Telling Different Dementias Apart

The Neurovascular-Neuroinflammatory Continuum: A Multi-Modal Diagnostic Framework for Distinguishing Pure Vascular Cognitive Impairment from Mixed-Pathology Dementia in the Absence of Specific Proteinopathies

Benjamin Aaron Gustafsson AdultCognitiveDisease.com

---

Abstract

The differential diagnosis of dementia in the elderly remains one of the most complex challenges in clinical neurology. While the dominance of the amyloid cascade hypothesis has provided robust biomarkers for Alzheimer’s Disease (AD), a significant diagnostic gap persists for patients presenting with cognitive decline who lack definitive amyloid or tau pathology, or in whom vascular injury co-exists with neurodegeneration. This doctoral thesis addresses the urgent clinical need to distinguish Pure Vascular Cognitive Impairment (VCI) from Mixed Dementia and non-amyloid mimics, specifically Limbic-predominant Age-related TDP-43 Encephalopathy (LATE). Grounded in the "Vascular-Immune Hypothesis," this research posits that VCI is characterized by a distinct inflammatory and microstructural signature driven by endothelial dysfunction and blood-brain barrier (BBB) breakdown, separable from the astrocytic and cortical neurodegeneration of AD and LATE. By synthesizing data on advanced neuroimaging metrics—specifically Peak Width of Skeletonized Mean Diffusivity (PSMD), Free-Water (FW) imaging, and Diffusion Tensor Imaging Analysis Along the Perivascular Space (DTI-ALPS)—with fluid biomarkers of neuroinflammation (Matrix Metalloproteinase-9 [MMP-9], Placental Growth Factor [PlGF], and Glial Fibrillary Acidic Protein [GFAP]), this thesis constructs a comprehensive diagnostic framework. The analysis demonstrates that Pure VCI is defined by a "High MMP-9/High PSMD/Low GFAP" profile, reflecting active vascular remodeling and diffuse white matter disintegration. In contrast, Mixed Dementia exhibits a synergistic "High MMP-9/High GFAP" profile, while LATE presents a distinct "Low Vascular/High Atrophy" signature. This work provides a theoretical and practical basis for precision phenotyping in dementia, facilitating targeted therapeutic interventions for vascular contributions to cognitive impairment.

Chapter 1: Introduction

1.1 The Evolving Nosology of Cognitive Impairment Historically, the classification of cognitive impairment has operated under a binary paradigm, separating "neurodegenerative" diseases, typified by Alzheimer’s Disease (AD), from "vascular" cognitive impairment (VCI) resulting from cerebrovascular disease (CVD).¹ This dichotomy was codified in early diagnostic criteria such as the NINDS-AIREN, which demanded a clear temporal relationship between clinical stroke and the onset of dementia to diagnose Vascular Dementia (VaD).¹ However, the advent of advanced neuroimaging and large-scale autopsy studies has dismantled this binary, revealing that "pure" pathologies are the exception rather than the rule in the aging brain. Current epidemiological data indicate that while pure VaD accounts for approximately 5-10% of dementia cases, cerebrovascular pathology contributes to cognitive decline in 30-40% of all dementia diagnoses, frequently co-occurring with AD pathology.¹ This overlap creates a spectrum of "Mixed Dementia," where vascular injury and proteinopathy act additively or synergistically to accelerate cognitive decline.² The complexity is further compounded by the recent characterization of Limbic-predominant Age-related TDP-43 Encephalopathy (LATE), a non-amyloid neurodegenerative condition that mimics AD clinically but lacks specific in vivo biomarkers, often leading to misdiagnosis in elderly populations.⁴ 1.2 The Diagnostic Gap: Absence of Specific Proteinopathies The central problem addressed by this thesis is the diagnostic ambiguity that arises in the absence of specific proteinopathies. The "A/T/N" framework (Amyloid/Tau/Neurodegeneration) has revolutionized AD research by providing biological definitions for the disease.⁶ However, this framework creates a diagnostic orphan category: patients who are amyloid-negative (A-) and tau-negative (T-) but exhibit significant neurodegeneration (N+) and cognitive impairment. These patients, often labeled as "Suspected Non-Alzheimer’s Pathophysiology" (SNAP), may suffer from pure VCI, LATE, or other rare etiologies. Standard structural imaging, such as T2-weighted MRI, identifies White Matter Hyperintensities (WMH), but these lesions are non-specific and correlate imperfectly with cognition.⁷ A patient with extensive WMH may be cognitively normal, while another with mild burden suffers severe executive dysfunction. This "clinico-radiological paradox" suggests that conventional imaging fails to capture the true extent of tissue injury and the underlying biological drivers.⁸ Furthermore, the mere presence of WMH does not confirm VCI as the sole driver of dementia, as WMH are also observed in AD and LATE.⁹ 1.3 The Vascular-Immune Hypothesis To resolve this diagnostic dilemma, this thesis adopts the "Vascular-Immune Hypothesis" as its theoretical lens. This hypothesis postulates that vascular contributions to cognitive impairment are not merely structural (e.g., infarcts) but active biological processes driven by neuroimmune dysregulation.¹⁰ According to this model, cardiovascular risk factors trigger endothelial dysfunction and blood-brain barrier (BBB) breakdown. This leads to the extravasation of serum proteins (e.g., fibrinogen) and the activation of perivascular immune cells, releasing pro-inflammatory cytokines (IL-6) and matrix-degrading enzymes (MMP-9).¹² This inflammatory cascade results in diffuse microstructural disintegration of the white matter—changes that are subtle on standard MRI but detectable via advanced diffusion metrics like PSMD and Free-Water imaging.¹⁴ Simultaneously, the shedding of inflammatory mediators into the cerebrospinal fluid (CSF) and plasma provides a distinct "fluid signature" of vascular injury.¹⁶ By integrating these domains, it becomes possible to fingerprint the vascular etiology of cognitive impairment even when specific proteinopathies are absent or ambiguous. 1.4 Research Objectives This doctoral thesis aims to rigorously answer the question: How can the integration of advanced neuroimaging and fluid biomarkers for neuroinflammation effectively distinguish pure VCI from mixed-pathology dementia? The specific research objectives are:

