysis of 28
randomized clinical trials, with a significantly stronger effect of conventional dose HT (0.625 mg
Page 166
conjugated equine estrogen and equivalent doses of other formulations) compared with low-
dose HT ( 0.3 mg conjugated equine estrogen and equivalent doses of other formulations)(9).
As the beneficial E2 effect on atherosclerosis is mediated though metabolic measures and
E2 has a differential effect on atherosclerosis according to the time-since-menopause when it is
initiated(10-13), we hypothesized that E2 has a differential effect on metabolic measures
according to time-since-menopause. We tested the effects of E2 dose and E2 levels on several
metabolic measures by time-since-menopause in a post-hoc analysis of data from the REPLENISH
trial(14).
4.3 MATERIALS AND METHODS
REPLENISH study
The REPLENISH trial was a phase 3, randomized, double-blinded, placebo-controlled,
multicenter trial (80 sites in the United States) testing a new formulation of HT, TX-001HR
(TherapeuticsMD, Inc, Boca Raton, FL), a combined E2 and bioidentical P4 in a single gelatin
capsule. TX-001HR is molecularly identical to endogenous E2 and P4 and has bioavailability
similar to E2 tablets and micronized P4 capsules administered together(15). The purpose of the
REPLENISH trial was to determine the efficacy of TX-001HR on reduction of vasomotor symptoms
(VMS) compared to placebo and endometrial safety. The trial was registered with
ClinicalTrials.gov as NCT01942668 and was conducted from August 2013 to October 2016.
Eligible participants were healthy postmenopausal women (defined as ≥12 months of
spontaneous amenorrhea or ≥6 months of spontaneous amenorrhea with FSH>40 mIU/ml or ≥6
weeks postsurgical bilateral oophorectomy) aged 40-65 years with a uterus who were seeking
treatment for VMS with serum E2 ≤ 50 pg/ml, body mass index ≤ 34 kg/m
2
, using ≤ 2
Page 167
antihypertensive drugs, a negative screening mammogram, normal breast examination and
endometrial biopsy. Exclusion criteria were contraindications to HT, heavy smokers, history of
endometrial hyperplasia or undiagnosed vaginal bleeding, history of cancer or clinically
significant physical or mental illness.
The primary REPLENISH study randomly assigned 1845 eligible participants to one of five
intervention groups of daily E2/P4 (1/100 mg, 0.5/100 mg, 0.5/50 mg, 0.25/50 mg) or placebo.
Randomization at each study site used a reproducible, computer-generated block randomization
schedule. Study participants and staff were blinded to the assigned intervention throughout the
study. Participants were followed for 6 post-randomization visits (at 1, 2, 3, 6, 9 and 12 months
or early termination) when specified safety outcomes were evaluated(14, 16-18).
The analysis reported here included REPLENISH participants who were in early (<6 years-
since-menopause) or late (≥10 years-since-menopause) postmenopause with available data on
received E2 dose, serum E2 levels and metabolic measures at baseline with at least one follow-
up visit. The cut point for early and late postmenopause used in this study was derived from prior
non-human primate(19) and human studies(13) indicating differential effect of HT on CVD among
women (<6 years-since-menopause and ≥10 years-since-menopause). The participants who were
6-9 years-since-menopause were not included in the analysis to allow for two distinct groups for
comparison.
Metabolic measures, estradiol doses, estradiol levels and progesterone levels
At least 8-hour fasting blood samples were collected and measured for CHOL, LDL-C, HDL-
C, TRIG and GLUC at baseline, 3, 6, 9 and 12 months. Serum E2 levels were centrally measured at
screening and at post-randomization months 1, 3, 6, 9 and 12 with a single GC-MS/MS method
Page 168
that was validated for human serum in the range of 2 to 500 pg/ml; the interassay coefficient of
variation was <8.5% (Ventiv Health Clinical Lab Inc;Princeton, NJ). Serum P4 levels were
measured at screening and at months 1, 3 and 12 with an LC-MS/MS assay that was validated for
human serum in the range of 0.05 to 50 ng/ml; the interassay coefficient of variation was <5.5%
(Ventiv Health Clinical Lab Inc;Princeton, NJ).
Statistical methods
Baseline characteristics were reported separately for each time-since-menopause
stratum (early and late postmenopause). Continuous variables were compared between time-
since-menopause strata using two-sample t tests. The comparisons of mean continuous variables
among E2 dose (1 mg, 0.5 mg, 0.25 mg and 0 mg) were performed with analysis of variance.
Categorical variables were compared between time-since-menopause strata and across different
E2 dose with chi-square tests. Pearson correlation was used to evaluate correlation between E2
dose and mean E2 levels.
On-trial levels of metabolic measures were compared between time-since-menopause
strata and across different E2 doses using mixed effects linear models. Per-participant changes
from baseline in metabolic measures over 12 months were analyzed as repeated measures
dependent variables; E2 dose and E2 levels were included as main independent variables, along
with time-since-randomization (number of months from randomization as a time variable in the
mixed model). Random effects were specified for participant-specific intercepts (baseline values
of metabolic measures). An interaction between E2 dose or E2 levels with time-since-menopause
strata tested whether the association of E2 dose and E2 levels with the change in metabolic
measures from baseline was modified by time-since-menopause. Analyses were adjusted for
Page 169
baseline metabolic levels, age and mean on-trial serum P4 levels. The effect of serum P4 level on
changes in metabolic measures was also evaluated. Use of lipid-lowering medication was also
tested as a possible confounder and effect modifier.
Model-based estimates (and standard errors) of the changes in metabolic measures were
calculated per 0.25 mg increase in E2 dose and per 1 pg/ml increase in serum E2 levels separately
for early and late postmenopause. All analyses used SAS (version 9.4; Cary, NC); a two-sided p
value less than 0.05 was considered as statistically significant.
4.4 RESULTS
Baseline characteristics
A total of 1216 early (<6 years-since-menopause) and 297 late (≥10 years-since-
menopause) postmenopausal women were included in the analysis. Mean age (SD) was 53.2 (3.7)
years vs. 58.4 (4.1) years. Mean time-since-menopause (SD) was 2.4 (1.7) years vs. 14.1 (3.9)
years in early and late postmenopause, respectively. At baseline, late postmenopausal women
had higher systolic blood pressure (p=0.001) and lower serum E2 levels (p=0.002) than early
postmenopausal women (Table 1). Baseline demographic characteristics across randomized E2
dose groups were similar.
On-trial levels of hormones
Early postmenopausal women had significantly higher mean (SE) on-trial E2 levels (27.4
(0.8) pg/ml compared with late postmenopausal women 22.9 (1.1) pg/ml; p=0.001) despite
similar compliance (76.2% in early vs. 75.4% in late postmenopause, p=0.72). On-trial E1 (p=0.83)
and P4 levels (p=0.65) were similar between early and late postmenopausal women. Metabolic
measures were similar between early and late postmenopausal women.
Page 170
Correlation between E2 dose and E2 levels
In the total cohort, E2 dose and mean E2 levels were significantly correlated (r=0.44,
p<0.001). Late postmenopausal women (r=0.52) had a stronger correlation between E2 dose and
mean E2 levels than early postmenopausal women (r=0.43). Mean E2 levels increased with higher
E2 doses; however, the range of E2 levels substantially overlapped among E2 doses.
Association between E2 dose and E2 levels with metabolic measures
E2 dose analysis.
Higher randomized E2 dose was significantly associated with lower CHOL (p=0.02), lower
LDL-C (p=0.002) and higher HDL-C (p=0.04) among early postmenopausal women but not among
late postmenopausal women. The association with E2 dose and all metabolic measures was
significantly different between early and late postmenopausal women (p interaction <0.03). With
longer time-since-menopause, the inverse associations of E2 dose with CHOL, LDL-C and GLUC
and the positive association of E2 dose with HDL-C and TRIG were attenuated. (Table 2, Figure
1).
E2 levels analysis.
Higher serum E2 levels were significantly associated with lower CHOL (p=0.004), lower
LDL-C (p=0.0001), lower GLUC (p=0.003) and higher TRIG (p=0.002) among early postmenopausal
women. The direction and magnitude of change in all measures with increasing E2 levels for both
early and late postmenopausal women were similar (p interaction>0.05). (Table 2, Figure 2)
P4 analysis.
Page 171
No significant main effect of serum P4 level on any metabolic measure was found. Serum
P4 level did not alter the association between E2 dose and serum E2 levels with any metabolic
measure (p>0.05).
Use of lipid-lowering medication
Current use of lipid-lowering medication was tested as a possible confounder of the
associations of E2 dose, E2 levels and P4 levels with metabolic measures. Use of lipid medication
did not confound any association of interest. In addition, we stratified the analysis based on
current use versus non-use of lipid-lowering medication; the associations of E2 dose, E2 levels
with metabolic measures were not different.
4.5 DISCUSSION
This study showed that there were significant associations of E2 dose and levels with lipid
and glucose levels among early postmenopausal women, the effect of E2 dose was not apparent
in late postmenopausal women 10 years or more since menopause. The E2 dose effect on lipid
levels found in this study is consistent with the dose-dependent E2 effect on changes in lipid
levels reported from a meta-analysis comparing the effect of conventional-dose HT (0.625 mg
conjugated equine estrogen and equivalent doses of other formulations) with low-dose HT ( 0.3
mg conjugated equine estrogen and equivalent doses of other formulations) on CHD risk factors
among 3,360 women(9). Although when compared with placebo, low-dose HT significantly
decreased CHOL and LDL-C levels, the effect of low-dose HT on lipid level reduction was not as
strong as conventional-dose HT, which showed the lowest CHOL and LDL-C when compared with
low-dose HT and placebo(9).
Page 172
E2 dose was positively associated with HDL-C level but inversely associated with CHOL
and LDL-C levels among early postmenopausal women. With increasing E2 dose, the decreased
CHOL, LDL-C and GLUC levels and increased HDL-C and TRIG levels from baseline statistically
significantly differed between early and late postmenopausal women (interaction p<0.03). The
more favorable effect of E2 dose on CHOL, LDL-C, HDL-C, TRIG and GLUC among early
postmenopausal women may explain, in part, the beneficial E2 effect on atherosclerosis in early
but not late postmenopausal women as reported in ELITE(13), a randomized, double-blinded,
placebo-controlled trial that showed differential effect of E2 on atherosclerosis progression
between early and late postmenopausal women.
Examining associations of E2 levels with metabolic variables, CHOL, LDL-C and GLUC were
decreased and HDL-C and TRIG were increased with higher E2 levels; estimates of associations
were of similar magnitude among both early and late postmenopausal women. The statistically
significant change among early but not late postmenopausal women may be due to the larger
sample size of the early postmenopausal group yielding higher power of statistical results.
However, examining associations of E2 dose with metabolic variables, associations were of
smaller magnitudes in early compared to late postmenopausal women, with significant tests of
interaction. These associations with E2 dose provide support for the conclusion that the effects
of E2 dose (but not E2 level) on these metabolic variables differ by time of initiation of HT relative
to menopause.
The associations between higher E2 dose and E2 levels with favorable lipid profile and
lowered GLUC level found in this study were consistent with the meta-analysis that HT had
favorable metabolic effects when compared with placebo(6). In addition, HT has been shown to
Page 173
significantly decrease incidence of diabetes by 12-35%(20-22). A pooled analysis demonstrated
that HT decreased new-onset diabetes by 30% (hazard ratio=0.7; 95%CI, 0.6-0.9) compared with
placebo(6). Our analysis showed significant GLUC reduction with higher E2 dose and E2 levels
were among early but not late postmenopausal women.
We were not able to identify explanatory factors related to these different results
between E2 dose and E2 levels in the REPLENISH data. Time-since-menopause strata, current
smoking, race, weight, body mass index, and use of lipid-lowering medication were equally
distributed among all E2 doses. An analysis restricted to women who were treated (excluding
women who received placebo) showed similar results to the analysis among the total sample.
The different effect of E2 dose and E2 levels on metabolic factors in this study may be due to
differences in pharmacokinetics of E2 between early and late postmenopausal women. The
change in lipid levels from statin therapy differs by age in both men and women, although the
precise mechanism is unknown(23). This different association between E2 dose and E2 levels on
metabolic factors requires further exploration; future studies will evaluate determinants of
attained E2 levels in response to HT.
Serum P4 showed no significant effect on metabolic measures nor did it have a
confounding effect on E2 dose and E2 levels on any metabolic measure. A pooled analysis of 248
prospective studies showed that progestin had some opposing effect on estrogen-induced
reduction of CHOL and LDL-C; also progestin attenuated the estrogen effect on increased HDL-C
and TRIG levels among postmenopausal women(24). Although progestin may attenuate the
estrogen effect on certain lipid levels, previous studies showed that P4 had no significant effect.
In the PEPI trial, CEE plus P4 increased HDL-C similarly to CEE alone, which was significantly higher
Page 174
than CEE plus MPA. There was no significant difference of LDL-C and TRIG levels between CEE
alone, CEE plus P4 and CEE plus MPA(8).
Among early postmenopausal women, the higher E2 dose and E2 levels were associated
with a more desirable metabolic profile. With higher E2 dose, the changes in lipids and glucose
level were attenuated with longer time-since-menopause. As E2 was reported to significantly
reduce atherosclerosis progression compared with placebo among early postmenopausal
women(13, 25), findings from this study suggest that the protective E2 effect on atherosclerosis
in early postmenopause may be explained, in part, through the E2 effect on changes in lipid and
glucose levels. According to the HT timing hypothesis, women benefit more from HT when it is
initiated closer to menopause; this study supports the idea that the benefit could be due to
favorable changes in lipid and glucose levels especially LDL-C and HDL-C.
Although the findings from this study support the HT timing hypothesis, further study is
needed to explain biological mechanisms that may underlie the differential effect of E2 dose and
E2 levels on metabolic measures in early and late postmenopausal women. Metabolism of
exogenous E2 occurs mainly in the liver through cytochrome P450, which is also involved in lipid
metabolism(26). Metabolism of E2 through several types of E2 receptors in the liver is complex
and needs further investigation(27). We propose that the differential effect between early and
late postmenopausal women may be due to reduced E2 responsiveness resulting from reduced
E2 receptor number and/or function with aging(28).
The strengths of this analysis include use of data from a randomized, placebo controlled
clinical trial with balanced baseline characteristics among 4 different randomized doses of E2.
