PET markers of tau and neuroinflammation are co-localized in progressive supranuclear palsy

Background: Progressive Supranuclear Palsy (PSP) is associated with tau-protein aggregation and neuroinflammation, but it remains unclear whether these pathogenic processes are related in vivo . Objectives: We examined the relationship between tau pathology and microglial activation using [ 18 F]AV-1451 (indexing tau burden) and [ 11 C]PK11195 (microglial activation) PET in n=17 patients with PSP- Richardson’s syndrome . Methods: Non-displaceable binding potential (BP ND ) for each ligand was quantified in 83 regions of interest (ROIs). [ 18 F]AV-1451 and [ 11 C]PK11195 BP ND values were correlated across all ROIs. The anatomical patterns of [ 18 F]AV-1451 and [ 11 C]PK11195 binding co-localization was determined across sets of regions derived from principal component analyses (PCAs). Finally, PCA-derived brain patterns of tau pathology and neuroinflammation were linked to clinical severity. Results: [ 18 F]AV-1451 and [ 11 C]PK11195 binding were positively related across all ROIs (r=0.577, p<0.0001). PCAs identified four components for each ligand, reflecting the relative expression of tau pathology or neuroinflammation in distinct groups of brain regions. Positive associations between [ 18 F]AV-1451 and [ 11 C]PK11195 components were found in sub-cortical (r=0.769, p<0.0001) and cortical components(r=0.836, p<0.0001). PCA-derived components reflecting tau burden (r=0.599, p=0.011) and neuroinflammation (r=0.713, p=0.001) in sub-cortical areas related to disease severity. Conclusions: We show that tau pathology and neuroinflammation co-localize in PSP, and that individual differences in subcortical tau pathology and neuroinflammation are linked to clinical severity. Although longitudinal studies are needed to determine how these molecular pathologies are causally linked, we suggest that the combination of tau- and immune-oriented strategies may be useful for effective disease-modifying treatments in PSP. the development of therapeutic strategies that synergistically target neuroinflammation and tau pathology in PSP. This study aimed to determine the correlation between microglial activation and tau burden in patients with PSP. We test the hypothesis that neuroinflammation and tau protein aggregation co-localise, and correlate with clinical severity. We assessed the topography of[ 11 C]PK11195 and [ 18 F]AV-1451 patterns of binding using: 1) a regions of interest (ROI) approach in which non-displaceable binding potential (BP ND ) values for each ligand were correlated across all ROIs; and 2) a principal component analysis (PCA) which reveals the set of spatially distributed patterns of pathogenic processes, representing the distribution and heterogeneity of pathology in PSP. activation may precede the formation of NFT (55) and then drive the spreading of pathological tau (56). Our findings suggest that the co-localization of neuroinflammation and tau pathology is an important pathogenetic mechanism in PSP, and both processes may be involved in defining the PSP clinical severity. A better understanding of the interaction between the pathological substrates in PSP and its effects on disease progression may crucially contribute to improving patients’ stratification and clinical trials. Specifically, our results encourage the application of [ 18 F]AV-1451 and [ 11 C]PK11195 PET as markers of co-localised pathological mechanisms in PSP to develop new targeting therapies and empower clinical trials.

the development of therapeutic strategies that synergistically target neuroinflammation and tau pathology in PSP.
This study aimed to determine the correlation between microglial activation and tau burden in patients with PSP. We test the hypothesis that neuroinflammation and tau protein aggregation colocalise, and correlate with clinical severity. We assessed the topography of[ 11

Participants
As part of the Neuroimaging of Inflammation in Memory and Other Disorders (NIMROD) study (29), we recruited 17 patients with a clinical diagnosis of probable PSP according to Movement Disorder Society (MDS) 1996 criteria (30). All patients also met the later MDS-PSP 2017 criteria for PSP-Richardson's syndrome (31). Patients underwent PET scanning with both [ 18 F]AV-1451 and [ 11 C]PK11195, to respectively assess tau pathology and neuroinflammation. To minimise radiation exposure in healthy people, two groups of control participants were enrolled: n=15 underwent For each subject, the aligned dynamic PET image series for each scan was rigidly coregistered to the T1-weighted MRI image. BPND was calculated in 83 cortical and subcortical ROIs using a modified version of the Hammers atlas (33,34), which includes brainstem parcellation and the cerebellar dentate nucleus. Prior to kinetic modelling, regional PET data were corrected for partial volume effects from cerebrospinal fluid by dividing by the mean regional grey-matter plus white-matter fraction determined from SPM segmentation. For [ 11 16)). The same data acquisition and analysis approach was applied for the two control groups.

