Amyloid-linked versus age-driven copathologies in Alzheimer’s dementia: differential associations with APOE ε4

Amyloid-linked versus age-driven copathologies in Alzheimer’s dementia: differential associations with APOE ε4 | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Amyloid-linked versus age-driven copathologies in Alzheimer’s dementia: differential associations with APOE ε4 Íñigo Rodríguez-Baz, Lidia Vaqué-Alcázar, Lucía Maure-Blesa, Lucía Pertierra, and 19 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9044264/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract The mechanisms by which apolipoprotein E ( APOE ) drives copathologies in established Alzheimer’s disease (AD) dementia via amyloid-dependent versus age-driven pathways remain unresolved. Analyzing data from 11,897 autopsied individuals from the National Alzheimer's Coordinating Center, with copathology analyses restricted to amyloid-positive AD dementia, we show that APOE effects followed two distinct trajectories. Cerebral amyloid angiopathy exhibited a striking ε4 dose-response (OR = 5.76, 95% CI: 4.20–7.96, p < 0.001; for ε4/ε4 compared to ε3/ε3), whereas arteriolosclerosis and atherosclerosis risk increased with age, independent of APOE haplotype. Lewy body pathology showed modest APOE associations restricted to limbic/amygdalar-predominant forms and was related to dementia duration, suggesting AD-mediated secondary synucleinopathy. TDP-43 pathology was associated with chronological age, demonstrating regional progression with minimal APOE dependence. These findings suggests that in amyloid-positive AD dementia, APOE ε4 selectively amplifies amyloid-related pathology, particularly cerebral amyloid angiopathy, while other copathologies accumulate through age-driven, APOE haplotype-independent processes. Health sciences/Neurology/Neurological disorders/Dementia/Alzheimer's disease Health sciences/Diseases/Neurological disorders/Dementia Biological sciences/Genetics/Clinical genetics/Disease genetics Biological sciences/Neuroscience/Genetics of the nervous system Biological sciences/Neuroscience/Diseases of the nervous system/Dementia/Alzheimer's disease Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Alzheimer's disease (AD) is the most common cause of dementia, neuropathologically characterized by the accumulation of amyloid plaques and tau neurofibrillary tangles. However, previous studies have demonstrated that the majority of patients with dementia present other neuropathologies at autopsy, frequently encompassing Lewy body pathology, TAR DNA-binding protein 43 (TDP-43) proteinopathy, cerebral amyloid angiopathy (CAA), and various forms of cerebrovascular disease. These studies have traditionally examined copathologies in dementia patients as largely independent contributors, often interpreted in the context of aging 1,2 . However, evidence from genetically determined AD, both in Autosomal Dominant AD (ADAD) 3 and Down syndrome-associated AD (DSAD) 4 , have shown that other neuropathological hallmarks beyond amyloid and tau are frequent despite the young age at which dementia manifests in these populations. The apolipoprotein E ( APOE ) gene represents the strongest common genetic risk factor for sporadic AD. The three major allelic variants (ε2, ε3, and ε4) confer differential risk, with ε4 associated with increased susceptibility and ε2 with relative protection against AD 5,6 . Beyond its established role in amyloid metabolism, some studies have shown that carriers of the APOE ε4 allele have more α-synuclein 7,8 and TDP-43 pathologies 9 , as well as increased CAA burden 10 . Recent evidence suggests that APOE ε4 homozygosity represents a genetically determined form of AD with near-complete penetrance of AD biology 11 . However, despite the much higher prevalence and the earlier presentation of AD symptoms in APOE ε4/ε4, when analyzing AD patients at dementia stage with amyloid confirmation, there were no differences across haplotypes, but younger individuals had higher tau positron emission tomography (PET) uptake 11 . This has been confirmed in other studies that have shown age-associated increases in tau PET burden in cognitively unimpaired individuals but decreases with age in patients with AD dementia 12,13 . Age is indeed a fundamental determinant of both neurodegenerative and vascular pathologies. Furthermore, the temporal evolution of copathologies in relation to dementia duration remains poorly understood. Whether APOE haplotype modifies the severity or distribution of Lewy body, TDP-43, and vascular pathologies in relation to dementia duration has not been systematically examined. Disentangling the relative contributions of APOE haplotype and chronological age in neuropathologically confirmed AD dementia patients to diverse neuropathological outcomes is therefore critical for understanding disease mechanisms. The objectives of this study were to evaluate whether APOE haplotypes are associated with neuropathological copathologies in individuals with neuropathologically confirmed AD dementia, while accounting for the potential confounding and modifying effects of age and dementia duration. Specifically, we aimed to: (1) examine whether APOE haplotypes exhibit different burden and regional specificity in their associations with Lewy body and TDP-43 pathologies; and (2) determine the relative contributions of APOE haplotype, age, and dementia duration to the severity of CAA, arteriolosclerosis, and atherosclerosis. By addressing these questions, this study seeks to refine our understanding of the role of APOE in shaping the heterogeneous neuropathological landscape of AD, with implications for mechanistic understanding, risk stratification, and the design of future clinical trials and therapeutic development. Results 2.1 Sample characteristics and distribution of APOE haplotype We analyzed neuropathological and clinical data from 11,897 autopsied individuals enrolled across National Institute on Aging-funded Alzheimer's Disease Research Centers and aggregated in the National Alzheimer's Coordinating Center (NACC) repository. Table 1 summarizes the demographic, clinical, and neuropathological features of the 11,897 individuals included in the study. The most common APOE haplotypes were ε3/ε3 (47.5%), followed by ε3/ε4 (33.4%) and ε4/ε4 (8.0%). Compared with ε3/ε3 carriers, both ε3/ε4 and ε4/ε4 carriers showed earlier ages at dementia diagnosis (median -1.7 and -5.6 years, respectively) and death (-2.2 and -6.7 years; all p < 0.001). Overall, 48.3% of the cohort were female and 89.4% were White. At the time of death, 73.1% had received a dementia diagnosis, and 57.6% had been clinically classified as having AD. Moderate or frequent neuritic plaques were present in 63.6% of participants. Amyloid positivity rates varied markedly across haplotypes in the full cohort. Among participants with a clinical diagnosis of AD dementia, amyloid positivity (defined as moderate or frequent neuritic plaques) was present in 88.7% of ε4/ε4 carriers, 79.3% of ε3/ε4 carriers, and 68.5% of ε3/ε3 carriers (Supplementary Figure 1, Supplementary Table 1). Similarly, concordant clinicopathological AD diagnosis was significantly more common among ε3/ε4 and ε4/ε4 carriers compared to ε3/ε3 carriers (Odds ratio [OR] = 2.73, with 95% confidence interval [95% CI] = 2.41–3.10 and OR 5.59, 95% CI 4.46–7.07, respectively; both p<0.001; Supplementary Table 2). To ensure analyses focused on neuropathologically confirmed AD and to avoid confounding by differential amyloid rates, subsequent copathology analyses (sections 2.2 and 2.3) were restricted to AD dementia cases with confirmed amyloid deposition. This restriction ensured that observed APOE effects on copathologies reflected associations within established AD pathology rather than differences in AD diagnostic accuracy across haplotypes. Of note, in both clinically diagnosed and amyloid-positive AD dementia, the prevalence of Braak stage V–VI decreased with increasing age at death, with the exception of ε4/ε4 carriers, who showed no appreciable age effect and consistently the highest prevalence of Braak stage V–VI across all age groups (Supplementary Figure 2, Supplementary Table 3). 2.2 Lewy body and TDP-43 pathologies Lewy body pathology was prevalent among individuals with amyloid-positive AD dementia, affecting more than one-third of ε3/ε3 carriers and over half of ε4/ε4 carriers overall (Figure 1, left panels). Prevalence increased with APOE ε4 dose, from 37.2% in ε3/ε3 to 46.2% in ε3/ε4 (OR = 1.39, 95% CI: 1.14–1.71, p = 0.001) and 52.6% in ε4/ε4 carriers (OR = 1.77, 95% CI: 1.33–2.36, p < 0.001). Across haplotypes, the most common regional pattern was limbic/amygdala-predominant involvement, followed by neocortical pathology, whereas brainstem-predominant Lewy pathology was comparatively infrequent. Regional distributions differed modestly by APOE haplotype. Compared with ε3/ε3 carriers, ε3/ε4 carriers had higher prevalence of olfactory bulb/region-unknown pathology (5.9% vs 3.1%; OR = 2.27, 95% CI: 1.14–4.93, p = 0.018) and neocortical involvement (14.5% vs 11.0%; OR = 1.50, 95% CI: 1.11–2.04, p = 0.008). ε4/ε4 carriers showed greater limbic/amygdalar involvement than ε3/ε3 carriers (29.0% vs 20.2%; OR = 1.63, 95% CI: 1.21–2.19, p = 0.001). However, despite these differences, no consistent dose-dependent shift toward more extensive regional involvement was observed across ε4 allele groups (Supplementary Table 4A). Lewy body pathology showed no significant association with age at death across APOE haplotypes (all likelihood ratio chi-squared tests [LRχ²] ≤ 2.54, p ≥ 0.111), consistent with the relatively stable prevalence across age strata shown in Figure 1. In regional analyses, increasing age was associated with lower amygdala involvement in ε3/ε3 carriers (OR per 10 years = 0.80, 95% CI: 0.68–0.93, p = 0.004), but no other significant age-related changes were observed in other regions or haplotypes (Supplementary Figure 3; Supplementary Table 5A). In contrast, Lewy pathology was associated with dementia duration. Longer duration was linked to greater limbic/amygdalar involvement in ε3/ε3 and ε3/ε4 carriers after adjustment for sex, education, age at dementia onset and Alzheimer's disease neuropathologic change (ADNC) classification (Figure 2, left panels; Supplementary Table 6A). This relationship was not observed in ε4/ε4 carriers, in whom (mainly limbic/amygdalar) Lewy pathology was already highly prevalent at shorter dementia durations, suggesting earlier or more rapid accumulation in this group. TDP-43 pathology was even more prevalent than Lewy pathology across all APOE haplotypes (Figure 1, right panels). Overall prevalence exceeded three-quarters of individuals in each group, except ε2/εX, and was higher in ε4/ε4 compared with ε3/ε3 carriers (81.9% vs 76.4%; OR = 1.99, 95% CI: 1.16–3.50, p = 0.012). Across haplotypes, amygdala involvement was the dominant regional pattern, with hippocampal, entorhinal/inferior temporal, and neocortical involvement occurring in progressively smaller proportions. Unlike Lewy pathology, TDP-43 showed a strong relationship with age at death. Overall prevalence increased with age in ε3/ε3 and ε3/ε4 carriers (both LRχ² ≥ 10.74, p ≤ 0.001), accompanied by regional progression beyond the amygdala. Specifically, increasing age was associated with greater entorhinal/inferior temporal involvement in ε3/ε3 carriers (OR per 10 years = 1.51, 95% CI: 1.17–1.99, p = 0.002) and greater neocortical involvement in ε3/ε4 carriers (OR per 10 years = 2.01, 95% CI: 1.38–3.01, p < 0.001). Conversely, amygdala-only involvement declined with age in ε3/ε4 (OR per 10 years = 0.73, 95% CI: 0.55–0.99, p = 0.044) and ε4/ε4 carriers (OR per 10 years = 0.57, 95% CI: 0.31–0.98, p = 0.042) (Supplementary Figure 4; Supplementary Table 5B). Despite its strong age dependency, TDP-43 pathology showed a weaker association with dementia duration. After adjustment for age at dementia onset, sex, education, and ADNC classification, longer duration remained associated with higher TDP-43 prevalence across most haplotypes (all LRχ² ≥ 5.04, p ≤ 0.025), although not significantly in ε4/ε4 carriers (LRχ² = 3.57, p = 0.059) (Figure 2, right panels). Regionally, longer duration was associated with increased entorhinal involvement in ε2/εX carriers and greater neocortical involvement in ε3/ε4 carriers, but these effects were modest compared with the age-related patterns (Supplementary Table 6B). 2.3 Vascular-related pathology CAA exhibited a striking APOE ε4 gene-dose effect, with increasing prevalence and severity across haplotypes. Compared with ε3/ε3 carriers, moderate-to-severe CAA was more common in ε3/ε4 carriers (OR = 1.54, 95% CI: 1.25–1.89, p < 0.001) and markedly elevated in ε4/ε4 carriers (OR = 5.76, 95% CI: 4.20–7.96, p < 0.001), with APOE ε4 homozygotes showing the highest proportions of both moderate and severe CAA (Supplementary Table 7A). This shift toward moderate and severe CAA was evident across all age strata, with ε4/ε4 carriers already showing a high burden at younger ages (Figure 3, left panels). Age further modified CAA burden in ε4 carriers. Among ε4/ε4 individuals, CAA prevalence increased with advancing age (OR per 10 years = 1.69, 95% CI: 1.21–2.38, p = 0.002), a pattern also observed for severe CAA in ε3/ε4 carriers (Supplementary Table 8A; Supplementary Figure 5, upper panels). In contrast, age-related increases were less pronounced in ε3/ε3 carriers. Dementia duration showed a distinct pattern. After full adjustment for age at dementia onset, sex, education, and ADNC classification, longer dementia duration was associated with increasing CAA prevalence in ε3/ε3 (OR per year = 1.09, 95% CI: 1.03–1.15, p = 0.002) and ε3/ε4 carriers (OR per year = 1.06, 95% CI: 1.02–1.11, p = 0.008; Figure 4, left panels). In ε4/ε4 carriers, however, longer duration was associated with lower prevalence of severe CAA (OR per year = 0.91, 95% CI: 0.84–0.98, p = 0.016), suggesting that severe CAA accumulates early in this haplotype (Supplementary Table 9A) and might be associated with survival. In contrast to CAA, cerebral arteriolosclerosis and atherosclerosis of the circle of Willis showed weaker associations with APOE haplotype and were driven primarily by age (Figure 3, middle and right panels). Arteriolosclerosis prevalence was modestly higher in ε4/ε4 compared with ε3/ε3 carriers (52.7% vs 46.0%; OR = 1.49, 95% CI: 1.10–2.01, p = 0.009), driven mainly by severe pathology (OR = 1.69, 95% CI: 1.12–2.55, p = 0.013). By contrast, moderate-to-severe atherosclerosis prevalence did not differ significantly across APOE haplotypes, although ε4/ε4 carriers showed higher rates of mild atherosclerosis (Supplementary Table 7B-C). Age was the dominant determinant of cerebral arteriolosclerosis and atherosclerosis of the circle of Willis. Prevalence of arteriolosclerosis increased with age across all haplotypes (all LRχ² ≥ 5.96, p ≤ 0.015), with parallel increases in moderate and severe strata (Supplementary Table 8B). Similarly, atherosclerosis prevalence increased with age in all haplotypes (all LRχ² ≥ 7.79, p ≤ 0.005), accompanied by progressive shifts toward greater severity (Supplementary Figure 5, middle and bottom panels). Associations with dementia duration were comparatively modest. In fully adjusted models, arteriolosclerosis prevalence increased with longer disease duration only in ε3/ε3 carriers (LRχ² = 18.84, p < 0.001), driven mainly by moderate pathology (Figure 4, middle panels; Supplementary Table 9B). Atherosclerosis prevalence increased with longer dementia duration in all haplotypes except ε2/εX (all LRχ² ≥ 9.04, p ≤ 0.003), although the magnitude of these effects was smaller than those observed with age (Figure 4, right panels; Supplementary Table 9C). Discussion In this large autopsy-based cohort of AD dementia with confirmed amyloid positivity, we found that APOE haplotype related to copathologies in a highly selective manner following two broad trajectories: amyloid-linked/ APOE -dependent versus age-driven/ APOE -independent. Lewy body and TDP-43 pathologies were both extremely frequent, but they diverged in their determinants: Lewy pathology showed modest APOE -related differences concentrated in limbic/amygdalar regions and was largely age-independent, yet it increased with dementia duration, positioning it as an “intermediate” process that may accrue alongside AD progression. TDP-43, in contrast, was less APOE -dependent and tracked primarily with age, with evidence of regional progression and comparatively weaker links to dementia duration. The vascular findings then provided the clearest dissociation: CAA was strongly ε4 dose-dependent, whereas non-CAA vascular disease (arteriolosclerosis and large-vessel atherosclerosis) was predominantly age-driven with limited haplotype effects. Notably, dementia duration related more to amyloid-linked pathology (CAA, and to some extent Lewy pathology) than to non-CAA vascular disease, reinforcing the interpretation that APOE ε4 chiefly amplifies amyloid-related mechanisms once dementia is established, while other copathologies accumulate largely through aging and disease-stage processes that are relatively APOE haplotype-independent. APOE strongly shapes core AD pathology even in dementia-stage disease. APOE ε4/ε4 carriers were enriched for clinicopathological AD, consistent with near-complete penetrance of AD biology 11 . Furthermore, the age-invariant frequency of Braak V-VI in ε4/ε4 suggests early tau saturation, in line with tau-PET evidence 12,13,24 . Conversely