CEO, Neuro-Innovators
Neural drug development has a math problem: roughly 5.6% of molecules entering Phase I reach FDA approval, and fewer still become commercial successes. After decades of single-molecule failures, Howison Schroeder is betting the answer isn’t a newer molecule—it’s a smarter combination of three existing ones, paired with structured rehabilitation.
As CEO of Neuro-Innovators, Schroeder is advancing a 505(b)(2) strategy built around metformin, cilostazol, and telmisartan, with a trial anchored at Spalding Rehabilitation Hospital—ranked #2 in the U.S. by U.S. News & World Report—and validated by a Fugl-Meyer outcome measure the FDA has already accepted in the stroke space. For a patient population with no standard of care and no ICD-10 diagnostic code, the stakes are significant.
Moe: How did the combination of drug therapy plus rehabilitation become the core of what Neuro-Innovators is building?
Howison Schroeder spent roughly 30 years in neurosciences, much of it working on neural functional measurement tools—specifically eye tracking. What he found was that when people performed these tests frequently, they functioned as exercises, even for reflexive or subconscious responses.
That observation led to an idea: combine a drug that promoted neuroplasticity—something like fluoxetine—with that kind of structured activity, and see if the result was an accelerated response. He didn’t pursue it at his prior company, but the patent was issued last year, and Neuro-Innovators built on essentially the same concept.
“The brain changes in response to stimuli or activity. We know that stroke rehabilitation works—it just doesn’t work well enough to change lives consistently. So if we can add a little fertilizer, maybe a little Miracle-Gro to the brain, can we help it rewire and rebuild itself through those necessary activities?”
The underlying principle is Hebbian—neurons that fire together wire together. The goal is to do the exercise while improving the brain’s environment to respond to that stimulus.
Moe: Why does the 505(b)(2) pathway fit this development strategy?
The challenge with repurposing a single existing drug, Schroeder explains, is commercial: without meaningful IP protection, there’s little incentive to fund a clinical trial. A physician might do it with an NIH grant, but generating the commercial momentum to make it widely adopted is a different problem.
The bigger insight, though, is scientific. The persistent failure of single new molecules points toward polypharmacology—the idea that the brain may need more than one mechanism of action working simultaneously.
“There are a few examples out there of people working with two or more drugs. And we find there’s a lot of synergy when we get up into the brain.”
For Neuro-Innovators, the combination settled at three: metformin, cilostazol, and telmisartan. Two felt insufficient for the variety of mechanisms needed. Four risked confusing the regulatory process and creating a different set of problems. Each of the three follows a different metabolic pathway out of the body—a deliberate safety design—and together they cover enough mechanisms of plasticity that Schroeder believes the probability of efficacy is meaningfully high. He also notes that, for example, the dose of metformin needed to induce neuroplastic responses in the brain is not correlated with the dose needed to affect insulin levels—which creates additional IP opportunity as the program matures.
The trial will be listed on ClinicalTrials.gov shortly.
“With the failure of all those new single molecules, it’s clear that we should be thinking about polypharmacology in the brain, not just monopharmacology.”
Moe: How did the Spalding Rehabilitation Hospital collaboration come together?
Through their scientific advisory board. Neuro-Innovators connected with George Wittenberg, a neurologist and neuro-rehab stroke specialist in Pittsburgh. Wittenberg pointed them toward colleagues at Spalding, where Dr. Paolo Bonato and Dr. Wang had already conducted a series of single-drug studies—adding one neuroplastic agent to see whether neural recovery from stroke could be enhanced.
Their findings were instructive. The single-drug approach couldn’t generate a statistical signal across the full population, but it did surface signals in specific subgroups. In a levodopa study, for instance, patients with low dopamine showed a measurable response—not large, but statistically present.
“They got very excited about what we are doing here and have been hugely supportive.”
Once a signed collaboration was in place, getting to IRB approval required navigating the Mass General Brigham pharmacy department, which had to review and approve the prescribing of all three compounds in combination. Schroeder acknowledges the bureaucracy has moved more slowly than the team might have liked, but frames it as a worthwhile tradeoff: “I think it’s offset by the pedigree we get by working with these just phenomenal and wonderful supportive people.” Spalding is ranked #2 in the U.S. for rehabilitation by U.S. News & World Report. The study is soon to be launched.
Moe: How does the trial’s measurement methodology address the known difficulties of blinding and objective assessment in neuro-rehab?
Schroeder traces his thinking on measurement back to his eye-tracking work—years spent trying to convince drug companies that sensitive functional measurement tools could objectively capture drug-induced changes in the brain. That experience sharpened his view of why neural drug trials fail.
He identifies four recurring problems: monopharmacology, blood-brain barrier penetration, toxicity in humans despite tolerability in animals, and—critically—the absence of an objective functional outcome measure the FDA will accept.
Chronic stroke survivors, he argues, solve that fourth problem.
“The correlation of the Fugl-Meyer score and the robotic output we’re going to get here titrate really, really well with outcomes.”
The FDA has already demonstrated willingness to accept the Fugl-Meyer as an outcome measure in the stroke space—specifically in the approval of the Vivistim™ device, which uses vagus nerve stimulation and was cleared using Fugl-Meyer as its primary endpoint. That precedent matters.
Moe: Why does this matter so much for the chronic stroke patient population specifically?
Stroke survivors, Schroeder notes, exist in a regulatory and clinical blind spot. There is no ICD-10 diagnostic code for them in the United States—or in the UK, based on conversations with a UK-based AI health entity. Without a diagnostic code, there is no database, no way to systematically track what has worked or hasn’t. He describes them as “ostensibly an abandoned population of disabled humans.”
The clinical threshold the program is targeting is a five-point change on the Fugl-Meyer scale, which runs to 66 points.
“If we can get a five-point change, we affect something clinically—something that patients can feel at home. They can suddenly put their hand up in the cupboard and pull out the jam jar to spread on their bread without help. That’s huge.”
Neuro-Innovators believes it can reach that threshold, or better.
Moe: How does the 505(b)(2) pathway change the economics of getting to market?
Because the three compounds already have extensive safety data on file with the FDA, Neuro-Innovators does not need to generate years of new preclinical safety work.
“We think we can make this drug commercial for—probably not more than $60 million.” He adds that the FDA may impose additional requirements that affect that figure, but even at that spend level, the economics are compelling: the U.S. market for chronic stroke survivors could be as high as $20 billion. “It’s not hard to get a good return on such a low capex investment and such a quick time to market.”
Howison Schroeder is CEO of Neuro-Innovators, a clinical-stage company developing combination neuroplasticity therapies for chronic stroke rehabilitation.
Moe Alsumidaie, MBA, MSF, is founder and Chief Editor of Vanguard Publications, which publishes Clinical Trial Vanguard, Pharma Vanguard and BullScope, and Head of Research at CliniBiz. He has two decades in clinical trial operations and data science, with earlier roles at Genentech, Abbott Vascular and Stanford University Medical Center, and is a guest lecturer in clinical trial sciences at Rutgers University.