This is the first study to explore the effect of different doses of combined oral E2 and P4 regimens
Page 175
on metabolic factors according to time-since-menopause. The analysis using repeated measures
over 12-month follow-up reduces the variability of measures yielding less measurement error.
Results from this study may be generalizable to healthy White or African American
postmenopausal women using HT who are within 6 years-since-menopause or 10 years-since-
menopause and body mass index ≤ 34 kg/m
2
. Limitations included limited sample size in late
postmenopausal group which could limit power to detect statistically significance result, lacking
of other CVD related metabolic measures such as apolipoprotein particles, insulin, homeostatic
model assessment of insulin resistance score. The generalization of results may be limited to
healthy postmenopausal women as we excluded participants with high risk for CVD. Late
postmenopausal women in this study were relatively young (<60 years old) when randomized to
HT, suggesting premature and early menopause in the study sample that may not be
generalizable to the majority of women who experience natural menopause. However, with
similar age among early and late postmenopausal women, these data allow for stronger test of
the hypothesis of time-since-menopause with less confounding by age.
4.6 CONCLUSION
This study showed that there were significant associations of both E2 dose and serum E2
levels with lipid and glucose levels among early postmenopausal women. The effect of E2 dose
was not apparent in late postmenopausal women with 10 or more years since menopause. The
differential effect of E2 dose on metabolic measures among early compared with late
postmenopausal women supports the timing hypothesis of E2 therapy on cardiovascular benefit.
Page 176
4.7 TABLES AND FIGURES
Page 177
Table 4.1 Baseline characteristics by postmenopausal strata; early (<6 years-since-menopause)
and late (≥10 years-since-menopause) postmenopause.
Early
postmenopause
Late
postmenopause p
N=1216 N=297
Age
53.2 (3.7) 58.4 (4.1) <.0001
Years-since-menopause
2.4 (1.7) 14.1 (3.9) <.0001
Body Mass Index (kg/m
2
)
26.8 (4.0) 26.7 (4.2) 0.90
Weight (kg)
72.3 (12.0) 71.4 (12.6) 0.25
Systolic blood pressure (mmHg)
120.8 (11.9) 123.3 (11.3) 0.001
Diastolic blood pressure (mmHg)
76.9 (8.0) 77.6 (7.6) 0.14
Race White
806 (66.45) 185 (62.71)
0.19 African American
377 (31.08) 107 (36.27)
Other
30 (2.47) 3 (2.02)
Estradiol (pg/ml)
6.64 (7.24) 5.28 (3.34) 0.002
Estrone (pg/ml)
23.47 (12.09) 23.24 (11.84) 0.77
Progesterone (pg/ml)
59.94 (94.45) 54.13 (15.45) 0.29
Total cholesterol (mg/dl)
207.90 (35.95) 208.50 (36.27) 0.79
Low-density lipoprotein cholesterol (mg/dl)
126.00 (33.02) 127.40 (31.60) 0.51
High-density lipoprotein cholesterol (mg/dl)
63.43 (15.90) 61.88 (15.81) 0.13
Triglyceride (mg/dl)
105.90 (50.10) 110.10 (45.58) 0.19
Glucose (mg/dl)
87.70 (9.31) 87.99 (8.47) 0.63
Continuous variables reported as mean (standard deviation), p value from t-test
Categorical variable reported as frequency (percent), p value from chi-square test
Page 178
Table 4.2 Estimated change from baseline of metabolic measures per 0.25 mg increase of E2 dose and 1 pg/ml serum increase of E2
levels by postmenopausal strata
Early postmenopause Late postmenopause Interaction
Estimate (95% CI) p Estimate (95% CI) p p
Total cholesterol (mg/dl) E2 dose
-1.34 (-2.45, -0.24) 0.02 -0.26 (-2.50, 1.98) 0.82 0.03
E2 level
-0.05 (-0.08, -0.02) 0.004 -0.05 (-0.16, 0.06) 0.39 0.43
HDL cholesterol (mg/dl) E2 dose
0.39 (0.01, 0.77) 0.04 0.04 (-0.72, 0.80) 0.92 0.001
E2 level
0.001 (-0.01, 0.01) 0.84 0.02 (-0.02, 0.05) 0.34 0.74
LDL cholesterol (mg/dl) E2 dose
-1.49 (-2.41, -0.56) 0.002 -0.37 (-2.24, 1.49) 0.70 0.01
E2 level
-0.05 (-0.08, -0.03) 0.0001 -0.09 (-0.18, 0.003) 0.06 0.06
Triglyceride (mg/dl) E2 dose
0.78 (-1.11, 2.67) 0.42 3.67 (-0.16, 7.49) 0.06 0.003
E2 level 0.09 (0.04, 0.15) 0.002 0.08 (-0.11, 0.27) 0.42 0.11
Glucose (mg/dl) E2 dose
-0.48 (-0.97, 0.003) 0.05 -0.84 (-1.83, 0.15) 0.10 0.001
E2 level
-0.02 (-0.04, -0.01) 0.003 -0.02 (-0.07, 0.03) 0.34 0.88
Estimates are from mixed effect model adjusted for baseline measure and P4 level
Interaction p value tests interaction between E2 dose or E2 level with time-since-menopause (years)
CHOL=total cholesterol, LDL-C=low density lipoprotein cholesterol, HDL-C=high density lipoprotein cholesterol, TRIG=triglyceride,
GLUC=glucose
Page 179
Figure 4.1 Estimated change from baseline of metabolic measures per 0.25 mg increase in
estradiol dose by postmenopausal strata
Estimates are from mixed effect model adjusted for baseline measure and P4 level
Interaction p value tests interaction between E2 dose with time-since-menopause (years);
CHOL=total cholesterol, LDL-C=low density lipoprotein cholesterol, HDL-C=high density
lipoprotein cholesterol, TRIG=triglyceride, GLUC=glucose
* significant change from baseline
** significant difference between early and late postmenopause
-4.00
-3.00
-2.00
-1.00
0.00
1.00
2.00
3.00
4.00
CHOL HDL LDL TRIG GLUC
Change from baseline
Early postmenopause Late postmenopause
** ** **
** **
*
*
*
Page 180
Figure 4.2 Estimated change from baseline of metabolic measures per 1 pg/ml increase in
estradiol levels by postmenopausal strata
Estimates are from mixed effect model adjusted for baseline measure and P4 level
Interaction p value tests interaction between E2 levels with time-since-menopause (years);
CHOL=total cholesterol, LDL-C=low density lipoprotein cholesterol, HDL-C=high density
lipoprotein cholesterol, TRIG=triglyceride, GLUC=glucose
*significant change from baseline
-0.10
-0.08
-0.06
-0.04
-0.02
0.00
0.02
0.04
0.06
0.08
0.10
CHOL HDL LDL TRIG GLUC
Change from baseline
Early postmenopause Late postmenopause
*
*
*
*
Page 181
4.8 REFERENCES
1. Heron M. Deaths: Leading Causes for 2015. Natl Vital Stat Rep 2017;66(5):1-76.
2. Benjamin EJ, Virani SS, Callaway CW, et al. Heart Disease and Stroke Statistics-2018
Update: A Report From the American Heart Association. Circulation 2018;137(12):e67-
e492.
3. Salpeter SR, Walsh JME, Greyber E, et al. Mortality associated with hormone
replacement therapy in younger and older women - A meta-analysis. Journal of General
Internal Medicine 2004;19(7):791-804.
4. Salpeter SR, Walsh JME, Greyber E, et al. Brief report: Coronary heart disease events
associated with hormone therapy in younger and older women. Journal of General
Internal Medicine 2006;21(4):363-6.
5. Masood DE, Roach EC, Beauregard KG, et al. Impact of sex hormone metabolism on the
vascular effects of menopausal hormone therapy in cardiovascular disease. Curr Drug
Metab 2010;11(8):693-714.
6. Salpeter SR, Walsh JM, Ormiston TM, et al. Meta-analysis: effect of hormone-
replacement therapy on components of the metabolic syndrome in postmenopausal
women. Diabetes Obes Metab 2006;8(5):538-54.
7. Walsh BW, Schiff I, Rosner B, et al. Effects of postmenopausal estrogen replacement on
the concentrations and metabolism of plasma lipoproteins. The New England journal of
medicine 1991;325(17):1196-204.
Page 182
8. Miller VT, Larosa J, Barnabei V, et al. Effects of Estrogen or Estrogen/Progestin Regimens
on Heart-Disease Risk-Factors in Postmenopausal Women - the Postmenopausal
Estrogen/Progestin Interventions (Pepi) Trial. Jama-Journal of the American Medical
Association 1995;273(3):199-208.
9. Casanova G, Bossardi Ramos R, Ziegelmann P, et al. Effects of low-dose versus placebo
or conventional-dose postmenopausal hormone therapy on variables related to
cardiovascular risk: a systematic review and meta-analyses of randomized clinical trials.
The Journal of clinical endocrinology and metabolism 2015;100(3):1028-37.
10. Hodis HN, Mack WJ, Lobo RA, et al. Estrogen in the prevention of atherosclerosis - A
randomized, double-blind, placebo-controlled trial. Annals of internal medicine
2001;135(11):939-53.
11. Karim R, Mack WJ, Lobo RA, et al. Determinants of the effect of estrogen on the
progression of subclinical atherosclerosis: Estrogen in the Prevention of Atherosclerosis
Trial. Menopause 2005;12(4):366-73.
12. Karim R, Hodis HN, Stanczyk FZ, et al. Relationship between serum levels of sex
hormones and progression of subclinical atherosclerosis in postmenopausal women. The
Journal of clinical endocrinology and metabolism 2008;93(1):131-8.
13. Hodis HN, Mack WJ, Henderson VW, et al. Vascular Effects of Early versus Late
Postmenopausal Treatment with Estradiol. New England Journal of Medicine
2016;374(13):1221-31.
Page 183
14. Lobo RA, Archer DF, Kagan R, et al. A 17beta-Estradiol-Progesterone Oral Capsule for
Vasomotor Symptoms in Postmenopausal Women: A Randomized Controlled Trial.
Obstetrics and gynecology 2018;132(1):161-70.
15. Pickar JH, Bon C, Amadio JM, et al. Pharmacokinetics of the first combination 17beta-
estradiol/progesterone capsule in clinical development for menopausal hormone
therapy. Menopause 2015;22(12):1308-16.
16. Constantine G, Revicki DA, Kagan R, et al. TX-001HR is Associated with a Clinically
Meaningful Effect on Vasomotor Symptoms (28th Annual Meeting of The North
American Menopause Society October 11-14, 2017, Philadelphia, PA.). Menopause
2017;24(12):1428.
17. Lobo RA, Archer DF, Constantine G, et al. 17b-Estradiol/Progesterone in a Single Oral
Softgel Capsule (TX-001HR) Significantly Reduced Moderate-to-Severe Vasomotor
Symptoms without Endometrial Hyperplasia (28th Annual Meeting of The North
American Menopause Society October 11-14, 2017, Philadelphia, PA.). Menopause
2017;24(12):1430.
18. Goldstein SR, Constantine G, Archer DF, et al. Effects of TX-001HR on Uterine Bleeding
Rates in Menopausal Women with Vasomotor Symptoms (28th Annual Meeting of The
North American Menopause Society October 11-14, 2017, Philadelphia, PA.).
Menopause 2017;24(12):1431.
19. Mikkola TS, Clarkson TB, Notelovitz M. Postmenopausal hormone therapy before and
after the women's health initiative study: what consequences? Ann Med
2004;36(6):402-13.
Page 184
20. Kanaya AM, Herrington D, Vittinghoff E, et al. Glycemic effects of postmenopausal
hormone therapy: the Heart and Estrogen/progestin Replacement Study. A randomized,
double-blind, placebo-controlled trial. Annals of internal medicine 2003;138(1):1-9.
21. Margolis KL, Bonds DE, Rodabough RJ, et al. Effect of oestrogen plus progestin on the
incidence of diabetes in postmenopausal women: results from the Women's Health
Initiative Hormone Trial. Diabetologia 2004;47(7):1175-87.
22. Bonds DE, Lasser N, Qi L, et al. The effect of conjugated equine oestrogen on diabetes
incidence: the Women's Health Initiative randomised trial. Diabetologia 2006;49(3):459-
68.
23. Karlson BW, Palmer MK, Nicholls SJ, et al. Effects of age, gender and statin dose on lipid
levels: Results from the VOYAGER meta-analysis database. Atherosclerosis 2017;265:54-
9.
24. Godsland IF. Effects of postmenopausal hormone replacement therapy on lipid,
lipoprotein, and apolipoprotein (a) concentrations: analysis of studies published from
1974-2000. Fertility and sterility 2001;75(5):898-915.
25. Sriprasert I, Hodis HN, Karim R, et al. Differential effect of plasma estradiol on subclinical
atherosclerosis progression in early versus late postmenopause. The Journal of clinical
endocrinology and metabolism 2018.
26. Tsuchiya Y, Nakajima M, Yokoi T. Cytochrome P450-mediated metabolism of estrogens
and its regulation in human. Cancer Lett 2005;227(2):115-24.
27. Palmisano BT, Zhu L, Stafford JM. Role of Estrogens in the Regulation of Liver Lipid
Metabolism. Adv Exp Med Biol 2017;1043:227-56.
Page 185
28. Post WS, Goldschmidt-Clermont PJ, Wilhide CC, et al. Methylation of the estrogen
receptor gene is associated with aging and atherosclerosis in the cardiovascular system.
Cardiovasc Res 1999;43(4):985-91.
CHAPTER 5
EFFECT OF ESTRADIOL DOSE AND ESTRADIOL LEVEL
ON THE COAGULATION MEASURES
IN EARLY AND LATE POSTMENOPAUSAL WOMEN
Page 187
5.1 ABSTRACT
Objective
This study evaluated associations of E2 dose and serum E2 levels on coagulation/anti-
coagulation measures in early (<6 years) compared with late (≥10 years) postmenopausal
women.
Methods
Postmenopausal women from the REPLENISH trial, testing four formulations of oral
combined E2 and progesterone (P4) compared with placebo. Mixed-effects linear models tested
the association of E2 dose and serum E2 levels with prothrombin time (PT), activated partial
thromboplastin time (APTT), antithrombin (ATHRM), fibrinogen (FIBRINO), protein C (PROTC),
and protein S (PROTS) assessed 5 times over 12 months.
