Neuropathology and Applied Neurobiology

ObjectiveAmyotrophic lateral sclerosis (ALS) is a devastating neuromuscular disease with limited therapeutic options. Diagnostic and surrogate endpoint biomarkers are needed for early disease detection, clinical trial design, and personalized medicine. MethodsWe tested the predictive power of a large set of primary skin fibroblast (n=443) from sporadic and familial ALS patients and healthy controls. We measured morphometric features of endoplasmic reticulum, mitochondria, and lysosomes by imaging with vital dyes. We also analysed immunofluorescence images of ALS-linked proteins, including TDP-43 and stress granule components. We studied fibroblasts under basal conditions and under metabolic (galactose medium), oxidative (arsenite), and heat stress conditions. We then employed machine learning (ML) techniques on the dataset to develop biomarkers. ResultsStress perturbations caused robust changes in the measured features, such as organellar morphology, stress granule formation, and TDP-43 mislocalization. ML approaches were able to predict the perturbation with near perfect performance (ROC-AUC > 0.99). However, when trying to predict disease state or disease groups (e.g., sporadic, or familial ALS), the performance of the ML algorithm was more modest (ROC-AUC Control vs ALS = 0.63). We also detected modest but significant scores when predicting clinical features, such as age of onset (ROC-AUC late vs early = 0.60). ConclusionsOur findings indicate that the ML morphometry we developed can accurately predict if human fibroblasts are under stress, but the differences between ALS and controls, while statistically significant, are small and pose a challenge for the development of biomarkers for clinical use by these approaches.

Diagnosing Amyotrophic Lateral Sclerosis (ALS) remains challenging due to its inherent heterogeneity. Cytoplasmic aggregation of TDP-43, observed in approximately 95% of ALS cases, has emerged as a key pathological hallmark. In this observational study, we investigated the feasibility of training deep learning models to classify TDP-43 pro-teinopathic samples versus healthy controls, with a particular focus on understanding how dataset limitations affect model performance. The dataset comprised super-resolution immunofluorescence images in which cytoplasmic and nuclear TDP-43 deposits were quantified using red and pink pixel counts. We formulated three classification tasks: TDP-43 pathology (binary), TDP-43 pathology grades (multiclass), and ALS diagnosis (binary). Initial deep learning experiments yielded inconclusive results, prompting dataset curation and the removal of problematic samples. Subsequent statistical analyses using t-tests, ANOVA, and hierarchical clustering revealed significant differences between healthy and pathological samples in terms of pixel distributions, total protein levels, and TDP-43 compart-mentalisation. These findings suggest that classification based on TDP-43 proteinopathy provides a more reliable framework for deep learning compared to ALS diagnosis, underscoring the importance of data quality and task strati-fication in model performance.

BackgroundAmyotrophic lateral sclerosis (ALS) displays considerable clinical, genetic and molecular heterogeneity. Machine learning approaches have shown potential to disentangle complex disease landscapes and they have been utilised for patient stratification in ALS. However, lack of independent validation in different populations and in pre-mortem tissue samples have greatly limited their use in clinical and research settings. We overcame such issues by performing a large-scale study of over 600 post-mortem brain and blood samples of people with ALS from four independent datasets from the UK, Italy, the Netherlands and the US. MethodsHierarchical clustering was performed on the 5000 most variably expressed autosomal genes identified from post-mortem motor cortex expression data of people with sporadic ALS from the KCL BrainBank (N=112). The molecular architectures of each cluster were investigated with gene enrichment, network and cell composition analysis. Methylation and genetic data were also used to assess if other omics measures differed between individuals. Validation of these clusters was achieved by applying linear discriminant analysis models based on the KCL BrainBank to the TargetALS US motor cortex (N=93), as well as Italian (N=15) and Dutch (N=397) blood expression datasets. Phenotype analysis was also performed to assess cluster-specific differences in clinical outcomes. ResultsWe identified three molecular phenotypes, which reflect the proposed major mechanisms of ALS pathogenesis: synaptic and neuropeptide signalling, excitotoxicity and oxidative stress, and neuroinflammation. Known ALS risk genes were identified among the informative genes of each cluster, suggesting potential for genetic profiling of the molecular phenotypes. Cell types which are known to be associated with specific molecular phenotypes were found in higher proportions in those clusters. These molecular phenotypes were validated in independent motor cortex and blood datasets. Phenotype analysis identified distinct cluster-related outcomes associated with progression, survival and age of death. We developed a public webserver (https://alsgeclustering.er.kcl.ac.uk) that allows users to stratify samples with our model by uploading their expression data. ConclusionsWe have identified three molecular phenotypes, driven by different cell types, which reflect the proposed major mechanisms of ALS pathogenesis. Our results support the hypothesis of biological heterogeneity in ALS where different mechanisms underly ALS pathogenesis in a subgroup of patients that can be identified by a specific expression signature. These molecular phenotypes show potential for stratification of clinical trials, the development of biomarkers and personalised treatment approaches.

