Advancing Precision Medicine in Neurology

1 Navigating Neurological Biomarker Research: Progress, Learnings and Future Directions Introduction The landscape of neurological disease research is rapidly evolving, with protein biomarkers playing a pivotal role in deepening our understanding, improving diagnosis, and guiding treatment of conditions such as Alzheimer’s disease (AD), Parkinson’s Disease (PD), and multiple sclerosis (MS). While cutting-edge proteomics platforms have enabled significant advancements in the field, challenges remain in translating discoveries into clinical applications. Bridging these gaps requires sharing key learnings from both successes and setbacks as the field advances toward precision medicine. In an effort to contribute to this knowledge sharing, this White Paper outlines key insights from a recent panel discussion which brought together leading academic and pharma experts as they covered: • Current and emerging biomarkers and their impact on diagnosing and treating neurological diseases. • Challenges, opportunities and practical strategies for integrating proteomics platforms into clinical research, from sample handling to data interpretation. • The need for collaborations and further partnerships between academia, industry, and clinicians to drive progress. • Future directions of proteomics and multi-omics in driving precision medicine in neurological disease research. State of the field: Neurological Protein Biomarker Research Neurodegenerative disorders In AD, pathophysiological changes begin long before the onset of clinical symptoms. These changes are accompanied by a cascade of biomarker alterations—initially involving amyloid accumulation, followed by tau aggregation. Ongoing research efforts aim to map these pathological trajectories before structural brain changes or cognitive decline become evident. Current therapeutic options are focused on early disease stages— specifically mild cognitive impairment due to AD and early-stage AD dementia—highlighted in the pink zone of Figure 1 (1), as well as prevention trials targeting pre-clinical AD (2, 3). In addition, a wide range of biological targets are currently under investigation, many extending beyond amyloid and tau pathology and spanning across phases I to III of clinical development, as depicted in Figure 2. Fluid-based target engagement markers were utilized in 26% of the analyzed trials (26% in phase I, 31% in phase II, and 15% in phase III;), with the highest use seen in trials targeting inflammation and amyloid. In contrast, use of fluidbased target engagement markers was limited in trials focused on synaptic plasticity and neuroprotection, despite biomarker availability. They were entirely absent across several other target classes, including neurotransmitter receptors, neurogenesis, vasculature, epigenetic regulators, proteostasis, the gut–brain axis, environmental factors, multi-target, and unknown targets (4). White paper Figure 1. Graph representing biomarker magnitude across the AD disease progression continuum (adapted from Sperling et al. Alzheimer’s Dement., 2011). Need to intervene early before disease pathology accelerates Aß accumulation (CSF/PET) Synaptic dysfunction (FDG-PET/MRI) Tau-mediated neuronal injury (CSF) Brain structure (VMRI) Cognition Function Detection threshold 2 Along those lines, there is a growing emphasis on the development of highly sensitive biomarkers capable of detecting and staging the earliest pathological changes in AD. In 2018, the National Institute on Aging and the Alzheimer's Association (NIA-AA) criteria introduced a biomarker-based framework for diagnosing AD, which was updated in 2024 to expand the range of fluid biomarkers available for diagnosis and staging, as shown in Table 1 (5). Today, cerebrospinal fluid (CSF) biomarkers— particularly the FDA approved, Aβ42/40, phosphorylated tau 181 (pTau181) and total Tau test, as well as tests for neurofilament light (NfL) and are used in expert clinical settings. The field is also rapidly moving toward the implementation of less invasive bloodbased assays to assess neuronal health, including Tau species and NfL, and demarcated by the 25 May 2025 FDA approval of a pTau217/Aβ42 AD blood test, with many new tests under evaluation. Figure 2. Number of trials with and without a fluid-based target engagement marker, by target class. Total number of trials (n= 272) was included in this analysis, darker shade indicates use of a fluid-based target engagement marker. Between brackets the total number of trials in that target class is listed (source: Oosthoek, Vermunt et al. Alzheimer Research & Therapy, 2024.). Intended use CSF Plasma Imaging Diagnosis A: (Aß proteinopathy) – – Amyloid PET T1: (phosphorylated and secreted AD tau) – p-tau217 – Hybrid ratios p-tau181/Aß42, t-tau/Aß42, Aß42/40 %p-tau217 – Staging, prognosis, as an indicator