Aims

Iron is widely observed in the amyloid-beta (Aβ) plaques in Alzheimer’s disease brain tissue. Here, the sensitivity of magnetic resonance imaging (MRI) R₂^* and quantitative susceptibility mapping (QSM) to measure the impact of ferritin-bound iron associating with aggregating Aβ (versus the equivalent concentration of dispersed ferritin), was tested using ultra-high-field (UHF) 9.4 T preclinical MRI.

Methods

In-vitro-formed preparations of Aβ(1-42) aggregated in a phosphate-free KH buffer with/without ferritin were constructed in a custom-designed agarose gel phantom for MRI analysis, and compared with Aβ-only, or dispersed ferritin, in agarose. High-resolution 3D gradient echo MRI was acquired for the phantoms to calculate the R₂^* maps, and to calculate the susceptibility maps via an optimised QSM processing pipeline for the 9.4 T preclinical UHF MRI system.

Results

Linear correlation of R₂^* and quantitative susceptibility on ferritin concentration was observed in the phantoms. The ferritin-dependent coefficient for the R₂^*-ferritin correlation in the aggregated Aβ/ferritin phantom (42.48 s^-1/mg·mL^-1) was 3-fold higher than for dispersed ferritin (14.65 s^-1/mg·mL^-1). The ferritin-dependent coefficient for the susceptibility-ferritin correlation in the aggregated Aβ/ferritin phantom was found to be 147.3 ppb/mg·mL^-1.

Conclusions

This work established a method for quantitative analysis of Aβ-associated ferritin-bound iron using QSM at 9.4 T, and revealed linear relationships between the ferritin-bound iron and the R₂^* and susceptibility measures. Both parameters proved sensitive to the distribution of the iron, with the clustering of ferritin associating with Aβ affecting the R₂^* and the susceptibility to a greater effect than dispersed ferritin.

Fig.1 Graphical abstract.

Aims

To investigate the resting-state functional connectivity (rs-FC) of the subthalamic nucleus (STN), the most frequently used deep brain stimulation (DBS) target for Parkinson’s disease (PD), in different PD phenotypes.

Methods

Clinical data and rs-functional MRI were acquired from 60 PD patients and 60 age- and sex-matched controls. PD patients were divided into two groups: 19 patients eligible for DBS (PD-DBS) and 41 not candidate for DBS (PD-noDBS). A seed-based RS-FC analysis was run between the bilateral STN and the rest of the brain, and compared between groups.

Results

PD-DBS showed a reduced connectivity between bilateral STN and bilateral sensorimotor areas relative to both controls and PD-noDBS patients. On the contrary, PD-DBS patients showed an increased connectivity between bilateral STN and globus pallidus, putamen and thalamus bilaterally compared to healthy controls. Similar pattern were found when PD-noDBS patients were compared to controls (albeit with lower connectivity levels than PD-DBS patients).

Conclusions

Our results suggest that functional connectivity of deep nuclei changes among PD phenotypes and confirm an important role of functional MRI as tool for selection of candidates for DBS. The idea that STN-DBS works by modulating and restoring functional connectivity between basal ganglia and sensorimotor areas is further corroborated.

Supported by. Ministry of Education, Science, and Technological Development of the Republic of Serbia [grant number #175090]; Italian Ministry of Health [grant number # RF-2018-12366746].

Aims

Parkinson’s disease (PD) patients are classified as tremor-dominant (TD) and postural instability and gait disorder (PIGD) phenotypes. The aim of this study was to investigate functional alterations within motor circuits of the cerebro-cerebellar system in PD-TD and PD-PIGD groups using stepwise functional connectivity (SFC) method.

Methods

32 PD-TD and 26 PD-PIGD patients performed clinical/cognitive evaluations and resting-state functional MRI (fMRI). 60 age- and sex-matched controls were also enrolled. SFC analysis aims to characterize regions that connect to specific seed brain areas at different levels of link-step distances. The cerebellar seed-region was identified using motor task in 23 controls. For each of the SFC maps, two-sample t-test comparisons between groups were performed.

Results

The performance of the fMRI-motor task was associated with activation of the lobule VI and vermis of the cerebellum. SFC analysis at one-link step distance showed, in both PD subtypes, a decreased connectivity between seed-region and thalamus and parietal lobe relative to controls; across intermediate link-steps, a reduced connectivity was observed with frontal, parietal and occipital lobes. Only PD-PIGD patients showed lower connectivity at intermediate link-step distances between the seed-cerebellar region and sensorimotor areas. Moreover, SFC pattern identified different localization of functional over‐connectivity in frontal lobe in both PD groups: inferior frontal gyrus and insula in PD-PIGD, and in orbitofrontal gyrus in PD-TD.

