Hillhurst Bio's HBI 002, an oral liquid that gives Parkinson's patients a controlled low dose of CO, has entered a small Phase 2a safety trial, with first data expected by the end of 2026.
For decades, researchers have tracked an uncomfortable epidemiological pattern: long-term cigarette smokers develop Parkinson's disease at markedly lower rates than people who have never smoked, even as tobacco shortens their lives in nearly every other way. The association is real and has been replicated, but no one has been able to turn it into a medicine. That is the science behind HBI-002, a new oral, at-home Parkinson's therapy that has just dosed its first patient. The drug is built around a controlled low dose of carbon monoxide, a molecule better known as a poison than as a treatment, and it comes from a small San Diego company called Hillhurst Biopharmaceuticals.
The company framed the milestone carefully: HBI-002 research "is distinct from, and does not mitigate, the well-established health risks associated with cigarette smoking," Hillhurst said in its June 15 announcement. That caveat matters. Cigarette smoke delivers carbon monoxide alongside thousands of other chemicals, including the carcinogens and particulates that drive lung cancer, heart disease, and emphysema. Pulling a single signaling molecule out of that toxic mix, at a dose many times lower than what a smoker inhales, is the bet underlying HBI-002.
The science behind that bet is older than the company. At low concentrations, carbon monoxide is not just a waste product of burning fuel. It is also one of the body's own signaling molecules, produced in small amounts during the breakdown of heme, the iron-bearing compound in red blood cells. In laboratory and animal models, low-dose CO has shown anti-inflammatory and anti-apoptotic effects, meaning it can dial back inflammation and help cells resist programmed cell death. Both processes are implicated in the slow loss of dopamine-producing neurons that defines Parkinson's disease. Researchers have also noted the striking epidemiology: some long-running studies have put the relative risk reduction among long-term smokers at up to 60%, an effect size large enough to suggest that the exposure is doing something biologically real, not just statistical noise.
HBI-002 is Hillhurst's attempt to test whether that biology can be delivered as a drug. The company describes HBI-002 as an oral liquid drug product — delivered at home, with a regimen designed for chronic use — based on its proprietary GLASS™ platform, which enables liquid delivery of therapeutic gases that were previously only available via inhalation. The format matters. Most existing Parkinson's drugs are taken as pills multiple times a day to manage symptoms, and the disease itself is progressive. An at-home oral therapy aimed at the underlying neuron loss, rather than at smoothing out the symptoms of it, is a different kind of intervention.
The new trial is a small, blinded, randomized, controlled Phase 2a study, with safety and tolerability as the primary endpoint, in 36 subjects receiving a 14-day dosing regimen. Pharmacokinetic data, which describe how the drug moves through the body, and biomarker readouts will be collected as secondary signals. Those biomarker and pharmacokinetic measures are the only early efficacy hints in this trial; clinical efficacy, meaning real changes in patient symptoms, is not the first question being asked. The trial is supported by The Michael J. Fox Foundation for Parkinson's Research (MJFF) Therapeutics Pipeline Program — which awarded $2 million according to Hillhurst's May 2025 grant announcement — and the Farmer Family Foundation, which together with Massachusetts General Hospital contributed $4.3 million to the study, according to the Hillhurst announcement, with a company-projected safety readout expected by the end of 2026 and a larger Phase 2b planned for 2027.
The scale of the disease is what makes even a safety-stage milestone worth watching. Parkinson's affects roughly 1.1 million people in the United States, and current therapies, primarily levodopa and dopamine agonists, treat the symptoms of dopamine loss rather than the neuron death that causes it. Decades of attempts to develop disease-modifying Parkinson's drugs have produced very few candidates that have held up in mid-stage trials. That long list of failures is the context against which HBI-002's mechanism will be judged: a low-dose signaling molecule, a chronic at-home pill, and a foundation-backed safety readout due in roughly six months.
"This milestone reflects our commitment to developing transformative therapies that address the underlying pathobiology of Parkinson's disease, where there remains a significant unmet need," said Andrew Gomperts, the company's CEO and President, in the release.
The honest version of the story is that one dosed patient is the earliest possible clinical milestone, not a treatment. The interesting version is the scientific bridge the trial is trying to build: a poison that the body also makes, an epidemiology that has bugged neurologists for decades, and a small oral pill designed to test whether the bridge will hold. The end-of-2026 safety readout will not answer the question, but it will tell researchers whether the bridge is safe to keep walking.