Aims

The ratio of phosphorylated tau (p-tau) over Aβ₄₂-amyloid in CSF demonstrate superior performance over single biomarkers for predicting Aβ-PET status using automated immunoassay platforms. Recent studies show clearly that plasma Aβ₄₂, and p-tau, can also predict Aβ-PET status, although only one study has explored the utility of the plasma p-tau181/Aβ₄₂ ratio in predicting Aβ-PET burden. The current study investigated the ability of plasma p-tau181, Aβ₄₀ and Aβ₄₂ to predict amyloid-PET status.

Methods

This study used p-tau181, p-tau231, Aβ₄₂, and Aβ₄₀ levels using prototype Simoa assays to measure plasma samples from the Australian Imaging, Biomarkers and Lifestyle (AIBL) study. PET, clinical, and demographic variables were also drawn from the AIBL study.

Results

The p-tau181/Aβ₄₂ ratio showed the best prediction of Aβ-PET in all participants (AUC =0.905, 95%CI: 0.86-0.95) and in cognitively unimpaired (AUC=0.873; 0.80–0.94), and cognitively impaired (AUC=0.908; 0.82–1) adults. Plasma p-tau181 was associated moderately with CSF-p-tau181 (Elecsys®, Spearmans ρ=0.53, P<0.001). Use of the p-tau181/Aβ₄₂ ratio improved this association significantly (Spearman’s ρ=0.74, P<0.0001). Plasma p-tau181 predicted abnormal CSF-Aβ levels with an AUC of 0.8 (0.73-0.88), which was also improved by use of the plasma p-tau181/Aβ₄₂ ratio (AUC=0.816, 0.74-0.89). The p-tau181/Aβ₄₂ ratio also predicted future rates of decline of cognition using the AIBL PACC and CDR-SoB (P<0.0001).

Conclusions

This high performing plasma p-tau181/Aβ₄₂ ratio has excellent potential for development as both a screening and prognostic assay for preclinical Alzheimer’s disease (AD). Its utility for specific diagnosis in prodromal and clinical AD remains to be determined.

Aims

Accepted biomarkers related to CSF Aβ and pTau lack insight for novel therapeutic strategies related to up- and down-stream biological pathways and processes. We have demonstrated the power of unbiased hybrid tissue/fluid proteomics to deliver biomarkers relevant to a novel mechanism of action that can guide decisions in later phases of clinical trials.

Methods

CSF (0.5 ml) from patients before and after 12 weeks treatment with Xpro at three different doses was analyzed using TMTcalibrator™ (Figure 1) where inclusion of high levels of TMTpro™-labelled AD brain cortex lysate with similarly labelled digests of CSF is used to boost sensitivity for disease-relevant peptides during mass spectrometry analysis. Off-line phosphopeptide enrichment and fractionation was combined with tandem mass spectrometry and sophisticated bioinformatics to identify biomarker sets relating to different aspects of disease modulated by Xpro treatment.

Results

Overall, 37,018 peptides associated with 4,966 protein groups and 5,202 individual phosphorylation sites were quantified. Functional analysis showed that treatment-responsive CSF proteins from a wide range of disease-relevant biological processes and pathways related to neuronal organization, synaptic function, immune and inflammatory response, cytoskeleton, metabolism, Rho GTPases signalling, and myelination. Notably, 12 weeks high-dose treatment led to decreased expression of several emerging neuronal markers - NFL↓84%; Tau pT217↓46%; Neurogranin ↓56% and increased for myelin regulatory factor ↑25%. We will show how these and other results are guiding design of Phase II studies during the presentation.

Conclusions

Unbiased proteomics revealed both known and unexpected effects of Xpro through measurement of CSF proteins and these panels will guide our future studies.

Aims

Recently, the p53 protein has gained attention for its role in the early evolution of Alzheimer’s disease (AD). p53 is regulated by post translational modifications (PTMs) which affect its conformation and function in several processes in AD. We have investigated the PTMs occurring in U-p53^AZ in patients suffering by different forms of dementia and at different stage of cognitive decline to show the specificity of U-p53^AZ in AD.

Methods

U-p53^AZ isolated from 48 patients (55 plasma samples) from AIBL biobank (70y av.) was analysed by AlzoSure®Confirm method to detect its PTMs. U-p53^AZ captured by 2D3A8Ab from plasma samples fwas sequenced by mass spectrometry technology (Orbitrap XL). The study cohort included 16AD, 9 Cognitive Normal (CN), 6 Mild Cognitive Impaired (MCI), 7 asymptomatic AD, 8 prodromal AD, 6 Frontotemporal Dementia (FTD), 1 Vascular Dementia and 1 Lewy Body Dementia patients.

Results

11 U-p53^AZ-PTMs of (acetylation, phosphorylation and ubiquitination in different aa-residues) were identified in total and each clinical group showed the same combination of PTMs: all AD patients showed the same PTMs pattern plus the truncation of the protein in the N-terminus region, whilst FTD showed only 3 PTMs and any truncation. Interestingly, asymptomatic and prodromal AD individuals showed a distinct PTMs pattern than CN and MCI.

