Nakagawa J, et al. – The researchers aimed to create a scoring system that predicts a better treatment for each rheumatoid arthritis (RA) patient before starting an IL–6 or a tumor necrosis factor inhibitor (TNF–i). They suggested that the score was easily calculated from common laboratory results. It seemed useful for identifying a better treatment at the time of selecting either an IL–6 or a TNF inhibitor.

Methods

• With DNA microarrays, the expression of IL–6 and TNF– α mRNA in peripheral blood from 45 newly diagnosed RA patients was measured to evaluate cytokine activation.
• Next, laboratory indices immediately before commencing treatment and disease activity score improvement ratio after 6 months in 98 patients treated with TCZ or TNF–i were retrospectively analyzed.
• Some indices correlated with TCZ efficacy were selected and their cutoff values were defined by receiver operating characteristic (ROC) analysis to develop a scoring system to discriminate between individuals more likely to respond to TCZ or TNF–i.
• The validity of the scoring system was verified in these 98 patients and an additional 228 patients.

Results

• Investigation revealed a significant inverse correlation between the expression of IL–6 and TNF– α mRNA in newly diagnosed RA patients.
• The analysis of 98 patients revealed significant correlation between TCZ efficacy and platelet counts, hemoglobin, aspartate aminotransferase, and alanine aminotransferase.
• However, there was no similar correlation in the TNF–i group.
• The cutoff values were defined by ROC analysis to develop a scoring system (1 point/item, maximum of 4 points).
• A good TCZ response was predicted if the score was ≥2; in contrast, TNF–i seemed to be preferable if the score was ≤1.
• Similar outcomes were obtained in a validation study of an additional 228 patients.
• If the case scored ≥3, the good responder rates of TCZ/TNF–i were 75.0%/37.9% (p < 0.01) and the non–responder rates were 3.1%/27.6% (p < 0.01), respectively.