of biological treatment effect A: (Aß proteinopathy) – – Amyloid PET T1: (phosphorylated and secreted AD tau) – – Hybrid ratios p-tau 181/Aß42, t-tau/Aß42, Aß42/40 %p-tau217 – T2: (AD tau proteinopathy) MTBR-tau243, other p-tau forms (e.g., p-tau205), nonphosphorylated mid-region tau fragments MTBR-tau243, other p-tau forms (e.g., p-tau205) Tau PET N (injury, dysfunction, or degeneration of neuropil) NfL NfL Anatomic MRI, FDG PET I (inflammation) Astrocytic activation GFAP GFAP – Identification of copathology N (injury, dysfunction, or degeneration of neuropil) NfL NfL Anatomic MRI, FDG PET V vascular brain injury – – Infarction on MRI or CT, WMH S a-synuclein aSyn-SAA Table 1. Intended uses for imaging, CSF, and plasma biomarker assays. 2024 AA revised criteria has expanded fluid biomarkers for diagnosis and staging of AD (source: Jack CR Jr, et al. Alzheimer’s Dement., 2024). 3 Figure 3. Individual trajectories show the heterogeneity of cognitive decline for an early AD cohort of 302 individuals, selected from the Alzheimer's Disease Neuroimaging Initiative database (ADNI). Clinical Dementia Rating scale–sum of boxes: the Alzheimer's Disease Assessment Scale–cognitive subscale (ADAS-Cog). Dotted vertical line presents scores at the 18 months time point. (adapted figure from Jutten RJ, et al. Neurology, 2021) Alzheimer’s disease is increasingly recognized as a heterogeneous condition. As shown in Figure 3, disease trajectories can vary significantly across individuals, underscoring the need for precision medicine approaches (6). Molecular subtyping— enabled by platforms such as the Olink Explore HT panel—has begun to reveal biologically meaningful subgroups within AD. These insights may help explain clinical variability and guide more targeted therapeutic strategies. Similar advances are being made in other neurodegenerative conditions, such as PD and Lewy body dementia, where trajectory mapping and molecular profiling are beginning to yield actionable insights. For example, in PD, the enzyme dopa decarboxylase (DDC)—a key player in the final step of dopamine synthesis—has emerged as a promising early-stage CSF biomarker (7). Following this recent discovery, diagnostic tests targeting DDC are already in development and may hold significant clinical utility in the near future. “Molecular subtyping—enabled by platforms such as the Olink Explore HT panel—has begun to reveal biologically meaningful subgroups within AD. These insights may help explain clinical variability and guide more targeted therapeutic strategies.” “To address the biological complexity of multiple sclerosis, multiplex protein panels are being developed to capture the multifaceted nature of disease activity and tissue damage. These panels not only support pathophysiological understanding but also show promise for aiding in early diagnosis.” Pallavi Sachdev, MPH, PhD Executive Director and Head, Translational Medicine, Eisai Inc. Ahmed Abdelhak, MD Assistant Professor of Neurology, University of California San Francisco Multiple sclerosis Diagnosis of MS has traditionally relied on magnetic resonance imaging (MRI); however, recent advances have brought fluid biomarkers—particularly NfL—into the spotlight. A rise in NfL levels has been observed in relation to MS activity and progression and has been noted before the appearance of clinical symptoms, especially following Epstein-Barr virus (EBV) infection. Various treatment strategies exert differing degrees of influence on NfL concentrations, offering a more refined framework for distinguishing high-efficacy from lower-efficacy therapies (8, 9). One of the major unmet needs in MS is the ability to monitor and predict disease progression. Glial fibrillary acidic protein (GFAP) has emerged as a potential marker in this context, with lower levels correlating with a reduced risk of disease progression (10). However, the biological complexity of MS—characterized by simultaneous processes such as immune cell activation, inflammation, and glial cell responses involving oligodendrocytes, microglia, and astrocytes—warrants more comprehensive biomarker strategies. To address this, multiplex protein panels are being developed to capture the multifaceted nature of disease activity and tissue damage. These panels not only support pathophysiological understanding but also show promise for aiding in early diagnosis. One such example is an 18-protein panel that includes NfL and myelin oligodendrocyte glycoprotein (MOG), a marker of myelin damage, developed using Olink’s Proximity Extension Assay (PEA) technology (11). Another notable example is a 21-protein panel capable of identifying individuals at risk of MS years before symptom onset, offering a crucial window for initiating preventive strategies to reduce long-term disability (12). Unlike Alzheimer’s disease and other dementias, MS lacks clearly defined