Tumor necrosis factor (TNF) antagonists are approved worldwide for the treatment of rheumatoid arthritis (RA). Clinical experience revealed that TNF-blocking therapy is effective for only approximately two thirds of patients, reflecting that there are 'responders' as well as 'nonresponders'. Given the destructive nature of RA, the risk of adverse effects, and considerable costs for therapy, there is a strong need to make predictions on success before the start of therapy. In the current issue of Arthritis Research & Therapy, Hueber and colleagues become the first to present a multi-parameter serum protein biomarker set that has predictive value prior to the start of anti-TNF treatment. Ultimately, this finding may contribute to a personalized form of medicine, whereby a specific therapy will be applied that is best suited to an individual patient.