Athérosclérose induite par l'apolipoprotéine B comme base causale de la maladie d'Alzheimer

Une perspective multidisciplinaire sur l'axe cardiorénal-cérébral

Résumé

Contexte. Le cadre neurocentrique dominant définit la maladie d'Alzheimer (MA) sur le plan biologique par l'amyloïde et la tau biomarqueurs et traite la pathologie vasculaire comme un modificateur fréquent mais secondaire. Les données génétiques, épidémiologiques et d'imagerie étayent de plus en plus une autre lecture dans laquelle l'exposition tout au long de la vie à apolipoprotéine BApoB)-contenant lipoprotéines est un facteur causal en amont de l'atteinte cérébro-vasculaire et des petits vaisseaux qui, dans un sous-groupe substantiel de patients, précède et accélère la pathologie caractéristique de la MA.

Objectif. Synthétiser les données probantes actuelles reliant l'exposition à vie à l'apoB aux lésions athérosclérotiques dans les lits vasculaires carotidien, intracrânien et rénal et, par conséquent, au syndrome cognitif cliniquement diagnostiqué comme la maladie d'Alzheimer.

Méthodes. Structuré revue narrative d'études primaires indexées sur PubMed, Randomisation mendélienne analyses, des cohortes d'imagerie longitudinale, des essais randomisés de prévention cardiovasculaire et de la démence, ainsi que des déclarations de consensus actuelles de l'American Heart Association, de la National Lipid Association et de l'Alzheimer’s Association. Les allégations de causalité ont été évaluées par rapport aux normes de reporting STROBE-MR.29]

Conclusions. L'exposition tout au long de la vie aux lipoprotéines contenant l'ApoB est établie de manière causale comme un facteur déterminant de la maladie vasculaire athérogène. Des données convergentes issues de la randomisation mendélienne ciblant les médicaments randomisation chez plus d'un million d'individus,7] imagerie TEP longitudinale intracrânienne sténose,[13] et des essais randomisés de réduction du risque vasculaire [14] soutient une contribution causale de cette lésion vasculaire au déclin cognitif, à troubles cognitifs vasculaires et démence, et aux phénotypes de démence à pathologie mixte qui dominent dans les séries d'autopsies de sujets âgés.2,32L'extension d'une lésion cérébrovasculaire à la maladie d'Alzheimer confirmée par biomarqueurs telle qu'définie sous le Critères révisés 2024 du NIA-AA[40] est pris en charge, mais avec une confiance proportionnellement plus faible, et la langue de ce manuscrit est nuancée en conséquence à ce point final. Les critères NIA-AA de 2024 sont un cadre diagnostique, et non une affirmation étiologique ; ils sont entièrement compatibles avec un rôle pathogène indépendant établi pour l'amyloïde-β (démontré le plus directement par la MA familiale autosomique dominante)[41] et un rôle causal en amont de la lésion cérébrovasculaire induite par l'ApoB dans la maladie sporadique à début tardif. La réduction à vie de l'ApoB — par des régimes alimentaires de haute qualité privilégiant les végétaux et, le cas échéant, une pharmacothérapie initiée au milieu de la vie — est une stratégie de prévention défendable contre le déclin cognitif et un composant plausible de la prévention de la démence et de la MA.

Introduction

Le modèle neurocentrique dominant de la maladie d'Alzheimer, qui privilégie la cascade amyloïde-β comme initiateur principal de la neurodégénérescence, est de plus en plus contesté par des données génétiques, épidémiologiques et d'imagerie indiquant que la santé vasculaire n'est pas seulement un facteur comorbide mais un moteur amont fondamental du processus pathologique.1,5,17Cette analyse étudie l'hypothèse selon laquelle l'exposition à long terme à des taux élevés de lipoprotéines contenant de l'apolipoprotéine B (ApoB) entraîne une athérosclérose progressive plaque accumulation dans les artères carotides, intracrâniennes et rénales, déclenchant une cascade de lésions systémiques et locales qui aboutissent à une démence chez un sous-ensemble substantiel de patients.4,7,20,22En examinant les mécanismes qui se chevauchent de la formation de plaques dans tous les lits vasculaires et les inférences causales offertes par la randomisation mendélienne, un cadre rigoureux émerge pour comprendre la maladie d'Alzheimer comme la manifestation neurologique tardive de décennies de lésions vasculaires, lipidiques et inflammatoires.2,3,9]

Le langage causal est utilisé délibérément mais est calibré en fonction de la force des preuves à chaque maillon de la chaîne. L'ApoB est établie comme causale pour l'athérosclérose maladie cardiovasculaire par triangulation via la randomisation mendélienne multivariée,21,25analyses d'exposition cumulée,22,33] et des essais randomisés de diminution des lipides.[26L'extension de cette voie causale aux troubles cognitifs vasculaires et à la démence à pathologie mixte est largement étayée par la méthode de randomisation mendélienne ciblée par médicament pour le non-HDL-C et la démence chez plus d'un million d'individus7et par l'imagerie TEP longitudinale reliant la sténose intracrânienne initiale à l'accumulation ultérieure d'amyloïde et de tau.13] L'extension supplémentaire à la maladie d'Alzheimer confirmée par biomarqueurs spécifiquement — la MA telle que définie par les critères révisés 2024 du NIA-AA [40— est étayé par une randomisation mendélienne spécifique à l'ApoB et orientée de manière cohérente [4] et par la compatibilité mécanistique de la lésion vasculaire avec la biologie de l'amyloïde et de la tau, [16,17,42] mais avec une intensité proportionnellement plus faible que le lien cardiovasculaire. Le manuscrit adopte en conséquence un langage nuancé concernant le critère d'évaluation de la MA et distingue explicitement la démence toutes causes confondues, les troubles cognitifs vasculaires, la démence à pathologie mixte et la MA confirmée par biomarqueurs tout au long du texte.

