The ALS Clinical Trials Landscape in 2026

Why biomarkers now define ALS drug development

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disorder that destroys the upper and lower motor neurons controlling voluntary movement, leading to muscle weakness, paralysis, and ultimately respiratory failure. Worldwide prevalence is estimated at roughly 290,000 to 360,000 people, a figure projected to climb toward 400,000 by 2040 as populations age.¹ Median survival from symptom onset is typically two to five years, and for most of the past three decades the therapeutic toolkit was limited to riluzole and edaravone — agents that extend survival only modestly.¹

The field is now in the middle of a structural shift. After a long run of failed trials built on broad, one-size-fits-all hypotheses, two developments have changed how sponsors design studies. First, the 2023 accelerated approval of QALSODY (tofersen) demonstrated that a genetically defined subtype of ALS could be treated, and that a fluid biomarker — not just a clinical rating scale — could anchor a regulatory decision.² Second, the 2024 withdrawal of Relyvrio after the large Phase 3 PHOENIX trial failed reminded the community how costly it is to advance a therapy without a biological readout that tracks target engagement.³

Together these events have pushed biomarkers from a supporting role to the center of ALS research. Sponsors increasingly want to stratify patients before enrollment, confirm that a drug reaches its molecular target, and read out a pharmacodynamic effect months before clinical decline would become measurable. This article maps the current ALS clinical trials landscape, reviews the staging systems and biomarkers that structure it, and examines why TDP-43 — the protein that defines the pathology of the overwhelming majority of ALS cases — remains the field’s most important unmet measurement challenge, and how neuron-derived exosome platforms such as NeuroDex’s ExoSORT™ aim to close that gap.

The ALS clinical trials landscape in 2026

From broad neuroprotection to genetic precision

Historically, ALS trials tested broadly acting compounds — anti-excitotoxic, anti-inflammatory, anti-apoptotic, and neurotrophic agents — across heterogeneous patient populations. A review of programs between 2020 and 2022 catalogued more than 50 candidates spanning at least eight mechanistic classes, with the revised ALS Functional Rating Scale (ALSFRS-R) serving as the dominant primary endpoint.¹ The signal-to-noise problem in such designs is severe: ALS progresses at very different rates between patients, so detecting a true drug effect against that variability requires large cohorts and long follow-up.

Tofersen reframed the question. An antisense oligonucleotide that degrades SOD1 messenger RNA, it targets the roughly 2% of ALS patients who carry a SOD1 mutation. The FDA granted accelerated approval in April 2023 on the basis of reductions in plasma neurofilament light chain (NfL), with the advisory committee voting unanimously that the NfL reduction was reasonably likely to predict clinical benefit.²⁴ Confirmatory evidence is being gathered in the presymptomatic ATLAS study. The lesson for the wider field was that a genetically stratified population plus a validated biomarker can compress the path to approval.

Platform trials and the current pipeline

Parallel testing has become the dominant efficiency strategy. The HEALEY ALS Platform Trial, run out of Massachusetts General Hospital, evaluates multiple investigational regimens against shared placebo controls, reducing cost and accelerating go/no-go decisions. One regimen that showed promise — pridopidine — has now advanced into a dedicated Phase 3 confirmatory study, PREVAiLS, which began recruiting in early 2026 to test the drug in people with early, rapidly progressive ALS.⁵

Representative directions in the 2026 pipeline include:

• Genetically targeted antisense and gene therapies, extending the tofersen model to other monogenic forms such as C9orf72– and FUS-associated ALS.
• Protein-homeostasis and aggregation-focused approaches, including antisense candidates aimed at upstream drivers of motor-neuron death.
• Neuroinflammation and combination strategies, reflecting the recognition that ALS is biologically heterogeneous and unlikely to yield to a single mechanism.
• Re-analysis of failed or withdrawn programs — such as the PHOENIX dataset — to mine survival and biomarker signals that can inform future trial design.³

The connecting thread is patient selection and objective readouts. Whether a sponsor is enrolling a monogenic subgroup or a sporadic-ALS cohort, the ability to define who is most likely to respond, and to measure a biological effect early, is now a competitive necessity rather than a scientific luxury.

ALS clinical staging systems and how they connect to biomarkers

“Staging” in ALS describes how far the disease has progressed, and it underpins both clinical communication and trial design. Two complementary systems are widely used.