APOE ε2 confers protection to AD pathology even in clinical AD dementia cases 5 . To reduce sample-composition bias when assessing copathologies, we focused our primary analyses on AD dementia with neuritic plaque-confirmed amyloid positivity, ensuring APOE effects were evaluated within established AD biology. Lewy body pathology was more frequent overall in ε3/ε4 and ε4/ε4 carriers. However, the most informative signal lied in the topography. The only consistent APOE -related pattern was limbic/amygdalar involvement, which was higher in ε4/ε4 than ε3/ε3 carriers (29.0% vs 20.2%), whereas associations in olfactory and neocortical regions were less consistent. This topographic selectivity places Lewy pathology in an “intermediate” position between the amyloid-linked and age-driven axes: it showed modest haplotype effects concentrated in a vulnerable region, with no robust (or decreasing) age gradient. Importantly, there was an amydgalar accumulation (but not neocortical) Lewy pathology with dementia duration. The concordant effects of APOE and dementia duration support the view that amygdala-predominant Lewy pathology often represents secondary, AD-mediated α-synucleinopathy rather than primary synuclein disease 25 . Notably, this amygdala-centered pattern is consistent with observations in genetically determined AD, including autosomal-dominant AD (ADAD) and Down syndrome-associated AD (DSAD), where Lewy pathology, when present, also shows a preferential limbic/amygdalar accumulation in the context of established AD biology 26,27 . By contrast, the lack of a consistent APOE dose-response for neocortical (diffuse) Lewy pathology is compatible with a distinct pathological trajectory more closely linked to the Lewy body dementia phenotype and a more aggressive course that can be relatively independent of AD burden 28 . TDP-43 pathology, by comparison, mapped much more cleanly onto an age-driven process. While ε4/ε4 carriers had slightly higher prevalence than ε3/ε3, we did not observe a significant ε3/ε4 effect within this amyloid-positive AD dementia cohort, differing from reports of ε4 dose-dependent associations in broader samples 9,29 . This discrepancy is plausibly explained by conditioning on established AD dementia with confirmed amyloidosis, which reduces heterogeneity but can also attenuate haplotype contrasts when baseline TDP-43 burden is already high. Consistent with this, ε3/ε4 effects on TDP-43 pathology emerged when we evaluated all-cause and unconfirmed-amyloid dementia groups (Supplementary Table 11). The dominant signals were again topographic and temporal: TDP-43 increased with age across most haplotypes and showed a shift from amygdala-predominant involvement toward broader limbic and neocortical spread, with dementia duration contributing more modestly than age. This pattern closely matches the defining clinical-pathological trajectory of Limbic-predominant age-related TDP-43 encephalopathy (LATE) 30 , supporting the interpretation that in amyloid-positive AD dementia, TDP-43 accumulation is driven primarily by aging and progressive regional spread, rather than by APOE haplotype per se. This age-driven, APOE -independent profile is further supported by genetic evidence showing that TDP-43 accumulation in aging brains follows a distinct genetic risk architecture: TMEM106B and GRN variants associate with TDP-43 pathology and hippocampal sclerosis independently of APOE haplotype 31 , and TNIP1, recently identified as a risk factor for frontotemporal lobar degeneration with TDP-43 pathology 32 , has been proposed to explain the TNIP1-AD genetic overlap through comorbid TDP-43 rather than amyloid mechanisms 33 . The amygdala’s broader vulnerability to misfolded protein accumulation across disorders provides a plausible substrate for this convergence 34 . One parsimonious explanation is that the amygdala, together with other phylogenetically older cortices, acts as an early “landing zone” for multiple age- and AD-linked proteinopathies, with stereotyped staging schemes often beginning in limbic structures (e.g., LATE/TDP-43) before extending cortically 30 . In AD specifically, previous works had shown that α-synuclein deposition frequently shows an amygdala-predominant pattern, suggesting region-specific susceptibility rather than uniform neocortical spread 35 . Mechanistically, this convergence could reflect a combination of selective neuronal/architectural vulnerability and interaction between pathologies, including cross-seeding or permissive proteostatic/inflammatory milieus generated by AD pathology, supported by experimental work demonstrating tau-α-synuclein cross-seeding and prion-like propagation 36 . Consistent with this, AD cases with amygdala-predominant α-synuclein have been reported to show particularly high co-occurring TDP-43 burden and even co-localization of aggregates within limbic neurons, reinforcing the idea of a limbic “co-proteinopathy hub” rather than independent parallel processes 37 . The strong APOE -related gradient we observed for CAA is well aligned with prior neuropathology literature showing that APOE ε4 preferentially amplifies amyloid deposition in the vessel wall 10 . In our amyloid-positive AD dementia cohort, haplotype outweighed both age and dementia duration as a correlate of CAA burden: ε4/ε4 carriers had an almost six-fold higher odds of moderate-to-severe CAA compared with ε3/ε3, and severity shifted upward with APOE ε4 gene dose. This pattern mirrors observations in clinicopathologically confirmed AD in which ε4 is more tightly coupled to severe CAA than to chronological age or disease duration 38 . The contrasting duration associations, positive in ε3/ε3 and ε3/ε4 but inverse for severe CAA in ε4/ε4, suggest that CAA is tightly coupled to AD biology; the ε4/ε4 pattern may be best explained by earlier attainment of severe CAA and consequent selection among longer survivors, as CAA is associated with increased mortality risk 39 . In contrast, non-CAA vascular lesions aligned predominantly with an age-driven trajectory, consistent with the progressive accumulation of cerebrovascular risk factors not captured in our analyses, with substantially weaker APOE effects than those seen for CAA. Cerebral arteriolosclerosis showed at most a modest enrichment in ε4/ε4 carriers relative to ε3/ε3, but the dominant signal across haplotypes was a progressive shift toward greater severity with advancing age. Differences from population-based reports that observed weaker or null APOE associations 40 may reflect both cohort composition (our restriction to neuropathologically confirmed AD dementia with amyloid positivity) and analytic choices (separating ε4 homozygotes rather than combining all ε4 carriers). Circle of Willis atherosclerosis similarly showed minimal haplotype-related differences, yet increased steadily with age across haplotypes, again aligning with prior neuropathological and cardiovascular pathology reports 40,41 . The additional, smaller association with dementia duration observed for atherosclerosis in our AD dementia sample may reflect cumulative exposure to systemic vascular risk and comorbidity during the clinical course, rather than an AD-specific genetic effect. Overall, these results sharpen a key distinction within amyloid-positive AD dementia: CAA behaves as an APOE /amyloid-linked vasculopathy, whereas arteriolosclerosis and large-vessel atherosclerosis largely reflect age- and risk-related cerebrovascular injury, with only limited modulation by APOE haplotype. This study has important strengths and some limitations that inform future research. The large autopsy-confirmed sample and standardized neuropathological assessments provide exceptional statistical power to detect APOE haplotype-specific effects across a broad spectrum of copathologies and reduce the diagnostic uncertainty inherent in purely clinical or biomarker-based cohorts. However, the cohort lacks diversity, being predominantly White, which limits generalizability to population-based and ancestrally diverse samples. In addition, the cross-sectional nature of autopsy data and evolving NACC collection protocols, including inter-site variability in pathological ratings, constrain inferences about temporal trajectories of pathology despite the use of harmonized criteria. Additionally, cerebrovascular risk factors such as hypertension, diabetes, or dyslipidemia were not included as covariates in our analyses, which may partially account for the age-driven vascular pathology patterns observed and should be addressed in future studies. Furthermore, our focus on amyloid-positive AD dementia, while ensuring diagnostic homogeneity, precludes assessment of APOE effects in amyloid-negative dementia or preclinical stages, where associations may differ. However, as shown in the supplementary material, the NACC database is not a population-based cohort and is highly enriched for individuals with dementia. Therefore, it might not be well-suited to address these questions. These considerations underscore the need for longitudinal, multimodal studies with serial fluid and imaging biomarkers as well as postmortem validation in more diverse and population-based cohorts. In conclusion, these data support a parsimonious model in which copathologies in amyloid-positive AD dementia distribute across distinct but overlapping axes. One axis is APOE /amyloid-linked, dominated by CAA severity and accompanied by selective limbic vulnerability, where Lewy pathology shows its most consistent (albeit modest) APOE signal. The other axis is age-driven and relatively APOE -independent, encompassing non-CAA vascular disease and TDP-43 pathology. This framework has practical implications for interpreting clinicopathologic variability and suggests that, in late-stage disease, trial stratification may benefit most from capturing amyloid-related vascular vulnerability and limbic-predominant co-proteinopathy patterns, rather than assuming broad APOE effects across all copathologies. Methods 4.1 Study cohort We analyzed data from the March 2025 freeze of the NACC repository ( https://www.naccdata.org/ ), which aggregates de-identified longitudinal clinical and neuropathological data from National Institute on Aging–funded Alzheimer’s Disease Research Centers (ADRCs). ADRCs enroll participants across the cognitive spectrum, from cognitively unimpaired individuals to those with dementia, although most participants have AD dementia and are typically recruited after symptom onset. All contributing ADRCs obtained local institutional review board approval and written informed consent from participants or their legal representatives, and procedures adhered to the Declaration of Helsinki. NACC maintains a harmonized clinical database based on the Uniform Data Set (UDS) and a complementary neuropathology database based on standardized Neuropathology Forms. For this study, we included participants who had died and undergone brain autopsy and excluded individuals without APOE genotyping or carrying a known pathogenic variant associated with familial AD (NPPDXP), frontotemporal dementia (NPPDXQ), or Down syndrome (NACCDOWN). 4.2 Demographic, cognitive, and genetic variables We extracted standard demographic variables from NACC. Sex (SEX) was coded as male or female. Race (RACE) was classified as White, Black or African American, American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, Asian, Other, or Unknown. Educational attainment was quantified as total years of formal education (EDUC). APOE haplotype was derived from the variable NACCAPOE and categorized as ε2/ε2, ε2/ε3, ε3/ε3, ε2/ε4, ε3/ε4, or ε4/ε4, with ε3/ε3 used as the reference group in all comparative analyses, given that it is the most common haplotype in the general population. Ages at diagnosis of mild cognitive impairment (MCI), dementia, and at death were calculated by combining the date of the visit at which MCI or dementia was first diagnosed (year, month, and day) with the recorded date of birth and date of death. Participants were then classified according to their last cognitive status as having dementia, MCI, or cognitively unimpaired (CU). Within the UDS, the Etiologic Diagnoses section allows clinicians to indicate, for each potential cause of cognitive impairment, whether it is the primary etiology, a contributing factor, or not contributing; each category is coded as present or absent based on clinical judgment and available diagnostic information. For this study, we focused on whether AD was recorded as the primary etiologic diagnosis of cognitive impairment. Participants were therefore classified at the visit closest to death as having clinical AD (AD as the primary etiology) or clinical non-AD (a non-AD primary etiology). 4.3 Neuropathological variables Neuropathological data were obtained from NACC Neuropathology Forms (versions 10–11) completed after autopsy 14 . ADNC was obtained from variable NPADNC, which integrates three components into a composite score (Not, Low, Intermediate, or High): Thal phase for amyloid plaque distribution (A score) 15 , Braak stage for neurofibrillary tangle burden (B score) 16 , and the Consortium to Establish a Registry for Alzheimer's Disease (CERAD) neuritic plaque density (C score) 17 . We used CERAD semi-quantitative assessments of neuritic plaque burden (NACCNEUR: No, Sparse, Moderate, or Frequent neuritic plaques) and Braak stage for neurofibrillary tangle burden (NACCBRAA: Stage 0–VI; recoded as None or stages I–II, III–IV, V–VI). Given the central role of amyloid biomarkers in research and clinical practice, CERAD neuritic plaque severity was collapsed into a binary amyloid status: individuals with "No" or "Sparse" neuritic plaques were considered amyloid-negative, and those with "Moderate" or "Frequent" plaques as amyloid-positive. This dichotomization was chosen to approximate amyloid PET positivity and align with current AD neuropathologic change frameworks 18–20 . Lewy body pathology was assessed using the NACC-derived variable NACCLEWY, which classifies cases as: no Lewy body pathology, brainstem-predominant, limbic or amygdala-predominant, neocortical, or Lewy bodies present in unspecified region/olfactory bulb. These categories are broadly consistent with the DLB Consortium neuropathological classification framework 21 , though NACC combines limbic and amygdala-predominant patterns into a single category. TDP-43 immunoreactive inclusions were assessed across five brain regions, spinal cord (NPTDPA), amygdala (NPTDPB), hippocampus (NPTDPC), entorhinal/inferior temporal cortex (NPTDPD), and neocortex (NPTDPE), as presence or absence; regional distribution patterns are consistent with LATE-NC staging criteria 22,23 . Cerebrovascular pathologies were graded using NACC semi-quantitative scales (none, mild, moderate, severe), including cerebral amyloid angiopathy (NACCAMY), arteriolosclerosis (NACCARTE), and Circle of Willis atherosclerosis (NACCAVAS) 14 . 