Results
Among 1215 early- and 297 late-postmenopausal women, E2 dose was statistically
significantly inversely associated with APTT in early-postmenopause, PROTC in late-
postmenopause, and with PT, ATHRM and PROTS in both groups. Serum E2 levels were
statistically significantly inversely associated with APTT, PROTS and FIBRINO among early-
postmenopause, PT among late-postmenopause, and ATHRM and PROTC in both groups. With
longer time-since-menopause, the inverse E2 dose effect and serum E2 effects became stronger.
Conclusion
Increasing E2 dose and serum E2 levels were associated with changes in coagulation/anti-
coagulation measures. The associations were stronger among women 10 years-since-
Page 188
menopause when initiating E2. Timing of E2 therapy, E2 dose and serum E2 levels relative to
time-since-menopause may modify venous thromboembolism risk.
Page 189
5.2 INTRODUCTION
The effect of combined estrogen/progestogen oral hormone therapy (HT) on coagulation
and anti-coagulation measures is complex, may be further complicated by time of initiation of
oral HT relative to menopause and may in part explain the effect of oral HT on thrombosis(1).
Oral HT is associated with alterations in coagulation, anti-coagulation and fibrinolysis measures
in a direction towards increased venous thrombosis risk manifested as venous thromboembolism
(VTE)(1-3).
Incidence of VTE in postmenopausal women increases with age and appears higher
among women using oral HT. In the Heart and Estrogen/Progestin Replacement Study, among
women randomized to conjugated equine estrogen (CEE) + medroxyprogesterone acetate (MPA),
women who were older than 65 years experienced higher incidence of VTE (HR 1.9, 95%CI 1.0-
3.6) than women who were 65 years or younger(4). These data suggest that the effects of oral
HT on thrombosis may be modified by age and/or time-since-menopause when oral HT is
initiated.
It is possible that oral HT has a different effect on VTE according to the time when oral HT
is initiated relative to menopause. As coagulation and anti-coagulation measures reflect
coagulation pathway activities and possibly VTE risk, we hypothesized that oral E2 has a
differential effect on coagulation and anti-coagulation measures according to time-since-
menopause when oral HT is initiated.
Four dosage strengths of TX-001HR were formulated as single, oral softgel capsules that
contained hormones that are biochemically identical to endogenous E2 and progesterone (P4)
and studied in the REPLENISH trial for treatment of moderate to severe vasomotor symptoms in
Page 190
postmenopausal women with a uterus(5). (The 1 mg/100 mg strength was approved by the FDA
in 2018 under the tradename Bijuva (TherapeuticsMD, Boca Raton, FL).) We tested the effects
of E2 dose and serum E2 levels on several coagulation and anti-coagulation measures by time-
since-menopause in a post-hoc analysis of data from the REPLENISH trial.
5.3 MATERIALS AND METHODS
REPLENISH study
The REPLENISH trial was a phase 3, randomized, double-blinded, placebo-controlled,
multicenter trial (80 sites in the United States) designed to determine the efficacy of 4 different
doses of TX-001HR on reduction of VMS and endometrial safety(5). The trial was conducted from
August 2013 to October 2016 and was registered with ClinicalTrials.gov as NCT01942668 in
September 2013. We excluded one participant from the analysis who was enrolled into the study
before the trial registration date.
Eligible participants were healthy postmenopausal women aged 40-65 years with a uterus
who were seeking treatment for VMS with serum E2 ≤ 50 pg/mL, body mass index ≤ 34 kg/m
2
,
using ≤ 2 antihypertensive drugs, a negative screening mammogram and normal breast
examination and endometrial biopsy. Exclusion criteria were contraindications to oral HT, heavy
smoking, history of endometrial hyperplasia or undiagnosed vaginal bleeding, history of cancer
or clinically significant physical or mental illness.
Of 1845 eligible women who were randomly assigned to one of five intervention groups
of daily E2 /P4 (1 mg/100 mg, 0.5 mg/100 mg, 0.5 mg/50 mg, 0.25 mg/50 mg) or placebo, 1835
women received at least one capsule of assigned intervention. Randomization at each study site
was achieved using a reproducible, computer-generated block randomization schedule. Study
Page 191
participants and staff were blinded to the assigned intervention throughout the study.
Participants were followed for 6 post-randomization visits (at 1, 2, 3, 6, 9 and 12 months or early
termination) when specified safety outcomes were evaluated.
Primary results of the REPLENISH trial showed that the two highest doses of TX-001HR
were effective in reducing both frequency and severity of VMS and had <1% incidence of
endometrial hyperplasia or cancer during the 12 months of intervention(6-9).
The analysis reported here included REPLENISH participants who were in early (<6 years-
since-menopause) or late (≥10 years-since-menopause) postmenopause with available data of
received E2 dose, serum E2 levels, coagulation and anti-coagulation measures at baseline and at
least one follow-up visit. The cut points of time-since-menopause between early and late
postmenopause derive from prior evidence of different effects of E2 on atherosclerosis in
primate(10, 11) and human(12) studies.
Measurement of metabolic parameters, coagulation and anti-coagulation measures, estradiol
doses, estradiol levels and progesterone levels
Fasting blood samples (at least 8 hours) were collected and measured for coagulation
measures; prothrombin time (PT), activated partial thromboplastin time (APTT), fibrinogen
(FIBRINO), and anti-coagulation measures; antithrombin (ATHRM), protein C (PROTC), and
protein S (PROTS) at baseline, 3, 6, 9 and 12 months.
PT, PTT and FIBRINO were measured with a turbidimetric clot detection method, ATHRM
was detected by a Chromogenic method, and PROTS was measured with a clot detection method.
All assays were completed on a STA-R Evolution Expert Series Hemostasis System (Diagnostica
Page 192
Stago Inc., USA), a fully automated, high-throughput hemostasis analyzer. PROTC was measured
with a clotting assay (CRYOCHECK
TM
Clot C
TM
, Precision Biologic Inc., Canada).
Serum E2 and P4 levels were analyzed at a central laboratory (inVentiv Health Clinical Lab
Inc; Princeton, NJ). Levels of E2 were measured at screening and at post-randomization months
1, 3, 6, 9 and 12 with a single GC-MS/MS method. The method was validated for E2 in human
serum in the range of 2 to 500 pg/mL. The interassay coefficient of variation was <8.5%. Levels
of P4 were measured at screening and at months 1, 3 and 12 with an LC-MS/MS assay that was
validated for human serum. The range of the assay for P4 was 0.05 to 50 ng/mL and the interassay
coefficient of variation was <5.5%.
Statistical methods
Baseline characteristics were reported separately for each time-since-menopause
stratum (early and late postmenopause). Continuous variables were compared between time-
since-menopause strata using two-sample t tests; comparisons among E2 doses (1 mg, 0.5 mg,
0.25 mg, and 0 mg) used ANOVA. Categorical variables were compared between time-since-
menopause strata and across different E2 dose with chi-square tests.
On-trial levels of coagulation and anti-coagulation measures were compared between
time-since-menopause strata and across different E2 doses using longitudinal modeling. Per-
participant changes from baseline in coagulation and anti-coagulation measures over 12 months
were analyzed with mixed effects linear models; E2 dose and E2 levels were included as main
independent variables, along with time-since-randomization (months) considered as a
continuous time variable in the mixed model. Random effects were specified for participant-
specific intercept (baseline values of coagulation and anti-coagulation measures). An interaction
Page 193
between E2 dose or E2 levels with time-since-menopause strata tested whether the association
of change in coagulation and anti-coagulation measures with E2 dose or E2 levels from baseline
was modified by time-since-menopause. Analyses were adjusted for baseline coagulation and
anti-coagulation levels and mean on-trial serum P4 levels. The effect of serum P4 level on changes
in coagulation and anti-coagulation measures was also evaluated.
Model-based estimates (and standard errors) of the changes in coagulation and anti-
coagulation measures were calculated per 0.25 mg increase in E2 dose and per 1 pg/mL increase
in serum E2 levels separately for early and late postmenopause. All analyses used SAS (version
9.4; Cary, NC); a two-sided p value less than 0.05 was considered statistically significant.
5.4 RESULTS
Baseline characteristics
A total of 1215 early (<6 years-since-menopause) and 297 late (≥10 years-since-
menopause) postmenopausal women who completed the trial and provided baseline and at least
one on-trial value of E2, P4, coagulation and anti-coagulation measures were included in the
analysis. Mean age (SD) was 53.2 (3.7) years vs. 58.4 (4.1) years, respectively. Mean time-since-
menopause (SD) was 2.4 (1.7) years vs. 14.1 (3.9) years in early and late postmenopause,
respectively. At baseline, late postmenopausal women had lower serum E2 levels and higher
systolic blood pressure compared to early postmenopausal women (p<0.05) (Table 1). Baseline
demographic characteristics across randomized E2 dose groups were similar (data not shown).
On-trial levels of serum hormones
After adjusting for dosage, early postmenopausal women had statistically significantly
higher mean (SE) on-trial E2 levels (27.4 (0.7) pg/mL compared with late postmenopausal women
Page 194
22.9 (1.1) pg/mL; p=0.001) (Table 2). Mean E1 and P4 levels were comparable between early and
late postmenopausal women. Mean on-trial E2 levels and E1 levels increased with higher E2 dose;
however, the range of E2 levels substantially overlapped among all E2 dose groups (data not
shown).
On-trial levels of coagulation and anti-coagulation measures
On-trial mean concentrations of PT, APTT, ANTHRM and PROTC were similar between
early and late postmenopausal women (p>0.25) (Table 2). However, PROTS was statistically
significantly lower (p=0.03) and FIBRINO was statistically significantly higher (p=0.01) among late
compared with early postmenopausal women.
Association between E2 dose and E2 levels with coagulation and anti-coagulation measures
E2 dose analysis.
Higher E2 dose was statistically significantly associated with lower PT, ATHRM and PROTS
in both early and late postmenopause. With higher E2 dose, APTT statistically significantly
decreased in early postmenopause (p=0.01) and PROTC statistically significantly decreased in late
postmenopause (p=0.005). With longer time-since-menopause, the inverse E2 dose effect on PT,
ANTHRM, PROTC, PROTS and FIBRINO became stronger (p interaction<0.001) (Table 3 and
Figure1).
E2 levels analysis.
Higher serum E2 levels were statistically significantly associated with lower ATHRM and
PROTC in both early and late postmenopause. With higher E2 levels, APTT, PROTS and FIBRINO
statistically significantly decreased in early postmenopause (p<0.005) and PT statistically
significantly decreased in late postmenopause (p=0.01). With longer time-since-menopause, the
Page 195
inverse serum E2 levels effects on PT, APTT, PROTC, PROTS and FIBRINO became stronger (p
interaction<0.03) (Table 3 and Figure2).
P4 analysis.
Serum P4 level was not associated with any coagulation or anti-coagulation measure
(P>0.17). Adjustment for serum P4 level did not alter the associations between E2 dose and
serum E2 levels with any coagulation or anti-coagulation measure.
5.5 DISCUSSION
In this post-hoc analysis from the REPLENISH trial, higher E2 dose and serum E2 levels
were associated with statistically significant reductions from baseline in several coagulation and
anti-coagulation measures including PT, APTT, ATHRM, PROTC, PROTS and FIBRINO over the 12-
month intervention and follow-up. The associations of E2 dose and level with coagulation and
anti-coagulation measures were of statistically greater magnitude in late versus early
postmenopause.
With higher E2 dose, PT, ATHRM, PROTC, PROTS and FIBRINO were statistically
significantly reduced from baseline and the changes differed between early and late
postmenopause. With higher serum E2 levels, PT, APTT, PROTC, PROTS and FIBRINO were
statistically significantly reduced from baseline and the changes differed between early and late
postmenopause. These differences indicate that changes toward coagulation risk is greater in
late compared with early postmenopausal women.
Serum P4 level did not have a statistically significant effect on coagulation or anti-
coagulation measures and did not confound the association between E2 dose and serum E2 level
with any coagulation or anti-coagulation measures.
Page 196
Findings in this study are consistent with the report from a meta-analysis of 46 clinical
trials among 31,203 postmenopausal women showing that oral HT was associated with
decreased levels of ATHRM, PROTC and PROTS(13). As oral HT has been shown to increase pro-
coagulation factors, and PT and APTT are used to measure activities of these factors in the
extrinsic, intrinsic and common coagulation pathways, oral HT may be associated with shorter PT
and APTT. Although prior studies reported no significant changes in PT or APTT with oral HT(14,
15), the data have either been analyzed solely in early postmenopausal women or with early and
late postmenopausal women combined thereby potentially masking the greater oral HT-
associated reduction in PT and APTT in late compared to early postmenopausal women as shown
in the REPLENISH data.
We demonstrated that both the E2 dose and E2 levels effects on coagulation as
determined through the extrinsic pathway of coagulation (measured by PT) were shortened in
late compared with early postmenopausal women. Although the effects of higher E2 dose and
E2 levels appeared to decrease PT in both early and late postmenopausal women, the estimates
indicated less reduction of PT in early postmenopausal women with the p-interaction supporting
a statistical difference between the two postmenopausal groups. In addition, higher E2 levels
were associated with greater reduction of APTT in late postmenopausal women with the p-
interaction supporting a statistical difference between the two postmenopausal groups.
With higher E2 dose and E2 levels, the anti-coagulation activities of PROTC and PROTS
were reduced to a statistically significantly less extent in early compared to late postmenopausal
women as supported by a significant p-interaction. As lower anti-coagulation activity implies
Page 197
greater tendency toward coagulation, early postmenopausal women had less blood coagulation
potential than late postmenopausal women as indicated by the variables in this study.
The reasons as to why E2-related alterations of coagulation and anti-coagulation
measures occur with stronger effects when oral HT is started in late postmenopause compared
to early postmenopause requires further mechanistic exploration. It is possible that the stronger
effect of E2 on coagulation and anti-coagulation measures among late postmenopausal women
was, in part, due to older age. The correlation between age and time-since-menopause was fairly
strong (Pearson correlation coefficient=0.54, p<0.001) and we did not adjust for age in the model
due to collinearity. As the correlation between changes in coagulation and anti-coagulation
measures and clinical VTE risk remains unclear, we may not expect a difference in clinical
consequences of VTE between early and late postmenopausal women. A Cochrane review of 19
clinical trials representing a total of 40,410 postmenopausal women showed similar VTE risk
when oral HT was initiated at different times-since-menopause. The VTE risk among
postmenopausal women within 10 years-since-menopause was 1.74 (95% CI, 1.11-2.73) and the
risk among those who were 10 or more years-since-menopause was somewhat higher at 1.96
(95% CI, 1.37-2.80)(16).
