Early Prediction of Clinical Response to Anti-TNF Treatment using Multi-omics and Machine Learning in Rheumatoid Arthritis

Abstract ObjectivesAdvances in immunotherapy by blocking TNF have remarkably improved treatment outcomes for rheumatoid arthritis patients. Although treatment specifically targets TNF, the downstream mechanisms of immune suppression are not completely understood. The present study was aimed to detect biomarkers and expression signatures of treatment response to TNF inhibition.MethodsPeripheral blood mononuclear cells were obtained from 39 patients collected before anti-TNF treatment initiation (day 0) and after three months. Response to treatment was defined based on the EULAR criteria and classified 23 patients as responders and 16 as non-responders. We investigated differences in gene expression in peripheral blood mononuclear cells, the proportion of cell types and cell phenotypes in peripheral blood using flow cytometry, and the level of proteins in plasma. Finally, using biological measurements, we run machine learning models to predict non-response. ResultsThe gene expression analysis in baseline samples revealed notably a higher expression of the gene EPPK1 in future responders. We also detected the suppression of genes and proteins following treatment, including suppression of expression of the gene, T-cell inhibitor CHI3L1, and its protein YKL-40. The gene expression results were replicated in an independent cohort. Finally, machine learning models mainly based on transcriptomics data showed high predictive utility (ROC AUC ± SEM: 0.81 ± 0.17) in classifying non-response to anti-TNF treatment in RA.ConclusionsOur integrative multi-omics analyses identified new biomarkers for prediction of response, found pathways influenced by treatment and suggested new predictive models of anti-TNF treatment in RA patients.