Blood test gives probabilistic onset window for ALS risk — with a known error bar

A blood test built on a 20-year NIH study of people at high genetic risk for ALS can estimate when symptoms will strike, within an average error of about 18 months, [according to a study in Nature Medicine](https://www.nature.com/articles/s41591-026-04528-x).

ALS is a fatal neurodegenerative disease that destroys the motor neurons controlling movement. By the time weakness or slurred speech appears, most patients have only a few years to live, and no treatment meaningfully slows the disease. The new model targets the moment before symptoms begin, when the biology is still shifting and a therapy might have room to work.

Researchers analyzed 516 plasma samples collected over nearly two decades from the NIH-funded Pre-fALS cohort, which follows carriers of ALS-linked gene variants. They identified 92 proteins whose levels changed before symptom onset, then narrowed the list to a 19-protein panel that predicted conversion to clinical ALS across 0.5- to 5-year windows with accuracy between 0.80 and 0.89.

The clock has a known margin: the model's mean absolute error for time to symptom onset is 1.6 years, wide enough that the team frames the work as a tool for prevention-trial enrollment rather than a personal forecast. Findings were "partially" replicated in UK Biobank data, the paper's abstract reports, though the replication ALS sub-cohort sample size is not available in the visible source excerpt; broader-population relevance is not yet established.

"It is one thing to know that you are at risk for ALS, but it is another thing to know when symptoms are likely to emerge," senior author Michael Benatar, M.D., Ph.D., a professor of neurology and public health sciences at the University of Miami, said in the NIH release. The number now has an error bar.