SBIR Phase II: AI-assisted identification of small molecules for targeted repair of vascular barrier dysfunctions

Akttyva Therapeutics, Inc., Watertown MA

Investigators

Abstract

The broader impact of this Small Business Innovation Research (SBIR) Phase II project will lead to an Artificial Intelligence (AI) assisted drug discovery platform that may shed light on unexplored interactions and mechanisms to quickly predict the best small molecules to design new therapies for diseases with unmet clinical needs. Currently, there is no targeted treatment addressing endothelial dysfunctions or vascular leaks. Through the AI-assisted platform the company will generate treatment candidates for disorders associated with vascular leaks and improve the efficiency of the drug discovery workflow. The first condition that the company will address is acute respiratory distress syndrome (ARDS), a life-threatening form of respiratory failure that affects approximately 200,000 patients each year in the US, resulting in nearly 75,000 deaths annually. The company’s therapeutic intervention for ARDS and other diseases with vascular leakage could significantly increase the survival rate, reduce costs associated with hospitalization, and improve the health outcomes of the patients. Leveraging AI-assisted drug discovery approach for ARDS drug development, the company will strive to bring forward effective strategies to screen approved and investigational therapeutic agents as well as novel chemical entities (NCEs) to derisk future clinical development. The proposed project seeks to develop an AI-assisted drug discovery platform to identify existing drugs as well as NCEs able to repair the vascular permeability barrier in several different conditions. In this project, the R&D efforts will be dedicated towards (1) further development of the Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADME-Tox) Engine to flag toxicophores and predict ADME-Tox endpoints, (2) development of the next level active-learning docking-scoring protocol for superior screening efficiency, (3) performing a Hit-to-lead campaign for protein activators for restoring disrupted endothelial barrier and identifying NCE candidates, and (4) performing advanced proof of concept studies of pre-lead protein activator compounds in ADME-Tox, human lung-on-chip model, and in vivo vascular leak mouse model. The successful completion of the Phase II activities could bring the development of an AI-assisted drug discovery platform with a novel pocket prediction engine, active learning-based docking-scoring workflow, ADME-Tox engine, and an electronic library of AI-generated synthesizable molecules collection. At the end of the SBIR Phase II project, the company will select the lead ARDS drug candidate for IND-enabling studies based on the battery of in vitro and in vivo studies. 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.