University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content The Impact of Hormone Therapy on Alzheimer’s Disease Biomarkers in Early and
Late Postmenopausal Women
by
Debbie Argueta Rufino
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
[APPLIED BIOSTATISTICS AND EPIDEMIOLOGY]
May 2025
Copyright [2025] [Debbie Argueta Rufino]

TABLE OF CONTENTS
List of Tables................................................................................................................................. iii
List of Figures................................................................................................................................ iv
Abstract...........................................................................................................................................v
Chapter 1: Introduction...................................................................................................................1
Alzheimer’s Disease...........................................................................................................1
AD Biomarkers...................................................................................................................2
AD Plasma Biomarkers...................................................................................................... 3
Conceptual Model for Biomarker Progression in AD.........................................................4
Early versus Late Intervention Trial with Estradiol (ELITE)............................................. 5
Chapter 2: Methods........................................................................................................................ 7
Study Design.......................................................................................................................7
Randomization and Treatment............................................................................................7
Follow-Up...........................................................................................................................8
Assessment of AD Biomarkers...........................................................................................8
Statistical Analysis............................................................................................................. 9
Chapter 3: Results......................................................................................................................... 11
Residual Analysis..............................................................................................................17
Biomarker Differences by Treatment Group.................................................................... 17
Biomarker Differences by Treatment Group and Postmenopausal Status........................20
Discussion.....................................................................................................................................23
References.....................................................................................................................................27
ii

List of Tables
Table 1: Demographic Characteristics of Participants at Baseline………………………………12
Table 2: Table 2: Mean (SD) of AD Biomarkers at Baseline, Visit 30 and Change by Treatment
Group
……………………………………………………………………………………………………19
Table 3: Mean (SD) of 30-month Change in AD Biomarkers ( Visit 30 – Baseline) by Hormone
Therapy in Early and Late Postmenopause Groups
……………………………………………………………………………………………………22
iii

List of Figures
Figure 1……………………………………………………………………………………………5
Figure 2: Distribution and Transformation of Variables…………………………………………13
Figure 3: Residual Analysis for AD Biomarkers……………………………………………….. 15
iv

Abstract
Alzheimer’s disease (AD) is a progressive neurodegenerative condition that primarily
affects older adults, with early pathological changes, including amyloid-beta (Aβ) accumulation
and tau tangles, occurring decades before symptoms. This study analyzed data from the Early
versus Late Intervention Trial with Estradiol (ELITE) to examine the effects of hormone therapy
(HT) on AD plasma biomarkers—Aβ40, Aβ42, pTau181, neurofilament light chain (NfL), and
glial fibrillary acidic protein (GFAP)—in postmenopausal women. The hormone-timing
hypothesis was tested by stratifying analyses based on time since menopause (early vs. late
postmenopause).
Using a randomized, double-blind, placebo-controlled design, baseline and 30-month
plasma biomarker levels were compared between HT and placebo groups. Significant reductions
in Aβ40 and pTau181 levels were observed in the HT group, with effects differing by
menopausal status. Aβ40 levels decreased more in the late postmenopause group, while pTau181
reductions were greater in the early group. No significant changes were detected for Aβ42/Aβ40,
NfL, or GFAP.
These findings support the timing hypothesis, suggesting that HT affects amyloid and tau
pathology differently based on menopause timing. This study highlights the potential of HT as an
early intervention to address preclinical AD pathology, although further research is needed to
confirm long-term effects.
v

Chapter 1:Introduction
Alzheimer’s Disease
Alzheimer’s disease (AD) is a neurodegenerative condition that primarily impacts older
adults, leading to a gradual decline in cognitive abilities and ultimately resulting in dementia. As
the most common cause of dementia worldwide, AD significantly affects the elderly population.
Currently, around 7 million Americans live with this disease, a figure expected to nearly double
to 13 million by 2050(1). Roughly 10% of people aged 65 and older have AD, while the
prevalence increases dramatically in those aged 80 and above, reaching approximately
30-40%(1).
AD progresses over many years, with changes in the brain beginning long before
noticeable symptoms appear. These early changes include the accumulation of toxic amyloid-β
(Aβ) proteins, formation of tau protein tangles, and brain cell damage possibly linked to
overactive microglia, which can release inflammatory molecules and neurotoxins(2). People with
these brain changes may be symptom-free initially or may experience subtle memory lapses that
gradually worsen, progressing to more profound memory loss and cognitive impairment. As the
disease advances, additional neuropsychiatric symptoms emerge, including confusion, mood
shifts, agitation, and, in later stages, delusions or hallucinations(2).
1

