# paper.pdf ## Metadata - PDFFormatVersion=1.7 - Language=English - IsLinearized=true - IsAcroFormPresent=false - IsXFAPresent=false - IsCollectionPresent=false - IsSignaturesPresent=false - Creator=Elsevier - Custom.CrossMarkDomains[1]=elsevier.com - Custom.CrossmarkMajorVersionDate=2010-04-23 - Custom.CreationDate--Text=6th May 2026 - Custom.ElsevierWebPDFSpecifications=7.0.1 - Custom.robots=noindex - Custom.doi=10.1016/j.tjpad.2026.100581 - Custom.CrossMarkDomains[2]=sciencedirect.com - Custom.CrossmarkDomainExclusive=true - ModDate=D:20260506114240Z - Author=Kitty K Lui - Title=Sleep complaints and genetic risk of Alzheimer’s disease in older women: associations with memory and tau deposition - Keywords=Alzheimer?s disease,Sleep,Polygenic hazard score,Apoe ?4,Memory,tau PET - CreationDate=D:20260506112045Z - Producer=Acrobat Distiller 8.1.0 (Windows) - Subject=The Journal of Prevention of Alzheimer's Disease, 13 (2026) 100581. doi:10.1016/j.tjpad.2026.100581 - ali:license_ref=http://creativecommons.org/licenses/by-nc-nd/4.0/ - crossmark:crossmarkdomains=elsevier.comsciencedirect.com - crossmark:crossmarkdomainexclusive=true - crossmark:doi=10.1016/j.tjpad.2026.100581 - crossmark:majorversiondate=2010-04-23 - dc:format=application/pdf - dc:identifier=10.1016/j.tjpad.2026.100581 - dc:publisher=Elsevier Masson SAS - dc:description=The Journal of Prevention of Alzheimer's Disease, 13 (2026) 100581. doi:10.1016/j.tjpad.2026.100581 - dc:subject=Alzheimer?s disease,Sleep,Polygenic hazard score,Apoe ?4,Memory,tau PET - dc:title=Sleep complaints and genetic risk of Alzheimer’s disease in older women: associations with memory and tau deposition - dc:creator=Kitty K Lui,Xin Wang,Melanie A Dratva,Ella T. Lifset,Jordan Stiver,Nadine C. Heyworth,Qian Shen,Michael Thomas,Pamela N. DeYoung,Atul Malhotra,Erin E. Sundermann,Sarah J. Banks - jav:journal_article_version=VoR - pdf:creationdate--text=6th May 2026 - pdf:producer=Acrobat Distiller 8.1.0 (Windows) - pdf:keywords=Alzheimer?s disease,Sleep,Polygenic hazard score,Apoe ?4,Memory,tau PET - pdfx:creationdate--text=6th May 2026 - pdfx:crossmarkdomains=sciencedirect.comelsevier.com - pdfx:crossmarkdomainexclusive=true - pdfx:crossmarkmajorversiondate=2010-04-23 - pdfx:elsevierwebpdfspecifications=7.0.1 - pdfx:doi=10.1016/j.tjpad.2026.100581 - pdfx:robots=noindex - prism:aggregationtype=journal - prism:coverdate=2026-08-01 - prism:coverdisplaydate=1 August 2026 - prism:doi=10.1016/j.tjpad.2026.100581 - prism:issn=2274-5807 - prism:number=7 - prism:pagerange=100581 - prism:publicationname=The Journal of Prevention of Alzheimer's Disease - prism:startingpage=100581 - prism:url=https://doi.org/10.1016/j.tjpad.2026.100581 - prism:volume=13 - tdm:policy=https://www.elsevier.com/tdm/tdmrep-policy.json - tdm:reservation=1 - xmp:createdate=2026-05-06T11:20:45 - xmp:creatortool=Elsevier - xmp:metadatadate=2026-05-06T11:42:40 - xmp:modifydate=2026-05-06T11:42:40 - xmpmm:documentid=uuid:1b499bed-4ce8-4c0c-b682-0d58baae1cbe - xmpmm:instanceid=uuid:b6ed5266-03cb-4855-90c0-636283357f42 - xmprights:marked=True ## Contents ### Page 1 ## Original Article # Sleep complaints and genetic risk of Alzheimer’s disease in older women: associations with memory and tau deposition Kitty K Lui ![orcid]() , Xin Wang, Melanie A Dratva, ![orcid]() , Ella T. Lifset, Jordan Stiver ![orcid]() , Nadine C. Heyworth, Qian Shen ![orcid]() , Michael Thomas, Pamela N. DeYoung, Atul Malhotra, Erin E. Sundermann, Sarah J. Banks,, ## A R T I C L E I N F O ### Key words: Alzheimer’s disease Sleep Polygenic hazard score Apoe ε4 Memory tau PET ## A B S T R A C T ### Background: Emerging evidence point to a bidirectional relationship between sleep disturbances and Alzheimer’s disease (AD). Poor sleep may be an overlooked risk factor for older women, who are disproportionately affected by AD and report worse subjective sleep quality than men. High genetic AD risk—characterized by the polygenic hazard score (PHS), including apolipoprotein (APOE) ε4 carriership—may further compound the effects of disrupted sleep on AD, particularly for older women. ### Objective: This study examined the moderating effect of genetic AD risk on subjective sleep as it related to memory and tau burden in a sample of older women. ### Participants: The sample consisted of older women (≥65 years old) from the Women Inflammation Tau Study. ### Measurement: Participants completed the Pittsburgh Sleep Quality Index (PSQI), Rey Auditory Learning Test, and Brief Visuospatial Memory Test-Revised. They also underwent [18]F-MK6240 positron emission tomography. Tau burden was calculated in composite regions across Braak stages. Genetic risk groups were characterized by PHS stratified at the 75th percentile. PSQI global score × PHS group interactions on memory composite scores (N= 69) and tau burden (N = 63) were examined. ### Results: PSQI global score × PHS group interactions were observed on visual memory and pathological tau in Braak regions III/IV (ps<0.10). Poorer subjective sleep was associated with worse visual memory and greater limbic tau deposition only among higher genetic risk women (ps<0.04). No significant associations were observed for verbal memory or tau in Braak regions I/II or V/VI. ### Conclusion: Older women with elevated genetic AD risk and subjective sleep difficulties may be at greater risk for visual memory deficits and tau burden in regions affected in early AD. This suggests that sleep complaints may represent a promising AD risk factor. Improving sleep may be a potential intervention target for AD mitigation and prevention, particularly for older women. ## 1. Introduction There is an urgent need to identify modifiable targets that will delay or prevent the onset of cognitive decline in Alzheimer’s disease (AD) []. Disrupted sleep is a promising modifiable risk factor that often predates expression of hallmark AD pathologies, predicts clinical symptoms, and increases the risk of AD dementia [,]. In particular, pathological tau in the form of neurofibrillary tangles (NFTs) is thought to play a key role in the link between sleep and AD. Decades before clinical symptoms of AD and prior to amyloid (Aβ) plaques deposition, phosphorylated tau starts to aggregate in subcortical regions important for regulating sleep-wake cycles—including the locus coeruleus (LC) and basal forebrain—and has been implicated in disrupting sleep and circadian rhythms []. Tau deposition in these regions occur in the earliest stages of AD, preceding ### Page 2 even the entorhinal cortex []. An expanding body of research supports a bidirectional relationship between sleep and AD, such that sleep dis-turbances have a role in NFT expression and memory dysfunction [,]. Notably, mice models have shown chronic sleep deprivation in-creases tau spreading in the LC []. Further, poor sleep quality has been associated with more tau deposition in temporal lobe regions affected in early AD [,]. Together, this bidirectional relationship creates a vi-cious cycle in which sleep disruption and AD pathogenesis mutually reinforce each other contributing to progressive cognitive and func-tional decline [,,]. Compared to men, women are disproportionately affected by AD, comprising two thirds of AD cases [,], and may also experience delayed mild cognitive impairment (MCI) diagnosis because of the fe-male advantage in verbal memory [,]. For example, even in the MCI stage, women outperform men in verbal memory measures despite equivalent levels of Aβ pathology, hippocampal volume, and hypo-metabolism [,,]. Women’s verbal memory reserve starts to diminish after MCI onset, with studies showing faster cognitive decline in women with MCI versus men with MCI [,]. In contrast, visual memory tests do not show a strong sex bias [,] and demonstrate similarly robust, if not stronger associations [,] with AD pathology, hippocampal