Citations

Objective Pre‐mortem diagnosis of parkinsonism is often challenging due to atypical presentations, overlapping syndromes, and co‐pathologies. This study aimed to develop a machine learning‐based algorithm predicting neuropathology in parkinsonism using chronological clinical presentations, which has previously been underexplored. Methods Clinical information was automatically abstracted from medical records of the Mayo Clinic Brain Bank using fine‐tuned Generative Pre‐trained Transformer 4 models. Patients who developed parkinsonism within 3 years of disease onset were included. Six machine learning models were trained with age, sex, family history, and 197 clinical presentations paired with onset information to predict neuropathologic diagnoses, including co‐pathologies. Results Among 7,825 donors, 949 met inclusion criteria, representing 9 neuropathologic categories: Lewy body disease (LBD; n = 128), LBD with Alzheimer's disease (AD; n = 136), progressive supranuclear palsy (PSP; n = 303), PSP with AD (n = 56), PSP with LBD (n = 27), multiple system atrophy (MSA; n = 120), corticobasal degeneration (CBD; n = 99), AD (n = 43), and frontotemporal lobar degeneration (FTLD; n = 37). The CatBoost algorithm achieved an area under the receiver operating characteristic curve of 0.83 across the 9 diagnostic categories at 3 years after onset. Important predictors included age at onset, restricted eye movement, and tremor. The model remained robust to incomplete data, requiring only 23 of 200 parameters for reliable predictions with an area under the curve of 0.80. The algorithm was implemented into a user‐friendly program providing diagnostic probabilities with visualizations of parameter contributions. Interpretation This neuropathology‐confirmed diagnostic algorithm provides a cost‐effective and interpretable screening tool for parkinsonism, bridging biomarker testing and molecular‐targeted therapies. ANN NEUROL 2026;99:1405–1414

Background TAR DNA-binding protein 43 (TDP-43) proteinopathy contributes to Alzheimer's disease progression, but diagnosis remains autopsy-dependent. Objective This study aimed to develop a predictive model using FDG-PET imaging to identify brain regions associated with TDP-43 pathology in vivo. Methods Penalized logistic regression analyzed data from 294 participants and a subset of 159 with Braak and Thal ≥4. Features included FDG standard uptake value ratios (SUVRs) from 123 brain regions and age. Participants were split into training (90%) and testing (10%) datasets, with significant predictors identified by non-zero coefficients at optimal lambda. Results The model achieved 68% accuracy (AUC = 0.70) in 294 participants, highlighting the middle temporal gyrus, parahippocampal gyrus, and hippocampus. In the subset (n = 159), accuracy was 65% (AUC = 0.71), with the medial amygdala and precentral gyrus as key predictors of TDP-43 pathology. Conclusions This study suggests a moderate predictive accuracy of FDG-PET to identify brain regions associated with Alzheimer's disease-related TDP-43 proteinopathy.

Abstract INTRODUCTION The tau positron emission tomography (PET) overlap index (OI) has shown promise in maximizing signal‐to‐noise for longitudinal tau PET imaging, particularly for early tau pathology, but requires validation against neuropathology. METHODS Fifty‐seven participants who underwent serial tau PET imaging (flortaucipir) and subsequent autopsy were included. Tau PET OI and standardized uptake value ratios (SUVRs) were compared across neuropathological diagnoses. RESULTS Tau PET OI showed greater concordance with neurofibrillary tangle (NFT) severity in the entorhinal cortex (a key region for Alzheimer's disease [AD] tauopathy) than SUVR, particularly in early Braak tangle stages (positivity: 52.2% for OI vs. 13.0% for SUVR). OI detected overlapping tau voxels that exhibited spatial correspondence with immunohistochemical and autoradiography measures of tau deposition across both AD and non‐AD tauopathies. DISCUSSION These findings demonstrate the enhanced capacity of OI in serial tau PET to robustly detect early and spatially localized tau pathology, supporting its application as a sensitive imaging metric in AD and select non‐AD tauopathies.

