Whole blood gene expression from infliximab treated Crohn's disease patients at three time points: pre-treatment, two weeks and fourteen weeks post first treatment

Personalized treatment of complex diseases is an unmet medical need pushing towards drug biomarker identification of one drug-disease combination at a time. Here, we used a novel computational approach for modeling cell-centered individual-level network dynamics from high-dimensional blood data to predict infliximab response and uncover individual variation of non-response. We identified and validated that the RAC1-PAK1 axis is predictive of infliximab response in inflammatory bowel disease. Intermediate monocytes, which closely correlated with inflammation state, play a key role in the RAC1-PAK1 responses, supporting their modulation as a therapeutic target. This axis also predicts response in Rheumatoid arthritis, validated in three public cohorts. Our findings support pan-disease drug response diagnostics from blood, implicating common mechanisms of drug response or failure across diseases. Whole blood samples from anti-TNF responding (n=15) and non-responding (n=9) IBD patients at three time points: pre-treatment, two weeks and fourteen weeks post first treatment