volume, and AD risk []. The growing body of research suggesting sex differences in early AD detection [] highlights the importance of incorporating both visual and verbal memory since visual memory may be a sensitive marker of incipient AD-related changes in women. Emerging evidence point to sleep disturbances as an overlooked AD risk factor that potentially contribute to the sex disparity in AD []. First, compared to men, women are more likely to have sleep complaints, including dissatisfaction with sleep quality, yet sleep problems are often both underdiagnosed and undertreated, particularly in women [,]. Moreover, Huang et al. reported that poor sleep was more strongly related to cognitive decline, hippocampal atrophy, and AD incidence in women than men. Our own work further showed that comorbid insomnia and sleep apnea (COMISA) attenuated the female advantage in verbal memory []. Sleep disturbances in middle age have also been linked to accelerated tau accumulation in later life []. This may be especially relevant for women who not only accumulate tau faster than men, but also experience significant changes in sleep patterns and poor sleep quality with the menopause transition []. Importantly, about 30 % of women continue to experience menopause-related sleep difficulties into older age []. Collectively, these studies implicate sleep disruption as a potentially relevant AD risk factor for women; although, the sleep and AD relationship remains poorly understood from a sex-specific perspective. Increasing evidence suggests that the effect of disrupted sleep on tau pathology may be a function of apolipoprotein (APOE) ε4 genotype []. APOE ε4 is posited to promote earlier and faster tau accumu-lation in sleep-wake regulating regions, thereby increasing vulnerability to dysregulated sleep [,,]. Supporting this, previous work has foundAPOE ε4 carriers exhibiting worse objective and subjective sleep pa-rameters as opposed to non-carriers []; however, this has not been consistently reported across studies []. Further, the relationship between sleep difficulties and AD have been found to be more pro-nounced in APOE ε4 carriers [,,]. APOE ε4 carriership also con-tributes to sex differences in AD as female ε4 carriers progress to AD earlier, accumulate tau more rapidly, and exhibit faster memory decline than male ε4 carriers and women non-carriers [,]. The cumu-lative effects of poor sleep and APOE ε4 may therefore disproportion-ately affect women. For example, Koo et al. found that female ε4 carriers across the AD continuum were more likely to experience sleep disruptions than female non-carriers. Yet, the interplay between sleep and genetic AD risk remain largely unknown, especially for women with elevated genetic risk who are most vulnerable AD. Research has primarily focused on APOE ε4 as the primary genetic risk factor for AD. More recently, the polygenic hazard score (PHS) have gained recognition as a more sensitive measure to determine genetic AD risk since it accounts not only for the cumulative effects of two APOEvariants (ε2 and ε4), but also other single nucleotide polymorphisms (SNPs) that been linked to small increases in disease risk []. For instance, the PHS include SNPs associated with BIN1, FERMT2, and others that have been linked to tau deposition [,,,]. AD PHS has showed to be more sensitive than APOE ε4 alone in predicting age of AD onset, Aβ and tau pathogenesis, and cognitive decline [,,,]. Intriguingly, even in absence of APOE ε4, PHS is still predictive of age of AD onset []. This unique sensitivity of the PHS in capturing conver-gence and progression of AD lends support in integrating the PHS in AD research, particularly in the context of how modifiable risk factors relate to AD biomarkers in those with high polygenic risk. To date, no study has leveraged the PHS to characterize the relationship between sleep and AD. In the Women: Inflammation Tau Study (WITS), we examined the relationship of self-reported sleep to memory and tau burden and the modifying role of AD genetic risk (i.e., PHS and APOE ε4 status). We hypothesized that poorer sleep would be reported in women with higher genetic AD risk. Further, worse subjective sleep would be related to higher tau pathology and poorer memory more strongly in women with higher genetic risk than women with lower genetic risk. ## 2. Methods ### 2.1. Participants WITS is an ongoing, prospective study at the University of California San Diego. This study focuses on older women at higher AD risk by way of older age (≥65 years), telephone Montreal Cognitive Assessment (T- MoCA) [] score of 13–20 out of 22 (suggestive of MCI), and either a PHS≥50th percentile and/or family history of dementia. Other inclusion criteria are ability to complete psychometric and other clinical assess-ments in English (i.e., adequate English language skills, vision, and hearing) and no major medical incident in the past 4 weeks (e.g., trauma or infection). Exclusion criteria are contraindication to lumbar puncture, chronic major psychiatric or neurodegenerative disorders, unstable or poorly controlled medical problems (e.g., heart failure, diabetes, hy-pertension, pulmonary disease with hypoxia or hypercapnia, significant liver problems or renal failure, major inflammatory disorders), current use of illegal substances or excessive alcohol use, taking specific medi-cations known to influence our measures of interest (cognition, inflammation, hormones, insulin resistance) including neuroleptics, anti-parkinsonism drugs, central nervous system stimulations, anticon-vulsants, insulin, coumadin, sedating antihistamines or hypnotics, and potent anti-inflammatory medications as well as magnetic resonance imaging (MRI) or PET contraindications, including non-MR safe elec-tronic devices or implants (e.g., cochlear implant), claustrophobia, adverse reactions to intravenous (IV)/oral contrasts, and body mass index (BMI) greater than 35 kg/m2. Seventy-three women have been enrolled in WITS so far. On average, all study procedures were completed within a 58-day period (SD=86 days, range=7–457 days). Of the 73 participants enrolled in the study, 4 participants had incomplete or invalid Pittsburgh Sleep Quality Index (PSQI) data and thus, were subsequently removed resulting in a final sample of N = 69 in analyses between PSQI and memory. Regarding tau PET data, 2 participants withdrew from the study prior to undergoing tau PET, 2 participants had been excluded from the study prior to neu-roimaging due to contraindication (but had completed PSQI and neu-rocognitive testing), and 1 participant had significant MRI artifact and did not complete PSQI, resulting in an N = 68. Within that subsample, 3 participants had invalid or incomplete PSQI data and an additional 2 participants had tau standard uptake value ratios (SUVRs) in Braak re-gions I/II and Braak regions III/IV that were statistical outliers (>3SD). Thus, these participants were removed from analyses, resulting in a final sample of N = 63 in analyses between PSQI and tau PET. A visual ### Page 3 depiction of participant exclusion is shown in. ### 2.2. Polygenic hazard score (PHS) Saliva samples were collected at-home as part of pre-enrollment to characterize AD risk for study eligibility. The PHS was developed to estimate genetic AD risk and is associated with age of AD onset []. Genetic sequencing was performed by Diagnomics Lab (San Diego, CA) and the AD PHS was computed based on 198,424 SNPs from the Illumina Global Screening Array, including two APOE variants (ε2 and ε4) []. Ambiguous SNPs were excluded. No additional quality control proced-ures were applied to assess genotype quality. Since WITS enriched for participants with already elevated genetic risk (PHS≥50th percentile) and given the sensitivity of the 75th percentile in identifying those with highest risk for AD