August 5, 2026 | Blood-based biomarkers predicting when symptoms of amyotrophic lateral sclerosis (ALS) will emerge are expected to catapult the medical research community forward in the quest to prevent the fatal neurodegenerative disorder. The effort is being spearheaded by Michael Benatar, M.D., Ph.D., professor of neurology and public health sciences, and research associate professor Joanne Wuu, Sc.M., both at the University of Miami Miller School of Medicine. The pair have been co-leading the Pre-Symptomatic Familial ALS (Pre-fALS) study collecting longitudinal data and biological samples from people at elevated genetic risk for ALS since the project’s 2007 inception.
Their optimism has most recently been fueled by the discovery of a 19-protein panel, including neurofilament light chain (NfL), which they recently found predicts future risk of phenoconversion—a term use to describe the transition point when a person who carries a genetic risk factor for ALS (e.g., SOD1 mutations) first develops outward symptoms of the disease (Nature Medicine, DOI: 10.1038/s41591-026-04528-x). Imminent phenoconversion was predicted across timeframes ranging from six months to five years, a relatively short but practical window for a future prevention trial, says Benatar.
“Therapeutic development efforts aimed at ALS and neurodegeneration more broadly were repeatedly hitting roadblocks,” says Benatar. The principal motivation for Pre-fALS has been to overcome the challenge of delayed therapeutic intervention.
As with other neurodegenerative disorders, ALS has a presymptomatic period when the underlying pathobiology of disease is active, but symptoms have not yet emerged, he says. By developing pre-symptomatic biomarkers that reflect the underlying pathobiology of disease, it becomes possible to treat at an earlier stage.
Pre-fALS enrolls healthy individuals who are at risk for developing genetic forms of ALS, which are known to account for 10-15% of all cases. Studying these people over time is necessarily a long-term undertaking, but also “currently the only and most powerful way to get at the question of what’s happening with biology before the disease comes over the clinical horizon,” Benatar says.
The first biological signal was found about a decade into the Pre-fALS study. “We hypothesized that the levels of neurofilament, a marker for neuro-axonal injury, would be elevated before phenoconversion and that turned out to be true,” says Benatar.
Specifically, the discovery established blood NfL as a susceptibility/risk biomarker for advancing ALS therapy development, with the potential to predict which people at high genetic risk for ALS would develop disease and when (Annals of Neurology, DOI: 10.1002/ana.25276). NfL in serum and cerebrospinal fluid (CSF) were quantified using an electrochemiluminescence immunoassay.
Benatar subsequently partnered with Biogen to design and launch the ATLAS trial, in which carriers of a pathogenic variant in the SOD1 gene are being treated with an SOD1-lowering antisense oligonucleotide (tofersen) before showing any clinical signs of ALS, he says. In 2023, the FDA approved the use of tofersen for people already affected with SOD1-ALS.
For ATLAS, NfL is being used primarily as an eligibility criterion among presymptomatic adults carrying a highly penetrant SOD1 gene mutation associated with rapidly progressive ALS. Monthly blood draws are being used to monitor participants and, if their NfL levels spike, the treatment phase is initiated, Benatar explains. At that point, study participants are randomized to receive either tofersen or a placebo via an injection into the spinal fluid. If anyone shows clinical signs or symptoms of ALS, they’re immediately moved into an open-label extension phase where they are guaranteed to receive the active drug, he notes.
“The goal of ATLAS is to figure out what’s the optimal timing of giving this gene therapy,” says Benatar. “Is it enough to give it to people when symptoms appear, or is it better to give it to people when their neurofilament concentration rises?”
The study began in 2021 and is expected to wrap up in another two years, he adds. ATLAS is effectively a phase 3 study that will hopefully provide the required confirmatory evidence of tofersen’s clinical benefit. The drug already has accelerated approval based, in part, on tofersen’s effect on lowering NfL which is regarded as a surrogate for improved clinical outcomes. If ATLAS is successful, tofersen might gain full, traditional regulatory approval.
