University of Southern California Dissertations and Theses

The associations of inflammatory markers with progression of subclinical atherosclerosis in early and late postmenopausal women

(USC Thesis Other)

The associations of inflammatory markers with progression of subclinical atherosclerosis in early and late postmenopausal women

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

The associations of inflammatory markers with progression of
subclinical atherosclerosis in early and late postmenopausal women

By

Zhen Weng

A Thesis Presented to the
FACULTY OF THE USC KECK SCHOOL OF MEDICINE
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(BIOSTATISTICS)

December 2020

Copyright 2020 Zhen Weng

ii
Acknowledgements
I would like to thank to my mentor Dr. Mack for guiding me to do such a meaningful and
interesting study and always providing me with supportive suggestions on my analysis. I would
also thank my other committee members, Dr. Hodis, Dr. Karim for their guidance and advice.
Thanks to my colleagues Xiaofu, Naoko, Intira for their kind help.

iii
TABLE OF CONTENTS

Acknowledgements ......................................................................................................................... ii
List of Tables ................................................................................................................................. iv
Abstract .......................................................................................................................................... vi
Introduction ..................................................................................................................................... 1
Methods........................................................................................................................................... 2
Study design ............................................................................................................................ 2
CIMT and GSM assessment ................................................................................................... 3
Statistical analysis ................................................................................................................... 4
Results ............................................................................................................................................. 5
1. Hormone Therapy effects on the inflammation measures .................................................. 5
2. Association between CIMT and each inflammatory marker in the total sample ................ 7
3. Association between GSM and each inflammatory marker in the total sample ............... 36
Discussion ..................................................................................................................................... 54
References ..................................................................................................................................... 56

iv
List of Tables
Table 1A: Effect of hormone therapy on biomarkers of inflammation in postmenopausal women6
Table 1B: Effect of Estradiol (Early Postmenopausal). .................................................................. 6
Table 1C: Effect of Estradiol (Late Postmenopausal). ................................................................... 6
Table 2 Mixed effects model estimates for IL-1⍺ in relation to CIMT .......................................... 9
Table 3 Mixed effects model estimates for MIP-1⍺ in relation to CIMT ..................................... 11
Table 4 Mixed effects model estimates for IFN-γ in relation to CIMT ........................................ 13
Table 5 Mixed effects model estimates for sICAM-1 in relation to CIMT .................................. 15
Table 6 Mixed effects model estimates for sVCAM-1 in relation to CIMT ................................ 17
Table 7 Mixed effects model estimates for IL-6 in relation to CIMT .......................................... 19
Table 8 Mixed effects model estimates for IL-8 in relation to CIMT .......................................... 21
Table 9 Mixed effects model estimates for IL-10 in relation to CIMT ........................................ 23
Table 10 Mixed effects model estimates for MCP-1 in relation to CIMT.................................... 25
Table 11 Mixed effects model estimates for P-selectin in relation to CIMT ................................ 27
Table 12 Mixed effects model estimates for IL-1β in relation to CIMT ...................................... 29
Table 13 Mixed effects model estimates for TNF- ⍺ in relation to CIMT .................................... 31
Table 15 Mixed effects model estimates for VEGF in relation to CIMT ..................................... 35
Table 16 Mixed effects model estimates for E-Selectin in relation to GSM ................................ 38
Table 17 Mixed effects model estimates for sICAM-1 in relation to GSM ................................. 40
Table 18 Mixed effects model estimates for IL-8 in relation to GSM.......................................... 41
Table 19 Mixed effects model estimates for P-selectin in relation to GSM ................................. 43
Table 20 Mixed effects model estimates for MCP-1 in relation to GSM ..................................... 44
Table 21 Mixed effects model estimates for SVCAM-1 in relation to GSM ............................... 45
v
Table 22 Mixed effects model estimates for VEGF in relation to GSM ..................................... 46
Table 23 Mixed effects model estimates for IL-1⍺ in relation to GSM ....................................... 47
Table 24 Mixed effects model estimates for IL-1β in relation to GSM....................................... 48
Table 25 Mixed effects model estimates for IL-6 in relation to GSM.......................................... 49
Table 26 Mixed effects model estimates for IL-10 in relation to GSM....................................... 50
Table 27 Mixed effects model estimates for IFN-γ in relation to GSM ....................................... 51
Table 28 Mixed effects model estimates for TNF- ⍺ in relation to GSM .................................... 52
Table 29 Mixed effects model estimates for MIP-1⍺ in relation to GSM ................................... 53

vi
Abstract
Introduction
Atherosclerosis is characterized by changes in the structure and composition of blood vessels.
Carotid artery intima-media thickness (CIMT) and grey scale median (GSM) measures capture
different aspects of atherosclerosis. The Early Versus Late Intervention Trial with Estradiol
(ELITE) showed that hormone therapy (HT) reduced CIMT progression in early but not late
postmenopausal women. Given these ELITE results supporting the HT timing effect, we sought
to study whether effects on inflammation may in part explain the benefit on atherosclerosis
progression. This thesis evaluated the associations of inflammatory biomarkers with
CIMT/GSM.

Methods
ELITE was a single-center, randomized, double-blinded, placebo-controlled trial of HT.
Participants were randomly assigned to receive HT or placebo treatment in a 1:1 ratio.
Randomization was stratified by CIMT (<0.75 or ≥0.75 mm), hysterectomy status (yes or no),
and timing of postmenopause status (early, <6 year; late ≥10 years). After randomization, women
attended study clinic visits every month for the first 6 months and every other month in the
follow-up period until trial completion.
In this post hoc trial analysis, the trial outcomes were CIMT and GSM. The independent
variables were each of the circulating levels of 14 inflammatory biomarkers measured three
times: at baseline prior to randomization, 12 months and 36 months post-randomization.
vii
Mixed-effects linear models were used to analyze per-participant CIMT/GSM progression.
Analysis was performed for the total sample, separately by two strata (early/late postmenopause),
and separately by four strata (early/late postmenopause and HT versus placebo randomized
treatment).

Results and conclusion
IFN-γ (p=0.011) , sICAM-1 (p=0.035) , and MCP-1 (p=0.004) were inversely associated with
CIMT progression in early postmenopause/placebo treatment group. sVCAM-1 was inversely
associated with CIMT progression in late postmenopause/HT group (p=0.018). IL-8 was
positively associated with CIMT progression in early postmenopause/HT group (p=0.006). IL-1β
was inversely associated with CIMT progression in early postmenopause/HT group (p=0.031).
E-selectin was positively associated with GSM progression in early postmenopause /placebo
treatment group (p=0.048). P-selectin was positively associated with GSM progression in the
early postmenopause stratum (p=0.026).
In conclusion, based on our current methods and analysis, IL-8 may in part explain the effect of
HT in reducing the progression of subclinical atherosclerosis in early postmenopausal women.

Introduction
Atherosclerosis is characterized by changes in the structure and composition of blood vessels.
Noninvasive vascular imaging captures underlying structural and compositional arterial
changes.
[1]
Carotid artery intima-media thickness (CIMT) is one such anatomic measure that has
been widely used. Grey scale median (GSM), measured from the echogenicity of the ultrasound
image, is a novel indicator of lipid deposition in the arterial wall.
[2]
Thus, CIMT and GSM
measures capture different aspects of atherosclerosis.
Cardiovascular disease (CVD) remains the number one cause of death in women.
[3]
It is widely
believed that outcome data from intervention trials in men are generalizable to women, providing
a framework for primary prevention of CVD in women. Over the past decade, a number of
studies have refuted this assumption of generalizability; these studies have concerned the sex-
specific efficacy of CVD primary prevention therapies and the age-related timing of initiation of
menopausal hormone therapy (HT) as modifiers of risk.
[4][5][6]
In this regard, the observation that
HT significantly reduces CVD and all-cause mortality in primary prevention when initiated in
women who are younger than age 60 years or are less than 10 years-since-menopause, compared
to older women and women distant from menopause, led to the formation of the menopause HT
timing hypothesis.
[7][8][9]
The Early Versus Late Intervention Trial with Estradiol (ELITE)
showed that hormone therapy (HT) administered within 6 years of menopause significantly
reduced subclinical atherosclerosis progression, measured by the change in carotid artery intima-
media thickness (CIMT), relative to placebo, whereas there was no effect in women who
received HT 10 or more years-since-menopause.
[10][11]
The ELITE study provided randomized
clinical trial evidence for a treatment-specific and age-related opportunity for reducing
2
subclinical atherosclerosis, related to cardiovascular disease and all-cause mortality trends in
women.
Given these ELITE results supporting an HT timing effect, we sought to study whether effects on
inflammation may in part, explain the benefit on atherosclerosis progression. This thesis
evaluated the associations of clinical biomarkers of inflammation including inflammatory
cytokines, chemokines and cell adhesion molecules associated with atherosclerosis (IL-1α, IL-
1β, IL-6, IL-8, IL-10, INF-γ, MCP-1, MIP-1α, TNF-α, sICAM-1, sVCAM-1, VEGF and E- and
P-selectins) on CIMT progression, the ELITE primary trial outcome. In addition, analyses
addressed associations with GSM, a relatively new marker of atherosclerosis progression.