devel-opment—evidenced by earliest age of AD and highest yearly incidence rate of AD []—the 75th percentile was used to stratify the sample into higher-risk PHS group (PHS≥75th percentile) and lower-risk PHS group (PHS<75th percentile). ### 2.3. Pittsburgh sleep quality index (PSQI) PSQI is a questionnaire that evaluates sleep quality over the past month across seven domains of sleep health: subjective sleep quality, sleep latency, sleep duration, sleep efficiency, sleep disturbances, use of sleep medication, and daytime dysfunction []. Nineteen items were transformed into seven component scores and treated as ordinal level data (range: 0–3). These component scores were then summed together to calculate a global score (range: 0–21) with higher scores reflecting poorer sleep. A global score >5 distinguished poor sleepers from good sleepers and has been validated as a sensitive measure of sleep quality in older adults [,]. ### 2.4. Memory tests Both visual and verbal memory were assessed in order to have measure of memory with and without a sex bias to account for the well- established female advantage in verbal memory observed across cognitively normal and MCI stages [,]. For visual memory, the Brief Visuospatial Memory Test-Revised (BVMT-R) was administered, which required learning six geometric shapes that were presented for 10s each over 3 trials []. After each learning trial, participants were asked to draw the geometric shapes. Following a 25-minute delay period, participants were asked to repro-duce those shapes again. Each shape was scored with 2 points (one for correct shape, the other for correct location). The maximum score for total learning across the 3 trials was 36 and delayed recall was 12. For verbal memory, the Rey Auditory Verbal Learning Test (RAVLT) was administered, which required learning a 15-word orally presented list read over 5 learning trials []. Participants were asked to immediately recall as many words as possible after each learning trial. Short-delayed recall was assessed after an interference list and long-delayed recall was assessed after a 20-minute delay period. The maximum score for total learning across 5 trials was 75 and delayed recall was 15. Visual and verbal memory composite scores were computed from the total learning and delayed recall raw scores from the BVMT-R and RAVLT, respectively. Each score was transformed to a z-score and then averaged together. ### 2.5. MRI MRI was collected in a 3.0T GE 750 with the sequence of inversion recovery fast spoiled gradient-echo sequence (IR-FSPGR; TI = 400 ms, FA = 11◦, voxel resolution = 1 × 1 × 1 mm3) scanner following ADNI protocols [] at the Altman Clinical Translational Research Institute (San Diego, CA). T1 weighted images were processed with FreeSurfer (v.7.1.1) to derive region of interests (ROIs) in each participant’s native ### Page 4 space using the Desikan-Killiany atlas. Quality control of FreeSurfer outputs were conducted by trained MRI analysts (coauthors: XW and MAD). Visual inspection was conducted to assess whether the pial and white matter surfaces accurately followed the gray–white matter boundaries, and whether the subcortical segmentation conformed to anatomical intensity boundaries. Manual edits (when deemed neces-sary) involved removing non-brain voxels from the pial surface and adding white matter control points to correct white matter segmentation errors by improving intensity normalization. Twelve participants had manual edits completed (n = 6 with PHS ≥75th percentile). ### 2.6. Tau PET imaging [18]F-MK6240 PET imaging was acquired in a GE Discovery 610 PET/computed tomography (CT) scanner at California Protons Cancer Therapy Center (San Diego, CA). The [18]F-MK6240 ligand has demonstrated better signal-to-noise ratio, has stronger affinity to NFTs, and reduced off-targeting binding compared to older generation tracers []. Participants were injected with 5 mCI (185 MBq) ±10 % dose of tracer for at least 60 secs. Scans started at 80 min post injection and lasted for 30 min (6 × 5 min frames). Smoothing, interframe realign-ment, and co-registration of 4D tau PET to T1 weighted MRI was per-formed using SPM12. The first 4 frames of PET data were then summed to generate SUVRs with the eroded inferior cerebellum grey matter as the reference region []. The eroded inferior cerebellum was created in the automated anatomical labeling (AAL) atlas in native T1 space from each individual’s segmented MRI []. Tau PET processing followed an established pipeline that was described in Betthauser et al. 2019 []. Since tau deposition typically follows a predictable temporal pattern [], SUVRs were calculated in composite regions associated with Braak regions I/II (entorhinal cortex and hippocampus), III/IV (limbic re-gions), and V/VI (neocortex) [,]. The regions used to calculate SUVRs in the Braak regions are shown in. ### 2.7. Statistical analysis Independent samples t-tests (i.e., Student’s t-test or Mann-Whitney Utest if normality was not met) or Chi-Square (Х²) tests were used to analyze differences in participant characteristics, PSQI global score, tau SUVR in composite ROIs, and memory scores between PHS groups. Normality of variables were examined with Shapiro-Wilk test and log- transformed variables were used if normality significantly improved. Linear regression was used to analyze the interaction between PSQI global score and PHS groups on tau SUVR or memory scores while controlling for age and years of education. BMI was included as a co-variate in models with tau SUVR as the outcome variable since there are reported associations between tau burden and BMI []. Interactions were probed at p < 0.10 and was followed by simple slope analyses with significance level set at p < 0.05. In models with a non-significant interaction at p ≥ 0.10, the interaction term was removed to assess the main effect of PSQI global score; significance was set at p < 0.05. In addition to p-values, effect sizes were reported as standardized beta-coefficients (β). β values between 0.10–0.29 were considered small, 0.30–0.49 were considered medium, and >0.50 were considered large []. Women with preexisting sleep disorders were not excluded from the study (n = 8). Two women had a diagnosis of insomnia disorder that was treated with medication (one with an antidepressant and the other with an anticonvulsant), five women had a known sleep apnea diagnosis (with two using continuous positive airway pressure [CPAP]), and one woman reported significant sleep problems that was treated with an antidepressant. Sensitivity analyses were conducted to exclude the 8 women with sleep disorders to examine whether these associations were present independent of clinically significant sleep problems. Further, to compare the results between PHS and APOE ε4, we examined the interactive effects of PSQI global score and APOE ε4 on memory and tau pathology while controlling for the aforementioned covariates. Lastly, for significant associations with PSQI global score, exploratory analyses were conducted to identify which of the 7 domains of sleep health were contributing to the effects. The statistical approach is described in theSupplementary Materials. All statistical analyses were performed using R Studio (V.2023.09.0 + 463). ## 3. Results ### 3.1. Participant characteristics Participant characteristics by PHS group can be found in. In this sample of 69 women (age: 72.5 ± 4.5, 93 % White [87 % Non- Hispanic/Latino; 6 % Hispanic/Latino], years of education: 16±2), 26 had a PHS≥75th percentile (higher-risk PHS group) and 43 had a PHS<75th percentile (lower-risk PHS group). Within the lower-risk PHS group, 19 women (34 %) had a PHS<50th percentile. Higher-risk PHS women were more likely to be APOE ε4 carriers (100 %) as opposed to those with lower-risk PHS (7 %; p < 0.001). Lower-risk PHS women had significantly