The strengths of this analysis include use of data from a randomized, placebo-controlled
clinical trial with balanced baseline characteristics among 4 randomized doses of E2. This is the
first study exploring the effect of different doses of a combined oral E2 and P4 regimen, so we
could also explore the P4-related effect. The analysis using repeated measures during the follow-
up reduces the variability of measures, yielding less measurement error. This analysis had some
limitations as we did not have information on sex hormone binding globulin to account for free
Page 198
and bound forms of E2. With the 12-month trial period, we could not evaluate longer-term
effects of E2 on coagulation and anti-coagulation measures; however, prior data showed that the
oral HT effect on VTE was highest within the first year of treatment(17), and we likely captured
most of the oral HT-related changes in these measures. Results from this study may be
generalizable to healthy postmenopausal women using oral HT who are within 6 years-since-
menopause or 10 years-since-menopause.
5.6 CONCLUSION
The results from this study show that E2 dose and serum E2 levels achieved with oral
combined E2 and P4 therapy are statistically significantly and inversely associated with anti-
coagulation measures among early and late postmenopausal women with stronger effects in
women further from menopause. The differential effect of oral HT when initiated in early versus
late postmenopausal women may explain the different thrombotic risk in older postmenopausal
women compared with younger postmenopausal women.
Page 199
5.7 TABLES AND FIGURES
Page 200
Table 5.1 Baseline characteristics by postmenopausal strata; early (<6 years-since-menopause)
and late (≥10 years-since-menopause) postmenopause.
Early
postmenopause
Late
postmenopause p
N=1215 N=297
Age (years)
53.2 (3.7) 58.4 (4.1) <.0001
Year-since-menopause (years)
2.4 (1.7) 14.1 (3.9) <.0001
Body Mass Index (kg/m
2
)
26.8 (4.0) 26.7 (4.2) 0.90
Weight (kg)
72.3 (12.0) 71.4 (12.6) 0.25
Systolic blood pressure (mmHg)
120.8 (11.9) 123.3 (11.3) 0.001
Diastolic blood pressure (mmHg)
76.9 (8.0) 77.6 (7.6) 0.14
Race White
806 (66.45) 185 (62.71)
0.19 African American
377 (31.08) 107 (36.27)
Other
30 (2.47) 3 (2.02)
Estradiol (pg/mL)
6.6 (7.2) 5.3 (3.3) 0.002
Estrone (pg/mL)
23.5 (12.1) 23.2 (11.8) 0.77
Progesterone (pg/mL)
59.9 (94.4) 54.1 (15.5) 0.29
Prothrombin Time (seconds)
11.3 (1.0) 11.2 (0.9) 0.07
Activated Partial Thromboplastin Time
(seconds)
36.6 (5.9) 35.8 (6.2) 0.06
Antithrombin (% activity)
117.0 (14.4) 116.8 (14.6) 0.87
Protein C (%)
134.2 (32.0) 138.4 (32.7) 0.05
Protein S (%)
101.4 (22.0) 101.6 (20.9) 0.86
Fibrinogen (Umol/L)
9.4 (2.1) 9.4 (2.1) 0.70
Continuous variables reported as mean (standard deviation), p value from t-test
Categorical variable reported as frequency (percent), p value from chi-square test
Page 201
Table 5.2 On-trial levels of hormone, coagulation and anti-coagulation measures by
postmenopausal strata; early (<6 years-since-menopause) and late (≥10 years-since-menopause)
postmenopause.
Early
postmenopause
Late
postmenopause
p
Estradiol (pg/mL)
27.4 (0.7) 22.9 (1.1) 0.001
Estrone (pg/mL)
127.5 (3.2) 125.8 (6.3) 0.83
Progesterone (pg/mL)
472.1 (149.6) 328.9 (45.7) 0.65
Prothrombin Time (seconds)
11.23 (0.02) 11.29 (0.04) 0.19
Activated Partial Thromboplastin Time (seconds) 37.70 (0.15) 37.56 (0.31) 0.70
Antithrombin (% activity)
111.64 (0.30) 111.81 (0.61) 0.80
Protein C (%)
122.42 (0.65) 120.79 (1.26) 0.25
Protein S (%)
98.11 (0.39) 96.34 (0.70) 0.03
Fibrinogen (Umol/L)
9.26 (0.04) 9.54 (0.10) 0.01
Continuous variables reported as least square mean (standard error), p value from mixed effect
linear model
Page 202
Table 5.3 Estimated change from baseline of coagulation and anti-coagulation measures per 0.25 mg increase of E2 dose and 1 pg/mL
serum increase of E2 levels by postmenopausal strata
Early postmenopause Late postmenopause Interaction
p
Estimate (95% CI) p Estimate (95% CI) p
Prothrombin Time (second)
E2 dose -0.04 (-0.08, -0.01) 0.02 -0.12 (-0.19, -0.04) 0.002 0.001
E2 level
-0.001 (-0.002, 0.001) 0.09 -0.01 (-0.01, -0.001) 0.01 0.0004
Activated Partial
Thromboplastin Time (second)
E2 dose
-0.40 (-0.69, -0.11) 0.01 -0.13 (-0.71, 0.45) 0.66 0.06
E2 level
-0.01 (-0.02, -0.001) 0.02 -0.02 (-0.05, 0.01) 0.12 0.01
Antithrombin (% activity) E2 dose
-1.22 (-1.76, -0.67) <.0001 -2.11 (-3.21, -1.01) 0.0002 <.0001
E2 level
-0.06 (-0.08, -0.04) <.0001 -0.06 (-0.12, -0.01) 0.02 0.27
Protein C (%) E2 dose
-0.23 (-1.43, 0.97) 0.70 -3.58 (-6.06, -1.10) 0.005 <.0001
E2 level
-0.04 (-0.08, -0.001) 0.04 -0.20 (-0.32, -0.07) 0.002 <.0001
Protein S (%) E2 dose
-0.92 (-1.61, -0.24) 0.01 -2.55 (-3.93, -1.17) 0.0003 <.0001
E2 level
-0.03 (-0.05, -0.01) 0.001 -0.05 (-0.12, 0.01) 0.11 0.02
Fibrinogen (Umol/L) E2 dose
-0.04 (-0.13, 0.04) 0.34 -0.04 (-0.21, 0.14) 0.69 <.0001
E2 level
-0.003 (-0.01, -0.0001) 0.04 -0.005 (-0.01, 0.004) 0.29 0.03
Estimates are from mixed effect linear model adjusted for baseline measure and P4 level
Interaction p value tests interaction between E2 dose or E2 levels with time-since-menopause (years)
Page 203
Figure 5.1 Estimated change from baseline of coagulation and anti-coagulation measures per
0.25 mg increase in estradiol dose by postmenopausal strata
Estimates are from mixed effect linear model adjusted for baseline measure and P4 level;
Interaction p value tests interaction between E2 dose with time-since-menopause (years);
prothrombin time (PT), activated partial thromboplastin time (APTT) antithrombin (ATHRM),
fibrinogen (FIBRINO), protein C (PROTC), and protein S (PROTS)
*significant change from baseline
** significant difference between early and late postmenopause
-4.00
-3.50
-3.00
-2.50
-2.00
-1.50
-1.00
-0.50
0.00
PT APTT ANTHRM PROTC PROTS FIBRINO
Change from baseline
Early postmenopause Late postmenopause
*
*
*
*
*
*
** ** **
** **
*
*
Page 204
Figure 5.2 Estimated change from baseline of coagulation and anti-coagulation measures per 1
pg/mL increase in estradiol level by postmenopausal strata
Estimates are from mixed effect linear model adjusted for baseline measure and P4 level;
Interaction p value tests interaction between E2 dose with time-since-menopause (years);
prothrombin time (PT), activated partial thromboplastin time (APTT) antithrombin (ATHRM),
fibrinogen (FIBRINO), protein C (PROTC), and protein S (PROTS)
*significant change from baseline
** significant difference between early and late postmenopause
-0.250
-0.200
-0.150
-0.100
-0.050
0.000
PT APTT ATHRM PROTC PROTS FIBRINO
Change from baseline
Early postmenopause Late postmenopause
*
*
*
*
*
*
** ** **
** **
*
*
Page 205
5.8 REFERENCES
1. Canonico M, Scarabin PY. Hormone therapy and risk of venous thromboembolism
among postmenopausal women. Climacteric : the journal of the International
Menopause Society 2009;12 Suppl 1:76-80.
2. Spencer FA, Becker RC. Diagnosis and management of inherited and acquired
thrombophilias. J Thromb Thrombolysis 1999;7(2):91-104.
3. Sandset PM. Mechanisms of hormonal therapy related thrombosis. Thromb Res
2013;131 Suppl 1:S4-7.
4. Grady D, Wenger NK, Herrington D, et al. Postmenopausal hormone therapy increases
risk for venous thromboembolic disease. The Heart and Estrogen/progestin
Replacement Study. Annals of internal medicine 2000;132(9):689-96.
5. Mirkin S, Amadio JM, Bernick BA, et al. 17beta-Estradiol and natural progesterone for
menopausal hormone therapy: REPLENISH phase 3 study design of a combination
capsule and evidence review. Maturitas 2015;81(1):28-35.
6. Constantine G, Revicki DA, Kagan R, et al. TX-001HR is Associated with a Clinically
Meaningful Effect on Vasomotor Symptoms (28th Annual Meeting of The North
American Menopause Society October 11-14, 2017, Philadelphia, PA.). Menopause
2017;24(12):1428.
7. Lobo RA, Archer DF, Constantine G, et al. 17b-Estradiol/Progesterone in a Single Oral
Softgel Capsule (TX-001HR) Significantly Reduced Moderate-to-Severe Vasomotor
Symptoms without Endometrial Hyperplasia (28th Annual Meeting of The North
Page 206
American Menopause Society October 11-14, 2017, Philadelphia, PA.). Menopause
2017;24(12):1430.
8. Lobo RA, Archer DF, Kagan R, et al. A 17beta-Estradiol-Progesterone Oral Capsule for
Vasomotor Symptoms in Postmenopausal Women: A Randomized Controlled Trial.
Obstetrics and gynecology 2018;132(1):161-70.
9. Goldstein SR, Constantine G, Archer DF, et al. Effects of TX-001HR on Uterine Bleeding
Rates in Menopausal Women with Vasomotor Symptoms (28th Annual Meeting of The
North American Menopause Society October 11-14, 2017, Philadelphia, PA.).
Menopause 2017;24(12):1431.
10. Williams JK, Anthony MS, Honore EK, et al. Regression of Atherosclerosis in Female
Monkeys. Arteriosclerosis Thrombosis and Vascular Biology 1995;15(7):827-36.
11. Mikkola TS, Clarkson TB, Notelovitz M. Postmenopausal hormone therapy before and
after the women's health initiative study: what consequences? Ann Med
2004;36(6):402-13.
12. Hodis HN, Mack WJ, Henderson VW, et al. Vascular Effects of Early versus Late
Postmenopausal Treatment with Estradiol. New England Journal of Medicine
2016;374(13):1221-31.
13. Acs N, Vajo Z, Miklos Z, et al. The effects of postmenopausal hormone replacement
therapy on hemostatic variables: a meta-analysis of 46 studies. Gynecol Endocrinol
2002;16(4):335-46.
Page 207
14. Lobo RA, Pickar JH, Wild RA, et al. Metabolic impact of adding medroxyprogesterone
acetate to conjugated estrogen therapy in postmenopausal women. The Menopause
Study Group. Obstetrics and gynecology 1994;84(6):987-95.
15. Winkler UH, Altkemper R, Kwee B, et al. Effects of tibolone and continuous combined
hormone replacement therapy on parameters in the clotting cascade: a multicenter,
double-blind, randomized study. Fertility and sterility 2000;74(1):10-9.
16. Boardman HMP, Hartley L, Eisinga A, et al. Hormone therapy for preventing
cardiovascular disease in post-menopausal women. Cochrane Database of Systematic
Reviews 2015(3).
17. Eisenberger A, Westhoff C. Hormone replacement therapy and venous
thromboembolism. The Journal of steroid biochemistry and molecular biology
2014;142:76-82.
Page 208
CHAPTER 6
FACTORS ASSOCIATED WITH SERUM ESTRADIOL LEVELS
AMONG POSTMENOPAUSAL WOMEN USING HORMONE THERAPY
Page 209
6.1 ABSTRACT
Objective
To identify factors associated with serum estradiol (E2) levels among healthy
postmenopausal women using hormone therapy (HT).
Methods
This is an unplanned post hoc analysis of data from the Early versus Late Intervention
Trial with Estradiol, a randomized controlled trial of 1 mg oral E2 with or without vaginal
progesterone in healthy postmenopausal women. We included results from visits when women
reported at least 80% compliance to HT. Mixed effects linear models identified factors
associated with serum E2 levels while taking HT assessed every 6 months over a median follow-
up of 4.8 years adjusted for baseline E2 level, visit and reduced E2 dose. Possible correlates
evaluated included demographics, clinical characteristics, medication use and biomarkers of
liver and kidney metabolic function.
Results
The analysis included 2160 E2 measurements in 275 postmenopausal women. Mean SD
age was 55.4 3.9 vs 64.4 5.5 years and mean SD time-since-menopause was 3.6 1.8 vs
16.0 5.6 years for early vs late postmenopausal women. Adjusted for pre-treatment E2 level,
visit and reduced dose indicator, higher serum E2 levels were associated with higher body mass
index (BMI), higher weight, surgical menopause, alcohol use, and antihypertensive medication
use. Current and past smoking and antifungal medication use were associated with lower
serum E2 levels. In the multivariable model, higher BMI and alcohol use were associated with
Page 210
higher serum E2 levels while current and past smoking were associated with lower serum E2
levels. These factors were similar between early and late postmenopausal women.
Conclusion
Factors associated with serum E2 levels among postmenopausal women taking HT
include BMI, alcohol use and smoking. As serum E2 levels relate to HT effect, achievement of
desirable E2 levels can be maximized through personalized intervention.
