Gunter, Jeffrey L.

Abstract Background Individuals with coexisting Alzheimer's disease (AD) and disorders of CSF dynamics such as normal‐pressure hydrocephalus (NPH) have lower CSF Aβ42 levels, which can confound interpretation of AD CSF‐biomarkers. However, little is known about how CSF dynamics influences plasma AD biomarkers. The objective of this study was to determine whether mass spectrometry‐based blood tests for AD are associated with disproportionately enlarged subarachnoid space hydrocephalus (DESH) on MRI. Method 509 Individuals from the Mayo Clinic Study of Aging, with MRI, 11C‐Pittsburgh compound B (Aβ) PET scans, and plasma phosphorylated‐tau protein 217, Aβ40 and Aβ42 were included. We performed an age‐adjusted logistic regression to determine whether abnormal values of p ‐Tau 217 (Abnormal if >1.55 pg/mL), %p‐tau217 defined as the ratio of phosphorylated Tau217 to non‐phosphorylated tau x 100% (Abnormal if >1.55%), Aβ42 (Abnormal if <41.5 pg/mL), and Aβ42/40 ratio (Abnormal if <0.094) were associated with the presence of DESH, as classified by a previously validated automated algorithm. Result Thirty‐two participants were classified as DESH‐positive. The median [IQR] age was 76.8 yrs [70.6, 82.9] for participants with DESH and 72.4 [59.3, 85.5] without DESH (Table 1). An abnormal plasma Aβ42 was associated with DESH (OR, 2.56 [1.19, 5.74], p = 0.018), but having an abnormal Aβ42/40 ratio (OR 2.01 [0.94, 4.56], p = 0.079) was not. There was no association between being DESH‐positive and an abnormal plasma p ‐tau217 (OR 1.30 [0.51, 3.09], p = 0.560) or with abnormal %plasma p ‐tau217 (OR 1.82 [0.74, 4.20], p = 0.17) (Figure 1). Subgroup analysis showed that only those participants who had an abnormal Aβ42 and positive amyloid PET (≥25 Centiloids) showed an association with DESH compared to biomarker‐ and PET‐negative participants (OR 3.10 [1.15, 8.43], p = 0.024 ) . Conclusion Understanding how CSF dynamics affect plasma biomarkers is critical for implementation. The relationship between lower Aβ42 and DESH is expected based on prior CSF studies, which can be addressed by using an amyloid ratio. p ‐tau217 shows robustness against DESH, making it a suitable biomarker in cases of abnormal cerebrospinal fluid dynamics although validation in a more severe CSF dynamics group is warranted.

Abstract Background Brain age models are increasingly being used as measures of brain health and neurodegeneration. These models can be deep learning (DL) based and/or classical methods and provide a brain age gap (BAG) which is the difference between chronological and predicted brain age. Our goal was to develop DL‐based scanner invariant brain age models that can be applied in aging and dementia studies even when there are scanner changes that may hinder longitudinal follow‐up. Method We identified 3374 cognitively unimpaired (CU) participants with T1 scans from the Mayo Clinic Study of Aging (MCSA). A subset of 863 CU amyloidosis negative (PiB‐PET cutoff < 25 Centiloid) and no history of diabetes and hypertension was selected (CU‐noVAβ). We built two models – Model‐A trained on CU‐noVAβ, and Model‐B trained on all CU. We included 453 MCSA MCI participants as test set for both models. We used 113 participants (74 CU, 28 MCI, 10 dementia) from MCSA and Mayo ADRC with scans from two different vendors (scanned within a week of each other) for testing cross‐vendor/scanner differences (cross‐vendor data). The two unique features of our model included histogram matching and scanner type as an input to the DenseNet model for training (Figure 1). Result In the data, the inclusion of scanner as an input decreased the prediction of mean age differences between the scanners (Model‐A=2.17 years; Model‐B=1.71 years) (Figure 2). The mean absolute error (MAE=absolute value of BAG) for Model‐A in MCI was 5.72 years and Model‐B in MCI was 4.06 years, consistent across scanners (Figure 3). Longitudinal prediction increased with age (Model‐A slope=0.53; Model‐B slope=0.63). Conclusion We found that training on all CU (Model B) rather than a subset CU‐noVAβ (Model A) allowed for the DenseNet model to backpropagate on more variable data which lowered MAE in prediction error between scanners and lowered MAE for MCIs. Histogram matching and scanner as an input to the brain age model accounted for scanner changes and resulted in scanner invariant longitudinal brain age estimation.