proteinopathies. As a result, exploratory proteomic approaches—such as those using large-scale datasets like the UK Biobank—are essential. These studies help to identify novel, differentially expressed proteins in MS patients compared to healthy controls (Figure 4), with the goal of uncovering new biological targets and broadening the biomarker toolkit available to researchers and clinicians (13). Fast progressors Slow progressors 4 The need for distinct categories of biomarkers across neurological diseases In 2016, the FDA–NIH Biomarker Working Group introduced a classification framework encompassing seven categories of biomarkers and related endpoints (14). These include diagnostic, monitoring, prognostic, predictive, pharmacodynamic/ response, safety, and susceptibility/risk biomarkers. Applying this framework to neurological diseases reveals persistent gaps that are impeding both clinical decision-making and therapeutic development. In multiple sclerosis (MS), a key urgent need lies in disease progression biomarkers. While disease-modifying therapies (DMTs) have significantly reduced disease activity—often measured via relapse rate or MRI lesions—effective tools to monitor and predict progression remain scarce. GFAP has shown promise as a progression marker, with lower concentrations associated with a reduced risk of deterioration (10). However, the complexity of MS pathology requires a more nuanced approach. Progression is heterogeneous across patients and can vary significantly depending on disease stage and individual characteristics. Therefore, a personalized biomarker strategy is critical. Large-scale proteomic studies have revealed considerable inter-individual variation in biological pathways that would have been difficult to detect using conventional technologies, further reinforcing the need for individualized biomarker profiles. In Alzheimer’s disease (AD), the early detection of pathology before clinical symptoms emerge remains a major unmet need. Diagnostic biomarkers capable of identifying and staging AD in its preclinical stage would allow for significantly earlier intervention. Equally important are biomarkers that can predict and track disease progression, which are essential for implementing precision medicine approaches and accelerating drug development. These tools would enable clinicians to deliver therapies before irreversible neurological damage occurs and to target individuals with more aggressive disease trajectories using molecular subtyping strategies. Beyond diagnostic and progression markers, there is a need for target engagement biomarkers. Broader proteomic approaches, combined with integration of preclinical data, can improve our understanding of the relevance of individual targets within specific biological pathways. This in turn enhances the reliability of these markers in measuring treatment response and informing trial success. Together, these advances underscore the importance of a multidimensional biomarker strategy—one that incorporates both disease-specific and patient-specific information, leverages emerging proteomic technologies, and supports the full therapeutic lifecycle from early diagnosis to late-stage intervention. Which pathways hold novel insights for biomarker discovery in neurological diseases? A critical component of effective biomarker development is understanding the biological pathways that a given marker represents. Without this context, it is difficult to interpret whether a biomarker reflects disease initiation, progression, or response to therapy. As the field of neurology moves toward more robust and clinically relevant fluid biomarkers, several biological pathways have emerged as promising focal points for discovery. As mentioned, in AD and other dementias, one of the foremost challenges is detecting and characterizing the early stages of disease—prior to irreversible neurological damage. Addressing this challenge will depend heavily on studying neuroinflammatory mechanisms, particularly the roles of astrocytes and microglia. These glial cells are increasingly understood to be active participants in disease progression and elucidating the timing and nature of their involvement may yield biomarkers that reflect early or preclinical disease states. Other pathways of interest include the complement cascade, as well as processes related to synaptic pruning and synaptic dysfunction. These pathways are believed to be disrupted early in disease and may contribute to the subtle neural changes that precede overt cognitive decline. By targeting these mechanisms, researchers may be able to uncover novel biomarkers capable of identifying the disease at its earliest stages or predicting its course more accurately. Figure 4. Plasma proteomic analysis of multiple sclerosis. Volcano plot displaying differences in plasma levels of proteins measured with Olink proteomics between UK Biobank participants with (n = 407) and without (n = 39,979) MS at the time of sample collection. The x-axis indicates the log-fold change of the protein (values above 0 indicate proteins present at higher levels in the MS cohort, while those with values below 0 present at higher levels in the control cohort). The y-axis indicates the negative log of the p value for each protein (higher values indicate a more statistically significant result). Proteins surpassing a Bonferroni-corrected threshold of 5% are shown in color; other results are shown in gray. (source: Jacobs et al., Annals of Clinical and Translational Neurology, 2024). 