Methods
Study design
The Early vs. Late Intervention Trial with Estradiol (ELITE;ClinicalTrials.gov NCT00114517)
was a single-center, randomized, double-blinded, placebo-controlled trial of hormone therapy.
[11]
ELITE aimed to test the hormone therapy (HT) timing hypothesis (i.e., whether the effects of HT
vary according to the timing of initiation in relation to menopause). Eligible participants were
healthy women in postmenopause, without clinical evidence of diabetes or cardiovascular
disease, with a serum E2 level lower than 25 pg/mL, and with cessation of menses for a
minimum of 6 months. Participants were randomly assigned to receive 1mg daily oral 17β-
estradiol or matching placebo in a 1:1 ratio. Randomization was stratified by baseline carotid-
artery intima–media thickness (CIMT) (<0.75 or ≥0.75 mm), hysterectomy status (yes or no),
and postmenopausal status (early, <6 year; late ≥10 years). After randomization, women attended
3
study clinic visits every month for the first 6 months and every other month in the follow-up
period until trial completion. The median duration of follow-up was 4.8 years (range, 0.5 to 6.7
years).
The ELITE primary trial outcome tested the HT effects compared to placebo on CIMT
progression; a statistical comparison of the HT effects in early versus late postmenopause strata
provided the test of the HT timing hypothesis. Briefly, HT reduced CIMT progression in early
but not late postmenopausal women.
In this post hoc trial analysis, the trial outcomes were CIMT and grey scale median (GSM). The
associations of inflammatory markers with the rates of progression of these trial outcomes were
evaluated. Independent variables were each of the circulating levels of 14 inflammatory
biomarkers (IL-1α, IL-1β, IL-6, IL-8, IL-10, INF-γ, MCP-1, MIP-1α, TNF-α, sICAM-1,
sVCAM-1, VEGF and E- and P-selectins) measured three times: at baseline prior to
randomization, 12 months and 36 months post-randomization. A prior analysis analyzed the
effects of HT on each of these inflammatory biomarkers; for context, these results will be
provided in the Results section of this thesis.

CIMT and GSM assessment
The rate of change in intima-media thickness of the far wall of the right distal common carotid
artery was assessed by computer image processing of B-mode ultrasonograms obtained at the
baseline examination and every 6 months during the follow-up period.
[12]
For the convenience of
analysis, the CIMT values were provided in mm units multiplied by 1000 for µm units. In the
same common carotid artery segment, the grey scale median (GSM) was determined from all
4
pixels within the intima-media complex on a scale from 0 (Black) to 255 (White) by automated
software.
[1]

Statistical analysis
There were 643 eligible participants in this study. Data imputation was done. There were 2
participant-visits at 12 months and 2 participant-visits at 36 months with inflammatory data but
without CIMT/GSM data. These CIMT/GSM values were imputed using the mean of on-trial
CIMT/GSM of the particular participant. By design, the levels of the inflammatory biomarkers
were measured only at baseline, 12 months and 36 months post-randomization; these markers
were therefore missing for CIMT/GSM visits at 6 months, 18 months, 24 months, 30 months, 42
months, 48 months, 54 months, 60 months, 66 months, and 72 months post-randomization.
These missing values were imputed with the mean of on-trial levels of each biomarker of the
particular participant. 3961 values for each biomarker were imputed. The on-trial mean was
used for imputation (rather than some time-dependent imputation), as analysis showed no post-
randomization time effects on the on-trial measures (i.e., the 12-month and 36-month means did
not differ). With the imputation of the inflammation measures, all of the CIMT/GSM data can
then be used to test the interaction of biomarker with time since randomization to test whether
the CIMT/GSM progression (association of CIMT/GSM with time) varied by levels of the
inflammation biomarker.
Mixed-effects linear models with restricted maximum likelihood (REML) method were used to
analyze per-participant CIMT/GSM progression. Levels of each biomarker were separately
included as a main independent variable along with time (years) since randomization. The
interaction of biomarker with time since randomization was included in the model to specifically
5
test if the levels of inflammation were associated with CIMT/GSM rates of progression.
Hysterectomy stratum and time--since-menopause stratum were included as covariates. Random
effects were specified for participant-specific intercept (baseline CIMT/GSM) and slope (rate of
change in CIMT/GSM). Analysis was performed for the total sample, separately by two strata
(early/late menopause), and separately by four strata (early/late menopause and HT versus
placebo randomized treatment).
A 2-sided significance level was chosen as 0.05. Statistical analysis was performed by Statistical
Analysis System software version 9.4 (SAS institute, Inc., Cary, North Carolina).

Results
1. Hormone Therapy effects on the inflammation measures
Analysis of the HT effects on the inflammation measures was conducted elsewhere (not as part
of this thesis). The results showed that in the total sample, average on-trial levels of E-selectin,
sICAM-1, IFNγ and IL-8 were significantly lower in the HT group compared with placebo-
treated women (all p-values <0.035). Stratified by time-since-menopause, women within 6 years
of menopause when randomized to HT showed significantly lower levels of E-selectin, sICAM-1
and IL-8 compared with placebo; only E-selectin was significantly lower among women
randomized to HT 10 or more years since menopause compared with placebo. （Table 1A-1C ）

6
Table 1A: Effect of hormone therapy on biomarkers of inflammation in postmenopausal women
All women (n = 535)
Early (n = 227)

Late (n = 308)

Cytokine
Difference (SE)
‡
p-value
†

Difference (SE) p-value Difference (SE) p-value
E-Selectin, ng/mL -0.131 (0.035) 0.0002
-0.135 (0.054) 0.01 -0.126 (0.046) 0.006
P-Selectin, ng/mL 0.009 (0.013) 0.52
0.021(0.021) 0.32 0.005 (0.039) 0.89
sICAM-1, ng/mL -0.037 (0.017) 0.03
-0.07 (0.026) 0.008 -0.012 (0.022) 0.57
sVCAM-1, ng/mL -0.004 (0.020) 0.86
-0.031(0.033) 0.35 0.018 (0.025) 0.48
MIP -1α, pg/mL -0.095 (0.087) 0.28
-0.04 (0.023) 0.08 -0.0005 (0.119) 0.99
IFNγ, pg/mL -0.144 (0.066) 0.03
-0.17 (0.103) 0.10 -0.123 (0.087) 0.16
MCP-1, pg/mL -0.002 (0.027) 0.94
0.066 (0.041) 0.11 -0.047 (0.036) 0.19
TNF-α, pg/mL -0.027 (0.024) 0.26
-0.037 (0.039) 0.35 -0.018 (0.03) 0.55
VEGF-A, pg/mL 0.042 (0.054) 0.43
0.028 (0.078) 0.72 0.057 (0.073) 0.43
IL-6, pg/mL -0.110 (0.060) 0.067
-0.145 (0.099) 0.14 -0.081 (0.075) 0.28
IL-8, pg/mL -0.249 (0.116) 0.03
-0.487 (0.169) 0.004 -0.061 (0.157) 0.69
IL-10, pg/mL -0.084 (0.117) 0.47
-0.306 (0.175) 0.08 0.083 (0.156) 0.59
* Treatment group comparisons on log-transformed biomarkers.
† P-values for (log) mean comparisons between the treatment groups.
‡ Differences are the (log) mean circulating level in the estradiol group minus that in the placebo group.
§ Baseline value adjustment was performed for P-selectin.

Table 1B: Effect of Estradiol (Early Postmenopausal).
Cytokine OR p-value
IL-1α, pg/mL 0.1747 (0.02785, 1.0963) 0.0625
IL-1β, pg/mL 0.2338 (0.03682, 1.4852) 0.1228
N=227

Table 1C: Effect of Estradiol (Late Postmenopausal).
Cytokine OR p-value
IL-1α, pg/mL 2.2094 (0.4220, 11.5660) 0.3467
IL-1β, pg/mL 1.6125 (0.2860, 9.0923) 0.5871
N=308

7
2. Association between CIMT and each inflammatory marker in the total sample
The tables below provide the mixed effects model estimates for each inflammatory marker in
relation to CIMT. Each marker was modeled separately. Each table provides results for the
analysis of the total sample, separately by early/late menopause, and separately by early/late
menopause and hormone therapy vs placebo treatment randomization. In each analysis, the
main effect of the marker tested the association with CIMT at time 0 (i.e., baseline CIMT). The
interaction of the marker with time (years since randomization) tested the association of marker
with the annual rate of progression of CIMT. In the total sample analysis, the baseline effect and
marker-time interaction tested the association only in the referent late menopause group. The
two-way interaction of marker-by-menopause category and the three-way interaction of marker-
by-menopause category-by time tested if the marker association with baseline CIMT and CIMT
progression differed in early vs late postmenopause. In the total sample analysis, the main effect
of menopause category tested whether baseline CIMT differed in early vs. late postmenopause;
the interaction of menopause category by years tested if the annual rate of CIMT progression
differed by menopause category. The interaction of marker by menopause category tested if the
baseline association of the marker with CIMT differed in early vs. late postmenopause.
In the total sample analysis, when the 3-way interaction of marker-by-menopause-by-time was
not significant (i.e., marker-years association did not differ by menopause group), the model for
the total sample analysis was rerun without the 3-way interaction term.

