Normal pressure hydrocephalus is a common cause of gait and cognitive impairment in older adults, marked by excessive CSF accumulation. Genetic studies suggest impaired fluid clearance, and clinical symptoms can improve after CSF diversion. However, no fluid biomarkers exist to explore CSF accumulation mechanisms, assist diagnosis, or predict response to treatment. Stable isotope labelling kinetics is a clinical research tool that uses non-radioactive isotopes to label newly translated proteins, enabling measurement of their appearance (synthesis) and disappearance (clearance) in compartments like CSF. This study aimed to develop a novel method to capture protein turnover in CSF and assess whether clearance disruption is evident in normal pressure hydrocephalus with extended follow-up. Proteins of interest were identified via mass spectrometry in human CSF and choroid plexus organoid-derived CSF-like fluid. Protein origin and synthesis rates were evaluated by labelling organoids with 13C6-leucine. Label incorporation was measured using targeted mass spectrometry to determine the ratio of labelled to unlabelled peptide. A proof-of-concept case-control study was then conducted in specialist neuroscience centres. Participants received intravenous 13C6-leucine and underwent serial CSF withdrawal via lumbar drain, with matched blood sampling for up to 72 h. Patients undergoing CSF drainage and controls were recruited sequentially. Targeted mass spectrometry was used to determine protein production and clearance rates. To determine the clinical relevance of these protein turnover rates to CSF flow, they were correlated with direct measurements of CSF production captured using a LiquoGuard machine linked to the lumbar CSF drain. We captured choroid plexus protein kinetics in human organoids and the CSF of participants undergoing CSF drainage (n = 10) or controls (ventricular CSF n = 4; lumbar CSF n = 5). The case and control cohorts varied in sex (NPH = 80% male and controls = 22% male) and in age. There was no significant age difference between NPH and the lumbar control cohort (n = 5) (NPH: 75 (71-78) versus 70 (63-84) years old; P = 0.2438). We found that transthyretin is abundantly secreted by choroid plexus organoids, and observed correlations with CSF transthyretin synthesis rates and volume of CSF production in vivo (P = 0.738; P < 0.05). Clearance rates of transthyretin are ∼10 fold slower in normal pressure hydrocephalus compared with controls, suggesting impaired CSF protein clearance. This method is a novel clinical tool for interrogating CSF protein dynamics and may have utility in tracking CSF flow clinically.

Alzheimer’s disease (AD) is one of the most common age-related causes of death, with limited effective disease-modifying treatments. Although metformin shows promise as a disease-modifying agent for AD, its molecular mechanism—specifically how it confers neuroprotection despite potentially increasing amyloid-beta (Aβ) load—remains obscure. Here, we demonstrate that metformin functions as a pharmacological activator of the ubiquitin-binding protease rngo/DDI2. Through a genetic screen in Drosophila , we identified rngo/DDI2 as a potent suppressor of Aβ toxicity. We provide in silico and genetic evidence that metformin interacts with the conserved D257 residue of the rngo/DDI2 RVP domain, inducing homodimerisation and subsequent protein stabilisation. This activation boosts proteasome activity in the presence of Aβ preferentially clearing highly abundant proteins to preserve proteostasis. Crucially, this intervention is broadly effective; rngo/DDI2 upregulation robustly suppresses toxicity in models of TDP-43 and C9orf72-repeat expansion pathology, indicating a generalised mechanism of neuroprotection. Supported by human iPSC data and evidence of DDI2 depletion in AD patient brains, our results identify rngo/DDI2 as a conserved regulator of neuronal resilience. We propose that directly targeting DDI2 stabilisation represents a novel, broadly applicable therapeutic strategy to counteract proteotoxic stress across a broad spectrum of neurodegenerative diseases.