BackgroundAmyotrophic lateral sclerosis (ALS) is a uniformly fatal neurodegenerative disease characterized by progressive cortical and spinal motor neuron loss, with most patients surviving only 2-5 years post-diagnosis. While approximately 10% of cases are familial (fALS), the remaining 90% are sporadic (sALS) with unknown genetic drivers. Importantly, clinical presentations are heterogeneous in both sporadic and familial ALS, underscoring the complexity of the disease. A pathological hallmark of ALS is the mislocalization of RNA-binding protein TDP-43 from the nucleus to the cytoplasm. This mislocalization produces both loss of function consequences, such as widespread RNA processing and splicing defects, as well as potential toxic gain of function effects associated with cytoplasmic aggregation. ResultsIn this study, we used RT-PCR data from induced pluripotent stem cell-derived motor neurons derived from 180 sALS and C9orf72 fALS patients from the Answer ALS collection to identify biological subgroups based on TDP-43 loss-of-function signatures. Spectral embedding revealed four distinct molecular clusters, including one subgroup genetically similar to controls and another with the most dysregulated mRNA expression, suggesting differing disease severity. Linear mixed models were then used to assess the longitudinal trajectory of over 90 clinical measures, and the between-cluster interaction effects were evaluated. Conclusions36 clinical outcomes showed significant differences across clusters, supporting the presence of biologically and clinically distinct ALS subtypes based on the TDP-43 associated pathogenic cascade. These findings demonstrate a critical role of RNA profiling in uncovering biologically meaningful subtypes of ALS, potentially allowing for more precise prognostic tools and the development of future personalized therapeutic approaches.

Amyotrophic lateral sclerosis (ALS) is a rapidly progressing neurodegenerative disease characterised by the loss of motor control. Current understanding of ALS pathology is largely based on post-mortem investigations at advanced disease stages. A systematic in-vivo description of the microstructural changes that characterise early-stage ALS, and their subsequent development, is so far lacking. Recent advances in ultra-high field (7T) MRI data modelling allow us to investigate cortical layers in-vivo. Given the layer-specific and topographic signature of pathology in ALS, we combined submillimeter structural 7T-MRI data (qT1, QSM), functional localisers of body parts (upper limb, lower limb, face) and automated layer modelling to systematically describe pathology in the primary motor cortex (M1), in 12 living ALS-patients with reference to 12 age-, gender-, handedness- and education-matched controls. Longitudinal sampling was performed for a subset of patients. We calculated multimodal pathology maps for each layer (superficial layer, layer 5a, layer 5b, layer 6) of M1 to identify hotspots of demyelination, iron and calcium accumulation in different cortical fields. We show preserved mean cortical thickness and layer architectures of M1, despite significantly increased iron in layer 6 and significantly increased calcium in layer 5a and superficial layer, in patients compared to controls. The behaviorally first-affected cortical field shows significantly increased iron in L6 compared to other fields, while calcium accumulation is atopographic and significantly increased in the low-myelin borders between cortical fields compared to the fields themselves. A subset of patients with longitudinal data shows that the low-myelin borders are particularly disrupted, and that calcium hotspots but to a lesser extent iron hotspots precede demyelination. Finally, we highlight that a very-slow progressing patient (P4) shows a distinct pathology profile compared to the other patients. Our data shows that layer-specific markers of in-vivo pathology can be identified in ALS-patients with a single 7T-MRI measurement after first diagnosis, and that such data provide critical insights into the individual disease state. Our data highlight the non-topographic architecture of ALS disease spread, and the role of calcium rather than iron accumulation in predicting future demyelination. We also highlight a potentially important role of low-myelin borders, that are known to connect to multiple areas within the M1 architecture, in disease spread. Importantly, the distinct pathology profile of a very-slow progressing patient (P4) highlights a distinction between disease duration and pathology progression. Our findings demonstrate the importance of in-vivo histology for the diagnosis and prognosis of neurodegenerative diseases such as ALS.