Abstract Background Neuropathologic evaluation is essential for diagnosing and treating dementia, but definitive diagnoses is only available after post‐mortem brain autopsy. Positron emission tomography (PET) offers a potential in vivo proxy for predicting neuropathology. This study proposes a semi‐supervised learning (SSL) model using a single FDG‐PET image to predict underlying neuropathology. Method The SSL model leverages a self‐organized graph structure based on a patient similarity network (PSN) created from 1,495 FDG‐PET images across the aging and neurodegenerative spectrum. Each patient is represented by a node, with edges indicating image similarity derived from principal component analysis features and cosine similarity weights. The PSN structure connects 20 nearest neighbors. Neuropathologic labels were available for 204 autopsied patients. The Poisson Learning Method was employed to propagate labels from these annotated nodes to unlabeled nodes, guided by the PSN structure. FDG‐PET images were collected from clinical and research participants at the Mayo Clinic and processed using an MRI‐free pipeline. Neuropathologic labels were assigned based on widely accepted diagnostic criteria. Leave‐one‐out cross‐validation (LOOCV) was used to assess model performance. Results The confusion matrix demonstrated high accuracy in predicting Alzheimer's disease, frontotemporal lobar degeneration with tau inclusions, and Lewy body disease pathology. A Sankey diagram and label distribution analysis revealed strong correlations between clinical and predicted neuropathologic labels. Conclusion This study highlights the potential of a graph‐based SSL pipeline to predict neuropathological changes using FDG‐PET imaging data, despite limited pathology labels. The findings underscore the value of SSL in bridging the gap between abundant neuroimaging data and scarce neuropathologic labels, contributing to improved dementia diagnosis and understanding.

Abstract Background Multi‐site aging and dementia studies are challenged by data heterogeneity caused by scanner (manufacturer and protocol) related variability. Quantitative analysis of white matter hyperintensity (WMH) from pooled samples, a traditional marker of cerebrovascular disease, is sensitive to machine variability. Our goal was to compare Deming (symmetric) linear regression and ComBat, a common method used for harmonizing neuroimaging measurements, for harmonizing WMH measurements on participants scanned on two different scanners (GE and Siemens) and protocols (2D FLAIR on GE and 3D FLAIR on Siemens). Methods Cognitively unimpaired (CU) ( n =53, age (mean (SD))=67 (15) years) and cognitively impaired (CI; MCI/dementia) ( n =34, age (mean (SD))=82 (9) years) participants from the Mayo Clinic Study of Aging and Mayo ADRC were scanned on GE and Siemens scanners within a week of each other (typically done on the same day). WMH as a percentage of total intracranial volume was measured using multispectral segmentation of FLAIR and T1‐weighted images (Gunter et. al. AAIC 2023). We used Deming regression as it includes errors in both input dimensions. ComBat was applied using age, sex, and diagnosis as covariates, with scanner vendor and protocol as the batch variable. We assessed the inter‐scanner similarity using Maximum Mean Discrepancy (MMD) analysis, where a value of zero indicates identical distributions, p >0.05 indicates no significant difference under H 0 . (Gretton et al., JMLR 2012). Results Our results showed that the distribution of WMH measurements, which were different in the CU group, were more similar after Deming than ComBat harmonization (Figure 1). This was confirmed by the lower MMD in the CU group from Deming regression (‐0.01 ( p =0.76) from 0.03 ( p =0.05)) compared to ComBat (0.04 ( p =0.02))(Figure 1). However, ComBat and Deming were unable to create distributions as compatible as original measurements for the CI group. Conclusion Using a unique dataset of the same participants scanned close in time on different scanner and protocols, we found that (1) MMD measurements were useful to test the need/improvement due to harmonization; (2) ComBat was not substantially better at creating compatible distributions between scanner vendors than Deming regression, and both were sometimes inadequate. Future work will require alternative models to linear scale and shift transformations.