4.5 Statistical analyses Categorical variables were summarized as counts and percentages by APOE haplotype (i.e., ε2/εX, ε3/ε3, ε3/ε4, ε4/ε4), and continuous variables were reported as medians with interquartile ranges. To examine associations between APOE haplotype and key clinical and pathological outcomes at death, we fitted binary logistic regression models using APOE ε3/ε3 as the reference category. Given the low representation of APOE ε2/ε2 (n = 44) and ε2/ε3 (n = 948), these groups were combined into a single category (ε2/εX) for neuropathological analyses to ensure adequate statistical power. Individuals with the ε2/ε4 haplotype (n = 331) were excluded due to the opposing effects of APOE ε2 and ε4 alleles on AD risk and the small sample size. Neuropathological outcomes were analyzed using regression models. For binary outcomes (presence or absence of pathology), we applied logistic regression. For ordered severity measures (Lewy body pathology distribution, TDP-43 pathology stage, and vascular pathology severity grades), we used proportional odds ordinal logistic regression. Models were fitted unadjusted and then progressively adjusted, first for sex and years of education, then additionally for age at death or age at dementia diagnosis as appropriate, and finally with the addition of AD neuropathologic change (ADNC) classification. ORs with 95% CIs were estimated from these models. Global effects of APOE haplotype, age at death, and dementia duration were assessed using LRχ². Given that progressive model adjustments represent sequential refinements of the same associations rather than independent tests, we did not apply corrections for multiple comparisons. Interpretation emphasized both effect magnitude (ORs with 95% CIs) and statistical significance. Analyses of Lewy body pathology, TDP-43 pathology, CAA, arteriolosclerosis, and Circle of Willis atherosclerosis were restricted to participants with a clinical diagnosis of AD dementia and neuropathologically confirmed amyloid positivity (moderate or frequent neuritic plaques by CERAD criteria). Age-related analyses used the following age-at-death categories: <65, 65–69, 70–74, 75–79, 80–84, 85–89, ≥90 years. Dementia duration was categorized as <2, 2–3, 4–6, 6–8, and ≥10 years to capture clinically distinct phases of disease progression while maintaining sufficient cases per group. Distributions of neuropathological features across these categories were assessed using chi-squared tests (χ²). Additional analyses are described in the Supplementary Methodology, including: (1) cohort characterization and rationale for restricting analyses to amyloid-positive AD dementia cases (Supplementary Figure 1, Supplementary Tables 1A-B and 2); (2) Braak neurofibrillary tangle staging across multiple diagnostic groups (all-cause dementia, AD dementia, and amyloid-positive AD dementia) (Supplementary Figure 2, Supplementary Tables 3 and 4); (3) APOE × age interaction testing for regional Lewy body, TDP-43, and cerebrovascular pathologies (Supplementary Figures 3, 4 and 5), (4) sex-stratified sensitivity analyses for all neuropathologies (Supplementary Table 10), and (5) TDP-43 pathology across diagnostic groups (Supplementary Table 11). All statistical analyses were performed using R (version 4.2.2). Tests were two-sided, and P < 0.05 was considered statistically significant. Declarations Data availability The data that support the findings of this study are available from the National Alzheimer's Coordinating Center (NACC; https://www.naccdata.org/) upon registration and approval. Acknowledgements The NACC database is funded by NIA/NIH Grant U24 AG072122. NACC data are contributed by the NIA-funded ADRCs: P30 AG062429 (PI James Brewer, MD, PhD), P30 AG066468 (PI Oscar Lopez, MD), P30 AG062421 (PI Bradley Hyman, MD, PhD), P30 AG066509 (PI Thomas Grabowski, MD), P30 AG066514 (PI Mary Sano, PhD), P30 AG066530 (PI Helena Chui, MD), P30 AG066507 (PI Marilyn Albert, PhD), P30 AG066444 (PI John Morris, MD), P30 AG066518 (PI Jeffrey Kaye, MD), P30 AG066512 (PI Thomas Wisniewski, MD), P30 AG066462 (PI Scott Small, MD), P30 AG072979 (PI David Wolk, MD), P30 AG072972 (PI Charles DeCarli, MD), P30 AG072976 (PI Andrew Saykin, PsyD), P30 AG072975 (PI David Bennett, MD), P30 AG072978 (PI Neil Kowall, MD), P30 AG072977 (PI Robert Vassar, PhD), P30 AG066519 (PI Frank LaFerla, PhD), P30 AG062677 (PI Ronald Petersen, MD, PhD), P30 AG079280 (PI Eric Reiman, MD), P30 AG062422 (PI Gil Rabinovici, MD), P30 AG066511 (PI Allan Levey, MD, PhD), P30 AG072946 (PI Linda Van Eldik, PhD), P30 AG062715 (PI Sanjay Asthana, MD, FRCP), P30 AG072973 (PI Russell Swerdlow, MD), P30 AG066506 (PI Todd Golde, MD, PhD), P30 AG066508 (PI Stephen Strittmatter, MD, PhD), P30 AG066515 (PI Victor Henderson, MD, MS), P30 AG072947 (PI Suzanne Craft, PhD), P30 AG072931 (PI Henry Paulson, MD, PhD), P30 AG066546 (PI Sudha Seshadri, MD), P20 AG068024 (PI Erik Roberson, MD, PhD), P20 AG068053 (PI Justin Miller, PhD), P20 AG068077 (PI Gary Rosenberg, MD), P20 AG068082 (PI Angela Jefferson, PhD), P30 AG072958 (PI Heather Whitson, MD), P30 AG072959 (PI James Leverenz, MD). We acknowledge the Support for Research Groups funding from the Department of Research and Universities from the Generalitat de Catalunya (2021 SGR 00979). Author contributions I.R.B. and J.F. conceived and designed the study, performed the statistical analyses, and wrote the manuscript. C.K.S., L.M.P., I.A., N.V.F., and R.S.P. supervised neuropathological data. L.V.A., L.M.B., L.P., J.A., L.V., I.B., M.C.I., M.B.S.S., J.S.G., O.D.I., M.R.A., S.S., C.A., I.I.G., D.A., and A.L. contributed to the interpretation of results and critically reviewed the manuscript. All authors read and approved the final version of the manuscript. Competing interests I. Rodríguez-Baz has received personal fees for service on advisory boards, speaker honoraria, or educational activities from Nutricia, Almirall, and Esteve. J. Arranz has received personal fees for service on advisory boards, speaker honoraria, or educational activities from Esteve, Lilly, and Roche Diagnostics. L. Molina-Porcel has provided consultancy services to Biogen. M. Carmona-Iragui has received personal fees for service on advisory boards, speaker honoraria, or educational activities from IMSERSO, Esteve, Lilly, Neuraxpharm, Adium Pharma, and Roche. M.R. Aranha is a co-founder and partner of Masima Soluções em Imagens Médicas LTDA (Porto Alegre, Brazil) and serves as an independent consultant to Ionis Pharmaceuticals. I. Illán-Gala has participated in advisory boards from UCB and Nutricia, and received speaker honoraria from Almirall, Esteve Pharmaceuticals S.A., Kern Pharma, Krka Farmacéutica S.L., Lilly, Nutricia, and Zambon S.A.U. D. Alcolea has received personal fees for advisory board services and/or speaker honoraria from Fujirebio-Europe, Roche, Nutricia, Krka Farmacéutica, Lilly, Zambon S.A.U., Grifols, and Esteve, outside the submitted work. A. Lleó has served as a consultant or on advisory boards for Fujirebio-Europe, Roche, Biogen, Grifols, Novartis, Eisai, Lilly, and Nutricia, outside the submitted work. C.K. Suemoto is funded by the Alzheimer's Association (AARG-20-678884 and 24CBIDR-1185483) and the São Paulo Research Foundation (FAPESP 2024/03917-7). J. Fortea has served on advisory boards, adjudication committees, or as a speaker for Roche, NovoNordisk, Esteve, Biogen, Laboratorios Carnot, Adamed, LMI, Novartis, Lundbeck, AC Immune, Alzheon, Zambon, Lilly, the Spanish Neurological Society, T21 Research Society, Lumind Foundation, Jérôme-Lejeune Foundation, Alzheimer's Association, National Institutes of Health USA, and Instituto de Salud Carlos III. D. Alcolea, A. Lleó, and J. Fortea hold a patent for markers of synaptopathy in neurodegenerative disease (licensed to ADx, EP8382175.0). A. Lleó is co-author of a patent on antibodies for amyloid precursor protein, methods and uses thereof (European priority No. EP25382226). No other competing interests were reported. Funding I. Rodríguez-Baz was supported by Instituto de Salud Carlos III (ISCIII), co-funded by the European Union (CM22/00052, CD25/00223), and by the Alzheimer's Association (AACSF-25-1486364). L. Vaqué-Alcázar was supported by a Sara Borrell postdoctoral fellowship from ISCIII, Spain (CD23/00235). L. Maure-Blesa was supported by ISCIII through a Río Hortega fellowship (CM23/00291), co-funded by the European Union. J. Arranz was supported by ISCIII through a Río Hortega fellowship (CM22/00243), co-funded by the European Union. R. Silva-Paradela was funded by the Alzheimer's Association Capacity Building in International Dementia Research (CBIDR) Program (24CBIDR-1185483) and the Global Brain Health Institute, Alzheimer's Association, and Alzheimer's Society (GBHI ALZ UK-25-1289657). M. Carmona-Iragui was supported by ISCIII (PI18/00335, PI22/00758, ICI23/00032), CIBERNED Program 1 (partly co-funded by FEDER, European Union), the Alzheimer's Association (AARG-22-973966), the Global Brain Health Institute (GBHI_ALZ-18-543740), and the Jérôme Lejeune Foundation (#1913 cycle 2019B; #2425 cycle 2024B). J. Selma-González is supported by the Contratos Predoctorales de Formación en Investigación en Salud program (FI25/00235) associated to project PI24/00598 (I. Illán-Gala), funded by ISCIII, Spain. C. Abdelnour received support from the Susan and Charles Berghoff Foundation and the ARISTOS program, funded by the European Union's Horizon Europe research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 101081334. O. Dols-Icardo receives funding from Fundación Española para el Fomento de la Investigación de la Esclerosis Lateral Amiotrófica (FUNDELA), Fundación HNA, the Alzheimer's Association (AARF-22-924456), and the Fondation Jérôme Lejeune (PDC-2023-51; #202307), and from ISCIII (PI21/01395, PI24/01087), co-funded by FEDER, European Union. I. Illán-Gala was supported by ISCIII (PI21/00791 and PI24/00598), co-funded by FEDER, European Union; the Alzheimer's Association (AACSF-21-850193); the Alzheimer Society (GBHI ALZ UK-21-72097); the Global Brain Health Institute as a senior Atlantic Fellow for Equity in Brain Health; and the Juan Rodés Contract (JR20/0018) from ISCIII, partly funded by the European Social Fund. D. Alcolea received funding from ISCIII (PI18/00435, PI22/00611, PI25/00422, INT23/00048), co-funded by FEDER, European Union, and from the Department of Health of the Generalitat de Catalunya through the PERIS programme (SLT006/17/125, SLT042/25/000034) and the Department of Research and Universities of the Generalitat de Catalunya (2021 SGR 00979). A. Lleó received funding from ISCIII (PI14/1561, PI20/01330), co-funded by FEDER, European Union. C.K. Suemoto was funded by the Alzheimer's Association (AARG-20-678884 and 24CBIDR-1185483) and the São Paulo Research Foundation (FAPESP 2024/03917-7). J. Fortea received funding from ISCIII (PI20/01473, PI23/01786, INT21/00073), co-funded by FEDER, European Union; CIBERNED Program 1; the National Institutes of Health (1R01AG056850-01A1, 3RF1AG056850-01S1, AG056850, R21AG056974, R01AG061566, 1R01AG081394-01, 1R61AG066543-01); the Department of Health of the Generalitat de Catalunya (SLT006/17/00119); Fundación Tatiana Pérez de Guzmán el Bueno (IIBSP-DOW-2020-151); the Horizon 2020 Research and Innovation Programme of the European Union (MES-CoBraD, GA 965422); BrightFocus Foundation; and Life Molecular Imaging (LMI). References Rabinovici, G. D. et al. Multiple comorbid neuropathologies in the setting of Alzheimer’s disease neuropathology and implications for drug development. Alzheimers Dement. 3 , 83–91 (2017). Boyle, P. A. et al. Attributable risk of Alzheimer’s dementia attributed to age-related neuropathologies. Ann. Neurol. 85 , 114–124 (2019). Sepulveda-Falla, D. et al. Comorbidities in early-onset sporadic versus presenilin-1 mutation-associated Alzheimer disease dementia: Evidence for dependency on Alzheimer disease neuropathological changes. J. Neuropathol. Exp. Neurol. 84 , 104–113 (2025). Ichimata, S. et al. Patterns of Mixed Pathologies in Down Syndrome. J. Alzheimers Dis. JAD 87 , 595–607 (2022). Reiman, E. M. et al. Exceptionally low likelihood of Alzheimer’s dementia in APOE2 homozygotes from a 5,000-person neuropathological study. Nat. Commun. 11 , 667 (2020). Li, Z., Shue, F., Zhao, N., Shinohara, M. & Bu, G. APOE2: protective mechanism and therapeutic implications for Alzheimer’s disease. Mol. Neurodegener. 15 , 63 (2020). Dickson, D. W. et al. APOE ε4 is associated with severity of Lewy body pathology independent of Alzheimer pathology. Neurology 91 , e1182–e1195 (2018). Jin, Y. et al. APOE4 exacerbates α-synuclein seeding activity and contributes to neurotoxicity in Alzheimer’s disease with Lewy body pathology. Acta Neuropathol. (Berl.) 143 , 641–662 (2022). Yang, H.-S. et al. Evaluation of TDP-43 proteinopathy and hippocampal sclerosis in relation to APOE ε4 haplotype status: a community-based cohort study. Lancet Neurol. 17 , 773–781 (2018). Nelson, P. T. et al. APOE-ε2 and APOE-ε4 Correlate with Increased Amyloid Accumulation in Cerebral Vasculature. J. Neuropathol. Exp. Neurol. 72 , 708–715 (2013). Fortea, J. et al. APOE4 homozygosity represents a distinct genetic form of Alzheimer’s disease. Nat. Med. 30 , 1284–1291 (2024). Ossenkoppele, R. et al. Tau PET positivity in individuals with and without cognitive impairment varies with age, amyloid-β status, APOE genotype and sex. Nat. Neurosci. 28 , 1610–1621 (2025). Cicognola, C. et al. APOE4 impact on soluble and insoluble tau pathology is mostly influenced by amyloid-β. Brain J. Neurol. 148 , 2373–2383 (2025). Besser, L. M. et al. The Revised National Alzheimer’s Coordinating Center’s Neuropathology Form-Available Data and New Analyses. J. Neuropathol. Exp. Neurol. 77 , 717–726 (2018). Thal, D. R., Rüb, U., Orantes, M. & Braak, H. Phases of A beta-deposition in the human brain and its relevance for the development of AD. Neurology 58 , 1791–1800 (2002). 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.) 112 , 389–404 (2006). Mirra, S. S. et al. The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology 41 , 479–486 (1991). Clark, C. M. et al. Cerebral PET with florbetapir compared with neuropathology at autopsy for detection of neuritic amyloid-β plaques: a prospective cohort study. Lancet Neurol. 11 , 669–678 (2012). La Joie, R. et al. Multisite study of the relationships between antemortem [ 11 C]PIB‐PET Centiloid values and postmortem measures of Alzheimer’s disease neuropathology. Alzheimers Dement. 15 , 205–216 (2019). Jack, C. R. et al. Revised criteria for diagnosis and staging of Alzheimer’s disease: Alzheimer’s Association Workgroup. Alzheimers Dement. alz.13859 (2024) doi:10.1002/alz.13859. McKeith, I. G. et al. Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology 65 , 1863–1872 (2005). Josephs, K. A. et al. Updated TDP-43 in Alzheimer’s disease staging scheme. Acta Neuropathol. (Berl.) 131 , 571–585 (2016). Nelson, P. T. et al. LATE-NC staging in routine neuropathologic diagnosis: an update. Acta Neuropathol. (Berl.) 145 , 159–173 (2023). Suemoto, C. K. et al. Apolipoprotein E and Alzheimer’s disease pathology in a diverse autopsy study. Alzheimers Dement. 21 , e70927 (2025). Baiardi, S., Hansson, O., Levin, J. & Parchi, P. In vivo detection of Alzheimer’s and Lewy body disease concurrence: Clinical implications and future perspectives. Alzheimers Dement. alz.14039 (2024) doi:10.1002/alz.14039. Levin, J. et al. α-Synuclein seed amplification assay detects Lewy body co-pathology in autosomal dominant Alzheimer’s disease late in the disease course and dependent on Lewy pathology burden. Alzheimers Dement. 20 , 4351–4365 (2024). Bernhardt, A. M. et al. Alpha-synuclein co-pathology in Down syndrome-associated Alzheimer’s disease. Alzheimers Dement. 21 , e70342 (2025). Graff-Radford, J. et al. Duration and Pathologic Correlates of Lewy Body Disease. JAMA Neurol. 74 , 310–315 (2017). Wennberg, A. M. et al. Association of Apolipoprotein E ε4 With Transactive Response DNA-Binding Protein 43. JAMA Neurol. 75 , 1347–1354 (2018). Nelson, P. T. et al. Limbic-predominant age-related TDP-43 encephalopathy (LATE): consensus working group report. Brain 142 , 1503–1527 (2019). Nelson, P. T. et al. Limbic-predominant age-related TDP-43 encephalopathy (LATE-NC): Co-pathologies and genetic risk factors provide clues about pathogenesis. J. Neuropathol. Exp. Neurol. 83 , 396–415 (2024). Pottier, C. et al. Deciphering distinct genetic risk factors for FTLD-TDP pathological subtypes via whole-genome sequencing. Nat. Commun. 16 , 3914 (2025). Dols-Icardo, O., Reus, L. M. & Ramirez, A. Genetic architecture of FTLD-TDP pathological subtypes. Trends Neurosci. 48 , 553–554 (2025). Nelson, P. T. et al. The Amygdala as a Locus of Pathologic Misfolding in Neurodegenerative Diseases. J. Neuropathol. Exp. Neurol. 77 , 2–20 (2018). Neubauer, A. et al. Alpha-synuclein deposition patterns in Alzheimer’s disease: association with cortical amyloid beta and variable tau load. Acta Neuropathol. (Berl.) 150 , 46 (2025). Williams, T., Sorrentino, Z., Weinrich, M., Giasson, B. I. & Chakrabarty, P. Differential cross-seeding properties of tau and α-synuclein in mouse models of tauopathy and synucleinopathy. Brain Commun. 2 , fcaa090 (2020). Gawor, K. et al. Amygdala-predominant α-synuclein pathology is associated with exacerbated hippocampal neuron loss in Alzheimer’s disease. Brain Commun. 6 , fcae442 (2024). Ringman, J. M. et al. Clinical predictors of severe cerebral amyloid angiopathy and influence of APOE genotype in persons with pathologically-verified Alzheimer’s disease. JAMA Neurol. 71 , 878–883 (2014). Bruce, S. S. et al. Prevalence of Cerebral Amyloid Angiopathy and Associated Risk of Subsequent Ischemic and Hemorrhagic Stroke and Mortality in a Nationwide Cohort. Ann. Neurol. 98 , 249–257 (2025). Lamar, M. et al. APOE genotypes as a risk factor for age-dependent accumulation of cerebrovascular disease in older adults. Alzheimers Dement. 15 , 258–266 (2019). Paradela, R. S. et al. Apolipoprotein E genotypes were not associated with intracranial atherosclerosis: a population-based autopsy study. Cardiovasc. Pathol. 