Letter: persistence of anti-infliximab antibodies after discontinuation of infliximab in patients with IBD


Sirs, In an interesting article, Ben-Horin et al. report that anti-infliximab (IFX) antibodies (Ab) are detectable in only a small minority of IBD patients at 1 year after discontinuation of IFX (3 of 16 patients, 19%).[1] We have recently monitored the decline of anti-IFX Ab in 56 IBD patients and found that 77%, 62%, 36%, 25% and 0%, respectively, had detectable anti-IFX Ab when tested 0–1, 1–2, 2–3, 3–4 and >4 years after discontinuation of IFX.[2]

There may be several explanations for these discrepancies, including the larger sample size of our study, a different composition of patients with Crohn's disease and ulcerative colitis (Ben-Horin included nine and seven, we included 46 and 10), and the extent of repeated anti-IFX Ab measurements in individual patients.

The use of different assays may also add to the conflicting findings. While our radioimmunoassay (RIA) uses anti-human λ Ab for detection, as in the capture ELISA used by Ben-Horin et al., binding in RIA takes place in fluid-phase and without addition of TNF-α.[3] This RIA has been extensively clinically validated in IBD and other diseases.[3-6] It has a higher sensitivity for anti-IFX Ab detection than bridging and capture ELISAs, and is capable of detecting all subtypes of anti-IFX Ab including monovalent IgG4 Ab, which predominate during long-term immunisations.[7]

This subclass of Ab, although detectable in capture ELISA, are not detected in bridging ELISAs.[7] Solid-phase assays may also give false negative results due to epitope masking.[7] Finally, anti-idiotypic Ab may go unnoticed in ELISAs where the capture phase consists of IFX bound to high-level TNF-α-coated plastic (Figure 1), a problem also noted by Ben-Horin et al.[8]

Figure 1.

(a) Capture ELISAs for detection of anti-drug Ab (ADA) against IFX. λ- light chain ADA, captured to IFX pre-bound to TNF-α-plated wells, are detected by enzyme-labeled anti-human λ-chain Ab (A). λ- and κ-light chain ADA, captured on F(ab') fragments of IFX or F(ab')2 fragments (not shown), are detected by enzyme-labeled anti-human Fc-γ Ab (B). (b) False-negative ADA testings may arise from matrix effects, e.g. epitope masking due to protein aggregation (not shown), failure to detect anti-idiotypic ADA (A), or high drugsensitivity (B). Note that functionally monovalent IgG4 ADA are not detected in bridging ELISA (not shown).[7]

Our conclusion is that anti-IFX Ab can persist for years after discontinuing the drug. The impact on efficacy and safety of pre-existing anti-IFX Ab prior to IFX retreatment remains to be definitively established.


Declaration of personal interests: Within the last two years, Steenholdt C has served as a speaker for MSD and Abbott, and as a consultant for MSD. Bendtzen K has served as a speaker for Pfizer, Wyeth, Roche, Novo-Nordisk, Bristol-Meyers Squibb and Biomonitor A/S, and owns stocks in Biomonitor A/S. Brynskov J has no disclosures. Declaration of funding interests: Casper Steenholdt's research programe is funded by independent grants from Aase and Ejnar Danielsen's Foundation, Beckett Foundation, Danish Biotechnology Program, Danish Colitis-Crohn Society, Danish Medical Association Research Foundation, Frode V. Nyegaard and wife's Foundation, Health Science Research Foundation of Region of Copenhagen, Herlev Hospital Research Council, Lundbeck Foundation and P. Carl Petersens Foundation.