Clinical Trial Registration
ClinicalTrials.gov, NCT00114517.
Page 211
6.2 INTRODUCTION
The North American Menopause Society (NAMS) in 2017 recommended that hormone
therapy (HT) should be individualized to maximize benefits and minimize risks(1). Although
there is no recommendation to monitor serum estradiol (E2) levels among postmenopausal
women taking HT, achieved E2 levels have been shown to be associated with potential benefits
on atherosclerosis as reported in Estrogen in the Prevention of Atherosclerosis Trial (EPAT) and
the Early vs Late Intervention Trial with Estradiol (ELITE)(2, 3). This association may explain the
reduced coronary heart disease and atherosclerosis progression only among early
postmenopausal women(4, 5). The estrogen threshold hypothesis postulates that each end
organ tissue varies in its sensitivity to E2(6). Some studies show that achieved E2 levels relate to
biological effects of HT treatment. For instance, in the Kronos Early Estrogen Prevention Study
(KEEPS), E2 level was related to intensity of hot flushes(7) and in the Ultra-Low-dose
Transdermal estRogen Assessment (ULTRA) trial, E2 level was related to bone mineral
density(8)
The achieved mean serum E2 levels among oral E2 compliant postmenopausal women
in ELITE varied widely from 9 to 360 pg/ml. Oral E2 is metabolized in the liver through the
cytochrome P450 (CYP) enzyme pathway(9, 10) and is excreted through urine and feces(11).
This study aims to explore the factors associated with E2 levels to reflect this wide range among
postmenopausal women taking HT. We hypothesized that demographics, clinical characteristics
and factors relating to liver and kidney metabolism may be potential factors associated with E2
level in postmenopausal women taking oral HT.
Page 212
6.3 MATERIALS AND METHODS
This was an unplanned post hoc analysis conducted among healthy postmenopausal
women participating in ELITE(2, 3). ELITE was a single-center, randomized, double-blinded,
placebo-controlled trial of HT in postmenopausal women, stratified by <6 years-since-
menopause (early postmenopause) and 10 or more years-since-menopause (late
postmenopause). ELITE was specifically designed to test the HT timing hypothesis, i.e., whether
the effects of HT on atherosclerosis progression vary according to the timing of HT initiation in
relation to menopause. Eligible women were healthy postmenopausal women with no clinical
history of cardiovascular disease or diabetes. A total of 643 postmenopausal women were
randomized to receive either HT or placebo according to time-since-menopause strata using a
1:1 ratio of stratified blocked randomization. The HT regimen included oral micronized 17-beta-
estradiol 1 mg/day with (in women with a uterus) or without 4% vaginal micronized
progesterone gel 45 mg/day for 10 days each month. After randomization, women attended
study clinic visits every month for the first 6 months and every other month thereafter until
trial completion. The trial was conducted from July 2005 to February 2013 with a median
duration of follow-up of 4.8 (range 0.5 to 6.7) years. The primary trial results showed that HT
was associated with less progression of subclinical atherosclerosis (measured as rate of change
in carotid artery intima-media thickness, CIMT) compared with placebo when therapy was
initiated in early, but not in late postmenopausal women(5). The ELITE trial was registered on
ClinicalTrials.gov (NCT00114517), funded by the National Institute on Aging, National Institutes
of Health (R01AG-024154) and was approved by the Institutional Review Board of the
University of Southern California.
Page 213
This analysis included ELITE visits with at least 80% compliance with active HT
determined by tablet count. Compliance was calculated as the percentage of active HT tablets
that should have been consumed over the inter-visit period, comparing number of pills
dispensed at the prior visit to those returned at the subsequent visit.
At baseline and at every 6 months during trial follow-up, serum levels of total E2 were
quantified by radioimmunoassay with preceding organic solvent extraction and Celite column
partition chromatography of samples(12). Assay sensitivity was 2 pg/ml and interassay
coefficients of variation were 11%, 13% and 12% at 15, 36 and 101 pg/ml, respectively.
Race–ethnicity, type of menopause (natural, surgical), multivitamin use, vitamin E use
and fish oil use were determined at baseline by self-report using structured questionnaires. At
baseline and each 6-month visit, we measured age, time-since-menopause, weight and body
mass index (BMI). Smoking status (never, past smoker, current smoker) and alcohol use (<1
drink, 1-2 drinks, >2 drinks per day) were self-reported at each visit. Total weekly metabolic
equivalent of energy expenditure (MET) calculated as weekly hours of moderate and vigorous
activity were determined from a structured 7-day physical activity recall(5). Current use of any
lipid-lowering medication, lipid lowering with statins in particular, antihypertensive, calcium
channel blocker, antifungal, diabetes and anticonvulsant medications were determined from
medications brought into each clinic visit. Creatinine, creatinine clearance, aspartate
aminotransferase (AST), alanine aminotransferase (ALT) and fasting blood glucose were
measured at baseline and at annual study visits by chemistry safety laboratory measures. A
reduced E2 dose indicator (yes, no) defined visits when reduced E2 dose (0.5 mg, 0.25 mg
instead of 1 mg oral E2) was taken.
Page 214
Baseline demographic and clinical characteristics were reported as means and standard
deviations (SD) for continuous variables and frequencies and percentages for categorical
variables. Baseline serum E2 levels and serum E2 levels while taking HT were reported as
mean SD and median (IQR). The levels were compared between early and late postmenopausal
women with Wilcoxon rank sum test.
Serum E2 levels were log transformed to achieve normality. As creatinine, creatinine
clearance, AST, and ALT were measured annually, these measures were imputed at the 6-, 18-,
30-, 42-, and 54-month visits using the method of multiply-imputed chained equations with 10
imputations(12).
Associations between log-transformed E2 levels while taking HT and each potential
correlate were first assessed using mixed effects linear models with a random intercept at the
participant level, and adjusted for baseline serum E2 level, follow-up visit (as indicator
variables) and reduced E2 dose indicator.
Potential E2 correlates with a p-value 0.15 were included in a multivariable model. A
manual backward selection approach was used to drop least significant variables from the
model; significant variables with p-value less than 0.05 were retained in the final model. The
final multivariable model was developed and presented among the total analysis sample and
stratified by early and late postmenopause. In the total analysis sample, the interaction term of
each significant factor by time-since-menopause strata was tested to determine whether the E2
association was modified by time-since-menopause.
The estimates from the model were back-transformed to show the association with E2
levels. The estimates of association with log E2 levels and model-estimated least square mean
Page 215
E2 levels for each variable in the multivariable model are presented in table 6.3 and table 6.5.
All statistical analyses were performed using SAS software version 9.4 (Cary, NC).
6.4 RESULTS
Of the 643 women in the ELITE trial, 323 women were randomized to HT. Among those
women, 275 women (123 early and 152 late postmenopause) had visits with at least 80%
compliance to HT, contributing 2160 E2 measurements over the trial follow-up for the analysis.
The mean SD age was 55.4 3.9 vs 64.4 5.5 years and mean time-since-menopause was
3.6 1.8 vs 16.0 5.6 years for early vs late postmenopausal women, respectively. The majority
of the women were non-Hispanic white (202/275, 73.5%) and had experienced natural
menopause (245/275, 89.1%). The mean SD BMI was 27.2 5.6 kg/m
2
, and mean creatinine,
creatinine clearance, AST, ALT and glucose levels were within normal ranges. More than half of
the women had never smoked (170/275, 61.8%) and were not currently consuming alcohol
(139/275, 50.5%) (Table 6.1).
Among the total analysis sample, the mean SD baseline serum E2 level was 10.8 3.3
pg/ml and the median (IQR) was 9.0 (3.0) pg/ml. Early and late postmenopausal women had
similar mean SD (10.6 2.9 vs 10.9 3.5 pg/ml) and median (IQR) (9.0 (2.0) vs 9.0 (3.0) pg/ml)
baseline serum E2 (Wilcoxon rank sum p=0.57). (Data not presented in a table)
The average serum E2 level while taking HT in the total analysis sample was 55.6 47.5
pg/ml and the median (IQR) was 45 (40) pg/ml. Early and late postmenopausal women had
similar mean (57.8 56.7 vs 53.6 36.9 pg/ml) and median (46 (40) vs 45 (39) pg/ml) serum E2
level while taking HT (Wilcoxon rank sum p=0.39). (Data not presented in a table)
Page 216
Adjusted for baseline serum E2 level, visit and a reduced E2 dose indicator variable,
higher serum E2 levels were significantly associated with higher BMI (p=0.002), surgical
menopause (p=0.04), consumption of > 2 alcoholic beverages per day (p<0.001), and
antihypertensive medication use (p=0.02). Lower serum E2 levels were associated with smoking
(p<0.001) and antifungal medication use (p=0.02) (Table 6.2 with back-transformed estimates
with E2 levels and Table 6.3 with estimates from model with log E2 levels). Alcohol use of >2
drinks per day was significantly associated with higher serum E2 level (p<0.001).
Compared to non-smokers, past smokers had significantly lower serum E2 levels
(p=0.005) and current smokers had the lowest serum E2 levels (p<0.001. Higher intake of
alcohol was significantly associated with higher serum E2 levels (p-value for trend=0.003).
Age, race–ethnicity, time-since-menopause, creatinine, creatinine clearance, AST, ALT,
fasting blood glucose, total weekly MET hours, hours of weekly moderate and vigorous activity,
use of lipid-lowering medication, calcium channel blocker, diabetic medication, anticonvulsant
medication, multivitamins, vitamin E and fish oil supplements were not significantly associated
with serum E2 levels.
In a multivariable mixed effects model in the total analysis sample (275 women, 2160
visits), higher serum E2 levels were significantly associated with higher BMI and consumption of
>2 alcohol beverages per day (p<0.001). Lower serum E2 levels were significantly associated
with current and past smoking (p<0.001) (Table 6.4 with back-transformed estimates with E2
levels and Table 6.5 with estimates from model with log E2 levels). Alcohol use of >2 drinks per
day was significantly associated with higher serum E2 levels (p<0.001). The beta coefficients
showed a trend in increasing serum E2 level with the amount of alcohol use (p for trend=0.002).
Page 217
The factors associated with serum E2 levels were similar between early (123 women,
1037 visits) and late (152 women, 1123 visits) postmenopausal women based on similar beta
estimates and non-significant interaction by time-since-menopause. While both current and
past smoking was associated with serum E2 level among early postmenopausal women, only
current smoking was associated with serum E2 level among late postmenopausal women. The
association of alcohol use on serum E2 level was statistically significant among early but not
late postmenopausal women (alcohol use by time-since-menopause interaction = 0.06).
Model-estimated mean serum E2 levels for each factor are presented in Table 6.6.
6.5 DISCUSSION
Our study identified BMI, alcohol use, and smoking as statistically significant correlates
of serum E2 levels among postmenopausal women taking oral E2 therapy. When considering
each alcohol use category, we found that alcohol use of > 2 drinks per day was significantly
associated with higher serum E2 levels. The beta coefficients showed a significant trend in
increasing serum E2 levels with the increasing alcohol use. The factors associated with serum E2
level were similar between early and late postmenopausal women. Though the association of
alcohol use and E2 levels appeared to be stronger in early compared to late postmenopausal
women, the interaction by time-since-menopause was not significant (p=0.06). The biological
explanation for BMI, alcohol use and smoking on E2 levels involves the complex pathway of E2
metabolism(10) as discussed below.
The positive association of E2 levels with BMI is consistent with several prior reports(13,
14). The EPAT showed that overweight and obese postmenopausal women using E2 therapy
attained significantly greater concentrations of both total E2 and free E2 adjusted for age
Page 218
(p=0.01)(15). In postmenopausal women, the aromatization from androstenedione to E1 and E2
occurs mainly in adipose tissue(16); hence elevated BMI which indicates increased fat mass
could explain the increased E2 levels(17).
The association between alcohol use and serum E2 levels found in this study confirmed
the findings from prior smaller studies. Acute alcohol consumption increased E2 levels among
12 postmenopausal women taking HT to a level of 300% higher than the targeted level in a
randomized, double-blind, placebo-controlled crossover study(18). A meta-analysis of 13
prospective cohort studies reported that higher levels of alcohol consumption were significantly
associated with increased E2 levels compared to non-drinkers (p trend=0.002)(19). A cross-
sectional analysis in the Women’s Health Initiative observational study reported a positive
association between alcohol consumption and E2, E2 metabolites and E1 among HT users. The
association was stronger with increasing dose of alcohol (p trend=0.02)(20). An alcohol effect
on E2 metabolism was hypothesized to occur through increased aromatase activity in the
conversion of androgens to E2 as alcohol drinkers had lower testosterone and higher E2 to
testosterone ratio compared to non-drinkers(21). Alcohol was also reported to promote
adrenal gland cell signaling of dehydroepiandrosterone sulfate production, a precursor of E2
production(22, 23). Alcohol consumption also increases the reduced nicotinamide-adenine
dinucleotide (NADH) to NAD
+
ratio in the liver(24) and leads to decreased catabolism of steroid
hormones through oxidation and inhibition of E2 conversion to E1(18).
A pharmacokinetic study of oral E2 among postmenopausal women reported that
smoking enhances the hepatic metabolism of oral E2, resulting in lower E2 levels(25). In a
randomized cross-over study of oral and transdermal E2 therapy, E2 levels were 40-70% lower
Page 219
among smokers compared with non-smokers; a statistically significant difference in E2 level was
seen among postmenopausal women taking oral, but not transdermal E2 therapy(26). Smoking
also reduces or completely cancels the efficacy of oral E2 therapy among postmenopausal
women on E2-related effects such as alleviation of hot flushes and urogenital symptoms,
beneficial effects on lipid metabolism, osteoporosis and cardiovascular disease(27, 28). As the
effect of smoking on E2 levels has been demonstrated only with oral E2, the mechanism could
involve the elevated hepatic clearance of E2 as smoking increases 2-hydroxylation of E2 and
leads to decreased bioavailability of E2(29). Additional mechanisms of a smoking effect on E2
include a smoking-related reduction of aromatase activity in granulosa cell and fatty tissue, a
reduction in steroid production from cholesterol through smoking-related inhibition of C-20, 22
desmolase, an increase in A-ring metabolism of the CYP450 enzyme system, increased sex-
hormone binding globulin (SHBG) capacity, stimulation of adrenal function and increased
hepatic and renal clearance(28).
