By Satish Casie Chetty, Daniele Bertolini, David Li-Bland

At Unlearn, pushing the boundaries of technology where artificial intelligence (AI) and medicine converge is at the forefront of our mission. Our commitment to this mission has led to significant advancements in the field. To date, we have successfully developed and launched our novel Digital Twin Generator (DTG) technology, which has been tailored for chronic progressive neurological diseases such as Alzheimer’s (AD), amyotrophic lateral sclerosis, Huntington’s disease, frontotemporal dementia (FTD), and Parkinson’s disease. Recently, we expanded our scope to include inflammation and immunology diseases, introducing our DTG for Crohn's disease. Today, we are thrilled to announce a significant step forward in our efforts to drive clinical innovation as we release our DTG for ischemic stroke within the realm of cerebrovascular diseases.

Globally, according to estimates from the World Health Organization, stroke is the second-leading cause of death and the third-leading cause of disability. In the United States alone, more than 795,000 people suffer from strokes yearly, with ischemic strokes comprising roughly 87% of cases. An ischemic stroke occurs when a blood clot obstructs a vessel that supplies blood to the brain, depriving brain tissue of oxygen and nutrients, ultimately leading to the death of brain cells, known as neurons. The brain's affected regions experience disruption in their functions, resulting in impairment and disability.

Importantly, ischemic stroke is an acute condition associated with rapidly developing symptoms, including body paralysis, speech and vision impediments, and vision loss. Patients with ischemic stroke experience a sudden debilitating event followed by a slow recovery with treatment and physical rehabilitation. Thus, the progression of stroke is vastly different from that of prior chronic progressive diseases we have worked on, such as AD and FTD.

Our Stroke DTG 1.0 focuses on acute treatment trials that target the removal of the clot either by chemical or mechanical intervention. These trials use global disability endpoints (such as the modified Rankin Scale) assessed 90 days after the initial stroke event.  While current standards of care, namely alteplase and endovascular thrombectomy, effectively treat patients in the acute stage of ischemic stroke, there is ongoing research to develop even more effective therapies with fewer side effects, particularly intracranial hemorrhages. The primary barrier to developing these new drugs is patient recruitment in stroke trials. Greater than 80% of stroke clinical trials have reported low recruitment, potentially leading to underpowered studies.

Our Stroke DTG 1.0 offers a promising solution to the recruitment issue in these trials. This state-of-the-art generative AI model is explicitly designed to forecast a comprehensive longitudinal trajectory of a patient’s recovery. In particular, it predicts common endpoints used in acute treatment trials, including the modified Rankin Scale, NIH Stroke Scale, and Barthel Index at 90 days post-randomization, as well as other safety measures.

A comprehensive list of inputs, outputs, and predictive performance of our model can be found in the Stroke DTG 1.0 specifications sheet on our website. When combined with our European Medicines Agency-qualified method, this DTG enables the design of acute treatment trials with fewer patients in the control arm, all while maintaining statistical power. Reduced enrollment targets lead to expedited timelines and reduced costs for trial sponsors while helping to bring new therapies to patients sooner.

Contact us to discover how our groundbreaking AI-powered DTG technology can significantly accelerate your clinical programs in ischemic stroke.