Statistical analyses
Age, years of education, ACE-R total and fluency scores were compared between patients and controls with independent-samples t-tests, while gender was compared with the Chi-square test. Linear regression models were applied to the longitudinal PSP-RS scores in individual patients to estimate the PSP-RS score at the time of each PET scan. Image analysis proceeded in four steps.
All statistical analyses were performed in SPSS Statistics version 25 (IBM).
First, to test whether microglial activation and tau pathology co-localised across the whole brain, we estimated the Pearson correlation of corresponding [ 11 C]PK11195 and [ 18 F]AV-1451 BPND group-average values across all 83 ROIs ( Figure 1).
Second, the number of ROIs was reduced from 83 to 46, averaging left and right regional BPND values, as in previous studies (16,28). This step reduces the degrees of freedom, increasing power, and is justified in PSP in which the motor syndrome is essentially symmetric. The differences between PSP and control groups in the 46 ROIs were tested for each ligand with independent t-tests with false discovery rate (FDR) correction for multiple comparisons.
Third, in PSP patients, BPND values in the 46 bilateral ROIs were included in separate PCAs for [ 11 C]PK11195 and [ 18 F]AV-1451. This reduces the data dimensionality further, identifying a small set of components that best explain the data variance which. The resulting component reveal anatomical patterns covary in terms of neuroinflammation or tau pathology. Varimax rotation was applied in the PCA to increase orthogonality across the different components (i.e., anatomical . CC-BY-NC-ND 4.0 International license It is made available under a author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the (which was not peer-reviewed) The copyright holder for this preprint . https://doi.org/10.1101/19010702 doi: medRxiv preprint patterns of neuroinflammation and tau pathology). The components with eigenvalues > 1 were retained, explaining >80% of the cumulative variance.
Finally, to test for co-localization of microglial activation and tau pathology in specific neuroanatomical patterns of ligand binding, we performed Pearson correlations between individual scores of each ligand-specific component extracted. The analyses adjusted for age differences and variability in the time interval between PET scans, included as covariates of no interest. For each ligand, we tested for correlations between regionally specific PCA clusters (i.e., anatomical patterns of neuroinflammation and tau pathology) and disease severity (the estimated PSP-RS score at the time of each scan). Bonferroni's method was used to correct for multiple comparisons.

Principal component analysis of [ 11 C]PK11195 and [ 18 F]AV-1451 BPND in PSP
For [ 11 C]PK11195 BPND, four components were identified which collectively explained 81.4% of the data variance ( Figure 2, left panel). Component 1 reflected [ 11 C]PK11195 binding in posterior cortical regions, the orbitofrontal cortex and cerebellar grey-matter (62.9% of the total variance).
Component 2 grouped together medial and superior regions of the temporal lobe including the amygdala, hippocampus and para-hippocampal gyrus, as well as other cortical areas such as the insula and temporo-parietal junction (9.2% variance). Component 3 was weighted to brainstem regions (i.e., midbrain and pons), the dentate nucleus, and the cerebellar white-matter (5.1% variance). Component 4 comprised superior and medial frontal regions (4.3% variance).

Correlation between [ 11 C]PK11195 and [ 18 F]AV-1451 principal components in PSP
After adjusting for Bonferroni correction (p=0.05/16 correlations between [ 18 F]AV-1451 and  CC-BY-NC-ND 4.0 International license It is made available under a author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the (which was not peer-reviewed) The copyright holder for this preprint