Conclusions

U-p53^AZ undergoes to different biochemical pathways impacting on its linear sequence specifically linked to dementia pathogenesis. U-p53^AZ-PTMs detected by AlzoSure® Confirm represents a simple, non-invasive tool to discriminate AD from other dementia and to identify the stage of cognitive decline.

Aims

Blood has been widely investigated to discover biomarkers and gain insights into the biology of aging and age-related diseases. Its protein composition provides information about complex biological processes, as proteins are often direct regulators of cellular pathways. Using recent methodological developments allowing the measurement of thousands of proteins with very high sensitivity and specificity, we sought to understand comprehensive proteomic changes in two AD clinical trials.

Methods

Phase 2 clinical trials (GRF6019-201 n=40 and GRF6019-202 n=26) testing the safety, tolerability, and feasibility of repeated infusions of the plasma fraction GRF6019 in Alzheimer's disease (AD) were used as the source to measure more than 7000 proteins in plasma using the SOMAscan and Olink assays. To evaluate the relevance of the proteomics changes induced by GRF6019, we compared these changes to those observed in a healthy aging cohort (~5000 proteins measured in 370 subjects).

Results

Standard statistical analysis at the protein levels lacked power due to the small sample size in phase 2 clinical trials. By analyzing trajectories of groups of proteins, clinical proteomics revealed multiple clusters of proteins responding to GRF6019. Remarkably, several pathways modulated by GRF6019 were particularly relevant for the biology of aging and AD – including the complement/coagulation cascades and neuronal pathways (q<0.05).

Conclusions

Altogether, our results suggest that the treatment of AD patients with a complex plasma fraction modulates biological pathways that are relevant to aging and AD. Our results establish deep proteomics as a powerful tool to study human response to treatment in clinical trials.

Aims

To identify cerebrospinal fluid (CSF) proteins associated with Alzheimer’s disease progression along with the clinical disease staging.

Methods

We measured the levels of 184 proteins in CSF samples from 556 subjective cognitive decline and mild cognitive impairment patients from three independent memory clinic longitudinal studies (Spanish ACE, n=410; German DCN, n=93; German Mannheim, n=53). We evaluated the association between protein levels and clinical stage, and the effect of protein levels on the progression from mild cognitive impairment to dementia of the Alzheimer’s type (DAT).

Results

Mild cognitive impairment subjects with increased CSF level of matrix metalloproteinase 10 showed a higher probability of progressing to DAT and a faster cognitive decline. CSF matrix metalloproteinase 10 increased the prediction accuracy of CSF Aβ₄₂, P-tau¹⁸¹, and T-tau for conversion to DAT. Including matrix metalloproteinase 10 to the [A/T/(N)] scheme improved considerably the prognostic value in mild cognitive impairment patients with abnormal Aβ₄₂, but normal P-tau¹⁸¹ and T-tau, and in mild cognitive impairment patients with abnormal Aβ₄₂, P-tau¹⁸¹, and T-tau. Matrix metalloproteinase 10 was correlated with age in subjects with normal Aβ₄₂, P-tau¹⁸¹, and T-tau levels.

Conclusions

Our findings support the use of CSF matrix metalloproteinase 10 as a prognostic marker for DAT and its inclusion to the [A/T/(N)] scheme to incorporate pathologic aspects beyond amyloid and tau. CSF level of matrix metalloproteinase 10 may reflect ageing and neuroinflammation.

Aims

A growing body of evidence has revealed substantial white matter (WM) degeneration in Alzheimer’s disease (AD). Nonetheless, robust characterization of WM changes has been challenging due to methodological limitations. Here, we applied a novel fixel-based analyses (FBA) framework to assess WM damage and its association with the underlying molecular pathology in the AD continuum.

Methods

304 amyloid-negative and 92 amyloid-positive cognitively unimpaired (CU) controls, 80 amyloid-positive individuals with mild cognitive impairment (MCI) and 88 patients with AD dementia (aged ≥ 50 years) were recruited from the Swedish BioFinder-2 study. Participants underwent diffusion-weighted MRI, tau and amyloid PET using [18F] RO948 and [18F] flutemetamol PET, respectively. Using FBA, we studied microscopic differences in fiber density (FD), macroscopic changes in fiber cross-section (FC) and the combination of both (FDC) across all WM fiber bundles.

Results

All FB metrics were decreased in cognitively impaired (CI) i.e., MCI and AD patients compared to both amyloid-negative and amyloid-positive CU individuals (Figure 1). The strongest reduction was observed in FC, demonstrating macroscopic morphological damage in several WM tracts. No between-group differences were found when comparing the amyloid-negative and amyloid-positive CU groups. Similarly, neither amyloid nor tau deposition were associated with FB metrics in CU individuals. However, in participants along the AD spectrum i.e., those with amyloid pathology widespread negative associations were found exclusively between FC and tau uptake (Figure 2).

Conclusions

These results indicate that WM damage is tightly linked to tau tangles rather than amyloid plaques in the AD continuum. This tau-induced degeneration only affects WM tracts macroscopically.

Aims

Methods

Results

Conclusions