AI Identifies 18 Existing Drugs That May Extend Survival for ALS Patients

A new study published in The Lancet on March 28 demonstrates how artificial intelligence can transform the way we discover treatments for incurable diseases.

Researchers at several Bay Area institutions used machine learning and causal inference analysis to examine health records of more than 11,000 ALS patients from the U.S. Department of Veterans Affairs. The results were striking: 18 already FDA-approved drugs were identified as potentially life-extending for patients with amyotrophic lateral sclerosis, the fatal neurodegenerative disease also known as Lou Gehrig's disease.

Three drug classes stood out in particular. Statins, type-5 phosphodiesterase inhibitors, and alpha-blockers all showed a positive association with longer survival. Since these medications are already approved with established safety profiles, physicians could in principle begin considering them for ALS patients now, while further clinical trials are being planned.

What is revolutionary here is not just the findings, but the method. Repurposing existing drugs saves years of approval processes. For a disease with no effective cure, where patients typically live two to five years after diagnosis, that matters enormously.

The study represents a concrete example of what many have long promised about AI in medicine: the ability to find patterns in enormous datasets that humans would never discover manually, and do it in a fraction of the time traditional research requires.

Further clinical studies are needed to confirm the findings, but the research team is optimistic. The combination of AI analysis and drug repurposing may prove to be one of the most important tools in the fight against neurodegenerative diseases.