5 In multiple sclerosis (MS), a similar shift in focus is occurring. While NfL has been widely adopted as a marker of axonal injury, recent findings suggest that glial-derived markers may correlate more strongly with disease progression. Microglia are gaining attention as a therapeutic target in emerging treatment strategies, underscoring the need to better characterize microglial signaling pathways and their relationship to disease dynamics. Importantly, many relevant biological pathways may remain undiscovered. Therefore, broad-scale discovery approaches, such as untargeted proteomics and multi-omics integration, remain indispensable. These methods not only enable the identification of novel biomarkers but also help to map previously uncharted biological networks involved in neurological disease. As our understanding of these pathways deepens, they will serve as the foundation for next-generation biomarkers—capable of informing diagnosis, stratifying patients, and guiding precision therapies in both research and clinical settings. What types of studies are needed to accelerate progress? As mentioned, one of the key challenges in neurological disease research is the heterogeneity of disease progression. This demands longitudinal, multi-omic cohort studies, particularly in diverse populations. Another important aspect is leveraging real-world evidence studies that not only integrate the data that is readily available, but also electronic health records and digital health data. In addition, it is essential to validate findings across centers and interpret them across diseases. This is largely enabled and supported through broader collaborative efforts, such as the Coral consortium. “Addressing the challenge of characterizing the early stages of AD and other dementias will depend heavily on studying neuroinflammatory mechanisms, particularly the roles of astrocytes and microglia. These glial cells are increasingly understood to be active participants in disease progression and elucidating the timing and nature of their involvement may yield biomarkers that reflect early or preclinical disease states.” Key considerations for running a successful proteomics study What is the Coral Consortium? Lisa Vermunt, MD, PhD Assistant Professor, Neurochemistry Laboratory, Amsterdam UMC The CORAL consortium (Community using Olink for Research on Alzheimer’s Disease and other neurological diseases) is a collaborative framework. The objective is to accelerate the identification of proteins and mechanisms for neurological diseases, as well as the translation of novel biomarkers for neurological diseases to the clinic. Consortium aims: • Sharing experiences with biomarker development workflows, starting from cerebrospinal fluid (CSF) and plasma proteins on the Olink Proteomics platform. • Performing collaborative meta-analyses and other studies to identify novel mechanisms and biomarkers for neurological diseases. CORAL was started in July 2021 and continuously welcomes new members. For more information, please contact Yanaika Hok-a-Hin (CORAL project coordinator) and Lisa Vermunt or Charlotte Teunissen (CORAL project chairs). Considering the substantial contributions of proteomics to neurological biomarker research and advancing precision neurology, it will undoubtedly play a crucial role in future clinical trials and academic research. Here are some key considerations for a successful proteomics study: • Designing a successful proteomics study starts with wellphenotyped groups and sufficiently large sample sizes. Extreme phenotypes can help to increase the study power if the sample sizes are not large, e.g. familial Alzheimer's disease compared to age- matched controls (15). Furthermore, studying longitudinal samples is powerful because individuals serve as their own controls, e.g. in clinical trials. • Handling large datasets requires understanding how the data was generated and choosing the right bioinformatics tools. It is therefore crucial to take the time to learn from the literature and other experts to make an informed decision on how to perform the data analysis, as well as interpret the results. • Validating targets in animal models is critical to confirm where targets are indeed expressed. Basic validation, like tissue staining, still has a role. • Collaboration is essential. Cross-functional teams with diverse expertise, shared goals, and transparent communication lead to success. Using project charters and regular check-ins, encouraging co-authorship, and identifying joint funding opportunities are essential. Also, advocating for open-access data repositories leads to shared resources and knowledge that will accelerate discovery and development. 