Amyotrophic lateral sclerosis (ALS) is a devastating neurodegenerative disease characterized by the degeneration of upper and lower motor neurons. A major neuropathological finding in ALS is the coexistence of glial activation and aggregation of the phosphorylated transactive response DNA-binding protein 43-kDa (pTDP43) in the motor cortex at the earliest stages of the disease. Despite this, the transcriptional alterations associated with these pathological changes in this major vulnerable brain region have yet to be fully characterized. Here, we have performed massive RNA sequencing of the motor cortex of ALS (n=11) and healthy controls (HC; n=8). We report extensive RNA expression alterations at gene and isoform levels, characterized by the enrichment of neuroinflammatory and synapse related pathways. The assembly of gene co-expression modules confirmed the involvement of these two principal transcriptomic changes, and showed a strong negative correlation between them. Furthermore, cell-type deconvolution using human single-nucleus RNA sequencing data as reference demonstrated that microglial cells are overrepresented in ALS compared to HC. Importantly, we also show for the first time in the human ALS motor cortex, that microgliosis is mostly driven by the increased proportion of a microglial subpopulation characterized by gene markers overlapping with the recently described disease associated microglia (DAM). Using immunohistochemistry, we further evidenced that this microglial subpopulation is overrepresented in ALS and that variability in pTDP43 aggregation among patients negatively correlates with the proportion of microglial cells. In conclusion, we report that neuroinflammatory changes in ALS motor cortex are dominated by microglia which is concomitant with a reduced expression of postsynaptic transcripts, in which DAM might have a prominent role. Microgliosis therefore represents a promising avenue for therapeutic intervention in ALS.

Motor neuron (MN) loss is a hallmark of neurodegenerative disorders, yet its assessment remains variable, confounding mechanistic and therapeutic interpretation. To address this, we conducted a systematic review and meta-analysis of spinal muscular atrophy (SMA) mouse studies, revealing 60% variability in reported MN loss, largely attributable to nonspecific spinal cord sampling. Using a whole-segment approach with tissue clearing, MN tracing, and multimodal imaging, we confirmed segment-dependent differences in MN counts. Common MN markers (SMI-32, Nissl) lacked specificity, whereas choline acetyltransferase (ChAT) provided robust labeling in murine and human spinal cords. Deep learning-based whole-mount segmentation enabled unbiased MN quantification and validated manual counts. Integrating analysis with computational modeling established segment sampling as a key driver of variability and revealed degeneration patterns: widespread MN loss in amyotrophic lateral sclerosis (ALS), selective MN loss in severe SMA, and preservation in mild SMA models. These findings establish a framework for reproducible MN quantification. HighlightsO_LISpinal cord segment-specific analysis reduces variability and allows accurate MN quantification C_LIO_LIChAT is the most reliable MN marker in murine and human spinal cords C_LIO_LIDeep learning-based segmentation enables unbiased MN quantification in intact spinal cords C_LIO_LIMN degeneration is widespread in ALS but restricted to pools innervating proximal muscles in severe SMA C_LI

The pathologies of oculopharyngeal muscular dystrophy (OPMD) and oculopharyngodistal myopathy (OPDM) are indistinguishable. We found that p62-positive intra-nuclear inclusions (INIs) in myonuclei (myo-INIs) were significantly more frequent in OPMD (11.4 {+/-} 4.1%, range 5.0- 17.5%) than in OPDM and other rimmed vacuolar myopathies (RVMs) (1-2% on average, range 0.0-3.5%, p<0.0001). In contrast, INIs in nonmuscle cells (nonmuscle-INIs) were present in OPDM, but absent in other RVMs, including OPMD. These results indicate that OPMD can be differentiated from OPDM and other RVMs by the frequent presence of myo-INIs ([&ge;]5%) and the absence of nonmuscle-INIs in muscle pathology.

