Deep Tech Opportunity: It’s time to develop better drugs to treat anxiety

There’s no mental health problem more widespread than debil­i­tating anxiety, a category encom­passing generalized anxiety disorder, social anxiety disorder, panic disorder, and phobias. According to the NIH, about 19 percent of adults had an anxiety disorder in the past year — more females (23.4 percent) than males (14.3 percent) — and about 31 percent of American adults will experience an anxiety disorder at some point in their lifetimes.

The World Health Orga­ni­za­tion says the condition affects more than 300 million people globally, but that only 28 percent of people who need treatment actually get it. Cognitive behavioral therapy (CBT) teaches important skills for dealing with anxiety. It works even better in combination with drug treatment such as selective serotonin reuptake inhibitors (SSRIs) and serotonin-norep­i­neph­rine reuptake inhibitors, also commonly used as anti­de­pres­sants. Benzo­di­azepines can also help for a limited period, but they can cause fatigue, cognitive impairment, and addiction.

But the truth is that no existing therapy is as effective as medical profes­sionals and patients would like. Most people relapse after treatment ends — in one study, anxiety returned for 26 to 45 percent of patients who discon­tinued anti­de­pres­sants, and another study showed that 48 percent of patients who received CBT were still symptomatic 2 to 14 years later.

“There’s a real need in this space for good anxiety drugs, without the side effect profiles of benzo­di­azepines and other medicines, that can treat general anxiety, social anxiety, PTSD, and a number of different indications,” says DCVC Bio Managing Partner Dr. John Hamer.

Our fund placed its first bet in this area in February 2025, partic­i­pating in a Series A financing round for Boston-based Newleos Ther­a­peu­tics. The company launched with four drugs in its pipeline, all licensed from Roche. Newleos’ lead drug candidate, NTX-1955, is a molecule that enhances signaling in the brain by gamma-aminobu­tyric acid, or GABA. GABA is an inhibitory neuro­trans­mitter that lets chloride ions into neurons and makes them less likely to fire. By slowing down this firing, the GABA system helps to keep us calm, which is why it’s already the target of numerous anti-anxiety drugs.

We’re also watching with high hopes as other companies explore new approaches to treating anxiety:

• A drug called etifoxine has dual effects on the GABA system. It binds to certain subunits of the GABA_A receptor directly, enhancing its action, while also stimulating production of neuros­teroids that themselves up-regulate the receptor. Etifoxine has long been approved for use against anxiety in France and a few smaller markets, and it’s considered as effective as benzo­di­azepines, but with fewer side effects. California-based GABA Ther­a­peu­tics has already moved into clinical trials of GRX-917, an analog of etifoxine that’s easier to metabolize.
• Swiss startup Synendos Ther­a­peu­tics is in phase 2 clinical trials of a selective endo­cannabi­noid reuptake inhibitor called SYT-510. Endo­cannabi­noids are natural substances in the body that activate the same receptors that are sensitive to phyto­cannabi­noids such as THC, the psychoac­tive molecule in cannabis. By limiting the reab­sorp­tion of endo­cannabi­noids by neurons, the drug would, in theory, help restore dysreg­u­lated neuro­trans­mis­sion in the brain, treating anxiety and mood disorders.
• Tezampanel, which blocks receptors for glutamate, the most common neuro­trans­mitter in vertebrates, was developed by Lilly as a treatment for migraines but later shelved. Animal studies show it may be an effective anti-anxiety drug. 

In general, we like the idea of refilling the pipeline of potential anxiety treatments by restarting research on shelved, abandoned, or orphan compounds. There’s even hope that AI could accelerate this repurposing process. One group of researchers writing last year in Nature Medicine showed that a graph neural network model, when allowed to comb through decades of medical literature, was efficient at exploring shared molecular pathways and pathologies and identifying candidate drugs for conditions that lack approved treatments.

We know that the world needs better anxiety treatments as soon as possible — and the advantage of the Newleos-style in-licensing approach is that when a drug has a known safety and efficacy profile, moving it to the clinic can be far faster and cheaper.