1. To Define the Microstructural Phenotype of VCI: To evaluate the diagnostic utility of Peak Width of Skeletonized Mean Diffusivity (PSMD) and Free-Water (FW) imaging in detecting diffuse vascular injury beyond visible WMH.
2. To Characterize the Vascular-Inflammatory Biomarker Profile: To assess the specificity of MMP-9, TIMP-1, and PlGF in distinguishing vascular-driven inflammation from the astrocytic reactivity (GFAP) associated with AD.
3. To Disentangle Co-Pathologies: To construct a diagnostic algorithm that differentiates Pure VCI from LATE-NC and Mixed Dementia, providing a roadmap for precision diagnosis in amyloid-negative cohorts.

Chapter 2: Literature Review and Theoretical

Framework

2.1 The Evolution of Vascular Cognitive Impairment Concepts The conceptualization of vascular contributions to dementia has undergone significant revision over the past century. Early 20th-century definitions focused on "hardening of the arteries" (arteriosclerosis) as the primary cause of senility. This was replaced in the 1970s and 80s by the "multi-infarct" paradigm, which posited that dementia resulted from the cumulative volume of necrotic tissue caused by large-vessel strokes.¹ However, the "multi-infarct" model proved inadequate for explaining the cognitive decline seen in patients with diffuse small vessel disease (SVD) but no history of acute stroke. This led to the introduction of the term "Vascular Cognitive Impairment" (VCI) by Hachinski and Bowler, a broad category encompassing everything from mild cognitive impairment (MCI) to frank vascular dementia.² More recently, the focus has shifted to "Vascular Contributions to Cognitive Impairment and Dementia" (VCID), acknowledging that vascular pathology is a ubiquitous co-conspirator in neurodegeneration.² Current diagnostic criteria, such as the VICCCS guidelines, emphasize the heterogeneity of VCI, categorizing it into post-stroke dementia, subcortical ischemic vascular dementia (SIVD), multi-infarct dementia, and mixed dementia.¹⁸ SIVD, characterized by extensive small vessel pathology, is the most common form and the most difficult to distinguish from AD in its early stages due to its insidious, progressive course.¹⁸ 2.2 The Biology of Cerebral Small Vessel Disease (SVD) Cerebral Small Vessel Disease (SVD) affects the small perforating arteries, arterioles, capillaries, and venules that supply the brain's deep structures.³ These vessels are uniquely vulnerable to hypertension and metabolic dysregulation. The hallmark pathologies are arteriolosclerosis (lipohyalinosis) and cerebral amyloid angiopathy (CAA).³ 2.2.1 The Neurovascular Unit (NVU) The central pathophysiological structure in VCI is the Neurovascular Unit (NVU), comprising the endothelial cell, pericyte, astrocytic end-feet, and the adjacent neuron.¹² In health, the NVU maintains the Blood-Brain Barrier (BBB), regulating the transport of nutrients and clearing metabolic waste via the glymphatic system.²¹ In SVD, endothelial dysfunction disrupts NVU coupling. Chronic hypoperfusion creates a hypoxic environment, upregulating Hypoxia-Inducible Factor 1-alpha (HIF-1α).²⁰ This triggers the release of Vascular Endothelial Growth Factor (VEGF) and Matrix Metalloproteinases (MMPs). While intended to promote angiogenesis, in the chronic phase, these factors increase BBB permeability, leading to fluid extravasation and "vasogenic edema".⁸ 2.3 The Vascular-Immune Hypothesis of Alzheimer's and VCI The "Vascular-Immune Hypothesis" represents a paradigm shift, proposing that vascular injury and immune activation are not merely downstream consequences of amyloid deposition but are primary drivers of synaptic failure.¹⁰ Recent reviews from 2024 and 2025 highlight that vascular risk factors (hypertension, diabetes) induce a state of "inflammaging"—chronic, low-grade systemic inflammation that degrades the BBB.²³ Once the barrier is breached, serum proteins like fibrinogen enter the parenchyma, activating microglia and converting them to a pro-inflammatory phenotype.¹² This creates a toxic cycle:

1. Vascular Injury: Arteriolosclerosis reduces compliance and blood flow.
2. BBB Leakage: MMPs degrade tight junctions.
3. Inflammation: Infiltrating leukocytes and activated microglia release ROS and cytokines.
4. White Matter Damage: Oligodendrocytes, highly sensitive to oxidative stress and inflammation, undergo apoptosis, leading to demyelination (WMH).²⁴ Crucially, this hypothesis provides a mechanism for distinguishing VCI from AD. While AD also involves inflammation, it is primarily driven by plaque-associated microglia and often involves impaired clearance of amyloid rather than the gross BBB disruption and mechanical pulsatility failure seen in VCI.¹¹ 2.4 The Challenge of LATE-NC Limbic-predominant Age-related TDP-43 Encephalopathy (LATE) has recently been defined as a distinct neurodegenerative entity affecting adults over 80.⁴ LATE is characterized by TDP-43 proteinopathy in the amygdala and hippocampus, causing an amnestic syndrome that mimics AD. Crucially, LATE patients are often amyloid-negative.²⁸ Because LATE is highly prevalent in the oldest-old, the same demographic most affected by SVD, co-occurrence is common. Identifying pure VCI requires excluding LATE, yet there are no FDA-approved fluid biomarkers for TDP-43.⁵ This thesis proposes that distinguishing VCI from LATE requires examining the discrepancy between vascular burden (high in VCI, variable in LATE) and specific patterns of atrophy (medial temporal lobe atrophy is severe in LATE, often mild in pure VCI).²⁹

Chapter 3: Methodology and Approach

3.1 Research Design This thesis employs a theoretical synthesis approach, integrating data from a wide array of recent clinical studies, meta-analyses, and neuropathological reviews (2020-2025) to construct a diagnostic framework. The methodology mimics a "virtual multi-modal cohort study," synthesizing findings from distinct domains—radiomics (imaging) and fluidomics (biomarkers)—to propose a unified diagnostic model. 3.2 Criteria for Biomarker Selection The selection of biomarkers for this framework was based on the following criteria:

1. Specificity: The marker must show differential expression in VCI vs. AD/LATE.
2. Mechanistic Relevance: The marker must reflect a core pathological process of the Vascular-Immune Hypothesis (e.g., BBB breakdown, demyelination).
3. Clinical Feasibility: The marker should be measurable in plasma or MRI, maximizing accessibility (though CSF data is also considered). Selected biomarkers include:

● Imaging: PSMD (SVD burden), Free-Water (Microstructure/Inflammation), DTI-ALPS

(Glymphatics). ● Fluid: MMP-9 (Vascular Inflammation), GFAP (Astrocytosis), NfL (Neuroaxonal injury), PlGF (Angiogenesis). 3.3 Data Integration Strategy The analysis integrates these markers by mapping them onto the two primary axes of pathology:

1. The Vascular Axis: Driven by SVD, ischemia, and BBB leakage.
2. The Neurodegenerative Axis: Driven by protein aggregation (Amyloid/Tau/TDP-43) and subsequent gliosis. By evaluating where a patient falls on these two axes, distinct phenotypes (Pure VCI, Pure AD/LATE, Mixed) can be resolved.

Chapter 4: Advanced Neuroimaging of White Matter and Glymphatics

Standard T2-weighted/FLAIR MRI is the workhorse of VCI diagnosis, visualizing White Matter Hyperintensities (WMH) of presumed vascular origin. However, WMH volume is a crude metric. It fails to distinguish between etiology (ischemic vs. inflammatory vs. degenerative) and lacks sensitivity to changes in the "Normal Appearing White Matter" (NAWM).⁷ This chapter explores advanced diffusion metrics that surpass these limitations. 4.1 Peak Width of Skeletonized Mean Diffusivity (PSMD) Peak Width of Skeletonized Mean Diffusivity (PSMD) has emerged as one of the most robust and biologically valid biomarkers for Cerebral Small Vessel Disease (SVD).¹⁴ 4.1.1 Technical Principles PSMD is derived from Diffusion Tensor Imaging (DTI). Unlike standard Mean Diffusivity (MD) which averages diffusion across a region, PSMD assesses the heterogeneity of diffusion. The process involves:

1. Skeletonization: FSL-TBSS software projects the subject’s MD map onto a standard white matter skeleton, eliminating contamination from CSF and gray matter partial volume effects.¹⁴
2. Histogram Analysis: A histogram of MD values within the skeleton is generated.
3. Calculation: PSMD is the difference between the 95th and 5th percentiles of this histogram.³² 4.1.2 Diagnostic Utility in VCI A wider peak (High PSMD) indicates greater variance in water diffusivity, reflecting a brain riddled with micro-lesions, demyelination, and lacunes. Studies consistently show that PSMD correlates more strongly with processing speed—the cognitive hallmark of VCI—than total WMH volume.³¹ Crucially, PSMD acts as a specific marker for the vascular contribution to cognitive decline. In comparisons between AD and SVD cohorts, PSMD is significantly higher in SVD.¹⁴ In mixed cohorts, PSMD mediates the relationship between vascular risk factors and cognitive impairment, acting as the structural correlate of "vascular burden".³³