62 , 107479 (2023). Table 1 Table 1 is available in the Supplementary Files section. Additional Declarations Yes there is potential Competing Interest. COMPETING INTERESTS I. Rodríguez-Baz has received personal fees for service on advisory boards, speaker honoraria, or educational activities from Nutricia, Almirall, and Esteve. J. Arranz has received personal fees for service on advisory boards, speaker honoraria, or educational activities from Esteve, Lilly, and Roche Diagnostics. L. Molina-Porcel has provided consultancy services to Biogen. M. Carmona-Iragui has received personal fees for service on advisory boards, speaker honoraria, or educational activities from IMSERSO, Esteve, Lilly, Neuraxpharm, Adium Pharma, and Roche. M.R. Aranha is a co-founder and partner of Masima Soluções em Imagens Médicas LTDA (Porto Alegre, Brazil) and serves as an independent consultant to Ionis Pharmaceuticals. I. Illán-Gala has participated in advisory boards from UCB and Nutricia, and received speaker honoraria from Almirall, Esteve Pharmaceuticals S.A., Kern Pharma, Krka Farmacéutica S.L., Lilly, Nutricia, and Zambon S.A.U. D. Alcolea has received personal fees for advisory board services and/or speaker honoraria from Fujirebio-Europe, Roche, Nutricia, Krka Farmacéutica, Lilly, Zambon S.A.U., Grifols, and Esteve, outside the submitted work. A. Lleó has served as a consultant or on advisory boards for Fujirebio-Europe, Roche, Biogen, Grifols, Novartis, Eisai, Lilly, and Nutricia, outside the submitted work. C.K. Suemoto is funded by the Alzheimer's Association (AARG-20-678884 and 24CBIDR-1185483) and the São Paulo Research Foundation (FAPESP 2024/03917-7). J. Fortea has served on advisory boards, adjudication committees, or as a speaker for Roche, NovoNordisk, Esteve, Biogen, Laboratorios Carnot, Adamed, LMI, Novartis, Lundbeck, AC Immune, Alzheon, Zambon, Lilly, the Spanish Neurological Society, T21 Research Society, Lumind Foundation, Jérôme-Lejeune Foundation, Alzheimer's Association, National Institutes of Health USA, and Instituto de Salud Carlos III. D. Alcolea, A. Lleó, and J. Fortea hold a patent for markers of synaptopathy in neurodegenerative disease (licensed to ADx, EP8382175.0). A. Lleó is co-author of a patent on antibodies for amyloid precursor protein, methods and uses thereof (European priority No. EP25382226). No other competing interests were reported. FUNDING I. Rodríguez-Baz was supported by Instituto de Salud Carlos III (ISCIII), co-funded by the European Union (CM22/00052, CD25/00223), and by the Alzheimer's Association (AACSF-25-1486364). L. Vaqué-Alcázar was supported by a Sara Borrell postdoctoral fellowship from ISCIII, Spain (CD23/00235). L. Maure-Blesa was supported by ISCIII through a Río Hortega fellowship (CM23/00291), co-funded by the European Union. J. Arranz was supported by ISCIII through a Río Hortega fellowship (CM22/00243), co-funded by the European Union. R. Silva-Paradela was funded by the Alzheimer's Association Capacity Building in International Dementia Research (CBIDR) Program (24CBIDR-1185483) and the Global Brain Health Institute, Alzheimer's Association, and Alzheimer's Society (GBHI ALZ UK-25-1289657). M. Carmona-Iragui was supported by ISCIII (PI18/00335, PI22/00758, ICI23/00032), CIBERNED Program 1 (partly co-funded by FEDER, European Union), the Alzheimer's Association (AARG-22-973966), the Global Brain Health Institute (GBHI_ALZ-18-543740), and the Jérôme Lejeune Foundation (#1913 cycle 2019B; #2425 cycle 2024B). J. Selma-González is supported by the Contratos Predoctorales de Formación en Investigación en Salud program (FI25/00235) associated to project PI24/00598 (I. Illán-Gala), funded by ISCIII, Spain. C. Abdelnour received support from the Susan and Charles Berghoff Foundation and the ARISTOS program, funded by the European Union's Horizon Europe research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 101081334. O. Dols-Icardo receives funding from Fundación Española para el Fomento de la Investigación de la Esclerosis Lateral Amiotrófica (FUNDELA), Fundación HNA, the Alzheimer's Association (AARF-22-924456), and the Fondation Jérôme Lejeune (PDC-2023-51; #202307), and from ISCIII (PI21/01395, PI24/01087), co-funded by FEDER, European Union. I. Illán-Gala was supported by ISCIII (PI21/00791 and PI24/00598), co-funded by FEDER, European Union; the Alzheimer's Association (AACSF-21-850193); the Alzheimer Society (GBHI ALZ UK-21-72097); the Global Brain Health Institute as a senior Atlantic Fellow for Equity in Brain Health; and the Juan Rodés Contract (JR20/0018) from ISCIII, partly funded by the European Social Fund. D. Alcolea received funding from ISCIII (PI18/00435, PI22/00611, PI25/00422, INT23/00048), co-funded by FEDER, European Union, and from the Department of Health of the Generalitat de Catalunya through the PERIS programme (SLT006/17/125, SLT042/25/000034) and the Department of Research and Universities of the Generalitat de Catalunya (2021 SGR 00979). A. Lleó received funding from ISCIII (PI14/1561, PI20/01330), co-funded by FEDER, European Union. C.K. Suemoto was funded by the Alzheimer's Association (AARG-20-678884 and 24CBIDR-1185483) and the São Paulo Research Foundation (FAPESP 2024/03917-7). J. Fortea received funding from ISCIII (PI20/01473, PI23/01786, INT21/00073), co-funded by FEDER, European Union; CIBERNED Program 1; the National Institutes of Health (1R01AG056850-01A1, 3RF1AG056850-01S1, AG056850, R21AG056974, R01AG061566, 1R01AG081394-01, 1R61AG066543-01); the Department of Health of the Generalitat de Catalunya (SLT006/17/00119); Fundación Tatiana Pérez de Guzmán el Bueno (IIBSP-DOW-2020-151); the Horizon 2020 Research and Innovation Programme of the European Union (MES-CoBraD, GA 965422); BrightFocus Foundation; and Life Molecular Imaging (LMI). Supplementary Files SUPPLEMENTARYMATERIALAPOEcopathology.docx SUPPLEMENTARY MATERIAL FOR: Amyloid-linked versus age-driven copathologies in Alzheimer’s dementia: differential associations with APOE ε4 Table1.docx Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9044264","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":605769667,"identity":"86639f86-801d-423f-99a2-c891a6d08904","order_by":0,"name":"Íñigo Rodríguez-Baz","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABDElEQVRIiWNgGAWjYBAC+fkHGCQYDEBM/gMMH4jRwiaRANPCk8A4A0RBZSTwawEDHgNmHuK0JB+8XVFQm9jPfybxs80fuzx76eZnEgw1dnU4tcgfS7Y8Y3A8cWbD+cPSuW3JxTwyx8wkGI4l47ZFnsdMssHgWOKGgz0J0rkNzIk9EglmEowNzLi1SPZAtOw/zGP82+JPPVBL+jeglno8fjkD0lKTuIGNx0yage0wUEsOyJbDeLSkJVs2GBwwnnGGJ82yt+14Ys+dM8UWCceOSzbg0CI/I/ngzYY/dbL9/ecP3/jxpzqxfXb7xhsfaqr5cdkCBYeR2CAXJRBQDwTIkYDTE6NgFIyCUTBSAQDwnlX0Tkb3UAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0003-3039-9115","institution":"Sant Pau Memory Unit, Department of Neurology, Hospital de la Santa Creu i Sant Pau, Institut de Recerca Sant Pau (IR SANT PAU)","correspondingAuthor":true,"prefix":"","firstName":"Íñigo","middleName":"","lastName":"Rodríguez-Baz","suffix":""},{"id":605769668,"identity":"04c16de6-c099-4685-96b2-76d86a269fdf","order_by":1,"name":"Lidia Vaqué-Alcázar","email":"","orcid":"https://orcid.org/0000-0002-6776-6559","institution":"University of Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Lidia","middleName":"","lastName":"Vaqué-Alcázar","suffix":""},{"id":605769669,"identity":"15c6fbcf-307b-495d-8b56-f34872914f2a","order_by":2,"name":"Lucía Maure-Blesa","email":"","orcid":"https://orcid.org/0000-0001-5643-7971","institution":"Biomedical Research Institute Sant Pau","correspondingAuthor":false,"prefix":"","firstName":"Lucía","middleName":"","lastName":"Maure-Blesa","suffix":""},{"id":605769670,"identity":"d2b44c97-222e-403c-be24-26616e8fce47","order_by":3,"name":"Lucía Pertierra","email":"","orcid":"","institution":"Sant Pau Memory Unit, Department of Neurology, Hospital de la Santa Creu i Sant Pau, Institut de Recerca Sant Pau (IR SANT PAU)","correspondingAuthor":false,"prefix":"","firstName":"Lucía","middleName":"","lastName":"Pertierra","suffix":""},{"id":605769671,"identity":"9e071a01-4ebb-488a-b7c8-f73f583dd37e","order_by":4,"name":"Javier Arranz Martinez","email":"","orcid":"https://orcid.org/0000-0003-0891-1215","institution":"Sant Pau","correspondingAuthor":false,"prefix":"","firstName":"Javier","middleName":"Arranz","lastName":"Martinez","suffix":""},{"id":605769672,"identity":"87d4137b-24b2-4caf-9b2c-ecfd30cca993","order_by":5,"name":"Laura Molina-Porcel","email":"","orcid":"https://orcid.org/0000-0003-4068-8578","institution":"Hospital Clínic Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Laura","middleName":"","lastName":"Molina-Porcel","suffix":""},{"id":605769673,"identity":"0219e8a8-ac99-456d-8d96-909519dd8279","order_by":6,"name":"Iban Aldecoa","email":"","orcid":"https://orcid.org/0000-0001-5774-7453","institution":"cHospital Clínic Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Iban","middleName":"","lastName":"Aldecoa","suffix":""},{"id":605769674,"identity":"39ef9321-56b7-4f14-8c50-b28668e1eb16","order_by":7,"name":"Naomi Ferreira","email":"","orcid":"https://orcid.org/0000-0002-1557-7350","institution":"University of Sao Paulo Medical School","correspondingAuthor":false,"prefix":"","firstName":"Naomi","middleName":"","lastName":"Ferreira","suffix":""},{"id":605769675,"identity":"b81121bf-424d-4ed3-b499-52468153049c","order_by":8,"name":"Regina Paradela","email":"","orcid":"","institution":"University of Sao Paulo","correspondingAuthor":false,"prefix":"","firstName":"Regina","middleName":"","lastName":"Paradela","suffix":""},{"id":605769676,"identity":"67cbe26e-ff49-4e38-b95d-8239c803b763","order_by":9,"name":"Laura Videla","email":"","orcid":"https://orcid.org/0000-0002-9748-8465","institution":"Biomedical Research Institute Sant Pau","correspondingAuthor":false,"prefix":"","firstName":"Laura","middleName":"","lastName":"Videla","suffix":""},{"id":605769677,"identity":"03b5825c-7ce4-4fe5-9198-5e7601ae4539","order_by":10,"name":"Isabel Barroeta","email":"","orcid":"","institution":"Biomedical Research Institute Sant Pau","correspondingAuthor":false,"prefix":"","firstName":"Isabel","middleName":"","lastName":"Barroeta","suffix":""},{"id":605769678,"identity":"01bc8ab8-c2c2-4b7c-9b8b-aae3653402db","order_by":11,"name":"María Carmona-Iragui","email":"","orcid":"","institution":"Biomedical Research Institute Sant Pau","correspondingAuthor":false,"prefix":"","firstName":"María","middleName":"","lastName":"Carmona-Iragui","suffix":""},{"id":605769679,"identity":"cf8672fa-9940-4cc8-8ea5-2e057a4d828c","order_by":12,"name":"Mª Belén Sánchez-Saudinós","email":"","orcid":"","institution":"Sant Pau Memory Unit, Department of Neurology, Hospital de la Santa Creu i Sant Pau, Institut de Recerca Sant Pau (IR SANT PAU)","correspondingAuthor":false,"prefix":"","firstName":"Mª","middleName":"Belén","lastName":"Sánchez-Saudinós","suffix":""},{"id":605769680,"identity":"3d79457f-f509-4c7d-92e8-58ad673cedd7","order_by":13,"name":"Judit Selma-González","email":"","orcid":"","institution":"Sant Pau Memory Unit, Department of Neurology, Hospital de la Santa Creu i Sant Pau, Institut de Recerca Sant Pau (IR SANT PAU)","correspondingAuthor":false,"prefix":"","firstName":"Judit","middleName":"","lastName":"Selma-González","suffix":""},{"id":605769681,"identity":"dba6e47f-102f-4814-a2e1-a1608a7ec413","order_by":14,"name":"Oriol Dols-Icardo","email":"","orcid":"https://orcid.org/0000-0003-2656-8748","institution":"Institut de recerca de l'hospital de Sant Pau","correspondingAuthor":false,"prefix":"","firstName":"Oriol","middleName":"","lastName":"Dols-Icardo","suffix":""},{"id":605769682,"identity":"ad8e7a7f-0219-4889-8047-a2cbeb7c9d25","order_by":15,"name":"Mateus Rozalem Aranha","email":"","orcid":"","institution":"Institut de Recerca Hospital de la Santa Creu i Sant Pau","correspondingAuthor":false,"prefix":"","firstName":"Mateus","middleName":"Rozalem","lastName":"Aranha","suffix":""},{"id":605769683,"identity":"25e18e01-1ca6-4443-b9b5-2d213ce691d7","order_by":16,"name":"Sònia Sirisi Dolcet","email":"","orcid":"","institution":"Institut de Recerca Sant Pau (IR SANT PAU), Hospital de la Santa Creu i Sant Pau","correspondingAuthor":false,"prefix":"","firstName":"Sònia","middleName":"Sirisi","lastName":"Dolcet","suffix":""},{"id":605769684,"identity":"93b67e91-b907-4bdc-a20c-5ca3daae7d35","order_by":17,"name":"Carla Abdelnour","email":"","orcid":"","institution":"Sant Pau Memory Unit, Department of Neurology, Hospital de la Santa Creu i Sant Pau, Institut de Recerca Sant Pau (IR SANT PAU)","correspondingAuthor":false,"prefix":"","firstName":"Carla","middleName":"","lastName":"Abdelnour","suffix":""},{"id":605769685,"identity":"201cadd9-db0e-40d0-ba9c-fd37585557ea","order_by":18,"name":"Ignacio Illán-Gala","email":"","orcid":"","institution":"Sant Pau Memory Unit, Hospital de la Santa Creu i Sant Pau - Biomedical Research Institute Sant Pau- Universitat Autònoma de Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Ignacio","middleName":"","lastName":"Illán-Gala","suffix":""},{"id":605769686,"identity":"cebe7b03-acca-4317-a836-44f31ca243a1","order_by":19,"name":"Daniel Alcolea","email":"","orcid":"https://orcid.org/0000-0002-3819-3245","institution":"Hospital de Sant Pau - IIB Sant Pau","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"","lastName":"Alcolea","suffix":""},{"id":605769687,"identity":"2f4b6e58-7654-4fbc-802e-163c5314f0ff","order_by":20,"name":"Alberto Lleo","email":"","orcid":"https://orcid.org/0000-0002-2568-5478","institution":"Institut de Recerca Sant Pau (IR SANT PAU), Hospital de la Santa Creu i Sant Pau","correspondingAuthor":false,"prefix":"","firstName":"Alberto","middleName":"","lastName":"Lleo","suffix":""},{"id":605769688,"identity":"e5d03ce4-78d3-4bcd-9a22-88c94db7e6e2","order_by":21,"name":"Claudia Suemoto","email":"","orcid":"https://orcid.org/0000-0002-5942-4778","institution":"University of Sao Paulo","correspondingAuthor":false,"prefix":"","firstName":"Claudia","middleName":"","lastName":"Suemoto","suffix":""},{"id":605769689,"identity":"eb033163-c669-41ae-afe9-28fa03f795aa","order_by":22,"name":"Juan Fortea","email":"","orcid":"https://orcid.org/0000-0002-1340-638X","institution":"Sant Pau Memory Unit, Hospital de la Santa Creu i Sant Pau - Biomedical Research Institute Sant Pau- Universitat Autònoma de Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Juan","middleName":"","lastName":"Fortea","suffix":""}],"badges":[],"createdAt":"2026-03-05 22:05:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9044264/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9044264/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104693179,"identity":"99de9125-4532-4f37-8e9e-1b94fd034d3c","added_by":"auto","created_at":"2026-03-16 06:42:47","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1329612,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-9044264/v1/1ea8a4398431ce7eeee11a25.png"},{"id":104693116,"identity":"8d3b33fb-269d-4352-b2e1-3bd968c7698b","added_by":"auto","created_at":"2026-03-16 06:42:34","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":661545,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-9044264/v1/5c8a2d027f3532d2e0c6ada5.png"},{"id":104693176,"identity":"d714848b-04fd-4e59-92ff-c11ec4522280","added_by":"auto","created_at":"2026-03-16 06:42:45","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":912111,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-9044264/v1/41b3c61712a3f36a47d76321.png"},{"id":104693130,"identity":"b5dc8e5c-76da-4822-9087-40a4484e7a3e","added_by":"auto","created_at":"2026-03-16 06:42:38","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1639328,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u0026nbsp;\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-9044264/v1/a29e6f3e1439c5f4115b7f68.png"},{"id":104693181,"identity":"a346f10d-4892-4a35-a9a6-8b1df3e6abd7","added_by":"auto","created_at":"2026-03-16 06:42:55","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5918218,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9044264/v1/7c2c5e31-b48e-4c66-b5c6-851985997aa2.pdf"},{"id":104693172,"identity":"fcfb9762-0264-4362-9d12-723164de63d6","added_by":"auto","created_at":"2026-03-16 06:42:45","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":12394875,"visible":true,"origin":"","legend":"SUPPLEMENTARY MATERIAL FOR: Amyloid-linked versus age-driven copathologies in Alzheimer\u0026#x2019;s dementia: differential associations with \u003ci\u003eAPOE\u003c/i\u003e \u0026#x03B5;4","description":"","filename":"SUPPLEMENTARYMATERIALAPOEcopathology.docx","url":"https://assets-eu.researchsquare.com/files/rs-9044264/v1/9a8726c3187d274ff750dbca.docx"},{"id":104693180,"identity":"856fd154-5094-4e4c-82c2-fc38912c37ec","added_by":"auto","created_at":"2026-03-16 06:42:47","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":23699,"visible":true,"origin":"","legend":"","description":"","filename":"Table1.docx","url":"https://assets-eu.researchsquare.com/files/rs-9044264/v1/b56d46f2a91525b95aa7530e.docx"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nCOMPETING INTERESTS\r\nI. Rodríguez-Baz has received personal fees for service on advisory boards, speaker honoraria, or educational activities from Nutricia, Almirall, and Esteve. J. Arranz has received personal fees for service on advisory boards, speaker honoraria, or educational activities from Esteve, Lilly, and Roche Diagnostics. L. Molina-Porcel has provided consultancy services to Biogen. M. Carmona-Iragui has received personal fees for service on advisory boards, speaker honoraria, or educational activities from IMSERSO, Esteve, Lilly, Neuraxpharm, Adium Pharma, and Roche. M.R. Aranha is a co-founder and partner of Masima Soluções em Imagens Médicas LTDA (Porto Alegre, Brazil) and serves as an independent consultant to Ionis Pharmaceuticals. I. Illán-Gala has participated in advisory boards from UCB and Nutricia, and received speaker honoraria from Almirall, Esteve Pharmaceuticals S.A., Kern Pharma, Krka Farmacéutica S.L., Lilly, Nutricia, and Zambon S.A.U. D. Alcolea has received personal fees for advisory board services and/or speaker honoraria from Fujirebio-Europe, Roche, Nutricia, Krka Farmacéutica, Lilly, Zambon S.A.U., Grifols, and Esteve, outside the submitted work. A. Lleó has served as a consultant or on advisory boards for Fujirebio-Europe, Roche, Biogen, Grifols, Novartis, Eisai, Lilly, and Nutricia, outside the submitted work. C.K. Suemoto is funded by the Alzheimer's Association (AARG-20-678884 and 24CBIDR-1185483) and the São Paulo Research Foundation (FAPESP 2024/03917-7). J. Fortea has served on advisory boards, adjudication committees, or as a speaker for Roche, NovoNordisk, Esteve, Biogen, Laboratorios Carnot, Adamed, LMI, Novartis, Lundbeck, AC Immune, Alzheon, Zambon, Lilly, the Spanish Neurological Society, T21 Research Society, Lumind Foundation, Jérôme-Lejeune Foundation, Alzheimer's Association, National Institutes of Health USA, and Instituto de Salud Carlos III. D. Alcolea, A. Lleó, and J. Fortea hold a patent for markers of synaptopathy in neurodegenerative disease (licensed to ADx, EP8382175.0). A. Lleó is co-author of a patent on antibodies for amyloid precursor protein, methods and uses thereof (European priority No. EP25382226). No other competing interests were reported.