NEW YORK (GenomeWeb) – Startup diagnostics firm CereDx is looking to establish rapid molecular testing to distinguish strokes from stroke mimics in a hospital setting.
Based on immune system-related biomarkers licensed from West Virginia University, the Morgantown-based firm has plans to launch a protein-based panel for point-of-care diagnostics platforms in the next two or three years.
"It's a yes or no diagnostic test for the moment a patient hits the door presenting with stroke symptoms," CereDx Co-founder and CSO Taura Barr told GenomeWeb. According to the American Stroke Association (ASA), approximately 800,000 people suffer from stroke each year. But more than 3 million patients present with stroke-like symptoms, Barr said. "There's a huge clinical need to sort out or triage these ambiguous patients," she said.
Patients not actually having a stroke need to be correctly diagnosed and patients suffering a stroke need to be placed in the proper care as soon as possible. "There's no single test or tool that does that right now," Barr said, adding that even in comprehensive stroke centers around the country, that evaluation is performed subjectively.
Strokes can be caused by either clotted (ischemic) or ruptured (hemorrhagic) blood vessels in the brain. Ischemic strokes, which can be treated with clot-busting drugs, come in several subtypes and account for more than 80 percent of cases, according to the ASA. While some previous stroke diagnostics focused on finding patients eligible for treatment, CereDx believes it has found a way to discern strokes, which are often mild, from other vague symptoms like dizziness.Â
If the company is right, it could help patients get better care and help those providers save money and more efficiently use resources like neuroimaging technologies, Barr said, adding that pinpointing the time of the stroke could be a future possibility for CereDx's technology.
At CereDx, she's building on biomarkers she discovered in her research at WVU, where's she's a professor at the Center for Neuroscience. Over a decade of research, first at the National Institutes of Health then at WVU, she has identified 50 protein biomarkers associated with stroke and has run clinical trials in 500 patients to validate biomarkers. Currently, CereDx is running a clinical trial to zero in on a subset of approximately 10 biomarkers that best identifies strokes with high sensitivity and specificity.
This new study, run at the University of Cincinnati, is looking at the biomarkers in real-world scenarios, Barr said. Any patient that comes into the emergency room with stroke-like symptoms  (and comes in less than 12 hours from the onset of symptoms) gets a blood draw.
The firm has already created the algorithm to confirm the biomarker subset, which will be tuned to reach sensitivity and selectivity in the 90 percent range. "Preliminary analysis shows we are in the high 80s," Barr said, noting that current stroke diagnosis currently has sensitivity of about 80 percent, but specificity of only 40 percent.
CereDx is certainly not the first company to try to address stroke diagnosis with an assay. For years, Proteome Sciences worked with research groups on stroke-related biomarkers, leading to studies that suggested diagnostics might even be able to pinpoint the time of stroke, critical data for determining who receives treatment. However, in September of this year, the firm told GenomeWeb it would focus more on services for pharmaceuticals and less on properties like its stroke biomarker portfolio.
Making a stroke diagnosis is especially challenging at community hospitals in the US. "Only 30 percent of hospitals have the stroke expertise to make these decisions easily," Barr said. Even in academic centers with the highest or second highest stroke ratings, Barr said a molecular diagnostic could help increase confidence a diagnosis. It's a challenge Barr is acutely aware of — in addition to holding a doctorate in translational neuroscience, she's also a registered nurse, with experience in critical care nursing, and holds an appointment as adjunct faculty at the University of Pittsburgh School of Nursing.
Because time is the critical factor in stroke care, CereDx is building a protein panel that can be run on a point-of-care platform. "Genomics approaches take too long," Barr said.
Richard Giersch, CereDx's CEO, said that the firm is looking at partnering with one of several antibody-based POC platform makers. A device already approved by the US Food and Drug Administration would be ideal, but the company is also scouting platforms that have yet to be cleared.
The Abbott i-Stat System, which is FDA-approved and implemented in many hospitals, is a top choice for CereDx, which is also vetting platforms from Minomic, Philips, and Roche.Â
CereDx's test will also need FDA approval. Giersch said the firm has a submission under review and will be meeting with FDA officials on Nov. 21 to discuss a de novo application. "The goal is to work with the FDA during this trial as well as in development and implementation of the pivotal trial," Barr said. Â
Barr said she expects the University of Cincinnati trial to wrap up in the next three to six months. Once the firm makes its biomarker selection, it can begin recruiting for the pivotal trial. Barr said recruitment could start about a year from now and the trial itself would take six to eight months, putting CereDx in the position to submit its application to the FDA in late 2018.
Giersch said that CereDx is closing its $1.75 million seed round and will be launching a Series A fundraising later this winter. "We are still determining the total amount needed, but if we raise between $8 million and $10 million that should get us through to FDA submission," he said.