higher PSQI global scores (more self-reported sleep diffi-culties) than higher-risk PHS women (p = 0.04). There were no signif-icant PHS group differences in age, years of education, t-MoCA scores, BMI, or memory scores. ### Page 5 Participant characteristics by PHS group of the subset of 63 women with tau PET data can be found in, which included 22 partici-pants with PHS≥75th percentile (all APOE ε4 carriers) and 43 partici-pants with PHS<75th percentile (3 APOE ε4 carriers). Consistent with past findings [], those in the higher-risk group had higher patholog-ical tau burden than those in the lower-risk group (ps=0.003–0.05). There were no significant differences in age, years of education, t-MoCA scores, BMI, PSQI global score, and memory scores between the higher-risk and lower-risk PHS group. In a subset of participants (N = 53), reproductive health information was available. The PHS groups did not significantly differ between age at menopause, type of menopause, sleep changes during the menopausal transition, and hormone replacement therapy use during/after the menopausal transition (ps>0.20; Table S1). ### 3.2. Interactions between PSQI global score and PHS on memory There was an interaction between PSQI global score and PHS group on the visual memory composite score at a trend level, despite a large effect size (B= 0.17, β= 0.52, SE=0.08, p = 0.05;). Simple slope analysis revealed that in only higher-risk PHS women, higher PSQI global score (i.e., more self-reported sleep difficulties) related to worse visual memory (B= 0.16, β= 0.51, SE=0.07, p = 0.02;A). This was not observed in lower-risk PHS women (p = 0.92). Results that examined the interactions between each PSQI component score and PHS on the visual memory composite score are presented in the Supple-mentary Materials. The interaction between PSQI global score and PHS group on the verbal memory composite score was not significant and effect size was small (B= 0.08, β= 0.27, SE=0.09, p = 0.33;B). There was also no main effect of PSQI global score on the verbal memory composite score which was further supported with a small effect size (B= 0.13,β= 0.14, SE=0.24, p = 0.59). ### 3.3. Interactions between PSQI global score and PHS on tau burden PSQI global score interacted with PHS group to predict tau SUVR in Braak regions III/IV at a trend level, although there was a medium effect size (B = 0.02, β=0.46, SE=0.01, p = 0.098;). Simple slopes analysis showed that only among those with higher-risk PHS, more subjective sleep difficulties were associated with higher tau deposition in Braak regions III/IV (B = 0.02, β=0.53, SE=0.01, p = 0.03). This association was not observed in those with lower-risk PHS (p = 0.61). Results that examined the interaction between each PSQI component and PHS on tau burden in Braak regions III/IV are presented in Sup-plementary Materials. PSQI global score did not interact with PHS to predict tau SUVR in Braak regions I/II, despite a medium effect size (β=0.32, p = 0.23). Similarly, no interaction was not observed for Braak regions V/VI and the effect size was small (β=0.17, p = 0.55). There was no main effect of PSQI global score on Braak regions I/II or V/VI, and the effect sizes were small (βs<0.23; ps>0.12). ### 3.4. Sensitivity analyses in women without sleep disorders We performed sensitivity analyses excluding the 8 women with sleep disorders (3 with PHS≥75th percentile) to account for whether clinical sleep problems (treated or untreated) may have confounded our find-ings. The PSQI global score and PHS group interaction on visual memory was no longer significant (B= 0.16, β= 0.45, SE=0.09, p = 0.11). Although, the medium effect size suggests that we may be underpow-ered to detect a significant effect given the reduced sample size. Despite the non-significant interaction, simple slopes similarly showed that in higher-risk PHS women, greater PSQI score was related to lower visual memory scores (B= 0.14, SE=0.08, p = 0.08), and not in lower-risk PHS women (p = 0.82). The PSQI global score × PHS group interaction on tau SUVR in Braak regions III/IV was at trend-level, with a large effect size (B = 0.02,β=0.54, SE=0.01, p = 0.06). Simple slope analysis continued to show that in those with higher-risk PHS, more report of sleep difficulties was related to higher tau burden in Braak III/IV (B = 0.023, SE=0.01, p =0.04), and not in lower-risk PHS women (p = 0.71). ### 3.5. Sensitivity analyses with APOE ε4 status To compare the results with PHS, APOE ε4 stratified analyses were also conducted. PSQI global score interacted with APOE ε4 to predict visual memory composite score (B= 1.63, β= 0.51, SE=0.09, p =0.06). Only among APOE ε4 carriers, more subjective sleep difficulties were related to worse visual memory performance (B= 0.16, β= 0.51, SE=0.07, p = 0.02), but not in non-carriers (p = 0.99). There was no interaction between PSQI global score and APOE ε4 status on verbal memory composite score (B= 0.05, β= 0.16, SE=0.09, p = 0.57). As for tau pathology, PSQI global score interacted with APOE ε4 status to predict tau SUVR in Braak regions III/IV (B = 0.02, β=0.57, SE=0.01, p = 0.04). In APOE ε4 carriers, more sleep complaints were related to greater tau burden in limbic regions (b = 0.02, β=0.61, SE=0.01, p = 0.01); which was not observed in non-carriers (p = 0.80). PSQI global score did not interact with APOE ε4 status to predict tau deposition in Braak regions I/II, although the effect size was medium (B= 0.03, β=0.42, SE=0.02, p = 0.13). There was no significant interac-tion between PSQI global score and tau burden on Braak regions V/VI and the effect size was small (B < 0.01, β=0.22, SE=0.01, p = 0.45). ### Page 6 ### Page 7 ## 4. Discussion In this study, lower-risk PHS women (PHS<75th percentile) showed higher PSQI global scores, reflecting worse subjective sleep, than higher- risk PHS women (PHS≥75th percentile), which was contrary to our hypotheses. Despite fewer sleep complaints overall, in only women with higher genetic AD risk, more subjective sleep difficulties were related to worse visual memory and greater tau burden in limbic regions. This was not observed in those with lower genetic risk. Self-reported sleep problems were not associated with verbal memory or tau deposition in Braak regions I/II and V/VI in a PHS-dependent manner or across the whole sample. It remains unclear whether genetic AD risk heightens susceptibility to sleep disturbances. Our results showed that women with lower ge-netic risk had worse subjective sleep difficulties compared to women with higher genetic risk. A similar finding has been previously reported in the context of APOE ε4 carriership, such that non-carriers reported worse sleep []. In contrast, other studies have reported that APOE ε4 carriers were more likely to report worse sleep measures of shorter duration, longer sleep latency, and more disruptions than non-carriers [,,], or no APOE ε4-dependent effects on subjective sleep mea-sures [,,]. Perhaps, in the higher genetic risk group, we may be capturing a discrepancy between self-reported and objective sleep measures, whereby those with MCI tend to overestimate their subjective sleep, possibly reflecting anosognosia or memory recall deficits, early clinical symptoms of AD [,]. While our results show that higher genetic risk does not appear to worsen self-reported sleep, it potentially points to an interactive effect between poor subjective sleep and early pathological and cognitive changes among individuals with elevated AD genetic risk. There are a few proposed mechanisms by which APOE ε4 contribute to the bidirectional relationship between sleep and AD. First, APOE ε4 accelerates early tau aggregation in the basal forebrain, a region involved in REM sleep-wake regulation, which may explain the marked reduction in REM sleep starting in midlife, particularly in ε4 