The introduction of monoclonal antibodies to anti-tumour necrosis factor α (TNFα) two decades ago has revolutionised the treatment of inflammatory bowel disease (IBD). Despite the advent of newer and more specific mechanisms, anti-TNFα agents remained commonly used mainly due to their efficacy, safety and rapid effect. Secondary loss of response (LOR) accounts for up to 50% of treatment discontinuation.1 The dynamics of LOR over time are poorly understood or predicted.

Schultheiss et al2 recently reported data from a retrospective cohort of more than 700 patients receiving anti-TNFα treatment. The median drug survival was 3.9 years, with 30% of patients treated for over 4 years. While the cumulative incidence of discontinuation during the entire follow-up was 50%, the incidence of LOR either with or without anti-drug antibodies (ADAs) varied significantly throughout therapy—higher during the first year (17%) and lower after 2 or more years. Longer treatment duration was also associated with lower rates of anti-TNFα withdrawal for side effects.

From a clinical point of view, these data are reassuring, especially for patients who have sustained benefit from anti-TNFα treatment for over 2 years, as prolonged treatment benefit was associated with a lower likelihood of treatment failure.

These findings are consistent with previous studies demonstrating that the main driver of LOR is the development of ADAs (immunogenicity). In most cases, immunogenicity occurs relatively early within the first year of treatment,3-5 and is associated with subsequent treatment failure and discontinuation. In this study, combination therapy with an immunomodulator mitigated LOR risk and ADA development, corroborating previous observations.4, 6 However, the authors did not find any modifiable predictors for LOR without ADAs or predictors for early versus late LOR.

Identifying predictors for anti-TNF LOR is a major unmet need in IBD, especially since more therapeutic options are at hand. The recently published PANTS study identified HLA-DQA1*05 as a predictor for anti-TNFα immunogenicity and LOR in the British population.7 Such genetic data may guide clinicians for better treatment selection (i.e., medical or surgical8) or personalise the need for combination therapy. Notwithstanding, in current clinical practice, the combination of an anti-TNF agent with an immunomodulator is the mainstay to prevent LOR. The initial assessment of patients with LOR remains drug optimisation based on drug levels and ADA. Smoking cessation is also advised.9 More modifiable risk factors need to be identified, and studies into mechanisms and predictors of LOR in the short- and long-term treatment course are warranted.

The relatively long follow-up in the study by Schultheiss et al, and the number of participants further add to the evidence in the area of secondary LOR to anti-TNFα agents in IBD that can be translated into clinical practice. Their study suggests that a tighter control approach and possibly proactive therapeutic drug monitoring10 may be prudent during the first or second year, and possibly less so for patients with longer sustained clinical response to anti TNFα agents.

Declaration of personal interests: Idan Goren has received research funding from Pfizer, European Organization of Crohn’s and colitis, and the International Organization for the Study of Inflammatory Bowel Diseases. Henit Yanai has served as a speaker, a consultant and an advisory board member for Abbvie, Ferring, Janssen, Neopharm Ltd., Pfizer, and Takeda; and has received research funding from Pfizer.