L'hypothèse vasculo-athérosclérotique de la maladie d'Alzheimer et de la démence

L'hypothèse vasculo-athérosclérotique postule que le syndrome clinique de la maladie d'Alzheimer est, chez un sous-ensemble substantiel de patients, la conséquence en aval d'une hypoperfusion cérébrale chronique et d'un dysfonctionnement de l'unité neurovasculaire initiés par la grande artère athérosclérose et propagées à travers le réseau des petits vaisseaux cérébraux.1,5,17Dans ce cadre, le rétrécissement focal des artères carotides et intracrâniennes, associé à une perte d'élasticité artérielle, crée un environnement hémodynamique hostile à la survie neuronale et protéine dégagement.12,18]

Les preuves neuropathologiques contemporaines démontrent que la distinction catégorielle classique entre la démence vasculaire pure et la maladie d'Alzheimer pure est empiriquement intenable chez les sujets âgés.3,32Des études d'autopsie en communauté rapportent que des neuropathologies mixtes — toute combinaison de types Alzheimer, corps de Lewy, TDP-43 et vasculaires lésions — sont présentes dans la majorité des cerveaux âgés, et la protéinopathie pure sans pathologie cérébrovasculaire est l'exception plutôt que la règle.3,32Athérosclérose des gros vaisseaux, y compris la sténose carotidienne et maladie athéosclérotique intracrânienne, coexiste fréquemment avec une maladie des petits vaisseaux cérébraux caractérisée par des hyperintensités de la matière blanche, des infarctus lacunaires et des microbleeds.3,8,9]

Rigidité artérielle — en particulier dans le proximal aorte et les artères carotides — transmettent la pulsatilité à haute pression dans la microvasculature délicate du cerveau.12Ce flux hyperpulsatile endommage la barrière hémato-encéphalique et perturbe la système glymphatique, un réseau périvasculaire responsable de l'élimination des déchets métaboliques, notamment l'amyloïde-β et la tau.15,18Dans ce cadre, l'accumulation de ces protéines est, en partie, la conséquence d'une élimination altérée plutôt que la seule conséquence d'une surproduction.16,17,18]

Cette hypothèse n'est pas exclusive du risque génétique canonique de la MA. L'apolipoprotéine E ε4 (APOE allèle ε4 — le plus grand facteur génétique commun facteur de risque pour la MA à début tardif — agit principalement par des voies cérébrovasculaires. Des preuves directes d'imagerie chez l'homme démontrent que les porteurs d'APOE ε4 présentent une rupture de la barrière hémato-encéphalique qui précède et prédit le déclin cognitif indépendamment de l'amyloïde et du tau,42fournissant un pont mécaniste entre la biologie des lipoprotéines systémiques, la lésion cérébrovasculaire et le phénotype clinique de la MA.5,16,42]

Nous nous engageons explicitement avec les critères révisés de l'Alzheimer’s Association de 2024,40] qui définissent biologiquement la MA par des biomarqueurs centraux de l'amyloïde et du tau, indépendamment du syndrome clinique. Ces critères constituent un cadre diagnostique et non une affirmation étiologique, et le groupe de travail reconnaît les co-pathologies vasculaires et autres comme des modificateurs courants de l'évolution de la maladie. Notre position n'est donc pas que les critères de 2024 sont erronés ou qu'ils nient les contributions vasculaires — ils ne font ni l'un ni l'autre. Notre position est que l'empreinte biomarqueur qu'ils opérationnalisent a, pour un sous-ensemble substantiel de patients, une cause amont : des décennies de lésions cérébrovasculaires induites par l'ApoB interagissant avec la biologie intrinsèque de l'amyloïde et du tau. Définir la MA par son empreinte biomarqueur aval est parfaitement compatible avec l'attribution de cette empreinte, en partie, à des causes vasculaires amont.

Tableau 1. Caractéristiques pathologiques et hémodynamiques qui se chevauchent dans les sous-types de démence

Fonctionnalité	Maladie d'Alzheimer (MA)	Démence vasculaire (DV)	Démence mixte
Protéinopathie primaire	plaques d'amyloïde-β ; enchevêtrements de tau	Minimal ou absent	Dépôts importants de Aβ et de tau
Lésions vasculaires	Souvent présent (CAA, micro-saignements)	Gros infarctus, lacunes, HBM	Vasculaire et protéinopathique combinés
Hémodynamique	Débit sanguin cérébral réduit ; défaillance glymphatique	Focal ou global ischémie	Hypoperfusion chronique + rigidité
Moteur lipidique	Lien de l'ApoB via la lésion cérébrovasculaire	Forte association LDL-C / ApoB	Charge élevée en ApoB tout au long de la vie
Caractéristiques d'imagerie	Atrophie hippocampique ; TEP-Aβ (+)	Infarctus multiples ; sténose carotidienne	Atrophie, HVB, charge athéromateuse

Adapté des références [3,5,9,32,40]. AAC = angiopathie amyloïde cérébrale ; LBM = hyperintensités de la substance blanche ; FDC = débit sanguin cérébral.

Dynamique des lipoprotéines et exposition causale tout au long de la vie

La concentration des lipoprotéines contenant de l'ApoB dans la circulation est au cœur du développement de l'athérosclérose.10,19,20] L'ApoB est une protéine structurelle présente à la surface de chaque particule de lipoprotéine athérogène. Les lipoprotéines de très basse densité d'origine hépatique (VLDL), des lipoprotéines de densité intermédiaire (IDL), les lipoprotéines de basse densité (LDL), et lipoprotéine(a) chacun transporte une molécule de apoB-100 par particule, tandis que chylomicrons et les restes de chylomicrons transportent apoB-48 provenant de l'intestin.20Chez les adultes en état post-absorptif, la charge athérogène dominante reflétée par la mesure de l'apoB plasmatique est principalement composée de particules porteuses d'apoB-100.20] Au-delà de la distinction numérique, les particules de LDL (apoB-100) ont des temps de séjour intimaux sensiblement plus longs que les restes de chylomicrons (apoB-48), qui circulent de manière plus transitoire ; la fraction d'apoB-100 contribue donc de manière disproportionnée aux lésions cumulées de la paroi artérielle et domine le “Années-ApoB”intègre dans le lit cérébro-vasculaire.

Alors que cholestérol LDL (LDL-C) mesure le cholestérol masse au sein des particules de LDL, l'ApoB fournit un décompte direct du total particules athérogènes.[20] Cette distinction est cruciale sur le plan mécanistique : la probabilité qu'une particule de lipoprotéine pénètre dans l'artère intima et le fait d'être retenu par les protéoglycanes extracellulaires relève de la concentration des particules plutôt que de la teneur en cholestérol par particule.19Les analyses de mendélisation mendélienne multivariables ont établi que l'ApoB est le paramètre lipidique dominant dans la relation entre les lipoprotéines et maladie coronarienne; lorsque l'ApoB est contrôlée, ni le LDL-C ni triglycérides conserver des informations causales indépendantes pour la maladie cardiovasculaire athéosclérotique.21,25]

Le risque conféré par l'ApoB est cumulatif, un concept désormais formalisé sous le nom de LDL/ApoB exposition cumulative hypothèse.22] athérogénèse est un processus lent qui commence généralement dans l'enfance par la formation de naissants stries lipidiques.[19] Comme particules contenant de l'ApoB restent élevées sur plusieurs décennies, la charge de plaque athérosclérotique s'accumule progressivement jusqu'à ce que des seuils critiques soient franchis, entraînant un rétrécissement luminal ou plaque de rupture.[22,33] Exposition cumulée tout au long de la vie (souvent décrite comme “années-cholestérol”ou “ ApoB-années ” est un prédicteur biologiquement plus approprié du risque cardiovasculaire et, plausiblement, de démence que toute mesure unique au milieu ou à la fin de la vie, qui peut être faussée par causalité inverse chez les personnes âgées développant une fragilité et perte de poids dans la phase préclinique de la démence.22,24,28]