ObjectiveNeurodegenerative diseases such as Parkinsons disease (PD), Alzheimers disease (AD) and amyotrophic lateral sclerosis (ALS) are traditionally considered strictly neurological disorders. However, clinical presentation is not restricted to neurological systems, and non-central nervous system (CNS) manifestations, particularly gastrointestinal (GI) symptoms, are common. Our objective was to understand the systemic distribution of pathology in archived non-CNS tissues, taken as part of routine clinical practice during life from people with ALS. DesignWe requested all surgical specimens of non-CNS tissue taken during life from 48 people with ALS, for whom evidence of the characteristic proteinopathy associated with ALS had been identified in the CNS after death (i.e., the pathological cytoplasmic accumulation of phosphorylated TDP-43 (pTDP-43) aggregates). Of the 48 patients, 13 had sufficient tissue for evaluation: 12 patients with sporadic ALS and 1 patient with a C9orf72 hexanucleotide repeat expansion. The final cohort consisted of 68 formalin-fixed paraffin embedded tissue samples from 22 surgical cases (some patients having more than one case over their lifetimes), representing 8 organ systems, which we examined for evidence of pTDP-43 pathology. The median age of tissue removal was 62.4 years old and median tissue removal to death was 6.3 years. ResultsWe identified pTDP-43 aggregates in multiple cell types of the GI tract (i.e., colon and gallbladder), including macrophages and dendritic cells within the lamina propria, as well as neuronal and glial cells of the myenteric plexus. Aggregates were also noted within lymph node parenchyma, blood vessel endothelial cells, and chondrocytes. We note that in all cases with non-CNS pTDP-43 pathology, aggregates were present prior to ALS diagnosis (median=3years) and, in some instances, preceded neurological symptom onset by more than 10 years. ConclusionThese data imply that patients with non-CNS symptoms may have occult protein aggregation that could be detected many years prior to neurological involvement. SummaryNeurodegenerative diseases such as Parkinsons disease (PD), Alzheimers disease (AD) and amyotrophic lateral sclerosis (ALS) are traditionally considered strictly neurological disorders. However, clinical presentation is not restricted to neurological systems, and non-central nervous system (CNS) manifestations, particularly gastrointestinal (GI) symptoms, are common. Our objective was to understand the systemic distribution of TDP-43 pathology in archived non-CNS tissues, taken as part of routine clinical practice during life from people with ALS. We identified pTDP-43 aggregates in multiple cell types of the GI tract (i.e., colon and gallbladder), and within lymph node parenchyma, blood vessel endothelial cells, and chondrocytes. We note that in all cases with non-CNS pTDP-43 pathology, aggregates were present prior to ALS diagnosis (median=24months) and, in some instances, preceded neurological symptom onset by more than 10years. These data imply that patients with non-CNS symptoms may have occult protein aggregation tha could be detected many years prior to neurological involvement. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=159 SRC="FIGDIR/small/484805v1_ufig1.gif" ALT="Figure 1"> View larger version (67K): org.highwire.dtl.DTLVardef@1f568c6org.highwire.dtl.DTLVardef@b1d99dorg.highwire.dtl.DTLVardef@45e643org.highwire.dtl.DTLVardef@1106d29_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOGraphical Abstract.C_FLOATNO Ante-mortem tissue cohort comprised of tissue taken from people with ALS demonstrates non-CNS accumulation of pTDP-43 aggregates prior to symptom onset. Schematic of workflow to identify pTDP-43 aggregates indicative of non-central nervous system (CNS) manifestations of ALS. Lower panel left: cartoon depicting organs and cell types that had evidence of pTDP-43 aggregation in ALS patient non-CNS ante-mortem tissue. Lower panel right: cartoon depicting organs with no evidence of pTDP-43 aggregation in ALS patient non-CNS ante-mortem tissue. C_FIG

Alzheimers Disease (AD) is a neurodegenerative disorder with complex neuropathological features, such as phosphorylated tau (p-tau) positive neurofibrillary tangles (NFTs) and neuritic plaques (NPs). The quantitative evaluation of p-tau pathology is a key element for the diagnosis of AD and other tauopathies. Assessment of tauopathies relies on semi-quantitative analysis and does not consider lesions heterogeneity (e.g., load and density of NFTs vs NPs). In this study, we developed a deep learning-based workflow for automated annotation and segmentation of NPs and NFTs from AT8-immunostained whole slide images (WSIs) of AD brain sections. Fifteen WSIs of frontal cortex from four biobanks with different tissue quality, staining intensity and scanning formats were used for the present study. We first applied an artificial intelligence (AI-)-driven iterative procedure to improve the generation of pathologist validated training datasets for NPs and NFTs. This procedure increased the annotation quality by more than 50%, especially for NPs when present in high density. Using this procedure, we obtained an expert validated annotation database with 5013 NPs and 5143 NFTs. As a second step, we trained two U-Net convolutional neural networks (CNNs) for accurate detection and segmentation of NPs or NFTs. The workflow achieved a high accuracy and consistency, with a mean Dice similarity coefficient of 0.81 for NPs and 0.77 for NFTs. The workflow also showed good generalization performance across different patients with different staining and tissue quality. Our study demonstrates that artificial intelligence can be used to correct and enhance annotation quality especially for complex objects, even when intermingled and present in high density, in brain tissue. Furthermore, the expert validated databases allowed to generate highly accurate models for segmenting discrete brain lesions using a commercial software. Our annotation database will be publicly available to facilitate human digital pathology applied to AD.