\r\n \r\nFUNDING\r\nI. Rodríguez-Baz was supported by Instituto de Salud Carlos III (ISCIII), co-funded by the European Union (CM22/00052, CD25/00223), and by the Alzheimer's Association (AACSF-25-1486364). L. Vaqué-Alcázar was supported by a Sara Borrell postdoctoral fellowship from ISCIII, Spain (CD23/00235). L. Maure-Blesa was supported by ISCIII through a Río Hortega fellowship (CM23/00291), co-funded by the European Union. J. Arranz was supported by ISCIII through a Río Hortega fellowship (CM22/00243), co-funded by the European Union. R. Silva-Paradela was funded by the Alzheimer's Association Capacity Building in International Dementia Research (CBIDR) Program (24CBIDR-1185483) and the Global Brain Health Institute, Alzheimer's Association, and Alzheimer's Society (GBHI ALZ UK-25-1289657). M. Carmona-Iragui was supported by ISCIII (PI18/00335, PI22/00758, ICI23/00032), CIBERNED Program 1 (partly co-funded by FEDER, European Union), the Alzheimer's Association (AARG-22-973966), the Global Brain Health Institute (GBHI_ALZ-18-543740), and the Jérôme Lejeune Foundation (#1913 cycle 2019B; #2425 cycle 2024B). J. Selma-González is supported by the Contratos Predoctorales de Formación en Investigación en Salud program (FI25/00235) associated to project PI24/00598 (I. Illán-Gala), funded by ISCIII, Spain. C. Abdelnour received support from the Susan and Charles Berghoff Foundation and the ARISTOS program, funded by the European Union's Horizon Europe research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 101081334. O. Dols-Icardo receives funding from Fundación Española para el Fomento de la Investigación de la Esclerosis Lateral Amiotrófica (FUNDELA), Fundación HNA, the Alzheimer's Association (AARF-22-924456), and the Fondation Jérôme Lejeune (PDC-2023-51; #202307), and from ISCIII (PI21/01395, PI24/01087), co-funded by FEDER, European Union. I. Illán-Gala was supported by ISCIII (PI21/00791 and PI24/00598), co-funded by FEDER, European Union; the Alzheimer's Association (AACSF-21-850193); the Alzheimer Society (GBHI ALZ UK-21-72097); the Global Brain Health Institute as a senior Atlantic Fellow for Equity in Brain Health; and the Juan Rodés Contract (JR20/0018) from ISCIII, partly funded by the European Social Fund. D. Alcolea received funding from ISCIII (PI18/00435, PI22/00611, PI25/00422, INT23/00048), co-funded by FEDER, European Union, and from the Department of Health of the Generalitat de Catalunya through the PERIS programme (SLT006/17/125, SLT042/25/000034) and the Department of Research and Universities of the Generalitat de Catalunya (2021 SGR 00979). A. Lleó received funding from ISCIII (PI14/1561, PI20/01330), co-funded by FEDER, European Union. C.K. Suemoto was funded by the Alzheimer's Association (AARG-20-678884 and 24CBIDR-1185483) and the São Paulo Research Foundation (FAPESP 2024/03917-7). J. Fortea received funding from ISCIII (PI20/01473, PI23/01786, INT21/00073), co-funded by FEDER, European Union; CIBERNED Program 1; the National Institutes of Health (1R01AG056850-01A1, 3RF1AG056850-01S1, AG056850, R21AG056974, R01AG061566, 1R01AG081394-01, 1R61AG066543-01); the Department of Health of the Generalitat de Catalunya (SLT006/17/00119); Fundación Tatiana Pérez de Guzmán el Bueno (IIBSP-DOW-2020-151); the Horizon 2020 Research and Innovation Programme of the European Union (MES-CoBraD, GA 965422); BrightFocus Foundation; and Life Molecular Imaging (LMI).","formattedTitle":"Amyloid-linked versus age-driven copathologies in Alzheimer’s dementia: differential associations with \u003ci\u003eAPOE\u003c/i\u003e ε4","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAlzheimer\u0026apos;s disease (AD) is the most common cause of dementia, neuropathologically characterized by the accumulation of amyloid plaques and tau neurofibrillary tangles. However, previous studies have demonstrated that the majority of patients with dementia present other neuropathologies at autopsy, frequently encompassing Lewy body pathology, TAR DNA-binding protein 43 (TDP-43) proteinopathy, cerebral amyloid angiopathy (CAA), and various forms of cerebrovascular disease. These studies have traditionally examined copathologies in dementia patients as largely independent contributors, often interpreted in the context of aging\u003csup\u003e1,2\u003c/sup\u003e. However, evidence from genetically determined AD, both in Autosomal Dominant AD (ADAD)\u003csup\u003e3\u003c/sup\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eand Down syndrome-associated AD (DSAD)\u003csup\u003e4\u003c/sup\u003e, have shown that other neuropathological hallmarks beyond amyloid and tau are frequent despite the young age at which dementia manifests in these populations.\u003c/p\u003e\n\u003cp\u003eThe apolipoprotein E (\u003cem\u003eAPOE\u003c/em\u003e) gene represents the strongest common genetic risk factor for sporadic AD. The three major allelic variants (\u0026epsilon;2, \u0026epsilon;3, and \u0026epsilon;4) confer differential risk, with \u0026epsilon;4 associated with increased susceptibility and \u0026epsilon;2 with relative protection against AD\u003csup\u003e5,6\u003c/sup\u003e. Beyond its established role in amyloid metabolism, some studies have shown that carriers of the \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;4 allele have more \u0026alpha;-synuclein\u0026nbsp;\u003csup\u003e7,8\u003c/sup\u003e and TDP-43 pathologies\u003csup\u003e9\u003c/sup\u003e, as well as increased CAA burden\u003csup\u003e10\u003c/sup\u003e. Recent evidence suggests that \u003cem\u003eAPOE\u0026nbsp;\u003c/em\u003e\u0026epsilon;4 homozygosity represents a genetically determined form of AD with near-complete penetrance of AD biology\u003csup\u003e11\u003c/sup\u003e.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eHowever, despite the much higher prevalence and the earlier presentation of AD symptoms in \u003cem\u003eAPOE\u0026nbsp;\u003c/em\u003e\u0026epsilon;4/\u0026epsilon;4, when analyzing AD patients at dementia stage with amyloid confirmation, there were no differences across haplotypes, but younger individuals had higher tau positron emission tomography (PET) uptake\u003csup\u003e11\u003c/sup\u003e.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eThis has been confirmed in other studies that have shown age-associated increases in tau PET burden in cognitively unimpaired individuals but decreases with age in patients with AD dementia\u003csup\u003e12,13\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eAge is indeed a fundamental determinant of both neurodegenerative and vascular pathologies.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eFurthermore, the temporal evolution of copathologies in relation to dementia duration remains poorly understood. Whether \u003cem\u003eAPOE\u0026nbsp;\u003c/em\u003ehaplotype modifies the severity or distribution of Lewy body, TDP-43, and vascular pathologies in relation to dementia duration has not been systematically examined. Disentangling the relative contributions of \u003cem\u003eAPOE\u003c/em\u003e haplotype and chronological age in neuropathologically confirmed AD dementia patients to diverse neuropathological outcomes is therefore critical for understanding disease mechanisms.\u003c/p\u003e\n\u003cp\u003eThe objectives of this study were to evaluate whether \u003cem\u003eAPOE\u003c/em\u003e haplotypes are associated with neuropathological copathologies in individuals with neuropathologically confirmed AD dementia, while accounting for the potential confounding and modifying effects of age and dementia duration. Specifically, we aimed to: (1) examine whether \u003cem\u003eAPOE\u003c/em\u003e haplotypes exhibit different burden and regional specificity in their associations with Lewy body and TDP-43 pathologies; and (2) determine the relative contributions of \u003cem\u003eAPOE\u0026nbsp;\u003c/em\u003ehaplotype, age, and dementia duration to the severity of CAA, arteriolosclerosis, and atherosclerosis. By addressing these questions, this study seeks to refine our understanding of the role of \u003cem\u003eAPOE\u0026nbsp;\u003c/em\u003ein shaping the heterogeneous neuropathological landscape of AD, with implications for mechanistic understanding, risk stratification, and the design of future clinical trials and therapeutic development.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003e2.1 Sample characteristics and distribution of \u003cem\u003eAPOE\u003c/em\u003e haplotype\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe analyzed neuropathological and clinical data from 11,897 autopsied individuals enrolled across National Institute on Aging-funded Alzheimer\u0026apos;s Disease Research Centers and aggregated in the National Alzheimer\u0026apos;s Coordinating Center (NACC) repository. Table 1 summarizes the demographic, clinical, and neuropathological features of the 11,897 individuals included in the study. The most common \u003cem\u003eAPOE\u003c/em\u003e haplotypes were \u0026epsilon;3/\u0026epsilon;3 (47.5%), followed by \u0026epsilon;3/\u0026epsilon;4 (33.4%) and \u0026epsilon;4/\u0026epsilon;4 (8.0%). Compared with \u0026epsilon;3/\u0026epsilon;3 carriers, both \u0026epsilon;3/\u0026epsilon;4 and \u0026epsilon;4/\u0026epsilon;4 carriers showed earlier ages at dementia diagnosis (median -1.7 and -5.6 years, respectively) and death (-2.2 and -6.7 years; all p \u0026lt; 0.001). Overall, 48.3% of the cohort were female and 89.4% were White. At the time of death, 73.1% had received a dementia diagnosis, and 57.6% had been clinically classified as having AD. Moderate or frequent neuritic plaques were present in 63.6% of participants.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAmyloid positivity rates varied markedly across haplotypes in the full cohort. Among participants with a clinical diagnosis of AD dementia, amyloid positivity (defined as moderate or frequent neuritic plaques) was present in 88.7% of \u0026epsilon;4/\u0026epsilon;4 carriers, 79.3% of \u0026epsilon;3/\u0026epsilon;4 carriers, and 68.5% of \u0026epsilon;3/\u0026epsilon;3 carriers (Supplementary Figure 1, Supplementary Table 1). Similarly, concordant clinicopathological AD diagnosis was significantly more common among \u0026epsilon;3/\u0026epsilon;4 and \u0026epsilon;4/\u0026epsilon;4 carriers compared to \u0026epsilon;3/\u0026epsilon;3 carriers (Odds ratio [OR] = 2.73, with 95% confidence interval [95% CI] = 2.41\u0026ndash;3.10 and OR 5.59, 95% CI 4.46\u0026ndash;7.07, respectively; both p\u0026lt;0.001; Supplementary Table 2).\u003c/p\u003e\n\u003cp\u003eTo ensure analyses focused on neuropathologically confirmed AD and to avoid confounding by differential amyloid rates, subsequent copathology analyses (sections 2.2 and 2.3) were restricted to AD dementia cases with confirmed amyloid deposition. This restriction ensured that observed \u003cem\u003eAPOE\u003c/em\u003e effects on copathologies reflected associations within established AD pathology rather than differences in AD diagnostic accuracy across haplotypes. Of note, in both clinically diagnosed and amyloid-positive AD dementia, the prevalence of Braak stage V\u0026ndash;VI decreased with increasing age at death, with the exception of \u0026epsilon;4/\u0026epsilon;4 carriers, who showed no appreciable age effect and consistently the highest prevalence of Braak stage V\u0026ndash;VI across all age groups (Supplementary Figure 2, Supplementary Table 3).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eLewy body and TDP-43 pathologies\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLewy body pathology was prevalent among individuals with amyloid-positive AD dementia, affecting more than one-third of \u0026epsilon;3/\u0026epsilon;3 carriers and over half of \u0026epsilon;4/\u0026epsilon;4 carriers overall (Figure 1, left panels). Prevalence increased with \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;4 dose, from 37.2% in \u0026epsilon;3/\u0026epsilon;3 to 46.2% in \u0026epsilon;3/\u0026epsilon;4 (OR = 1.39, 95% CI: 1.14\u0026ndash;1.71, p = 0.001) and 52.6% in \u0026epsilon;4/\u0026epsilon;4 carriers (OR = 1.77, 95% CI: 1.33\u0026ndash;2.36, p \u0026lt; 0.001). Across haplotypes, the most common regional pattern was limbic/amygdala-predominant involvement, followed by neocortical pathology, whereas brainstem-predominant Lewy pathology was comparatively infrequent.\u003c/p\u003e\n\u003cp\u003eRegional distributions differed modestly by \u003cem\u003eAPOE\u003c/em\u003e haplotype. Compared with \u0026epsilon;3/\u0026epsilon;3 carriers, \u0026epsilon;3/\u0026epsilon;4 carriers had higher prevalence of olfactory bulb/region-unknown pathology (5.9% vs 3.1%; OR = 2.27, 95% CI: 1.14\u0026ndash;4.93, p = 0.018) and neocortical involvement (14.5% vs 11.0%; OR = 1.50, 95% CI: 1.11\u0026ndash;2.04, p = 0.008). \u0026epsilon;4/\u0026epsilon;4 carriers showed greater limbic/amygdalar involvement than \u0026epsilon;3/\u0026epsilon;3 carriers (29.0% vs 20.2%; OR = 1.63, 95% CI: 1.21\u0026ndash;2.19, p = 0.001). However, despite these differences, no consistent dose-dependent shift toward more extensive regional involvement was observed across \u0026epsilon;4 allele groups (Supplementary Table 4A).\u003c/p\u003e\n\u003cp\u003eLewy body pathology showed no significant association with age at death across \u003cem\u003eAPOE\u003c/em\u003e haplotypes (all likelihood ratio chi-squared tests [LR\u0026chi;\u0026sup2;] \u0026le; 2.54, p \u0026ge; 0.111), consistent with the relatively stable prevalence across age strata shown in Figure 1. In regional analyses, increasing age was associated with lower amygdala involvement in \u0026epsilon;3/\u0026epsilon;3 carriers (OR per 10 years = 0.80, 95% CI: 0.68\u0026ndash;0.93, p = 0.004), but no other significant age-related changes were observed in other regions or haplotypes (Supplementary Figure 3; Supplementary Table 5A).\u003c/p\u003e\n\u003cp\u003eIn contrast, Lewy pathology was associated with dementia duration. Longer duration was linked to greater limbic/amygdalar involvement in \u0026epsilon;3/\u0026epsilon;3 and \u0026epsilon;3/\u0026epsilon;4 carriers after adjustment for sex, education, age at dementia onset and Alzheimer\u0026apos;s disease neuropathologic change (ADNC) classification (Figure 2, left panels; Supplementary Table 6A). This relationship was not observed in \u0026epsilon;4/\u0026epsilon;4 carriers, in whom (mainly limbic/amygdalar) Lewy pathology was already highly prevalent at shorter dementia durations, suggesting earlier or more rapid accumulation in this group.\u003c/p\u003e\n\u003cp\u003eTDP-43 pathology was even more prevalent than Lewy pathology across all \u003cem\u003eAPOE\u003c/em\u003e haplotypes (Figure 1, right panels). Overall prevalence exceeded three-quarters of individuals in each group, except \u0026epsilon;2/\u0026epsilon;X, and was higher in \u0026epsilon;4/\u0026epsilon;4 compared with \u0026epsilon;3/\u0026epsilon;3 carriers (81.9% vs 76.4%; OR = 1.99, 95% CI: 1.16\u0026ndash;3.50, p = 0.012). Across haplotypes, amygdala involvement was the dominant regional pattern, with hippocampal, entorhinal/inferior temporal, and neocortical involvement occurring in progressively smaller proportions.\u003c/p\u003e\n\u003cp\u003eUnlike Lewy pathology, TDP-43 showed a strong relationship with age at death. Overall prevalence increased with age in \u0026epsilon;3/\u0026epsilon;3 and \u0026epsilon;3/\u0026epsilon;4 carriers (both LR\u0026chi;\u0026sup2; \u0026ge; 10.74, p \u0026le; 0.001), accompanied by regional progression beyond the amygdala. Specifically, increasing age was associated with greater entorhinal/inferior temporal involvement in \u0026epsilon;3/\u0026epsilon;3 carriers (OR per 10 years = 1.51, 95% CI: 1.17\u0026ndash;1.99, p = 0.002) and greater neocortical involvement in \u0026epsilon;3/\u0026epsilon;4 carriers (OR per 10 years = 2.01, 95% CI: 1.38\u0026ndash;3.01, p \u0026lt; 0.001). Conversely, amygdala-only involvement declined with age in \u0026epsilon;3/\u0026epsilon;4 (OR per 10 years = 0.73, 95% CI: 0.55\u0026ndash;0.99, p = 0.044) and \u0026epsilon;4/\u0026epsilon;4 carriers (OR per 10 years = 0.57, 95% CI: 0.31\u0026ndash;0.98, p = 0.042) (Supplementary Figure 4; Supplementary Table 5B).\u003c/p\u003e\n\u003cp\u003eDespite its strong age dependency, TDP-43 pathology showed a weaker association with dementia duration. After adjustment for age at dementia onset, sex, education, and ADNC classification, longer duration remained associated with higher TDP-43 prevalence across most haplotypes (all LR\u0026chi;\u0026sup2; \u0026ge; 5.04, p \u0026le; 0.025), although not significantly in \u0026epsilon;4/\u0026epsilon;4 carriers (LR\u0026chi;\u0026sup2; = 3.57, p = 0.059) (Figure 2, right panels). Regionally, longer duration was associated with increased entorhinal involvement in \u0026epsilon;2/\u0026epsilon;X carriers and greater neocortical involvement in \u0026epsilon;3/\u0026epsilon;4 carriers, but these effects were modest compared with the age-related patterns (Supplementary Table 6B).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.3 Vascular-related pathology\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCAA exhibited a striking \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;4 gene-dose effect, with increasing prevalence and severity across haplotypes. Compared with \u0026epsilon;3/\u0026epsilon;3 carriers, moderate-to-severe CAA was more common in \u0026epsilon;3/\u0026epsilon;4 carriers (OR = 1.54, 95% CI: 1.25\u0026ndash;1.89, p \u0026lt; 0.001) and markedly elevated in \u0026epsilon;4/\u0026epsilon;4 carriers (OR = 5.76, 95% CI: 4.20\u0026ndash;7.96, p \u0026lt; 0.001), with \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;4 homozygotes showing the highest proportions of both moderate and severe CAA (Supplementary Table 7A). This shift toward moderate and severe CAA was evident across all age strata, with \u0026epsilon;4/\u0026epsilon;4 carriers already showing a high burden at younger ages (Figure 3, left panels).