Tableau 2. Valeur prédictive comparative des biomarqueurs lipidiques pour le risque athérogène

Biomarqueur	Définition	Relation avec la formation de plaque	Utilité
LDL-C	Masse de cholestérol dans les particules de LDL	Fournit un substrat de cholestérol pour la croissance de la plaque	Marqueur standard ; omet le risque dans discordance
ApoB	Total count of atherogenic particles (one per particle)	Dictates rate of particle entry into the intima	Superior predictor of particle retention
Cholestérol non-HDL	Total cholesterol minus HDL-C	Captures LDL, VLDL, IDL, and Lp(a) cholesterol	Practical surrogate for atherogenic burden
Lp(a)	LDL-like particle with apolipoprotein(a)	Pro-thrombotic and pro-inflammatory	Independent genetic risk factor
Cumulative ApoB exposure	Integral of ApoB over time (“ApoB-years”)	Integrated measure of total vascular injury	Best conceptual predictor of lifetime risk

Adapté des références [19,20,21,22,25,33].

Review of Mendelian Randomization Evidence in Dementia

Mendelian randomization (MR) leverages the random assortment of genetic alleles during gametogenesis to estimate the causal effect of modifiable exposures on outcomes, mimicking the design of a randomized controlled trial while avoiding many sources of observational confounding.[29] All MR claims summarized below have been evaluated against the STROBE-MR statement on transparent reporting of Mendelian randomization investigations.[29]

MR analyses of genetic instruments for common lipid-lowering drug targets — HMGCR (statine target), PCSK9 (PCSK9 inhibitor target), NPC1L1 (ézétimibe target), and CETP — provide some of the strongest current causal evidence linking lifelong lower non-HDL-C to lower dementia risk. In a one-sample MR meta-analysis of 1,091,775 individuals, the odds ratios per 1 mmol/L (≈39 mg/dL) lower non-HDL-C for all-cause dementia were 0.24 (95% CI 0.18–0.31) for HMGCR, 0.18 (0.12–0.25) for NPC1L1, 0.30 (0.26–0.34) for CETP, and 0.97 (0.70–1.35) for PCSK9.[7] An earlier Danish MR analysis reported risk ratios per 1 mmol/L lower LDL-C of 0.57 (0.27–1.17) for Alzheimer’s disease and 0.81 (0.34–1.89) for vascular dementia — directionally consistent with protection but underpowered for individual outcomes.[27]

Two points warrant explicit acknowledgement. First, the primary outcome in Nordestgaard et al. is all-cause dementia rather than biomarker-confirmed Alzheimer’s disease under the 2024 NIA-AA criteria.[40] Second, the PCSK9 estimate is compatible with the null, in contrast to robust signals for HMGCR, NPC1L1, and CETP.[7] This target-specific heterogeneity argues against a simple model in which any pharmacologic mechanism that lowers ApoB-containing lipoproteins will equivalently lower dementia risk, and points toward target-specific cerebrovascular biology that future trials must address.

MR evidence specifically for ApoB and Alzheimer’s disease is more recent. Univariable two-sample MR using UK Biobank ApoB instruments and IGAP AD summary statistics shows higher genetically proxied ApoB is associated with shortened healthspan and — in the authors’ own hedged language — may possibly increase risk for Alzheimer’s disease.[4] Multivariable MR conditioning on LDL-C indicates that ApoB retains its association with shortened healthspan even when LDL-C is null in the joint model.[4] We adopt the authors’ own characterization: this is supportive but preliminary causal evidence at the AD endpoint, which should be replicated against biomarker-confirmed AD outcomes as such data become available.

Limitations of Mendelian Randomization

Several limitations apply to all MR evidence presented above and must inform causal interpretation:

• Survival bias: individuals with genetically high ApoB may die of premature cardiovascular events before reaching the age at which dementia would clinically manifest, potentially attenuating the observed lipid–AD association.[4,7]
• Competitive risk of death: more specifically than general survival bias, ApoB is so strongly causal for infarctus du myocarde et accident vasculaire cérébral that individuals with genetically elevated ApoB who would otherwise have developed AD are disproportionately removed from the AD-eligible population by earlier cardiovascular mortality. This competing-risk structure systematically attenuates the observable lipid–AD association in standard Mendelian randomization designs, and is one specific reason the AD-endpoint signal in current MR is weaker than the cardiovascular signal — not necessarily because the causal pathway is weak.[4,27]
• Pleiotropy: genetic variants may affect the outcome through pathways other than the intended exposure. MR-Egger, weighted median, and multivariable MR estimators were used by the studies cited to assess robustness; ApoB-conditional analyses retain a directionally consistent signal.[4,21,25]
• Outcome heterogeneity: clinically diagnosed AD in administrative datasets frequently includes substantial vascular copathology;[3,32] biomarker-confirmed AD endpoints under the 2024 NIA-AA framework[40] would strengthen future MR work. The drug-target MR evidence for all-cause dementia[7] is methodologically stronger but applies to dementia broadly, not to biomarker-confirmed AD specifically.
• Ancestry and generalizability: most MR studies of lipids and dementia have been conducted in populations of European ancestry; causal estimates may differ across ancestry groups.[7]
• Lifelong-exposure interpretation: MR estimates the cumulative effect of lifelong genetic exposure, which typically exceeds the effect achievable by pharmacotherapy initiated in late life. This difference is informative for prevention timing but does not directly translate to expected effect sizes from late-life dementia trials.[22,26]

Regional Pathophysiology of Atherosclerotic Plaque Formation

Atherosclerosis is a chronic inflammatory disease initiated by lipid retention in focal areas of arteries, particularly regions of disturbed, non-laminar flow.[10,19] The molecular and cellular mechanisms of plaque formation are broadly conserved across major vascular beds, yet local anatomy and hemodynamics impart distinctive characteristics to disease in each location.[10,11,31]

Carotid Artery Atherosclerosis

The carotid bifurcation is especially vulnerable to plaque formation owing to its complex geometry and flow reversal during the cardiac cycle.[10,11] Low and oscillatory wall shear stress at the origin of the internal carotid artery promotes endothelial dysfunction and increases vessel-wall permeability to ApoB-containing lipoproteins.[10,11,19] A distinctive feature of carotid plaque is its tendency to embolize: high-velocity flow dislodges thrombi forming on ulcerated plaque surfaces, generating recurrent microemboli rather than complete proximal occlusion.[11,34] Cumulative microembolic injury, lacunar infarction, and microbleeds are core substrates of vascular cognitive impairment.[2,9,34]