Abstract γ-secretase is a multi-subunit enzyme complex responsible for cleaving hundreds of substrates in diverse cellular contexts. Variation in subunit composition - including the use of alternate catalytic subunits Presenilin 1 (PSEN1) and Presenilin 2 (PSEN2) - results in diverse γ-secretase complexes. Point mutations in PSEN1 and PSEN2 cause familial forms of Alzheimer’s disease, while loss-of-function mutations in the γ-secretase subunits PSEN1, PSENEN and NCSTN cause acne inversa. To advance therapeutic strategies targeting γ-secretase in Alzheimer’s disease, a better understanding of individual γ-secretase complexes is required. In this study, we used CRISPR-Cas9 genome engineering to generate PSEN2-knockout iPSCs in order to compare the consequence of PSEN2 knockout versus PSEN1 knockout in iPSC-derived brain cells. In contrast to PSEN1-knockout, PSEN2-knockout did not alter APP cleavage or Aβ generation in iPSC-neurons, nor did it disrupt Nicastrin maturation. Similarly, PSEN2-knockout had little impact on TREM2 processing in iPSC-microglia. Instead, our data indicate that loss of PSEN2 primarily impacts the endo-lysosomal system in iPSC-neurons, causing an accumulation of early endosome markers and a reduction in lysosomal markers – phenotypes not observed in PSEN1-knockout neurons. Taken together, these findings highlight distinct and non-redundant functions of PSEN1 and PSEN2 in human brain cells, reinforcing findings in animal models and subcellular localisation studies. This work advances our understanding of distinct γ-secretase complex functions and provides insights that will support future therapeutic efforts to inhibit, modulate or stabilise γ-secretase.

Alzheimer's disease symptoms include gradual cognitive decline and memory loss that is correlated with progressive loss of neuronal connections due to an imbalance of excitatory and inhibitory synaptic functions. These have been shown in various rodent models but direct measurements of excitatory-inhibitory changes have yet to be performed in human neurons. Therefore, our project aims to construct a human-induced pluripotent stem cell co-culture model representing important brain circuitry which captures synaptic dysfunction. Familial Alzheimer's disease patient induced pluripotent stem cells carrying mutant APP V717I and their isogenic controls were differentiated into cortical glutamatergic neurons and astrocytes using dual-SMAD inhibition followed by in vitro corticogenesis. Building upon this, we differentiated inhibitory interneurons expressing parvalbumin and somatostatin via ventral patterning with sonic hedgehog activation. Then, we co-cultured these cells with differentiated cortical neurons and astrocytes. The properties of the co-culture model were validated using immunohistochemistry, confocal microscopy combined with electrophysiological whole-cell recordings. Confocal microscopy validated the presence of excitatory cortical neurons, astrocytes, and two inhibitory interneuron types, parvalbumin and somatostatin expressing interneurons within the co-culture. Whole-cell recordings revealed intrinsic membrane properties from individual excitatory and inhibitory neurons in this co-culture from day 70 onwards. Spontaneous synaptic activity recorded from the APP V717I-induced pluripotent stem cell model showed synaptic hyperexcitability correlated with altered morphological changes, which was expected in contrast to the isogenic control co-culture. Our novel co-culture model, including astrocytes, excitatory and inhibitory neurons, represents a strong model of brain circuitry in Alzheimer's disease. These models will enable investigation of Alzheimer's disease causative mutations on neuronal connectivity in human neurons allowing for confirmation of network dysfunction in Alzheimer's disease in human neurons. It also has the potential of becoming a valuable preclinical tool to screen novel targeted therapies. This is a methods paper validating a human-induced pluripotent stem cell-derived excitatory-inhibitory neuron-astrocyte co culture with electrophysiological readouts and immunostaining, focusing on Alzheimer's disease relevant network physiology.