Table 4.1: PSMD as a Discriminator

Condition PSMD Value Biological Correlate

Healthy Control Low Intact Myelin, Homogeneous Diffusion Pure VCI (SVD) Very High Diffuse Demyelination, Micro-infarcts, Edema

Pure AD Low/Moderate Tract-specific

degeneration (e.g., Cingulum), less diffuse

Mixed Dementia High Additive effect of vascular

injury 4.2 Free-Water (FW) Imaging: Mapping Neuroinflammation Free-Water (FW) imaging is a bi-tensor DTI model that separates the diffusion signal into two compartments:

1. Tissue Compartment: Water molecules restricted by axons and myelin (providing FA/MD metrics corrected for edema).
2. Free-Water Compartment: Extracellular water that is free to diffuse (representing edema, atrophy, or inflammation).¹⁵ 4.2.1 Deep vs. Juxtacortical Signatures The spatial distribution of Free Water offers a powerful tool for differentiation.

● VCI Pattern: In VCI, BBB leakage and venous hypertension lead to interstitial fluid accumulation primarily in the deep white matter and periventricular regions. This

"vasogenic edema" results in a widespread elevation of FW in the deep tracts.³⁴

● AD/Mixed Pattern: In AD, FW elevation is prominently observed in the juxtacortical white matter (U-fibers adjacent to the cortex). This reflects the underlying cortical

neurodegeneration, atrophy, and microscopic tau pathology spreading from the gray matter into the white matter.³⁶ Therefore, a ratio of Deep-to-Juxtacortical FW could serve as a phenotypic marker. High deep FW suggests VCI; high juxtacortical FW suggests neurodegeneration. 4.3 DTI-ALPS: Assessing Glymphatic Clearance The Diffusion Tensor Imaging Analysis Along the Perivascular Space (DTI-ALPS) measures the diffusivity of water along the perivascular spaces (PVS) of the medullary veins, serving as a proxy for glymphatic function.²¹ 4.3.1 Glymphatics in VCI vs. AD While both VCI and AD involve glymphatic failure, the mechanisms differ.

● VCI: Arteriolosclerosis leads to vessel stiffening. Since glymphatic flow is driven by

arterial pulsatility, vascular stiffness directly halts clearance. Low DTI-ALPS scores are strongly predictive of VCI severity and future dementia risk in stroke patients.²¹

● AD: Glymphatic failure is often related to Aβ deposition blocking the PVS (CAA) or

aquaporin-4 dysregulation. However, the combination of PSMD and DTI-ALPS is potent. VCI is characterized by the

specific combination of High PSMD (tissue damage) and Low DTI-ALPS (clearance failure).³⁷

This "Vascular Failure" profile contrasts with early AD, where tissue heterogeneity (PSMD) might be lower despite localized clearance issues.

Chapter 5: Fluid Biomarkers: The Inflammatory

Fingerprint

While neuroimaging visualizes the structural consequences of disease, fluid biomarkers provide a real-time readout of the molecular pathophysiology. To distinguish VCI from mixed pathologies in the absence of amyloid/tau, we must rely on biomarkers that reflect the specific inflammatory cascades of the neurovascular unit. 5.1 Matrix Metalloproteinase-9 (MMP-9): The Vascular Remodeler Matrix Metalloproteinase-9 (MMP-9) is arguably the most specific circulating biomarker for vascular brain injury available today. It is an inducible zinc-metalloproteinase responsible for degrading the extracellular matrix, particularly the basal lamina components (Type IV collagen, laminin) that support the BBB.¹³ 5.1.1 MMP-9 in VCI vs. AD Multiple studies have established that MMP-9 levels are significantly elevated in the CSF and plasma of patients with VCI compared to AD and healthy controls.¹⁶

● Mechanism: In VCI, chronic hypoxia and endothelial activation trigger the release of

MMP-9 from endothelial cells and neutrophils. This leads to BBB opening and white matter degradation.