Intracranial Atherosclerotic Disease (ICAD)

ICAD refers to plaque formation in the large arteries at the base of the brain — including the middle cerebral artère, basilar artery, and the Circle of Willis.[11,13] Unlike extracranial carotid disease, ICAD has been directly linked, in prospective longitudinal imaging, to subsequent deposition of amyloid-β and tau in the brain parenchyma. In a prospective cohort with serial PiB-PET (amyloid) and AV-1451-PET (tau) imaging, the presence of any intracranial stenosis at baseline was significantly associated with greater amyloid-β accumulation over four years, and stenosis in two or more arteries predicted greater tau deposition over two years.[13] This longitudinal directionality — vascular injury preceding accelerated proteinopathy — is among the strongest current human imaging evidence that intracranial vascular health is causally upstream of AD-defining pathology rather than merely co-occurring with it.

Renal Artery Atherosclerosis and the Cardiorenal-Cerebral Axis

Renal artery stenosis (RAS) is overwhelmingly atherosclerotic in origin and concentrates at the aortic orifice or within the proximal one-third of the main renal artery.[31,37] The kidney, like the brain, is a low-resistance vascular bed sensitive to hyperpulsatile flow and ischemia.[6,12] Plaque progression in the renal artery is augmented by dysfunctional perivascular adipose tissue and expansion of the vasa vasorum, which release pro-inflammatory cytokines and extracellular vesicles.[31]

Atherosclerotic RAS activates the renin-angiotensin-aldosterone system, producing secondary hypertension that further accelerates damage to the cerebral microvasculature and establishes a feed-forward loop linking kidney, heart, and brain.[6,31,36,39] Chronic kidney disease is itself characterized by systemic inflammation, oxidative stress, and uremic endothelial toxicity, all of which are independently associated with cognitive decline through the cardiorenal-cerebral axis.[6,38,39]

Table 3. Comparison of Plaque Characteristics Across Vascular Beds

Vascular bed	Primary anatomical site	Unique hemodynamic driver	Primary clinical outcome
Carotid artery	Carotid bifurcation; ICA origin	Flow reversal and turbulence; low/oscillatory shear	Embolic stroke and cognitive impairment
Intracranial arteries	MCA, basilar, Circle of Willis	Complex branching; loss of autoregulation	Hypoperfusion and accelerated proteinopathy
Renal artery	Aortic orifice / proximal one-third	Orifice-related shear stress	RAS, hypertension, CKD
Coronary artery	Proximal LAD and bifurcations	Cyclic compressive stress; bending	Myocardial infarction and angina

Adapté des références [10,11,13,31,37].

The Mechanistic Pathway: From Lipids to Neurodegeneration

The transition from elevated circulating ApoB to clinical dementia involves a multi-stage cascade in which subendothelial retention of ApoB-containing lipoproteins serves as the initiating event.[19] Retained particles undergo oxidative and enzymatic modification, generating potent pro-inflammatory signals that activate endothelial cells, recruit monocytes, and drive macrophage foam-cell formation.[10,19,53]

Lipid Retention, Endothelial Activation, and Hypoperfusion

Endothelial cells respond to oxidized lipoproteins by expressing adhesion molecules (VCAM-1, ICAM-1) and releasing chemokines (MCP-1, CXCL1) and cytokines (IL-1β, IL-6).[10,19,53] Progressive plaque growth produces arterial narrowing and stiffening,[10,12] which in the cerebrovascular tree manifests as reduced cerebral blood flow and chronic ischemia.[12,17] Chronic hypoperfusion induces metabolic stress in neurons and glia, activates the inflammasome NLRP3, and primes the brain for accelerated neurodegeneration.[17,53]

Amyloid-β in the Vascular Pathway

A central element of the vascular framing is that amyloid-β (Aβ) accumulation is, in part, both a cause and a consequence of cerebrovascular injury.[16,17] Ischemia and oxidative stress upregulate amyloid precursor protein expression and shift its processing toward the amyloidogenic pathway,[16,17] while vascular Aβ deposition (cerebral amyloid angiopathy, CAA) impairs vessel-wall contractility and reduces protein efflux from the brain.[16] This generates a feed-forward loop: vascular injury promotes Aβ accumulation, which causes additional vascular damage and neuroinflammation through innate immune activation.[16,17,53]

We are explicit, however, that amyloid is not merely downstream debris from vascular injury. Autosomal-dominant familial AD caused by mutations in APP, PSEN1, or PSEN2 establishes amyloid as a sufficient pathogenic factor in its own right — these mutations produce AD through altered amyloid biology without requiring upstream vascular insult.[41] Anti-amyloid monoclonal antibody trials in sporadic AD, discussed below in the Counterarguments section, further demonstrate that amyloid removal produces statistically robust clinical benefit, which is incompatible with a model in which amyloid is epiphenomenal. Our claim is therefore narrower and more defensible: in sporadic late-onset AD — the overwhelming majority of cases — decades of ApoB-driven cerebrovascular injury are a major upstream determinant of the cerebral environment in which intrinsic amyloid and tau biology become pathogenic. The vascular and amyloid pathways are interacting and bidirectionally amplifying rather than mutually exclusive.

Glymphatic Dysfunction and Protein Clearance

Clearance of metabolic waste from the brain is mediated in part by the glymphatic system, a perivascular CSF–interstitial fluid exchange network whose driving force depends on arterial pulsatility and stable venous outflow.[15,18] Arterial stiffening — a direct mechanical consequence of atherosclerosis and vascular aging — impairs the pulsatile driving force,[12] and rodent and human work have shown that hypertension and aging reduce CSF influx along the perivascular space.[18] When glymphatic clearance is impaired, neurotoxic proteins including Aβ and tau accumulate, contributing to synaptic dysfunction and cognitive decline.[12,15,18] We treat the arterial-stiffness → impaired-glymphatic-clearance → AD chain as a mechanistically plausible and partially supported hypothesis rather than an established clinical pathway. Human glymphatic physiology remains incompletely validated, and the diffusion-tensor-image analysis along the perivascular space (DTI-ALPS) index is an imaging proxy for perivascular diffusivity rather than a direct measure of glymphatic flux; its reproducibility and biological specificity are the subject of active methodological work.[35]

Testing the Lifelong Protection Hypothesis

Early, sustained lipid lowering is critical to preventing irreversible neurodegeneration.[22,26] MR studies effectively provide a natural model of lifelong lower LDL-C and ApoB, showing that individuals with genetically lower exposure carry substantially lower lifetime cardiovascular risk than those who initiate lipid-lowering therapy only in late life.[22,25]

Quantitatively, clinical trials in the Cholesterol Treatment Trialists’ Collaboration demonstrate approximately a 22% relative reduction in major vascular events per 1 mmol/L (≈39 mg/dL) reduction in LDL-C achieved with statin therapy, irrespective of baseline risk.[26] By contrast, MR estimates a markedly larger reduction in coronary heart disease per 1 mmol/L genetically lower lifelong LDL-C — roughly threefold the per-mmol/L benefit observed with late-life statin therapy.[22,33] We label this difference clearly as an inference about the value of duration: it argues that for dementia prevention, interventions should ideally occur in midlife, before mature plaques and the cerebrovascular substrate for cognitive decline have developed.[22,40] Direct extrapolation of the cardiovascular per-mmol/L estimate to dementia outcomes is not warranted in the absence of trial data, and we present the timing inference rather than a numerical prediction of dementia-event reduction.

