Abstract Background The low serum concentration of anti-tumor necrosis factor-alpha (anti-TNFα) agents and the presence of anti-drug antibodies are related to poor therapeutic outcomes in inflammatory bowel disease (IBD). We aimed to evaluate the prognosis of patients having both antidrug antibodies and supratherapeutic drug concentrations. Methods IBD patients on maintenance infliximab (IFX) or adalimumab (ADA) therapy were prospectively recruited. Serum drug concentrations and antidrug antibodies were measured (Anser, Prometheus). The drug concentrations were classified as subtherapeutic (IFX: <3 µg/mL; ADA: <5 µg/mL), therapeutic (IFX: 3-8 µg/mL; ADA 5-10 µg/mL), and supratherapeutic (IFX > 8 µg/mL; ADA > 10 µg/mL) concentrations regardless of the timing of blood sample collection. We analysed the relationship between i) drug concentrations and ii) the presence and absence of anti-drug antibodies with use of systemic corticosteroids and anti-TNF discontinuation. Results Of 172 patients (122 CD [70.9%], 44 UC [25.6%], 3 IBD-U [1.7%], and 3 IBD-pouch [1.7%]), the median age was 32.3 years (range, 24.8-42.5), and the median duration of IBD was 9.7 years (range 5.3-15.6). IFX and ADA were used in 81 (47.1%) and 91 (52.9%) patients, respectively, and concurrent usage of immunomodulators and systemic corticosteroids was observed in 39 (22.7%) and 23 (13.4%) patients, respectively. Supratherapeutic, therapeutic, and subtherapeutic concentrations were observed in 74 (43.0%), 32 (18.6%), and 66 (38.4%), respectively. The patients with antidrug antibodies (n = 79, 45.9%) showed a higher frequency of having subtherapeutic drug concentrations (68.4% vs 12.9%, P < 0.001) compared to those without antibodies, and the presence of antibodies was an independent predictor of discontinuation of anti-TNFα agents (hazard ratio 3.50, 95% confidence interval 1.88-6.51, P < 0.001, Fig. 1A). In subgroup analysis, patients with antidrug antibodies and supratherapeutic drug concentration exhibited a tendency to have improved drug persistence compared to those with subtherapeutic drug concentration, albeit insignificantly (P = 0.099, Fig. 1B). However, the persistence in this group was still poorer than that observed in patients without antibodies (P = 0.003). Finally, neither anti-TNFα concentrations nor the presence of antidrug antibodies were associated with the need for additional systemic corticosteroid use within 6 months. Conclusion The presence of antidrug antibodies can predict the discontinuation of anti-TNFα agents despite supratherapeutic drug levels. While supratherapeutic drug concentration may improve the treatment persistence of anti-TNFα agents in patients with anti-drug antibodies, it does not last as long as in those without antibodies.