Polymyositis with mitochondrial pathology (PM-Mito) was first identified in 1997 as a subtype of idiopathic inflammatory myopathy. Recent findings demonstrated significant molecular similarities between PM-Mito and Inclusion Body Myositis (IBM), suggesting a trajectory from early to late IBM and prompting the inclusion of PM-Mito as an IBM precursor (early IBM) within the IBM spectrum. Both PM-Mito and IBM show mitochondrial abnormalities, suggesting mitochondrial disturbance is a critical element of IBM pathogenesis. The primary objective of this cross-sectional study was to characterize the mitochondrial phenotype in PM-Mito at histological, ultrastructural, and molecular levels and to study the interplay between mitochondrial dysfunction and inflammation. Skeletal muscle biopsies of 27 patients with PM-Mito and 27 with typical IBM were included for morphological and ultrastructural analysis. Mitochondrial DNA (mtDNA) copy number and deletions were assessed by qPCR and long-range PCR, respectively. In addition, full-length single-molecule sequencing of the mtDNA enabled precise mapping of deletions. Protein and RNA levels were studied using unbiased proteomic profiling, immunoblotting, and bulk RNA sequencing. Cell-free mtDNA (cfmtDNA) was measured in the serum of IBM patients. We found widespread mitochondrial abnormalities in both PM-Mito and IBM, illustrated by elevated numbers of COX-negative and SDH-positive fibers and prominent ultrastructural abnormalities with disorganized and concentric cristae within enlarged and dysmorphic mitochondria. MtDNA copy numbers were significantly reduced, and multiple large-scale mtDNA deletions were already evident in PM-Mito, compared to healthy age-matched controls, similar to the IBM group. The activation of the canonical cGAS/STING inflammatory pathway, possibly triggered by the intracellular leakage of mitochondrial DNA, was evident in PM-Mito and IBM. Elevated levels of circulating cfmtDNA also indicated leakage of mtDNA as a likely inflammatory trigger. In PM-Mito and IBM, these findings were accompanied by dysregulation of proteins and transcripts linked to the mitochondrial membranes. In summary, we identified that mitochondrial dysfunction with multiple mtDNA deletions and depletion, disturbed mitochondrial ultrastructure, and defects of the inner mitochondrial membrane are features of PM-Mito and IBM, underlining the concept of an IBM-spectrum disease (IBM-SD). The activation of inflammatory pathways related to mtDNA release indicates a significant role of mitochondria-associated inflammation in the pathogenesis of IBM-SD. Thus, mitochondrial abnormalities precede tissue remodeling and infiltration by specific T-cell subpopulations (e.g., KLRG1+) characteristic of late IBM. This study highlights the critical role of early mitochondrial abnormalities in the pathomechanism of IBM, which may lead to new approaches to therapy.

ObjectiveTo evaluate the diagnostic performance of serum neurofilament light chain (sNfL) and cardiac troponin T (cTnT) as biomarkers for amyotrophic lateral sclerosis (ALS) and to determine whether their combination improves diagnostic accuracy. MethodsWe retrospectively analyzed 293 ALS patients, 47 neurodegenerative disease controls and 24 healthy controls. An independent validation cohort of 501 ALS patients was additionally analyzed to confirm reproducibility of the results. Receiver operating characteristic (ROC) curve analysis was performed for sNfL, cTnT and their combination, and the area under the curve (AUC) was compared across groups. An ALS-specific cTnT cut-off of 8.35 ng/L was determined using the Youden index and applied in subgroup analyses, in which biomarker-negative ALS patients (normal sNfL and cTnT) were compared to biomarker-positive patients regarding disease duration and progression rate. ResultssNfL alone showed excellent performance in discriminating ALS patients from healthy controls (AUC = 0.951), but only moderate performance in discriminating neurodegenerative disease controls (AUC = 0.789). Combining sNfL and cTnT improved diagnostic accuracy for ALS over neurodegenerative disease controls, with a combined AUC of 0.866. Similar AUCs were observed in the validation cohort. Biomarker-negative ALS patients had a longer disease duration (73.0 vs. 18.0 months, p=0.0003) and a lower progression rate (0.19 vs. 0.70 points per months, p<0.0001) than biomarker-positive patients. InterpretationWhile sNfL alone performs well in distinguishing ALS from healthy controls, cTnT provides additional value in distinguishing ALS from disease controls. The combination of sNfL and cTnT improves diagnostic accuracy and may help identify clinically distinct ALS subgroups.

ObjectiveMotor neuron disease (MND) is a debilitating neurodegenerative disease with profound unmet need. In pre-symptomatic mutation carriers, elevations in neurofilament light (NfL) precede symptom onset, however, the presence and timing of elevation is much more difficult to study in sporadic cases. MethodsUsing the UK Biobank cohort, we tested whether plasma NfL predicted risk of diagnosis of sporadic MND using survival analysis. ResultsWe identified 241 MND patients with pre-diagnosis NfL data, of which 203 (84%) lacked predicted loss of function or deleterious missense variants in established ALS genes. A total of 42,752 controls without MND were obtained from a random sample of UK Biobank participants. At two years pre-diagnosis, we found that NfL levels in patients exceeded the 95th percentile of controls and that patients could be discriminated from controls at high accuracy (AUC = 0.95 (95% CI 0.89-1.01)). In participants with hospital record follow-up after study enrollment, a 2-fold increase in NfL levels was associated with a 3.4 fold risk of receiving an MND diagnosis per year (95% CI 2.9-3.9, P = 4 x 10-64)). DiscussionOur findings show that NfL can identify sporadic MND as early as 2 years prior to diagnosis.

