Aims

[¹⁸F]AV133 PET is a biomarker of nigral-striatal dopaminergic function that putatively measures reduced vesicular monoamine transporter (VMAT-2) density in progressing Parkinson’s disease. This study aimed to identify the optimal target and reference regions for quantifying [¹⁸F]AV133 so as to maximise power in future clinical imaging trials of novel therapies.

Methods

The Parkinson Progression Marker Initiative (PPMI) is a multicentre, international observational clinical study to assess PD biomarkers, including [¹⁸F]AV133 PET and Ioflupane (DaT). Longitudinal [¹⁸F]AV133 PET imaging data was available for 38 subjects, each of whom were followed for up to 4 years. Emission data were acquired for 10 mins between 80--90 post-injection of the tracer. Each subject had an associated T1 MRI which was used for inter-subject registration into stereotaxic space (MNI152). Data were quantified using an SUVR approach – permutations of five reference and six target regions were evaluated to determine maximal power of longitudinal signal change (S:N) (Fig1). [¹⁸F]AV133 data were also compared with the DaT data obtained from the same subjects.

Results

[¹⁸F]AV133 quantification using the cerebral white matter as the reference region and putamen as the target region showed the highest power (S:N = 1.1). [¹⁸F]AV133 showed a 75% and 71% increase in statistical power in year 1 and year 2 follow up when compared with DaT leading to a ~46% reduction in sample size required to detect the same pharmacodynamic changes (Fig2).

Conclusions

Cerebral white matter and putamen were selected as the optimal reference and target regions for [¹⁸F]AV133 quantification. With the limitation of small sample size, [¹⁸F]AV133 provides increased sensitivity to longitudinal changes in PD over existing DaT imaging, increasing power to detect pharmacodynamic responses and likely reduced sample sizes in future clinical imaging trials.

Aims

Frontotemporal dementia (FTD) is in the majority of cases neuropathologically characterized by accumulation of either tau or TDP-43. PET tracers targeted at the protein tau could thus potentially predict underlying protein pathology. We examined the PET tracer [¹⁸F]RO948 retention in FTD, aiming to include cases with predictable underlying protein pathology.

Methods

We included 35 patients with FTD: 21 behavioural variant FTD (bvFTD) cases, 11 symptomatic C9orf72 mutations carriers, one patient with non-genetic bvFTD-amyotrophic lateral sclerosis (ALS), one individual with bvFTD due to a GRN mutation and one individual due to a MAPT mutation (R406W). All underwent [¹⁸F]RO948 PET and MRI. Two patients also underwent postmortem neuropathological examination. PET tracer retention was examined using a region-of-interest (ROI) and voxel-wise approaches. Comparison subjects were cases of Alzheimer´s disease (AD, n =13) and Aβ-negative cognitively unimpaired individuals (n =13). Tracer binding was also assessed using [³H]RO948autoradiography in six cases not from the present cohort.

Results

[¹⁸F]RO948 retention across ROIs was comparable to that in Aβ-negative cognitively unimpaired individuals and clearly lower than in AD. Only minor loci of tracer retention were seen in FTD. Autoradiography did not show any [³H]RO948 binding. AD-like retention levels and specific in-vitro binding was however seen in the R406W MAPT mutation carriers.

Conclusions

[¹⁸F]RO948 uptake is not significantly increased in FTD patients and thus cannot predict underlying protein pathology in the majority of cases, with the clear exception being R406W MAPT mutation carriers. [¹⁸F]RO948 has however a promising specificity in differentiating AD from FTD.

Aims

CSF production and drainage are vital brain functions with deficiencies contributing to amyloid lesions in murine models of aging and Alzheimer’s disease (AD). We developed a human PET based technology to measure CSF drainage (de Leon et al 2017). Using [18F]-tau tracers we replicated our findings of reduced ventricular CSF drainage in AD and the association with [11C]-PiB PET amyloid accumulation (Li et al 2022). We recently developed an improved PET radiotracer for CSF dynamics [11C]-Butanol, a highly permeable alcohol that does not bind to brain tissues. [11C]-Butanol with a 20min half-life enables unbiased estimates of brain CSF egress. We hypothesized that CSF turnover is reduced in normal elders, associated with amyloid, and that anatomical CSF egress pathways could be identified.

Methods

Thirty-two normal subjects between the ages of 20-85y were examined. Dynamic list mode [11C]-Butanol PET, [18F]MK6240, and [11C]-PiB were sequentially obtained and co-registered to MRI. Tracer time activity data were used to model the egress of CSF from brain and extra-cranial sites. We examined known CSF draining regions including: brain, ventricle, nasal turbinates, superior sagittal sinus, optic nerve sheath, and control regions.