8
IL-1 ⍺ (Table 2)
In the total sample at baseline, a negative association of IL-1⍺ with CIMT was not statistically
significant (p=0.84). A positive association of IL-1⍺ with CIMT progression was not
statistically significant (p=0.80). The IL-1 ⍺ association with baseline CIMT (p=0.60) and CIMT
progression (p=0.69) did not significantly differ by menopause category. When analyzed by
early/late menopause separately, a positive association of IL-1⍺ with baseline CIMT (p=0.56)
and a negative association of IL-1⍺ with CIMT progression (p=0.72) in the early menopause
group were not statistically significant. In the late menopause group, a negative association of
IL-1 ⍺ with baseline CIMT (p=0.78) and a positive association of IL-1 ⍺ with CIMT progression
(p=0.71) were not statistically significant.
When analyzed by four strata of early/late postmenopause and hormone therapy/placebo, all
strata showed non-significant associations of IL-1⍺ with baseline CIMT and non-significant
associations of IL-1⍺ with CIMT progression.

9

Table 2 Mixed effects model estimates for IL-1⍺ in relation to CIMT
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline CIMT
Main effect of
Marker
Beta -0.005 0.014 -0.007 0.021 -0.005 -0.048 0.018
SE 0.024 0.024 0.024 0.028 0.051 0.044 0.030
p-value 0.84 0.56 0.78 0.44 0.92 0.27 0.55
Interaction
with early vs
late
Postmenopaus
e
Beta 0.018

SE 0.034
p-value 0.60
Marker
Association
no interaction
Beta -0.001
SE 0.022
p-value 0.96
CIMT progression
Main effect of
Marker
Beta 0.002 -0.003 0.003 -0.007 0.007 0.007 -0.003
SE 0.009 0.009 0.008 0.011 0.019 0.012 0.011
p-value 0.80 0.72 0.71 0.54 0.71 0.59 0.80
Interaction
with early vs
late
Postmenopaus
e
Beta -0.005

SE 0.013
p-value 0.69
Marker
Association
no interaction
Beta
-
0.0001
SE 0.006
p-value 0.98
10
MIP-1⍺ (Table 3)
In the total sample, a negative association of MIP-1⍺ with CIMT baseline (p=0.73) and a
positive association of MIP-1⍺ with CIMT progression (p=0.87) were not statistically
significant. The MIP-1⍺ association with baseline CIMT (p=0.86) and CIMT progression
(p=0.56) did not significantly differ by menopause category. When analyzed by early/late
menopause separately, the negative association of MIP-1⍺ with baseline CIMT was non-
significant in both early (p=0.64) and late (p=0.74) menopause. The positive association of MIP-
1⍺ with CIMT progression was non-significant in both early (p=0.42) and late (p=0.88)
menopause.
When analyzed by four strata of early/late postmenopause and hormone therapy/placebo, all
strata showed non-significant associations of MIP-1⍺ with baseline CIMT and non-significant
associations of MIP-1⍺ with CIMT progression.

11

Table 3 Mixed effects model estimates for MIP-1⍺ in relation to CIMT
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline CIMT
Main effect of
Marker
Beta -0.001 -0.002 -0.001 -0.006 0.007 -0.004 0.001
SE 0.003 0.004 0.003 0.005 0.007 0.005 0.004
p-value 0.73 0.64 0.74 0.21 0.30 0.46 0.87
Interaction
with early vs
late
Postmenopaus
e
Beta -0.001

SE 0.005
p-value 0.86
Marker
Association
no interaction
Beta -0.002
SE 0.003
p-value 0.53
CIMT progression
Main effect of
Marker
Beta 0.0002 0.001 0.0002 0.003 -0.002 0.0004 0.00002
SE 0.001 0.002 0.001 0.002 0.003 0.002 0.002
p-value 0.87 0.42 0.88 0.12 0.41 0.83 0.99
Interaction
with early vs
late
Postmenopaus
e
Beta 0.001

SE 0.002
p-value 0.56
Marker
Association
no interaction
Beta 0.001
SE 0.001
p-value 0.50

12
IFN- γ (Table 4)
In the total sample, a negative association of IFN-γ with CIMT baseline (p=0.07) and a positive
association of IFN-γ with CIMT progression (p=0.34) were not statistically significant. The IFN-
γ association with baseline CIMT significantly differed by menopause category (p=0.028).
However, the IFN-γ association with CIMT progression (p=0.56) did not significantly differ by
menopause category. When analyzed by early/late menopause separately, a positive association
of IFN-γ with baseline CIMT (p=0.13) in the early menopause group and a negative association
of IFN-γ with baseline CIMT (p=0.08) in the late menopause group were not statistically
significant. A negative association of IFN-γ with CIMT progression in the early menopause
group (p=0.08) and a positive association of IFN-γ with CIMT progression (p=0.34) in the late
menopause group were not statistically significant.
When analyzed by four strata of early/late postmenopause and hormone therapy/placebo, IFN-γ
was positively associated with baseline CIMT (p=0.003) and inversely associated with CIMT
progression (p=0.011) in the early menopause/placebo group. All other strata showed non-
significant associations of IFN-γ with baseline CIMT and non-significant associations of IFN
with CIMT progression.

13
Table 4 Mixed effects model estimates for IFN-γ in relation to CIMT
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline CIMT
Main effect of
Marker
Beta -0.330 0.105 -0.332 0.058 0.740 -0.368 -0.247
SE 0.185 0.069 0.188 0.075 0.247 0.228 0.363
p-value 0.07 0.13 0.08 0.44 0.003 0.11 0.50
Interaction
with early vs
late
Postmenopau
se
Beta 0.436

SE 0.198
p-value 0.028
Marker
Association
no interaction
Beta -0.161
SE 0.151
p-value 0.28
CIMT progression
Main effect of
Marker
Beta 0.105 -0.084 0.106 -0.063 -0.276 0.177 0.014
SE 0.110 0.048 0.110 0.057 0.109 0.152 0.174
p-value 0.34 0.08 0.34 0.27 0.011 0.24 0.93
Interaction
with early vs
late
Postmenopau
se
Beta -0.189

SE 0.120
p-value 0.12
Marker
Association
no interaction
Beta -0.052
SE 0.045
p-value 0.25

14
sICAM-1 (Table 5)
In the total sample, a negative association of sICAM-1 with baseline CIMT (p=0.22) and a
negative association of sICAM-1 with CIMT progression (p=0.82) were not statistically
significant. The sICAM-1 association with baseline CIMT (p=0.45) and CIMT progression
(p=0.74) did not significantly differ by menopause category. When analyzed by early/late
menopause separately, all strata showed non-significant negative associations of sICAM-1 with
baseline CIMT and negative associations of sICAM-1 with CIMT progression.
When analyzed by four strata of early/late postmenopause and hormone therapy/placebo,
sICAM-1 was positively associated with baseline CIMT (p=0.040) and inversely associated with
CIMT progression (p=0.035) in the early menopause/placebo group. All other strata showed non-
significant associations of sICAM-1 with baseline CIMT and non-significant associations of
sICAM-1 with CIMT progression.

15
Table 5 Mixed effects model estimates for sICAM-1 in relation to CIMT
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline CIMT
Main effect of
Marker
Beta -0.012 -0.001 -0.010 -0.026 0.033 -0.021 0.007
SE 0.010 0.011 0.010 0.015 0.016 0.013 0.014
p-value 0.22 0.91 0.29 0.08 0.040 0.10 0.61
Interaction
with early vs
late
Postmenopaus
e
Beta 0.011

SE 0.015
p-value 0.45
Marker
Association
no interaction
Beta -0.010
SE 0.008
p-value 0.23
CIMT progression
Main effect of
Marker
Beta -0.001 -0.003 -0.002 0.006 -0.015 -0.001 -0.006
SE 0.005 0.005 0.005 0.007 0.007 0.007 0.007
p-value 0.82 0.54 0.63 0.38 0.035 0.84 0.39
Interaction
with early vs
late
Postmenopaus
e
Beta -0.002

SE 0.007
p-value 0.74
Marker
Association
no interaction
Beta -0.002
SE 0.003
p-value 0.51

16
sVCAM-1 (Table 6)
In the total sample, a negative association of sVCAM-1 with baseline CIMT (p=0.69) and a
negative association of sVCAM-1 with CIMT progression (p=0.25) were not statistically
significant. The sVCAM-1 association with baseline CIMT (p=0.60) and CIMT progression
(p=0.72) did not significantly differ by menopause category. When analyzed by early/late
menopause separately, all strata showed non-significant negative associations of sVCAM-1 with
baseline CIMT and non-significant negative associations of sCAM-1 with CIMT progression.
When analyzed by four strata of early/late postmenopause and hormone therapy/placebo,
sVCAM-1 was inversely associated with baseline CIMT in early menopause/HT group
(p=0.002). sVCAM-1 was inversely associated with CIMT progression in late menopause/HT
group (p=0.018). All other strata showed non-significant associations of sVCAM-1 with baseline
CIMT and non-significant associations of sVCAM-1 with CIMT progression.