carriers [,,,]. This significant change in REM sleep architecture could then lead to acute disruptions in neural functions that support memory consolidation as well as next-day learning and retrieval of new infor-mation []. Secondly, mice models have demonstrated synergistic ef-fects of APOE ε4 and sleep disturbances in accelerating Aβ plaque deposition and subsequently promoting Aβ-related tau seeding and spreading []. In line with this, a human observational study reported that worse sleep quality was more strongly associated with worse cognition, greater hippocampal atrophy, and higher AD incidence inAPOE ε4 carriers []. Similarly, older adults with APOE ε4 positivity and poor sleep consolidation demonstrated the fastest cognitive decline, highest AD incidence, and greatest post-mortem NFT density; however, good sleep consolidation appeared to attenuate the adverse effect ofAPOE ε4 and those individuals resembled AD-related outcomes similar to non-carriers (irrespective of their sleep pattern) []. Together, these findings implicate that poor sleep could accelerate AD pathogenesis and cognitive impairment in those with elevated genetic risk, however, improving sleep may confer some resilience against the disease. Lastly, sleep disturbances could also increase susceptibility to AD through metabolic dysfunction and increased inflammatory response with APOEε4 exacerbating these processes []. Specifically, Huang et al. (2025) identified GDF15, a marker of cellular stress, mitochondrial dysfunction, and inflammation, as a key protein linking poor sleep and AD in APOE ε4 carriers []. In prior work, the mechanistic links between sleep and genetic AD risk have largely focused on APOE ε4. Leng et al. reported that AD polygenic risk score was associated with higher odds of short sleep duration, with the effects attenuated when removing APOE ε4. The re-sults of our study cannot parse out the effect of APOE ε4 in the higher genetic risk group since 89 % of the women were APOE ε4 carriers. Further, our sensitivity analyses also showed that in only APOE ε4 carriers, PSQI global score was related to limbic tau pathology and vi-sual memory. It remains unknown whether SNPs associated with other AD-related genes like BIN1, play a prominent role in the sleep and AD relationships, independent of APOE. For example, while BIN1 has been implicated in tau pathogenesis [], whether BIN1 disrupts neurobio-logical mechanisms important for regulating sleep-wake functions have not been examined. Rodent and human studies enriching for higher genetic risk without APOE ε4 positivity are warranted to disentangle the strong influence of APOE ε4 versus the polygenic component on sleep disruption. Our findings indicated that a PHS-dependent association of poor sleep was specific to visual memory, suggesting that visual memory could potentially be a sensitive marker of sleep-related memory impairment for older women with higher genetic AD risk. The absence of verbal memory findings could reflect the female verbal advantage [,], potentially conferring some protection against the effects of poor sleep in preclinical and prodromal AD, even in those with elevated ge-netic risk. Together, it raises the possibility that poor subjective sleep may act synergistically with genetic risk to increase vulnerability to AD and cognitive decline, with visual memory deficits being more sensitive to this combined influence. These results differ from the scant studies that have examined the modifying role of APOE ε4 on the association between subjective sleep measures and memory, particularly with visual memory. Past research have not observed a moderating role of APOE ε4 in the relationship of self-reported sleep to either verbal and/or visual memory in samples of both women and men [,]. In those studies, the absence of visual memory findings could be attributable to different assessment formats (i.e., incidental encoding of a complex figure as opposed to repeat learning trials of shapes) [] and combining verbal and visual memory into a memory composite score []. While other studies have shown stronger associations between insomnia and verbal memory in APOE ε4 carrier than non-carriers, they did not assess for visual memory nor conduct sex-stratified analyses. Further, those samples were generally younger (mean ages: mid to late 50s) than this sample (mean age: 72.5 years) [,]. Though, a possible expla-nation is that specific symptoms to insomnia (e.g., troubles with latency and maintenance and early awakenings) could be more closely linked to verbal memory, which somewhat aligns with our past work that found that women with COMISA exhibited an attenuated female verbal memory advantage when compared to women with obstructive sleep apnea only []. While this present study showed that poor subjective sleep quality was not related to verbal memory, it does not preclude the possibility that other dimensions of sleep disturbances (e.g., insomnia symptoms, sleep-disordered breathing) may be associated with visual and/or verbal memory deficits across sexes, highlighting the need for further research. We found that among women with higher genetic risk, more sleep complaints were associated with greater limbic tau pathology. This ex-tends previous work that showed actigraphy-derived measures of less consolidated sleep, less consistent sleep efficiency, and shorter sleep duration were associated with elevated AD pathology, including NFTs and cortical Aβ, in APOE ε4 carriers [,,]. Although our findings were limited to tau deposition in Braak III/IV regions, this is consistent with earlier work from our group that demonstrated in Aβ+ older male Veterans, PSQI global scores was associated with tau burden in only Braak regions III/IV []. These specific relationships to Braak regions III/IV might reflect that poor self-reported sleep is a risk factor for more widespread tau buildup in limbic regions and adjacent cortical regions, beyond the entorhinal cortex and hippocampus (Braak regions I/II), particularly during the early symptomatic stages of AD. A few limitations of this study need to be addressed. Generalizability of these results to the wider population are limited due to a sample comprised of mostly White non-Hispanic/Latino and highly educated women and on a genetic marker that was developed in primarily White European cohorts []. The sample size also placed some constraints on statistical power, and we did not correct for multiple comparisons. Also, ### Page 8 removing those with sleep disorders (n = 8) resulted in nonsignificant and trend-level interactions, despite comparable effect sizes, potentially reflecting reduced power in the subsample. Notably, across interactions with p < 0.10, the effect sizes were medium to large, indicating a considerable effect. To address this power issue, we will replicate these analyses once data collection is complete, and also assess longitudinal relationships. Even though the PSQI does not concretely map on to a specific sleep disorder, it is a validated questionnaire that has high sensitivity and specificity in capturing sleep quality. Also, while it could be argued that subjective sleep measurements may be less accurate or reliable compared to objective sleep, there is compelling evidence that self-reported sleep represents a unique dimension of sleep and has also been linked to AD pathology independent of the effects of objective sleep []. A comprehensive approach incorporating both objective and self-report measures may better capture the complexity of sleep in order to improve our understanding of its relationship with AD. In conclusion, our findings suggest a link between poor sleep and AD risk, specifically limbic tau pathology and visual memory, particularly in older women with greater genetic predisposition to AD. Our findings support the use of PHS in combination with