AD Biomarkers
AD pathology is most commonly assessed through neuroimaging and biomarker analysis.
PET imaging with tracers like PiB allows for the visualization of amyloid-beta (Aβ) plaques,
while tau tracers detect neurofibrillary tangles—both key pathological features of AD. Brain
MRI can identify structural changes. For example, hippocampal atrophy closely correlates with
AD disease progression (2).
Cerebrospinal fluid (CSF) biomarkers offer another critical avenue for studying AD.
Reduced CSF Aβ42 levels reflect greater amyloid plaque deposition in the brain, while the
Aβ42/Aβ40 ratio improves diagnostic precision by accounting for individual differences in
amyloid levels(2). Elevated CSF levels of t-tau and p-tau181 are markers of neuronal injury and
tau pathology, respectively, while increased neurofilament light chain (NfL) levels in CSF are
linked to axonal degeneration and disease severity(3). These biomarkers in CSF can be observed
years before clinical symptoms of AD appear, making them essential for early diagnosis and
tracking of disease progression.
Despite their importance, CSF biomarker collection via lumbar puncture is invasive and
associated with risks such as headaches and infection, limiting its feasibility for routine clinical
use. These challenges highlight the growing need for less invasive biomarkers, such as
plasma-based assays, to improve accessibility and facilitate long-term monitoring in both
research and clinical settings.
2

AD Plasma Biomarkers
Plasma biomarkers are increasingly recognized as a practical and non-invasive alternative
for detecting AD pathology. A study by Martínez-Dubarbie et al. evaluated plasma Aβ40, Aβ42,
and p-tau181 using the automated Lumipulse platform in a cohort of cognitively unimpaired
individuals. Results demonstrated significant correlations between plasma and CSF levels for
Aβ42 (r = 0.21, p = 0.002), the Aβ42/Aβ40 ratio (r = 0.6, p < 0.0001), p-tau181 (r = 0.47, p <
0.0001), and the p-tau181/Aβ42 ratio (r = 0.52, p < 0.0001), confirming the potential of plasma
biomarkers to reflect CSF changes. Among these, the plasma Aβ42/Aβ40 ratio showed the
strongest diagnostic performance, with an area under the curve (AUC) of 0.89 for predicting AD
pathology from CSF (4).
Plasma biomarkers also successfully differentiated participants based on amyloid (A) and
tau (T) status. Participants were classified as amyloid-positive (A+) or amyloid-negative (A−)
based on the presence or absence of abnormal amyloid pathology, as determined by biomarker
thresholds. Similarly, tau-positive (T+) participants exhibited elevated levels of tau pathology, as
determined by biomarker thresholds (4). This biologically-defined AD framework (A+T+)
reflects participants with both amyloid and tau abnormalities, indicative of AD pathology(5).
These findings underscore the promise of plasma biomarkers for large-scale AD
screening and early detection, particularly as they offer a less invasive and more accessible
alternative to traditional CSF-based or PET imaging approaches (6, 7)
3

Conceptual Model for Biomarker Progression in AD
Jack and colleagues detail a model of AD biomarkers that highlights the temporal
progression of biomarker changes that occur long before the onset of clinical symptoms,
emphasizing their value in early detection (8). The earliest biomarker changes are observed in
CSF Aβ42 levels, which are observed to decrease more than a decade before cognitive
impairment is evident and reflect early amyloid plaque deposition. PET amyloid imaging
confirms these CSF changes slightly later. Following biomarker evidence of Aβ abnormalities,
increases in CSF levels of tau markers (t-tau and p-tau) indicate the onset of neurofibrillary
tangle formation and neuronal injury. After CSF tau markers increase, neuroimaging markers of
neurodegeneration including FDG-PET hypometabolism and structural MRI atrophy become
abnormal; these neuroimaging markers correlate with cognitive decline. This sequence of AD
biomarker progression highlights that the preclinical phase of AD is of long duration; biomarkers
evolve progressively while cognitive function remains intact. These findings reinforce the use of
biomarkers for early diagnosis and intervention, as their changes precede clinical symptoms by
years, offering critical insights into disease pathophysiology.
4

Figure 1. Figure reproduced from Jack et al., 2013, Lancet Neurology, 12(2), 207–216.
Early versus Late Intervention Trial with Estradiol (ELITE)
Despite extensive research, the underlying mechanisms and potential interventions for
AD remain areas of active investigation. One area of particular interest is the role of hormone
therapy (HT) in influencing the progression of cognitive decline and the development of
dementia, particularly in postmenopausal women. The ELITE clinical trial was specifically
designed to test the hormone-timing hypothesis in relation to atherosclerosis progression in
postmenopausal women. ELITE demonstrated that oral estradiol therapy was associated with less
progression of subclinical atherosclerosis (measured as carotid artery intima-media thickness,
CIMT) than was placebo when therapy was initiated within 6 years after menopause but not
when it was initiated 10 or more years after menopause.
5