17
Table 6 Mixed effects model estimates for sVCAM-1 in relation to CIMT
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline CIMT
Main effect of
Marker
Beta -0.002 -0.005 -0.001 -0.020 0.013 0.001 -0.003
SE 0.004 0.005 0.004 0.006 0.007 0.006 0.006
p-value 0.69 0.25 0.73 0.002 0.06 0.89 0.61
Interaction
with early vs
late
Postmenopaus
e
Beta -0.003

SE 0.006
p-value 0.60
Marker
Association
no interaction
Beta -0.002
SE 0.004
p-value 0.53
CIMT progression
Main effect of
Marker
Beta -0.002 -0.001 -0.002 0.004 -0.007 -0.006 0.002
SE 0.002 0.002 0.002 0.003 0.003 0.003 0.003
p-value 0.25 0.67 0.22 0.19 0.06 0.018 0.56
Interaction
with early vs
late
Postmenopaus
e
Beta 0.001

SE 0.003
p-value 0.72
Marker
Association
no interaction
Beta -0.002
SE 0.001
p-value 0.22

18
IL-6 (Table 7)
In the total sample, a negative association of IL-6 with baseline CIMT (p=0.67) and a positive
association of IL-6 with CIMT progression (p=0.28) were not statistically significant. The IL-6
association with baseline CIMT (p=0.53) and CIMT progression (p=0.33) did not significantly
differ by menopause category. When analyzed by early/late menopause separately, in the early
menopause group, the positive association of IL-6 with baseline CIMT (p=0.48) and the negative
association of IL-6 with CIMT progression (p=0.11) were non-significant. In the late menopause
group, IL-6 had non-significant negative associations with both baseline CIMT (p=0.65) and
CIMT progression (p=0.27). When analyzed by four strata of early/late postmenopause and
hormone therapy/placebo, in the late menopause/HT group, IL-6 was inversely associated with
baseline CIMT (p=0.018). All other strata showed non-significant associations of IL-6 with
baseline CIMT and non-significant associations of IL-6 with CIMT progression.

19
Table 7 Mixed effects model estimates for IL-6 in relation to CIMT
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline CIMT
Main effect of
Marker
Beta -0.018 0.088 -0.019 0.171 0.062 -0.921 -0.007
SE 0.042 0.126 0.042 0.265 0.140 0.389 0.041
p-value 0.67 0.48 0.65 0.52 0.66 0.018 0.86
Interaction
with early vs
late
Postmenopaus
e
Beta 0.084

SE 0.134
p-value 0.53
Marker
Association
no interaction
Beta -0.014
SE 0.042
p-value 0.74
CIMT progression
Main effect of
Marker
Beta 0.735 -0.090 -0.016 -0.158 -0.030 0.267 -0.017
SE 0.015 0.057 0.015 0.097 0.075 0.163 0.015
p-value 0.28 0.11 0.27 0.10 0.69 0.10 0.24
Interaction
with early vs
late
Postmenopaus
e
Beta -0.058

SE 0.059
p-value 0.33
Marker
Association
no interaction
Beta -0.021
SE 0.015
p-value 0.16

20
IL-8 (Table 8)
In the total sample, a positive association of IL-8 with baseline CIMT (p=0.23) and a negative
association of IL-8 with CIMT progression (p=0.33) were not statistically significant. The IL-8
association with baseline CIMT (p=0.06) did not significantly differ by menopause category.
However, the IL-8 association with CIMT progression significantly differed by menopause
category (p=0.047). When analyzed by early/late menopause separately, IL-8 had a negative
non-significant association with baseline CIMT(p=0.14) in early menopause group and a positive
non-significant association with baseline CIMT (p=0.23) in late menopause group. IL-8 had a
positive non-significant association with CIMT progression (p=0.08) in early menopause group
and negative non-significant association CIMT progression (p=0.34) in late menopause group.
When analyzed by four strata of early/late postmenopause and hormone therapy/placebo, in early
menopause/HT group, IL-8 was inversely associated with baseline CIMT (p<0.001) and
positively associated with CIMT progression (p=0.006). In early menopause/placebo group, IL-8
was positively associated with baseline CIMT (p=0.003). All other strata showed non-
significant associations of IL-8 with baseline CIMT and non-significant associations of IL-8 with
CIMT progression.

21
Table 8 Mixed effects model estimates for IL-8 in relation to CIMT
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline CIMT
Main effect of
Marker
Beta 0.005 -0.008 0.005 -0.028 0.023 0.010 0.0003
SE 0.004 0.005 0.004 0.007 0.008 0.006 0.005
p-value 0.23 0.14 0.23 <0.001 0.003 0.11 0.95
Interaction
with early vs
late
Postmenopaus
e
Beta -0.012

SE 0.007
p-value 0.06
Marker
Association
no interaction
Beta

SE
p-value
CIMT progression
Main effect of
Marker
Beta -0.002 0.004 -0.002 0.009 -0.001 -0.004 0.0004
SE 0.002 0.002 0.002 0.003 0.004 0.003 0.002
p-value 0.33 0.08 0.34 0.006 0.74 0.13 0.86
Interaction
with early vs
late
Postmenopaus
e
Beta 0.006

SE 0.003
p-value 0.047
Marker
Association
no interaction
Beta

SE
p-value

22
IL-10 (Table 9)
In the total sample, IL-10 had a non-significant positive association with both baseline CIMT
and CIMT progression. The IL-10 association with baseline CIMT (p=0.96) and with CIMT
progression (p=0.80) did not significantly differ by menopause category. When analyzed by
early/late menopause separately, IL-10 had a positive non-significant association with baseline
CIMT (p=0.95) in early menopause group and negative non-significant association with baseline
CIMT (p=0.88) in late menopause group. IL-10 had a positive non-significant association with
CIMT progression in both early and late menopause groups. When analyzed by four strata of
early/late postmenopause and hormone therapy/placebo, all strata showed non-significant
associations of IL-10 with baseline CIMT and non-significant associations of IL-10 with CIMT
progression.

23

Table 9 Mixed effects model estimates for IL-10 in relation to CIMT
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline CIMT
Main effect of
Marker
Beta 0.0002 0.0004 -0.0009 0.011 -0.002 -0.005 0.002
SE 0.0057 0.0064 0.0060 0.016 0.007 0.011 0.007
p-value 0.973 0.945 0.882 0.51 0.82 0.66 0.79
Interaction
with early vs
late
Postmenopau
se
Beta 0.0005

SE 0.0084
p-value 0.956
Marker
Association
no interaction
Beta 0.0001
SE 0.0053
p-value 0.979
CIMT progression
Main effect of
Marker
Beta 0.0050 0.0040 0.0055 0.00002 0.005 0.009 0.002
SE 0.0029 0.0032 0.0030 0.006 0.004 0.004 0.004
p-value 0.086 0.210 0.065 0.997 0.19 0.052 0.65
Interaction
with early vs
late
Postmenopau
se
Beta
-
0.0011

SE 0.0042
p-value 0.798
Marker
Association
no interaction
Beta 0.0048
SE 0.0022
p-value 0.0299

24
MCP-1 (Table 10)
In the total sample, a negative association of MCP-1 with baseline CIMT (p=0.18) and a positive
association of MCP-1 with CIMT progression (p=0.57) were not statistically significant. The
MCP-1 association with baseline CIMT significantly differed by menopause category (p=0.039).
However, the MCP-1 association with CIMT progression did not significantly differ by
menopause category (p=0.08). When analyzed by early/late menopause separately, MCP-1 had a
positive non-significant association with baseline CIMT (p=0.11) in early menopause group and
negative non-significant association with baseline CIMT (p=0.22) in late menopause group.
MCP-1 had a negative non-significant association with CIMT progression (p=0.053) in early
menopause group and positive non-significant association CIMT progression (p=0.72) in late
menopause group. When analyzed by four strata of early/late postmenopause and hormone
therapy/placebo, in early menopause/placebo group, MCP-1 was positively associated with
baseline CIMT (p=0.001) and inversely associated with CIMT progression (p=0.004). In late
menopause/HT group, MCP-1 was inversely associated with baseline CIMT (p=0.028). All other
strata showed non-significant associations of MCP-1 with baseline CIMT and non-significant
associations of MCP-1 with CIMT progression.

25

Table 10 Mixed effects model estimates for MCP-1 in relation to CIMT
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline CIMT
Main effect of
Marker
Beta -0.006 0.008 -0.006 -0.012 0.021 -0.016 0.004
SE 0.005 0.005 0.005 0.008 0.006 0.007 0.006
p-value 0.18 0.11 0.22 0.13 0.001 0.028 0.57
Interaction
with early vs
late
Postmenopau
se
Beta 0.014

SE 0.007
p-value 0.039
Marker
Association
no interaction
Beta -0.002
SE 0.004
p-value 0.56
CIMT progression
Main effect of
Marker
Beta 0.001 -0.005 0.001 0.001 -0.009 0.003 -0.001
SE 0.002 0.002 0.002 0.004 0.003 0.003 0.003
p-value 0.57 0.053 0.72 0.73 0.004 0.45 0.86
Interaction
with early vs
late
Postmenopau
se
Beta -0.006

SE 0.003
p-value 0.08
Marker
Association
no interaction
Beta -0.002
SE 0.002
p-value 0.35

26
P-selectin (Table 11)
In the total sample, a positive association of P-selectin with baseline CIMT was not statistically
significant (p=0.60). A negative association of P-selectin with CIMT progression was not
statistically significant (p=0.96). The P-selectin association with baseline CIMT (p=0.91) and
CIMT progression (p=0.29) did not significantly differ by menopause category. When analyzed
by early/late menopause separately, P-selectin had a non-significant positive association with
baseline CIMT in both early and late menopause groups and had a non-significant negative
association with CIMT progression in both early and late menopause groups. When analyzed by
four strata of early/late postmenopause and hormone therapy/placebo, all strata showed non-
significant associations of P-selectin with baseline CIMT and non-significant associations of P-
selectin with CIMT progression.

