Aims

Brain glucose hypometabolism, indexed by [¹⁸F]FDG-PET, is considered a biomarker of neurodegeneration in Alzheimer’s disease (AD). However, other brain cells, such as astrocytes and microglia, also consume considerable amounts of glucose and may contribute to the [¹⁸F]FDG-PET signal in the brain. Thus, the cellular source of the [¹⁸F]FDG-PET signal remains controversial. ​​Here, we aimed to evaluate whether canonical markers of different brain cell types (neuron, oligodendrocyte, astrocyte, and microglia) associate with brain [¹⁸F]FDG-PET signal in humans and a rat model of human amyloidosis.

Methods

[¹⁸F]FDG-PET images were acquired in ten-month-old APP/PS1 (TgF344-AD,n=8) and wild-type (WT,n=6) rats. Next, we evaluated the gene expression of NeuN, Olig2, GFAP, S100B, and IBA1 in the frontal and temporoparietal cortices and cerebellum. Association maps integrating mRNA with brain [¹⁸F]FDG-PET signal were conducted at the voxel level using RMINC. [¹⁸F]FDG-PET and brain tissue mRNA (GFAP, ACSA2, S100B, and GLAST) of cognitively unimpaired individuals were obtained from ADNI and Allen Brain Atlas database, respectively. In humans, mean regional [¹⁸F]FDG-PET SUVr (reference region: pons) and mRNA expression were used to evaluate their association using a Gaussian model. Differences were considered statistically significant at p<0.05 (t>2).

Results

GFAP and S100B mRNA levels in the temporoparietal and frontal cortices positively correlated with brain [¹⁸F]FDG-PET signal (local maxima, t_GFAP=9.4, t_S100B=6.6;Fig1B-C). No positive associations were found with IBA1, Olig2 and NeuN (t<2;Fig1D-F). We expanded our analysis by comparing gene astrocyte markers and astrocyte gene cell markers in humans. At the regional level, we found that FDG-PET SUVr and astrocyte mRNA were positively associated: S100B (R-squared=0.34), GFAP (R-squared=0.33), GLAST (R-squared=0.31) and ACSA-2 (R-squared=0.31; Fig1G).

Conclusions

​​Our findings suggest astrocyte mRNA markers are more closely associated with brain [¹⁸F]FDG-PET signal than neuronal, oligodendrocyte, and microglial mRNA markers.

Aims

Investigations into the amyloid, tau, and neurodegeneration (AT(N)) framework, which is used to stage Alzheimer disease (AD), typically consist of clinic-based non-Hispanic White (NHW) populations. The present study sought to cross-sectionally characterize amyloid PET (A), tau PET (T), AD cortical thickness signature (N) and white matter hyperintensities (V) and determine the relative importance of each marker in predicting cognitive status in a large ethnically and racially diverse sample.

Methods

Data were obtained from the community-based Health and Aging Brain Study – Health Disparities (HABS-HD) and included Mexican Americans (MAs, n=231), African Americans (AAs, n=602), and NHWs (n=272). Group differences in AT(N)-V biomarkers were assessed using analysis of covariance. Age, sex, cognitive status (Clinical Dementia Rating Scale [CDR]), and years of education were included as covariates. Dominance analysis determined the relative ‘importance’ of each biomarker in predicting global cognitive status in the entire cohort and separately for each group.

Results

Tau was significantly greater in MAs compared to AAs (p<0.05) and NHWs (p<0.001), and in AAs compared to NHWs (p<0.001). AAs also had greater neurodegeneration compared to NHWs (p<0.001) and MAs (p<0.05). No other significant differences were observed (ps>0.05). For the entire cohort, V was the best biomarker for predicting cognitive functioning (greater R² values; see Figure 1). In MAs, V was the most important while in AAs, A was the most important. In NHWs, T and V were of equal importance.

Conclusions

These results aid in the evaluation of the newly proposed AT(N) criteria and demonstrate that the magnitude and overall contributions of AT(N)-V imaging biomarkers for cognitive outcomes varies among ethno-racial groups. Possible differential effects of AD biomarkers should be considered in research, clinical trials, and response to treatment.

