IIIM-17/13

BOSTON--( BUSINESS WIRE )-- DeepCure , a therapeutics company using AI to discover novel drugs for inflammation and immune diseases, today announced the selection of its first development candidate, DC-9476. DC-9476 is a third generation BRD inhibitor that is selective for the BD2 domain of Brd4, which is expected to enter clinical trials in 2025. The BD2 domain of Brd4 regulates multiple, clinically validated, cytokine-related pathways, such as TNF-alpha, IL6, IL-17, and IL-1b, that are important in a variety of autoimmune diseases. Many autoimmune diseases have epigenetic changes, including histone modifications that alter chromatin structure and gene expression. The BET family of proteins, which includes Brd4, that “read” these histone modification can cause patterns of gene expression in innate and adaptive immune cells that persist long after the stimulus is gone. In addition, Brd4 BD2 activity can also directly regulate transcription factors. Inhibition of Brd4 BD2 by small molecule drugs can break a self-perpetuating cycle of gene expression that can occur across multiple pathways that is detrimental to normal tissues and reinstate self-resolving processes that should follow inflammatory triggers. DC-9476 has shown promising efficacy in multiple preclinical models of autoimmune diseases, including rheumatoid arthritis (RA) and Still’s disease. In RA models, DC-9476 was superior to standard of care therapies, TNF-alpha inhibitors, IL-6 inhibitors, and the JAK inhibitor tofacitinib, and it also improved the effectiveness of TNF-alpha inhibitors when used in combination therapy. Unlike earlier generations of non-selective inhibitors, DC-9476 has an excellent safety profile and did not cause thrombocytopenia. The novel candidate was discovered using DeepCure’s proprietary AI platform by exploiting differences in the structure and function among the BET protein family and the two bromodomains (BD1 and BD2). DeepCure’s AI platform integrates advanced ML and physics-based tools to find interaction sites on the protein surface (PocketExpander™) and to design novel, diverse compounds that are synthetically feasible (MolGen™). “In the past years of translational and clinical research we have learnt that autoimmune diseases are the effect of multiple deranged pathways. Targeted therapies against specific cytokines, kinases or receptors have offered a first important boost in how we manage these conditions but patients would benefit enormously from pharmacological inhibition of multiple inflammatory pathways,” said Professor Francesco Del Galdo, M.D., Ph.D., Professor of Experimental Medicine NIHR Biomedical Research Centre and Institute of Rheumatic and Musculoskeletal Medicine University of Leeds. “However, current attempts with simply combining clinically existing therapies that target individual pathways have shown limited success, mostly due to increased toxicity. New therapeutics such as DC-9476 that have a novel mechanism to address simultaneously multiple pathways while also having an excellent safety profile, are extremely promising for the field.” “We’re excited to announce our first AI-generated candidate, which we plan to advance into clinical trials next year. This validates our generative AI and physics-based drug discovery engine,” said Kfir Schreiber, CEO & Co-Founder, DeepCure. “DC-9476 is the first of many novel development candidates to emerge from our platform, with the potential to transform the treatment of inflammation and immune diseases.” About DeepCure DeepCure was founded by researchers at MIT to accelerate breakthrough therapies using artificial intelligence (AI) and AI-enabling technologies to innovate small molecule discovery. The company is based in Boston, MA, and its engineers, chemists, and biologists collaborate on hard problems to find solutions that will have an enormous impact on patient health. For more information, visit www.deepcure.ai .