27
Table 11 Mixed effects model estimates for P-selectin in relation to CIMT
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline CIMT
Main effect of
Marker
Beta 0.015 0.007 0.016 -0.027 0.048 0.013 0.024
SE 0.028 0.034 0.028 0.043 0.054 0.037 0.044
p-value 0.60 0.84 0.57 0.54 0.37 0.72 0.59
Interaction
with early vs
late
Postmenopau
se
Beta -0.005

SE 0.044
p-value 0.91
Marker
Association
no interaction
Beta 0.027
SE 0.025
p-value 0.29
CIMT progression
Main effect of
Marker
Beta -0.001 -0.020 -0.002 -0.007 -0.027 -0.007 0.001
SE 0.013 0.016 0.013 0.020 0.025 0.017 0.020
p-value 0.96 0.20 0.90 0.73 0.28 0.69 0.97
Interaction
with early vs
late
Postmenopau
se
Beta -0.021

SE 0.020
p-value 0.29
Marker
Association
no interaction
Beta -0.009
SE 0.010
p-value 0.33

28
IL- 1β (Table 12)
In the total sample, a positive association of IL-1β with baseline CIMT (p=0.35) and with CIMT
progression (p=0.97) were not statistically significant. The IL-1β association with baseline CIMT
(p=0.55) and CIMT progression (p=0.71) did not significantly differ by menopause category.
When analyzed by early/late menopause separately, IL-1β had a non-significant positive
association with baseline CIMT in both early and late menopause groups. IL-1β had non-
significant negative association with CIMT progression in early menopause group (p=0.52) and
non-significant positive association with CIMT progression in late menopause group (p=0.96).
When analyzed by four strata of early/late menopause and hormone therapy/placebo, IL-1β was
inversely associated with CIMT progression in early menopause/HT group (p=0.031) .
All other strata showed non-significant associations of IL-1β with baseline CIMT and non-
significant associations of IL-1β with CIMT progression.

29
Table 12 Mixed effects model estimates for IL-1β in relation to CIMT
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline CIMT
Main effect of
Marker
Beta 0.006 0.001 0.005 0.014 -0.003 -0.0004 0.013
SE 0.006 0.006 0.006 0.014 0.007 0.008 0.008
p-value 0.35 0.84 0.36 0.30 0.66 0.96 0.13
Interaction
with early vs
late
Postmenopau
se
Beta -0.005

SE 0.009
p-value 0.55
Marker
Association
no interaction
Beta 0.007
SE 0.005
p-value 0.21
CIMT progression
Main effect of
Marker
Beta
0.0000
9
-0.002 0.0001 -0.013 0.002 0.002 -0.002
SE 0.003 0.003 0.003 0.006 0.003 0.004 0.004
p-value 0.97 0.52 0.96 0.031 0.52 0.59 0.62
Interaction
with early vs
late
Postmenopau
se
Beta -0.001

SE 0.004
p-value 0.71
Marker
Association
no interaction
Beta -0.001
SE 0.002
p-value 0.76

30
TNF- ⍺ (Table 13)
In the total sample, a negative association of TNF- ⍺ with baseline CIMT (p=0.81) and with
CIMT progression (p=0.86) were not statistically significant. The TNF- ⍺ association with
baseline CIMT (p=0.25) and CIMT progression (p=0.73) did not significantly differ by
menopause category. When analyzed by early/late menopause separately, TNF- ⍺ had non-
significant positive association with baseline CIMT in early menopause group (p=0.17) and
negative association with baseline CIMT in late menopause group (p=0.86). TNF- ⍺ had non-
significant negative associations with CIMT progression in both early and late menopause
groups. When analyzed by four strata of early/late postmenopause and hormone therapy/placebo,
TNF- ⍺ was positively associated with baseline CIMT in early menopause/placebo group
(p<0.001). All other strata showed non-significant associations of TNF- ⍺ with baseline CIMT
and non-significant associations of TNF- ⍺ with CIMT progression.

31
Table 13 Mixed effects model estimates for TNF- ⍺ in relation to CIMT
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline CIMT
Main effect of
Marker
Beta -0.036 0.206 -0.027 -0.083 1.217 -0.108 -0.002
SE 0.153 0.151 0.155 0.176 0.316 0.296 0.178
p-value 0.81 0.17 0.86 0.64 <0.001 0.72 0.99
Interaction
with early vs
late
Postmenopau
se
Beta 0.252

SE 0.218
p-value 0.25
Marker
Association
no interaction
Beta -0.006
SE 0.125
p-value 0.96
CIMT progression
Main effect of
Marker
Beta -0.016 -0.050 -0.025 0.049 -0.331 -0.081 0.024
SE 0.094 0.064 0.095 0.072 0.201 0.157 0.118
p-value 0.86 0.43 0.79 0.49 0.10 0.61 0.84
Interaction
with early vs
late
Postmenopau
se
Beta -0.039

SE 0.114
p-value 0.73
Marker
Association
no interaction
Beta -0.043
SE 0.054
p-value 0.43

32
E-selectin (Table 14)
In the total sample, E-selectin was inversely associated with baseline CIMT (p=0.00003).
However, E-selectin had a non-significant association with CIMT progression (p=0.38). The E-
selectin association with baseline CIMT (p=0.36) and with CIMT progression (p=0.68) did not
significantly differ by menopause category. When analyzed by early/late menopause separately,
E-selectin was inversely associated with baseline CIMT in both early (p=0.014) and late
(p=0.00005) menopause group. However, a positive association of E-selectin with CIMT
progression in early menopause group (p=0.78) and in late menopause group (p=0.52) were not
statistically significant. When analyzed by four strata of early/late postmenopause and hormone
therapy/placebo, E-selectin was inversely associated with baseline CIMT in early menopause/HT
group (p<0.001), and in the late menopause/HT group (p<0.001). However, E-selectin had non-
significant associations with CIMT progression in all strata.

33

Table 14 Mixed effects model estimates for E-selectin in relation to CIMT
Effect
Paramete
r
Total
Sample
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline CIMT
Main effect of
Marker
Beta -0.328 -0.228 -0.318 -0.402 0.116 -0.551 0.116
SE 0.078 0.093 0.078 0.118 0.151 0.101 0.127
p-value
0.0000
3
0.014 0.00005 <0.001 0.44 <0.001 0.36
Interaction
with early vs
late
Postmenopaus
e
Beta 0.112

SE 0.122
p-value 0.36
Marker
Association
no interaction
Beta -0.315
SE 0.071
p-value
0.0000
1
CIMT progression
Main effect of
Marker
Beta 0.031 0.013 0.023 0.074 -0.105 0.073 -0.084
SE 0.035 0.046 0.035 0.061 0.069 0.047 0.053
p-value 0.38 0.78 0.52 0.22 0.13 0.12 0.11
Interaction
with early vs
late
Postmenopaus
e
Beta -0.024

SE 0.057
p-value 0.68
Marker
Association
no interaction
Beta 0.022
SE 0.028
p-value 0.43

34
VEGF (Table 15)
In the total sample, a negative association of VEGF with CIMT baseline (p=0.18) and a positive
association of VEGF with CIMT progression (p=0.95) were not statistically significant. The
VEGF association with baseline CIMT (p=0.25) and with CIMT progression (p=0.90) did not
significantly differ by menopause category. When analyzed by early/late menopause separately,
in early menopause group, VEGF had a positive non-significant association with both baseline
CIMT (p=0.61) and CIMT progression (p=0.92). However, in late menopause group, VEGF had
a negative non-significant association with both baseline CIMT (p=0.19) and CIMT progression
(p=0.99). When analyzed by four strata of early/late postmenopause and hormone
therapy/placebo, all strata showed non-significant associations of VEGF with baseline CIMT and
non-significant associations of VEGF with CIMT progression.

35
Table 15 Mixed effects model estimates for VEGF in relation to CIMT
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline CIMT
Main effect of
Marker
Beta -0.007 0.004 -0.007 0.0003 0.008 -0.005 -0.008
SE 0.006 0.007 0.006 0.009 0.011 0.009 0.007
p-value 0.18 0.61 0.19 0.98 0.49 0.56 0.24
Interaction
with early vs
late
Postmenopaus
e
Beta 0.010

SE 0.009
p-value 0.25
Marker
Association
no interaction
Beta -0.008
SE 0.005
p-value 0.13
CIMT progression
Main effect of
Marker
Beta 0.0001 0.0003 -0.00003 0.0002 0.002 -0.0002 -0.0003
SE 0.002 0.003 0.002 0.004 0.004 0.003 0.003
p-value 0.95 0.92 0.99 0.97 0.69 0.95 0.91
Interaction
with early vs
late
Postmenopaus
e
Beta 0.0005

SE 0.004
p-value 0.90
Marker
Association
no interaction
Beta 0.0003
SE 0.002
p-value 0.86
3. Association between GSM and each inflammatory marker in the total sample
The tables below provide the mixed effects model estimates for each inflammatory marker in
relation to GSM. Each marker was modeled separately. Each table provides results for the
analysis of the total sample, separately by early/late menopause, and separately by early/late
menopause and hormone therapy vs placebo treatment randomization. In each analysis, the
main effect of the marker tested the association with GSM at time 0 (i.e., baseline GSM). The
interaction of the marker with time (years since randomization) tested the association of
marker with the annual rate of progression of GSM. In the total sample analysis, the baseline
effect and marker-time interaction tested the association only in the referent late menopause
group. The two-way interaction of marker-by-menopause category and the three-way
interaction of marker-by-menopause category-by time tested if the marker association with
baseline GSM and GSM progression differed in early vs late postmenopause. In the total
sample analysis, the main effect of menopause category tested whether baseline GSM
differed in early vs. late postmenopause; the interaction of menopause category by years
tested if the annual rate of GSM progression differed by menopause category. The
interaction of marker by menopause category tested if the baseline association of the marker
with GSM differed in early vs. late postmenopause.
In the total sample analysis, when the 3-way interaction of marker-by-menopause-by-time
was not significant (i.e., marker-years association did not differ by menopause group), the
model for the total sample analysis was rerun without the 3-way interaction term.