APOE ε4 status in subse-quent work. Notably, self-reported sleep measures, like the PSQI, are cost-effective and low burden, and could be a promising marker to detect early AD changes, offering a practical tool to identify individuals at elevated AD risk in wider populations, although further investigation is warranted. Further, this study highlights sleep as a potential modifi-able target to promote healthy brain aging in older women. We acknowledge that we did not test whether these are sex-specific findings and may be observed in older men as well. Thus, replicating these an-alyses with sex-stratified approaches are necessary in larger, ethno-culturally diverse samples in order to elucidate sleep disturbances as an AD risk factor for each sex. ## Funding This study was supported by the National Institute of Aging (Co PIs: S. J. Banks and E. E. Sundermann; R01AG080663-01/02), Women’s Alzheimer’s Movement, the Alzheimer’s Association, and UCSD Stein Institute for Research on Aging and Center for Healthy Aging. WITS was also supported by funding from the California Department of Public Health, Chronic Disease Control Branch, Alzheimer’s Disease Program (19-10613). The findings and conclusions in this report are those of the author(s) and do not necessarily represent the views or opinions of the California Department of Public Health or the California Health and Human Services Agency. ## Declaration of the use of generative AI and AI-assisted technologies in scientific writing and in figures, images and artwork AI (i.e., ChatGPT) was only used to improve clarity of certain sentences. ## Ethical statement All study protocols were approved by the Institutional Review Board (IRB) at the University of California, San Diego (under protocol number 200383). All participants provided written informed consent to partic-ipate in this study. ## Data statement Please contact PI Dr. Sarah Banks (sbanks@health.ucsd.edu) for data requests. ## CRediT authorship contribution statement Kitty K Lui: Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Data curation. Xin Wang: Writing – re-view & editing, Methodology, Data curation. Melanie A Dratva:Writing – review & editing, Project administration, Investigation, Formal analysis, Data curation. Ella T. Lifset: Writing – review &editing, Data curation. Jordan Stiver: Writing – review & editing.Nadine C. Heyworth: Project administration, Data curation. Qian Shen: Writing – review & editing, Formal analysis. Michael Thomas:Writing – review & editing, Formal analysis. Pamela N. DeYoung:Writing – review & editing, Data curation. Atul Malhotra: Writing –review & editing, Supervision, Conceptualization. Erin E. Sunder-mann: Writing – review & editing, Supervision, Conceptualization.Sarah J. Banks: Writing – review & editing, Supervision, Conceptualization. ## Declaration of competing interest The authors declare the following financial interests/personal re-lationships which may be considered as potential competing interests: Sarah J. Banks reports was provided by National Institute of Aging. Sarah J. Banks reports financial support was provided by Women’s Alzheimer’s Movement. Sarah J. Banks reports was provided by Alz-heimer’s Association. Sarah J. Banks reports financial support was provided by UCSD Stein Institute for Research on Aging and Center for Healthy Aging. Sarah J. Banks reports financial support was provided by California Department of Public Health. Dr. Malhotra is funded by NIH. He reports income from Eli Lilly, Livanova, Sunrise, Zoll and Powell Mansfield. He is co-founder of Clairyon, a small startup unrelated to this topic. ResMed provided a philanthropic donation to UCSD. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. ## Acknowledgements We would like to thank the research participants of WITS and the WITS team for contributing to this study. ## Supplementary materials Supplementary material associated with this article can be found, in the online version, at. ## References - [1] Hickman R, Faustin A, Wisniewski T. Alzheimer Disease and its growing epidemic: risk factors, biomarkers and the urgent need for therapeutics. Neurol Clin 2016;34 (4):941–53. https://doi.org/10.1016/j.ncl.2016.06.009. - [2] Johnson CE, Duncan MJ, Murphy MP. Sex and sleep disruption as contributing factors in Alzheimer’s disease. J Alzheimers Dis JAD 2024;97(1):31–74. https:// doi.org/10.3233/JAD-230527. - [3] Shi L, Chen SJ, Ma MY, et al. Sleep disturbances increase the risk of dementia: a systematic review and meta-analysis. Sleep Med Rev 2018;40:4–16. https://doi. org/10.1016/j.smrv.2017.06.010. - [4] Wang C, Holtzman DM. Bidirectional relationship between sleep and Alzheimer’s disease: role of amyloid, tau, and other factors. Neuropsychopharmacology 2020; 45(1):104–20. https://doi.org/10.1038/s41386-019-0478-5. - [5] Braak H, Del Tredici K. Where, when, and in what form does sporadic Alzheimer’s disease begin? Curr Opin Neurol 2012;25(6):708. https://doi.org/10.1097/ WCO.0b013e32835a3432. - [6] Kang SS, Ahn EH, Liu X, et al. ApoE4 inhibition of VMAT2 in the locus coeruleus exacerbates Tau pathology in Alzheimer’s disease. Acta Neuropathol (Berl) 2021; 142(1):139–58. https://doi.org/10.1007/s00401-021-02315-1. - [7] Lew CH, Petersen C, Neylan TC, Grinberg LT. Tau-driven degeneration of sleep- and wake-regulating neurons in Alzheimer’s disease. Sleep Med Rev 2021;60: 101541. https://doi.org/10.1016/j.smrv.2021.101541. - [8] Holth JK, Fritschi SK, Wang C, et al. The sleep-wake cycle regulates brain interstitial fluid tau in mice and CSF tau in humans. Science (1979) 2019;363 (6429):880–4. https://doi.org/10.1126/science.aav2546. ### Page 9 - [9] Mander BA, Winer JR, Jagust WJ, Walker MP. Sleep: a novel mechanistic pathway, biomarker, and treatment target in the pathology of Alzheimer’s disease? Trends Neurosci 2016;39(8):552–66. https://doi.org/10.1016/j.tins.2016.05.002. - [10] Saletin JM, Walker MP. Nocturnal mnemonics: sleep and hippocampal memory processing. Front Neurol 2012;3:59. https://doi.org/10.3389/fneur.2012.00059. - [11] Andrews M, Ross R, Malhotra A, et al. Sleep and tau pathology in Vietnam war veterans with preclinical and prodromal Alzheimer’s disease. J Alzheimers Dis Rep 2021;5(1):41–8. https://doi.org/10.3233/ADR-200245. - [12] Winer JR, Morehouse A, Fenton L, et al. Tau and β-amyloid burden predict actigraphy-measured and self-reported impairment and misperception of Human sleep. J Neurosci 2021;41(36):7687–96. https://doi.org/10.1523/ JNEUROSCI.0353-21.2021. - [13] Wang C, Nambiar A, Strickland MR, et al. APOE-ε4 synergizes with sleep disruption to accelerate Aβ deposition and aβ-associated tau seeding and spreading. J Clin Invest 2023;133(14). https://doi.org/10.1172/JCI169131. - [14]. - [15] Bubu OM, Andrade AG, Umasabor-Bubu OQ, et al. Obstructive sleep apnea, cognition and Alzheimer’s disease: a systematic review integrating three decades of multidisciplinary research. Sleep Med Rev 2020;50:101250. https://doi.org/ 10.1016/j.smrv.2019.101250. - [16] 2025 Alzheimer’s disease facts and figures. Alzheimers Dement 2025;21(4): e70235. https://doi.org/10.1002/alz.70235. - [17] Nebel RA, Aggarwal NT, Barnes LL, et al. Understanding the impact of sex and gender in Alzheimer’s disease: a call to action. Alzheimers Dement 2018;14(9): 1171–83. https://doi.org/10.1016/j.jalz.2018.04.008. - [18] Sundermann EE, Biegon A, Rubin LH, Lipton RB, Landau S, Maki PM. Does the female advantage in verbal memory contribute to underestimating AD pathology in women versus men? J Alzheimers Dis JAD 2017;56(3):947–57. https://doi.org/ 10.3233/JAD-160716. - [19] Emrani S, Sundermann EE. Sex/gender differences in the clinical trajectory of Alzheimer’s disease: insights into diagnosis and cognitive reserve. Front Neuroendocrinol 2025;77:101184. https://doi.org/10.1016/j.yfrne.2025.101184. - [20] Sundermann EE, Maki PM, Rubin LH, et