Changes to the relative abundance of amyloid-beta (Aβ) peptides are hallmarks of Alzheimer's disease. Induced pluripotent stem cell (iPSC)-derived neurons offer a physiological model of Aβ production. We employed unbiased, data-driven analyses to investigate combinations of Aβ peptides as Alzheimer's disease biomarkers and the relative contribution of peptides to Alzheimer's disease pathogenesis. We measured Aβ37, Aβ38, Aβ40, Aβ42 and Aβ43 in 10 iPSC-neuronal cultures from PSEN1 mutation carriers. We combined these data with published cell model data and used linear weighted combinations to (i) distinguish Alzheimer's disease from controls, and (ii) predict age-at-onset for PSEN1 mutations. Data-driven approaches distinguished Aβ42 and Aβ43 from shorter peptides, providing unbiased evidence for a greater association of Aβ42 and Aβ43 to disease pathogenesis, compared with shorter peptides (Aβ37, Aβ38 and Aβ40). Weighted linear combinations of Aβ peptides outperform Aβ42/40 and provide insights into relative peptide contribution as biomarkers. A representative weighted composite value ratio (wCVR) derived from all data, balancing both disease classification and age-at-onset prediction, was ( 21 ⋅ A β 37 + 10 ⋅ A β 38 + 69 ⋅ A β 40 ) /   ( 94 ⋅ A β 42 + 6 ⋅ A β 43 ) . This work suggests a practical non-parametric harmonization approach to employing Aβ ratios as biomarkers for Alzheimer's disease, from multiple sites and assays. Building on this foundation, we applied a new model using weighted composite value ratios, which outperform existing biomarkers across all tasks. This underscores the value of integrating multiple peptides and assigning optimized weightings. The study confirms the association of Aβ42 and Aβ43 with Alzheimer's disease pathogenesis in a data-driven manner. Peptide weights further provide mechanistic insights into the relative contribution of each peptide to disease, such as a greater contribution of Aβ37 compared to Aβ38. The algorithm used herein can be further refined to improve biomarkers for Alzheimer's disease.

Parkinson's disease-associated proteins PINK1 and Parkin collaboratively regulate stress-induced mitophagy. While in vitro human neuronal cultures are valuable for studying the roles of PINK1 and Parkin in a disease-relevant context, the impact of culture conditions on these processes remains largely underexplored. Here, it is shown that human induced neurons (iNeurons) cultured in N2B27 and BrainPhys medium exhibit distinct PINK1-Parkin-dependent mitophagy phenotypes. Specifically, BrainPhys-cultured iNeurons show greater resistance to PINK1-dependent mitophagy initiation, linked to a reduction in glucose availability and reduced PINK1 protein availabilities, leading to decreases in stress-induced and basal mitophagy fluxes. These findings highlight the critical impact of culture conditions on mitophagy dynamics and emphasize the need to account for media-specific differences when using in vitro models to investigate mitophagy mechanisms in human neurons.

While induced pluripotent stem cells (iPSCs) have gained popularity in studying neurodegenerative diseases, the heterogeneity of stem cells used across studies impacts cross-study comparison. The iPSC Neurodegenerative Disease Initiative (iNDI) selected the KOLF2.1J cell line and prioritized its use as a reference standard for studying the effects of pathogenic variants on cell biology due to its stability and neutral neurodegenerative disease genetic risk. This cell line, and its derivatives expressing over 100 variants related to Alzheimer's disease, Parkinson's disease, and other neurological diseases, are available for academic and industry access. Current genomic data analyses are limited by the use of a human reference genome that does not capture the complete genetic background of a given iPSC line. While in the future this issue may be partially mitigated by the creation of a comprehensive human pangenome, previous work has shown that generating custom genomes is of value both to characterize the variation present and to serve as a more appropriate genomic reference. Here, we generated and characterized a custom complete genome assembly from KOLF2.1J. Mapping of sequencing reads to a personalized diploid assembly results in more comprehensive mapping compared to traditional linear references (i.e GRCh38). In addition, we provide a comprehensive custom gene annotation along with isoform expression and differential methylation analyses across multiple cell types. The assembly and all additional data is browsable and publicly available. This resource will enable more accurate investigation of the KOLF2.1J cell line and any genomics data generated compared to using traditional generalized references, while also serving as a foundational approach for establishing custom reference assemblies for other high-value iPSC lines.

Memory-based cognition relies on the integrity of cortico-hippocampal circuits, which are compromised in Alzheimer’s disease (AD) as β-amyloid (Aβ) and tau accumulate. However, the mechanisms linking this pathology to circuit dysfunction remain unclear. In mouse models, using in vivo two-photon and Neuropixels recordings, we show that Aβ-tau pathology promotes both region- and layer-specific impairments, involving reduced burst firing in superficial cortical layers and CA1 and reduced mean firing of excitatory and inhibitory neurons in deep cortical layers and CA1. Exposure to Aβ primed the susceptibility of neuronal populations to tau-induced impairment. Combined Aβ-tau reduced synaptic NMDA receptor (NMDAR) density in both mouse and human tissue, while Aβ-tau co-reduction restored NMDARs and firing patterns and improved contextual memory. NMDAR antagonism in healthy mice phenocopied regional and laminar deficits. Our findings implicate synaptic NMDAR hypofunction as a reversible mechanism linking Aβ-tau synergy to cortico-hippocampal dysfunction in AD.