Aims

We evaluated for possible racial differences in baseline and longitudinal plasma Aβ42/40 between self-reported Black and White participants from a multi-center observational study.

Methods

Longitudinal plasma samples collected at three AD Research Centers (Washington University, University of Pennsylvania, and University of Alabama-Birmingham) from a total of 193 Black and 753 White participants were centrally processed using C₂N Diagnostics’ PrecivityAD™ blood test for Aβ42, Aβ40, and Aβ42/40. General linear mixed effects models were used to estimate the mean baseline levels of amyloid biomarkers as well as their annual rates of change for each racialized group and to compare these measures between racial groups.

Results

Participants were 43 to 91 years at baseline, and longitudinally followed for a mean of 6.38 years. Among 193 Black participants, 160 were CDR 0 and 33 had a CDR>0 at baseline. Among 753 White participants, 629 were CDR 0 and 124 had a CDR>0 at baseline. Regardless of the CDR at baseline, Black participants had a higher level of Aβ42/40, compared to White participants, especially in individuals with baseline ages between 60 and 75 years. Black participants had a significantly lower rate of plasma Aβ42/40 positivity at baseline, consistent with brain amyloid. Longitudinally, Black participants had a faster rate of increase in both Aβ42 and Aβ40 than White participants, but no differences were found in the annual rate of change in Aβ42/40.

Conclusions

Black individuals participating in research studies may have a lower rate of amyloid positivity at baseline, which could decrease the proportion of Black individuals eligible for AD prevention trials and treatment trials for AD dementia. However, there were no significant racial differences in the longitudinal rates of change for plasma Aβ42/40.

Aims

Impaired kidney function has a potential confounding effect on peripheral Alzheimer’s Disease (AD) biomarker levels. This is potentially important when certain blood biomarkers will be employed in the routine clinical assessment of suspected AD to initiate treatment. This study aimed to assess the concentrations of plasma AD biomarkers by stratifying individuals by renal function by using estimated glomerular filtration rate (eGFR) values.

Methods

We included 228 amyloid-PET negative participants from the Translational Biomarkers in Aging and Dementia (TRIAD) cohort (Young: 35; CU-: 134; MCI-: 25; MCI+: 2; Non-AD: 32). TRIAD confirmed amyloid negativity with [18F]AZD-4694 SUVR >1.5. Plasma samples measured Aβ 42/40, GFAP, NfL, p-tau217, and p-tau231 using Simoa technology at the University of Gothenburg, Sweden. Plasma creatinine was assessed on the Cobas c501 module. Glomerular barrier function was evaluated using eGFR calculation (mL/min/1.73 m2) based on creatinine, age, and gender, dividing participants into high eGFR (CKD-stage 1, eGFR ≥90, group A: 157), medium eGFR (CKD-stage 2, eGFR 60-89, group B: 63), and low eGFR (CKD-stage ≥3, eGFR <60, group C: 8) groups.

Results

Plasma biomarker concentrations were highest in CKD-stage 3 (Aβ42:9.78±2.04 pg/mL, Aβ40:146.2±33.45 pg/mL, GFAP:242.2±129.3 pg/mL, NfL: 44.36±24.62, p-tau217:0.858±0.981, p-tau231:24.91±21.11).They were also elevated in CKD-stage 2(Aβ42:7.39±2.01 pg/mL,Aβ40:100.3±22.69 pg/mL, GFAP:171.9±62.04 pg/mL, NfL:27.18±14.54, p-tau217:0.351±0.322, p-tau231:14.75±5.80) compared to CKD-stage 1 (Aβ42:6.88±1.91 pg/mL, Aβ40:87.02±22.44 pg/mL, GFAP:128.4±81.30 pg/mL, NfL:18.42±15.03, p-tau217:0.270±0.351, p-tau231:14.04±9.31).Additionally, Aβ42/β40 and Aβ42/p-tau217 ratios were highest in CKD-stage 1 (0.08±0.01 and 41.50±49.50) compared to CKD-stage 2 (0.07±0.01 and 30.97±19.49) and CKD-stage 3 (0.06±0.01 and 18.23±10.11).

