Journal Article

Human cardiovascular disease model predicts xanthine oxidase inhibitor cardiovascular risk

Wamhoff, Brian R., Feaver, Ryan E., Henke, Brad R., Hoang, Steve, Zhao, Lin, Figler, Robert A., Cole, Banumathi K., Taylor, Justin, Bowers, M. Scott, Lawson, Mark J., Nyborg, Andrew C., Johns, Brian A., LaMoreaux, Brian D.

2023

Overview

Some health concerns are often not identified until late into clinical development of drugs, which can place participants and patients at significant risk. For example, the United States Food and Drug Administration (FDA) labeled the xanthine oxidase inhibitor febuxostat with a”boxed” warning regarding an increased risk of cardiovascular death, and this safety risk was only identified during Phase 3b clinical trials after its approval. Thus, better preclinical assessment of drug efficacy and safety are needed to accurately evaluate candidate drug risk earlier in discovery and development. This study explored whether an in vitro vascular model incorporating human vascular cells and hemodynamics could be used to differentiate the potential cardiovascular risk associated with molecules that have similar on-target mechanisms of action. We compared the transcriptomic responses induced by febuxostat and other xanthine oxidase inhibitors to a database of 111 different compounds profiled in the human vascular model. Of the 111 compounds in the database, 107 are clinical-stage and 33 are FDA-labelled for increased cardiovascular risk. Febuxostat induces pathway-level regulation that has high similarity to the set of drugs FDA-labelled for increased cardiovascular risk. These results were replicated with a febuxostat analog, but not another structurally distinct xanthine oxidase inhibitor that does not confer cardiovascular risk. Together, these data suggest that the FDA warning for febuxostat stems from the chemical structure of the medication itself, rather than the target, xanthine oxidase. Importantly, these data indicate that cardiovascular risk can be evaluated in this in vitro human vascular model, which may facilitate understanding the drug candidate safety profile earlier in discovery and development.

• Publisher: Public Library of Science,Public Library of Science (PLoS)
• Subject: Algorithms / Atherosclerosis / Biology and Life Sciences / Cardiovascular disease / Cardiovascular diseases / Care and treatment / Clinical outcomes / Clinical trials / Diagnosis / Drug development / Drug efficacy / Drugs / Evaluation / Feature selection / Health aspects / Health risks / Hemodynamics / Inhibitors / Lipoproteins / Magnetic resonance imaging / Medicine and Health Sciences / Modelling / Physical Sciences / Physiology / Research and analysis methods / Risk assessment / Safety / Shear stress / Support vector machines / Transcriptomics / Tumor necrosis factor-TNF / Xanthine / Xanthine oxidase