During deep co-evolution of viruses and host cells, viruses have selected specific host cellular proteins redirected from physiological functions to viral needs, thereby disturbing cellular proteostasis and increasing the risk of triggering protein misfolding diseases (PMDs). Identifying virus-specific, repurposed host proteins also allows the study of fundamental cellular events in "sporadic" PMDs, independent of the virus. Here, we identify a small molecule with very strong activity against neurotropic herpes simplex virus 1 (HSV-1), modulating an allosteric site of macrophage migration inhibitory factor (MIF). The compound efficiently reduces both HSV-1-mediated and non-mediated tau phosphorylation or aggregation in vitro and in vivo. The lead compound, as well as conformation-sensitive antibodies, specifically interacts with an oxidized conformer of MIF (oxMIF) enriched in postmortem brain homogenates of patients with Alzheimer's disease (AD). OxMIF thus participates in a host-viral interface connecting HSV-1 infection, and possibly other external stressors, with tau cellular pathology characteristic for PMDs, including AD.

Protocol for cortical organoid generation in 96-well plates. Based on Pasca et al. 2015 (https://doi.org/10.1038/nmeth.3415) with some modifications.

INTRODUCTION:Africa, home to 1.4 billion people and the highest genetic diversity globally, harbors unique genetic variants crucial for understanding complex diseases like neurodegenerative disorders. However, African populations remain underrepresented in induced pluripotent stem cell (iPSC) collections, limiting the exploration of population-specific disease mechanisms and therapeutic discoveries. METHODS:To address this gap, we established an open-access African Somatic and Stem Cell Bank. RESULTS:In this initial phase, we generated 10 rigorously characterized iPSC lines from fibroblasts representing five Nigerian ethnic groups and both sexes. These lines underwent extensive profiling for pluripotency, genetic stability, differentiation potential, and Alzheimer's disease and Parkinson's disease risk variants. Clustered regularly interspaced palindromic repeats (CRISPR)/CRISPR-associated protein 9 technology was used to introduce frontotemporal dementia-associated MAPT mutations (P301L and R406W). DISCUSSION:This collection offers a renewable, genetically diverse resource to investigate disease pathogenicity in African populations, facilitating breakthroughs in neurodegenerative research, drug discovery, and regenerative medicine. HIGHLIGHTS:We established an open-access African Somatic and Stem Cell Bank. 10 induced pluripotent stem cell lines from five Nigerian ethnic groups were rigorously characterized. Clustered regularly interspaced palindromic repeats (CRISPR)/CRISPR-associated protein 9 technology was used to introduce frontotemporal dementia-causing MAPT mutations. The African Somatic and Stem Cell Bank is a renewable, genetically diverse resource for neurodegenerative research.

Neurofilament light protein (NfL) is a promising biomarker of neuronal injury and neurodegeneration. NfL levels in cerebrospinal fluid (CSF) and blood provide information about disease progression and are increasingly relied on as outcome measure in clinical trials. Understanding NfL kinetics in vivo is critical for interpreting NfL in response to new events where a steady state cannot be assumed, such as acute injury, disease onset or progression, or response to disease-modifying therapies. We infused human participants with diagnosed primary tauopathies (progressive supranuclear palsy, n = 5; corticobasal syndrome, n = 3; behavioural variant frontotemporal dementia, n = 2) with a stable isotope tracer ( 13 C 6 -leucine) and collected CSF by lumbar puncture at 4, 14, 20, 60 and 120 days post-labelling. In addition, post-mortem brain tissue from three participants who were infused with the tracer 18, 44 and 50 months earlier were homogenised and biochemically fractionated to separate the soluble and insoluble fraction. NfL was enriched in all samples via immunoprecipitation. The ratio of labelled to unlabelled NfL was measured monitoring proteotypic peptides using established targeted mass spectrometry workflows to quantitate the tracer-to-tracee ratio (TTR). Analysis of CSF detected low labelling of NfL (0.04 – 0.36% TTR) by 120 days that was comparable to the TTR levels detected in the soluble brain fraction. There was NfL present in the insoluble fraction (2.87% of the total NfL at 18-, 8.87% at 44- and 14.03% at 50-months post-mortem , averaged across several NfL peptides), with different relative abundances of NfL domains between these fractions. Interestingly, the 13 C 6 -labelled NfL signal detected in the insoluble fraction was more abundant, despite a higher recovery of total NfL in the soluble fraction. NfL turnover in vivo is remarkably slow as it is scarcely captured by 120 days post-labelling. The labelling results from brain tissue suggest that in these cases of primary tauopathies newly-synthesized NfL might be delocalized to a pool of protein of low solubility that does not contribute to NfL levels in CSF. Current experiments are addressing whether newly-synthesized NfL might be sequestered into pathological inclusions.