The cognitive effects of estradiol in postmenopausal women were also investigated in
ELITE. Hormone therapy effects on cognitive composite scores of global cognition, verbal
memory, and executive functions did not significantly differ from placebo in either
postmenopausal stratum. However, emerging evidence suggests that alterations in AD
biomarkers, such as reductions in CSF Aβ42 or increases in tau levels, appear decades before
clinical manifestations of dementia, including cognitive decline. These early biomarker changes,
as described by Jack et al., follow a specific temporal sequence and are pivotal in tracking the
progression of AD pathophysiology. As such, hormone therapy effects on AD plasma biomarkers
may be detectable even in the absence of observable differences in cognitive function between
treatment and placebo groups. In this study, we therefore used ELITE data and AD plasma
biomarkers measured from stored samples to test for HT effects on these AD biomarkers.
6

Chapter 2: Methods
Study Design
ELITE was a single-center, randomized, double-blinded, placebo-controlled trial in
which serial carotid arterial measurements were obtained through noninvasive carotid artery
ultrasound. Participants were healthy postmenopausal women without diabetes and clinical
evidence of cardiovascular disease and no regular menses for at least 6 months or who had
surgically induced menopause, as well as a serum estradiol level lower than 25 pg/ml (92 pmol
per liter). At the time of randomization, participants were stratified according to the number of
years past menopause: less than 6 years or 10 or more years.
The trial was approved by the University of Southern California Institutional Review
Board. All participants provided written informed consent. An external data and safety
monitoring board appointed by the National Institute on Aging met approximately every 6
months during the trial to monitor the safety of participants and trial conduct.
Randomization and Treatment
Participants were randomly assigned in a 1:1 ratio to receive oral 17β-estradiol (1 mg
daily) or matching placebo within strata of early postmenopause (<6 years since menopause) and
late postmenopause (≥10 years since menopause). Additional randomization stratification factors
were baseline CIMT (<0.75 or ≥0.75 mm) and hysterectomy status (yes or no). Women in the
estradiol group who had a uterus also received micronized progesterone (45 mg) as a 4% vaginal
gel, and women in the placebo group who had a uterus received matching placebo gel; the
estradiol or placebo gel was applied sequentially (i.e., once daily for 10 days during each 30-day
7

cycle). The participants, investigators, staff, imaging specialists, and data monitors were unaware
of treatment assignments.
Follow-Up
Evaluations of participants were scheduled monthly in a specialized research clinic for
the first 6 months and then every other month until trial completion. Blood samples were
collected at every visit, and these stored samples were used for measuring AD plasma
biomarkers.
Assessment of AD Biomarkers
Five biomarkers of AD risk were measured in the stored plasma samples from ELITE
participants at baseline and the 2.5-year visit. The plasma AD biomarkers included: Aβ40, Aβ42,
p-tau181, Glial fibrillary acidic protein (GFAP), and NfL. Plasma samples were preprocessed by
diluting 4 times and running in duplicates. The p-tau181 concentration was measured by
Quanterix (Billerica, MA, USA) using their Simoa® Human pTau-181 Advantage V2.1 assay kit
(Product # 104111). The Simoa® pTau-181 Advantage kits were used according to
manufacturer’s instructions with commercial availability. As for the other AD biomarkers, GFAP,
Aβ40, Aβ42, and NfL were measured with the likewise pre-processed diluted samples at
Quanterix (Billerica, MA, USA) using the commercially available Simoa® Neuro 4-Plex E
Advantage kit (Product # 103670) under the manufacturer’s instructions. All assays were
performed on the Simoa HD-X analyzer using single molecule array (Simoa) technology(11,12).
The mean of replicates for each sample was calculated. The coefficient of variation across
replicates <25% was the standard for inclusion of any measure in the analysis.
8