● Differentiation: In pure AD (without significant vascular pathology), MMP-9 levels are

often normal or only mildly elevated. The "Vascular-Immune" signature of VCI is defined by a massive upregulation of proteolytic activity.¹⁶ 5.1.2 The MMP-9/TIMP-1 Ratio MMP-9 is regulated by Tissue Inhibitor of Metalloproteinase-1 (TIMP-1). In VCI, this balance is

disrupted. A high MMP-9/TIMP-1 ratio indicates unchecked proteolysis and active BBB

breakdown.³⁸ This ratio correlates with white matter lesion volume and poor cognitive outcomes in stroke patients.³⁹ 5.2 Glial Fibrillary Acidic Protein (GFAP): The Astrocytic Discriminator Glial Fibrillary Acidic Protein (GFAP) is a marker of reactive astrogliosis. While traditionally viewed as a general injury marker, recent ultrasensitive plasma assays have revealed a distinct specificity for amyloid pathology. 5.2.1 GFAP in AD vs. VCI Plasma GFAP is markedly elevated in amyloid-positive individuals (AD and Mixed Dementia) due to the intense astrocytic reaction to amyloid plaques.⁴⁰ Crucially, in patients with Pure VCI (amyloid-negative), GFAP levels are significantly lower, even in the presence of extensive WMH.⁴¹

● The Discrimination: This creates a powerful dichotomy. VCI drives NfL (axonal injury) and MMP-9 (vascular injury) but induces only a moderate GFAP response. AD drives

GFAP aggressively.

● Study Data: The Mayo Clinic study on plasma biomarkers in VCI found that GFAP

separated VCI from AD better in amyloid-negative groups, establishing it as a negative predictor for Pure VCI.⁴¹ 5.3 Placental Growth Factor (PlGF) and Angiogenesis Placental Growth Factor (PlGF), a member of the VEGF family, is another biomarker gaining traction for VCI.

● Vascular Specificity: PlGF is upregulated in response to cerebral

hypoperfusion/ischemia. Higher levels correlate with greater WMH volume and cognitive decline.⁸

● Mechanism: It represents the brain's attempt at angiogenesis in the face of capillary

rarefaction. Elevated PlGF, particularly when combined with high MMP-9, signals an active "angiogenic-inflammatory" state characteristic of VCI.⁴² 5.4 Neurofilament Light (NfL): The Non-Specific Intensity Marker Neurofilament Light (NfL) is elevated in almost all neurodegenerative conditions, reflecting axonal destruction. While not specific for etiology, it serves as a "severity" marker. ● In VCI, NfL correlates strongly with PSMD and WMH burden.⁴⁰

● Diagnostic Utility: NfL confirms the presence of neurodegeneration. Its utility lies in

combination: High NfL + High GFAP suggests AD/Mixed. High NfL + Low GFAP suggests VCI or LATE.

Chapter 6: The Confounder of LATE-NC and Differential Diagnosis

The "absence of specific proteinopathies" often implies a patient who is amyloid-negative. In the oldest-old, this clinical picture is frequently caused by Limbic-predominant Age-related TDP-43 Encephalopathy (LATE). Because LATE lacks a specific fluid biomarker, it is the primary mimic of VCI in amyloid-negative cohorts. Distinguishing Pure VCI from LATE (or VCI + LATE) is essential for prognosis, as LATE progresses more slowly but is resistant to vascular interventions.⁴ 6.1 Clinical and Radiological Profiles of LATE LATE neuropathological change (LATE-NC) affects the medial temporal lobe (MTL), causing an amnestic syndrome.

● Atrophy Pattern: LATE is associated with severe, disproportionate atrophy of the

hippocampus and amygdala.⁵

● Cognitive Profile: Predominantly episodic memory loss, contrasting with the

dysexecutive profile of VCI.⁴ 6.2 Distinguishing VCI from LATE Since we cannot yet measure TDP-43 directly in plasma (though NULISA assays are in development ³⁰), we must use a "Process of Elimination" based on the biomarkers established in previous chapters.

The "Vascular-Atrophy Mismatch"

● Pure VCI: High Vascular Burden (High PSMD, High MMP-9) + Mild/Moderate

Hippocampal Atrophy (often proportional to global atrophy).

● Pure LATE: Low Vascular Burden (Low PSMD, Normal MMP-9) + Severe Hippocampal

Atrophy (Hippocampal Sclerosis). ● Mixed VCI + LATE: High Vascular Burden + Severe Hippocampal Atrophy. This distinction is vital. A patient with severe memory loss and WMH might be diagnosed as Mixed Dementia. However, if their MMP-9 is normal and PSMD is low, the WMH may be incidental, and the primary driver is likely LATE. Conversely, high MMP-9 confirms the vascular lesions are active and contributing to the pathology.²⁹

Chapter 7: Integrated Diagnostic Framework and Conclusions

7.1 The Integrated Algorithm Synthesizing the neuroimaging and fluid biomarker data, this thesis proposes a multi-modal diagnostic algorithm for stratifying cognitively impaired patients in the absence of amyloid/tau biomarkers.