6 Looking ahead in neurological biomarker research The momentum in neurological protein biomarker research is undeniable. Fueled by technological innovation in proteomics and strengthened by growing collaboration between academia and industry, the field is making strides toward earlier detection, more precise diagnosis, and targeted treatment strategies for complex conditions like Alzheimer’s disease and multiple sclerosis. Molecular subtyping, enabled by deep proteomics and multi-omics, reflects underlying pathophysiological differences and offers a framework for precision medicine strategies and can guide rational combination therapies tailored to molecular subtypes, increasing the likelihood of clinical success and patient benefit. Yet, to realize the full potential of precision neurology, the path forward must be anchored in longitudinal, multi-omic, and collaborative study designs, with special attention to patient heterogeneity and underrepresented populations. As biomarker discovery continues to evolve, the integration of biological insights, rigorous validation, and real-world applicability will be critical to translating scientific progress into meaningful clinical impact. The next generation of biomarker-driven innovation will depend not only on discovery—but on our shared commitment to data transparency, methodological rigor, and cross-disciplinary partnership. Olink’s PEA™ platform can support your protein biomarker research journey Olink’s product portfolio, backed by the innovative PEA technology, offers an end-to-end solution for biomarker research. It supports researchers from exploratory studies to clinical translation by allowing for simultaneous measurement of thousands to a single digit number of analytes with a single, uniquely scalable platform. Olink’s portfolio includes a selection of preconfigured and customizable panels, while offering a simple, wash-free workflow and the lowest sample volume requirement among multiplex immunoassay platforms. Discover the new solution – Olink® Target 48 Neurodegeneration panel Get access to PEA Start leveraging the PEA technology by selecting an option that works best for you: • Run your samples at Olink Analysis Service • Find an Olink Certified Service Provider • Set up Olink in your own lab Contact us at enquiries@olink.com so we can guide you through your choice. Value of Analyzing CSF vs Plasma and alternative matrices in Neurological Protein Biomarker Research The comparative analysis of cerebrospinal fluid (CSF) and plasma offers a critical window into the biological processes underlying neurological and neurodegenerative diseases. While CSF provides a more direct reflection of central nervous system (CNS) processes due to its close anatomical proximity to the brain, plasma represents a more accessible and non-invasive medium that, when properly interpreted, can yield valuable systemic and CNS-related insights. Matched CSF–plasma studies are particularly powerful, as they enable researchers to track the same biological processes across compartments. However, it is important to note that signals in CSF and plasma may not always align. This discrepancy itself can be informative, highlighting differential expression or compartmentalization of disease processes. These paired samples can also facilitate the development of blood-based proxies for CSF biomarkers—an essential step toward more scalable diagnostics and monitoring strategies in both research and clinical settings. Despite their potential, CSF–plasma matched studies remain underutilized, primarily due to the logistical challenges and invasiveness associated with CSF collection. As such, expanding access to these samples represents a critical opportunity for the field. In diseases such as MS, plasma biomarkers hold particular promise, given the strong systemic immune component that characterizes the condition. Signals related to inflammation and immune activation may be captured more readily in blood than in CSF in some contexts. Furthermore, the practicality of repeated blood sampling makes it a preferred matrix for longitudinal monitoring of treatment response and disease stability, especially in real-world clinical settings. In parallel, extracellular vesicles (EVs) originating from the brain have emerged as a promising, though still evolving, matrix for neuroproteomic research. EVs may offer a means to access