King’s clinical staging

The King’s system assigns five stages. Stages 1 to 3 reflect the number of central nervous system regions involved — bulbar, upper limb, and lower limb — as the disease spreads anatomically. Stage 4 marks nutritional failure (requiring gastrostomy) or respiratory failure (requiring non-invasive ventilation), and Stage 5 is death.⁶ King’s staging has its greatest resolution in early-to-mid disease, making it useful for enrolling patients into trials at comparable points of anatomical spread.⁷

Milano-Torino (MiToS) functional staging

The MiToS system derives six stages from loss of independent function across domains captured by the ALSFRS-R — swallowing, communication, movement, and breathing. It provides finer resolution in later disease, complementing King’s staging; analyses of trial datasets recommend reporting both.⁷

Crucially for diagnostics developers, staging and biomarkers are linked. Higher King’s stages correlate with higher cerebrospinal-fluid NfL levels and with neuroimaging measures of white-matter degeneration, and staging systems are explicitly used to stratify patients when validating new biomarkers and as exploratory outcome measures in trials.⁶ A biomarker that tracks the underlying pathology — rather than only downstream axonal injury — would let sponsors map molecular change onto clinical stage with far greater precision, which is exactly the gap that TDP-43 measurement could fill.

Existing ALS biomarkers: the NfL anchor and its limits

Neurofilament light chain (NfL)

NfL is the most validated fluid biomarker in ALS. It is a structural scaffolding protein of myelinated axons; when neurons are injured or die, NfL is released and becomes detectable at elevated levels in cerebrospinal fluid and blood.⁸ In ALS, NfL functions as a prognostic marker (higher levels predict faster progression and shorter survival), a response marker, and a potential safety signal. In presymptomatic SOD1 carriers, serum NfL can rise up to a year before clinical symptoms appear.²

Its strategic value was confirmed by tofersen, which lowered plasma NfL by roughly 40–50% over six months — the data that supported accelerated approval.⁴ Because NfL can reveal a drug effect earlier and with greater sensitivity than clinical scales, it enables shorter trials with smaller cohorts.⁹ A formal FDA biomarker-qualification effort is underway, though as of 2026 there is still no FDA-cleared NfL assay for routine clinical use, and challenges of assay standardization and reference ranges remain.⁸⁹

What NfL cannot tell you

NfL’s great strength — sensitivity to neuro-axonal damage — is also its limitation. It is a marker of generalneurodegeneration, not of any specific disease mechanism. Elevated NfL appears across multiple sclerosis, Alzheimer’s disease, traumatic brain injury, and other conditions, so it cannot by itself confirm ALS pathology or report on the protein that actually drives it. For a disease-modifying therapy aimed at a specific pathogenic protein, sponsors need a biomarker tied directly to that protein. In ALS, that protein is TDP-43.

TDP-43: the central pathology and the measurement gap

TAR DNA-binding protein 43 (TDP-43) is a nuclear protein that normally regulates RNA metabolism, splicing, and transport. Under pathological stress it mislocalizes from the nucleus into the cytoplasm, becomes hyperphosphorylated, and aggregates — a hallmark found in more than 95% of ALS cases and in 40–45% of frontotemporal dementia (FTD), placing the two diseases on a shared molecular spectrum.¹⁰¹¹ TDP-43 pathology also appears in limbic-predominant age-related TDP-43 encephalopathy (LATE) and as a co-pathology in Alzheimer’s disease, widening the potential clinical relevance of a reliable assay.¹²

Despite being the defining feature of ALS pathology, TDP-43 has been notoriously difficult to measure in living patients. Attempts to quantify TDP-43 or its phosphorylated and fragmented forms directly in cerebrospinal fluid or plasma have struggled with low sensitivity and poor reproducibility, while peripheral blood is dominated by TDP-43 from non-neuronal sources, producing high background and weak brain specificity.¹¹ This “diagnostic invisibility” is a genuine bottleneck: without a brain-specific TDP-43 readout, trials cannot easily stratify patients by TDP-43 burden, and developers of TDP-43-directed therapies have no clean pharmacodynamic marker.

