Abstract
Drug repurposing is an attractive strategy for therapy development, particularly in rare diseases where traditional drug development approaches may be challenging owing to high cost and small numbers of patients. In this study, we used a drug identification and repurposing pipeline to identify candidate targetable drivers of disease and corresponding therapies through application of causal reasoning using a combination of open-access resources and transcriptomics data. We optimized our approach on psoriasis as a disease model, demonstrating the ability to identify known and, to date, unrecognized molecular drivers of psoriasis and link them to current and emerging therapies. Application of our approach to a cohort of tissue samples of necrobiosis lipoidica (an unrelated; rare; and, to date, molecularly poorly characterized cutaneous inflammatory disorder) identified a unique set of upstream regulators, particularly highlighting the role of IFNG and the Jak-signal transducer and activator of transcription pathway as a likely driver of disease pathogenesis and linked it to Jak inhibitors as potential therapy. Analysis of an independent cohort of necrobiosis lipoidica samples validated these findings, with the overall agreement of drug-matched upstream regulators above 96%. These data highlight the utility of our approach in rare diseases and offer an opportunity for drug discovery in other rare diseases in dermatology and beyond.