\u003c/p\u003e\n\u003cp\u003eAge further modified CAA burden in \u0026epsilon;4 carriers. Among \u0026epsilon;4/\u0026epsilon;4 individuals, CAA prevalence increased with advancing age (OR per 10 years = 1.69, 95% CI: 1.21\u0026ndash;2.38, p = 0.002), a pattern also observed for severe CAA in \u0026epsilon;3/\u0026epsilon;4 carriers (Supplementary Table 8A; Supplementary Figure 5, upper panels). In contrast, age-related increases were less pronounced in \u0026epsilon;3/\u0026epsilon;3 carriers.\u003c/p\u003e\n\u003cp\u003eDementia duration showed a distinct pattern. After full adjustment for age at dementia onset, sex, education, and ADNC classification, longer dementia duration was associated with increasing CAA prevalence in \u0026epsilon;3/\u0026epsilon;3 (OR per year = 1.09, 95% CI: 1.03\u0026ndash;1.15, p = 0.002) and \u0026epsilon;3/\u0026epsilon;4 carriers (OR per year = 1.06, 95% CI: 1.02\u0026ndash;1.11, p = 0.008; Figure 4, left panels). In \u0026epsilon;4/\u0026epsilon;4 carriers, however, longer duration was associated with lower prevalence of severe CAA (OR per year = 0.91, 95% CI: 0.84\u0026ndash;0.98, p = 0.016), suggesting that severe CAA accumulates early in this haplotype (Supplementary Table 9A) and might be associated with survival.\u003c/p\u003e\n\u003cp\u003eIn contrast to CAA, cerebral arteriolosclerosis and atherosclerosis of the circle of Willis showed weaker associations with \u003cem\u003eAPOE\u003c/em\u003e haplotype and were driven primarily by age (Figure 3, middle and right panels). Arteriolosclerosis prevalence was modestly higher in \u0026epsilon;4/\u0026epsilon;4 compared with \u0026epsilon;3/\u0026epsilon;3 carriers (52.7% vs 46.0%; OR = 1.49, 95% CI: 1.10\u0026ndash;2.01, p = 0.009), driven mainly by severe pathology (OR = 1.69, 95% CI: 1.12\u0026ndash;2.55, p = 0.013). By contrast, moderate-to-severe atherosclerosis prevalence did not differ significantly across \u003cem\u003eAPOE\u003c/em\u003e haplotypes, although \u0026epsilon;4/\u0026epsilon;4 carriers showed higher rates of mild atherosclerosis (Supplementary Table 7B-C).\u003c/p\u003e\n\u003cp\u003eAge was the dominant determinant of cerebral arteriolosclerosis and atherosclerosis of the circle of Willis. Prevalence of arteriolosclerosis increased with age across all haplotypes (all LR\u0026chi;\u0026sup2; \u0026ge; 5.96, p \u0026le; 0.015), with parallel increases in moderate and severe strata (Supplementary Table 8B). Similarly, atherosclerosis prevalence increased with age in all haplotypes (all LR\u0026chi;\u0026sup2; \u0026ge; 7.79, p \u0026le; 0.005), accompanied by progressive shifts toward greater severity (Supplementary Figure 5, middle and bottom panels).\u003c/p\u003e\n\u003cp\u003eAssociations with dementia duration were comparatively modest. In fully adjusted models, arteriolosclerosis prevalence increased with longer disease duration only in \u0026epsilon;3/\u0026epsilon;3 carriers (LR\u0026chi;\u0026sup2; = 18.84, p \u0026lt; 0.001), driven mainly by moderate pathology (Figure 4, middle panels; Supplementary Table 9B). Atherosclerosis prevalence increased with longer dementia duration in all haplotypes except \u0026epsilon;2/\u0026epsilon;X (all LR\u0026chi;\u0026sup2; \u0026ge; 9.04, p \u0026le; 0.003), although the magnitude of these effects was smaller than those observed with age (Figure 4, right panels; Supplementary Table 9C).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this large autopsy-based cohort of AD dementia with confirmed amyloid positivity, we found that \u003cem\u003eAPOE\u003c/em\u003e haplotype related to copathologies in a highly selective manner following two broad trajectories: amyloid-linked/\u003cem\u003eAPOE\u003c/em\u003e-dependent versus age-driven/\u003cem\u003eAPOE\u003c/em\u003e-independent. Lewy body and TDP-43 pathologies were both extremely frequent, but they diverged in their determinants: Lewy pathology showed modest \u003cem\u003eAPOE\u003c/em\u003e-related differences concentrated in limbic/amygdalar regions and was largely age-independent, yet it increased with dementia duration, positioning it as an \u0026ldquo;intermediate\u0026rdquo; process that may accrue alongside AD progression. TDP-43, in contrast, was less \u003cem\u003eAPOE\u003c/em\u003e-dependent and tracked primarily with age, with evidence of regional progression and comparatively weaker links to dementia duration. The vascular findings then provided the clearest dissociation: CAA was strongly \u0026epsilon;4 dose-dependent, whereas non-CAA vascular disease (arteriolosclerosis and large-vessel atherosclerosis) was predominantly age-driven with limited haplotype effects. Notably, dementia duration related more to amyloid-linked pathology (CAA, and to some extent Lewy pathology) than to non-CAA vascular disease, reinforcing the interpretation that \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;4 chiefly amplifies amyloid-related mechanisms once dementia is established, while other copathologies accumulate largely through aging and disease-stage processes that are relatively \u003cem\u003eAPOE\u003c/em\u003e haplotype-independent.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAPOE\u003c/em\u003e strongly shapes core AD pathology even in dementia-stage disease. \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;4/\u0026epsilon;4 carriers were enriched for clinicopathological AD, consistent with near-complete penetrance of AD biology\u003csup\u003e11\u003c/sup\u003e. Furthermore, the age-invariant frequency of Braak V-VI in \u0026epsilon;4/\u0026epsilon;4 suggests early tau saturation, in line with tau-PET evidence\u003csup\u003e12,13,24\u003c/sup\u003e. Conversely \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;2 confers protection to AD pathology even in clinical AD dementia cases\u003csup\u003e5\u003c/sup\u003e. To reduce sample-composition bias when assessing copathologies, we focused our primary analyses on AD dementia with neuritic plaque-confirmed amyloid positivity, ensuring \u003cem\u003eAPOE\u003c/em\u003e effects were evaluated within established AD biology.\u003c/p\u003e\n\u003cp\u003eLewy body pathology was more frequent overall in \u0026epsilon;3/\u0026epsilon;4 and \u0026epsilon;4/\u0026epsilon;4 carriers. However, the most informative signal lied in the topography. The only consistent \u003cem\u003eAPOE\u003c/em\u003e-related pattern was limbic/amygdalar involvement, which was higher in \u0026epsilon;4/\u0026epsilon;4 than \u0026epsilon;3/\u0026epsilon;3 carriers (29.0% vs 20.2%), whereas associations in olfactory and neocortical regions were less consistent. This topographic selectivity places Lewy pathology in an \u0026ldquo;intermediate\u0026rdquo; position between the amyloid-linked and age-driven axes: it showed modest haplotype effects concentrated in a vulnerable region, with no robust (or decreasing) age gradient. Importantly, there was an amydgalar accumulation (but not neocortical) Lewy pathology with dementia duration.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe concordant effects of \u003cem\u003eAPOE\u003c/em\u003e and dementia duration support the view that amygdala-predominant Lewy pathology often represents secondary, AD-mediated \u0026alpha;-synucleinopathy rather than primary synuclein disease\u003csup\u003e25\u003c/sup\u003e. Notably, this amygdala-centered pattern is consistent with observations in genetically determined AD, including autosomal-dominant AD (ADAD) and Down syndrome-associated AD (DSAD), where Lewy pathology, when present, also shows a preferential limbic/amygdalar accumulation in the context of established AD biology\u0026nbsp;\u003csup\u003e26,27\u003c/sup\u003e. By contrast, the lack of a consistent \u003cem\u003eAPOE\u003c/em\u003e dose-response for neocortical (diffuse) Lewy pathology is compatible with a distinct pathological trajectory more closely linked to the Lewy body dementia phenotype and a more aggressive course that can be relatively independent of AD burden\u003csup\u003e28\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eTDP-43 pathology, by comparison, mapped much more cleanly onto an age-driven process. While \u0026epsilon;4/\u0026epsilon;4 carriers had slightly higher prevalence than \u0026epsilon;3/\u0026epsilon;3, we did not observe a significant \u0026epsilon;3/\u0026epsilon;4 effect within this amyloid-positive AD dementia cohort, differing from reports of \u0026epsilon;4 dose-dependent associations in broader samples\u003csup\u003e9,29\u003c/sup\u003e. This discrepancy is plausibly explained by conditioning on established AD dementia with confirmed amyloidosis, which reduces heterogeneity but can also attenuate haplotype contrasts when baseline TDP-43 burden is already high. Consistent with this, \u0026epsilon;3/\u0026epsilon;4 effects on TDP-43 pathology emerged when we evaluated all-cause and unconfirmed-amyloid dementia groups (Supplementary Table 11). The dominant signals were again topographic and temporal: TDP-43 increased with age across most haplotypes and showed a shift from amygdala-predominant involvement toward broader limbic and neocortical spread, with dementia duration contributing more modestly than age. This pattern closely matches the defining clinical-pathological trajectory of Limbic-predominant age-related TDP-43 encephalopathy (LATE)\u003csup\u003e30\u003c/sup\u003e, supporting the interpretation that in amyloid-positive AD dementia, TDP-43 accumulation is driven primarily by aging and progressive regional spread, rather than by \u003cem\u003eAPOE\u003c/em\u003e haplotype per se.\u0026nbsp;This age-driven, \u003cem\u003eAPOE\u003c/em\u003e-independent profile is further supported by genetic evidence showing that TDP-43 accumulation in aging brains follows a distinct genetic risk architecture: TMEM106B and GRN variants associate with TDP-43 pathology and hippocampal sclerosis independently of \u003cem\u003eAPOE\u003c/em\u003e haplotype\u003csup\u003e31\u003c/sup\u003e, and TNIP1, recently identified as a risk factor for frontotemporal lobar degeneration with TDP-43 pathology\u003csup\u003e32\u003c/sup\u003e, has been proposed to explain the TNIP1-AD genetic overlap through comorbid TDP-43 rather than amyloid mechanisms\u003csup\u003e33\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe amygdala\u0026rsquo;s broader vulnerability to misfolded protein accumulation across disorders provides a plausible substrate for this convergence\u003csup\u003e34\u003c/sup\u003e. One parsimonious explanation is that the amygdala, together with other phylogenetically older cortices, acts as an early \u0026ldquo;landing zone\u0026rdquo; for multiple age- and AD-linked proteinopathies, with stereotyped staging schemes often beginning in limbic structures (e.g., LATE/TDP-43) before extending cortically\u003csup\u003e30\u003c/sup\u003e. In AD specifically, previous works had shown that \u0026alpha;-synuclein deposition frequently shows an amygdala-predominant pattern, suggesting region-specific susceptibility rather than uniform neocortical spread\u003csup\u003e35\u003c/sup\u003e. Mechanistically, this convergence could reflect a combination of selective neuronal/architectural vulnerability and \u003cstrong\u003einteraction between pathologies,\u0026nbsp;\u003c/strong\u003eincluding cross-seeding or permissive proteostatic/inflammatory milieus generated by AD pathology, supported by experimental work demonstrating tau-\u0026alpha;-synuclein cross-seeding and prion-like propagation\u003csup\u003e36\u003c/sup\u003e. Consistent with this, AD cases with amygdala-predominant \u0026alpha;-synuclein have been reported to show particularly high co-occurring TDP-43 burden and even co-localization of aggregates within limbic neurons, reinforcing the idea of a limbic \u0026ldquo;co-proteinopathy hub\u0026rdquo; rather than independent parallel processes\u003csup\u003e37\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe strong \u003cem\u003eAPOE\u003c/em\u003e-related gradient we observed for CAA is well aligned with prior neuropathology literature showing that \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;4 preferentially amplifies amyloid deposition in the vessel wall\u003csup\u003e10\u003c/sup\u003e. In our amyloid-positive AD dementia cohort, haplotype outweighed both age and dementia duration as a correlate of CAA burden: \u0026epsilon;4/\u0026epsilon;4 carriers had an almost six-fold higher odds of moderate-to-severe CAA compared with \u0026epsilon;3/\u0026epsilon;3, and severity shifted upward with \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;4 gene dose. This pattern mirrors observations in clinicopathologically confirmed AD in which \u0026epsilon;4 is more tightly coupled to severe CAA than to chronological age or disease duration\u003csup\u003e38\u003c/sup\u003e. The contrasting duration associations, positive in \u0026epsilon;3/\u0026epsilon;3 and \u0026epsilon;3/\u0026epsilon;4 but inverse for severe CAA in \u0026epsilon;4/\u0026epsilon;4, suggest that CAA is tightly coupled to AD biology; the \u0026epsilon;4/\u0026epsilon;4 pattern may be best explained by earlier attainment of severe CAA and consequent selection among longer survivors, as CAA is associated with increased mortality risk\u003csup\u003e39\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eIn contrast, non-CAA vascular lesions aligned predominantly with an age-driven trajectory, consistent with the progressive accumulation of cerebrovascular risk factors not captured in our analyses, with substantially weaker \u003cem\u003eAPOE\u003c/em\u003e effects than those seen for CAA. Cerebral arteriolosclerosis showed at most a modest enrichment in \u0026epsilon;4/\u0026epsilon;4 carriers relative to \u0026epsilon;3/\u0026epsilon;3, but the dominant signal across haplotypes was a progressive shift toward greater severity with advancing age. Differences from population-based reports that observed weaker or null \u003cem\u003eAPOE\u003c/em\u003e associations\u003csup\u003e40\u003c/sup\u003e may reflect both cohort composition (our restriction to neuropathologically confirmed AD dementia with amyloid positivity) and analytic choices (separating \u0026epsilon;4 homozygotes rather than combining all \u0026epsilon;4 carriers). Circle of Willis atherosclerosis similarly showed minimal haplotype-related differences, yet increased steadily with age across haplotypes, again aligning with prior neuropathological and cardiovascular pathology\u0026nbsp;reports\u003csup\u003e40,41\u003c/sup\u003e.\u0026nbsp;The additional, smaller association with dementia duration observed for atherosclerosis in our AD dementia sample may reflect cumulative exposure to systemic vascular risk and comorbidity during the clinical course, rather than an AD-specific genetic effect. Overall, these results sharpen a key distinction within amyloid-positive AD dementia: CAA behaves as an \u003cem\u003eAPOE\u003c/em\u003e/amyloid-linked vasculopathy, whereas arteriolosclerosis and large-vessel atherosclerosis largely reflect age- and risk-related cerebrovascular injury, with only limited modulation by \u003cem\u003eAPOE\u003c/em\u003e haplotype.\u003c/p\u003e\n\u003cp\u003eThis study has important strengths and some limitations that inform future research. The large autopsy-confirmed sample and standardized neuropathological assessments provide exceptional statistical power to detect \u003cem\u003eAPOE\u003c/em\u003e haplotype-specific effects across a broad spectrum of copathologies and reduce the diagnostic uncertainty inherent in purely clinical or biomarker-based cohorts. However, the cohort lacks diversity, being predominantly White, which limits generalizability to population-based and ancestrally diverse samples. In addition, the cross-sectional nature of autopsy data and evolving NACC collection protocols, including inter-site variability in pathological ratings, constrain inferences about temporal trajectories of pathology despite the use of harmonized criteria. Additionally, cerebrovascular risk factors such as hypertension, diabetes, or dyslipidemia were not included as covariates in our analyses, which may partially account for the age-driven vascular pathology patterns observed and should be addressed in future studies. Furthermore, our focus on amyloid-positive AD dementia, while ensuring diagnostic homogeneity, precludes assessment of \u003cem\u003eAPOE\u003c/em\u003e effects in amyloid-negative dementia or preclinical stages, where associations may differ. However, as shown in the supplementary material, the NACC database is not a population-based cohort and is highly enriched for individuals with dementia. Therefore, it might not be well-suited to address these questions. These considerations underscore the need for longitudinal, multimodal studies with serial fluid and imaging biomarkers as well as postmortem validation in more diverse and population-based cohorts.\u003c/p\u003e\n\u003cp\u003eIn conclusion, these data support a parsimonious model in which copathologies in amyloid-positive AD dementia distribute across distinct but overlapping axes. One axis is \u003cem\u003eAPOE\u003c/em\u003e/amyloid-linked, dominated by CAA severity and accompanied by selective limbic vulnerability, where Lewy pathology shows its most consistent (albeit modest) \u003cem\u003eAPOE\u003c/em\u003e signal. The other axis is age-driven and relatively \u003cem\u003eAPOE\u003c/em\u003e-independent, encompassing non-CAA vascular disease and TDP-43 pathology. This framework has practical implications for interpreting clinicopathologic variability and suggests that, in late-stage disease, trial stratification may benefit most from capturing amyloid-related vascular vulnerability and limbic-predominant co-proteinopathy patterns, rather than assuming broad \u003cem\u003eAPOE\u003c/em\u003e effects across all copathologies.