Randomized vascular-risk-reduction trials provide complementary, if smaller-effect, evidence. In SPRINT MIND, intensive blood-pressure control (target SBP <120 mmHg) versus standard control (<140 mmHg) reduced the composite of mild cognitive impairment or probable dementia (hazard ratio 0.85, 95% CI 0.74–0.97) and reduced MCI specifically (HR 0.81, 95% CI 0.69–0.94), although the probable-dementia primary endpoint did not reach statistical significance.[14] In the FINGER multidomain prevention trial, a two-year combined intervention of diet, exercise, cognitive training, and vascular risk monitoring improved cognitive composite scores versus control in at-risk older adults.[50] These trials demonstrate that randomized modification of vascular risk factors can produce measurable cognitive benefit, supporting the broader logic of lifelong vascular-risk reduction for dementia prevention.

The Role of Plant-Based Dietary Patterns

High-quality plant-based dietary patterns are a powerful non-pharmacological strategy for reducing lifelong ApoB exposure and systemic vascular inflammation.[23,46,49] The quality of the diet matters: a healthful plant-based diet index (hPDI) emphasizing whole grains, fruits, vegetables, legumes, and nuts is associated with lower dementia risk, whereas an unhealthful plant-based pattern (uPDI) high in refined grains and sugar-sweetened beverages is associated with higher risk.[43,46]

Meta-analytic evidence from prospective cohorts (n≈208,000) shows the highest tertile of hPDI is associated with approximately 21% lower dementia incidence relative to the lowest tertile, while the highest tertile of uPDI is associated with approximately 26% higher dementia incidence.[43] In adults with established cardiometabolic disease — a population for whom the cardiorenal-cerebral axis is already activated — a healthful plant-based diet pattern combined with broader healthy-lifestyle factors substantially attenuates dementia risk in UK Biobank follow-up.[45] Among plant-forward dietary patterns, Mediterranean and MIND diets currently have the strongest cumulative evidence base, including randomized cardiovascular outcome data (PREDIMED)[49] and consistent observational associations with lower dementia risk;[44] strict whole-food plant-based or vegan patterns have weaker direct evidence for dementia or AD endpoints specifically. We therefore frame plant-forward dietary patterns broadly as a biologically well-grounded prevention strategy supported by strong observational data and (for régime méditerranéen) randomized cardiovascular evidence, and we do not assert the superiority of strict whole-food plant-based or vegan diets over Mediterranean or MIND patterns on the basis of currently available dementia data.

Beyond ApoB lowering itself, neuroprotective effects of high-quality plant-based dietary patterns are mediated by complementary mechanisms: improved fonction endothéliale and reduced oxidative stress;[23,46] favorable modulation of the gut microbiome with reduced production of trimethylamine N-oxide (TMAO), a metabolite linked to atherosclerosis and cognitive impairment;[47,48] and improved tension artérielle, insulin sensitivity, and inflammatory tone.[23,49,50] Mediterranean and MIND dietary patterns share many of these mechanisms and are supported by both observational and randomized evidence for cardiovascular outcomes and cognitive endpoints.[44,49,50]

Table 4. Impact of Dietary Patterns on Vascular and Cognitive Risk Factors

Dietary pattern	Key components	Effect on ApoB	Effect on inflammation	Dementia risk
Healthful plant-based (hPDI)	Whole plants, legumes, nuts	Decreased	Decreased	Lower [43,45]
Mediterranean	Olive oil, plants, fish, nuts	Moderately decreased	Moderately decreased	Lower [44,49]
MIND	Plants, berries, fish, olive oil	Moderately decreased	Decreased	Lower [44]
Unhealthful plant-based (uPDI)	Refined grains, sweets, fried potatoes	Variable / neutral	Increased	Higher [43]
Western	Red and processed meat, refined sugars	Increased	Increased	Higher [23,46]

Adapté des références [23,43,44,45,46,49].

Counterarguments and Reconciliation with the Biological Definition of AD

The hypothesis that ApoB-driven atherosclerosis is a causal upstream contributor to Alzheimer’s disease must engage three legitimate counterarguments: (1) the late-life low-cholesterol paradox, (2) cases of neurodegeneration without overt vascular disease, and (3) the partial therapeutic success of anti-amyloid monoclonal antibodies, which appear to support an amyloid-causal model.

The Late-Life Low-Cholesterol Paradox

Observational studies frequently show that low cholesterol in adults aged ≥ 75 years is associated with higher dementia risk.[24,28] The most parsimonious explanation is reverse causation: preclinical dementia is preceded by years of unintentional weight loss and frailty, and in late-stage neurodegeneration sometimes by hepatic synthetic dysfunction, all of which lower circulating cholesterol independent of any protective effect of low lifelong ApoB exposure.[24,28] This paradox underscores why midlife cholesterol measurements and lifelong genetic instruments — not geriatric cholesterol levels — are the appropriate substrate for causal inference about the ApoB–dementia relationship.[22,29]

Non-Vascular AD and Genetic Architecture

Predominantly neurodegenerative Alzheimer’s disease occurs through specific autosomal-dominant mutations in APP, PSEN1, and PSEN2,[41] and additional contributors include chronic infection, mitochondrial dysfunction, sleep-disordered breathing, and chronic sleep deprivation.[41] These are not refutations of a vascular-causal model; they are simply reminders that AD is a heterogeneous syndrome in which atherosclerosis is a major causal contributor in a substantial subset of patients, not the exclusive cause in all.[3,32,41] The APOE ε4 allele increases AD risk in significant part through cerebrovascular mechanisms — including blood-brain barrier breakdown that precedes cognitive decline,[42] impaired Aβ clearance, and increased CAA[16] — illustrating that genetic and vascular risk are intertwined rather than competing.

