Serum Neurofilament Light Chain Predicts Stroke Outcome and is a Potential Marker for Treatment Effects of Neural Stem Cell-derived Extracellular Vesicles in a Rat Stroke Model

Translational Stroke Research (2026) 17:54 https://doi.org/10.1007/s12975-026-01445-6 RESEARCH Serum Neurofilament Light Chain Predicts Stroke Outcome and is a Potential Marker for Treatment Effects of Neural Stem Cell-derived Extracellular Vesicles in a Rat Stroke Model Megan K. Cannon1 · Alinde R. Fojtik1 · Charles M. White1 · Raymond L. Swetenburg1 · Steven L. Stice1,2,3 · Sean I. Savitz4 · Emily W. Baker1 Received: 6 February 2026 / Revised: 3 April 2026 / Accepted: 28 April 2026 © The Author(s) 2026 Abstract Acute ischemic stroke (AIS) remains a leading cause of disability worldwide, and effective treatments are urgently needed beyond reperfusion therapy. Translating preclinical success to clinical impact has been hindered by variability in animal models and the lack of translational biomarkers that predict outcomes across species. To overcome these barriers, we developed a robust rat AIS model optimized for consistency and severity, enabling rigorous therapeutic testing. Additionally, we tested a panel of common clinical serum biomarkers to improve translation from rodents to humans. We demonstrated that serum neurofilament light chain (NfL) -a biomarker widely used in clinical stroke studies- strongly correlated with functional outcomes, establishing a translational link that has not been previously reported in rats. Notably, NfL’s predictive capabilities performed as well as infarct volume while outperforming serum biomarkers intercellular adhesion molecule-1 (ICAM-1) and S100 calcium binding protein (S100B). Using this platform, we evaluated the therapeutic impact of neural stem cell-derived extracellular vesicles (NSC EV), a novel biologic therapy poised for clinical trials, on stroke outcome in our rat AIS model. A three-dose regimen of NSC EV over 48 h confirmed our previous studies and produced the best outcomes in stroked animals evidenced by smaller infarct volume, improved neurologic score, and reduced serum NfL. In addition, single-dose and two-dose regimens of NSC EV were both effective at some endpoints. These findings not only validate NfL as a cross-species biomarker but also provide critical dosing insights for NSC EV therapy, accelerating the path from bench to bedside for AIS treatment. Keywords Ischemic Stroke · Neural stem cell derived extracellular vesicles · Biomarkers · Neurofilament light chain Introduction While reperfusion therapy is the gold standard intervention for acute ischemic stroke (AIS), less than half will be functionally independent at Day 90 even when complete Emily W. Baker 1 Aruna Bio Inc, Athens, GA, USA 2 Regenerative Bioscience Center, University of Georgia, Athens, GA, USA 3 Department of Animal and Dairy Science, University of Georgia, Athens, GA, USA 4 Institute for Stroke and Cerebrovascular Disease, UTHealthHouston, Houston, TX, USA cerebral reperfusion is achieved [1–3]. Furthermore, reperfusion therapy does not address the complex cascade of cellular and molecular events triggered by ischemia. Additional treatment options are an urgent, global unmet medical need. Neural stem cell-derived extracellular vesicles (NSC EV) are a novel biopharmaceutical product with powerful, multi-mechanistic target biology, and previous studies published by our lab demonstrate that NSC EV reduces infarct size and mitigates the systemic immune response leading to preserved motor function in multiple stroke models [4–6]. Additionally, NSC EV provides unique advantages to cellbased therapies due to its manufacturing scalability, ease of administration, and the ability to cross the blood brain barrier and home to the infarct site [7]. While other biologic therapies including NSC EV show promise in stroke 54 Page 2 of 13 models, it is challenging to translate preclinical results to clinical trials. Identifying reliable biomarkers that translate from preclinical to clinical studies is a gap that hinders new stroke therapies. Robust and consistent blood-based biomarkers could aid in AIS prognosis and serve as a surrogate biomarker for treatment effect, while being less expensive and more accessible than neuroimaging or cerebrospinal fluid (CSF)-based biomarkers [8]. Establishing a biomarker that is linked, either directly or indirectly, to the investigational drug’s action is critical for clinical trial success, as it enables highly quantitative pharmacological assessment and strengthens the biological rationale for therapeutic efficacy [9]. There have been promising candidates for determining AIS prognosis as well as differentiating between stroke types [10, 11]. Recent clinical studies of anti-inflammatory and neuroprotective investigational drugs for AIS have measured serum levels of pro-inflammatory mediators, such as interleukin-6 (IL-6), interleukin-1 beta (IL-1β), and tumor necrosis factor-alpha (TNFα) [12, 13]. However, no recent trials have successfully met their primary endpoint after Phase 2 despite promising preclinical blood-based biomarker data [12, 13]. This could possibly be due to species differences in post-AIS inflammatory signaling and cell death mechanisms [14]. The temporal, and often transient, dynamic changes in the serum concentration of these proteins in AIS patients, which may vary based on stroke severity and/or type, could also contribute to a lack of meaningful readouts. The primary objective of this study was to build upon our previous pharmacology studies by identifying a prognostic biomarker that correlates with NSC EV activity and remains informative across relevant therapeutic windows for extended dosing paradigms. A robust rat transient middle cerebral artery occlusion (tMCAO) model presents an opportunity to develop a biomarker that can directly bridge preclinical data to clinical endpoints. In the absence of validated biomarkers for stroke outcome, we selected S100 calcium-binding protein B (S100B), intercellular adhesion molecule-1 (ICAM-1), and neurofilament light chain (NfL) based on our NSC EV pharmacology data as well as studies published by other groups. Previous studies have demonstrated that S100B correlates with infarct size and neurological outcomes in both preclinical rat models and human patients [15–17]. Similarly, elevated ICAM-1 levels have been associated with worse outcomes in AIS patients [18]. In addition, ICAM-1 plays a key role in blood–brain barrier disruption and leukocyte infiltration, supporting its relevance as a candidate biomarker given our data showing that NSC EV treatment mitigates hemorrhagic transformation in a pig stroke model [5]. Finally, NfL has emerged as a promising biomarker for AIS prognosis. NfL is a well-established 13 Translational Stroke Research (2026) 17:54 marker for neuroaxonal injury or degradation and has been studied in many neurological diseases including multiple sclerosis [19], amyotrophic lateral sclerosis (ALS) [20], traumatic brain injury (TBI) [21], and Alzheimer’s disease [22–26]. Clinical studies have shown a strong asso (...truncated)