Effects on cognition of regional astrogliosis assessed by 18F-SMBT-1 PET in non- demented individuals

Technical Summary: Effects on Cognition of Regional Astrogliosis Assessed by 18F-SMBT-1 PET in Non-Demented Individuals

Problem Statement
Alzheimer's Disease (AD) is characterized by the accumulation of amyloid-beta (Aβ) and hyperphosphorylated tau, yet these proteins alone do not fully explain the clinical heterogeneity observed in cognitive decline. Neuroinflammation, specifically astrogliosis (the activation of astrocytes), is a critical factor in AD pathology. While plasma glial fibrillary acidic protein (GFAP) has been linked to worse cognition, it lacks spatial resolution. Furthermore, the mechanisms by which astrogliosis influences cognitive decline, particularly the role of neuroprotective growth factors (e.g., BDNF, IGFR1) in mediating this relationship, remain unclear. Existing MAO-B tracers for imaging astrogliosis have limitations in kinetics and sensitivity. This study addresses the need to evaluate regional astrogliosis using the novel tracer 18F-SMBT-1 in non-demented individuals and to explore whether peripheral inflammatory markers and growth factors mediate the relationship between astrogliosis and cognitive performance.

Methodology
The study utilized a cross-sectional analysis of 94 non-demented participants from the INROAADS (INflammation ROles in Aging and Alzheimer's disease Study) cohort.

• Participants: Mean age 68±7 years; 52% women; 89% White. Participants were stratified by Aβ status into A- (n=76) and A+ (n=18) groups based on 11C-PiB PET imaging.
• Neuroimaging:
  • Astrogliosis: Assessed via 18F-SMBT-1 PET (summed 70–90 min post-injection), a tracer targeting overexpressed Monoamine Oxidase-B (MAO-B) in reactive astrocytes.
  • Aβ Status: Assessed via 11C-PiB PET (summed 50–70 min post-injection), converted to Centiloid (CL) units.
  • Structural: 3.0T MRI (MPRAGE and FLAIR) used for parcellation and registration.
• Biomarkers: Plasma samples were analyzed using the NULISAseq CNS kit to quantify GFAP, peripheral inflammatory markers (IL-6, CRP, TNF, IL-4, IL-10, IL-13), and growth factors (BDNF, VGF, VEGFA, VEGFD, IGF1R, IGFBP7).
• Cognitive Assessment: Evaluated using the Montreal Cognitive Assessment (MoCA) for overall cognition, and specific tests for verbal memory (Word List Delayed Recall), visual memory (Benson Complex Figure Delayed Recall), attention (Digit Span Forward), and executive function (Letter Fluency and Clock Drawing Test).
• Statistical Analysis: Robust linear regression models adjusted for age, sex, and education were used to assess relationships between astrogliosis, inflammation, growth factors, and cognition. Causal mediation analyses were performed to test if BDNF and IGFR1 mediated the association between astrogliosis and visual memory, stratified by Aβ status.

Key Contributions

1. Application of 18F-SMBT-1: This study applies the reversible MAO-B tracer 18F-SMBT-1 to assess regional astrogliosis in a non-demented population, demonstrating its utility in detecting early pathological changes.
2. Stratification by Aβ Status: The analysis explicitly differentiates between Aβ-negative and Aβ-positive non-demented individuals, revealing that the impact of astrogliosis on cognition is contingent upon the presence of amyloid pathology.
3. Mediation Analysis: The study investigates the specific role of growth factors (BDNF and IGFR1) as mediators between astrogliosis and cognitive decline, identifying a suppressive mediation effect by IGFR1.

Results

• Astrogliosis and Aβ Status: 18F-SMBT-1 binding in the precuneus was significantly higher in A+ individuals compared to A- individuals. No other regional differences were found between the groups.
• Astrogliosis and Cognition:
  • Overall Cognition: No significant association was found between 18F-SMBT-1 binding and MoCA scores in either A- or A+ groups.
  • Domain-Specific Cognition (A+ Group Only): Among A+ individuals, higher 18F-SMBT-1 binding in multiple regions (including precuneus, posterior cingulate, inferior/superior parietal, and DLPFC) was significantly associated with worse verbal memory, visual memory, and executive function.
  • A- Group: No significant associations were found between astrogliosis and any cognitive domain.
• Peripheral Markers:
  • In A+ individuals, higher IL-6 was associated with worse overall cognition and executive function, while higher IL-10 was associated with better cognition.
  • In A- individuals, higher IL-4 was associated with better visual memory, and higher IGFBP7 was associated with better executive function.
• Mediation Analysis:
  • BDNF: Did not significantly mediate the relationship between astrogliosis and visual memory.
  • IGFR1: Showed a significant suppressive mediation effect on the association between 18F-SMBT-1 astrogliosis and worse visual memory, but only in A+ individuals. IGFR1 accounted for more than 50% of the suppression of this negative effect. This effect was not observed when astrogliosis was measured via plasma GFAP.

Significance and Claims
The authors conclude that in a non-demented population with low levels of AD pathology, regional astrogliosis (measured by 18F-SMBT-1) is associated with worse cognitive performance, but this relationship is restricted to individuals who are Aβ-positive. This suggests that astrogliosis may be a relevant early mechanism in the progression of AD specifically in the presence of amyloid pathology.

Furthermore, the study claims that IGFR1 plays a neuroprotective, suppressive role, mitigating the negative impact of Aβ-related astrogliosis on visual memory. The authors posit that this finding highlights a potential homeostatic response where IGFR1 signaling may counteract the memory impairment caused by reactive astrocytes overexpressing MAO-B. The authors note that 18F-SMBT-1 appears to be a more specific indicator of AD-relevant astrogliosis than plasma GFAP, as the mediation effect was only observed with the PET tracer.

The paper modestly frames these findings as a baseline assessment of a small cohort, acknowledging that the low prevalence of pathology and the cross-sectional nature of the data limit the ability to draw definitive causal conclusions regarding the long-term evolution of these mechanisms. The authors emphasize the need for future longitudinal studies to confirm the role of IGFR1 and the trajectory of astrogliosis in the AD spectrum.