37
E-selectin （Table 16 ）
In the total sample, E-selectin was inversely associated with baseline GSM (p<.001). However,
E-selectin had a non-significant association with GSM progression (p=0.61). E-selectin was
inversely associated with baseline GSM in all strata except early postmenopause/HT stratum,
whereas the association between E-selectin and GSM progression was non-significant in all
strata except the early postmenopause/placebo stratum (p=0.048).
38

Table 16 Mixed effects model estimates for E-Selectin in relation to GSM
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline GSM
Main effect of
Marker
Beta -0.167 -0.215 -0.165 -0.126 -0.326 -0.134 -0.213
SE 0.048 0.058 0.048 0.081 0.085 0.062 0.076
p-value <.001 <.001 0.001 0.12 0.0001 0.030 0.005
Interaction
with early vs
late
Postmenopaus
e
Beta -0.044

SE 0.075
p-value 0.56
Marker
Association
no interaction
Beta -0.163
SE 0.047
p-value 0.001
GSM progression
Main effect of
Marker
Beta 0.005 0.001 0.005 -0.022 0.034 0.007 0.007
SE 0.011 0.013 0.010 0.019 0.017 0.015 0.014
p-value 0.61 0.92 0.64 0.27 0.048 0.67 0.64
Interaction
with early vs
late
Postmenopaus
e
Beta -0.006

SE 0.017
p-value 0.69
Marker
Association
no interaction
Beta 0.003
SE 0.008
p-value 0.73
39
sICAM-1 （Table 17 ）
In the total sample, sICAM-1 was inversely associated with baseline GSM (p=0.012). However,
the association between sICAM-1 and GSM progression was non-significant. (p=0.25) sICAM-1
was significantly associated with baseline GSM in late postmenopause stratum (p=0.012), early
postmenopause/placebo treatment stratum (p=0.029), and late postmenopause/placebo treatment
stratum (p<0.001) . The association between E-selectin and GSM progression was non-
significant in all strata.

40
Table 17 Mixed effects model estimates for sICAM-1 in relation to GSM
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline GSM
Main effect of
Marker
Beta -0.017 -0.013 -0.017 -0.005 -0.022 -0.012 -0.022
SE 0.007 0.007 0.007 0.011 0.010 0.009 0.011
p-value 0.012 0.08 0.012 0.67 0.029 0.16 0.038
Interaction
with early vs
late
Postmenopau
se
Beta 0.004

SE 0.010
p-value 0.69
Marker
Association
no interaction
Beta -0.015
SE 0.007
p-value 0.021
GSM progression
Main effect of
Marker
Beta 0.002 0.000 0.002 -0.001 0.001 0.003 0.001
SE 0.002 0.002 0.002 0.003 0.002 0.002 0.003
p-value 0.25 0.80 0.25 0.84 0.54 0.24 0.78
Interaction
with early vs
late
Postmenopau
se
Beta -0.003

SE 0.002
p-value 0.28
Marker
Association
no interaction
Beta 0.001
SE 0.001
p-value 0.58

41
IL-8 （Table 18 ）
IL-8 was positively associated with baseline GSM (p=0.014) whereas the association between
IL-8 and GSM progression was negative and non-significant (p=0.42). IL-8 was significantly
associated with baseline GSM only in late postmenopause stratum (p=0.013). The associations of
IL-8 with GSM progression were non-significant among all strata.
Table 18 Mixed effects model estimates for IL-8 in relation to GSM
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline GSM
Main effect of
Marker
Beta 0.006 -0.002 0.006 -0.003 -0.0003 0.005 0.006
SE 0.002 0.003 0.002 0.004 0.005 0.003 0.003
p-value 0.014 0.59 0.013 0.53 0.95 0.10 0.08
Interaction
with early vs
late
Postmenopaus
e
Beta -0.007

SE 0.004
p-value 0.07
Marker
Association
no interaction
Beta 0.005
SE 0.002
p-value 0.019
GSM progression
Main effect of
Marker
Beta
-
0.0004
0.0000 0.0000
-
0.00003
0.002 -0.001 -0.0001
SE 0.001 0.001 0.000 0.001 0.001 0.001 0.001
p-value 0.42 0.56 0.37 0.98 0.15 0.42 0.82
Interaction
with early vs
late
Postmenopaus
e
Beta 0.001

SE 0.001
p-value 0.37
Marker
Association
no interaction
Beta
-
0.0001
SE 0.000
p-value 0.72
42
P-selectin （Table 19 ）
The associations of P-selectin with baseline GSM and GSM progression were both positive and
non-significant. However, the association of P-selectin with GSM progression in the model
without the interaction term of early versus late menopause was positive and significant
(p=0.028). Among all strata, the association of P-selectin with baseline GSM were non-
significant. However, P-selectin was positively associated with GSM progression only in early
postmenopause stratum (p=0.026).

43
Table 19 Mixed effects model estimates for P-selectin in relation to GSM
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline GSM
Main effect of
Marker
Beta 0.007 -0.003 0.007 0.022 -0.041 -0.005 0.029
SE 0.018 0.020 0.017 0.027 0.032 0.022 0.030
p-value 0.69 0.89 0.68 0.41 0.20 0.81 0.33
Interaction
with early vs
late
Postmenopau
se
Beta -0.009

SE 0.027
p-value 0.74
Marker
Association
no interaction
Beta 0.003
SE 0.017
p-value 0.88
GSM progression
Main effect of
Marker
Beta 0.003 0.009 0.004 0.007 0.004 0.008 -0.005
SE 0.004 0.004 0.004 0.006 0.006 0.005 0.006
p-value 0.39 0.026 0.34 0.19 0.53 0.11 0.36
Interaction
with early vs
late
Postmenopau
se
Beta 0.006

SE 0.005
p-value 0.30
Marker
Association
no interaction
Beta 0.006
SE 0.003
p-value 0.028

44
MCP-1（Table 20 ）
MCP-1 was positively associated with baseline GSM only in early menopause/ placebo stratum
(p=0.026). MCP-1 was not significantly associated with GSM progression in any analysis.
Table 20 Mixed effects model estimates for MCP-1 in relation to GSM
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline GSM
Main effect of
Marker
Beta -0.004 0.003 -0.004 -0.004 0.011 -0.006 -0.003
SE 0.003 0.004 0.003 0.005 0.005 0.004 0.005
p-value 0.21 0.35 0.21 0.47 0.026 0.20 0.53
Interaction
with early vs
late
Postmenopau
se
Beta 0.007

SE 0.005
p-value 0.12
Marker
Association
no interaction
Beta -0.004
SE 0.003
p-value 0.19
GSM progression
Main effect of
Marker
Beta 0.0002 0.0000 0.0000 0.0002 -0.0004 -0.0001 0.0005
SE 0.001 0.001 0.001 0.001 0.001 0.001 0.001
p-value 0.83 0.60 0.83 0.86 0.73 0.95 0.60
Interaction
with early vs
late
Postmenopau
se
Beta 0.0002

SE 0.001
p-value 0.87
Marker
Association
no interaction
Beta 0.0002
SE 0.001
p-value 0.66

45
Non-significant biomarkers
The remaining biomarkers (i.e. sVCAM-1, VEGF, IL-1 ⍺, IL-1β, MCP-1, IL-6, IL-10) showed
no significant associations with baseline GSM or GSM progression in the total sample or any
stratified analysis (Table 21-Table 29).