We further explored the association between past smokers and E2 levels by conducting
sensitivity analysis removing women who had recently stopped smoking and women with
secondhand smoking exposure. The magnitude and statistical significance of the association of
past smoking with E2 level did not change. One possible explanation could be that the induction
of liver enzymatic activity by smoking may be permanent or may take many decades to return
to baseline levels, similar to the decreasing risk over decades of lung cancer seen in individuals
who stop smoking. However, this is conjecture and will require further study to untangle this
interesting but complex finding.
Page 220
We found that antifungal medication use was not statistically associated with E2 level in
the final model. Other studies have reported no effect of co-administration of oral E2 and
antifungal medication on E2 levels(30-32). CYP3A4 inhibitors such as ketoconazole(32) and
grapefruit juice(30) have been reported to increase E1 (but not E2) levels among women taking
oral E2. We evaluated other CYP3A4-related medications which could result in a drug
interaction with oral E2 through liver metabolism including antiepileptic, antihypertensive,
lipid-lowering and antidiabetic medications (33-35) and commonly used supplements
(multivitamin, vitamin E and fish oil). None of these medications were associated with serum
E2 level.
A post hoc analysis from the EPAT trial showed that exercise and physical activity were
associated with lower levels of E2 over 2 years of intervention with oral E2 1 mg/day(36); we
did not find this association in the ELITE sample. As EPAT participants were older, had higher
BMI and higher baseline E2 levels and E2 levels while taking HT than participants in ELITE, the
difference may be due to different population characteristics.
Endogenous serum E2 levels among premenopausal and perimenopausal women
significantly differed across race–ethnicity at menopausal transition in the Study of Women
Across the Nation (SWAN), with highest serum E2 levels in Hispanic women and lowest levels in
Asian women. The difference in serum E2 level disappeared after adjustment for BMI(37). In
ELITE participants who were postmenopausal women taking HT, race–ethnicity were not
related to achieved serum E2 levels while BMI was significantly associated with serum E2 level.
These findings may suggest that metabolism of both endogenous and exogenous E2 is related
to BMI rather than race–ethnicity.
Page 221
The strength of this study is the randomized clinical trial data providing prospective
repeated measurements of E2 levels and possible correlates that decreased variability of
measurements and allowed determination of active factors associated with E2 levels among
postmenopausal women taking oral E2 therapy with good compliance. Despite the overall
significant trend in serum E2 levels by level of daily alcohol use, the estimate of association with
alcohol intake of more than 2 drinks per day was based on a small sample size, thus needs
further exploration with a larger sample size. The analysis was also limited to total serum E2
levels and did not account for other estrogen components including free E2, E1 and SHBG. Only
11% of women were surgically menopausal and 18 women used antifungal medications during
the study, limiting analyses of these specific associations.
6.6 CONCLUSION
Although titration of HT to specified serum E2 levels is not recommended by the
American College of Obstetricians and Gynecologists (ACOG), American Society for
Reproductive Medicine (ASRM) or NAMS, this study has important public health implications for
postmenopausal women taking HT, as achieved E2 levels relate to the biological response of
HT(6-8). In particular, we have previously reported that the effect of HT on atherosclerosis
progression among postmenopausal women is related to achieved serum E2 levels(2, 3). Higher
E2 levels were associated with reduced atherosclerosis progression when initiated in early
postmenopause (<6 years since menopause), however, higher E2 levels were associated with
greater atherosclerosis progression when initiated in late postmenopause (≥10 years since
menopause)(3). Many significant factors associated with E2 levels in this study such as weight,
alcohol and smoking are modifiable. Postmenopausal women should control their weight and
Page 222
refrain from smoking and limit alcohol use with a goal of obtaining the lowest effective dose of
HT. Postmenopausal women who are obese, drink more than 2 alcoholic beverages per day and
are 10 years since menopause may be at higher risk for atherosclerosis due to higher E2 levels
when taking HT. Postmenopausal women who continue to smoke may need higher E2 dosages
to maintain benefits in clinical outcomes. Health care professionals prescribing HT need to
consider these modifiable life-style factors since these factors may require adjustment for the
most appropriate treatment dose for each individual woman to acquire the desired outcome
with minimal side effects and maximum compliance.
Page 223
6.7 TABLES
Page 224
Table 6.1 Baseline demographic and clinical characteristics
Variable Mean/Frequency Standard deviation/Percent
Age (years)
Total analysis sample 60.7 6.8
Early postmenopause 55.4 3.9
Late postmenopause 64.4 5.5
Time since menopause (years)
Total analysis sample 10.2 7.6
Early postmenopause 3.6 1.8
Late postmenopause 16 5.6
Race/Ethnicity
White, non-Hispanic 202 73.5%
African American, non-Hispanic 21 7.6%
Hispanic 30 10.9%
Asian, non-Hispanic 22 8.0%
Menopause type
Natural 245 89.1%
Surgical 30 10.9%
Body mass index (kg/m2) 27.2 5.6
Creatinine (mg/dL) 0.85 0.15
Creatinine clearance (mL/min) 80.7 22.0
Aspartate aminotransferase (U/L) 21.3 6.8
Alanine aminotransferase (U/L) 20.5 9.0
Fasting blood glucose (mg/dL) 96.3 11.0
Smoking status
Never 170 61.8%
Past smoker 94 34.2%
Current smoker 11 4.0%
Alcohol use
None 139 50.5%
<1 drink per day 97 35.3%
1-2 drinks per day 30 10.9%
>2 drinks per day 9 3.3%
Total weekly metabolic equivalent hours 247 23.3
Medications and supplements use
Lipid-lowering 55 20%
Statin 51 18.5%
Antihypertensive medication 58 21.1%
Calcium channel blocker 14 5.1%
Antifungal 2 0.7%
Anticonvulsant 8 2.9%
Multivitamin 163 59.3%
Vitamin E 46 16.7%
Fish oil 83 30.3%
Page 225
Continuous variables are presented as mean and standard deviation; categorical variables are presented
as frequency, percent.
Page 226
Table 6.2 Association of serum estradiol levels while taking hormone therapy with demographic and
clinical characteristics and median estradiol level by categorical variable
Variable
Women
Visits
Median
estradio
l level
(pg/ml)
Beta
(95%Confidence
interval)
P-value
Overall
P-
value
Age* (years) 275 2160
0.9978 (0.9899 – 1.0059) 0.60
Race/Ethnicity 275 2160
White, non-Hispanic 202 1615
45.0
Reference
0.99
African American, non-
Hispanic 21 141
43.0
0.9807 (0.7985 – 1.2046) 0.85
Hispanic 30 225
54.0
1.0164 (0.8529 – 1.2114) 0.86
Asian, non-Hispanic 22 179
41.0
1.0047 (0.8224 – 1.2276) 0.96
Body mass index* (kg/m
2
) 275 2160
1.0139 (1.0052 – 1.0227) 0.002
Weight* (kg) 275 2147
1.0020 (1.0007 – 1.0034) 0.006
Time since menopause (years) 259 2046
0.9977 (0.9903 – 1.0052) 0.55
Time since menopause* (years) 259 2046
0.9976 (0.9903 – 1.0051) 0.52
Time since menopause 275 2160
Early postmenopause
123 1037
46.0
Reference
Late postmenopause
152 1123
45.0
0.9615 (0.8634 – 1.0707) 0.47
Menopause type 275 2160
Natural 245 1916
43.0
Reference
Surgical 30 244
64.0
1.1982 (1.0096 – 1.4221) 0.04
Menopause type 275 2160
Natural, early 117 981 44.0 Reference
0.12
Natural, late 128 935 41.0 0.9458 (0.8449 – 1.0589) 0.33
Surgical, early 6 56 61.5 1.3209 (0.9115 – 1.9143) 0.14
Surgical, late 24 188 64.5 1.1303 (0.9283 – 1.3765) 0.22
Creatinine* (mg/dL) 258 2143
1.1149 (0.8359 – 1.4873) 0.46
Creatinine clearance* (ml/min) 258 2130
1.0011 (0.9994 – 1.0029) 0.25
Aspartate aminotransferase* (U/L)
258 2143
0.9996 (0.994 – 1.0053) 0.89
Alanine aminotransferase* (U/L)
258 2143
0.9989 (0.9943 – 1.0037) 0.64
Fasting blood glucose* (mg/dl)
256 1042
1.0020 (0.9975 – 1.0066) 0.37
Page 227
Smoking status* 275 2160
Never 170 1338
50.0
Reference
<0.001
Past smoker 101 741
40.0
0.8535 (0.7647 - 0.9528) 0.005
Current smoker 14 81
26.0
0.6694 (0.5516 - 0.8125) <0.001
Alcohol use* 275 2160
None 167 1059
46.0
Reference
<0.001
<1 drink per day 165 828
42.0
1.0359 (0.9667 - 1.1102) 0.32
1-2 drinks per day 56 231
51.0
1.0941 (0.9798 - 1.2218) 0.11
>2 drinks per day 11 42
94.5
1.6780 (1.3386 - 2.1035) <0.001
Total weekly MET hours* 275 2155
0.9997 (0.9988 - 1.0007) 0.63
Moderate + vigorous activity
weekly hours*, tertiles 275 2156
0 - 2.60 193 719
47.0
Reference
0.35
2.62 - 6.75 222 727
44.0
0.9624 (0.569 - 1.6281) 0.15
6.80+ 197 710
44.0
0.9707 (0.9143 - 1.0308) 0.33
Medications and supplements
Lipid lowering medication* 275 2160
No 224 1632
44.0
Reference
Yes 86 528
49.0
1.0166 (0.9299 - 1.1115) 0.72
Statin use* 275 2160
No 228 1684
45.0
Reference
Yes 79 476
46.0
1.0232 (0.9324 - 1.1228) 0.63
Antihypertensive medication* 275 2160
No 223 1677
45.0
Reference
Yes 82 483
46.0
1.1155 (1.0166 - 1.2241) 0.02
Calcium channel blocker* 275 2160
No 263 2065
45.0
Reference
Yes 20 95
54.0
1.0592 (0.8841 - 1.269) 0.53
Diabetes medication* 275 2160
No 273 2136
45.0
Reference
Yes 3 24
43.0
0.9157 (0.6014 - 1.3942) 0.68
Antifungal medication* 275 2160
No 275 2115
45.0
Reference
Page 228
Yes 18 45
46.0
0.8049 (0.6839 - 0.9476) 0.02
Anticonvulsant medication* 275 2160
No 271 2079
45.0
Reference
Yes 17 81
46.0
0.9026 (0.7685 - 1.0602) 0.21
Multivitamin use 275 2160
0.70
Never 51 396
48.0
Reference
Took regularly in the past 61 463
47.0
0.9743 (0.8229 - 1.1537) 0.76
Take regularly now 163 1301
44.0
0.9457 (0.8202 - 1.0906) 0.44
Vitamin E use 275 2160
0.27
Never 148 1178
43.0
Reference
Took regularly in the past 81 613
50.0
1.1035 (0.9756 - 1.2484) 0.12
Take regularly now 46 369
48.0
1.0733 (0.9246 - 1.2459) 0.35
Fish oil use 274 2156
0.53
Never 162 1309
43.0
Reference
Took regularly in the past 29 243
48.0
1.0949 (0.9161 - 1.3088) 0.32
Take regularly now 83 604
48.0
1.0466 (0.9267 - 1.1821) 0.46
Beta estimates and p-values are from mixed effects linear model, adjusted for baseline serum estradiol
level, follow-up visit and reduced estradiol dose indicator; Beta estimates and 95% confidence interval
are back-transformed from the model with log estradiol level.; Back transformed betas represent a fold
change in means from reference category for categorical variable or per increasing unit of continuous
variable.; p-trend for alcohol use p=0.003
*Time varying variable, assessed at each visit
Page 229
Table 6.3 Association of Log Serum Estradiol Levels While Taking Hormone Therapy with Demographic
and Clinical Characteristics
Variable Beta
Standard
error
Age* (years) -0.0022 0.0041
Race/Ethnicity
White, non-Hispanic Reference
African American, non-Hispanic -0.0195 0.1049
Hispanic 0.0163 0.0895
Asian, non-Hispanic 0.0047 0.1022
Body mass index* (kg/m
2
) 0.0138 0.0044
Weight* (kg) 0.0020 0.0007
Time since menopause (years) -0.0023 0.0038
Time since menopause* (years) -0.0024 0.0038
Time since menopause
Early postmenopause
Reference
Late postmenopause
-0.0393 0.0549
Menopause type
Natural Reference
Surgical 0.1808 0.0874
Menopause type
Natural, early Reference
Natural, late -0.0557 0.0576
Surgical, early 0.2783 0.1893
Surgical, late 0.1225 0.1005
Creatinine* (mg/dL) 0.1088 0.1470
Creatinine clearance* (ml/min) 0.0011 0.0009
Aspartate aminotransferase* (U/L)
-0.0004 0.0029
Alanine aminotransferase* (U/L)
-0.0011 0.0024
Fasting blood glucose* (mg/dl)
0.0020 0.0023
Smoking status*
Never Reference
Past smoker -0.1584 0.0561
Page 230
Current smoker -0.4013 0.0988
Alcohol use*
None Reference
<1 drink per day 0.0353 0.0353
1-2 drinks per day 0.0899 0.0563
>2 drinks per day 0.5176 0.1153
Total weekly MET hours* -0.0003 0.0005
Moderate + vigorous activity weekly hours*, tertiles
0 - 2.60 Reference
2.62 - 6.75 -0.0383 0.2682
6.80+ -0.0297 0.0306
Medications and supplements
Lipid lowering medication*
No Reference
Yes 0.0165 0.0455
Statin use*
No Reference
Yes 0.0229 0.0474
Antihypertensive medication*
No Reference
Yes 0.1093 0.0474
Calcium channel blocker*
No Reference
Yes 0.0575 0.0922
Diabetes medication*
No Reference
Yes -0.0881 0.2145
Antifungal medication*
No Reference
Yes -0.2170 0.0832
Anticonvulsant medication*
No Reference
Yes -0.1025 0.0821
Page 231
Multivitamin use
Never Reference
Took regularly in the past -0.0260 0.0862
Take regularly now -0.0558 0.0727
Vitamin E use
Never Reference
Took regularly in the past 0.0985 0.0629
Take regularly now 0.0707 0.0761
Fish oil use
Never Reference
Took regularly in the past 0.0907 0.0910
Take regularly now 0.0455 0.0621
Beta estimates and standard errors are from mixed effects linear model, adjusted for baseline serum
estradiol level, follow-up visit and reduced estradiol dose indicator.