Amyotrophic lateral sclerosis (ALS) is a neurogenerative disease resulting from progressive degeneration of motor neurons leading to systemic consequences. Despite being the most common motor neuron disease, with increasing global prevalence, limited treatment options exist. Emerging evidence from genetic studies and pathology analyses implicates RNA dysregulation in ALS pathogenesis, however, deep, comprehensive RNA sequencing studies have not been carried out. Here, we analysed >240 ALS and control whole blood transcriptome samples. Cross-sectional (Ncases=121, Ncontrols=53) and longitudinal (Nobservations=103) cohorts supported complementary expression analyses of disease mechanisms across disease stages. Both short (N=241) and long-read (N=16) technologies were utilised to discover splicing changes. Total RNA was extracted from PAXgene whole blood RNA tubes before libraries (Illumina Stranded Total RNA RiboZero Plus) were prepared and sequenced ([~]50M PE reads per sample). Long-read sequencing was performed using the Mas-Seq protocol with Kinnex full-length RNA prep kit and sequenced (PacBio Revio platform, 10M reads per sample) for full-length transcripts. Case-control cohort analyses identified 50 significantly differentially expressed genes, with pathway analyses implicating RNA processing and immune system regulation. Findings were corroborated using existing ALS RNAseq datasets from blood (correlation >0.4), iPSC-MN and post-mortem tissues. Alternative splicing (AS) analyses (LeafCutter) identified 62 clusters. Within-case analyses involved ALS cases with multiple (2-4) visits, detected 144 genes associated with disability progression over time. The long-read sequencing (Ncases=8, Ncontrols=8) provided novel discovery insights, in particular in the HLA region. This comprehensive blood-based transcriptomic dataset reveals both known and novel disease mechanisms in ALS, offering valuable insights that could inform future research and therapeutic development. The results of this study may inform and refine the prioritization of candidate genes and loci in future ALS research.

ATTR amyloidosis is a systemic disease characterized by the deposition of amyloid fibrils made of transthyretin, a protein integral to transporting retinol and thyroid hormones. Transthyretin is primarily produced by the liver and circulates in blood as a tetramer. The retinal epithelium also secretes transthyretin, which is secreted to the vitreous humor of the eye. Because of mutations or aging, transthyretin can dissociate into amyloidogenic monomers triggering amyloid fibril formation. The deposition of transthyretin amyloid fibrils in the myocardium and peripheral nerves causes cardiomyopathies and neuropathies, respectively. Using cryo-electron microscopy, here we determined the structures of amyloid fibrils extracted from cardiac and nerve tissues of an ATTRv-V30M patient. We found that fibrils from both tissues share a consistent structural conformation, similar to the previously described structure of cardiac fibrils from an individual with the same genotype, but different from the fibril structure obtained from the vitreous humor. Our study hints to a uniform fibrillar architecture across different tissues within the same individual, only when the source of transthyretin is the liver. Moreover, this study provides the first description of ATTR fibrils from the nerves of a patient and enhances our understanding of the role of deposition site and protein production site in shaping the fibril structure in ATTRv-V30M amyloidosis.

BackgroundAdvances in our technological capacity to interrogate brain tumour biology has led to the ever-increasing use of genomic sequencing in routine diagnostic decision making. Presently, brain tumours are routinely classified based on their epigenetic signatures, leading to a paradigm shift in diagnostic pathways. Such testing can be performed so rapidly using nanopore sequencing that results can be provided intraoperatively. This information greatly improves upon the fidelity of smear diagnosis and can help surgeons tailor their approach, balancing the risks of surgery with the likely benefit. Nevertheless, full integrated diagnosis may require subsequent additional assays to detect pathognomonic somatic mutations and structural variants, thereby delaying the time to final diagnosis. MethodsHere, we present ROBIN, a tool based upon PromethION nanopore sequencing technology that can provide both real-time, intraoperative methylome classification and next-day comprehensive molecular profiling within a single assay. ROBIN uniquely integrates three methylation classifiers 1-3 to improve diagnostic performance in the intraoperative setting. FindingsWe demonstrate classifier performance on 50 prospective intraoperative cases, achieving a diagnostic turnaround time under 2 hours and generating robust tumour classifications within minutes of sequencing. Furthermore, ROBIN can detect single nucleotide variants (SNVs), copy number variants (CNVs) and structural variants (SVs) in real-time, and is able to inform a complete integrated diagnosis within 24 hours. Classifier performance demonstrated concordance with final integrated diagnosis in 90% of prospective cases. InterpretationNanopore sequencing can greatly improve upon the turnaround times for standard of care diagnostic testing, including sequencing, and is furthermore able to reliably provide clinically actionable intraoperative tumour classification. FundingThe Jean-Shanks Foundation, the Pathological Society of Great Britain and Ireland, the British Neuropathological Society, and the Wellcome Trust.