Abstract Background Use of software to “de‐face” research brain images has grown as advances in automated face recognition have made it increasingly possible to re‐identify participants. Imaging studies of Alzheimer’s Disease and related diseases have included more participants from historically underrepresented groups, but de‐facing software have primarily been validated using images from mostly non‐Hispanic White participants, and its performance has not been directly compared across ethnoracial groups. Method From The Health and Aging Brain Study: Health Disparities (HABS‐HD) study, we constructed a sample of 305 T1‐weighted MRI from HABS‐HD baseline visits, matched across three ethnoracial groups (non‐Hispanic White (NHW, n = 104), non‐Hispanic African American (AA, n = 104), Mexican American (MA, n = 97)) according to age‐decade, sex, cognitive status (impaired/unimpaired), and MRI scanner model. We ran mri_reface 0.3.5 on each image, and we ran FreeSurfer 7.4.1 on each unmodified image and on each de‐faced output. We then compared regional gray matter (GM) volume and cortical thickness measurements by subtracting each de‐faced image measurement from the corresponding unmodified image measurement. We compared these differences between de‐faced and unmodified images across the three matched groups using GLM analyses adjusted for age, sex, and cognitive status, and Bonferroni corrected. Result When aggregating across all brain regions, differences in cortical thickness measures due to de‐facing were significantly different between the NHW and MA groups, and between AA and MA, but not between NHW and AA (Figure 1). Differences between these group means were <0.005mm. For GM volume, differences were significant between AA and MA (magnitude < 0.2%), but not between NHW and AA or NHW and MA (Figure 2). Examining each region individually (Figure 3), cortical thickness differences were significant for only 4 regions (all with magnitudes <0.03mm), and GM volume differences were significant for only 6 regions (all with magnitudes <2%). Conclusion Effects of mri_reface on FreeSurfer brain measurements were significantly different across ethnoracial groups. Our future work will optimize mri_reface to eliminate these differences. However, the magnitudes of these differences are sufficiently small that they are unlikely to have a practical effect on analyses of data using current versions.

Abstract INTRODUCTION Plasma biomarkers’ utility for predicting incident mild cognitive impairment (MCI) remains unclear. We evaluated associations of plasma Alzheimer's disease (AD) biomarkers and amyloid positron emission tomography (PET) with transitions from cognitively unimpaired (CU) to MCI in the Mayo Clinic Study of Aging (MCSA) and BioFINDER‐2 studies. METHODS Associations of continuous baseline plasma biomarker levels and amyloid PET Centiloid with progression to MCI, adjusting for age, sex, and education, were evaluated with Cox proportional hazards models. RESULTS The study included 381 MCSA and 584 BioFINDER‐2 participants. Amyloid PET and percent phosphorylated to non‐phosphorylated tau217 (%p‐tau217) were strong predictors of progression to MCI in both cohorts: hazard ratios of 1.49 and 1.23 in the MCSA and 1.72 and 1.65 in BioFINDER, respectively. Amyloid beta 42/40 was a significant predictor in BioFINDER‐2 only (hazard ratio 2.20). DISCUSSION Plasma %p‐tau217 was associated with progression from CU to MCI in both cohorts, although differences in biomarker associations may be related to differences in the two cohorts. Highlights Mass‐spectrometry‐based plasma phosphorylated tau217 was associated with cognitively unimpaired to mild cognitive impairment (MCI) progression. Plasma amyloid beta 42/40 was a significant predictor in BioFINDER but not the Mayo Clinic Study of Aging (MCSA). Amyloid positron emission tomography (PET) was the strongest predictor of progression to MCI in the MCSA. Plasma had added value to amyloid PET in BioFINDER but not the MCSA. Biomarker performance may vary with cohort and biomarker measurement differences.

