What does this research mean for the field?

Global (Q alb) and regional (K trans) measures of blood-brain barrier permeability capture complementary dimensions of dysfunction in dementia, and plasma biomarkers cannot be assumed to substitute for CSF markers in assessing these processes. Novelty: ClaimNovelty.INCREMENTAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To determine how effectively CSF, plasma, and MRI biomarkers reflect neuroinflammation and blood–brain barrier disruption across various forms of cognitive impairment and dementia.

September 14, 2026Frontiers in AgingOpen Access

Explanatory limitations of CSF, plasma, and MRI biomarkers for inflammation and blood–brain barrier disruption in dementia

View Full Paper

Ask AI

Bookmark

Authors

AGAzadeh Golduzian EEErik B. Erhardt ACArvind Caprihan

Discussion

Member takes

Overview

Cross-sectional study reveals disparate fluid and imaging measures of blood–brain barrier breakdown in dementia, indicating that plasma cannot simply substitute for cerebrospinal fluid.

Key Points

To determine how effectively CSF, plasma, and MRI biomarkers reflect neuroinflammation and blood–brain barrier disruption across various forms of cognitive impairment and dementia.
Analyzed 149 participants with Alzheimer’s disease, leukoaraiosis, mixed dementia, subcortical ischemic vascular dementia, or memory impairment from the MarkVCID and UNM ADRC cohorts.
Measured global BBB permeability via CSF albumin index (Qalb) and regional permeability via dynamic contrast-enhanced MRI (Ktrans), comparing them to fluid biomarkers (cytokines, MMPs, GFAP, NfL) and diffusion/structural MRI metrics.
Assessed univariate and multivariable associations using AIC-based stepwise selection, bootstrap selection frequencies, penalized regression, and cross-validated R².
Global permeability (Qalb) and regional permeability (Ktrans) did not significantly correlate (Spearman ρ = 0.14, p = 0.28), demonstrating they capture distinct BBB dysfunction dimensions.
Multivariable CSF models explained 53% of variance (R² = 0.53) in Qalb (retaining MMP-2, Flt-1, IL-8, GFAP, and NfL), compared to R² = 0.32 for plasma; for Ktrans, CSF, plasma, and MRI models explained R² of 0.36, 0.39, and 0.14, respectively.
CSF-to-plasma biomarker correlations varied widely from r = -0.03 (MMP-9) to r = 0.74 (NfL) and r = 0.70 (IL-13), indicating plasma biomarkers cannot routinely substitute for CSF measurements.

Cite This Study

Golduzian et al. (2026) studied this question.

synapsesocial.com/papers/6aa7b2e10926e14a848b1799 https://doi.org/10.3389/fragi.2026.1923046

View Full Paper

Ask AI

Bookmark

September 14, 2026Frontiers in AgingOpen Access

Explanatory limitations of CSF, plasma, and MRI biomarkers for inflammation and blood–brain barrier disruption in dementia

View Full Paper

Ask AI

Bookmark

Authors

AGAzadeh Golduzian EEErik B. Erhardt ACArvind Caprihan

Discussion

Member takes

Overview

Cross-sectional study reveals disparate fluid and imaging measures of blood–brain barrier breakdown in dementia, indicating that plasma cannot simply substitute for cerebrospinal fluid.

Key Points

To determine how effectively CSF, plasma, and MRI biomarkers reflect neuroinflammation and blood–brain barrier disruption across various forms of cognitive impairment and dementia.
Analyzed 149 participants with Alzheimer’s disease, leukoaraiosis, mixed dementia, subcortical ischemic vascular dementia, or memory impairment from the MarkVCID and UNM ADRC cohorts.
Measured global BBB permeability via CSF albumin index (Qalb) and regional permeability via dynamic contrast-enhanced MRI (Ktrans), comparing them to fluid biomarkers (cytokines, MMPs, GFAP, NfL) and diffusion/structural MRI metrics.
Assessed univariate and multivariable associations using AIC-based stepwise selection, bootstrap selection frequencies, penalized regression, and cross-validated R².
Global permeability (Qalb) and regional permeability (Ktrans) did not significantly correlate (Spearman ρ = 0.14, p = 0.28), demonstrating they capture distinct BBB dysfunction dimensions.
Multivariable CSF models explained 53% of variance (R² = 0.53) in Qalb (retaining MMP-2, Flt-1, IL-8, GFAP, and NfL), compared to R² = 0.32 for plasma; for Ktrans, CSF, plasma, and MRI models explained R² of 0.36, 0.39, and 0.14, respectively.
CSF-to-plasma biomarker correlations varied widely from r = -0.03 (MMP-9) to r = 0.74 (NfL) and r = 0.70 (IL-13), indicating plasma biomarkers cannot routinely substitute for CSF measurements.

Cite This Study

Golduzian et al. (2026) studied this question.

synapsesocial.com/papers/6aa7b2e10926e14a848b1799 https://doi.org/10.3389/fragi.2026.1923046

View Full Paper

Ask AI

Bookmark