Anti-Amyloid Monoclonal Antibody Trials

Lecanemab et donanemab provide the most direct test of the amyloid pathway in symptomatic disease. In CLARITY AD, lecanemab slowed CDR-SB decline by 0.45 points relative to placebo over 18 months on an 18-point scale (mean 1.21 vs 1.66; 27% relative slowing), with amyloid-related imaging abnormalities (ARIA-E) in 12.6% and symptomatic ARIA-E in 2.8%.[51] In TRAILBLAZER-ALZ 2, donanemab slowed iADRS decline by 35% and CDR-SB decline by 36% over 76 weeks; ARIA-E occurred in 24.0% (symptomatic 6.1%), with three donanemab-related deaths attributable to ARIA.[52]

We accept these trials as substantive evidence that amyloid is mechanistically important in symptomatic AD: removing amyloid produces measurable, statistically robust, and consistent slowing of cognitive decline across primary and secondary endpoints. That result is not compatible with a model in which amyloid is epiphenomenal vascular debris, and we do not adopt such a model. Whether the absolute magnitude of slowing reaches a minimal clinically important difference (MCID) is genuinely contested rather than settled, with proposed CDR-SB MCIDs for early AD ranging across roughly 0.5–2.0 points and different working groups landing in different places along that range. Reasonable people disagree on whether the lecanemab and donanemab effects clear that threshold.

What the trials do not show is that amyloid removal in late-stage symptomatic disease is sufficient to halt or reverse AD progression: the effects are incremental, the ARIA safety profile is non-trivial and concentrated in APOE ε4 homozygotes, and the trials test late-stage amyloid removal rather than upstream prevention. These results are fully compatible with our framing — amyloid is mechanistically important and a legitimate therapeutic target; decades of upstream ApoB-driven cerebrovascular injury are an additional and modifiable major determinant of the substrate on which amyloid pathology operates; and a complete dementia-prevention strategy reasonably invests in both pathways.

Engagement with the 2024 NIA-AA Biological Definition of AD

The 2024 Alzheimer’s Association revised criteria operationalize AD biologically by core amyloid and tau biomarkers, independent of clinical syndrome.[40] These criteria are a diagnostic framework, not an etiologic theory. The working group does not claim that amyloid and tau are the sole upstream causes of AD and does not deny vascular contributions — indeed, vascular and other co-pathologies are explicitly recognized as common modifiers of disease course. We are therefore not arguing against the 2024 criteria; we are making a complementary etiologic argument compatible with them. For a substantial subset of patients meeting the biological criteria for AD, the upstream cause of the amyloid/tau biomarker fingerprint is, in our reading, lifelong ApoB-driven cerebrovascular atherogenesis interacting with intrinsic amyloid biology. Longitudinal evidence that intracranial atherosclerosis predicts subsequent amyloid and tau accumulation[13] is directly compatible with this reading.

Proposed Research Designs to Test the Hypothesis

Rigorous testing of the vascular-atherosclerotic causal hypothesis requires longitudinal studies that integrate cardiovascular and neurodegenerative markers across the life course.[9,40] Studies should follow individuals from their 30s and 40s with serial measurement of ApoB, Lp(a), and inflammatory markers.[20,22] High-resolution imaging — including contrast-enhanced ultrasound and vessel-wall MRI — should monitor plaque growth and regression in the carotid and intracranial arteries.[11,13] Advanced neuroimaging, including amyloid- and tau-PET and emerging perivascular-diffusivity proxies of glymphatic function, would map how vascular injury precedes protein accumulation.[13,15,18,35]

New MR studies should isolate ApoB from other lipid traits using multivariable instruments and examine its independent effect against biomarker-confirmed AD endpoints under the 2024 NIA-AA criteria.[4,40] Randomized prevention trials are needed to test whether early, aggressive ApoB lowering — initiated in midlife rather than in late life — reduces subsequent cognitive decline and neurodegenerative-protein burden.[7,22,26] Trials should stratify by APOE ε4 status and baseline cerebrovascular burden,[16,42] and should comply with STROBE-MR or CONSORT reporting as appropriate.[29]

Final Thesis and Conclusion

Cognitive decline, vascular cognitive impairment and dementia, mixed-pathology dementia, and — in a substantial subset of patients — biologically defined Alzheimer’s disease can be productively understood as late neurological manifestations of decades-long vascular-lipid-inflammatory injury operating alongside intrinsic amyloid and tau biology.[1,2,3,17,40] The central modifiable driver of this vascular pathway is lifelong exposure to ApoB-containing lipoproteins, which initiates and propagates atherosclerotic plaque formation in critical cerebral-supplying and renal arteries.[7,20,22,31] These plaques cause chronic cerebral hypoperfusion, arterial stiffening, blood-brain barrier dysfunction, and impaired glymphatic clearance, contributing to the accumulation of amyloid and tau and to clinical cognitive decline in interaction with intrinsic neurodegenerative biology.[12,13,16,17,18,42]

Our argument is therefore not that Alzheimer’s disease should be reclassified as vascular disease or that amyloid is mechanistically unimportant. Familial AD,[41] anti-amyloid trial results in sporadic AD,[51,52] and the 2024 NIA-AA biological criteria[40] each establish that amyloid biology is a legitimate and necessary component of any complete model of AD. Our argument is more focused: for a substantial subset of patients meeting biological criteria for AD, the upstream cause of the amyloid/tau fingerprint is, in important part, decades of ApoB-driven cerebrovascular injury. This framing is fully compatible with the 2024 NIA-AA biological criteria, which define AD diagnostically by amyloid and tau biomarkers without making etiologic claims about the upstream cause of those biomarkers.

The practical consequence is that prevention strategy for Alzheimer’s disease and related dementias should expand to give substantially greater weight to lifelong preservation of vascular integrity — through high-quality plant-forward dietary patterns,[43,44,45] randomized vascular-risk-reduction strategies such as those validated in SPRINT MIND and FINGER,[14,50] and, where indicated, lipid-lowering pharmacotherapy initiated in midlife rather than in late life[7,22,26] — alongside continued investment in late-stage protein-targeted therapy. Both arms of intervention are warranted by current evidence; the upstream vascular arm offers the larger plausible public-health gain on the longest time horizons, and is the focus of this manuscript.