Figure 7.1: The VCI vs. Mixed Diagnostic Matrix

Phenotype Pure VCI Mixed (VCI + Pure LATE Mixed (VCI +

AD) LATE)

Pathophysiol Endothelial Ischemia + TDP-43 Ischemia + ogy Dysfunction, Amyloid/Tau Proteinopathy TDP-43

Ischemia

PSMD (DTI) High (>95th High Normal/Mild High

%ile)

Free Water Deep WM Juxtacortical Medial Deep + Medial

Dominant + Deep Temporal Temporal

MMP-9 Elevated Elevated Normal Elevated

Plasma GFAP Low/Normal Very High Low/Normal Low/Normal

PlGF High Variable Normal High

Atrophy Subcortical/Dif Hippocampal + Severe Severe fuse Cortical Hippocampal Hippocampal

7.2 Interpretation of the "Absence of Proteinopathies" In a patient who is Amyloid-Negative (A-):

1. High PSMD + High MMP-9 + Low GFAP: Confirms Pure VCI. The cognitive decline is driven by white matter disintegration and vascular inflammation.
2. Low PSMD + Normal MMP-9 + Severe Atrophy: Suggests Pure LATE. The vascular markers rule out VCI as the primary driver; the atrophy points to TDP-43.
3. High PSMD + High MMP-9 + Severe Atrophy: Suggests VCI + LATE. Both pathologies are active. 7.3 Conclusion and Future Directions The "Vascular-Immune Hypothesis" provides the missing link in the differential diagnosis of non-amyloid dementia. By moving beyond simple structural MRI and adopting advanced microstructural metrics (PSMD, Free-Water) alongside specific inflammatory biomarkers (MMP-9, PlGF), clinicians can effectively phenotype patients who fall into the diagnostic "gray zone." This framework has profound implications for clinical trials. Currently, "vascular dementia" trials likely recruit significant numbers of LATE patients (diluting efficacy), while AD trials exclude patients with WMH who might benefit from vascular optimization. The precision phenotyping proposed here allows for the separation of "Angiogenic-Inflammatory" dementias (VCI) from "Neurodegenerative-Astrocytic" dementias (AD), paving the way for targeted therapies that address the specific biological drivers of cognitive decline—whether they be the vessel wall, the blood-brain barrier, or the neuron itself. Future research must focus on the standardization of PSMD acquisition protocols across scanners ⁴³ and the validation of plasma TDP-43 assays ³⁰ to eventually replace the inferential diagnosis of LATE. Until then, the multi-modal integration of white matter imaging and inflammatory fluid markers remains our most powerful tool for dissecting the vascular contribution to the dementia continuum.

Citations

¹

Works cited

1. Nelson NW. Differential Diagnosis of Alzheimer’s Dementia and Vascular Dementia. Disease-a-Month 2007;53(3):148-151. DOI 10.1016/j.disamonth.2007.04.004.

2. Smith EE, Aparicio HJ, Gottesman RF, Goyal MS, Greenberg SM, Schneider JA, 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). DOI 10.1161/STR.0000000000000494.

3. Karamchandani R, Barbas NR. Vascular Cognitive Impairment. Oxford Medicine Online 2017. DOI 10.1093/med/9780199937837.003.0021.

4. Nelson PT, Schneider JA, Jicha GA, Duong MT, Wolk DA. When Alzheimer's is LATE: Why Does it Matter?. Ann Neurol 2023;94(2):211-222. DOI 10.1002/ana.26711.

5. New diagnostic guidelines aim to distinguish common memory disorder from Alzheimer's, accessed December 12, 2025, https://www.news-medical.net/news/20250128/New-diagnostic-guidelines-aim-t o-distinguish-common-memory-disorder-from-Alzheimere28099s.aspx

6. Mendes AJ, Ribaldi F, Pievani M, Boccalini C, Garibotto V, Frisoni GB. Validating the Amyloid Cascade Through the Revised Criteria of Alzheimer's Association Workgroup 2024 for Alzheimer Disease. Neurology 2025;104(11). DOI 10.1212/WNL.0000000000213675.

7. Heiss WD, Rosenberg GA, Thiel A, Berlot R, de Reuck J. Neuroimaging in vascular cognitive impairment: a state-of-the-art review. BMC Med 2016;14(1):174. DOI 10.1186/s12916-016-0725-0.

8. Edwards NC, Lao P, Alshikho MJ, Mazen J, Huber B, Hale C, et al.. Alzheimer Disease, Vascular Disease, and Blood-Brain Barrier Permeability Biomarkers in Middle-Aged Adults. Neurology 2025;105(9). DOI 10.1212/WNL.0000000000214220.

9. Newton P, Tchounguen J, Pettigrew C, Lim C, Lin Z, Lu H, et al.. Regional White Matter Hyperintensities and Alzheimer's Disease Biomarkers Among Older Adults with Normal Cognition and Mild Cognitive Impairment. J Alzheimers Dis 2023;92(1):323-339. DOI 10.3233/JAD-220846.

10. New study reveals how blood vessel cells shape dementia risk - Imperial College

London, accessed December 12, 2025, https://www.imperial.ac.uk/news/271357/new-study-reveals-blood-vessel-cells/ 11. (PDF) The Vascular-Immune Hypothesis of Alzheimer's Disease - ResearchGate, accessed December 12, 2025, https://www.researchgate.net/publication/367555939_The_Vascular-Immune_Hyp othesis_of_Alzheimer's_Disease

12. Rosenberg GA. Matrix Metalloproteinase-Mediated Neuroinflammation in Vascular Cognitive Impairment of the Binswanger Type. Cell Mol Neurobiol 2016;36(2):195-202. DOI 10.1007/s10571-015-0277-4.