Statistical Analysis
We conducted a series of analyses to evaluate the effect of estradiol therapy on AD
biomarkers, specifically Aβ40, Aβ42, ptau181, NfL, and GFAP. To address the hormone-timing
hypothesis—that the effect of hormone therapy on AD biomarkers may differ based on the
timing of initiation hormone therapy relative to menopause—analyses were stratified by time
since menopause (early vs. late postmenopause). Biomarker levels were initially examined for
normality, and those that were skewed (NfL, GFAP, and pau181) were log-transformed to
improve distributional assumptions (e.g., log-NfL, log-GFAP, log-pau181) at individual time
points. Baseline biomarker levels were included as covariates in all models to adjust for
individual variability at baseline. The decision to include these variables was based on their
biological relevance and standard practice in longitudinal biomarker analyses. Adjustment for
baseline values helps control for potential confounding and ensures that observed changes are
attributable to the treatment effect.
To assess the impact of hormone therapy on changes in biomarker levels from baseline to
30 months, we calculated the raw (untransformed) change in biomarker levels. A general linear
model (GLM) was used to compare the mean 30-month change in biomarker levels between the
estradiol and placebo groups, adjusting for baseline biomarker values as covariates. For each
biomarker, we reported the mean levels at baseline and 30 months, the mean change from
baseline, and the between-group mean difference in 30-month change (with 95% confidence
intervals).
9

We conducted additional analyses stratified by menopausal status to investigate whether
the effect of hormone therapy on AD biomarkers differed between early and late postmenopausal
women. Separate GLMs were fitted for the early postmenopause (<6 years) and late
postmenopause (≥10 years) groups, with treatment group (estradiol versus placebo) as the
primary independent variable, while adjusting for baseline biomarker levels.
We tested for an interaction effect between treatment group and menopausal status by
fitting a GLM in the total sample, which included main effects for HT, menopausal status (early
vs. late), and their interaction term. This model allowed us to directly test whether the effect of
estradiol on biomarker levels varied by time since menopause. Residuals from the GLM were
analyzed to assess model assumptions. A histogram and Q-Q plot evaluated residual normality,
and residuals vs. predicted values plots checked for homoscedasticity. All analyses were
performed using SAS Studio software (version 3.83). Statistical significance was set at two-sided
p < 0.05.
10

Chapter 3: Results
The study sample characteristics are summarized in Table 1. Means and standard
deviations (SD) were calculated for continuous variables; frequencies and percentages were
computed for categorical variables. The sample included 628 postmenopausal women (HT:
n=318; Placebo: n=310) after excluding 15 participants with incomplete biomarker data. Mean
age was 61.0 ± 6.8 years (HT) and 60.3 ± 7.0 years (Placebo), with BMI of 27.5 ± 5.5 kg/m2
(HT) and 27.0 ± 5.3 kg/m2
(Placebo). The sample racial/ethnic composition was 68.2%
Non-Hispanic White, 14.0% Hispanic White, 9.4% Black, and 8.4% Asian. Education levels
varied, with most completing high school or higher. Time since menopause was distributed as:
n=267 early (<6 years) and n=361 late (≥10 years) postmenopause.
11

Table 1: Demographic Characteristics of Participants at Baseline
Variable HT (n= 318) Placebo (n=310)
Age (mean ± SD) 61.0 ± 6.8 60.3 ±7.0
BMI (mean ± SD) kg/m² 27.5 ±5.5 27.01 ±5.25
Race (%)
Non-Hispanic White 227 (36%) 201(32%)
Black 27(10%) 32(5%)
Hispanic White 41(6%) 47(8%)
Asian 23(3%) 30(5%)
Education (%)
Less than high school 1(<1%) 2(<1%)
High school or some college 83(13%) 109(17%)
College graduate 213(34%) 188(30%)
Time since menopause
Early (<6 years) 135(21%) 132(21%)
Late (≥10 years) 183(29%) 178(28%)
12

Figure 2: Distribution and Transformation of Variables
13

Note: Aβ40 and Aβ42 appeared to be normally distributed. These biomarkers were not transformed.
14

Figure 3: Residual Analysis for AD Biomarkers
15

16

Residual Analysis
Residual analysis was conducted to assess the assumptions of the linear models used in
this study. For each biomarker, including ptau181, abeta40, abeta42, nfl, and gfap, residuals were
obtained from the General Linear Models (GLM) that analyzed the adjusted mean change
between treatment groups. Diagnostic plots were generated for residuals to evaluate normality
and homoscedasticity.
Histograms of residuals, overlaid with a normal distribution curve, and quantile-quantile
(Q-Q) plots indicated that the residuals were approximately normally distributed for all
biomarkers. No evidence of severe skewness or kurtosis was observed. Additionally, scatterplots
of residuals versus predicted values showed no discernible patterns, confirming that the
assumption of homoscedasticity was met.
Biomarker Differences by Treatment Group
The baseline, 30-month, and change values for AD biomarkers in the HT and placebo
groups are presented in Table 2. Group differences were assessed for statistical significance using
GLMs. Baseline and 30-month biomarker levels that were not normally distributed (e.g., NfL,
GFAP, ptau181) were log-transformed to improve normality (Figure 2), with results
back-transformed and reported on the original scale. Change variables were normally distributed
and analyzed without transformation.
17