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003e4.1 Study cohort\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe analyzed data from the March 2025 freeze of the NACC repository (\u003cu\u003ehttps://www.naccdata.org/\u003c/u\u003e), which aggregates de-identified longitudinal clinical and neuropathological data from National Institute on Aging\u0026ndash;funded Alzheimer\u0026rsquo;s Disease Research Centers (ADRCs). ADRCs enroll participants across the cognitive spectrum, from cognitively unimpaired individuals to those with dementia, although most participants have AD dementia and are typically recruited after symptom onset.\u003c/p\u003e\n\u003cp\u003eAll contributing ADRCs obtained local institutional review board approval and written informed consent from participants or their legal representatives, and procedures adhered to the Declaration of Helsinki. NACC maintains a harmonized clinical database based on the Uniform Data Set (UDS) and a complementary neuropathology database based on standardized Neuropathology Forms. For this study, we included participants who had died and undergone brain autopsy and excluded individuals without \u003cem\u003eAPOE\u003c/em\u003e genotyping or carrying a known pathogenic variant associated with familial AD (NPPDXP), frontotemporal dementia (NPPDXQ), or Down syndrome (NACCDOWN). \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.2 Demographic, cognitive, and genetic variables\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe extracted standard demographic variables from NACC. Sex (SEX) was coded as male or female. Race (RACE) was classified as White, Black or African American, American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, Asian, Other, or Unknown. Educational attainment was quantified as total years of formal education (EDUC). \u003cem\u003eAPOE\u003c/em\u003e haplotype was derived from the variable NACCAPOE and categorized as \u0026epsilon;2/\u0026epsilon;2, \u0026epsilon;2/\u0026epsilon;3, \u0026epsilon;3/\u0026epsilon;3, \u0026epsilon;2/\u0026epsilon;4, \u0026epsilon;3/\u0026epsilon;4, or \u0026epsilon;4/\u0026epsilon;4, with \u0026epsilon;3/\u0026epsilon;3 used as the reference group in all comparative analyses, given that it is the most common haplotype in the general population. Ages at diagnosis of mild cognitive impairment (MCI), dementia, and at death were calculated by combining the date of the visit at which MCI or dementia was first diagnosed (year, month, and day) with the recorded date of birth and date of death. Participants were then classified according to their last cognitive status as having dementia, MCI, or cognitively unimpaired (CU).\u003c/p\u003e\n\u003cp\u003eWithin the UDS, the Etiologic Diagnoses section allows clinicians to indicate, for each potential cause of cognitive impairment, whether it is the primary etiology, a contributing factor, or not contributing; each category is coded as present or absent based on clinical judgment and available diagnostic information. For this study, we focused on whether AD was recorded as the primary etiologic diagnosis of cognitive impairment. Participants were therefore classified at the visit closest to death as having clinical AD (AD as the primary etiology) or clinical non-AD (a non-AD primary etiology).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.3 Neuropathological variables\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNeuropathological data were obtained from NACC Neuropathology Forms (versions 10\u0026ndash;11) completed after autopsy\u003csup\u003e14\u003c/sup\u003e. ADNC was obtained from variable NPADNC, which integrates three components into a composite score (Not, Low, Intermediate, or High): Thal phase for amyloid plaque distribution (A score)\u003csup\u003e15\u003c/sup\u003e, Braak stage for neurofibrillary tangle burden (B score)\u003csup\u003e16\u003c/sup\u003e, and the Consortium to Establish a Registry for Alzheimer\u0026apos;s Disease (CERAD) neuritic plaque density (C score)\u003csup\u003e17\u003c/sup\u003e. We used CERAD semi-quantitative assessments of neuritic plaque burden (NACCNEUR: No, Sparse, Moderate, or Frequent neuritic plaques) and Braak stage for neurofibrillary tangle burden (NACCBRAA: Stage 0\u0026ndash;VI; recoded as None or stages I\u0026ndash;II, III\u0026ndash;IV, V\u0026ndash;VI). Given the central role of amyloid biomarkers in research and clinical practice, CERAD neuritic plaque severity was collapsed into a binary amyloid status: individuals with \u0026quot;No\u0026quot; or \u0026quot;Sparse\u0026quot; neuritic plaques were considered amyloid-negative, and those with \u0026quot;Moderate\u0026quot; or \u0026quot;Frequent\u0026quot; plaques as amyloid-positive. This dichotomization was chosen to approximate amyloid PET positivity and align with current AD neuropathologic change frameworks\u003csup\u003e18\u0026ndash;20\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eLewy body pathology was assessed using the NACC-derived variable NACCLEWY, which classifies cases as: no Lewy body pathology, brainstem-predominant, limbic or amygdala-predominant, neocortical, or Lewy bodies present in unspecified region/olfactory bulb. These categories are broadly consistent with the DLB Consortium neuropathological classification framework\u003csup\u003e21\u003c/sup\u003e, though NACC combines limbic and amygdala-predominant patterns into a single category. TDP-43 immunoreactive inclusions were assessed across five brain regions, spinal cord (NPTDPA), amygdala (NPTDPB), hippocampus (NPTDPC), entorhinal/inferior temporal cortex (NPTDPD), and neocortex (NPTDPE), as presence or absence; regional distribution patterns are consistent with LATE-NC staging criteria\u003csup\u003e22,23\u003c/sup\u003e. Cerebrovascular pathologies were graded using NACC semi-quantitative scales (none, mild, moderate, severe), including cerebral amyloid angiopathy (NACCAMY), arteriolosclerosis (NACCARTE), and Circle of Willis atherosclerosis (NACCAVAS)\u003csup\u003e14\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.5 Statistical analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCategorical variables were summarized as counts and percentages by \u003cem\u003eAPOE\u003c/em\u003e haplotype (i.e., \u0026epsilon;2/\u0026epsilon;X, \u0026epsilon;3/\u0026epsilon;3, \u0026epsilon;3/\u0026epsilon;4, \u0026epsilon;4/\u0026epsilon;4), and continuous variables were reported as medians with interquartile ranges. To examine associations between \u003cem\u003eAPOE\u003c/em\u003e haplotype and key clinical and pathological outcomes at death, we fitted binary logistic regression models using \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;3/\u0026epsilon;3 as the reference category. \u003c/p\u003e\n\u003cp\u003eGiven the low representation of \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;2/\u0026epsilon;2 (n = 44) and \u0026epsilon;2/\u0026epsilon;3 (n = 948), these groups were combined into a single category (\u0026epsilon;2/\u0026epsilon;X) for neuropathological analyses to ensure adequate statistical power. Individuals with the \u0026epsilon;2/\u0026epsilon;4 haplotype (n = 331) were excluded due to the opposing effects of \u003cem\u003eAPOE\u003c/em\u003e \u0026epsilon;2 and \u0026epsilon;4 alleles on AD risk and the small sample size.\u003c/p\u003e\n\u003cp\u003eNeuropathological outcomes were analyzed using regression models. For binary outcomes (presence or absence of pathology), we applied logistic regression. For ordered severity measures (Lewy body pathology distribution, TDP-43 pathology stage, and vascular pathology severity grades), we used proportional odds ordinal logistic regression. Models were fitted unadjusted and then progressively adjusted, first for sex and years of education, then additionally for age at death or age at dementia diagnosis as appropriate, and finally with the addition of AD neuropathologic change (ADNC) classification. ORs with 95% CIs were estimated from these models. Global effects of \u003cem\u003eAPOE \u003c/em\u003ehaplotype, age at death, and dementia duration were assessed using LR\u0026chi;\u0026sup2;. Given that progressive model adjustments represent sequential refinements of the same associations rather than independent tests, we did not apply corrections for multiple comparisons. Interpretation emphasized both effect magnitude (ORs with 95% CIs) and statistical significance.\u003c/p\u003e\n\u003cp\u003eAnalyses of Lewy body pathology, TDP-43 pathology, CAA, arteriolosclerosis, and Circle of Willis atherosclerosis were restricted to participants with a clinical diagnosis of AD dementia and neuropathologically confirmed amyloid positivity (moderate or frequent neuritic plaques by CERAD criteria). Age-related analyses used the following age-at-death categories: \u0026lt;65, 65\u0026ndash;69, 70\u0026ndash;74, 75\u0026ndash;79, 80\u0026ndash;84, 85\u0026ndash;89, \u0026ge;90 years. Dementia duration was categorized as \u0026lt;2, 2\u0026ndash;3, 4\u0026ndash;6, 6\u0026ndash;8, and \u0026ge;10 years to capture clinically distinct phases of disease progression while maintaining sufficient cases per group. Distributions of neuropathological features across these categories were assessed using chi-squared tests (\u0026chi;\u0026sup2;).\u003c/p\u003e\n\u003cp\u003eAdditional analyses are described in the Supplementary Methodology, including: (1) cohort characterization and rationale for restricting analyses to amyloid-positive AD dementia cases (Supplementary Figure 1, Supplementary Tables 1A-B and 2); (2) Braak neurofibrillary tangle staging across multiple diagnostic groups (all-cause dementia, AD dementia, and amyloid-positive AD dementia) (Supplementary Figure 2, Supplementary Tables 3 and 4); (3) \u003cem\u003eAPOE\u003c/em\u003e \u0026times; age interaction testing for regional Lewy body, TDP-43, and cerebrovascular pathologies (Supplementary Figures 3, 4 and 5), (4) sex-stratified sensitivity analyses for all neuropathologies (Supplementary Table 10), and (5) TDP-43 pathology across diagnostic groups (Supplementary Table 11).\u003c/p\u003e\n\u003cp\u003eAll statistical analyses were performed using R (version 4.2.2). Tests were two-sided, and P \u0026lt; 0.05 was considered statistically significant.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data that support the findings of this study are available from the National Alzheimer\u0026apos;s Coordinating Center (NACC; https://www.naccdata.org/) upon registration and approval.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe NACC database is funded by NIA/NIH Grant U24 AG072122. NACC data are contributed by the NIA-funded ADRCs: P30 AG062429 (PI James Brewer, MD, PhD), P30 AG066468 (PI Oscar Lopez, MD), P30 AG062421 (PI Bradley Hyman, MD, PhD), P30 AG066509 (PI Thomas Grabowski, MD), P30 AG066514 (PI Mary Sano, PhD), P30 AG066530 (PI Helena Chui, MD), P30 AG066507 (PI Marilyn Albert, PhD), P30 AG066444 (PI John Morris, MD), P30 AG066518 (PI Jeffrey Kaye, MD), P30 AG066512 (PI Thomas Wisniewski, MD), P30 AG066462 (PI Scott Small, MD), P30 AG072979 (PI David Wolk, MD), P30 AG072972 (PI Charles DeCarli, MD), P30 AG072976 (PI Andrew Saykin, PsyD), P30 AG072975 (PI David Bennett, MD), P30 AG072978 (PI Neil Kowall, MD), P30 AG072977 (PI Robert Vassar, PhD), P30 AG066519 (PI Frank LaFerla, PhD), P30 AG062677 (PI Ronald Petersen, MD, PhD), P30 AG079280 (PI Eric Reiman, MD), P30 AG062422 (PI Gil Rabinovici, MD), P30 AG066511 (PI Allan Levey, MD, PhD), P30 AG072946 (PI Linda Van Eldik, PhD), P30 AG062715 (PI Sanjay Asthana, MD, FRCP), P30 AG072973 (PI Russell Swerdlow, MD), P30 AG066506 (PI Todd Golde, MD, PhD), P30 AG066508 (PI Stephen Strittmatter, MD, PhD), P30 AG066515 (PI Victor Henderson, MD, MS), P30 AG072947 (PI Suzanne Craft, PhD), P30 AG072931 (PI Henry Paulson, MD, PhD), P30 AG066546 (PI Sudha Seshadri, MD), P20 AG068024 (PI Erik Roberson, MD, PhD), P20 AG068053 (PI Justin Miller, PhD), P20 AG068077 (PI Gary Rosenberg, MD), P20 AG068082 (PI Angela Jefferson, PhD), P30 AG072958 (PI Heather Whitson, MD), P30 AG072959 (PI James Leverenz, MD). We acknowledge the Support for Research Groups funding from the Department of Research and Universities from the Generalitat de Catalunya (2021 SGR 00979).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eI.R.B. and J.F. conceived and designed the study, performed the statistical analyses, and wrote the manuscript. C.K.S., L.M.P., I.A., N.V.F., and R.S.P. supervised neuropathological data. L.V.A., L.M.B., L.P., J.A., L.V., I.B., M.C.I., M.B.S.S., J.S.G., O.D.I., M.R.A., S.S., C.A., I.I.G., D.A., and A.L. contributed to the interpretation of results and critically reviewed the manuscript. All authors read and approved the final version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eI. Rodr\u0026iacute;guez-Baz has received personal fees for service on advisory boards, speaker honoraria, or educational activities from Nutricia, Almirall, and Esteve. J. Arranz has received personal fees for service on advisory boards, speaker honoraria, or educational activities from Esteve, Lilly, and Roche Diagnostics. L. Molina-Porcel has provided consultancy services to Biogen. M. Carmona-Iragui has received personal fees for service on advisory boards, speaker honoraria, or educational activities from IMSERSO, Esteve, Lilly, Neuraxpharm, Adium Pharma, and Roche. M.R. Aranha is a co-founder and partner of Masima Solu\u0026ccedil;\u0026otilde;es em Imagens M\u0026eacute;dicas LTDA (Porto Alegre, Brazil) and serves as an independent consultant to Ionis Pharmaceuticals. I. Ill\u0026aacute;n-Gala has participated in advisory boards from UCB and Nutricia, and received speaker honoraria from Almirall, Esteve Pharmaceuticals S.A., Kern Pharma, Krka Farmac\u0026eacute;utica S.L., Lilly, Nutricia, and Zambon S.A.U. D. Alcolea has received personal fees for advisory board services and/or speaker honoraria from Fujirebio-Europe, Roche, Nutricia, Krka Farmac\u0026eacute;utica, Lilly, Zambon S.A.U., Grifols, and Esteve, outside the submitted work. A. Lle\u0026oacute; has served as a consultant or on advisory boards for Fujirebio-Europe, Roche, Biogen, Grifols, Novartis, Eisai, Lilly, and Nutricia, outside the submitted work. C.K. Suemoto is funded by the Alzheimer\u0026apos;s Association (AARG-20-678884 and 24CBIDR-1185483) and the S\u0026atilde;o Paulo Research Foundation (FAPESP 2024/03917-7). J. Fortea has served on advisory boards, adjudication committees, or as a speaker for Roche, NovoNordisk, Esteve, Biogen, Laboratorios Carnot, Adamed, LMI, Novartis, Lundbeck, AC Immune, Alzheon, Zambon, Lilly, the Spanish Neurological Society, T21 Research Society, Lumind Foundation, J\u0026eacute;r\u0026ocirc;me-Lejeune Foundation, Alzheimer\u0026apos;s Association, National Institutes of Health USA, and Instituto de Salud Carlos III. D. Alcolea, A. Lle\u0026oacute;, and J. Fortea hold a patent for markers of synaptopathy in neurodegenerative disease (licensed to ADx, EP8382175.0). A. Lle\u0026oacute; is co-author of a patent on antibodies for amyloid precursor protein, methods and uses thereof (European priority No. EP25382226). No other competing interests were reported.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eI. Rodr\u0026iacute;guez-Baz was supported by Instituto de Salud Carlos III (ISCIII), co-funded by the European Union (CM22/00052, CD25/00223), and by the Alzheimer\u0026apos;s Association (AACSF-25-1486364). L. Vaqu\u0026eacute;-Alc\u0026aacute;zar was supported by a Sara Borrell postdoctoral fellowship from ISCIII, Spain (CD23/00235). L. Maure-Blesa was supported by ISCIII through a R\u0026iacute;o Hortega fellowship (CM23/00291), co-funded by the European Union. J. Arranz was supported by ISCIII through a R\u0026iacute;o Hortega fellowship (CM22/00243), co-funded by the European Union. R. Silva-Paradela was funded by the Alzheimer\u0026apos;s Association Capacity Building in International Dementia Research (CBIDR) Program (24CBIDR-1185483) and the Global Brain Health Institute, Alzheimer\u0026apos;s Association, and Alzheimer\u0026apos;s Society (GBHI ALZ UK-25-1289657). M. Carmona-Iragui was supported by ISCIII (PI18/00335, PI22/00758, ICI23/00032), CIBERNED Program 1 (partly co-funded by FEDER, European Union), the Alzheimer\u0026apos;s Association (AARG-22-973966), the Global Brain Health Institute (GBHI_ALZ-18-543740), and the J\u0026eacute;r\u0026ocirc;me Lejeune Foundation (#1913 cycle 2019B; #2425 cycle 2024B). J. Selma-Gonz\u0026aacute;lez is supported by the Contratos Predoctorales de Formaci\u0026oacute;n en Investigaci\u0026oacute;n en Salud program (FI25/00235) associated to project PI24/00598 (I. Ill\u0026aacute;n-Gala), funded by ISCIII, Spain. C. Abdelnour received support from the Susan and Charles Berghoff Foundation and the ARISTOS program, funded by the European Union\u0026apos;s Horizon Europe research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 101081334. O. Dols-Icardo receives funding from Fundaci\u0026oacute;n Espa\u0026ntilde;ola para el Fomento de la Investigaci\u0026oacute;n de la Esclerosis Lateral Amiotr\u0026oacute;fica (FUNDELA), Fundaci\u0026oacute;n HNA, the Alzheimer\u0026apos;s Association (AARF-22-924456), and the Fondation J\u0026eacute;r\u0026ocirc;me Lejeune (PDC-2023-51; #202307), and from ISCIII (PI21/01395, PI24/01087), co-funded by FEDER, European Union. I. Ill\u0026aacute;n-Gala was supported by ISCIII (PI21/00791 and PI24/00598), co-funded by FEDER, European Union; the Alzheimer\u0026apos;s Association (AACSF-21-850193); the Alzheimer Society (GBHI ALZ UK-21-72097); the Global Brain Health Institute as a senior Atlantic Fellow for Equity in Brain Health; and the Juan Rod\u0026eacute;s Contract (JR20/0018) from ISCIII, partly funded by the European Social Fund. D. Alcolea received funding from ISCIII (PI18/00435, PI22/00611, PI25/00422, INT23/00048), co-funded by FEDER, European Union, and from the Department of Health of the Generalitat de Catalunya through the PERIS programme (SLT006/17/125, SLT042/25/000034) and the Department of Research and Universities of the Generalitat de Catalunya (2021 SGR 00979). A. Lle\u0026oacute; received funding from ISCIII (PI14/1561, PI20/01330), co-funded by FEDER, European Union. C.K. Suemoto was funded by the Alzheimer\u0026apos;s Association (AARG-20-678884 and 24CBIDR-1185483) and the S\u0026atilde;o Paulo Research Foundation (FAPESP 2024/03917-7). J. Fortea received funding from ISCIII (PI20/01473, PI23/01786, INT21/00073), co-funded by FEDER, European Union; CIBERNED Program 1; the National Institutes of Health (1R01AG056850-01A1, 3RF1AG056850-01S1, AG056850, R21AG056974, R01AG061566, 1R01AG081394-01, 1R61AG066543-01); the Department of Health of the Generalitat de Catalunya (SLT006/17/00119); Fundaci\u0026oacute;n Tatiana P\u0026eacute;rez de Guzm\u0026aacute;n el Bueno (IIBSP-DOW-2020-151); the Horizon 2020 Research and Innovation Programme of the European Union (MES-CoBraD, GA 965422); BrightFocus Foundation; and Life Molecular Imaging (LMI).