*Time varying variable
Page 232
Table 6.4 Multivariable association of estradiol levels while taking hormone therapy with demographic and clinical characteristics among total
analysis sample and by postmenopausal strata
Total analysis sample Early postmenopause Late postmenopause
Variable Women Visits Beta
(95%
Confidence
interval)
P-
value
Women Visits Beta
(95%
Confidence
interval)
P-
value
Women Visits Beta
(95%
Confidence
interval)
P-
value
n 275 2160
123 1037
152 1123
Body mass index
(kg/m2)* 275 2160 1.015 (1.0067 - 1.0235) <0.001 123 1037 1.0109 (0.9991 - 1.0229) 0.07 152 1123 1.0181 (1.0064 - 1.03) 0.002
Smoking status*
<0.001
<0.001
0.04
Never 170 1338 Reference
80 690 Reference
90 648 Reference
Past smoker 101 741 0.8516 (0.7651 - 0.9481) 0.003 43 301 0.7806 (0.664 - 0.9178) 0.003 58 440 0.9261 (0.8022 - 1.0692) 0.30
Current smoker 14 81 0.6709 (0.5545 - 0.8119) <0.001 8 46 0.6515 (0.5109 - 0.8309) <0.001 6 35 0.6576 (0.4742 - 0.9122) 0.01
Alcohol use*
<0.001
<0.001
0.42
None 167 1059 Reference
69 467 Reference
98 592 Reference
<1 drink per day 165 828 1.0414 (0.9722 - 1.1157) 0.25 76 420 1.0163 (0.915 - 1.129) 0.76 89 408 1.0556 (0.9631 - 1.157) 0.25
1-2 drinks per day 56 231 1.0999 (0.9862 - 1.2268) 0.09 28 128 1.1190 (0.9561 - 1.3097) 0.16 28 103 1.0661 (0.9132 - 1.2447) 0.42
>2 drinks per day 11 42 1.6979 (1.3572 - 2.1243) <0.001 6 22 2.2371 (1.6276 - 3.0751) <0.001 5 20 1.2607 (0.9189 - 1.7299) 0.15
Beta estimates and p-values are from mixed effects linear model, adjusted for baseline serum estradiol level, follow-up visit and reduced
estradiol dose indicator; Beta estimates and 95% confidence interval are back-transformed from the model with log estradiol level.; Back
transformed betas represent a fold change in means from reference category for categorical variable or per increasing unit of continuous
variable.; p-trend for alcohol use for total analysis sample p=0.002, early postmenopause p=0.005 and late postmenopause p=0.15; Interaction
of alcohol use by time-since-menopause p=0.06; *Time varying variable
Page 233
Table 6.5 Multivariable Association of Log-Transformed Estradiol Levels While Taking Hormone Therapy
with Demographic and Clinical Characteristics Among Total Sample and by Postmenopausal Strata
Total analysis
sample
Early
postmenopause
Late
postmenopause
Variable Beta
Standard
error
Beta
Standard
error
Beta
Standard
error
Body mass index
(kg/m2)* 0.0149 0.0042 0.0108 0.0060 0.0179 0.0059
Smoking status*
Never Reference
Past smoker -0.1606 0.0547
-
0.2477 0.0826 -0.0768 0.0733
Current smoker -0.3991 0.0973
-
0.4285 0.1241 -0.4191 0.1669
Alcohol use*
None Reference
<1 drink per day 0.0406 0.0351 0.0162 0.0536 0.0541 0.0468
1-2 drinks per day 0.0952 0.0557 0.1124 0.0803 0.0640 0.0790
>2 drinks per day 0.5294 0.1143 0.8052 0.1623 0.2317 0.1614
Beta estimates and standard errors are from mixed effects linear model, adjusted for baseline serum
estradiol level, follow-up visit and reduced estradiol dose indicator;
*Time varying variable
Page 234
Table 6.6 Estimated Serum Estradiol Levels for Each Multivariable Model Determinant
Variable
Least square mean
(pg/ml)
95% confidence interval
(pg/ml)
Body mass index category < 25 kg/m
2
37.9 35.2-40.8
25 to < 30 kg/m
2
41.1 38.2-44.2
30 kg/m
2
43.2 39.6-47.1
Smoking status Never 43.2 40.3-46.3
Past smoker 36.9 33.8-40.3
Current smoker 28.9 24.1-34.7
Alcohol use None 39.0 36.5-41.6
<1 drink per day 40.4 37.8-43.2
1-2 drinks per day 42.7 38.5-47.3
>2 drinks per day 65.5 52.4-81.7
Least square means are from mixed effects linear model, adjusted for baseline serum estradiol level,
follow-up visit and reduced estradiol dose indicator
Page 235
6.8 REFERENCES
1. The 2017 hormone therapy position statement of The North American Menopause
Society. Menopause 2018;25(11):1362-87.
2. Karim R, Hodis HN, Stanczyk FZ, et al. Relationship between serum levels of sex
hormones and progression of subclinical atherosclerosis in postmenopausal women. J
Clin Endocrinol Metab 2008;93(1):131-8.
3. Sriprasert I, Hodis HN, Karim R, et al. Differential effect of plasma estradiol on subclinical
atherosclerosis progression in early versus late postmenopause. J Clin Endocrinol Metab
2018.
4. Boardman HMP, Hartley L, Eisinga A, et al. Hormone therapy for preventing
cardiovascular disease in post-menopausal women. Cochrane Database of Systematic
Reviews 2015(3).
5. Hodis HN, Mack WJ, Henderson VW, et al. Vascular Effects of Early versus Late
Postmenopausal Treatment with Estradiol. New England Journal of Medicine
2016;374(13):1221-31.
6. Barbieri RL. Hormone treatment of endometriosis: the estrogen threshold hypothesis.
Am J Obstet Gynecol 1992;166(2):740-5.
7. Santoro N, Allshouse A, Neal-Perry G, et al. Longitudinal changes in menopausal
symptoms comparing women randomized to low-dose oral conjugated estrogens or
transdermal estradiol plus micronized progesterone versus placebo: the Kronos Early
Estrogen Prevention Study. Menopause 2017;24(3):238-46.
Page 236
8. Ettinger B, Ensrud KE, Wallace R, et al. Effects of ultralow-dose transdermal estradiol on
bone mineral density: a randomized clinical trial. Obstet Gynecol 2004;104(3):443-51.
9. O'Connell MB. Pharmacokinetic and pharmacologic variation between different
estrogen products. J Clin Pharmacol 1995;35(9S):18S-24S.
10. Kuhl H. Pharmacology of estrogens and progestogens: influence of different routes of
administration. Climacteric 2005;8 Suppl 1:3-63.
11. Stanczyk FZ, Archer DF, Bhavnani BR. Ethinyl estradiol and 17beta-estradiol in combined
oral contraceptives: pharmacokinetics, pharmacodynamics and risk assessment.
Contraception 2013;87(6):706-27.
12. Probst-Hensch NM, Ingles SA, Diep AT, et al. Aromatase and breast cancer susceptibility.
Endocr Relat Cancer 1999;6(2):165-73.
13. McTiernan A, Wu L, Chen C, et al. Relation of BMI and physical activity to sex hormones
in postmenopausal women. Obesity (Silver Spring) 2006;14(9):1662-77.
14. Baglietto L, English DR, Hopper JL, et al. Circulating steroid hormone concentrations in
postmenopausal women in relation to body size and composition. Breast Cancer Res
Treat 2009;115(1):171-9.
15. Karim R, Mack WJ, Hodis HN, et al. Influence of age and obesity on serum estradiol,
estrone, and sex hormone binding globulin concentrations following oral estrogen
administration in postmenopausal women. J Clin Endocrinol Metab 2009;94(11):4136-
43.
Page 237
16. Hetemaki N, Savolainen-Peltonen H, Tikkanen MJ, et al. Estrogen Metabolism in
Abdominal Subcutaneous and Visceral Adipose Tissue in Postmenopausal Women. J Clin
Endocrinol Metab 2017;102(12):4588-95.
17. Gruber CJ, Tschugguel W, Schneeberger C, et al. Production and actions of estrogens. N
Engl J Med 2002;346(5):340-52.
18. Ginsburg ES, Mello NK, Mendelson JH, et al. Effects of alcohol ingestion on estrogens in
postmenopausal women. JAMA 1996;276(21):1747-51.
19. Endogenous H, Breast Cancer Collaborative G, Key TJ, et al. Circulating sex hormones
and breast cancer risk factors in postmenopausal women: reanalysis of 13 studies. Br J
Cancer 2011;105(5):709-22.
20. Playdon MC, Coburn SB, Moore SC, et al. Alcohol and oestrogen metabolites in
postmenopausal women in the Women's Health Initiative Observational Study. Br J
Cancer 2018;118(3):448-57.
21. Rinaldi S, Peeters PH, Bezemer ID, et al. Relationship of alcohol intake and sex steroid
concentrations in blood in pre- and post-menopausal women: the European Prospective
Investigation into Cancer and Nutrition. Cancer Causes Control 2006;17(8):1033-43.
22. Onland-Moret NC, Peeters PH, van der Schouw YT, et al. Alcohol and endogenous sex
steroid levels in postmenopausal women: a cross-sectional study. J Clin Endocrinol
Metab 2005;90(3):1414-9.
23. Shafrir AL, Zhang X, Poole EM, et al. The association of reproductive and lifestyle factors
with a score of multiple endogenous hormones. Horm Cancer 2014;5(5):324-35.
Page 238
24. Sarkola T, Makisalo H, Fukunaga T, et al. Acute effect of alcohol on estradiol, estrone,
progesterone, prolactin, cortisol, and luteinizing hormone in premenopausal women.
Alcohol Clin Exp Res 1999;23(6):976-82.
25. Lobo RA, Cassidenti DL. Pharmacokinetics of oral 17 beta-estradiol. J Reprod Med
1992;37(1):77-84.
26. Geisler J, Omsjo IH, Helle SI, et al. Plasma oestrogen fractions in postmenopausal
women receiving hormone replacement therapy: influence of route of administration
and cigarette smoking. J Endocrinol 1999;162(2):265-70.
27. Mueck AO, Seeger H. Smoking, estradiol metabolism and hormone replacement
therapy. Curr Med Chem Cardiovasc Hematol Agents 2005;3(1):45-54.
28. Ruan X, Mueck AO. Impact of smoking on estrogenic efficacy. Climacteric 2015;18(1):38-
46.
29. Michnovicz JJ, Hershcopf RJ, Naganuma H, et al. Increased 2-hydroxylation of estradiol
as a possible mechanism for the anti-estrogenic effect of cigarette smoking. N Engl J
Med 1986;315(21):1305-9.
30. Schubert W, Eriksson U, Edgar B, et al. Flavonoids in grapefruit juice inhibit the in vitro
hepatic metabolism of 17 beta-estradiol. Eur J Drug Metab Pharmacokinet
1995;20(3):219-24.
31. Annas A, Carlstrom K, Alvan G, et al. The effect of ketoconazole and diltiazem on
oestrogen metabolism in postmenopausal women after single dose oestradiol
treatment. Br J Clin Pharmacol 2003;56(3):334-6.
Page 239
32. Wiesinger H, Berse M, Klein S, et al. Pharmacokinetic interaction between the CYP3A4
inhibitor ketoconazole and the hormone drospirenone in combination with
ethinylestradiol or estradiol. Br J Clin Pharmacol 2015;80(6):1399-410.
33. Menon RM, Badri PS, Wang T, et al. Drug-drug interaction profile of the all-oral anti-
hepatitis C virus regimen of paritaprevir/ritonavir, ombitasvir, and dasabuvir. J Hepatol
2015;63(1):20-9.
34. Terada T, Hira D. Intestinal and hepatic drug transporters: pharmacokinetic,
pathophysiological, and pharmacogenetic roles. J Gastroenterol 2015;50(5):508-19.
35. Hassan LS, Monson RS, Danielson KK. Oestradiol levels may differ between
premenopausal women, ages 18-50, with type 1 diabetes and matched controls.
Diabetes Metab Res Rev 2017;33(2).
36. Choudhury F, Bernstein L, Hodis HN, et al. Physical activity and sex hormone levels in
estradiol- and placebo-treated postmenopausal women. Menopause 2011;18(10):1079-
86.
37. Randolph JF, Jr., Sowers M, Gold EB, et al. Reproductive hormones in the early
menopausal transition: relationship to ethnicity, body size, and menopausal status. J Clin
Endocrinol Metab 2003;88(4):1516-22.
CHAPTER 7
SUMMARY AND FUTURE DIRECTIONS
Page 241
7.1 SUMMARY
This dissertation tested the hormone therapy (HT) timing hypothesis on atherosclerosis
and related vascular risk factors. The HT timing hypothesis postulates that there is a differential
effect of HT, particularly estradiol (E2) on cardiovascular disease by time-since-menopause. In
the first paper using the Early versus Late Intervention Trial with Estradiol (ELITE) data(1), we
demonstrated that E2 levels are differentially associated with atherosclerosis progression
determined by common carotid intima media thickness (CIMT) according to timing of hormone
therapy initiation. With higher E2 levels, the CIMT progression rate was decreased among early
postmenopausal women, but increased among late postmenopausal women(2). The second and
third studies demonstrated that both E2 dose and E2 levels were associated with changes in
metabolic, coagulation and anticoagulation measures in REPLENISH data(3). These associations
differed in early and late postmenopausal women, which may explain the differential effect of E2
on atherosclerosis and thromboembolism risk. The association between E2 level and decreased
lipid levels found in early postmenopausal women was attenuated by longer time since
menopause. The association between E2 level and decreased coagulation levels found in early
postmenopausal women was enhanced by longer time since menopause. Noting the large
variability in achieved E2 levels among women using HT in both the ELITE and REPLENISH trials,
study four identified determinants of serum E2 levels among postmenopausal women taking HT,
which included body mass index, surgical menopause, alcohol use, smoking and antifungal
medication.
The cumulative results from these studies indicate that time-since-menopause modifies
the effect of E2 levels with atherosclerosis progression and the mechanism may occur through
Page 242
the differential changes in metabolic, coagulation and anticoagulation measures between early
and late postmenopausal women. Achievement of certain E2 levels according to time since
menopause may be influenced through modifiable determinants for possible additional
cardiovascular benefits.