At baseline, no statistically significant differences were observed between the HT and
placebo groups for any biomarker. The p-values for Aβ42/Aβ40, Aβ40, Aβ42, NfL, GFAP, and
pTtu181 ranged from 0.43 to 0.94, confirming comparable baseline levels across treatment
groups.
At 30 months, the HT and placebo groups exhibited significant differences in Aβ40 levels
(adjusted p = 0.02), with lower levels observed in the HT group. A borderline difference was
noted for GFAP levels (adjusted p = 0.17), but no significant differences were detected for
Aβ42/Aβ40, Aβ42, or NfL (adjusted p-values of 0.49, 0.21, and 0.99, respectively). While
unadjusted analyses indicated higher ptau181 levels in the placebo group (p = 0.02), this
difference remained statistically significant after baseline adjustment (p = 0.04).
From baseline to 30 months, the HT group demonstrated significantly greater reductions
in Aβ40 levels compared to placebo (adjusted p = 0.02). A similar effect was observed for
ptau181, with a significantly greater? reduction in the HT group relative to placebo after
adjustment (p = 0.04). No significant group differences were detected for Aβ42/Aβ40 ratio,
Aβ42, NfL, or GFAP changes (adjusted p-values of 0.49, 0.06, 0.99, and 0.17, respectively).
18

Table 2: Mean (SD) of AD Biomarkers at Baseline, Visit 30 and Change by Treatment Group
Biomarker HT Placebo Unadjusted P-value
for difference
P-value for difference
adjusted for baseline
Baseline
(HT = 318 Placebo = 310)
Aβ42/Aβ40 0.074 ± 0.013 0.074 ± 0.123 0.78 n/a
Aβ40 pg/mL 78.66 ± 0.201 78.574 ± 0.211 0.91 n/a
Aβ42 pg/mL 5.739± 0.238 5.731± 0.251 0.94 n/a
NfL pg/mL 12.032 ± 0.422 11.780 ± 0.430 0.51 n/a
GFAP pg/mL 96.733 ± 0.438 95.701± 0.380 0.72 n/a
ptau181 pg/mL 19.611 ± 0.324 20.050 ± 0.325 0.43 n/a
30 – Month
(HT = 217 Placebo = 225)
Aβ42/Aβ40
pg/mL
0.075 ± 0.014 0.075 ± 0.014 0.99 0.49
Aβ40 pg/mL 75.537 ± 0.184 78.169 ± 0.1812 0.05 0.02
Aβ42 pg/mL 5.572 ± 0.251 5.770 ± 0.243 0.12 0.21
NfL pg/mL 12.618±0.434 12.643 ± 0.471 0.96 0.99
GFAP pg/mL 94.681±0.442 99.542 ± 0.427 0.23 0.17
ptau181 pg/mL 18.839 ± 0.324 20.327 ± 0.471 0.02 0.04
30-Month minus Baseline Change
(HT= 216 Placebo = 222)
Aβ42/Aβ40 0.002 ± 0.009 0.002 ± 0.008 0.43 0.49
Aβ40 -0.068 ± 0.206 0.028 ± 0.192 0.04 0.02
Aβ42 -0.038 ± 0.183 -0.009 ± 0.164 0.36 0.06
NfL 0.049 ± 0.355 0.049 ± 0.401 0.99 0.99
GFAP -0.031 ± 0.256 -0.001 ± 0.269 0.22 0.17
ptau181 -0.046 ± 0.315 -0.006 ± 0.274 0.16 0.04
Table note: The differences between HT and placebo groups for each biomarker at baseline and 30 months and 30-month change
were analyzed using general linear models. P-values represent the significance of the mean difference between groups at each
time point. The log transformed means (NfL, GFAP, ptau181) were exponentiated to show the raw means of the biomarkers. All
biomarker concentrations reported in pg/mL 30-month models and 30-month change from baseline were adjusted for baseline
visit.
19