Abstract Background XK disease is a multisystem neurodegenerative disorder caused by mutations in the XK gene that codes for the lipid scramblase XK. Objective The aim was to describe the lipidomic spectrum in postmortem brain tissue from XK patients. Methods We measured the levels of 593 lipid species in the caudate nucleus (CN), putamen, and dorsolateral prefrontal cortex (DLPFC) from postmortem tissues of 5 XK patients and 6 controls. Results In XK patients, we observed increased levels of triacylglycerol, monoacylglycerol, phosphatidylserine, and ceramide in the CN. Acylated phosphatidylglycerol levels were reduced in both the CN and putamen. Acyl carnitine, dihydrosphingomyelin, and monosialodihexosylganglioside were reduced, whereas N ‐acyl phosphatidylethanolamine was increased in the DLPFC. N ‐Acyl serine was reduced in all three regions. Conclusions Our findings provide initial evidence of abnormal sphingolipid and phospholipid concentrations in the brains of XK patients and may provide insights into mechanisms of neurodegeneration in this disease. © 2026 International Parkinson and Movement Disorder Society.

Abstract INTRODUCTION Globular glial tauopathy (GGT) is a rare type of frontotemporal lobar degeneration (FTLD) characterized by deposition of 4‐repeat tau. Detecting GGT with tau positron emission tomography (tau‐PET) is challenging. We aim to determine the associations between tau‐PET, autoradiography, and neuropathology methods in GGT. METHODS We identified three patients with GGT who had completed antemortem tau‐PET. Healthy control and two patients with Alzheimer's disease (AD) were also included for comparison. We analyzed gray matter (GM) and white matter (WM) from the superior/middle frontal gyrus (S/M‐FG) and the superior/middle temporal gyrus (S/M‐TG). Immunohistochemical staining was performed with antibodies against phospho‐tau (AT8, PHF‐1, RD4), glial fibrillary acidic protein (GFAP), and resting microglia (ionized calcium‐binding adaptor molecule 1; Iba1). Immunofluorescence with a fluorescent tau‐PET analog (T726) was performed. Regional tau‐PET standardized uptake value ratios (SUVRs) were calculated for comparative GM and WM regions. Autoradiographic studies with 18F‐AV1451 were also conducted. RESULTS Tau‐PET showed increased uptake in the WM regions of the S/M‐FG and S/M‐TG in GGT, and the GM regions in AD. Co‐localization was observed between T726 and PHF‐1, RD4, and GFAP in the WM in the patients with GGT, whereas co‐localization was observed with PHF‐1 predominantly in the GM in AD. Little co‐localization was observed with Iba1. In vitro 18F‐AV1451 autoradiography studies demonstrated minimal binding in GGT. Tau‐PET differentiated underlying GGT from AD based on the relative involvement of the WM and GM. CONCLUSION Histopathologic findings suggest that some flortaucipir uptake in GGT may represent underlying 4R tau, whereas autoradiographic analysis suggests that uptake is likely due to off‐target binding. Further studies with larger cohorts are needed to determine the pathological basis of flortaucipir uptake in GGT.

Dominant-inactivating mutations in the colony stimulating factor-1 receptor ( CSF1R ) cause CSF-1R–related leukoencephalopathy (CRL), an adult-onset neurodegenerative disease that is modeled in the Csf1r +/– mouse. CRL is caused by microglial dysfunction. However, the primary microglial deficit is unknown. To address this question, we employed single-nucleus RNA sequencing of brains from young Csf1r +/– mice without pathological or behavioral alterations. Reduction of CSF-1R signaling caused metal ion accumulation in brain macrophages, with concomitant activation of cell death and stress response pathways in oligodendrocytes and neuronal subpopulations. Reduction of metallothionein 1 ( Mt1 ) and 3 ( Mt3 ) gene expression was a common feature in glial and neuronal cells of Csf1r +/– mice. Overexpression of Mt1 restored metal ion homeostasis, normalized ROS production in microglia, and prevented the development of behavioral deficits, while Mt3 deletion had disease-enhancing effects. These findings demonstrate CSF-1R regulation of metal ion homeostasis via metallothioneins in the brain.

