High-Throughput Proteomic Profiling of Hepatocyte Response to Oncologic Compounds for Toxicity and Immune Risk Prediction in Drug Development 2259796

Abstract Introduction The clinical success of immune modulatory compounds is inevitably accompanied by new risks of toxicity. Even well established therapies such as corticosteroids can result in paradoxical liver inflammation when delivered systemically, whereas immune checkpoint inhibitor therapy can result in a range of immune-related adverse events (irAEs), including rare but fatal myocarditis. The development of new immune modulating therapies requires scalable strategies for early toxicity screening. We previously leveraged the Nomic platform, capable of absolute quantification of 1000 proteins in high-throughput and low cost, to identify ∼500 compounds limiting adipocyte inflammation. Here, we leveraged Nomic’s Omni 1000 to capture potential toxicities. Methods Cell cultures of human hepatocytes or cardiomyocytes were treated with 510 bioactive compounds at 3 doses; after 48h, supernatants were collected and 1000 proteins were quantified. Results Consistent with clinical observations, corticosteroids such as methylprednisolone displayed anti-inflammatory properties, while paradoxically increasing the expression of pro-inflammatory SAA and IL-6 in hepatocytes. Cytotoxic compounds were characterized by the release of normally intracellular proteins (ex: GAPDH, MAPK3, IRF3, etc.) into the supernatant, including PHA-793887, a CDK inhibitor that led to fatal liver failures due to toxic effects that went undetected pre-clinically, but were captured here. Mechanistic information was also captured: for example, cycloheximide, an inhibitor of protein synthesis, resulted in reduced levels of hundreds of proteins, and so-called “superinduction” of IL-6, at low doses; at high doses, widespread cell death occurred. Conclusion This ability of high-plex proteomics to distinguish bioactivity from toxicity enabled us to discern anti-inflammatory compounds with acceptable toxicity profiles from compounds with severe toxicity risks, highlighting its value for the development of new immunomodulatory therapies. Funding Source n/a Topic Categories Computational and Systems Immunology (COMP)