Many proteins associated with amyotrophic lateral sclerosis (ALS) remain poorly characterized, in part because validated reagents for protein-level studies are scarce. We previously established knockout (KO)-based antibody characterization workflows and showed that widely used antibodies against the ALS-associated protein C9orf72 lacked specificity (Laflamme et al., 2019), and subsequently scaled this framework to systematically benchmark research antibodies, revealing that up to 61% do not perform as recommended by manufacturers (Ayoubi et al., 2023). Here, we extend this approach by establishing the ALS-Reproducible Antibody Platform (ALS-RAP) to evaluate antibodies against proteins encoded by ALS risk genes. We characterized 303 antibodies targeting 33 ALS-associated proteins using KO-based antibody characterization workflows to identify high-quality reagents for common experimental applications. Using validated antibodies, we profiled protein levels across human induced pluripotent stem cell (iPSC)-derived and primary neurological cell types, revealing diverse cellular distributions and higher protein levels for several ALS-associated proteins in glial and immune populations. Together, ALS-RAP provides a validated antibody toolbox and protein expression resource for studying ALS-associated proteins, supporting the view that ALS genetics converges on multicellular disease mechanisms involving both neuronal and glial populations.

BackgroundAmyotrophic lateral sclerosis (ALS) is a fatal heterogeneous neurodegenerative disease that typically leads to death from respiratory failure within two to five years. Despite the identification of several genetic risk factors, the biological processes involved in ALS pathogenesis remain poorly understood. The motor cortex is an ideal region to study dysregulated pathological processes in ALS as it is affected from the earliest stages of the disease. In this study, we investigated motor-cortex gene expression of cases and controls to gain new insight into the molecular footprint of ALS. MethodsWe performed a large case-control differential expression analysis of two independent post-mortem motor cortex bulk RNA-sequencing (RNAseq) datasets from the Kings College London BrainBank (N = 171) and TargetALS (N = 132). Differentially expressed genes from both datasets were subjected to gene and pathway enrichment analysis. Genes common to both datasets were also reviewed for their involvement with known mechanisms of ALS pathogenesis to identify potential candidate genes. Finally, we performed a correlation analysis of genes implicated in pathways enriched in both datasets with clinical outcomes such as the age of onset and survival. ResultsDifferential expression analysis identified 2,290 and 402 differentially expressed genes in KCL BrainBank and TargetALS cases, respectively. Enrichment analysis revealed significant synapse-related processes in the KCL BrainBank dataset, while the TargetALS dataset carried an immune system-related signature. There were 44 differentially expressed genes which were common to both datasets, which represented previously recognised mechanisms of ALS pathogenesis, such as lipid metabolism, mitochondrial energy homeostasis and neurovascular unit dysfunction. Differentially expressed genes in both datasets were significantly enriched for the neuropeptide signalling pathway. By looking at the relationship between the expression of neuropeptides and their receptors with clinical measures, we found that in both datasets NPBWR1, TAC3 and SSTR1 correlated with age of onset, and GNRH1, TACR1 with survival. We provide access to gene-level expression results to the broader research community through a publicly available web application (https://alsgeexplorer.er.kcl.ac.uk). ConclusionThis study identified motor-cortex specific pathways altered in ALS patients, potential molecular targets for therapeutic disease intervention and a set of neuropeptides and receptors for investigation as potential biomarkers.