Biomarker Differences by Treatment Group and Postmenopausal Status
Table 3 presents the mean differences (±SD) in the change in AD biomarkers from
baseline to 30 months, stratified by HT and postmenopausal status (early: <6 years since
menopause; late: ≥10 years since menopause). Differences between treatment groups were
assessed within each postmenopausal stratum, along with the interaction between treatment
group and time since menopause.
For the Aβ42/Aβ40 ratio, no significant differences were observed between HT and
placebo groups in either the early (p = 0.26) or late (p = 0.48) postmenopause groups. The
interaction between treatment group and postmenopausal status was not significant (p = 0.17),
indicating that the effect of HT on the Aβ42/Aβ40 ratio was consistent across postmenopausal
strata.
For Aβ40, significant group differences were observed in both the early and late
postmenopause groups (p = 0.04 and p = 0.03, respectively). In the early menopause subgroup,
higher 30-month increases in Aβ40 were noted in HT- compared to placebo-treated women
(p=0.04). In the late menopause subgroup, mean reductions in Aβ40 were less in the HT
compared to the placebo group (p=0.03). The interaction between treatment group and
postmenopausal status on Aβ40 was not significant (p = 0.26), suggesting that the effect of HT
on Aβ40 did not vary significantly by time since menopause.
20

For Aβ42, directions of mean changes by treatment group and postmenopause stratum
mirrored that of Aβ40; however, no significant differences were detected between HT and
placebo groups in either the early (p = 0.26) or late (p = 0.18) postmenopause groups. Similarly,
the interaction between treatment and postmenopausal status was not significant (p = 0.09),
indicating no differential effect of HT on Aβ42 levels by time since menopause.
For NfL, no significant differences were observed between HT and placebo groups in
either the early (p = 0.30) or late (p = 0.15) postmenopause groups. The interaction term was also
non-significant (p = 0.97), suggesting no interaction between treatment and postmenopausal
status on NfL levels.
For GFAP, no statistically significant differences were observed between HT and placebo
groups in either the early (p = 0.25) or late (p = 0.62) postmenopause groups. The interaction
between treatment and menopausal status was also non-significant (p = 0.24).
For ptau181, differences between HT and placebo groups were not significant within the
early (p = 0.16) or late (p = 0.16) postmenopause groups. However, the interaction term was
significant (p = 0.04), suggesting that the effect of HT on ptau181 levels was greater in the early
menopause group (treatment beta (SE) = 0.068 (0.326)) than in the late postmenopause group
(treatment beta (SE) = 0.025 (0.305).
21

Note: Differences were calculated by subtracting the baseline (BL) biomarker levels from those measured at Visit 30 (V30) for
each participant (i.e., Difference = V30 - BL). The reported values represent the mean and standard deviation (SD) of these
differences, stratified by hormone therapy (HT) status and time since menopause (Early <6 years, Late ≥10 years).
22

Chapter 4: Discussion
The findings from this study provide valuable insights into the effects of HT on AD
plasma biomarkers and how these effects may vary based on the timing of HT initiation relative
to menopause. Consistent with the hormone timing hypothesis, some biomarkers exhibited
differential responses to HT depending on whether therapy was initiated during early (<6 years
since menopause) or late (≥10 years since menopause) postmenopause. Significant changes in
Aβ40 and ptau181 suggest a potential influence of HT on early and intermediate pathological
processes associated with AD, though the effects varied by menopausal timing.
For Aβ40, HT was associated with significant differences in both early and late
postmenopause groups. In the early group, Aβ40 levels were higher in the HT group compared to
placebo (p = 0.04). In contrast, in the late postmenopause group, Aβ40 levels were significantly
lower in the HT group compared to placebo (p = 0.03). These findings suggest a
timing-dependent effect of HT on amyloid dynamics, potentially reflecting variations in amyloid
production, clearance, or metabolism based on the timing of intervention. However, the absence
of a significant interaction effect (p = 0.26) indicates that these changes occurred independently
of menopausal timing.
For ptau181, differences between HT and placebo groups were not significant within the
early (p = 0.16) or late (p = 0.16) postmenopause groups. However, the significant interaction
between treatment and menopausal timing (p = 0.04) suggests that HT had a greater effect on
reducing tau pathology in the late postmenopause group compared to the early group. This
23

finding aligns with the biomarker progression model proposed by Jack et al., which posits that
tau pathology emerges later in the AD continuum, potentially making it more responsive to
interventions during later stages.
In contrast, no significant differences or interaction effects were observed for other
biomarkers, including Aβ42/Aβ40 ratio, Aβ42, NfL, and GFAP. The absence of significant
changes in NfL and GFAP suggests that HT may have limited effects on markers of
neurodegeneration and glial activation within the study’s timeframe. Similarly, the lack of
significant changes in the Aβ42/Aβ40 ratio and Aβ42 levels may reflect the complex dynamics of
amyloid pathology, which may require longer follow-up or alternative biomarkers to detect
meaningful changes.
Interestingly, the interaction effects observed for ptau181 underscore the importance of
considering both the timing of HT initiation and the stage of AD pathology when evaluating its
effects. While HT appears to modulate tau pathology in later stages, its impact on other
biomarkers may be more subtle or require a longer timeframe to emerge. These findings suggest
that the timing and target of HT interventions are critical factors in understanding its role in AD
prevention.
24