Table 21 Mixed effects model estimates for SVCAM-1 in relation to GSM
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline GSM
Main effect of
Marker
Beta 0.001 0.002 0.001 0.005 -0.002 -0.0002 0.003
SE 0.003 0.003 0.003 0.004 0.005 0.004 0.004
p-value 0.70 0.60 0.72 0.30 0.63 0.95 0.56
Interaction
with early vs
late
Postmenopau
se
Beta 0.0005

SE 0.004
p-value 0.91
Marker
Association
no interaction
Beta 0.001
SE 0.003
p-value 0.62
GSM progression
Main effect of
Marker
Beta 0.001 0.000 0.001 0.001 -0.001 0.001 -0.001
SE 0.001 0.001 0.001 0.001 0.001 0.001 0.001
p-value 0.40 0.83 0.43 0.59 0.58 0.16 0.52
Interaction
with early vs
late
Postmenopau
se
Beta
-
0.0005

SE 0.001
p-value 0.65
Marker
Association
no interaction
Beta 0.0004
SE 0.000
p-value 0.46

46

Table 22 Mixed effects model estimates for VEGF in relation to GSM
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline GSM
Main effect of
Marker
Beta -0.003 0.001 -0.003 -0.0004 0.003 -0.004 -0.002
SE 0.003 0.003 0.003 0.005 0.005 0.004 0.004
p-value 0.22 0.72 0.23 0.93 0.58 0.31 0.56
Interaction
with early vs
late
Postmenopau
se
Beta 0.005

SE 0.004
p-value 0.28
Marker
Association
no interaction
Beta -0.003
SE 0.003
p-value 0.27
GSM progression
Main effect of
Marker
Beta
-
0.0002
-0.001 0.0000 -0.001 -0.001 0.001 -0.001
SE 0.001 0.001 0.001 0.001 0.001 0.001 0.001
p-value 0.73 0.27 0.74 0.65 0.21 0.12 0.07
Interaction
with early vs
late
Postmenopau
se
Beta -0.001

SE 0.001
p-value 0.46
Marker
Association
no interaction
Beta
-
0.0004
SE 0.000
p-value 0.30

47

Table 23 Mixed effects model estimates for IL-1⍺ in relation to GSM
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline GSM
Main effect of
Marker
Beta 0.001 0.0000 0.001 -0.002 0.006 -0.006 0.007
SE 0.010 0.015 0.010 0.020 0.024 0.015 0.014
p-value 0.92 0.99 0.91 0.93 0.79 0.70 0.63
Interaction
with early vs
late
Postmenopau
se
Beta 0.000

SE 0.018
p-value 0.98
Marker
Association
no interaction
Beta -0.001
SE 0.010
p-value 0.96
GSM progression
Main effect of
Marker
Beta -0.002 0.002 -0.002 0.001 0.003 -0.003 -0.001
SE 0.002 0.003 0.002 0.004 0.004 0.003 0.002
p-value 0.25 0.53 0.23 0.78 0.45 0.31 0.62
Interaction
with early vs
late
Postmenopau
se
Beta 0.004

SE 0.003
p-value 0.22
Marker
Association
no interaction
Beta -0.001
SE 0.002
p-value 0.57

48

Table 24 Mixed effects model estimates for IL-1β in relation to GSM
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline GSM
Main effect of
Marker
Beta -0.002 -0.002 -0.002 0.0004 -0.002 -0.002 -0.001
SE 0.004 0.004 0.004 0.008 0.005 0.005 0.006
p-value 0.67 0.65 0.68 0.96 0.66 0.75 0.88
Interaction
with early vs
late
Postmenopau
se
Beta
-
0.0002

SE 0.006
p-value 0.98
Marker
Association
no interaction
Beta -0.002
SE 0.004
p-value 0.55
GSM progression
Main effect of
Marker
Beta -0.001 0.0000 -0.001 0.002 -0.001 -0.001 -0.001
SE 0.001 0.001 0.001 0.002 0.001 0.001 0.001
p-value 0.46 0.94 0.40 0.15 0.34 0.47 0.66
Interaction
with early vs
late
Postmenopau
se
Beta 0.001

SE 0.001
p-value 0.48
Marker
Association
no interaction
Beta
-
0.0002
SE 0.001
p-value 0.72

49

Table 25 Mixed effects model estimates for IL-6 in relation to GSM
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline GSM
Main effect of
Marker
Beta -0.014 0.108 -0.014 0.147 0.112 -0.366 -0.013
SE 0.017 0.092 0.017 0.179 0.109 0.296 0.018
p-value 0.41 0.24 0.41 0.41 0.30 0.22 0.46
Interaction
with early vs
late
Postmenopaus
e
Beta 0.120

SE 0.092
p-value 0.19
Marker
Association
no interaction
Beta -0.013
SE 0.017
p-value 0.45
GSM progression
Main effect of
Marker
Beta 0.002 -0.027 0.002 -0.056 -0.012 -0.006 0.002
SE 0.003 0.020 0.003 0.045 0.020 0.094 0.003
p-value 0.55 0.18 0.55 0.22 0.56 0.95 0.47
Interaction
with early vs
late
Postmenopaus
e
Beta -0.030

SE 0.020
p-value 0.13
Marker
Association
no interaction
Beta 0.001
SE 0.003
p-value 0.71

50

Table 26 Mixed effects model estimates for IL-10 in relation to GSM
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline GSM
Main effect of
Marker
Beta 0.004 0.005 0.005 0.001 0.007 0.007 0.003
SE 0.004 0.005 0.004 0.012 0.006 0.006 0.005
p-value 0.28 0.38 0.26 0.96 0.22 0.30 0.55
Interaction
with early vs
late
Postmenopau
se
Beta 0.0005

SE 0.007
p-value 0.95
Marker
Association
no interaction
Beta 0.004
SE 0.004
p-value 0.30
GSM progression
Main effect of
Marker
Beta -0.001 0.0000 -0.001 -0.0004 0.0002 -0.001 -0.0004
SE 0.001 0.001 0.001 0.003 0.001 0.002 0.001
p-value 0.47 0.97 0.43 0.90 0.88 0.43 0.74
Interaction
with early vs
late
Postmenopau
se
Beta 0.001

SE 0.002
p-value 0.75
Marker
Association
no interaction
Beta -0.001
SE 0.001
p-value 0.50

51
Table 27 Mixed effects model estimates for IFN-γ in relation to GSM
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline GSM
Main effect of
Marker
Beta 0.012 -0.130 0.017 -0.107 -0.230 0.052 -0.043
SE 0.159 0.067 0.159 0.074 0.211 0.198 0.265
p-value 0.94 0.054 0.91 0.15 0.28 0.79 0.87
Interaction
with early vs
late
Postmenopau
se
Beta -0.142

SE 0.173
p-value 0.41
Marker
Association
no interaction
Beta -0.050
SE 0.155
p-value 0.75
GSM progression
Main effect of
Marker
Beta -0.054 0.029 -0.055 0.005 0.094 -0.099 -0.002
SE 0.045 0.028 0.043 0.033 0.061 0.066 0.056
p-value 0.22 0.30 0.20 0.88 0.13 0.14 0.97
Interaction
with early vs
late
Postmenopau
se
Beta 0.085

SE 0.052
p-value 0.10
Marker
Association
no interaction
Beta 0.009
SE 0.023
p-value 0.69

52

Table 28 Mixed effects model estimates for TNF- ⍺ in relation to GSM
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline GSM
Main effect of
Marker
Beta -0.081 -0.048 -0.079 -0.003 -0.468 -0.130 -0.071
SE 0.136 0.134 0.135 0.165 0.275 0.230 0.168
p-value 0.55 0.72 0.56 0.98 0.09 0.57 0.67
Interaction
with early vs
late
Postmenopau
se
Beta 0.043

SE 0.190
p-value 0.82
Marker
Association
no interaction
Beta -0.117
SE 0.130
p-value 0.37
GSM progression
Main effect of
Marker
Beta -0.024 0.037 -0.028 0.034 -0.042 -0.074 0.008
SE 0.047 0.044 0.045 0.056 0.076 0.075 0.056
p-value 0.61 0.41 0.53 0.54 0.58 0.32 0.88
Interaction
with early vs
late
Postmenopau
se
Beta 0.056

SE 0.064
p-value 0.38
Marker
Association
no interaction
Beta 0.006
SE 0.032
p-value 0.84

53

Table 29 Mixed effects model estimates for MIP-1⍺ in relation to GSM
Effect
Paramete
r
Total
Sampl
e
Early
Menopaus
e
Late
Menopaus
e
Early/H
T
Early/Placeb
o
Late/H
T
Late/Placeb
o
Baseline GSM
Main effect of
Marker
Beta 0.002 0.001 0.002 0.003 -0.003 0.004 0.001
SE 0.002 0.002 0.002 0.003 0.004 0.002 0.002
p-value 0.17 0.49 0.16 0.31 0.43 0.13 0.61
Interaction
with early vs
late
Postmenopaus
e
Beta -0.001

SE 0.003
p-value 0.80
Marker
Association
no interaction
Beta 0.002
SE 0.002
p-value 0.19
GSM progression
Main effect of
Marker
Beta
-
0.0002
0.0000 0.0000 -0.0003 0.001 -0.0004 0.00003
SE 0.000 0.000 0.000 0.0005 0.001 0.0005 0.0004
p-value 0.55 0.77 0.53 0.58 0.09 0.43 0.95
Interaction
with early vs
late
Postmenopaus
e
Beta 0.0003

SE 0.000
p-value 0.54
Marker
Association
no interaction
Beta
-
0.0001
SE 0.000
p-value 0.78