Mutations in PSEN1 cause familial Alzheimer’s disease with almost complete penetrance. Age at onset is highly variable between different PSEN1 mutations and even within families with the same mutation. Current research into late onset Alzheimer’s disease implicates inflammation in both disease onset and progression. PSEN1 is the catalytic subunit of γ-secretase, responsible for regulated intramembrane proteolysis of numerous substrates that include cytokine receptors. For this reason, we tested the hypothesis that mutations in PSEN1 impact inflammatory responses in astrocytes, thereby contributing to disease progression. We developed patient-derived models of iPSC-astrocytes, representing three lines harbouring PSEN1 mutations and six control lines (including two isogenic controls). Transcriptomic and biochemical assays were used to investigate differential inflammatory responses to TNFα, IL1α and C1Q. We show that PSEN1 is upregulated in response to inflammatory stimuli, and this upregulation is disrupted by pathological PSEN1 mutations. Using transcriptomic analyses, we demonstrate that PSEN1 mutant astrocytes have an augmented inflammatory profile in their basal state, concomitant with gene expression signatures revealing dysregulated intramembrane proteolysis and JAK-STAT signalling. Detailed investigation of the JAK-STAT2 signalling pathway showed reduced cell surface expression of IFNAR2, lower STAT2 phosphorylation cascades and delayed NFκB nuclear localisation in PSEN1 mutant astrocytes in response to inflammatory stimuli, thereby implicating the notion of altered cytokine signalling cascades. Finally, we use small molecule modulators of γ-secretase to confirm a role for PSEN1/γ-secretase in regulating the astrocytic response to inflammatory stimuli. Together, these data suggest that mutations in PSEN1 enhance cytokine signalling via impaired regulated intramembrane proteolysis, thereby predisposing astrocytic inflammatory profiles. These findings support a two-hit contribution of PSEN1 mutations to fAD pathogenesis, not only impacting APP and Aβ processing but also altering the cellular response to inflammation.

This study provides the first quantification of neurofilament light chain (NfL) kinetics in the human CNS using stable isotope labelling kinetics. NfL is elevated in CSF and blood across a range of traumatic, inflammatory and neurodegenerative diseases of the CNS, and has been increasingly included in clinical trials as a secondary or exploratory outcome measure of target engagement. Interpreting trajectories of NfL post-treatment has been challenging, prompting a greater need and focus on understanding its pathophysiology. We set out to measure NfL kinetics in the human CNS using stable isotope labelling kinetics. In human neurons derived from induced pluripotent stem cells, we show that NfL turnover is relatively slow, comparable to other long-lived proteins such as tau. We detected a delay of 3 to 6 days in the release of NfL into the media, unexpected for a passively released protein and supporting that controlled mechanisms of release could contribute to the appearance of NfL in the extracellular milieu. We optimized the kinetic NfL assay to measure the turnover of NfL in the human CNS. Participants with diagnosed primary tauopathies (n = 10) were recruited to the Human CNS Tau Kinetics in Tauopathies study and a cohort of cognitively unimpaired or with mild cognitive impairment (Clinical Dementia Rating score ≤0.5; n = 22) to the Tau Stable Isotope Labelling Kinetics study. Patients with suspected normal pressure hydrocephalus (n = 3) and primary tauopathy cases (n = 3) were examined to assess labelling in the brain parenchyma and ventricular CSF. In brain tissue, isotopically labelled in vivo and sampled ex-vivo and post-mortem, NfL is rapidly labelled but remains stable 18 months after, indicating very slow turnover and likely incorporation into very stable NfL networks. In line with a controlled mechanism of release in vivo, appearance of labelled NfL in CSF was detectable between 53 and 162 days post-labelling, during which NfL labelling did not reach its peak, suggestive of a half-life in CSF >3 months. These findings support the interpretation that acute rises in CSF NfL concentration are likely to be related to passive release or CSF clearance failure. We also show that active but delayed release of newly translated NfL can contribute to the concentration of NfL in CSF, but this would not be expected for at least 8 weeks. Clinical trials using NfL as an outcome measure will benefit from substantially longer follow-up periods and isotopic labelling to understand the NfL response to therapeutic intervention.