Results

Ventricular [11C]-Butanol production and clearance reductions are age associated (r=-.83, p<.01) and reduced in subjects above the median age of 67y (~23%, p<.05). Ventricular CSF egress reductions are associated with global brain amyloid (p<.05), but not global brain tau estimates. Using [11C]-Butanol we identified the optic nerve sheath as a CSF egress site affected by aging.

Conclusions

PET imaging of CSF may provide quantitative evidence for drainage pathways carrying misfolded proteins impacted by age.

Aims

In Alzheimer´s disease (AD), brain glucose hypometabolism, indexed by [¹⁸F]FDG-PET, is considered a biomarker of neurodegeneration. However, other brain cells, such as astrocytes and microglia, also consume considerable amounts of glucose and may contribute significantly 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, astrocyte, and microglia) associate with brain [¹⁸F]FDG-PET signal in a rat model of human amyloidosis.

Methods

[¹⁸F]FDG-PET imaging was conducted in ten-month-old APP/PS1 (TgF344-AD, n=8) and wild-type (WT, n=6) rats. Next, we evaluated the gene expression of GFAP, NeuN, and IBA1 in the frontal and temporoparietal cortices and cerebellum. GFAP protein levels were also quantified in the same brain regions, cerebrospinal fluid (CSF), and plasma. Association maps integrating protein or mRNA with brain [¹⁸F]FDG-PET were conducted at the voxel level using RMINC. Differences were considered statistically significant at p < 0.05 (t > 2).

Results

GFAP mRNA levels in the temporoparietal (local maxima, t₍₁₃₎=9.4; Fig1A) and frontal (local maxima, t₍₁₃₎= 6.6; Fig1A) cortices positively correlated with brain [¹⁸F]FDG-PET. No associations were found with neuronal and microglial markers mRNA levels (t₍₁₃₎<2; Fig1B and Fig1C, respectively). Furthermore, we found positive associations between plasma GFAP and brain [¹⁸F]FDG-PET signal (local maxima, t₍₁₃₎= 10.62; Fig1E), but not with CSF GFAP (t₍₁₃₎<2; Fig1F).

Conclusions

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

Figure 1. Correlation between gene expression, protein immunocontent, fluid biomarkers and [¹⁸F]FDG-PET SUVr.

Aims

Plasma markers of tau, such as phosphorylated (pTau) or total tau detect the protein levels in the blood. They could potentially be useful in the clinical and clinical trial settings to assess changes in tau over time. Our aim was to investigate whether plasma pTau epitopes and N-terminal tau fragment (NTA) predict future accumulation of tau-PET in specific regions.

Methods

We included cognitively unimpaired, mild cognitive impairment and AD participants from the TRIAD cohort. Tau progression was calculated as the relative change (Δ) in [¹⁸F]MK6240. We conducted linear regressions between plasma levels at baseline and Δ[¹⁸F]MK6240, in Braak regions I/II, III/IV and V/VI and voxel-wise. Finally, we assessed which plasma marker was the best predictor of tau accumulation at specific Braak stages by comparing β-values of the linear regressions between plasma markers (z-scores) and Δ[¹⁸F]MK6240, correcting for age and sex.

Results

We observed a strong correlation between baseline plasma pTau181, 231 and 217 and ΔBraakIII/IV and ΔBraakV/VI, when NTA correlated with ΔBraakV/VI only (Figure1). Voxel-wise analyses showed a positive relationship between Δ[¹⁸F]MK6240 and plasma pTau181, 231 and 217 in the superior frontal, orbitofrontal and cingulate cortices. Plasma NTA signal was observed in the cuneus, paracentral and precentral gyri (Figure2). Finally, among the significant associations, the stronger β-value was for plasma pTau231 for ΔBraakIII/IV and plasma NTA for ΔBraakV/VI (Figure3).

Conclusions

Plasma pTau biomarkers(231, 181, 217) are great predictors of future tau accumulation in regions outside of the medial temporal lobe (BraakIII and above), while NTA is related to accumulation at late Braak stages(V/VI). Plasma pTau231 was the best predictor of early tau, while NTA was for late Braak regions. Plasma markers could be useful at predicting future tau accumulation, at a stage-specific level.

Aims

To compare the performances of two novel visual read-based methods for flortaucipir-PET in patients with Alzheimer’s Disease (AD) to identify high tau accumulation, defined as AD-signature weighted neocortical Standardized Uptake Value ratio (SUVr)>1.46.

Methods

Baseline flortaucipir-PET scans displaying advanced-AD tau-patterns (increased neocortical activity beyond posterior lateral temporal and occipital regions with bindings 1.65-times higher [1.65x-visual threshold] than cerebellar average) were utilized from the confirmatory phase of A05 study (N=78, NCT02016560). Four expert human-readers conducted reads using two novel visual read-based methods while being blinded to each other and all quantitative results. The first approach (SUVrmax-method) requires a reader to visually identify and hand-draw focal regions-of-interest on frontal areas with the highest bindings and calculate an SUVr between the identified frontal maximum and average cerebellum binding. The second approach (Enhanced Read-method) requires the reader to evaluate binding elevation compared to cerebellum with an enhanced visual threshold, either across gray matter (global, 3.795x-visual threshold) or only in the frontal lobe (2.805x-visual threshold), compared to the standard 1.65x-visual threshold specified in the FDA-approved Tauvid™ label. Methods were compared to the standard quantitative high tau determination as standard-of-truth.