54
Discussion
The Early Versus Late Intervention Trial with Estradiol (ELITE) showed that HT reduced CIMT
progression in early but not late postmenopausal women. However, the biological mechanisms
underlying the atheroprotective effects of HT are not clear. The hypothesis that the effect of HT
on atherosclerosis progression may be in part, attributed to modulation of serum inflammatory
markers by HT has not been evaluated in large, long-term randomized controlled trials.
In a recent study in the same cohort of postmenopausal women in the current study, we showed
that the mean circulating levels of IL-8, sICAM-1, E-selectin and IFNγ were significantly
reduced in the HT group relative to the placebo group with the former two biomarkers
specifically reduced in the early postmenopause group, the women who showed a reduction of
atherosclerosis progression with HT. In this study, we found among a panel of 14 serum
inflammatory biomarkers, IL-8 was positively associated with CIMT progression in the early
postmenopause women randomized to HT (p=0.006 , Table 8). The other biomarkers shown to
be reduced with HT relative to placebo, sICAM-1, E-selectin and IFNγ failed to show this
association with atherosclerosis.
IL-8 is a potentially important cytokine in the development of atherosclerosis.
[13]
IL-8 is
produced by macrophages and is stored in Weibel-Palade bodies of endothelial cells.
[14][15]
Following arterial damage, endothelial cells and macrophages release IL-8 that induces a series
of physiological changes that primarily induce chemotaxis and recruitment of neutrophils to the
site of arterial damage. IL-8 has been shown to induce the migration and proliferation of
endothelial and smooth muscle cells.
[16][17]
Thus, the data suggest the possibility that the effects
of HT on CIMT progression may be in part, attributed to the modulation of IL-8 in the
55
inflammatory microenvironment of the damaged vascular wall reflected as a reduction of IL-8
circulating levels by HT.
GSM was used as another measure of atherosclerosis progression. GSM and CIMT progression
had different associations with a few inflammatory markers. GSM progression was positively
associated with E-selectin in the early postmenopause women who received placebo and was
positively associated with P-selectin in women in the early postmenopause stratum, whereas E-
selectin and P-selectin had no association with CIMT progression.
There are many strengths of this study, including the large sample size, randomized design and
long-term longitudinal follow-up with subclinical atherosclerosis measurements. By design, the
serum levels of the inflammatory biomarkers were determined only at baseline, 12 months and
36 months post-randomization; these biomarker levels were therefore not available for
association with the CIMT/GSM measurements at other 6-month visits from 6-month through
72-months post-randomization visits. To avoid losing CIMT/GSM information, 3961 values for
each biomarker were imputed. This technique could result in less precise associations between
CIMT/GSM and each biomarker.
In conclusion, based on our current methods and analysis, IL-8 may partially explain the effect
of HT on reducing the progression of subclinical atherosclerosis in early postmenopausal
women.

56
References
[1]. Jung M, Parrinello CM, Xue X, et al. Echolucency of the carotid artery intima-media complex
and intima-media thickness have different cardiovascular risk factor relationships: the
Women's Interagency HIV Study. J Am Heart Assoc. 2015;4(2):e001405.
[2]. Karim R, Xu W, Hodis HN, et al. Abstract MP09: Effect of Estradiol Therapy on Arterial Wall
Echomorphology in the Early versus Late Intervention Trial With Estradiol(elite).
Circulation :journal of the American Heart Association. 2020;141(Suppl_1).
doi:10.1161/circ.141.suppl_1.MP09
[3]. A Report From the American Heart Association. Heart Disease and Stroke Statistics—2015
Update. Circulation. 2015;131:447-454.
[4]. Petretta M, Costanzo P, Perrone-Filardi P, Chiariello M. Impact of gender in primary
prevention of coronary heart disease with stain therapy: a meta-analysis. Int J Cardiol
2010;138:25-31.
[5]. Berger JS, Roncaglioni MC, Avanzini F, Pangrazzi I, Tognoni G, Brown DL. Aspirin for the
primary prevention of cardiovascular events in women and men: a sex-specific meta-analysis
of randomized controlled trials. JAMA 2006;295:306-313.
[6]. Ridker PM, Cook NR, Lee IM, Gordon D, Gaziano JM, Manson JE, Hennekens CH, Buring
JE. A randomized trial of low-dose aspirin in the primary prevention of cardiovascular disease
in women. N Engl J Med 2005;352:1293-1304.
[7]. Hodis HN, Mack WJ. The timing hypothesis and hormone replacement therapy: a paradigm
shift in the primary prevention of coronary heart disease in women. Part 1: comparison of
therapeutic efficacy. J Am Geriatr Soc 2013;61:1005-1010.
[8]. Hodis HN, Mack WJ. The timing hypothesis and hormone replacement therapy: a paradigm
57
shift in the primary prevention of coronary heart disease in women. 2. Comparative risks. J
Am Geri- atr Soc 2013;61:1011-8.
[9]. Hodis HN, Mack WJ. A window of opportunity: the reduction of coronary heart disease and
total mortality with menopausal therapies is age and time dependent. Brain Res
2011;1379:244-252.
[10]. Hodis HN, Mack WJ, Azen SP, et al. Hormone therapy and the progression of coronary-
artery atherosclerosis in post- menopausal women. N Engl J Med 2003; 349:535-45.
[11]. Hodis HN, Mack WJ, Henderson VW, et al. Vascular Effects of Early versus Late
Postmenopausal Treatment with Estradiol. The New England Journal of Medicine.
2016;374(13):1221-1231.
[12]. 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 2015; 22: 391-401.
[13]. Stavros Apostolakis, Konstantina V ogiatzi, Virginia Amanatidou, Demetrios A. Spandidos,
Interleukin 8 and cardiovascular disease, Cardiovascular Research, V olume 84, Issue 3, 1
December 2009, Pages 353–360
[14]. Wolff B, Burns AR, Middleton J, Rot A (1998). "Endothelial cell "memory" of
inflammatory stimulation: human venular endothelial cells store interleukin 8 in Weibel-
Palade bodies". J. Exp. Med. 188 (9): 1757–62.
[15]. Utgaard JO, Jahnsen FL, Bakka A, Brandtzaeg P, Haraldsen G (1998). "Rapid secretion of
prestored interleukin 8 from Weibel-Palade bodies of microvascular endothelial cells". J. Exp.
Med. 188 (9): 1751-6.
[16]. Yue TL, Wang X, Sung CP, Olson B, McKenna PJ, Gu JL, Feuerstein GZ. Interleukin-8. A
58
mitogen and chemoattractant for vascular smooth muscle cells. Circ Res. 1994 Jul;75(1):1-7.
[17]. Aihua Li, Seema Dubey, Michelle L. Varney, Bhavana J. Dave, Rakesh K. Singh The
Journal of Immunology March 15, 2003, 170 (6) 3369-3376.

Abstract (if available)

Abstract Introduction: Atherosclerosis is characterized by changes in the structure and composition of blood vessels. Carotid artery intima-media thickness (CIMT) and grey scale median (GSM) measures capture different aspects of atherosclerosis. The Early Versus Late Intervention Trial with Estradiol (ELITE) showed that hormone therapy (HT) reduced CIMT progression in early but not late postmenopausal women. Given these ELITE results supporting the HT timing effect, we sought to study whether effects on inflammation may in part explain the benefit on atherosclerosis progression. This thesis evaluated the associations of inflammatory biomarkers with CIMT/GSM. ❧ Methods: ELITE was a single-center, randomized, double-blinded, placebo-controlled trial of HT. Participants were randomly assigned to receive HT or placebo treatment in a 1:1 ratio. Randomization was stratified by CIMT (<0.75 or ≥0.75 mm), hysterectomy status (yes or no), and timing of postmenopause status (early, <6 year

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Effect of estradiol on circulating levels of inflammatory cytokines in postmenopausal women

PDF

Association of arterial stiffness progression with subclinical atherosclerosis measurements in postmenopausal women

$Effect of menopausal hormone replacement therapy on lipoprotein particle fractions and association with carotid artery intima-media thickness and grey-scale median, independent measurements of su...$

PDF

Effect of 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

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

PDF

Relationship between progression of atherosclerosis and coagulation measures in a randomized-controlled trial

PDF

Bone mineral density is associated with carotid atherosclerosis in healthy postmenopausal women: a longitudinal analysis of randomized clinical trial data

PDF

The impact of hormone therapy on Alzheimer’s disease biomarkers in early and late postmenopausal women

PDF

Associations between inflammatory markers and change in cognitive endpoints

PDF

The effects of hormone therapy on carotid artery intima-media thickness and serum lipids by ApoE4 genotype

PDF

Subclinical carotid atherosclerosis, psychosocial measures, and cognitive function in middle- to older-aged adults

PDF

Subclinical atherosclerosis in overweight Latino youth: influence of cardiometabolic risk factors

PDF

Effects of post-menopausal hormone therapy on arterial stiffness in the ELITE trial

PDF

Hormone therapy timing hypothesis and atherosclerosis

PDF

A randomized controlled trial of hormone therapy on postmenopausal women’s quality of life

PDF

Sex hormones and atherosclerosis in postmenopausal women

PDF

Association of carotid artery stiffness with measures of cognition in older adults

PDF

Association of oophorectomy with progression of carotid atherosclerosis

PDF

Association of bone density and bone metabolism markers: the Women’s Isoflavone Soy Health (WISH) trial

PDF

The association of apolipoprotein E genotype with cholesterol levels and atherosclerosis: three randomized clinical trials

PDF

The association of asthma and asthma-related medications on subclinical atherosclerosis: a cross-sectional analysis of four randomized clinical trials

Asset Metadata

Creator Weng, Zhen (author)

Core Title The associations of inflammatory markers with progression of subclinical atherosclerosis in early and late postmenopausal women

School Keck School of Medicine

Degree Master of Science

Degree Program Biostatistics

Publication Date 12/01/2020

Defense Date 12/01/2020

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag atherosclerosis,carotid-artery intima–media thickness,grey scale median,inflammatory markers,OAI-PMH Harvest

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Mack, Wendy (committee chair), Hodis, Howard (committee member), Karim, Roksana (committee member)

Creator Email 543405512@qq.com,zhenweng@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-402370

Unique identifier UC11668681

Identifier etd-WengZhen-9175.pdf (filename),usctheses-c89-402370 (legacy record id)

Legacy Identifier etd-WengZhen-9175.pdf

Dmrecord 402370

Document Type Thesis

Rights Weng, Zhen

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