The amyloid precursor protein ( APP ) is processed by multiple enzymes to generate biologically active peptides, including amyloid-β (Aβ), which aggregates to form the hallmark pathology of Alzheimer’s disease (AD). Aβ is produced through an initial β-secretase cleavage of APP, generating a 99-amino acid C-terminal fragment (APP-C99). Subsequent cleavage of APP-C99 by γ-secretase produces Aβ peptides of varying lengths. To better understand the transcriptional regulation of Aβ production, we employed long-read RNA sequencing and identified previously unannotated transcripts encoding APP-C99 with an additional methionine residue (APP-C100), generated independently of β-secretase cleavage. These transcripts are expressed separately from full-length APP , and we observed that cells lacking full-length APP can still produce Aβ through these shorter isoforms. Importantly, mass spectrometry analysis of cerebrospinal fluid (CSF) revealed peptides consistent with the methionine-extended Aβ species, supporting the in vivo translation of these transcripts. Our findings reveal an alternative pathway for Aβ generation and aggregation, highlighting a potential new target for modulating Aβ accumulation in AD. ### Competing Interest Statement E.K.G. has received paid consultancy from Isogenix limited within the last 12 months. J.H. has consulted with Eisai and Eli Lilly. BrightFocus Foundation, A2021009F UK Research and Innovation, N008324/1 Alzheimer's Society, AS-JF-18-008 Alzheimer’s Association, 23AARFD-1029918 Alzheimer’s Research UK, ARUK-SRF2016B-2 UCL Biomedical Research Centre, https://ror.org/03r9qc142 Fidelity Foundation Cure Alzheimer’s Fund Dolby Foundation

Protocol for cortical organoid generation in 96-well plates. Based on Pasca et al. 2015 (https://doi.org/10.1038/nmeth.3415) with some modifications.

Genome-wide association studies have identified Glycoprotein Nmb ( GPNMB ) as a risk factor for Parkinson’s disease. The risk allele increases GPNMB transcription and GPNMB protein levels in the CSF highlighting GPMNB as a potential biomarker for Parkinson’s disease. However, a lack of knowledge of GPNMB’s function and mechanism of secretion has hindered an interpretation of secreted GPNMB levels. In this study, we assessed the mechanism of GPNMB secretion by macrophages, the primary cell type expressing GPNMB in the brain. We show that GPNMB is secreted in response to lysosomal stress via lysosomal exocytosis and highlight the Parkinson’s disease risk factor LRRK2 as a strong modulator of GPNMB secretion.

The blood-brain barrier (BBB) preserves brain health through selective permeability, and its disruption is a hallmark of many neurological disorders. Mechanical stimuli such as shear stress and cyclic strain are increasingly recognised to influence BBB integrity and function, while alterations in tissue stiffness and extracellular matrix composition contribute to its breakdown during ageing and disease. Despite its importance, BBB mechanobiology remains underexplored. Here we highlight the central role of mechanics in BBB development, pathology, and ageing, identify key knowledge gaps, and argue that combining innovative BBB model systems with mechanical probing techniques could transform therapeutic strategies targeting brain vascular dysfunction.