The specific NfL assay used in the clinical trial has yet to be translated into clinical practice, and “it matters what assay you use,” says Benatar. The regulatory rigors involved in this transition need to happen before doctors can start ordering the test to monitor disease progression in ALS patients.
Several NfL assays are available, he shares. ATLAS uses the Siemens Healthineers assay, which is approved in multiple international markets and as a laboratory-developed test (LDT) in the U.S. Others include the Roche Elecsys assay that recently received CE Mark approval allowing its sale in much of Europe. Both tests have been granted Breakthrough Device Designation by the Food and Drug Administration, as has the Simoa assay, an LDT of Quanterix.
Pre-symptomatic Prediction
With the latest study, Benatar and his colleagues took a less targeted approach to look beyond NfL for other protein biomarkers they might find utilizing Olink, a widely used proteomics platform that is now part of Thermo Fisher Scientific. For this analysis, participant samples and data were drawn from the Pre-fALS study as well as a companion biomarker study that recruited healthy controls and patients with clinically manifest ALS, and a multi-center natural history study run by the CReATe Consortium that included individuals with the disease.
More than 500 serially collected plasma samples were used to look for about 5,000 proteins and ask how concentrations of them changed over time, says Benatar. This identified 92 proteins whose levels differed in people before they eventually showed symptoms. The more manageable 19-protein core panel that could accurately predict future risk of phenoconversion was determined using machine learning techniques.
Study findings were partially replicated in UK Biobank data, confirming pre-symptomatic increases in several proteins and that a multi-protein panel outperformed NfL alone in estimating time to phenoconversion. The UK Biobank dataset was an imperfect, best-available replication cohort for the unique Pre-fALS study group, Benatar says.
Among the key differences with the UK Biobank cohort is that proteomic analyses were done on a one-time blood draw at the time of enrollment, all using an earlier version of Olink that only looked at about 3,000 proteins, he continues. By contrast, the Pre-fALS and related studies entailed longitudinal follow-up, providing critical insights into the temporal trajectories of protein data.
Characterizing the timing of phenoconversion among people in the population-based UK Biobank is also tricky because phenotypic data comes from linked medical and hospital records, says Benatar. “The best we have from UK Biobank is hospitalization with an ALS ICD-10 diagnosis code, so it’s an approximation of when people develop symptoms of disease.”
Beyond the availability of presymptomatic data, the value in using UK Biobank data includes the fact that most people who develop ALS have non-genetic forms of the disease. This speaks to the generalizability of the study findings to the broader at-risk population.
Identification of the biomarker set is “a very good start, but there is still a lot of work to do,” Benatar readily acknowledges. “Further work is definitely needed to validate these findings.”
The Olink platform combines antibody-based protein recognition with DNA readouts and, while widely used and powerful, is not the only way to interrogate the proteome, says Benatar. The research team has already started exploring other techniques that may see different parts of the proteome, to minimize bias from inadvertently creeping into their investigation.
They have also started looking at CSF samples provided by individuals in the Pre-fALS study, he adds. The hope is that expanding the search to other bodily fluids will uncover additional targets or confirm the ones already seen in blood.
These next steps should help narrow down the protein panel further, after which a decision must be made about the best way to do the measuring for the test to be adopted in real-world practice, says Benatar. Fantastic assays currently exist to measure NfL but it will be essential to develop similarly high-quality assays for the additional proteins we’ve identified, says Benatar.
Ultimately, a blood-based biomarker is likely to be most helpful in clinical practice because of ease of sampling, he continues, but CSF analyses are also incredibly valuable because they involve neural fluid in direct contact with the brain and spinal cord. The development of clinical tests will require significant work from collaborators, partners, and other members of the scientific community.
“Progress in the development of biomarkers with well-defined contexts of use will be essential to future successful drug discovery efforts,” says Benatar. “The most recent data to emerge from the Pre-fALS study will help to bring us closer to the goal of preventing ALS.”