13. Turner RJ, Sharp FR. Implications of MMP9 for Blood Brain Barrier Disruption and Hemorrhagic Transformation Following Ischemic Stroke. Front Cell Neurosci 2016;10:56.016.00056.

    Find this paper

14. Zanon Zotin MC, Yilmaz P, Sveikata L, Schoemaker D, van Veluw SJ, Etherton MR, et al.. Peak Width of Skeletonized Mean Diffusivity: A Neuroimaging Marker for White Matter Injury. Radiology 2023;306(3). DOI 10.1148/radiol.212780.

15. Free-water: A promising structural biomarker for cognitive decline in aging and mild cognitive impairment | Imaging Neuroscience | MIT Press, accessed December 12, 2025, https://direct.mit.edu/imag/article/doi/

    Find this paper

16. Adair JC, Charlie J, Dencoff JE, Kaye JA, Quinn JF, Camicioli RM, et al.. Measurement of gelatinase B (MMP-9) in the cerebrospinal fluid of patients with vascular dementia and Alzheimer disease. Stroke 2004;35(6):e159-62. DOI 10.1161/01.STR.0000127420.10990.76.

17. Dichgans M, Leys D. Vascular Cognitive Impairment. Circulation Research 2017;120(3):573-591. DOI 10.1161/circresaha.116.308426.

18. Chang Wong E, Chang Chui H. Vascular Cognitive Impairment and Dementia. Continuum (Minneap Minn) 2022;28(3):750-780. DOI 10.1212/CON.0000000000001124.

19. Iadecola C, Duering M, Hachinski V, Joutel A, Pendlebury ST, Schneider JA, et al.. Vascular Cognitive Impairment and Dementia: JACC Scientific Expert Panel. J Am Coll Cardiol 2019;73(25):3326-3344. DOI 10.1016/j.jacc.2019.04.034.

20. Rosenberg GA. Inflammation and White Matter Damage in Vascular Cognitive Impairment. Stroke 2009;40(3_suppl_1). DOI 10.1161/strokeaha.108.533133.

21. Grigoras ML, Bondar AC, Bratosin F, Bogdan IG, Marc F. Diffusion Tensor Imaging Along the Perivascular Space (DTI-ALPS) in Ischemic Stroke: A Systematic Review of Diagnostic and Prognostic Performance for Post-Stroke Cognitive Impairment. Diagnostics (Basel) 2025;15(22). DOI 10.3390/diagnostics15222905.

22. Matrix Metalloproteinases and Blood-Brain Barrier Disruption in Acute Ischemic Stroke - Frontiers, accessed December 12, 2025, https://www.frontiersin.org/journals/neurology/articles/ full

23. Rundek T, Tolea M, Ariko T, Fagerli EA, Camargo CJ. Vascular Cognitive Impairment (VCI). Neurotherapeutics 2022;19(1):68-88. DOI 10.1007/s13311-021-01170-y.

24. Tian W, Kyriakides TR. Matrix metalloproteinase-9 deficiency leads to prolonged foreign body response in the brain associated with increased IL-1beta levels and leakage of the blood-brain barrier. Matrix Biol 2009;28(3):148-59. DOI 10.1016/j.matbio.2009.02.002.

25. Asahi M, Wang X, Mori T, Sumii T, Jung JC, Moskowitz MA, et al.. Effects of matrix metalloproteinase-9 gene knock-out on the proteolysis of blood-brain barrier and white matter components after cerebral ischemia. J Neurosci 2001;21(19):7724-32. DOI 10.1523/JNEUROSCI.21-19-07724.2001.

26. Nelson PT, Dickson DW, Trojanowski JQ, Jack CR, Boyle PA, Arfanakis K, et al.. Limbic-predominant age-related TDP-43 encephalopathy (LATE): consensus working group report. Brain 2019;142(6):1503-1527.awz099.

    Find this paper

27. Nelson PT, Schneider JA, Jicha GA, Duong MT, Wolk DA. When Alzheimer's is LATE: Why Does it Matter?. Ann Neurol 2023;94(2):211-222. DOI 10.1002/ana.26711.

28. LATE, Explained - Alzheimer's Disease Research Center - University of Wisconsin–Madison, accessed December 12, 2025, https://www.adrc.wisc.edu/dementia-matters/late-explained

29. Plasma TDP-43 is a potential biomarker for advanced limbic-predominant age-related TDP-43 encephalopathy neuropathologi - medRxiv, accessed December 12, 2025, DOI 10.1101/2025.07.09.25331146">https://www.medrxiv.org/content/10.1101/2025.07.09.25331146

30. Luckey AM, Ghosh S, Wang CP, Beiser A, Bernal R, Li Z, et al.. Biological validation of peak-width of skeletonized mean diffusivity as a VCID biomarker: The MarkVCID Consortium. Alzheimers Dement 2024;20(12):8814-8824. DOI 10.1002/alz.14345.