Conclusions

These findings highlight subtype-specific PD changes in cerebellar functional connectivity, providing novel insights into the pathophysiological mechanism underlying different motor phenotypes.

Funding: Ministry of Education and Science Republic of Serbia (#175090), Italian Ministry of Health (#RF-2018-12366746).

Aims

This study aimed to clinically validate a new brain age prediction model as a step towards integration in routine clinical practice for the improvement of Alzheimer's disease (AD) diagnosis and related neurodegenerative diseases.

Methods

The study population (REMEMBER, n=746) consisted of cognitively healthy (HC, n=91), subjective cognitive decline (SCD, n=66), mild cognitive impairment (MCI, n=319) and AD dementia (ADD, n=270) subjects (Table 1). Automated brain volumetry of global, cortical and subcortical brain structures computed by icobrain dm (v.4.4) were extracted from T1-weighted MRI sequences that were acquired at participating memory clinics from the Belgian Dementia Council (BeDeCo) and used to predict ‘brain age’ and ‘brain-predicted age difference’ (BPAD) for every subject [Denissen et al, 2021].

Results

There was a weak-to-moderate correlation between MMSE score and both brain age (r=-0.35, p<0.001) and BPAD (r=-0.22, p<0.001)(Figure 1). When comparing BPAD and brain age to chronological age, MCI and ADD patients showed an older than normal brain age estimation (Figure 2), decreasing with age (Figure 3), indicating BPAD and brain age can serve as early biomarkers for AD. Longitudinal data (n=475, excluding ADD) showed that 76% of subjects converting to a more severe stage in the AD continuum (n=116) had a significantly higher brain age than chronological age (p<0.001), with over half (54%) showing a BPAD > 10. No significant correlation between BPAD and conversion time was found.

Conclusions

Brain age and BPAD have the potential to serve as complementary metrics in the understanding of cognitive status and conversion in different stages of the AD continuum.

Aims

Pathology-sensitive MRI biomarkers for predicting and monitoring cognitive decline in Parkinson’s disease (PD) are a well-recognized need. Neurofilament is a marker of axonal degeneration, and associates with cognitive decline in PD. Here, we investigated whether cortical mean diffusivity (MD) is sensitive to cortical axonal degeneration and pathological hallmarks in clinically-defined and pathologically-confirmed PD, PD with dementia (PDD) and non-neurological controls.

Methods

We included 16 PD (7 PD and 9 PDD) and 17 control brain donors with post-mortem in-situ 3DT1 and DTI from which cortical MD was calculated with Freesurfer. After autopsy, PD and Alzheimer’s disease (AD) pathological hallmarks, together with neurofilaments, were analyzed in cortical regions using QuPath. Associations were calculated using linear mixed models.

Results

Phosphorylated-neurofilament medium-chain (p-NfM) and heavy-chain (p-NfH) were increased in both PD (p<.001,+28%) and PDD (p=.001,+23%) compared to controls, and they negatively correlated with amyloid-beta accumulation in PDD (r=-.64,p=.001). In turn, NfL was not increased in PD nor PDD, but positively correlated with pSer129 alpha-synuclein (p-aSyn) (r=.36,p=.006) and p-tau (r=.55,p<.001) in PDD, most significantly in the parahippocampal gyrus (r=.87,p<.001). MD did not correlate with NfL, NfM or NfH, but correlated with p-aSyn (r=-.37,p=.021) in the parahippocampal gyrus in the whole cohort, even though the correlation didn’t survive when excluding the controls (r=-.40,p=.201).

Conclusions

Although axonal degeneration occurs in PD and PDD, which associated with PD and AD pathological hallmarks, MD was only sensitive enough to measure p-aSyn load in the parahippocampal gyrus. Therefore, different MRI biomarkers are needed for cortical axonal degeneration in PD.

Aims

We examine the effect of XPro1595, a next generation tumor necrosis factor (TNF) inhibitor that selectively neutralizes soluble TNF, on white matter metrics of neuroinflammation, axonal density and myelin and compared these to traditional CSF biomarkers in a Phase1b trial in AD patients with biomarkers of inflammation (ADi).