Results

Based on a previously found ROC-determined cut point using computer-generated SUVrmax values, the SUVrmax-method showed median 88% Positive Percent Agreement (PPA), 87% Negative Percent Agreement (NPA) and 87% overall accuracy across raters. The Enhanced Read-method showed median 75% PPA, 88% NPA and 85% overall accuracy and median 81% PPA, 91% NPA and 88% overall accuracy for global and frontal approaches, respectively.

Conclusions

Both novel flortaucipir-PET visual read-based methods showed comparable and promising performances in identifying AD with high tau, do not require computationally intensive approaches and can be implemented with different imaging software.

Aims

The recent FDA approval of [¹⁸F]flortaucipir (FTP) PET reader guidelines has enabled standardized visual assessment. This study aimed to understand how FTP Alzheimer’s disease (AD) visual patterns relate to measures of neurodegeneration from structural MRI scans.

Methods

We included1786 participants, ranging from cognitively normal to demented, with available FTP-PET and T1-weighted MR scans from five cohorts (Table1). The group consisted of 1202 cognitively normal (CN) and 584 cognitively impaired (CI) participants, not selected for amyloid-β-positivity. Three trained readers, blinded to clinical information, assessed FTP scans according to the approved guidelines¹. Majority read was used to classify scans as either negative, moderate, or advanced visual FTP pattern. For voxel-wise analyses, T1-weighted MR scans were segmented, spatially normalized into MNI space, and smoothed with 8mm FWHM using SPM12. Cortical volumes were compared between CN tau-negative and CN or CI tau-moderate and tau-advanced groups, corrected for age, sex and intracranial volume.

Results

FTP scans were classified as 75.2% negative, 6.4% moderate and 18.4% advanced. The CN group was classified as 91.9% negative, 3.3% moderate and 4.8% advanced and the CI group as 51.7% negative, 7.2% moderate and 41.1% advanced. Widespread differences in cortical volume were observed between CN tau-negatives and CI tau-advanced groups, with highest effect sizes in the precuneus and temporal lobe. For CN tau-negatives compared with CI tau-moderate, largest differences were observed in the (medial) temporal lobe and for CN tau-negatives compared with CN tau-advanced participants, small volumetric differences were observed in precuneal, frontal and superior temporal regions (Fig.1a-c).

'

Conclusions

The moderate and advanced visual FTP patterns were related to distinct volumetric changes in CI patients and aligned with previously reported associations between regional FTP signal and measures of neurodegeneration.

Aims

Neurofibrillary tangle pathology (NFTp) in Alzheimer’s disease (AD) is closely related to neurodegeneration, which can be measured in-vivo using structural MRI or FDG-PET. We aimed to assess the relative sensitivities of FDG-PET-measured hypometabolism and MRI-measured gray matter (GM) atrophy to early stages of NFTp as assessed by post-mortem neuropathological examination.

Methods

We studied 64 individuals from the Alzheimer’s Disease Neuroimaging Initiative autopsy cohort who had Braak NFTp staging performed at autopsy and had FDG-PET and structural T1-MRI scans acquired before death (imaging-to-death interval: 3.4±2.3 years). Associations of Braak stages with regional FDG-PET SUVRs and GM volumes on MRI were assessed in exploratory brain-wide Spearman correlation analyses across 52 cortical and subcortical brain regions. For regions showing a significant association (p<0.05, FDR-corrected), we then performed pair-wise comparisons between grouped Braak stages 0/I (N=7), II-IV (N=14), and V/VI (N=43).

Results

Higher Braak stages were significantly associated with lower FDG-PET SUVR in several temporo-parietal cortical regions typically affected by AD (Fig. 1A). Compared to the Braak 0/I reference group, most of these regions showed significant and pronounced (Cohen’s d>0.9) hypometabolism in early Braak stages II-IV, and severity of hypometabolism further increased in Braak stages V/VI. Although GM volume on MRI showed a similar regional association with Braak stages (Fig. 1B), effect sizes were considerably lower and differences to the Braak 0/I reference group were only significant for advanced Braak stages V/VI.

Conclusions

Glucose hypometabolism as measured by FDG-PET is a sensitive neuroimaging marker of the neurodegenerative changes that accompany progressive stages of neurofibrillary tangle pathology. Earliest NFTp-related neurodegenerative changes captured by FDG-PET hypometabolism appear to precede macrostructural gray matter atrophy as measured by MRI.