Abstract INTRODUCTION Tauopathies are a heterogeneous group of neurodegenerative disorders defined by abnormal aggregation of tau protein. Although cryogenic electron microscopy (cryo‐EM) has uncovered disease‐specific tau structures, translating these insights into diagnostic tools remains difficult. METHODS We developed a heparin‐free, salt‐modulated real‐time quaking‐induced conversion (RT‐QuIC) assay using K12 and K11 tau substrates, targeting aggregation‐prone regions. This current method improves on previous methodology by minimising the number of required substrates by modulating reaction salt content in order to differentiate yet‐undistinguished tauopathy strains. Thioflavin T fluorescence kinetics and attenuated total reflectance Fourier transform infrared spectroscopy (ATR‐FTIR) spectroscopy were used to classify tau aggregates from human brain homogenates. RESULTS This method differentiated eight tauopathies, including Alzheimer's disease, Pick disease, progressive supranuclear palsy (PSP), corticobasal degeneration (CBD), argyrophilic grain disease (AGD), frontotemporal dementia with parkinsonism associated with chromosome 17 with N279K mutation (FTDP‐17 N279K), and globular glial tauopathies types II and III. Subclassification of 4R tauopathies was achieved by modulating salt conditions and analyzing aggregation profiles. FTIR confirmed preservation of conformational differences. DISCUSSION This salt‐modulated, heparin‐free RT‐QuIC platform enables sensitive tauopathy classification based on strain‐specific kinetics and structure. It offers a practical tool for diagnostic development, mechanistic studies, and therapeutic screening.

Abstract INTRODUCTION The locus coeruleus (LC) degenerates early in Alzheimer's disease (AD). However, the extent of rostrocaudal degeneration across clinicopathologic heterogeneity remains underexplored in AD. METHODS Using digital pathology, we quantified LC neuronal density and area at three neuroanatomic levels in a large AD series. RESULTS Analysis of neuropathologic AD subtypes revealed greater middle LC vulnerability in hippocampal sparing AD compared to typical and limbic predominant AD. Regression analyses identified distinct predictor variables associated with the degeneration of rostral and middle LC. Age at onset predicted 24% of the variability in rostral LC neuronal density, whereas Braak stage, brain weight, Lewy body disease, and age at onset accounted for 15% of the variability in middle LC. Analyses of clinical presentations revealed lower rostral LC neuronal density in non‐amnestic compared to amnestic AD cases. DISCUSSION These insights demonstrate greater LC degeneration in atypical clinicopathologic forms of AD, including hippocampal sparing, young‐onset, and non‐amnestic presentations.

Abstract Recent advances in long-read transcriptome sequencing enable high-throughput profiling of full-length RNA isoforms in bulk, single-cell, and single-nucleus samples. However, long-read datasets typically contain a mixture of complete and partial transcripts, leading to pervasive ambiguity in read-to-isoform assignment and complicating accurate isoform identification and quantification, particularly in the absence of reliable reference annotations. These challenges are further amplified in single-cell and single-nucleus samples, where coverage is sparse and transcriptional heterogeneity is high. Here, we present the Long Read Alignment Assembler (LRAA), a unified and versatile computational framework for isoform identification and quantification from long-read RNA sequencing data across bulk, single-cell, and single-nucleus transcriptomic samples. LRAA combines splice-graph based structural modeling with expectation maximization based optimization to probabilistically resolve ambiguous read assignments and improve isoform abundance estimation. The framework supports quantification-only, reference-guided, and fully reference-free (de novo) modes of analysis within a single methodological paradigm. We benchmarked LRAA using both simulated and genuine long-read datasets spanning sequencing standards and whole transcriptomes. Central to this evaluation is a novel benchmarking strategy based on Multiplexed Overexpression of Regulatory Factors (MORFs), which provides biologically expressed, barcoded isoforms with unambiguous read-level ground truth. Across all benchmarks, including MORFs, synthetic spike-ins, and whole-transcriptome datasets, LRAA consistently outperformed state-of-the-art methods in isoform identification accuracy, sensitivity, and expression quantification. Finally, we demonstrate the biological utility of LRAA by resolving cell-type-specific isoform usage across peripheral blood immune cell populations and by detecting a pathogenic cryptic isoform of STMN2 with associated transcriptional changes in single-nucleus RNA-seq data from frontal cortex tissue of an individual with frontotemporal dementia (FTD). Together, these results establish LRAA as a robust and general solution for resolving transcript diversity in complex biological systems, from development to disease.