Familial Alzheimer's disease (fAD), arising from mutations in amyloid-precursor-protein (APP) and presenilin (PSEN1/2) genes, leads to the production of longer, aggregation-prone amyloid-beta (Aβ) peptides—a hallmark of Alzheimer's disease. Age-at-onset (AAO) varies among carriers of different mutations. Recent evidence challenges the Aβ42:40 ratio as the leading and predictor of AAO between different pathogenic variants, prompting exploration of peptide combinations as potential biomarkers for these tasks. Data consisted of patient-derived in-vitro iPSC models of fAD and Ab measurements via enzyme-linked immunosorbent assay (ELISA). With novel data of our own (N=39) and published data from Liu et al. 2022 (N=134) (Table 1), we performed a data-driven analysis of Aβ37/38/40/42/43 combinations, employing area under the curve (AUC) and coefficient of determination metrics for classification (WT vs AD) and AAO regression. A linear model optimized for these metrics was also explored. Each peptide was divided by the average of the control group of the same assay. Then, we weighted each peptide in accordance with the average profile of the control population in Liu. Data transformation using controls proved necessary for to produce comparable ratios (Figure 1). Table 2 presents diagnostic and predictive performance for each peptide-ratio biomarker relative to biomarkers from the literature (top 3 rows). Our new data-driven biomarkers improved upon the literature in both diagnostic AUC (not statistically significant: DeLong’s test p>0.47) and predicting AAO (Spearman’s R^2 significant differences in UCL data; Hittner’s p<0.005; p>0.5 in Liu’s). Our results do not recapitulate Liu et al. findings, as Aβ37:42 did not outperform Aβ42:40 in our data; one explanation could be that our lines represent a physiological model of Ab production that is free from plaque deposition. While no significant differences emerged in AD classification, specific peptide ratios demonstrated varying correlations with AAO: short-to-long ratio > Aβ42:40, and Aβ42:40 > Aβ37:42, but other ratios were found to perform better in each set, suggesting mutation-specific profiles. Data-driven linear combination can improve the results, and notably, both optimizing for AUC or AAO found the same ratio of short/long, albeit with different weights. Standardization through control-based data was essential for robust measures.

Normal pressure hydrocephalus is a common cause of gait and cognitive impairment in later life, characterised by accumulation of excessive cerebrospinal fluid (CSF). Clinical improvement can occur following CSF diversion. No biomarkers are available to mechanistically investigate fluid accumulation, support diagnosis or predict response to CSF diversion. We developed a stable isotope labeling kinetics (SILK) method to capture the function of the main site of production of CSF in humans, the choroid plexus (ChP), in vitro and in vivo. We captured ChP protein kinetics in human ChP organoids and the CSF of participants with suspected NPH undergoing CSF drainage (n=10) or controls (n=9). We found that transthyretin is abundantly secreted by ChP organoids, and we observe correlations with CSF transthyretin synthesis rates and volume of CSF production in vivo (ρ=0.738; p<0.05). Clearance rates of transthyretin are ~10 fold slower in NPH compared to controls, demonstrating impaired CSF clearance. ChP SILK is a novel clinical tool for interrogating CSF flow. ### Competing Interest Statement KM is a shareholder of Guilford Street Laboratories Ltd. RJB has received research funding from Avid Radiopharmaceuticals, Janssen, Roche/Genentech, Eli Lilly, Eisai, Biogen, AbbVie, Bristol Myers Squibb, and Novartis. Washington University and RJB have equity ownership interest in C2N Diagnostics and receive income based on technology (stable isotope labeling kinetics, blood plasma assay, methods of diagnosing AD with phosphorylation changes, neurofilament light chain assays and materials) licensed by Washington University to C2N Diagnostics. RJB receives income from C2N Diagnostics for serving on the scientific advisory board. RJB serves as an unpaid member on scientific advisory boards for Roche and Biogen. ### Funding Statement RWP is the recipient of an Alzheimer's Association Clinician Scientist fellowship AACSF-20-685780, Alzheimer's disease Part the Cloud award PTC-22-982162 and a Rosetrees Race Against Dementia Team award supported by the Q Charitable Trust. He is supported by the UCLH Queen Square BRC. TAG was supported by research funding from the Alzheimer's Association (23AARFD-1029918). ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: The NHS Health Research Authority, Research Ethics Committee London-Bloomsbury gave ethical approval for the UCL NPH SILK study. All individuals provided informed written consent. Individuals with suspected idiopathic normal pressure hydrocephalus were recruited from the specialist hydrocephalus service at the National Hospital for Neurology and Neurosurgery, Queen Square. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes Data available from the corresponding author on reasonable request.