Methods

16 participants were enrolled in an open-label multi-dose trial of XPro1595. They needed one or more biomarkers of inflammation to enroll. MRI was assessed every three months using 3T structural and diffusion MRI. Quantitative tractometry was done using indices of free-water (a proxy of neuroinflammation), apparent fiber density (AFD – a proxy of axon integrity) and tissue radial diffusivity (RDt – a proxy to myelin content) in the AD bundles as composite biomarkers of neuroinflammation, axon regeneration and remyelination respectively.

Results

MRI metrics showed a progressive improvement over 12 months in all white matter metrics; including: i) a 46% reduction in free-water (neuroinflammation), ii) 17% increase in AFD, iii) a 17% reduction in RDt, iv) and an increase in total WM volume and left temporal lobe volume increased. These measures correlated with multiple protein CSF biomarkers of neuroinflammation and neurodegeneration.

Conclusions

In this Phase 1b trial, we showed that XPro1595 improves white matter structure within bundles that are specifically affected in AD. High correlation with gold standard CSF protein neuroinflammatory and neurodegenerative biomarkers validate the use of non-invasive MRI tools to measure WM pathology. Together, these data demonstrate the superiority of WM metrics over traditional CSF analysis of WM biomarkers.

Aims

To investigate the association between the main Alzheimer’s neuropathological hallmarks and a novel set of cortical microstructural Diffusion Tensor Imaging measures.

Methods

Forty-one autopsy confirmed cases [7=cognitively normal, 10= Mild cognitive Impairment, 24=Alzheimer’s Disease (AD)] from the National Alzheimer’s Coordinating Center (NACC) database were included. (The NACC database is funded by NIA/NIH Grant U24AG072122; with NACC data contributed by the NIA-funded ADRCs). In-vivo T1 structural and diffusion tensor imaging (DTI) scans were analysed to calculate 3 novel cortical grey matter diffusion measures (AngleR, PerpPD and ParlPD) [McKavanagh et al., 2019;] and mean diffusivity (MD). For neuropathological assessment of dementia cases, following NIA recommendations, the Thal phase, CERAD neuritic plaque assessment, Braak NFT and cerebral amyloid angiopathy scores were used as classical neuropathological hallmarks to investigate the relationship with novel cortical diffusivity measures.

The relationships between cortical diffusivity and severity of white matter rarefaction, atherosclerosis of the circle of Willis, and locus coeruleus hypopigmentation were also investigated. Statistical significance included false discovery rate correction (FDR).

Results

Novel cortical diffusivity measures were significantly associated with Thal phase, Braak NFT and locus coeruleus hypopigmentation scores (Figure 1) (standard MD was not significant).

Conclusions

These findings suggest that cortical diffusivity measures are sensitive to AD neuropathological changes in the microstructure of cortical grey matter. In vivo markers of neurodegeneration (N) that are sensitive to the range of histopathological features of amyloid, tau and monoamine pathology in Alzheimer’s disease offer a useful complement to existing non-invasive markers of amyloid (A) and tau (T).

Aims

How intrinsic properties of the brain, such as anatomical connectivity and gene expression combine to determine the pattern of disease progression in de novo Parkinson’s disease (PD) remains unknown. Here we used MRI data from PPMI to map the atrophy progression over four years and related it to structural and functional connectivity and gene ontology (GO) enrichment analysis.

Methods

We derived atrophy maps for four time points using deformation-based morphometry (DBM) applied to T1-weighted MRI from 74 de novo PD patients (50M) and 157 HC (115M). W-score maps were computed from the DBM maps to account for age and sex using the HC group as a reference. Atrophy differences between time points were compared with rANOVA and pTFCE approach. We next investigated if the regional atrophy progression was correlated with the atrophy progression of its structurally and functionally connected neighborhood using networks from healthy adults. We also verify if the atrophy progression was associated with specific biological processes using the GOrilla and PANTHER platforms.

Results

We found that atrophy significantly progressed over two and four years in the caudate, nucleus accumbens, hippocampus, and posterior cortical regions. This progression was shaped by both structural and functional brain connectivity and was more pronounced in regions with a higher expression of genes related to synaptic activity.

Conclusions

In addition to identifying vulnerable brain regions, we demonstrate that the progression of atrophy in PD is in line with the prion-like propagation hypothesis of alpha-synuclein along the brain connectome and the synaptic spreading hypothesis.