When the long cabling of nerve cells die off in neurodegenerative diseases, they don’t just wither away passively. Instead, there’s an enzymatic signaling cascade that coordinates axon die-off. One key executioner is a protein called SARM1, which responds to sensing a dip in NAD level by chewing up the rest of the NAD in the cell, tanking the level of this essential coenzyme.
Pharmaceutical companies are beginning to test the therapeutic hypothesis that blocking SARM1 could save neurons that would otherwise be doomed to die. They hope to intervene in difficult-to-treat diseases like ALS and chemotherapy-induced neuropathy, multiple sclerosis, and glaucoma.
Two companies, Eli Lilly and Company and Nura Bio, have completed Phase 1 trials of SARM1-inhibiting molecules and announced that they are advancing them in clinical trials.
But this class of active site inhibitors also has a crucial possible liability. At low doses, when mixed with a little active SARM1, some have turned out to inadvertently speed up axon loss in animal models. So a second generation of inhibitors, which capture the enzyme in an inactive state, is beginning to enter first-in-human trials now.
“Where we began, 8 years ago, it was a novel biology story,” says Shilpa Sambashivan, CEO of Nura Bio. “Today it’s a very exciting, novel clinical story.”
SARM1 breaks down NAD, killing axons
While normal axons die when severed from their cell bodies, axons from animals with no SARM1 can survive. When researchers made that observation, in several labs concurrently, they weren’t sure what SARM1 did. “For the longest time, everyone actually thought SARM was going to be an immune signaling protein,” says Sambashivan, because of SARM1’s amino acid sequence, which resembles innate immune proteins of the inflammasome.
“Where we began, 8 years ago, it was a novel biology story. Today it’s a very exciting, novel clinical story.”
But based on biochemistry carried out in Aaron DiAntonio and Jeffrey Milbrandt’s labs at Washington University in St. Louis, researchers found that SARM1 itself is an enzyme. It cleaves the critical coenzyme NAD, leaving ADP-ribose (ADPR). (SARM1 also can generate cyclic ADPR.) The enzyme’s active site is usually concealed, but when cellular NAD drops—which can happen after axon injury or infection—it unfurls, breaking down more NAD in a deadly positive-feedback loop. Without NAD, mitochondria in the axon falter. “You can kill just part of a cell by poisoning its metabolism,” DiAntonio says, “and that’s what SARM does.”
Researchers still aren’t sure whether increased ADPR or reduced NAD is more important to cell killing—but they knew that blocking SARM1 might be beneficial.
SARM1 base-exchange inhibitors are a mixed bag
One of the reactions the enzyme SARM1 can catalyze is hydrolysis of the coenzyme nicotinamide adenine dinucleotide (NAD) into its component nucleosides, nicotinamide (NAM) and ADP-ribose (ADPR).
“Everyone’s screens always find the same kind of molecule,” says DiAntonio, who cofounded a company, Disarm, to hunt for SARM1 inhibitors based on his academic work. (Lilly acquired Disarm in 2020).
These inhibitors can behave in puzzling ways; although assays showed they block SARM1 activity, they did not bind to the protein on their own. Eventually, they proved to act as prodrugs (Neuron 2022, DOI: 10.1016/j.neuron/2022.08.017). SARM1 can catalyze a covalent bond between these inhibitors and the NAD cleavage product. The ADPR-drug conjugate, researchers found, lodges in SARM1’s active site and blocks its activity.
“Chemists don’t like it when they don’t know what’s going on,” Sambashivan says. “It was definitely a switch that was flipped when we understood the mechanism.”
Numerous companies have launched active site inhibitor programs—but not everyone had success. In a narrow dosing window, insufficient blockage of SARM1 can, paradoxically, activate the enzyme. DiAntonio says this is thought to happen when just one out of several active sites in a multimer binds to the inhibitor; that may stabilize a more active form of the others.
In a paper in Communications Chemistry, a team from AstraZeneca described the phenomenon as “a pharmacological Hydra,” alluding to the mythological monster that grows a new head each time one is severed (2026, DOI: 10.1038/s42004-026-02074-8). According to that AstraZeneca paper, one would expect to detect inhibitor-initiated activation and hastened neurodegeneration only when SARM1 is already activated. In other words, side effects could show up in a patient trial, but not in healthy volunteers.
AstraZeneca, Schrodinger, and Genentech have laid SARM1 active site inhibitors to rest in light of the risk that they could worsen neurodegenerative disease, but Nura and Lilly continue to advance their inhibitors.
In September 2025, an ALS center at Massachusetts General Hospital announced a plan to test Lilly’s SARM1-targeting candidate LY3873862 as part of an ongoing, multiarmed clinical trial. That study has not begun. Lilly spokesperson Allison Howell says in a statement emailed to C&EN that the company has pivoted and is working with regulators to plan a stand-alone study. “Our move to a standalone study was driven by FDA trial design considerations,” the statement says.
On Monday, Nura announced that they had secured $73.8 million in series B fundraising and have begun to treat people with ALS with their active site inhibitor, NB-4746, in a global study. Nura is also pursuing allosteric inhibitors of SARM1 that bind and stabilize the inactive form of the enzyme.
Allosteric inhibitors may be a safer bet
While active site inhibitors of SARM1 depend on an active enzyme to work, allosteric inhibitors lodge the enzyme in a closed, autoinhibited conformation.
This week, Nura launched a first-in-human study of its allosteric inhibitor, NB-9402. Meanwhile, the biotech start-up Sironax is running trials of an allosteric SARM1 inhibitor called SIR2501 for chemotherapy-induced neuropathy after SIR2501 received fast track designation from the FDA in May. And in April, Tenvie started first-in-human studies of TNV108, an allosteric SARM1 inhibitor that the company hopes will treat peripheral neuropathy.
These inhibitors may prove to have liabilities of their own; still, SARM1 has shown up as a possible target in a long list of diseases, spanning the peripheral, central, and ocular nervous systems. “Mice and men are not created equal,” Sambashivan says. “But with the data that we’ve seen preclinically . . . I think there is really a belief that SARM1 inhibition has pluripotent therapeutic benefit.”
Laurel Oldach is a senior editor and life sciences reporter at C&EN.