al. Female advantage in verbal memory: evidence of sex-specific cognitive reserve. Neurology 2016;87(18):1916–24. https://doi.org/10.1212/WNL.0000000000003288. - [21] Sundermann EE, Biegon A, Rubin LH, et al. Better verbal memory in women than men in MCI despite similar levels of hippocampal atrophy. Neurology 2016;86(15): 1368–76. https://doi.org/10.1212/WNL.0000000000002570. - [22] Chapman RM, Mapstone M, Gardner MN, et al. Women have farther to fall: gender differences between normal elderly and Alzheimer’s disease in verbal memory engender better detection of AD in Women. J Int Neuropsychol Soc JINS 2011;17 (4):654–62. https://doi.org/10.1017/S1355617711000452. - [23] Sundermann EE, Banks SJ, Bondi MW, et al. Sex differences in the relationship of biomarker change to memory decline in early Alzheimer’s disease: an observational cohort study. Biol Sex Differ 2026;17(1):38. https://doi.org/ 10.1186/s13293-025-00820-6. - [24] Brunet HE, Caldwell JZK, Brandt J, Miller JB. Influence of sex differences in interpreting learning and memory within a clinical sample of older adults. Neuropsychol Dev Cogn B Aging Neuropsychol Cogn 2020;27(1):18–39. https:// doi.org/10.1080/13825585.2019.1566433. - [25] Gale SD, Baxter L, Connor DJ, Herring A, Comer J. Sex differences on the Rey Auditory Verbal Learning Test and the Brief Visuospatial Memory Test-revised in the elderly: normative data in 172 participants. J Clin Exp Neuropsychol 2007;29 (5):561–7. https://doi.org/10.1080/13803390600864760. - [26] Bonner-Jackson A, Mahmoud S, Miller J, Banks SJ. Verbal and non-verbal memory and hippocampal volumes in a memory clinic population. Alzheimers Res Ther 2015;7(1):61. https://doi.org/10.1186/s13195-015-0147-9. - [27] Ziontz J, Harrison TM, Fonseca C, et al. Connectivity, pathology, and ApoE4 interactions predict longitudinal tau spatial progression and memory. Hum Brain Mapp 2024;45(17):e70083. https://doi.org/10.1002/hbm.70083. - [28] Zammit AR, Ezzati A, Zimmerman ME, Lipton RB, Lipton ML, Katz MJ. Roles of hippocampal subfields in verbal and visual episodic memory. Behav Brain Res 2017;317:157–62. https://doi.org/10.1016/j.bbr.2016.09.038. - [29] Seo EH, Lim HJ, Yoon HJ, et al. Visuospatial memory impairment as a potential neurocognitive marker to predict tau pathology in Alzheimer’s continuum. Alzheimers Res Ther 2021;13(1):167. https://doi.org/10.1186/s13195-021- 00909-1. - [30] Geiger AR, Guevara JE, King JB, Hoffman JM, Duff K. Using neuropsychological test scores to predict beta-amyloid deposition in older adults across the late-life cognitive continuum. J Clin Exp Neuropsychol 2025;47(7):692–700. https://doi. org/10.1080/13803395.2025.2578345. - [31] Ferretti MT, Iulita MF, Cavedo E, et al. Sex differences in Alzheimer disease — The gateway to precision medicine. Nat Rev Neurol 2018;14(8):457–69. https://doi. org/10.1038/s41582-018-0032-9. - [32] Van Den Berg JF, Miedema HME, Tulen JHM, Hofman A, Neven AK, Tiemeier H. Sex differences in subjective and actigraphic sleep measures: a population-based study of elderly persons. Sleep 2009;32(10):1367–75. https://doi.org/10.1093/ sleep/32.10.1367. - [33] Huang LY, Tan L, Tan CC, et al. Sleep quality, APOE ε4, and Alzheimer’s disease: associations from two prospective cohort studies and mechanisms by plasma proteomic analysis. BMC Med 2025;23(1):462. https://doi.org/10.1186/s12916- 025-04255-z. - [34] Holloway BM, Harding CD, DeYoung P, et al. Comorbid insomnia and sleep apnea is associated with worse verbal episodic memory in older females. J Clin Sleep Med JCSM Off Publ Am Acad Sleep Med 2025;21(12):2129–38. https://doi.org/ 10.5664/jcsm.11902. - [35] Shaver JL, Woods NF. Sleep and menopause: a narrative review. Menopause 2015; 22(8):899. https://doi.org/10.1097/GME.0000000000000499. - [36] Freedman RR, Roehrs TA. Sleep disturbance in menopause. Menopause 2007;14 (5):826. https://doi.org/10.1097/gme.0b013e3180321a22. - [37] Harrington YA, Parisi JM, Duan D, Rojo-Wissar DM, Holingue C, Spira AP. Sex hormones, sleep, and memory: interrelationships across the adult female lifespan. Front Aging Neurosci 2022;14. https://doi.org/10.3389/fnagi.2022.800278. - [38] Nemes S, Logan PE, Manchella MK, et al. Sex and APOE ε4 carrier effects on atrophy, amyloid PET, and tau PET burden in early-onset Alzheimer’s disease. Alzheimers Dement 2023;19(S9):S49–63. https://doi.org/10.1002/alz.13403. - [39] Buckley RF, Mormino EC, Rabin JS, et al. Sex differences in the association of global amyloid and regional tau deposition measured by positron emission tomography in clinically normal older adults. JAMA Neurol 2019;76(5):542–51. https://doi.org/10.1001/jamaneurol.2018.4693. - [40] Rajan KB, Weuve J, Barnes LL, McAninch EA, Wilson RS, Evans DA. Population estimate of people with clinical AD and mild cognitive impairment in the United States (2020–2060). Alzheimers Dement J Alzheimers Assoc 2021;17(12): 1966–75. https://doi.org/10.1002/alz.12362. - [41] Farrer LA. Effects of age, Sex, and ethnicity on the association between Apolipoprotein E genotype and Alzheimer Disease: a meta-analysis. JAMA 1997; 278(16):1349. https://doi.org/10.1001/jama.1997.03550160069041. - [42] Wang X, Zhou W, Ye T, Lin X, Zhang J. Sex difference in the association of APOE4 with memory decline in mild cognitive impairment. J Alzheimer’s Dis 2019;69(4): 1161–9. https://doi.org/10.3233/JAD-181234. - [43] Lim ASP, Yu L, Kowgier M, Schneider JA, Buchman AS, Bennett DA. Modification of the relationship of the apolipoprotein E ε4 allele to the risk of Alzheimer disease and neurofibrillary tangle density by sleep. JAMA Neurol 2013;70(12):1544–51. https://doi.org/10.1001/jamaneurol.2013.4215. - [44] Blackman J, Love S, Sinclair L, Cain R, Coulthard E. APOE ε4, Alzheimer’s disease neuropathology and sleep disturbance, in individuals with and without dementia. Alzheimers Res Ther 2022;14(1):47. https://doi.org/10.1186/s13195-022-00992- y. - [45] Drogos LL, Gill SJ, Tyndall AV, et al. Evidence of association between sleep quality and APOE ε4 in healthy older adults. Neurology 2016;87(17):1836–42. https:// doi.org/10.1212/WNL.0000000000003255. - [46] Burke SL, Maramaldi P, Cadet T, Kukull W. Associations between depression, sleep disturbance, and apolipoprotein E in the development of Alzheimer’s disease: dementia. Int Psychogeriatr 2016;28(9):1409–24. https://doi.org/10.1017/ S1041610216000405. - [47] Pivac LN, Brown BM, Sewell KR, et al. Suboptimal self-reported sleep efficiency and duration are associated with faster accumulation of brain amyloid beta in cognitively unimpaired older adults. Alzheimers Dement Diagn Assess Dis Monit 2024;16(2):e12579. https://doi.org/10.1002/dad2.12579. - [48] Palpatzis E, Bass N, Jones R, Mukadam N. Longitudinal association of apolipoprotein E and sleep with incident dementia. Alzheimers Dement 2022;18 (5):888–98. https://doi.org/10.1002/alz.12439. - [49] Kahya M, Vidoni E, Burns JM, Thompson AN, Meyer K, Siengsukon CF. The relationship between apolipoprotein ε4 carrier status and sleep characteristics in cognitively normal older adults. J Geriatr Psychiatry Neurol 2017;30(5):273–9. https://doi.org/10.1177/0891988717720301. - [50] L´opez-García S, Lage C, Pozueta A, et al. Sleep time estimated by an actigraphy watch correlates with CSF tau in cognitively unimpaired elders: the modulatory role of APOE. Front Aging Neurosci 2021;13. https://doi.org/10.3389/ fnagi.2021.663446. - [51] Fenton L, Isenberg AL, Aslanyan V, et al. Variability in objective sleep is associated with Alzheimer’s pathology and cognition. Brain Commun 2023;5(2):fcad031. https://doi.org/10.1093/braincomms/fcad031. - [52] Coughlan GT, Klinger HM, Boyle R, et al. Sex differences in longitudinal tau-PET in preclinical Alzheimer disease: a meta-analysis. JAMA Neurol 2025;82(4):364–75. https://doi.org/10.1001/jamaneurol.2025.0013. - [53] Neu SC, Pa J, Kukull W, et