7.2 RELATED WORK
In addition to the work in this dissertation, we have conducted several studies related to
the hormone therapy timing hypothesis on atherosclerosis using data from the ELITE and
REPLENISH trials.
7.2.1 Modifying Effect of ApoE4 Genotype on the Association Between Metabolic Phenotype
and Subclinical Atherosclerosis in Postmenopausal Women
In ELITE, we examined the interaction between metabolic phenotype and apolipoprotein
E4 (ApoE4) genotype on atherosclerosis measured by CIMT(4). Metabolic phenotypes were
determined by a K-means clustering algorithm using 9 metabolic and vascular biomarkers:
glucose, the HOMA score, ketones, triglycerides, high density lipoprotein cholesterol (HDL-C), low
density lipoprotein cholesterol (LDL-C), hemoglobin A1C (HbA1c) and systolic and diastolic blood
pressure. Using these measures, 3 clusters were identified that were labeled as healthy, high
blood pressure and poor metabolic phenotypes. The healthy and poor metabolic phenotypes
significantly differed on all metabolic biomarkers. Healthy and high blood pressure phenotypes
significantly differed on all biomarkers except glucose and HbA1c; high blood pressure and poor
metabolic phenotype significantly differed on all biomarkers except ketones and LDL-C(5).
In this cross-sectional analysis, we used general linear models to test whether the
association between metabolic phenotypes and common carotid intima-media thickness (CIMT)
Page 243
differed by ApoE4 genotype. The results showed that ApoE4+ women with poor metabolic
phenotype (elevated LDL-C, triglycerides, glucose, HbA1C, HOMA score, and lower HDL-C) had
the highest CIMT compared with all other groups. In ApoE4+ women, CIMT was significantly
higher in those with poor metabolic phenotype compared with healthy (p = 0.0003) and high
blood pressure (p = 0.001) phenotypes. These results indicate that metabolic phenotype had a
negative effect on CIMT in women with ApoE4+ but not ApoE4- (interaction p = 0.001). These
effects were not observed on CIMT progression in longitudinal analysis using mixed effects linear
models.
These study findings have significant clinical and public health implications as preventive
intervention strategies targeted to postmenopausal women at high metabolic risk (in particular
those who have ApoE4+ genotype) can potentially reduce the burden of CHD.
7.2.2 Determinants of attained estradiol levels in response to oral estradiol plus progesterone
therapy
Using REPLENISH trial data, we evaluated baseline determinants of attained E2 levels in
response to combined oral E2 and progesterone. The determinants of E2 level in REPLENISH trial
data were similar as in ELITE data. After adjusting for E2 dose and baseline E2 level, on-trial E2
levels were significantly negatively associated with time-since-menopause, current smoking
while positively associated with current alcohol use.
Results from this study supplemented the information from ELITE on determinants of E2
level among postmenopausal women using oral combined HT regimen of E2 and progesterone.
The significant determinants of E2 levels in both REPLENISH and ELITE data were smoking and
alcohol use. However, there were several inconsistent determinants between two studies. In
Page 244
ELITE, time since menopause did not significantly associate with E2 level whereas it did in
REPLENISH data. This could be due to the unique population sampling in the ELITE, with stratified
sampling by time-since-menopause. We also identified significant association between E2 level
with body mass index, creatinine level, alanine aminotransferase level, surgical menopause and
antifungal medication use in the ELITE data. A possible explanation could be that the determinant
variables in REPLENISH were limited to baseline instead of on-trial assessments as in the ELITE
data.
7.2.3 Modifying effect of time since menopause and age on the association of estradiol dose
and estradiol level on coagulation measures
Using REPLENISH data, we explored the modifying effect of time-since-menopause and
age on the association of E2 dose and E2 level on coagulation and anticoagulation measures;
prothrombin time (PT), activated partial thromboplastin time (APTT) antithrombin (AT), protein
C (PROTC), protein S (PROTS) and fibrinogen (FIB).
Results showed that with longer time-since-menopause, higher E2 dose was associated
with lower levels of all coagulation measures tested. However, the time since menopause
modifying effect was reduced and no longer significant after adjustment for age. With longer
time-since-menopause, higher E2 level was associated with lower levels of PT, APTT, PROTC and
PROTS. After adjustment for age, the significant modifying effect of time-since-menopause on E2
level remained only for PT.
With older age, higher E2 dose and higher E2 level was associated with lower levels of all
coagulation and anticoagulation measures tested. Age significantly modified the association
between E2 dose and level with coagulation measures in directions similar to the modifications
Page 245
by time since menopause; the age modifying effect remained significant after adjustment for
time since menopause.
As age and time since menopause are usually positively correlated, results from this study
indicated that age may be a stronger chronological parameter than time since menopause to
represent the modifying effect of E2 on changes in coagulation and anticoagulation measures. As
age is a known risk factor for venous thromboembolism, our findings may be able to explain the
risk of venous thromboembolism among older postmenopausal women using HT.
7.3 FUTURE DIRECTIONS
The findings from this dissertation and our related work showed that time-since-
menopause significantly modified the effect of HT on atherosclerosis and vascular-related
metabolic and coagulation measures. In addition to the work reported here, we have collected
additional data to postulate that the differential effect of HT on atherosclerosis progression by
time-since-menopause may be explained by several factors such as the extend of underlying
atherosclerosis at the time of initiation of HT, and differential effects of HT on metabolic
measures, lipoprotein subclasses, and inflammation markers by time since menopause.
7.3.1 Effect of hormone therapy on atherosclerosis by extent of underlying atherosclerosis at
time of hormone therapy initiation
To test the hypothesis of HT effect on atherosclerosis status at HT initiation, we will
conduct a pooled analysis of data from several randomized clinical studies testing HT intervention
in postmenopausal women. These trials were all conducted at the USC Atherosclerosis Research
Unit and used the same protocols for assessment of CIMTprogression over trial follow-up. As
these trials included postmenopausal women with and without pre-existing cardiovascular
Page 246
disease, the combined data provide a wide range of atherosclerosis levels at the time of HT
initiation at randomization. The trial data to be included in this pooled analysis in addition to
ELITE(1) are the Women’s Estrogen-progestin Lipid-Lowering Hormone Atherosclerosis
Regression Trial(6) (WELL-HART, conducted in women with coronary artery disease) and the
Estrogen in Prevention of Atherosclerosis Trial(7) (EPAT, conducted in women without coronary
artery disease).
The objective of this pooled analysis will be to test the hypothesis that the effect of HT on
atherosclerosis progression differs by extent of atherosclerosis at HT initiation; Specifically, we
hypothesize that HT may not be beneficial in women with high levels of atherosclerosis, but will
be beneficial in women with lower levels of atherosclerosis, measured by CIMT.
7.3.2 Possible mechanisms of HT effect on atherosclerosis progression
To explore the mechanisms of an HT effect on atherosclerosis progression, we are using
ELITE data and stored samples to conduct studies evaluating the HT effect on metabolic
measures, lipoprotein subclasses, inflammation markers. We will be testing: (1) whether HT has
an effect on these measures and whether the HT effect differs by time since menopause; (2)
whether these measures are associated with CIMT progression; and (3) whether these measures
mediate the effect of HT on CIMT progression, particularly in early menopause.
For example, in one study we will examine the associations between CIMT progression
and metabolic profiles (total cholesterol, triglyceride, LDL-C, HDL-C, glucose, HbA1c) in early
versus late postmenopausal women and quantify the extent of the beneficial effect of HT on
CIMT progression that can be explained by treatment-related changes of these metabolic
Page 247
variables in early versus late postmenopausal women. We will conduct similar analyses using
lipoprotein subclasses, inflammatory markers, mRNA expression and DNA methylation.
7.3.3 A new atherosclerosis measure
To date, we have primarily used CIMT to measure atherosclerosis as an indicator of CHD
risk. We have recently used ELITE ultrasound images to obtain grey scale median (GSM) as a
relatively newer measure for atherosclerosis. GSM was analyzed from ultrasonographic images
to assess arterial wall tissue characteristics. Lower GSM values indicate potentially greater
arterial wall lipid accumulation.
We compared the associations of CHD risk factors with CIMT and GSM of the carotid
artery intima-media complex in ELITE postmenopausal women. Results showed that while GSM
and CIMT share some common cardiovascular risk factors, GSM is associated with a greater
number of CHD risk factors such as lipids, whereas cimt was more associated with smoking and
blood pressure This study indicated that we may use GSM as a compliment to CIMT to measure
atherosclerosis risk among postmenopausal women.
We plan to compare cardiovascular risk factors of CIMT and GSM in postmenopausal
women using ELITE data. We will evaluate the effect of HT on progression of GSM compared with
placebo stratified by time since menopause. We will also study the association between E2 level
with progression of GSM stratified by time since menopause.
7.3.4 Future studies
As CHD remains the most common cause of death among postmenopausal women and
HT reduces risk of atherosclerosis and all-cause mortality in certain women, we will continue to
use the rich ELITE data to explore the mechanisms underlying the effect of E2 on atherosclerosis
Page 248
according to time-since-menopause through several mechanisms including endocrine,
metabolic, coagulation, inflammatory pathways as well as gene expression and methylation
pathways. We are currently actively working on several analyses of metabolic measures,
lipoprotein subclasses and inflammatory markers. We will obtain additional data on reproductive
hormonal profiles, gene expression and methylation in the near future.
This dissertation contributed some information on possible mechanisms of the
differential HT benefit on atherosclerosis among postmenopausal women. With additional data,
we hope to provide insights on the mechanisms of HT on atherosclerosis and CHD in younger
postmenopausal women.
Page 249
7.4 REFERENCES
1. Hodis HN, Mack WJ, Henderson VW, et al. Vascular Effects of Early versus Late
Postmenopausal Treatment with Estradiol. New England Journal of Medicine
2016;374(13):1221-31.
2. Sriprasert I, Hodis HN, Karim R, et al. Differential effect of plasma estradiol on subclinical
atherosclerosis progression in early versus late postmenopause. The Journal of clinical
endocrinology and metabolism 2018.
3. Lobo RA, Archer DF, Kagan R, et al. A 17beta-Estradiol-Progesterone Oral Capsule for
Vasomotor Symptoms in Postmenopausal Women: A Randomized Controlled Trial.
Obstetrics and gynecology 2018;132(1):161-70.
4. Sriprasert I, Mack WJ, Hodis HN, et al. Effect of ApoE4 Genotype on the Association
Between Metabolic Phenotype and Subclinical Atherosclerosis in Postmenopausal
Women. The American journal of cardiology 2019.
5. Rettberg JR, Dang H, Hodis HN, et al. Identifying postmenopausal women at risk for
cognitive decline within a healthy cohort using a panel of clinical metabolic indicators:
potential for detecting an at-Alzheimer's risk metabolic phenotype. Neurobiol Aging
2016;40:155-63.
6. Hodis HN, Mack WJ, Azen SP, et al. Hormone therapy and the progression of coronary-
artery atherosclerosis in postmenopausal women. New England Journal of Medicine
2003;349(6):535-45.
Page 250
7. Hodis HN, Mack WJ, Lobo RA, et al. Estrogen in the prevention of atherosclerosis - A
randomized, double-blind, placebo-controlled trial. Annals of internal medicine
2001;135(11):939-53.
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Effects of post-menopausal hormone therapy on arterial stiffness in the ELITE trial
PDF
Sex hormones and atherosclerosis in postmenopausal women
PDF
Effect of menopausal hormone replacement therapy on lipoprotein particle fractions and association with carotid artery intima-media thickness and grey-scale median, independent measurements of su...
PDF
Effect of estradiol on circulating levels of inflammatory cytokines in postmenopausal women
PDF
The effects of hormone therapy on carotid artery intima-media thickness and serum lipids by ApoE4 genotype
PDF
A randomized controlled trial of hormone therapy on postmenopausal women’s quality of life
PDF
Dietary fats, atherosclerosis, and cardiovascular disease
PDF
Subclinical carotid atherosclerosis, psychosocial measures, and cognitive function in middle- to older-aged adults
PDF
Association of oophorectomy with progression of carotid atherosclerosis
PDF
Association of carotid artery stiffness with measures of cognition in older adults
PDF
Associations between physical activities with bone mineral density in postmenopausal women
PDF
Relationship of blood pressure and antihypertensive medications to cognitive change in the BVAIT, WISH, and ELITE clinical trials
PDF
The associations of inflammatory markers with progression of subclinical atherosclerosis in early and late postmenopausal women
PDF
Associations between isoflavone soy protein (ISP) supplementation and breast cancer in postmenopausal women in the Women’s Isoflavone Soy Health (WISH) clinical trial
PDF
Cardiovascular disease risk factors and cognitive function
PDF
The role of heritability and genetic variation in cancer and cancer survival
PDF
Genes and hormonal factors involved in the development or recurrence of breast cancer
PDF
Factors that influence mammographic density: role of estrogen metabolism genes, biomarkers of inflammation, and lifestyle
PDF
Antibiotic resistance in a large ICU cohort: 1995-2002
PDF
The effect of vitamin D supplementation on the progression of carotid intima-media thickness and arterial stiffness in elderly African American women: Results of a randomized placebo-controlled trial
Asset Metadata
Creator
Sriprasert, Intira
(author)
Core Title
Hormone therapy timing hypothesis and atherosclerosis
School
Keck School of Medicine
Degree
Doctor of Philosophy
Degree Program
Epidemiology
Publication Date
08/03/2020
Defense Date
04/23/2020
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
atherosclerosis,cardiovascular disease,estradiol,hormone therapy,OAI-PMH Harvest
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Mack, Wendy Jean (
committee chair
), Hodis, Howard Neil (
committee member
), Karim, Roksana (
committee member
), McKean-Cowdin, Roberta (
committee member
), Stanczyk, Frank (
committee member
)
Creator Email
intiras@yahoo.com,sriprase@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c89-358195
Unique identifier
UC11666737
Identifier
etd-Sriprasert-8877.pdf (filename),usctheses-c89-358195 (legacy record id)
Legacy Identifier
etd-Sriprasert-8877.pdf
Dmrecord
358195
Document Type
Dissertation
Rights
Sriprasert, Intira
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
atherosclerosis
cardiovascular disease
estradiol
hormone therapy