Conclusions

This study demonstrates the effect of kidney impairment by means of eGFR on blood-based biomarkers, which is independent of AD pathology. Kidney impairment should be considered and ruled out when interpreting blood biomarker levels for possible AD pathology.

Aims

To compare cerebrospinal fluid (CSF) concentrations of amyloid peptides, BACE1, neurogranin (Ng), tau proteins between amyloid (A) and tau (T) proteinopathy Alzheimer's disease (AD) profiles.

Methods

Population corresponded to 425 AD and MCI participants of the BALTAZAR cohort (NCT01315639). CSF biomarkers Tau, pTau181, Aβ40 and Aβ42 were determined with routine assays, sAPPβ, Aβ38 with multiplex kits, BACE1 and Ng with simplex immunoassays. Concentrations were compared between A-T-, A+T-, A-T+, and A+T+ profiles (based on CSF Aβ42/Aβ40 and pTau181 values).

Results

CSF BACE1 and Ng remained stable in A+T- but increased significantly in A+T+, as well as in A-T+. Aβ peptides showed a decrease in A+T-. The non-amyloidogenic forms Aβ38 and Aβ40 showed a significant increase (>30%) in A+T+, as well as in A-T+. sAPPβ variations were similar to those of BACE1. These results have been confirmed in independent cohorts (including ADNI). Conversion of MCI to dementia was low in A-T- and A-T+, and increased in A+T- and A+T+. CSF BACE1, Ng and Ng/BACE1 higher values were also associated increased conversion. ApoE4 was associated with A+T- and even more with A+T+, but not with A-T+.

Conclusions

Isolated amyloidopathy (A+T-) has a limited pathological impact and is associated with lower concentrations of Aß peptides in the CSF, probably due to aggregation phenomena. A+T+ profile corresponded to AD with synaptic dysfunction, neurodegeneration and cognitive decline, but also with higher concentrations of non-amyloidogenic Aß peptides in CSF. This surprising increase that could be linked to BACE1, has also been observed in isolated tau proteinopathy (A-T+) (reminiscent of Primary Age-Related Tauopathy). In conclusion, we believe that amyloid and tau proteinopathies are two synergistic pathogenic phenomena: amyloidopathy induces phosphorylation of tau protein, while tau proteinopathy increases amyloid peptide production.

Aims

Investigate the role of neuroinflammation in the progression of Alzheimer's disease (AD) by identifying immune-related proteins in cerebrospinal fluid (CSF) and plasma and their association to imaging biomarkers of neuroinflammation.

Methods

Participants from the TRIAD cohort incorporating within the AD spectrum, and with available TSPO PET data (positivity = 3 SD > mean ROI-SUVR of young participants), had cerebrospinal fluid (CSF) (n=100) and plasma (n=169) samples analyzed using Proximity Extension Assay (PEA) technology targeting 368 inflammation-related proteins (Olink). LIMMA models evaluated the differential protein expression according to TSPO PET status adjusting for age and amyloid load (PET). Associations were FDR corrected for multiple comparisons. Gene ontology (GO) was used for enrichment analysis on relevant results. Spearman rank was used on a cross-correlation matrix that included the top 10 results from the LIMMA analysis.

Results

Preliminary results identified several proteins that are in higher concentrations in the CSF of TSPO positive in comparison with TSPO negative participants (Figure 1). The functional enrichment analysis unveiled terms related with “positive regulation of MAPK cascade”, “cytokine-mediated signaling pathway”, “leukocyte migration” and “positive regulation of ERK1 and ERK2 cascade” (Figure 2). Cross-correlation matrix (Figure 3) showed that these proteins are in high correlation with each other, with TSPO PET and in low degree of correlation with amyloid and tau PET. Furthermore, the expression of these proteins increased in a stepwise manner from lower to higher TSPO PET binding (figure 4).

