Amyotrophic lateral sclerosis (ALS) and related neurodegenerative disorders are increasingly conceptualized by their underlying biology rather than solely by clinical phenotype. This shift has highlighted pre-symptomatic stages as critical windows for intervention. A key obstacle to ALS prevention trials is the inability to predict which unaffected carriers of ALS-associated pathogenic variants will phenoconvert to clinically manifest disease and the timing of that conversion. Prior work identified pre-symptomatic increases in blood neurofilament light (NfL) in some high-penetrance genotypes; however, because NfL primarily reflects axonal loss, it may be insufficient across the broader range of genotypes and disease tempos. The discovery of additional susceptibility/risk biomarkers that capture upstream pathobiology is therefore important to inform prevention-trial design and individual risk estimation.
This longitudinal, high-throughput proteomic study used the Olink Explore platform and analyzed 516 serially collected plasma samples. The samples derived from distinct groups: 33 individuals who phenoconverted to clinically manifest ALS, 35 people with clinically manifest ALS, 10 pre-symptomatic pathogenic-variant carriers who had not phenoconverted during follow-up, and 59 control participants. The Pre-Symptomatic Familial ALS (Pre-fALS) study provided an extended prospective natural-history resource in which participants were followed from pre-symptomatic states through phenoconversion, enabling time-resolved biomarker discovery.
From the longitudinal proteomic data, investigators identified 92 proteins with concentrations that changed prior to phenoconversion. The study characterized the longitudinal trajectories of these proteins to determine temporal relationships to symptom onset. Specific protein names beyond the highlighted examples were presented in the full report; the article emphasizes that biochemical signals detectable in blood precede clinical manifestations and can reflect diverse pathobiological mechanisms rather than only downstream axonal degeneration.
A core panel of 19 proteins was derived from the 92 candidates. When considered collectively, this 19-protein panel predicted phenoconversion across time horizons ranging from 0.5 to 5 years. The panel achieved crossvalidated areas under the receiver-operating-characteristic curve (AUCs) of 0.80–0.89 across those horizons and provided estimates of time to phenoconversion with a mean absolute error (MAE) of 1.6 years. These performance metrics indicate that a multi-protein approach can meaningfully stratify risk and estimate timing of clinical onset in pre-symptomatic, pathogenic-variant carriers.
The authors report partial replication of findings using UK Biobank data. This external dataset confirmed pre-symptomatic increases in several proteins identified in the discovery cohort, including NEFL (neurofilament light), EDA2R and CA3. Importantly, the multi-protein panel outperformed NEFL alone in estimating time to phenoconversion, supporting the added value of integrated proteomic signatures over single-marker approaches.
These results illuminate pre-symptomatic biology in ALS by identifying proteins whose levels shift before clinical onset. A validated panel that predicts phenoconversion and provides time-to-event estimates could improve selection and stratification of participants for prevention trials and inform optimal timing for intervention. Because biofluid biomarkers can reflect systemic and upstream pathobiology and are amenable to standardized collection and high-throughput assays, they are attractive tools for multi-center prevention efforts. The work reinforces that susceptibility/risk biomarkers beyond NfL are attainable and necessary to broaden trial eligibility across diverse genotypes and disease courses.
The source article reports the principal discovery and partial replication; detailed analytic methods, the full list of proteins and their individual trajectories, and additional replication metrics are presented in the full paper. As with any discovery effort, further validation in independent longitudinal cohorts and refinement of the panel for different genetic subgroups and disease tempos will be required before clinical deployment. The study underscores the importance of continued natural-history sampling of pre-symptomatic carriers and of external validation to strengthen confidence in biomarkers intended for use in prevention trials.
Longitudinal plasma proteomics in a well-characterized pre-symptomatic cohort identified 92 proteins changing before phenoconversion and a core 19-protein panel that predicted phenoconversion across 0.5–5 years (crossvalidated AUC 0.80–0.89) with a mean absolute error of 1.6 years for estimated time to onset. Partial replication in UK Biobank confirmed increases in several proteins, including NEFL, EDA2R and CA3, and showed that a multi-protein signature outperformed NEFL alone. These findings advance understanding of pre-symptomatic ALS biology and support the development of susceptibility/risk biomarkers to guide prevention-trial design and timing of early intervention.