\u003c/p\u003e\n"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eRabinovici, G. D. \u003cem\u003eet al.\u003c/em\u003e Multiple comorbid neuropathologies in the setting of Alzheimer\u0026rsquo;s disease neuropathology and implications for drug development. \u003cem\u003eAlzheimers Dement.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 83\u0026ndash;91 (2017).\u003c/li\u003e\n\u003cli\u003eBoyle, P. A. \u003cem\u003eet al.\u003c/em\u003e Attributable risk of Alzheimer\u0026rsquo;s dementia attributed to age-related neuropathologies. \u003cem\u003eAnn. Neurol.\u003c/em\u003e \u003cstrong\u003e85\u003c/strong\u003e, 114\u0026ndash;124 (2019).\u003c/li\u003e\n\u003cli\u003eSepulveda-Falla, D. \u003cem\u003eet al.\u003c/em\u003e Comorbidities in early-onset sporadic versus presenilin-1 mutation-associated Alzheimer disease dementia: Evidence for dependency on Alzheimer disease neuropathological changes. \u003cem\u003eJ. Neuropathol. Exp. Neurol.\u003c/em\u003e \u003cstrong\u003e84\u003c/strong\u003e, 104\u0026ndash;113 (2025).\u003c/li\u003e\n\u003cli\u003eIchimata, S. \u003cem\u003eet al.\u003c/em\u003e Patterns of Mixed Pathologies in Down Syndrome. \u003cem\u003eJ. Alzheimers Dis. JAD\u003c/em\u003e \u003cstrong\u003e87\u003c/strong\u003e, 595\u0026ndash;607 (2022).\u003c/li\u003e\n\u003cli\u003eReiman, E. M. \u003cem\u003eet al.\u003c/em\u003e Exceptionally low likelihood of Alzheimer\u0026rsquo;s dementia in APOE2 homozygotes from a 5,000-person neuropathological study. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 667 (2020).\u003c/li\u003e\n\u003cli\u003eLi, Z., Shue, F., Zhao, N., Shinohara, M. \u0026amp; Bu, G. APOE2: protective mechanism and therapeutic implications for Alzheimer\u0026rsquo;s disease. \u003cem\u003eMol. Neurodegener.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 63 (2020).\u003c/li\u003e\n\u003cli\u003eDickson, D. W. \u003cem\u003eet al.\u003c/em\u003e APOE \u0026epsilon;4 is associated with severity of Lewy body pathology independent of Alzheimer pathology. \u003cem\u003eNeurology\u003c/em\u003e \u003cstrong\u003e91\u003c/strong\u003e, e1182\u0026ndash;e1195 (2018).\u003c/li\u003e\n\u003cli\u003eJin, Y. \u003cem\u003eet al.\u003c/em\u003e APOE4 exacerbates \u0026alpha;-synuclein seeding activity and contributes to neurotoxicity in Alzheimer\u0026rsquo;s disease with Lewy body pathology. \u003cem\u003eActa Neuropathol. (Berl.)\u003c/em\u003e \u003cstrong\u003e143\u003c/strong\u003e, 641\u0026ndash;662 (2022).\u003c/li\u003e\n\u003cli\u003eYang, H.-S. \u003cem\u003eet al.\u003c/em\u003e Evaluation of TDP-43 proteinopathy and hippocampal sclerosis in relation to APOE \u0026epsilon;4 haplotype status: a community-based cohort study. \u003cem\u003eLancet Neurol.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 773\u0026ndash;781 (2018).\u003c/li\u003e\n\u003cli\u003eNelson, P. T. \u003cem\u003eet al.\u003c/em\u003e APOE-\u0026epsilon;2 and APOE-\u0026epsilon;4 Correlate with Increased Amyloid Accumulation in Cerebral Vasculature. \u003cem\u003eJ. Neuropathol. Exp. Neurol.\u003c/em\u003e \u003cstrong\u003e72\u003c/strong\u003e, 708\u0026ndash;715 (2013).\u003c/li\u003e\n\u003cli\u003eFortea, J. \u003cem\u003eet al.\u003c/em\u003e APOE4 homozygosity represents a distinct genetic form of Alzheimer\u0026rsquo;s disease. \u003cem\u003eNat. Med.\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 1284\u0026ndash;1291 (2024).\u003c/li\u003e\n\u003cli\u003eOssenkoppele, R. \u003cem\u003eet al.\u003c/em\u003e Tau PET positivity in individuals with and without cognitive impairment varies with age, amyloid-\u0026beta; status, APOE genotype and sex. \u003cem\u003eNat. Neurosci.\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 1610\u0026ndash;1621 (2025).\u003c/li\u003e\n\u003cli\u003eCicognola, C. \u003cem\u003eet al.\u003c/em\u003e APOE4 impact on soluble and insoluble tau pathology is mostly influenced by amyloid-\u0026beta;. \u003cem\u003eBrain J. Neurol.\u003c/em\u003e \u003cstrong\u003e148\u003c/strong\u003e, 2373\u0026ndash;2383 (2025).\u003c/li\u003e\n\u003cli\u003eBesser, L. M. \u003cem\u003eet al.\u003c/em\u003e The Revised National Alzheimer\u0026rsquo;s Coordinating Center\u0026rsquo;s Neuropathology Form-Available Data and New Analyses. \u003cem\u003eJ. Neuropathol. Exp. Neurol.\u003c/em\u003e \u003cstrong\u003e77\u003c/strong\u003e, 717\u0026ndash;726 (2018).\u003c/li\u003e\n\u003cli\u003eThal, D. R., R\u0026uuml;b, U., Orantes, M. \u0026amp; Braak, H. Phases of A beta-deposition in the human brain and its relevance for the development of AD. \u003cem\u003eNeurology\u003c/em\u003e \u003cstrong\u003e58\u003c/strong\u003e, 1791\u0026ndash;1800 (2002).\u003c/li\u003e\n\u003cli\u003eBraak, H., Alafuzoff, I., Arzberger, T., Kretzschmar, H. \u0026amp; Del Tredici, K. Staging of Alzheimer disease-associated neurofibrillary pathology using paraffin sections and immunocytochemistry. \u003cem\u003eActa Neuropathol. (Berl.)\u003c/em\u003e \u003cstrong\u003e112\u003c/strong\u003e, 389\u0026ndash;404 (2006).\u003c/li\u003e\n\u003cli\u003eMirra, S. S. \u003cem\u003eet al.\u003c/em\u003e The Consortium to Establish a Registry for Alzheimer\u0026rsquo;s Disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer\u0026rsquo;s disease. \u003cem\u003eNeurology\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 479\u0026ndash;486 (1991).\u003c/li\u003e\n\u003cli\u003eClark, C. M. \u003cem\u003eet al.\u003c/em\u003e Cerebral PET with florbetapir compared with neuropathology at autopsy for detection of neuritic amyloid-\u0026beta; plaques: a prospective cohort study. \u003cem\u003eLancet Neurol.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 669\u0026ndash;678 (2012).\u003c/li\u003e\n\u003cli\u003eLa Joie, R. \u003cem\u003eet al.\u003c/em\u003e Multisite study of the relationships between \u003cem\u003eantemortem\u003c/em\u003e [\u003csup\u003e11\u003c/sup\u003e C]PIB‐PET Centiloid values and \u003cem\u003epostmortem\u003c/em\u003e measures of Alzheimer\u0026rsquo;s disease neuropathology. \u003cem\u003eAlzheimers Dement.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 205\u0026ndash;216 (2019).\u003c/li\u003e\n\u003cli\u003eJack, C. R. \u003cem\u003eet al.\u003c/em\u003e Revised criteria for diagnosis and staging of Alzheimer\u0026rsquo;s disease: Alzheimer\u0026rsquo;s Association Workgroup. \u003cem\u003eAlzheimers Dement.\u003c/em\u003e alz.13859 (2024) doi:10.1002/alz.13859.\u003c/li\u003e\n\u003cli\u003eMcKeith, I. G. \u003cem\u003eet al.\u003c/em\u003e Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. \u003cem\u003eNeurology\u003c/em\u003e \u003cstrong\u003e65\u003c/strong\u003e, 1863\u0026ndash;1872 (2005).\u003c/li\u003e\n\u003cli\u003eJosephs, K. A. \u003cem\u003eet al.\u003c/em\u003e Updated TDP-43 in Alzheimer\u0026rsquo;s disease staging scheme. \u003cem\u003eActa Neuropathol. (Berl.)\u003c/em\u003e \u003cstrong\u003e131\u003c/strong\u003e, 571\u0026ndash;585 (2016).\u003c/li\u003e\n\u003cli\u003eNelson, P. T. \u003cem\u003eet al.\u003c/em\u003e LATE-NC staging in routine neuropathologic diagnosis: an update. \u003cem\u003eActa Neuropathol. (Berl.)\u003c/em\u003e \u003cstrong\u003e145\u003c/strong\u003e, 159\u0026ndash;173 (2023).\u003c/li\u003e\n\u003cli\u003eSuemoto, C. K. \u003cem\u003eet al.\u003c/em\u003e Apolipoprotein E and Alzheimer\u0026rsquo;s disease pathology in a diverse autopsy study. \u003cem\u003eAlzheimers Dement.\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, e70927 (2025).\u003c/li\u003e\n\u003cli\u003eBaiardi, S., Hansson, O., Levin, J. \u0026amp; Parchi, P. In vivo detection of Alzheimer\u0026rsquo;s and Lewy body disease concurrence: Clinical implications and future perspectives. \u003cem\u003eAlzheimers Dement.\u003c/em\u003e alz.14039 (2024) doi:10.1002/alz.14039.\u003c/li\u003e\n\u003cli\u003eLevin, J. \u003cem\u003eet al.\u003c/em\u003e \u0026alpha;-Synuclein seed amplification assay detects Lewy body co-pathology in autosomal dominant Alzheimer\u0026rsquo;s disease late in the disease course and dependent on Lewy pathology burden. \u003cem\u003eAlzheimers Dement.\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 4351\u0026ndash;4365 (2024).\u003c/li\u003e\n\u003cli\u003eBernhardt, A. M. \u003cem\u003eet al.\u003c/em\u003e Alpha-synuclein co-pathology in Down syndrome-associated Alzheimer\u0026rsquo;s disease. \u003cem\u003eAlzheimers Dement.\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, e70342 (2025).\u003c/li\u003e\n\u003cli\u003eGraff-Radford, J. \u003cem\u003eet al.\u003c/em\u003e Duration and Pathologic Correlates of Lewy Body Disease. \u003cem\u003eJAMA Neurol.\u003c/em\u003e \u003cstrong\u003e74\u003c/strong\u003e, 310\u0026ndash;315 (2017).\u003c/li\u003e\n\u003cli\u003eWennberg, A. M. \u003cem\u003eet al.\u003c/em\u003e Association of Apolipoprotein E \u0026epsilon;4 With Transactive Response DNA-Binding Protein 43. \u003cem\u003eJAMA Neurol.\u003c/em\u003e \u003cstrong\u003e75\u003c/strong\u003e, 1347\u0026ndash;1354 (2018).\u003c/li\u003e\n\u003cli\u003eNelson, P. T. \u003cem\u003eet al.\u003c/em\u003e Limbic-predominant age-related TDP-43 encephalopathy (LATE): consensus working group report. \u003cem\u003eBrain\u003c/em\u003e \u003cstrong\u003e142\u003c/strong\u003e, 1503\u0026ndash;1527 (2019).\u003c/li\u003e\n\u003cli\u003eNelson, P. T. \u003cem\u003eet al.\u003c/em\u003e Limbic-predominant age-related TDP-43 encephalopathy (LATE-NC): Co-pathologies and genetic risk factors provide clues about pathogenesis. \u003cem\u003eJ. Neuropathol. Exp. Neurol.\u003c/em\u003e \u003cstrong\u003e83\u003c/strong\u003e, 396\u0026ndash;415 (2024).\u003c/li\u003e\n\u003cli\u003ePottier, C. \u003cem\u003eet al.\u003c/em\u003e Deciphering distinct genetic risk factors for FTLD-TDP pathological subtypes via whole-genome sequencing. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 3914 (2025).\u003c/li\u003e\n\u003cli\u003eDols-Icardo, O., Reus, L. M. \u0026amp; Ramirez, A. Genetic architecture of FTLD-TDP pathological subtypes. \u003cem\u003eTrends Neurosci.\u003c/em\u003e \u003cstrong\u003e48\u003c/strong\u003e, 553\u0026ndash;554 (2025).\u003c/li\u003e\n\u003cli\u003eNelson, P. T. \u003cem\u003eet al.\u003c/em\u003e The Amygdala as a Locus of Pathologic Misfolding in Neurodegenerative Diseases. \u003cem\u003eJ. Neuropathol. Exp. Neurol.\u003c/em\u003e \u003cstrong\u003e77\u003c/strong\u003e, 2\u0026ndash;20 (2018).\u003c/li\u003e\n\u003cli\u003eNeubauer, A. \u003cem\u003eet al.\u003c/em\u003e Alpha-synuclein deposition patterns in Alzheimer\u0026rsquo;s disease: association with cortical amyloid beta and variable tau load. \u003cem\u003eActa Neuropathol. (Berl.)\u003c/em\u003e \u003cstrong\u003e150\u003c/strong\u003e, 46 (2025).\u003c/li\u003e\n\u003cli\u003eWilliams, T., Sorrentino, Z., Weinrich, M., Giasson, B. I. \u0026amp; Chakrabarty, P. Differential cross-seeding properties of tau and \u0026alpha;-synuclein in mouse models of tauopathy and synucleinopathy. \u003cem\u003eBrain Commun.\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, fcaa090 (2020).\u003c/li\u003e\n\u003cli\u003eGawor, K. \u003cem\u003eet al.\u003c/em\u003e Amygdala-predominant \u0026alpha;-synuclein pathology is associated with exacerbated hippocampal neuron loss in Alzheimer\u0026rsquo;s disease. \u003cem\u003eBrain Commun.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, fcae442 (2024).\u003c/li\u003e\n\u003cli\u003eRingman, J. M. \u003cem\u003eet al.\u003c/em\u003e Clinical predictors of severe cerebral amyloid angiopathy and influence of APOE genotype in persons with pathologically-verified Alzheimer\u0026rsquo;s disease. \u003cem\u003eJAMA Neurol.\u003c/em\u003e \u003cstrong\u003e71\u003c/strong\u003e, 878\u0026ndash;883 (2014).\u003c/li\u003e\n\u003cli\u003eBruce, S. S. \u003cem\u003eet al.\u003c/em\u003e Prevalence of Cerebral Amyloid Angiopathy and Associated Risk of Subsequent Ischemic and Hemorrhagic Stroke and Mortality in a Nationwide Cohort. \u003cem\u003eAnn. Neurol.\u003c/em\u003e \u003cstrong\u003e98\u003c/strong\u003e, 249\u0026ndash;257 (2025).\u003c/li\u003e\n\u003cli\u003eLamar, M. \u003cem\u003eet al.\u003c/em\u003e APOE genotypes as a risk factor for age-dependent accumulation of cerebrovascular disease in older adults. \u003cem\u003eAlzheimers Dement.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 258\u0026ndash;266 (2019).\u003c/li\u003e\n\u003cli\u003eParadela, R. S. \u003cem\u003eet al.\u003c/em\u003e Apolipoprotein E genotypes were not associated with intracranial atherosclerosis: a population-based autopsy study. \u003cem\u003eCardiovasc. Pathol.\u003c/em\u003e \u003cstrong\u003e62\u003c/strong\u003e, 107479 (2023).\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table 1","content":"\u003cp\u003eTable 1 is available in the Supplementary Files section.\u003c/p\u003e\n"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9044264/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9044264/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The mechanisms by which apolipoprotein E (\u003ci\u003eAPOE\u003c/i\u003e) drives copathologies in established Alzheimer’s disease (AD) dementia via amyloid-dependent versus age-driven pathways remain unresolved. Analyzing data from 11,897 autopsied individuals from the National Alzheimer's Coordinating Center, with copathology analyses restricted to amyloid-positive AD dementia, we show that \u003ci\u003eAPOE\u003c/i\u003e effects followed two distinct trajectories. Cerebral amyloid angiopathy exhibited a striking ε4 dose-response (OR = 5.76, 95% CI: 4.20–7.96, p \u003c 0.001; for ε4/ε4 compared to ε3/ε3), whereas arteriolosclerosis and atherosclerosis risk increased with age, independent of \u003ci\u003eAPOE\u003c/i\u003e haplotype. Lewy body pathology showed modest \u003ci\u003eAPOE\u003c/i\u003e associations restricted to limbic/amygdalar-predominant forms and was related to dementia duration, suggesting AD-mediated secondary synucleinopathy. TDP-43 pathology was associated with chronological age, demonstrating regional progression with minimal \u003ci\u003eAPOE\u003c/i\u003e dependence. These findings suggests that in amyloid-positive AD dementia, \u003ci\u003eAPOE\u003c/i\u003e ε4 selectively amplifies amyloid-related pathology, particularly cerebral amyloid angiopathy, while other copathologies accumulate through age-driven, \u003ci\u003eAPOE\u003c/i\u003e haplotype-independent processes.","manuscriptTitle":"Amyloid-linked versus age-driven copathologies in Alzheimer’s dementia: differential associations with APOE ε4","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-16 06:41:19","doi":"10.21203/rs.3.rs-9044264/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"b6ee6f7e-df3a-45fc-ab8b-89e7046ef707","owner":[],"postedDate":"March 16th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"This content is not available.","date":"2026-05-11T08:50:46+00:00","index":3,"fulltext":"This content is not available."},{"type":"editorInvitedReview","content":"This content is not available.","date":"2026-05-08T14:52:24+00:00","index":2,"fulltext":"This content is not available."}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":64472488,"name":"Health sciences/Neurology/Neurological disorders/Dementia/Alzheimer's disease"},{"id":64472489,"name":"Health sciences/Diseases/Neurological disorders/Dementia"},{"id":64472490,"name":"Biological sciences/Genetics/Clinical genetics/Disease genetics"},{"id":64472491,"name":"Biological sciences/Neuroscience/Genetics of the nervous system"},{"id":64472492,"name":"Biological sciences/Neuroscience/Diseases of the nervous system/Dementia/Alzheimer's disease"}],"tags":[],"updatedAt":"2026-03-16T06:41:20+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-16 06:41:19","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9044264","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9044264","identity":"rs-9044264","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}