BackgroundThe disease entity known as amyotrophic lateral sclerosis (ALS) is now known to represent a collection of overlapping syndromes. A better understanding of this heterogeneity and the ability to distinguish ALS subtypes would improve the clinical care of patients and enhance our understanding of the disease. Subtype profiles could be incorporated into the clinical trial design to improve our ability to detect a therapeutic effect. A variety of classification systems have been proposed over the years based on empirical observations, but it is unclear to what extent they genuinely reflect ALS population substructure. MethodsWe applied machine learning algorithms to a prospective, population-based cohort consisting of 2,858 Italian patients diagnosed with ALS for whom detailed clinical phenotype data were available. We replicated our findings in an independent population-based cohort of 1,097 Italian ALS patients. FindingsWe found that semi-supervised machine learning based on UMAP applied to the output of a multi-layered perceptron neural network produced the optimum clustering of the ALS patients in the discovery cohort. These clusters roughly corresponded to the six clinical subtypes defined by the Chio classification system (bulbar ALS, respiratory ALS, flail arm ALS, classical ALS, pyramidal ALS, and flail leg ALS). The same clusters were identified in the replication cohort. A supervised learning approach based on ensemble learning identified twelve clinical parameters that predicted ALS clinical subtype with high accuracy (area under the curve = 0{middle dot}94). InterpretationOur data-driven study provides insight into the ALS populations substructure and demonstrates that the Chio classification system robustly identifies ALS subtypes. We provide an interactive website (https://share.streamlit.io/anant-dadu/machinelearningforals/main) so that clinical researchers can predict the clinical subtype of an ALS patient based on a small number of clinical parameters. FundingNational Institute on Aging and the Italian Ministry of Health. RESEARCH IN CONTEXTO_ST_ABSEvidence before this studyC_ST_ABSWe searched PubMed for articles published in English from database inception until January 5, 2021, about the use of machine learning and the identification of clinical subtypes within the amyotrophic lateral sclerosis (ALS) population, using the search terms "machine learning", AND "classification", AND "amyotrophic lateral sclerosis". This inquiry identified twenty-nine studies. Most previous studies used machine learning to diagnose ALS (based on gait, imaging, electromyography, gene expression, proteomic, and metabolomic data) or improve brain-computer interfaces. One study used machine learning algorithms to stratify ALS postmortem cortex samples into molecular subtypes based on transcriptome data. Kueffner and colleagues crowdsourced the development of machine learning algorithms to approximately thirty teams to obtain a consensus in an attempt to identify ALS patients subpopulation. In addition to clinical trial information in the PRO-ACT database (www.ALSdatabase.org), this effort used data from the Piedmont and Valle dAosta Registry for ALS (PARALS). Four ALS patient categories were identified: slow progressing, fast progressing, early stage, and late stage. This approachs clinical relevance was unclear, as all ALS patients will necessarily pass through an early and late stage of the disease. Furthermore, no attempt was made to discern which of the existing clinical classification systems, such as the El Escorial criteria, the Chio classification system, and the Kings clinical staging system, can identify ALS subtypes. We concluded that there remained an unmet need to identify the ALS populations substructure in a data-driven, non-empirical manner. Building on this, there was a need for a tool that reliably predicts the clinical subtype of an ALS patient. This knowledge would improve our understanding of the clinical heterogeneity associated with this fatal neurodegenerative disease. Added value of this studyThis study developed a machine learning algorithm to detect ALS patients clinical subtypes using clinical data collected from the 2,858 Italian ALS patients in PARALS. Ascertainment of these patients within the catchment area was near complete, meaning that the dataset truly represented the ALS population. We replicated our approach using clinical data obtained from an independent cohort of 1,097 Italian ALS patients that had also been collected in a population-based, longitudinal manner. Semi-supervised learning based on Uniform Manifold Approximation and Projection (UMAP) applied to a multilayer perceptron neural network provided the optimum results based on visual inspection. The observed clusters equated to the six clinical subtypes previously defined by the Chio classification system (bulbar ALS, respiratory ALS, flail arm ALS, classical ALS, pyramidal ALS, and flail leg ALS). Using a small number of clinical parameters, an ensemble learning approach could predict the ALS clinical subtype with high accuracy (area under the curve = 0{middle dot}94). Implications of all available evidenceAdditional validation is required to determine these algorithms accuracy and clinical utility in assigning clinical subtypes. Nevertheless, our algorithms offer a broad insight into the clinical heterogeneity of ALS and help to determine the actual subtypes of disease that exist within this fatal neurodegenerative syndrome. The systematic identification of ALS subtypes will improve clinical care and clinical trial design.

Tofersen is the first effective and approved therapy for familial ALS caused by pathogenic variants in the SOD1 gene. Following treatment with tofersen, neurofilaments in patients CSF and serum display a faster response than clinical parameters, underlining their importance as a biomarker for treatment response in clinical trials. This evidence led us to hypothesize that this novel treatment might represent an opportunity to identify additional therapy-responsive biomarkers for ALS. We chose the commercial NUcleic acid Linked Immuno-Sandwich Assay (NULISA), to investigate a predefined panel of 120 neural, glial and inflammatory markers in CSF and serum samples longitudinally collected from SOD1-ALS patients at baseline and three months after tofersen treatment. We identified a set of proteins (beyond pNfH and NfL) whose levels differed between SOD1-ALS and the matched control group and that were responsive to treatment with tofersen, including A{beta}42, NPY and UCHL1. Even though our results warrant validation in larger cohorts and longer follow-up time, they may pave the way for a panel of responsive proteins solidifying biomarker endpoints in clinical trials.