Figure 1

Figure 2

Figure 3

Figure 4

Conclusions

More detailed analysis will be conducted both in CSF and in plasma to further investigate the promising targets, their potential as biomarkers and their coverage between amyloid and tau deposition.

Aims

pTau217 is one of the most promising blood-based biomarkers of Alzheimer's disease (AD). We compared the performance of two pTau217 immunoassays and a mass spectrometry (MS) assay for prediction of abnormal amyloid-PET.

Methods

Plasma samples from 109 cognitively unimpaired (CU), 85 MCI, and 33 AD dementia individuals were evaluated. Plasma pTau217 concentrations were measured using ALZpath Simoa pTau217 and Fujirebio Lumipulse G pTau217(N-Terminal) (ADx research prototype) immunoassays, and a pTau217 ratio MS assay (C₂N Diagnostics). Amyloid-PET was performed with Pittsburg ¹¹C-Compound B; abnormal amyloid (A+) was defined as Centiloid ≥25 (SUVR≥1.52) (n=137 A+, 90 A−). Area under the ROC curve (AUC) was calculated using amyloid-PET status or a combination of amyloid-PET status and clinical diagnosis as the outcome and plasma pTau217 as the predictor.

Results

All assays were correlated with amyloid-PET (Spearman rho of 0.68 ALZpath, 0.60 Lumipulse, and 0.69 C₂N, P < 0.001). Fold changes in the median concentration of A+ versus A− groups were 2.6 for ALZpath, 2.5 for Lumipulse, and 3.4 for C₂N. For prediction of A+ vs. A− among the MCI/dementia group, the AUC was 0.87 (95% CI 0.80–0.94) for ALZpath, 0.90 (0.84–0.96) for Lumipulse, and 0.90 (0.84–0.95) for C₂N. In the CU group, the AUC for prediction of A+ vs. A- was 0.81 (0.73-0.89) for ALZpath, 0.75 (0.66-0.85) for Lumipulse, and 0.77 (0.68-0.86) for C₂N.

Conclusions

In our study, the pTau217 immunoassays showed correlation with amyloid-PET and diagnostic accuracy for prediction of amyloid pathology in the MCI/dementia group similar to the C₂N pTau217 ratio. Upon further validation on a larger sample cohort, performance of these immunoassays may demonstrate suitability for screening and/or diagnosis of Alzheimer's disease, when used alongside clinical assessment.

Aims

To investigate the association of a 6-level biological system with clinical staging for Alzheimer's Disease (AD). Specifically, we aim to assess the potential impact of plasma GFAP and P-tau217 for predicting disease progression and compare this 6-level classification to the NIA-AA Revised Clinical Criteria for AD (draft, July-2023) 4-level biological staging.

Methods

From the TRIAD database, amyloid-beta positive (AB-PET) participants were designated as part of the AD continuum. These individuals were then classified based on two systems: one involving 6 biological and clinical independent stages and one based on the 2023 NIA criteria (Table1). We determined cut-off points for plasma biomarkers as described in Table1 to achieve better sensitivity. The classification was sequential: Tau Braak stages (18F-MK-6240 TAU-PET) took precedence over p-tau217, and p-tau217 over GFAP. Statistical analysis included linear models for cross-sectional results and logistic regression for clinical progression (CDR increase at follow-up).

Results

We classified 162 individuals across the AD-continuum. The distribution of clinical stages across biological stages in 6-level (6L) and NIA-AA (4L) systems are represented in Figure1. Notably, the progressive distribution suggests the higher clinical burden of sequential AD pathology as measured by the biological stages. Prediction models revealed that, even after adjusting for age, sex, APOE, and years of education, biological stages strongly predicted clinical stages in both 4L and 6L systems (p<0.001). In contrast, the two additional stages in 6L-system showed clear difference in cross-sectional and longitudinal (CDR, p<0.01) cognitive changes.

Conclusions

The addition of fluid biomarkers GFAP and P-tau217 as intermediate biological stages enhances prediction model of clinical staging and disease progression. This system could be used for population enrichment in clinical trials. More studies are needed to assess its clinical utility.