Comparison of two ELISA versions for infliximab serum levels in patients diagnosed with ankylosing spondylitis

There are various immunosorbent assays which can be used to determine infliximab (IFX) levels. Results vary between assays complicating reliability in everyday clinical practice. The aim of this study was to determine whether quantitative or qualitative assay data prove more accurate in the assessment of infliximab levels in AS patients. We analyzed 40 serum samples, taken prior to infusion, from AS patients who had been undergoing IFX therapy as a first-line of biological treatment for more than a year. IFX levels and IFX–anti-drug antibodies (ADA) were measured using two different ELISA assays [Promonitor^® IFX R1 and R2 (version 1), Promonitor^® IFX and anti-IFX (version 2) (Progenika Biopharma, Spain)] strictly following the manufacturer’s guidelines. Cohen’s unweighted kappa and the intraclass correlation coefficient determined qualitative and quantitative agreement for serum levels in version 1 and version 2. Bland–Altman plots were drawn to compare both assays. The comparison of data measuring IFX levels for version 1 and version 2 resulted in questionable quantitative agreement (ICC 0.659; 95 % CI 0.317–0.830) and moderate qualitative agreement (κ 0.607; 95 % CI 0.387–0.879) owing to systematically higher values in version 2 than version 1. Version 2 consistently detected higher levels of infliximab, particularly when analyzed in a quantitative context. Further research is needed to synchronize cutoff levels between essays and diseases so therapeutic drug ranges can be established.