Progress is accelerating on two fronts. Ultrasensitive immunoassays — for example, the NULISA platform — have detected elevated phosphorylated TDP-43 (pTDP-43) in the plasma of ALS patients and have linked plasma TDP-43 to features of advanced LATE pathology.¹³¹¹ In parallel, attention has turned to extracellular vesicles (EVs) as carriers that can shuttle TDP-43 out of neurons and across the blood–brain barrier into the circulation, where it can be sampled non-invasively.¹²

Neuron-derived exosomes and the ExoSORT™ platform

Why neuron-derived exosomes change the equation

Exosomes — a class of small extracellular vesicles, typically under ~150 nm — are released by virtually all cells, including neurons, and carry a protected cargo of proteins and nucleic acids that reflects the state of their cell of origin. Because they cross the blood–brain barrier, neuron-derived exosomes (NDEs, also called neuron-derived EVs) offer a way to interrogate central nervous system biology from a simple blood draw.¹⁴ The catch is enrichment: NDEs are only a minor fraction of total circulating EVs, so the entire approach depends on isolating the neuronal subset with high specificity. Early methods relied on the L1CAM adhesion molecule for immunocapture, but the specificity of L1CAM as a neuronal marker has been debated.¹⁵

How ExoSORT™ isolates the neuronal signal

NeuroDex, a CLIA-certified diagnostics company based in Natick, Massachusetts, built its ExoSORT™ platformspecifically to address that enrichment problem. Rather than relying on a single, contested marker, the published ExoSORT method captures NDEs using proprietary antibodies against the neuronal proteins GAP43 and NLGN3 (neuroligin-3), conjugated to magnetic beads after a total-EV precipitation step. The captured vesicles show the expected EV characteristics — vesicular morphology under electron microscopy, an average diameter around 127 nm, enrichment of canonical EV markers such as CD9 and FLOT1, and strong depletion of contaminating plasma proteins like albumin and ApoA1.¹⁵ The result is a brain-enriched compartment in which biomarkers can be measured with the cell-of-origin specificity that whole-plasma assays lack.

TDP-43 inside neuron-derived exosomes

This is where NDE isolation and the TDP-43 problem intersect. By measuring TDP-43 within neuron-derived vesicles instead of in bulk plasma, the approach sidesteps much of the peripheral background that has frustrated direct plasma assays. NeuroDex has reported that TDP-43 — alongside LC3, an autophagy marker — can be measured in NDEs and differs between ALS patients and controls.¹⁶ The independent literature is consistent with this direction: longitudinal studies have found that an exosomal TDP-43 ratio changes measurably over months of ALS follow-up, and a 2024 Nature Medicine study established plasma EV tau and TDP-43 as diagnostic biomarkers across the FTD–ALS spectrum.¹⁷¹⁰

The platform’s clinical relevance has already been tested inside a therapeutic trial. In NeuroSense Therapeutics’ Phase IIa study of PrimeC in ALS, ExoSORT™ was used to measure NDE biomarkers including TDP-43 and LC3, with the treatment associated with statistically significant changes in those markers — an early read on biological activity of exactly the kind sponsors increasingly require.¹⁶ For a deeper primer on the biology, NeuroDex maintains a dedicated resource on TDP-43 in ALS and neurodegeneration.

Beyond ALS: a generalizable diagnostic engine

The same neuron-derived-EV strategy extends across neurodegeneration. NeuroDex has received grants from The Michael J. Fox Foundation to apply ExoSORT™ to alpha-synuclein and lysosomal-function biomarkers for Parkinson’s disease, including work alongside the landmark PPMI cohort.¹⁸ Reported NDE alpha-synuclein assays have distinguished synucleinopathies from healthy controls with an area under the curve of 0.86, and Lewy body dementia from controls at 0.91.¹⁹ More recently, NeuroDex’s ExoSORT work featured in a 2026 Journal of Clinical Endocrinology & Metabolism paper showing that CNS Akt–mTOR pathway engagement by GLP-1 and PPARγ drugs can be detected from neuron-derived vesicles in plasma — extending the platform from diagnosis toward pharmacodynamic monitoring of therapies.²⁰

Where the field is heading

Three trends will define the next phase of ALS therapeutic development, and each strengthens the case for brain-specific molecular biomarkers.

• Biomarker-defined trials. Patient stratification by molecular subtype — genetic status today, TDP-43 burden tomorrow — will become standard for enrollment and endpoint design.
• Surrogate endpoints. Tofersen showed regulators will accept a fluid biomarker as reasonably likely to predict benefit; the next generation of surrogates will need to be tied to specific pathogenic proteins, not only to general axonal loss.
• Theranostics and pharmacodynamics. As therapies target TDP-43 and related pathways directly, the same assay used to diagnose and stage a patient can confirm that a drug is hitting its target in the brain.

