STTR Phase II: Optimized manufacturing and machine learning based automation of Endothelium-on-a-chip microfluidic devices for drug screening applications.

Biochip Labs, Inc., Cleveland OH

Investigators

Abstract

The broader impact/commercial potential of this Small Business Technology Transfer (STTR) Phase II project is to address the unmet need in the companion diagnostic guided therapy market for Sickle Cell Disease (SCD). SCD is a lifelong disease affecting millions of people worldwide. Emerging therapies are estimated to be $150k-$200k per patient each year. A companion diagnostic cost in SCD anti-adhesive therapies is estimated at least $3,000 per patient. With improved accessibility to patients living in low- and middle-income countries and scalable curative therapies, the global SCD treatment market size is projected to increase to $8.75B by 2029. Additionally, companion diagnostic-guided drugs have an increased regulatory approval probability of 50% in Phase III clinical trials. The proprietary Endothelium-on-a-chip platform with human donor cells provides a physiologically relevant means to study blood-endothelium interaction. This platform can be integrated into preclinical studies to screen the effect of novel drug candidates as well as for assessment of drug toxicity. In Phase I of the STTR project, standards and quality control criteria for experimental conditions on the Endothelium-on-a-chip were established. The continuing projects with pharmaceutical companies have highlighted the need to scale the manufacturing process. This Small Business Technology Transfer (STTR) Phase II project will focus on optimizing the manufacturing process to enable the scale-up of the assay and develop the machine learning system for automated data analysis. Currently, this assay involves in-house fabrication and is limited to the central laboratory at the company’s location. Manufacturing will be optimized with proper selection of material, fabrication methods, scalable techniques, and systematic integration of different elements of the assay. This will potentially enable the commercialization and implementation of the technology at a larger scale. Current methods of analysis include counting adhesion events manually. This method will be replaced by a machine learning-based system to identify and classify adhesion events and separate those from the endothelial cells in the background. Automated data analysis will enable faster outcomes and remove user bias. Since this approach relies on enhancing the capabilities of the existing platform for scale-up and streamlined analysis, it is anticipated that it will improve its accessibility to the broader research community. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.