al. Apolipoprotein E genotype and sex risk factors for Alzheimer Disease: a meta-analysis. JAMA Neurol 2017;74(10):1178–89. https:// doi.org/10.1001/jamaneurol.2017.2188. - [54] Wang YTT, Pascoal TA, Therriault J, et al. Interactive rather than independent effect of APOE and sex potentiates tau deposition in women. Brain Commun 2021; 3(2):fcab126. https://doi.org/10.1093/braincomms/fcab126. - [55] Koo KYG, Schweizer TA, Fischer CE, Munoz DG. Abnormal sleep behaviours across the spectrum of Alzheimer’s Disease severity: influence of APOE genotypes and lewy bodies. Curr Alzheimer Res 2019;16(3):243–50. https://doi.org/10.2174/ 1567205016666190103161034. - [56] Desikan RS, Fan CC, Wang Y, et al. Genetic assessment of age-associated Alzheimer disease risk: development and validation of a polygenic hazard score. PLoS Med 2017;14(3):e1002258. https://doi.org/10.1371/journal.pmed.1002258. - [57] Tan CH, Fan CC, Mormino EC, et al. Polygenic hazard score: an enrichment marker for Alzheimer’s associated amyloid and tau deposition. Acta Neuropathol (Berl) 2018;135(1):85–93. https://doi.org/10.1007/s00401-017-1789-4. - [58] European Alzheimer’s Disease Initiative (EADI). Genetic and environmental risk in Alzheimer’s Disease (GERAD), Alzheimer’s Disease Genetic Consortium (ADGC), et al. Meta-analysis of 74,046 individuals identifies 11 new susceptibility loci for Alzheimer’s disease. Nat Genet 2013;45(12):1452–8. https://doi.org/10.1038/ ng.2802. - [59] Fan CC, Banks SJ, Thompson WK, et al. Sex-dependent autosomal effects on clinical progression of Alzheimer’s disease. Brain 2020;143(7):2272–80. https://doi.org/ 10.1093/brain/awaa164. ### Page 10 - [60] Wang X, Broce I, Qiu Y, et al. A simple genetic stratification method for lower cost, more expedient clinical trials in early Alzheimer’s disease: a preliminary study of tau PET and cognitive outcomes. Alzheimers Dement 2023;19(7):3078–86. https://doi.org/10.1002/alz.12952. - [61] Reas ET, Shadrin A, Frei O, et al. Improved multimodal prediction of progression from MCI to Alzheimer’s disease combining genetics with quantitative brain MRI and cognitive measures. Alzheimers Dement 2023;19(11):5151–8. https://doi.org/ 10.1002/alz.13112. - [62] Katz MJ, Wang C, CO Nester, et al. T-MoCA: a valid phone screen for cognitive impairment in diverse community samples. Alzheimers Dement Diagn Assess Dis Monit 2021;13(1):e12144. https://doi.org/10.1002/dad2.12144. - [63] Reas ET, Alderson-Myers A, Solders SK, et al. Long COVID-related blood-brain barrier breakdown and microstructure in older adults are modified by sex and Alzheimer’s disease genetic risk. Imaging Neurosci 2025;3:23. https://doi.org/ 10.1162/IMAG.a.23. IMAG.a. - [64] Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res 1989;28(2):193–213. https://doi.org/10.1016/0165-1781(89)90047-4. - [65] Vitiello MV, Larsen LH, Moe KE. Age-related sleep change: gender and estrogen effects on the subjective–objective sleep quality relationships of healthy, noncomplaining older men and women. J Psychosom Res 2004;56(5):503–10. https://doi.org/10.1016/S0022-3999(04)00023-6. - [66] Benedict RHB, Schretlen D, Groninger L, Dobraski M, Shpritz B. Revision of the Brief Visuospatial Memory Test: studies of normal performance, reliability, and validity. Psychol Assess 1996;8(2):145–53. https://doi.org/10.1037/1040- 3590.8.2.145. - [67]. - [68] Jack Jr CR, Bernstein MA, Fox NC, et al. The Alzheimer’s disease neuroimaging initiative (ADNI): MRI methods. J Magn Reson Imaging 2008;27(4):685–91. https://doi.org/10.1002/jmri.21049. - [69] Pascoal TA, Therriault J, Benedet AL, et al. 18F-MK-6240 PET for early and late detection of neurofibrillary tangles. Brain 2020;143(9):2818–30. https://doi.org/ 10.1093/brain/awaa180. - [70] Betthauser TJ, Cody KA, Zammit MD, et al. In vivo characterization and quantification of neurofibrillary tau PET radioligand 18F-MK-6240 in humans from Alzheimer Disease dementia to young controls. J Nucl Med Off Publ Soc Nucl Med 2019;60(1):93–9. https://doi.org/10.2967/jnumed.118.209650. - [71] Braak H, Alafuzoff I, Arzberger T, Kretzschmar H, Del Tredici K. Staging of Alzheimer disease-associated neurofibrillary pathology using paraffin sections and immunocytochemistry. Acta Neuropathol (Berl) 2006;112(4):389–404. https:// doi.org/10.1007/s00401-006-0127-z. - [72] Landau S, Ward TJ, Murphy A, Jagust W. Flortaucipir (AV-1451) processing methods. Published online 2021.. - [73] Wang X, Sundermann EE, Buckley RF, Reas ET, McEvoy LK, Banks SJ. Sex differences in the associations of obesity with tau, amyloid PET, and cognitive outcomes in preclinical Alzheimer’s disease: cross-sectional A4 study. J Alzheimers Dis 2023;95(2):615–24. https://doi.org/10.3233/JAD-230466. - [74] Nieminen P. Application of standardized regression coefficient in meta-analysis. BioMedInformatics 2022;2(3):434–58. https://doi.org/10.3390/ biomedinformatics2030028. - [75] Spira AP, An Y, Peng Y, et al. APOE genotype and nonrespiratory sleep parameters in cognitively intact older adults. Sleep 2017;40(8):zsx076. https://doi.org/ 10.1093/sleep/zsx076. - [76] DiNapoli EA, Gebara MA, Kho T, et al. Subjective-objective sleep discrepancy in older adults with MCI and subsyndromal depression. J Geriatr Psychiatry Neurol 2017;30(6):316–23. https://doi.org/10.1177/0891988717731827. - [77] Tadokoro K, Ohta Y, Hishikawa N, et al. Discrepancy of subjective and objective sleep problems in Alzheimer’s disease and mild cognitive impairment detected by a home-based sleep analysis. J Clin Neurosci 2020;74:76–80. https://doi.org/ 10.1016/j.jocn.2020.01.085. - [78] Andr´e C, Martineau-Dussault M`E, Baril AA, et al. Reduced rapid eye movement sleep in late middle-aged and older apolipoprotein E ε4 allele carriers. Sleep 2024; 47(7):zsae094. https://doi.org/10.1093/sleep/zsae094. - [79] Huang S, Wang D, Zhou H, et al. Neuroimaging consequences of cerebral small vessel disease in patients with obstructive sleep apnea–hypopnea syndrome. Brain Behav 2019;9(8). https://doi.org/10.1002/brb3.1364. - [80] Mukherjee U, Sehar U, Brownell M, Reddy PH. Mechanisms, consequences and role of interventions for sleep deprivation: focus on mild cognitive impairment and Alzheimer’s disease in elderly. Ageing Res Rev 2024;100:102457. https://doi.org/ 10.1016/j.arr.2024.102457. - [81] Parhizkar S, Holtzman DM. APOE mediated neuroinflammation and neurodegeneration in Alzheimer’s disease. Semin Immunol 2022;59:101594. https://doi.org/10.1016/j.smim.2022.101594. - [82] Leng Y, Ackley SF, Glymour MM, Yaffe K, Brenowitz WD. Genetic risk of Alzheimer’s disease and sleep duration in non-demented elders. Ann Neurol 2021; 89(1):177–81. https://doi.org/10.1002/ana.25910. - [83] Chapuis J, Hansmannel F, Gistelinck M, et al. Increased expression of BIN1 mediates Alzheimer genetic risk by modulating tau pathology. Mol Psychiatry 2013;18(11):1225–34. https://doi.org/10.1038/mp.2013.1. - [84] Tsapanou A, Gu Y, O’Shea DM, et al. Sleep quality and duration in relation to memory in the elderly: initial results from the Hellenic Longitudinal Investigation of Aging and Diet. Neurobiol Learn Mem 2017;141:217–25. https://doi.org/ 10.1016/j.nlm.2017.04.011. - [85] Siddarth P, Thana-udom K, Ojha R, et al. Sleep quality, neurocognitive performance, and memory self-appraisal in middle-aged and older adults with memory complaints. Int Psychogeriatr 2021;33(7):703–13. https://doi.org/ 10.1017/S1041610220003324. - [86] Baril AA, Beiser AS, Sanchez E, et al. Insomnia symptom severity and cognitive performance: moderating role of APOE genotype. Alzheimers Dement 2022;18(3): 408–21. https://doi.org/10.1002/alz.12405. - [87] Grau-Rivera O, Operto G, Falc´on C, et al. Association between insomnia and cognitive performance, gray matter volume, and white matter microstructure in cognitively unimpaired adults. Alzheimers Res Ther 2020;12:4. https://doi.org/ 10.1186/s13195-019-0547-3.