Background Anti-drug antibodies can be elicited by infliximab and adalimumab, but the rate of their decay after therapy is stopped is unknown. Aim To investigate the decline of anti-drug antibody titre after anti-TNF cessation, and to evaluate the clinical utility of anti-drug antibody measurement before anti-TNF re-induction. Methods Inflammatory bowel disease (IBD) patients who stopped anti-TNF therapy and had measurable anti-drug antibodies were prospectively followed up by serial blood measurements of antibodies levels. The clinical outcome of a second cohort of patients who received re-induction by infliximab or adalimumab after a drug holiday >4 months was determined vis-à-vis their anti-drug antibodies status before re-induction. Results The first cohort included 22 patients with anti-drug antibodies who were prospectively followed up after cessation of anti-TNF. Sixteen had antibodies-to-infliximab (ATI) and six had antibodies-to-adalimumab (ATA). ATI titres declined within 12 months to below detection levels in 13/16 infliximab-treated patients, whereas ATA titres became undetectable in only 2/6 adalimumab-treated patients (P = 0.04). The second cohort comprised 27 patients who resumed anti-TNFs (24 infliximab, 3 adalimumab). Of these, 3/5 patients with measurable anti-drug antibodies before re-induction experienced severe hypersensitivity reaction and/or nonresponse mandating drug-discontinuation, compared to 11/22 patients who were re-induced without measurable anti-drug antibodies (OR = 1.5, 95% CI 0.2-11, P = 0.7). Conclusions Antibodies to infliximab titres decline to undetectable levels within one year of cessation of infliximab in the majority of patients, whereas antibodies to adalimumab seem to persist longer after adalimumab discontinuation. Measuring antibodies to infliximab prior to infliximab re-induction is probably of little clinical utility, especially if more than a 12-month drug-holiday has elapsed.