31. Low A, Mak E, Stefaniak JD, Malpetti M, Nicastro N, Savulich G, et al.. Peak Width of Skeletonized Mean Diffusivity as a Marker of Diffuse Cerebrovascular Damage. Front Neurosci 2020;14:238. DOI 10.3389/fnins.2020.00238.

32. Hu Q, Xiao Z, Zhou X, Zhao Q, Ding D, Zhang J. Peak width of skeletonized mean diffusivity mediates the relationship between cerebral small vessel disease burden and cognitive impairment in community-dwelling older adults. Brain Commun 2025;7(3):fcaf233. DOI 10.1093/braincomms/fcaf233.

33. Maillard P, Fletcher E, Singh B, Martinez O, Johnson DK, Olichney JM, et al.. Cerebral white matter free water. Neurology 2019;92(19). DOI 10.1212/WNL.0000000000007449.

34. Bergamino M, McElvogue MM, Stokes AM, Alzheimer's Disease Neuroimaging Initiative. Distinguishing Early from Late Mild Cognitive Impairment Using Magnetic Resonance Free-Water Diffusion Tensor Imaging. NeuroSci 2025;6(1). DOI 10.3390/neurosci6010008.

35. Diffusion MRI reveals new pattern underlying Alzheimer's disease | AuntMinnie, accessed December 12, 2025, https://www.auntminnie.com/clinical-news/mri/article/15768821/diffusion-mri-reve als-new-pattern-underlying-alzheimers-disease

    Find this paper

36. Relationship of Perivascular Space Markers With Incident Dementia in Cerebral Small Vessel Disease | Stroke - American Heart Association Journals, accessed December 12, 2025, https://www.ahajournals.org/doi/ OKEAHA.123.045857

    Find this paper

37. Bašić J, Milošević V, Djordjević B, Stojiljković V, Živanović M, Stefanović N, et al.. Matrix Remodeling Enzymes as Potential Fluid Biomarkers of Neurodegeneration in Alzheimer's Disease. Int J Mol Sci 2024;25(11). DOI 10.3390/ijms25115703.

38. Włodarczyk L, Cichon N, Karbownik MS, Saluk J, Miller E. Exploring the Role of MMP-9 and MMP-9/TIMP-1 Ratio in Subacute Stroke Recovery: A Prospective Observational Study. Int J Mol Sci 2024;25(11). DOI 10.3390/ijms25115745.

39. Raghavan S, Graff-Radford J, Hofrenning E, Fought AJ, Reid RI, Kamykowski MG, et al.. Plasma NfL and GFAP for predicting VCI and related brain changes in community and clinical cohorts. Alzheimers Dement 2025;21(6):e70381. DOI 10.1002/alz.70381.

40. Raghavan S, Graff-Radford J, Hofrenning E, Fought AJ, Reid RI, Kamykowski MG, et al.. Plasma NfL and GFAP for predicting VCI and related brain changes in community and clinical cohorts. Alzheimers Dement 2025;21(6):e70381. DOI 10.1002/alz.70381.

41. Edwards NC, Lao P, Alshikho MJ, Mazen J, Huber B, Hale C, et al.. Alzheimer Disease, Vascular Disease, and Blood-Brain Barrier Permeability Biomarkers in Middle-Aged Adults. Neurology 2025;105(9):e214220. DOI 10.1212/WNL.0000000000214220.

42. Biesbroek JM, de Kort FA, Anblagan D, Bastin ME, Beiser A, Brodaty H, et al.. Signature White Matter Hyperintensity Locations Associated With Vascular Risk Factors Derived From 15 653 Individuals. Stroke 2025;56(10):3047-3059. DOI 10.1161/STROKEAHA.125.051159.

43. Custodero C, Ciavarella A, Panza F, Gnocchi D, Lenato GM, Lee J, et al.. Role of inflammatory markers in the diagnosis of vascular contributions to cognitive impairment and dementia: a systematic review and meta-analysis. Geroscience 2022;44(3):1373-1392. DOI 10.1007/s11357-022-00556-w.

44. Dumont M, Roy M, Jodoin PM, Morency FC, Houde JC, Xie Z, et al.. Free Water in White Matter Differentiates MCI and AD From Control Subjects. Front Aging Neurosci 2019;11:270.19.00270.

    Find this paper

45. Metti AL, Cauley JA. How predictive of dementia are peripheral inflammatory markers in the elderly?. Neurodegener Dis Manag 2012;2(6):609-622. DOI 10.2217/NMT.12.68.

46. Campaña Perilla LA, Mahecha Carvajal ME, Cardona Ortegón JD, Barragán Corrales C, Barrera Patiño AM. Diffusion Tensor Imaging Protocol: The Need for Standardization of Measures. Radiology 2023;308(2):e230470. DOI 10.1148/radiol.230470.

The Validity Ledger

The argument above is only as strong as its weakest load-bearing joint, and the reader is owed an explicit accounting of where it stands on the ground and where it stands on inference.