References

1. Iadecola C. The pathobiology of vascular dementia. Neuron. 2013;80(4):844-866. doi:10.1016/j.neuron.2013.10.008
2. Smith EE, Aparicio HJ, Gottesman RF, et al. Vascular Contributions to Cognitive Impairment and Dementia in the United States: Prevalence and Incidence: A Scientific Statement From the American Heart Association. Stroke. 2025;56(10):e317-e330. doi:10.1161/STR.0000000000000494
3. Kapasi A, DeCarli C, Schneider JA. Impact of multiple pathologies on the threshold for clinically overt dementia. Acta Neuropathol. 2017;134(2):171-186. doi:10.1007/s00401-017-1717-7
4. Martin L, Boutwell BB, Messerlian C, Adams CD. Mendelian randomization reveals apolipoprotein B shortens healthspan and possibly increases risk for Alzheimer’s disease. Commun Biol. 2024;7(1):230. Published 2024 Feb 24. doi:10.1038/s42003-024-05887-2
5. Sweeney MD, Sagare AP, Zlokovic BV. Blood-brain barrier breakdown in Alzheimer disease and other neurodegenerative disorders. Nat Rev Neurol. 2018;14(3):133-150. doi:10.1038/nrneurol.2017.188
6. Bugnicourt JM, Godefroy O, Chillon JM, Choukroun G, Massy ZA. Cognitive disorders and dementia in CKD: the neglected kidney-brain axis. J Am Soc Nephrol. 2013;24(3):353-363. doi:10.1681/ASN.2012050536
7. Nordestgaard LT, Hanson A, Sanderson E, et al. Cholesterol-lowering drug targets reduce risk of dementia: Mendelian randomization and meta-analyses of 1 million individuals. Alzheimers Dement. 2025;21(10):e70638. doi:10.1002/alz.70638
8. Bos D, Vernooij MW, de Bruijn RF, et al. Atherosclerotic calcification is related to a higher risk of dementia and cognitive decline. Alzheimers Dement. 2015;11(6):639-47.e1. doi:10.1016/j.jalz.2014.05.1758
9. Wardlaw JM, Smith C, Dichgans M. Small vessel disease: mechanisms and clinical implications. Lancet Neurol. 2019;18(7):684-696. doi:10.1016/S1474-4422(19)30079-1
10. Libby P, Buring JE, Badimon L, et al. Atherosclerosis. Nat Rev Dis Primers. 2019;5(1):56. Published 2019 Aug 16. doi:10.1038/s41572-019-0106-z
11. Yahagi K, Kolodgie FD, Lutter C, et al. Pathology of Human Coronary and Carotid Artery Atherosclerosis and Vascular Calcification in Diabetes Mellitus. Arterioscler Thromb Vasc Biol. 2017;37(2):191-204. doi:10.1161/ATVBAHA.116.306256
12. Mitchell GF, van Buchem MA, Sigurdsson S, et al. Arterial stiffness, pressure and flow pulsatility and brain structure and function: the Age, Gene/Environment Susceptibility–Reykjavik study. Brain. 2011;134(Pt 11):3398-3407. doi:10.1093/brain/awr253
13. Kang KM, Park C, Nam H, et al. Intracranial stenosis and longitudinal progression of Alzheimer’s disease pathologies. Alzheimers Dement. 2025;21(12):e71010. doi:10.1002/alz.71010
14. SPRINT MIND Investigators for the SPRINT Research Group, Williamson JD, Pajewski NM, et al. Effect of Intensive vs Standard Blood Pressure Control on Probable Dementia: A Randomized Clinical Trial. JAMA. 2019;321(6):553-561. doi:10.1001/jama.2018.21442
15. Iliff JJ, Wang M, Liao Y, et al. A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid β. Sci Transl Med. 2012;4(147):147ra111. doi:10.1126/scitranslmed.3003748
16. Greenberg SM, Bacskai BJ, Hernandez-Guillamon M, Pruzin J, Sperling R, van Veluw SJ. Cerebral amyloid angiopathy and Alzheimer disease – one peptide, two pathways. Nat Rev Neurol. 2020;16(1):30-42. doi:10.1038/s41582-019-0281-2
17. Iadecola C. The Neurovascular Unit Coming of Age: A Journey through Neurovascular Coupling in Health and Disease. Neuron. 2017;96(1):17-42. doi:10.1016/j.neuron.2017.07.030
18. Mestre H, Tithof J, Du T, et al. Flow of cerebrospinal fluid is driven by arterial pulsations and is reduced in hypertension. Nat Commun. 2018;9(1):4878. Published 2018 Nov 19. doi:10.1038/s41467-018-07318-3
19. Tabas I, Williams KJ, Borén J. Subendothelial lipoprotein retention as the initiating process in atherosclerosis: update and therapeutic implications. Circulation. 2007;116(16):1832-1844. doi:10.1161/CIRCULATIONAHA.106.676890
20. Soffer DE, Marston NA, Maki KC, et al. Role of apolipoprotein B in the clinical management of cardiovascular risk in adults: An Expert Clinical Consensus from the National Lipid Association. J Clin Lipidol. 2024;18(5):e647-e663. doi:10.1016/j.jacl.2024.08.013
21. Richardson TG, Sanderson E, Palmer TM, et al. Evaluating the relationship between circulating lipoprotein lipids and apolipoproteins with risk of coronary heart disease: A multivariable Mendelian randomisation analysis. PLoS Med. 2020;17(3):e1003062. Published 2020 Mar 23. doi:10.1371/journal.pmed.1003062
22. Ference BA, Braunwald E, Catapano AL. The LDL cumulative exposure hypothesis: evidence and practical applications. Nat Rev Cardiol. 2024;21(10):701-716. doi:10.1038/s41569-024-01039-5
23. Sacks FM, Lichtenstein AH, Wu JHY, et al. Dietary Fats and Cardiovascular Disease: A Presidential Advisory From the American Heart Association. Circulation. 2017;136(3):e1-e23. doi:10.1161/CIR.0000000000000510
24. Mielke MM, Zandi PP, Sjögren M, et al. High total cholesterol levels in late life associated with a reduced risk of dementia. Neurology. 2005;64(10):1689-1695. doi:10.1212/01.WNL.0000161870.78572.A5
25. Zuber V, Gill D, Ala-Korpela M, et al. High-throughput multivariable Mendelian randomization analysis prioritizes apolipoprotein B as key lipid risk factor for coronary artery disease. Int J Epidemiol. 2021;50(3):893-901. doi:10.1093/ije/dyaa216
26. Cholesterol Treatment Trialists’ (CTT) Collaboration, Baigent C, Blackwell L, et al. Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet. 2010;376(9753):1670-1681. doi:10.1016/S0140-6736(10)61350-5