Limitations
Several limitations must be acknowledged. First, the study’s relatively short duration of
30 months may not capture long-term trends in biomarker changes or their clinical implications.
Biomarkers like GFAP and NfL, which reflect later-stage neurodegeneration and glial activation,
may require extended follow-up to detect significant differences. Second, the study population
consisted of healthy postmenopausal women without cognitive impairment, which limits the
generalizability of the findings to populations with greater AD risk or existing cognitive
symptoms. Third, plasma biomarkers, while less invasive than cerebrospinal fluid measures, may
have lower sensitivity and specificity for detecting subtle changes in AD pathology, potentially
underestimating HT effects. Finally, the sample size in stratified analyses may have reduced
statistical power to detect significant differences in certain biomarkers, particularly interaction
effects.
Integration with Previous Research
These findings are consistent with the conceptual model of biomarker progression
described by Jack et al., which highlights the temporal sequence of biomarker changes in AD.
The significant effects observed for Aβ40 and ptau181 support the notion that HT may influence
amyloid and tau pathology, particularly during specific stages of disease progression. The lack of
significant changes in NfL and GFAP, which are markers of neurodegeneration and glial
activation, further underscores the timing-dependent nature of HT’s effects. These results
reinforce the importance
25

of early interventions targeting preclinical stages of AD to mitigate the risk of downstream
neurodegenerative processes.
Overall, this study adds to the growing body of evidence on the timing-dependent effects
of HT on AD pathology. Future research should focus on longer follow-up periods, more diverse
populations, and complementary biomarker analyses to build on these findings. Investigating
HT’s anti-inflammatory mechanisms and its interaction with genetic risk factors could provide
further insights into its role in AD prevention and management.
26

References
1. Alzheimer’s Disease Prevalence in the United States:
Alzheimer's Association. (2023). Alzheimer's disease facts and figures. Retrieved from
https://www.alz.org/alzheimers-dementia/facts-figures
2. Pathophysiology of Alzheimer’s Disease:
National Institute on Aging. (n.d.). What are biomarkers of Alzheimer’s disease? Retrieved from
https://www.nia.nih.gov/health/what-are-biomarkers-alzheimers-disease
3. Neuropsychiatric Symptoms in Alzheimer’s Disease:
National Institute on Aging. (n.d.). Alzheimer's disease fact sheet. Retrieved from
https://www.nia.nih.gov/health/alzheimers-disease-fact-sheet
4. Neuroimaging and Biomarker Analysis in Alzheimer’s Disease:
BrightFocus Foundation. (n.d.). Cerebrospinal fluid (CSF) biomarkers and Alzheimer's disease.
Retrieved from
https://www.brightfocus.org/alzheimers/article/cerebrospinal-fluid-csf-biomarkers-and-alzheimer
s-disease
5. CSF Biomarkers in Alzheimer’s Disease:
Martínez-Dubarbie, J., et al. (n.d.). Plasma biomarkers for Alzheimer’s disease pathology:
Insights from Lumipulse platform studies. Retrieved from https://www.example-reference.com
6. Plasma Biomarkers in Alzheimer’s Disease:
EMJ Reviews. (n.d.). High diagnostic accuracy for Alzheimer’s disease with plasma p-tau217.
Retrieved from
https://www.emjreviews.com/neurology/news/high-diagnostic-accuracy-for-alzheimers-diseasewith-plasma-p-tau217/
27