brainderived proteins in blood, bridging the gap between CNS pathology and peripheral detection. While methodological consensus on EV isolation and analysis is still developing, this area is expected to yield valuable tools for biomarker discovery and disease monitoring. Finally, direct analysis of brain tissues and animal models remains essential for validating findings and understanding the mechanistic context of circulating biomarkers. These models provide a crucial link between observed protein expression and the underlying cellular and molecular processes. In summary, CSF, plasma, extracellular vesicles, and brain tissue each offer unique advantages and insights. The choice of matrix should be guided by the specific research or clinical question, the biological compartment of interest, and the feasibility of sample collection in the intended application. 7 References 1. Sperling, R.A., Aisen, P.S., Beckett, L.A., et al. Toward defining the preclinical stages of Alzheimer's disease: Recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. (2011) Alzheimer's & Dementia. doi: 10.1016/j.jalz.2011.03.003 2. Rafii MS, Sperling RA, Donohue MC, et al. The AHEAD 3-45 Study: Design of a prevention trial for Alzheimer's disease. (2023) Alzheimers Dement. doi: 10.1002/alz.12748. 3. Sperling RA, Donohue MC, Raman R, et al. Association of Factors With Elevated Amyloid Burden in Clinically Normal Older Individuals. (2020) JAMA Neurol. doi: 10.1001/jamaneurol.2020.0387 4. Oosthoek M, Vermunt L, de Wilde A, et al. Utilization of fluid-based biomarkers as endpoints in disease-modifying clinical trials for Alzheimer's disease: a systematic review. (2024) Alzheimer Research & Therapy. doi: 10.1186/s13195-024-01456-1. 5. Jack CR, Andrews JS, Beach TG, et al. Revised criteria for diagnosis and staging of Alzheimer's disease: Alzheimer's Association Workgroup. (2024) Alzheimer's Dement. doi: 10.1002/alz.13859 6. Jutten RJ, Sikkes SAM, Van der Flier WM, et al. Alzheimer's Disease Neuroimaging Initiative. Finding Treatment Effects in Alzheimer Trials in the Face of Disease Progression Heterogeneity. (2021) Neurology. doi: 10.1212/WNL.0000000000012022 7. del Campo, M., Vermunt, L., Peeters, et al. CSF proteome profiling reveals biomarkers to discriminate dementia with Lewy bodies from Alzheimer´s disease. (2023) Nat Commun. doi: 10.1038/s41467-023- 41122-y 8. Bjornevik K., Cortese M., Healy B.C., et al. Longitudinal analysis reveals high prevalence of Epstein-Barr virus associated with multiple sclerosis. (2022) Science. doi:10.1126/science.abj8222; 9. Benkert P, Meier S, Schaedelin S, et al. NfL Reference Database in the Swiss Multiple Sclerosis Cohort Study Group. Serum neurofilament light chain for individual prognostication of disease activity in people with multiple sclerosis: a retrospective modelling and validation study. (2022) Lancet Neurology. doi: 10.1016/S1474-4422(22)00009-6 10. Abdelhak A., Maceski A.M., Schädelin S., et al. Treatment-associated changes in serum glial fibrillary acidic protein and neurofilament light chain levels and risk of disability progression independent of relapse activity in multiple sclerosis (2024) ECTRIMS. doi:10.1177/1352458524126921 11. Chitnis T, Qureshi F, Gehman VM, et al., Inflammatory and neurodegenerative serum protein biomarkers increase sensitivity to detect clinical and radiographic disease activity in multiple sclerosis. (2024) Nature Communication. doi: 10.1038/s41467-024-48602-9. 12. Abdelhak et al, in revision 13. Jacobs, B.M., Vickaryous, N., Giovannoni, G., et al. Plasma proteomic profiles of UK Biobank participants with multiple sclerosis. (2024). doi: 10.1002/acn3.51990 14. FDA-NIH Biomarker Working Group. BEST (Biomarkers, EndpointS, and other Tools). Silver Spring (MD): Food and Drug Administration (US); 2016 15. van der Ende EL, In 't Veld SGJG, Hanskamp I, et al. CSF proteomics in autosomal dominant Alzheimer's disease highlights parallels with sporadic disease. (2023) Brain. doi: 10.1093/brain/awad213. www.olink.com © 2025 Olink Proteomics AB, part of Thermo Fisher Scientific. Olink products and services are For Research Use Only. Not for use in diagnostic procedures. All information in this document is subject to change without notice. This document is not intended to convey any warranties, representations and/or recommendations of any kind, unless such warranties, representations and/or recommendations are explicitly stated. Olink assumes no liability arising from a prospective reader’s actions based on this document. OLINK, NPX, PEA, PROXIMITY EXTENSION, INSIGHT and the Olink logotype are trademarks registered, or pending registration, by Olink Proteomics AB. All third-party trademarks are the property of their respective owners. Olink products and assay methods are covered by several patents and patent applications https://www.olink.com/patents/. 1656, v1.0, 2025-09-04