27. Benn M, Nordestgaard BG, Frikke-Schmidt R, Tybjærg-Hansen A. Low LDL cholesterol, PCSK9 and HMGCR genetic variation, and risk of Alzheimer’s disease and Parkinson’s disease: Mendelian randomisation study. BMJ. 2017;357:j1648. Published 2017 Apr 24. doi:10.1136/bmj.j1648
28. Iwagami M, Qizilbash N, Gregson J, et al. Blood cholesterol and risk of dementia in more than 1·8 million people over two decades: a retrospective cohort study. Lancet Healthy Longev. 2021;2(8):e498-e506. doi:10.1016/S2666-7568(21)00150-1
29. Skrivankova VW, Richmond RC, Woolf BAR, et al. Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization: The STROBE-MR Statement. JAMA. 2021;326(16):1614-1621. doi:10.1001/jama.2021.18236
30. Schwartz GG, Steg PG, Szarek M, et al. Alirocumab and Cardiovascular Outcomes after Acute Coronary Syndrome. N Engl J Med. 2018;379(22):2097-2107. doi:10.1056/NEJMoa1801174
31. Safian RD. Renal artery stenosis. Prog Cardiovasc Dis. 2021;65:60-70. doi:10.1016/j.pcad.2021.03.003
32. Schneider JA, Arvanitakis Z, Bang W, Bennett DA. Mixed brain pathologies account for most dementia cases in community-dwelling older persons. Neurology. 2007;69(24):2197-2204. doi:10.1212/01.wnl.0000271090.28148.24
33. Domanski MJ, Tian X, Wu CO, et al. Time Course of LDL Cholesterol Exposure and Cardiovascular Disease Event Risk. J Am Coll Cardiol. 2020;76(13):1507-1516. doi:10.1016/j.jacc.2020.07.059
34. Bos D, Arshi B, van den Bouwhuijsen QJA, et al. Atherosclerotic Carotid Plaque Composition and Incident Stroke and Coronary Events. J Am Coll Cardiol. 2021;77(11):1426-1435. doi:10.1016/j.jacc.2021.01.038
35. Taoka T, Ito R, Nakamichi R, et al. Reproducibility of diffusion tensor image analysis along the perivascular space (DTI-ALPS) for evaluating interstitial fluid diffusivity and glymphatic function: CHanges in Alps index on Multiple conditiON acquIsition eXperiment (CHAMONIX) study. Jpn J Radiol. 2022;40(2):147-158. doi:10.1007/s11604-021-01187-5
36. Toyoda K. Cerebral small vessel disease and chronic kidney disease. J Stroke. 2015;17(1):31-37. doi:10.5853/jos.2015.17.1.31
37. Hansen KJ, Edwards MS, Craven TE, et al. Prevalence of renovascular disease in the elderly: a population-based study. J Vasc Surg. 2002;36(3):443-451. doi:10.1067/mva.2002.127351
38. Sedaghat S, Sorond F, Yaffe K, et al. Decline in kidney function over the course of adulthood and cognitive function in midlife. Neurology. 2020;95(17):e2389-e2397. doi:10.1212/WNL.0000000000010631
39. Drew DA, Weiner DE, Sarnak MJ. Cognitive Impairment in CKD: Pathophysiology, Management, and Prevention. Am J Kidney Dis. 2019;74(6):782-790. doi:10.1053/j.ajkd.2019.05.017
40. Jack CR Jr, Andrews JS, Beach TG, et al. Revised criteria for diagnosis and staging of Alzheimer’s disease: Alzheimer’s Association Workgroup. Alzheimers Dement. 2024;20(8):5143-5169. doi:10.1002/alz.13859
41. Knopman DS, Amieva H, Petersen RC, et al. Alzheimer disease. Nat Rev Dis Primers. 2021;7(1):33. Published 2021 May 13. doi:10.1038/s41572-021-00269-y
42. Montagne A, Nation DA, Sagare AP, et al. APOE4 leads to blood-brain barrier dysfunction predicting cognitive decline. Nature. 2020;581(7806):71-76. doi:10.1038/s41586-020-2247-3
43. Shen J, Chen H, Gong Y, et al. Association between plant-based diets and incident dementia: results from prospective cohort studies and a meta-analysis. J Prev Alzheimers Dis. 2026;13(2):100457. doi:10.1016/j.tjpad.2025.100457
44. de Crom TOE, Mooldijk SS, Ikram MK, Ikram MA, Voortman T. MIND diet and the risk of dementia: a population-based study. Alzheimers Res Ther. 2022;14(1):8. Published 2022 Jan 12. doi:10.1186/s13195-022-00957-1
45. Dunk MM, Dove A, Wang J, Sakakibara S, Carballo-Casla A, Xu W. Plant-Based Diet Quality, Healthy Lifestyle, and Dementia Risk in Older Adults With Cardiometabolic Diseases. JACC Adv. 2025;4(11 Pt 1):102229. doi:10.1016/j.jacadv.2025.102229
46. Satija A, Bhupathiraju SN, Spiegelman D, et al. Healthful and Unhealthful Plant-Based Diets and the Risk of Coronary Heart Disease in U.S. Adults. J Am Coll Cardiol. 2017;70(4):411-422. doi:10.1016/j.jacc.2017.05.047
47. Tang WH, Wang Z, Levison BS, et al. Intestinal microbial metabolism of phosphatidylcholine and cardiovascular risk. N Engl J Med. 2013;368(17):1575-1584. doi:10.1056/NEJMoa1109400
48. Vogt NM, Romano KA, Darst BF, et al. The gut microbiota-derived metabolite trimethylamine N-oxide is elevated in Alzheimer’s disease. Alzheimers Res Ther. 2018;10(1):124. Published 2018 Dec 22. doi:10.1186/s13195-018-0451-2
49. Estruch R, Ros E, Salas-Salvadó J, et al. Primary Prevention of Cardiovascular Disease with a Mediterranean Diet Supplemented with Extra-Virgin Olive Oil or Nuts. N Engl J Med. 2018;378(25):e34. doi:10.1056/NEJMoa1800389
50. Ngandu T, Lehtisalo J, Solomon A, et al. A 2 year multidomain intervention of diet, exercise, cognitive training, and vascular risk monitoring versus control to prevent cognitive decline in at-risk elderly people (FINGER): a randomised controlled trial. Lancet. 2015;385(9984):2255-2263. doi:10.1016/S0140-6736(15)60461-5
51. van Dyck CH, Swanson CJ, Aisen P, et al. Lecanemab in Early Alzheimer’s Disease. N Engl J Med. 2023;388(1):9-21. doi:10.1056/NEJMoa2212948
52. Sims JR, Zimmer JA, Evans CD, et al. Donanemab in Early Symptomatic Alzheimer Disease: The TRAILBLAZER-ALZ 2 Randomized Clinical Trial. JAMA. 2023;330(6):512-527. doi:10.1001/jama.2023.13239
53. Heneka MT, Carson MJ, El Khoury J, et al. Neuroinflammation in Alzheimer’s disease. Lancet Neurol. 2015;14(4):388-405. doi:10.1016/S1474-4422(15)70016-5