Abstract Objective To compare iron deposition in the substantia nigra (SN) as measured with quantitative susceptibility mapping (QSM) on antemortem magnetic resonance imaging (MRI) between individuals with and without Lewy‐related pathology at autopsy. Methods We performed a retrospective cohort study including 54 participants who underwent autopsy and antemortem MRI with QSM. Cases were classified as Lewy body disease (LBD)‐present (n = 13) if they had Lewy‐related pathology and LBD‐absent (n = 41) if they did not have Lewy‐related pathology. QSM was calculated for the whole SN and the two subregions: pars compacta (SNc) and pars reticulata (SNr). Nonparametric Wilcoxon rank‐sum tests were used to compare SN QSM values between LBD‐present and LBD‐absent cases. Area under the receiver operating characteristic (ROC) curve (AUC) analyses tested the accuracy of SN QSM values to distinguish the two groups. Associations of QSM values in the SN and its subregions with clinical features were tested with Spearman's correlations. Results The LBD‐present group had higher QSM values in the SNc ( P = 0.008) than the LBD‐absent group with no differences in SNr. QSM values of the SNc distinguished LBD‐present and LBD‐absent cases with good accuracy (AUC = 0.74) and correlated with the presence of parkinsonism and parkinsonism severity. Conclusions This study provides neuropathological confirmation of the utility of SNc QSM as an in vivo biomarker of Lewy‐related pathology. Imaging evidence of abnormal iron deposition in the SNc could potentially serve as a biomarker for inclusion in emerging research frameworks aimed at defining individuals with LBD based on their biological characteristics. Ultimately, this could facilitate more precise diagnoses and guide treatment strategies in LBD. © 2025 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Abstract Normal pressure hydrocephalus is typically defined by the triad of gait disturbance, cognitive impairment, and urinary incontinence, and is characterized on MRI by disproportionately enlarged subarachnoid-space. Gait disturbance is also a commonly reported symptom in Parkinsonian disorders, especially progressive supranuclear palsy, although the frequency, clinical significance and mechanisms of hydrocephalus in these disorders are unclear. We aimed to assess the prevalence of hydrocephalic MRI parameters in a large cohort of Parkinsonian disorders and evaluate associations with clinical features and abnormalities on MRI and PET. Two hundred and thirty-eight participants with a Parkinsonian disorder, including 181 progressive supranuclear palsy, 36 corticobasal syndrome and 21 Parkinson’s disease, were enrolled from Mayo Clinic by the Neurodegenerative Research Group between September 2009 to October 2023. Automated detection of disproportionately enlarged subarachnoid-space hydrocephalus (D) was applied and using Evans’ index (E) cut-off point &gt;0.3, participants were classified based on both measures as imaging-suggestive of hydrocephalus (D+E+), enlarged subarachnoid-space only (D+E−), large Evans’ index only (D−E+) and no imaging evidence of hydrocephalus (D−E−). Demographic, clinical and imaging features, including magnetic resonance parkinsonism index, cortical and subcortical volumes, white matter hyperintensities, diffusion tractography metrics, and metabolism on PET, were compared across groups. Among the 238 participants, 24 had borderline subarachnoid space scores and were excluded. The remaining 214 participants were classified as: D+E+ (n = 20, 9%); D+E− (n = 8, 4%); D−E+ (n = 71, 33%) and D−E− (n = 115, 54%). Among the progressive supranuclear palsy participants, 11% were D+E+, 4% D+E−, 34% D−E+ and 51% D−E−. Most cases (n = 18) in the imaging-suggestive of hydrocephalus D+E+ group had progressive supranuclear palsy. The D+E+ participants were older, had more disorientation, more downgaze palsy, greater midbrain and cortical atrophy, lower striatal metabolism, greater degeneration of long-range white matter tracts, larger cistern areas and more periventricular and deep white matter hyperintensities compared to the D−E− participants without imaging evidence of hydrocephalus. The D+E− participants had the highest metabolism in the paracentral lobule and superior parietal gyrus. The D−E+ participants showed worse disease severity and greater midbrain and cortical atrophy compared to the D−E− participants. These findings demonstrate that disproportionately enlarged subarachnoid-space hydrocephalus occurs in ∼15% of progressive supranuclear palsy participants, and is associated with worse clinical and imaging outcomes, as well as white matter hyperintensities. We hypothesize that disproportionately enlarged subarachnoid-space may be a mechanistic byproduct of degeneration and subsequent cerebrospinal fluid flow re-distribution in progressive supranuclear palsy.

Abstract INTRODUCTION Limited data exist on the utility of plasma biomarkers to predict incident abnormal amyloid positron emission tomography (PET). In this study we evaluate the association of plasma Alzheimer's disease (AD) biomarkers with amyloid PET progression among initially amyloid PET negative (A−) individuals. METHODS We included 290 A−, cognitively unimpaired Mayo Clinic Study of Aging participants. We estimated the association of each baseline plasma biomarker with progression from A− to A+ and with rate of amyloid PET change. RESULTS Interquartile range differences in amyloid beta 42/40, percent phosphorylated tau 217 (%p‐tau217), and Amyloid Probability Score 2 were associated with 1.29 ( P = 0.09), 1.38 ( P < 0.001), and 1.20 ( P = 0.05) increases, respectively, in the hazard of progression from A− to A+ and 0.27 ( P = 0.16), 0.50 ( P = 0.007), and 0.28 ( P = 0.15) Centiloid/year increases, respectively, in annual rate of amyloid PET change. DISCUSSION Plasma %p‐tau217 may be a useful screening tool to enrich for participants with increased likelihood of progressing from normal to abnormal amyloid PET in a primary prevention trial. Highlights Plasma phosphorylated tau 217 was associated with amyloid positron emission tomography progression, negative to positive. The associations were weaker for amyloid beta 42/40 and Amyloid Probability Score 2. Age and apolipoprotein E ε4 carriership were also important predictors. These markers may be useful for enrichment of a primary prevention trial.