NfL will remain the workhorse for tracking neurodegeneration, but it cannot report on mechanism. The decisive advances will come from biomarkers that read the disease’s defining biology — above all TDP-43 — with the cell-of-origin specificity that neuron-derived exosome platforms are designed to deliver. For sponsors building the next wave of ALS programs, the question is shifting from whether to use brain-specific blood biomarkers to which platform can measure the right protein reliably. To discuss NDE biomarker support for an ALS or neurodegeneration program, explore the NeuroDex ExoSORT™ platform.

References

1. Jiang J, Wang Y, Deng M. New developments and opportunities in drugs being trialed for amyotrophic lateral sclerosis from 2020 to 2022. Front Pharmacol. 2022. doi:10.3389/fphar.2022.1054006
2. Biogen. FDA Grants Accelerated Approval for QALSODY™ (tofersen) for SOD1-ALS (2023). investors.biogen.com
3. Amylyx Pharmaceuticals, Inc.; Withdrawal of Approval of NDA for RELYVRIO. Federal Register (2025); see also The ALS Association statement on the PHOENIX trial. federalregister.gov
4. Quanterix. FDA Accelerated Approval of Tofersen Highlights Importance of Blood Neurofilament Light Chain as Surrogate Endpoint (2023). quanterix.com
5. Mass General Brigham. What’s the latest on ALS research and clinical trials? (PREVAiLS / pridopidine, 2026). massgeneralbrigham.org
6. Balendra R, Al-Chalabi A, et al. A standard operating procedure for King’s ALS clinical staging. Amyotroph Lateral Scler Frontotemporal Degener. 2019. doi:10.1080/21678421.2018.1556696
7. Fang T, et al. Comparison of the King’s and MiToS staging systems for ALS. Amyotroph Lateral Scler Frontotemporal Degener. 2017. PubMed 28054828
8. Advancing Clinical Use of Neurofilament Light Chain: Translational Insights From Research to Routine Practice. 2025. PMC12575937
9. Benatar M, et al. Biomarker Qualification for Neurofilament Light Chain in ALS: Theory and Practice. Ann Neurol. 2024. doi:10.1002/ana.26860
10. Chatterjee M, et al. Plasma extracellular vesicle tau and TDP-43 as diagnostic biomarkers in FTD and ALS. Nat Med. 2024;30:1771–1783. nature.com
11. Plasma TDP-43 is a potential biomarker for advanced LATE neuropathologic change (NULISA pTDP-43 elevated in ALS). Mol Neurodegener. 2025. doi:10.1186/s13024-025-00910-4
12. Extracellular vesicles in TDP-43 proteinopathies: pathogenesis and biomarker potential. 2025. PMC12153176
13. TDP-43-proteinopathy at the crossroads of tauopathy: on copathology and current and prospective biomarkers. Front Cell Neurosci. 2025. doi:10.3389/fncel.2025.1671419
14. Zhao X, Huang S. Plasma extracellular vesicle: a novel biomarker for neurodegenerative disease diagnosis. 2024. doi:10.20517/evcna.2024.56
15. Synaptic proteins in neuron-derived extracellular vesicles as biomarkers for Alzheimer’s disease: novel methodology and clinical proof of concept (NeuroDex ExoSORT GAP43/NLGN3 method). 2023. PMC10568955
16. NeuroSense Therapeutics. Peer-Reviewed Publication of PrimeC Phase IIa ALS Study — ExoSORT™ measurement of NDE TDP-43 and LC3 (2022). prnewswire.com
17. Chen P-C, et al. Exosomal TAR DNA-binding protein-43 and neurofilaments in plasma of ALS patients: a longitudinal follow-up study. J Neurol Sci. 2020. sciencedirect.com
18. NeuroDex Inc. Receives Grant from The Michael J. Fox Foundation to Advance Biomarker Research for Parkinson’s Disease (2024). biospace.com
19. Neuron-Derived Extracellular Vesicles as a Blood-Based Biomarker for Alpha-Synuclein Across Synucleinopathies (ExoSORT). 2025. PMC12725625
20. Evers BM, Watson C, Abbasi M, Haque K, Verma A, Eitan E, Rasgon N, et al. Detection of CNS Akt-mTOR engagement by GLP-1 and PPARγ agonists via neuron-derived vesicles. J Clin Endocrinol Metab. 2026. doi:10.1210/clinem/dgag176