Discussion
This study suggests that tau protein aggregation and microglial activation are anatomically colocalized in PSP, and relate to disease severity. The relationship is observed across widespread brain regions although it is most evident in some cortical (i.e. insula and temporo-parietal junction) and . CC-BY-NC-ND 4.0 International license It is made available under a author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
is the (which was not peer-reviewed) The copyright holder for this preprint . https://doi.org/10.1101/19010702 doi: medRxiv preprint Nevertheless, we acknowledge that the affinity of [ 18 F]AV-1451 to the 4-repeat tau in non-AD tauopathies as PSP is lower than its affinity to 3/4-repeat AD-related tau pathology (38,39). In PSP patients, increased [ 18 F]AV-1451 binding has been shown sub-cortical rather than cortical regions, consistently with previous neuropathological evidence (16, 18,19,23,24,26). This evidence supports the use of [ 18 F]AV-1451 PET to quantify and localise tau pathology in tauopathies with clear and known pathologic substrates, such as PSP. Our previous study (16) also showed that it is possible to discriminate, with machine-learning approaches and multivariate pattern analyses, the neuroanatomical pattern of [ 18 F]AV-1451 binding in PSP from the one seen in AD. This corroborates the use of [ 18 F]AV-1451 PET as a marker of disease-specific pathological changes.
[ 11 C]PK11195 has been criticised for its relatively low signal to noise ratio and low brain penetration which can affect its sensitivity to activated microglia. However, this would only reduce the effect sizes and increase type II error, rather than leading to false positive findings. Several second-generation PET radioligands for TSPO are characterised by higher signal-noise ratio than [ 11 C]PK11195 but their binding is affected by single nucleotide polymorphisms (rs6971) that cause heterogeneity in PET data and requires genetic screening (45). In contrast, [ 11 C]PK11195 binding is not affected by this polymorphism (45), and it has well established methods of kinetic analysis (35).
Hence, [ 11 C]PK11195 PET remains the most used method to study microglia activation in neurodegenerative diseases (17), and it has been successfully applied in PSP (27,28).
With these caveats in mind, we now discuss our principal results. To study the in vivo colocalization between microglial activation and tau pathology in PSP, we applied correlation analyses between the binding of the two ligands 1) across all brain regions, and 2) between principle components of set of bilateral brain regions, extracted to reduce the complexity of the imaging data.
With the first approach, we found a positive correlation between [ 18 F]AV-1451 and [ 11 C]PK11195 binding, across the whole brain ( Figure 1). This indicates a close association between microglial activation and tau pathology in PSP that extensively involves both subcortical and cortical regions.
This finding also aligns with in vivo correlation between neuroinflammation and tau aggregation in AD and frontotemporal dementia (46,47). Collectively, these multi-tracer PET studies support previous in vitro evidence of the association between microglial activation and tau aggregation in different tauopathies (see review (48)). The spatial distribution of the in vivo association between microglial activation and tau pathology in PSP also mirrors previous findings about neurodegeneration and tau pathology affecting not only subcortical but also cortical regions in PSP found to be impaired in PSP, alongside theory of mind and social cognition (51,52). The recognition of happiness was reported to be preserved in patients with PSP, while the recognition of negative emotions (i.e. anger, disgust, surprise, fear and sadness) was affected in these patients (51). Basal ganglia, insula and amygdala have been reported to be implicated in the recognition of predominately negative emotions, and to be pathologically affected in PSP (49, 52,53). The association between microglial activation and tau pathology that we found in limbic regions may complement the biological explanation of emotion-related and social deficits in PSP, however, longitudinal studies are needed to clarify the timing of these interacting effects on the pathological and clinical disease progression.
In addition, our finding of tau and neuroinflammation co-localization in the cortex of patients with PSP Richardson's syndrome is in keeping with previous post-mortem evidence showing tau pathology and atrophy not only in subcortical and limbic regions, but also in the parietal lobe (49,54). Specifically, the supramarginal gyrus has been described as the most affected brain region in . CC-BY-NC-ND 4.0 International license It is made available under a author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
is the (which was not peer-reviewed) The copyright holder for this preprint . https://doi.org/10.1101/19010702 doi: medRxiv preprint two independent pathological cohorts of patients with PSP Richardson's syndrome (49,54). The absence of in vivo evidence about supramarginal atrophy in the literature may enhance the importance of the association between neuroinflammation and tau accumulation in this region as an early biomarker of a later-stage neuronal loss.
Our study has some limitations. First, we acknowledge the limited power of the analyses related to the relatively small size of our sample. Although our cohort is larger than many previous multi-tracer PET studies on rare neurodegenerative diseases like PSP, our findings will benefit from larger and independent replication samples. Second, the recruitment was based on clinical diagnosis, which was confirmed at each follow-up visit; however, post-mortem pathological confirmation was available for only 7 patients. Third, our results are based on a cross-sectional design, which cannot be used to infer causal relationship between tau and microglial activation. To conclude, our results confirm the relevance of neuroinflammation to PSP-Richardson's syndrome and a close association with tau pathology in the core regions for this disease. Our findings indicate a topographical overlapping between neuroinflammation and tau pathology in those regions previously described by post-mortem studies as mainly involved in PSP pathophysiology. Although we cannot infer the causal direction in the relationship between pathological mechanisms, we speculate that microglial activation may be activated by an initial tau misfolding and contribute to tau pathology and propagation. The latter, in turn, may lead to further neuroinflammation, as previously suggested in AD (see review (48)). A growing literature from preclinical research suggests that microglial activation may precede the formation of NFT (55) and then drive the spreading of pathological tau (56). Our findings suggest that the co-localization of neuroinflammation and tau pathology is an important pathogenetic mechanism in PSP, and both processes may be involved in defining the PSP clinical severity. A better understanding of the interaction between the pathological substrates in PSP and its effects on disease progression may . CC-BY-NC-ND 4.0 International license It is made available under a author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
is the (which was not peer-reviewed) The copyright holder for this preprint . CC-BY-NC-ND 4.0 International license It is made available under a author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
is the (which was not peer-reviewed) The copyright holder for this preprint . https://doi.org/10.1101/19010702 doi: medRxiv preprint Figure 1. Whole brain correlation between regional mean non-displaceable binding potential (BPND) of [ 11 C]PK11195 and [ 18 F]AV-1451 in the PSP group. Each point represents the average value across all patients for a specific brain region, while colours indicate brain macro-areas.
. CC-BY-NC-ND 4.0 International license It is made available under a author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
is the (which was not peer-reviewed) The copyright holder for this preprint . https://doi.org/10.1101/19010702 doi: medRxiv preprint

Figure 2. First four principal components for [ 11 C]PK11195 non-displaceable binding potential (BPND) and [ 18 F]AV-1451 BPND in the PSP group. The colours represent the rotated weights of all brain regions for each component.
. CC-BY-NC-ND 4.0 International license It is made available under a author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
is the (which was not peer-reviewed) The copyright holder for this preprint . https://doi.org/10.1101/19010702 doi: medRxiv preprint  . CC-BY-NC-ND 4.0 International license It is made available under a author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
is the (which was not peer-reviewed) The copyright holder for this preprint . https://doi.org/10.1101/19010702 doi: medRxiv preprint