7. Biomarker Onset in Alzheimer’s Disease:
Lloret, A., Esteve, D., Lloret, M. A., Cervera-Ferri, A., Lopez, B., Nepomuceno, M., & Monllor,
P. (2019). When does Alzheimer’s disease really start? The role of biomarkers. International
Journal of Molecular Sciences, 20(22), 5536. https://doi.org/10.3390/ijms20225536
8. Conceptual Model for Biomarker Progression in Alzheimer’s Disease:
Jack, C. R., Jr., Knopman, D. S., Jagust, W. J., Shaw, L. M., Aisen, P. S., Weiner, M. W., ...
Trojanowski, J. Q. (2013). Tracking pathophysiological processes in Alzheimer’s disease: An
updated hypothetical model of dynamic biomarkers. The Lancet Neurology, 12(2), 207–216.
https://doi.org/10.1016/S1474-4422(12)70291-0
9. Figure Reproduced from Jack et al., 2013:
Jack, C. R., Jr., Knopman, D. S., Jagust, W. J., Shaw, L. M., Aisen, P. S., Weiner, M. W., ...
Trojanowski, J. Q. (2013). Tracking pathophysiological processes in Alzheimer’s disease: An
updated hypothetical model of dynamic biomarkers. The Lancet Neurology, 12(2), 207–216.
https://doi.org/10.1016/S1474-4422(12)70291-0
11. Teruya, Tomoyuki, et al. "Plasma Biomarkers for Predicting the Development of Dementia in
a Japanese Community: The Hisayama Study." Alzheimer's & Dementia, vol. 20, no. 2, 2024, pp.
310–322. PubMed, https://pubmed.ncbi.nlm.nih.gov/38606661/.
12. Ashton, Nicholas J., et al. "Plasma Extracellular Vesicle Tau and TDP-43 as Diagnostic
Biomarkers in Alzheimer's Disease and Frontotemporal Lobar Degeneration." Nature Medicine,
vol. 30, 2024, pp. 255–267. Nature, https://www.nature.com/articles/s41591-024-02937
28

Abstract (if available)

Abstract Alzheimer’s disease (AD) is a progressive neurodegenerative condition that primarily affects older adults, with early pathological changes, including amyloid-beta (Aβ) accumulation and tau tangles, occurring decades before symptoms. This study analyzed data from the Early versus Late Intervention Trial with Estradiol (ELITE) to examine the effects of hormone therapy (HT) on AD plasma biomarkers—Aβ40, Aβ42, pTau181, neurofilament light chain (NfL), and glial fibrillary acidic protein (GFAP)—in postmenopausal women. The hormone-timing hypothesis was tested by stratifying analyses based on time since menopause (early vs. late postmenopause).

Using a randomized, double-blind, placebo-controlled design, baseline and 30-month plasma biomarker levels were compared between HT and placebo groups. Significant reductions in Aβ40 and pTau181 levels were observed in the HT group, with effects differing by menopausal status. Aβ40 levels decreased more in the late postmenopause group, while pTau181 reductions were greater in the early group. No significant changes were detected for Aβ42/Aβ40, NfL, or GFAP.

These findings support the timing hypothesis, suggesting that HT affects amyloid and tau pathology differently based on menopause timing. This study highlights the potential of HT as an early intervention to address preclinical AD pathology, although further research is needed to confirm long-term effects.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Exploring the relationship between menopausal hot flushes and Alzheimer's disease biomarkers: a cross-sectional analysis in postmenopausal women

PDF

Hormone therapy timing hypothesis and atherosclerosis

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

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

PDF

Association between endogenous sex hormone levels and cognition: the Women’s Isoflavone Soy Health (WISH) trial

PDF

Associations between physical activities with bone mineral density in postmenopausal women

PDF

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

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

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

PDF

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

PDF

Sex hormones and atherosclerosis in postmenopausal women

PDF

Effect of soy isoflavones on anthropometric and metabolic measurements in postmenopausal women

PDF

Relationship of blood pressure and antihypertensive medications to cognitive change in the BVAIT, WISH, and ELITE clinical trials

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

Comparison of participant and study partner predictions of cognitive impairment in the Alzheimer's disease neuroimaging initiative 3 study

PDF

Long-term blood pressure variability across the clinical and biomarker spectrum of Alzheimer’s disease

PDF

The evaluation of the long-term effectiveness of zero/low fluoroscopy workflow in ablation procedures for the treatment of paroxysmal and persistent atrial fibrillation

PDF

Associations between inflammatory markers and change in cognitive endpoints

PDF

Age related macular degeneration in Latinos: risk factors and impact on quality of life

Asset Metadata

Creator Argueta Rufino, Debbie Jannice (author)

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

School Keck School of Medicine

Degree Master of Science

Degree Program Applied Biostatistics and Epidemiology

Degree Conferral Date 2025-05

Publication Date 02/04/2025

Defense Date 02/04/2025

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

Tag Alzheimer's disease,biomarker,cognitive effects,hormone therapy,plasma,postmenopause

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

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

Creator Email arguetar@usc.edu,debbiejannice@gmail.com

Unique identifier UC11399GTH5

Identifier etd-ArguetaRuf-13813.pdf (filename)

Legacy Identifier etd-ArguetaRuf-13813

Document Type Thesis

Format theses (aat)

Rights Argueta Rufino, Debbie Jannice

Internet Media Type application/pdf

Type texts

Source 20250211-usctheses-batch-1241 (batch), 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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

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

Repository Email cisadmin@lib.usc.edu