Despite the proven efficacy of biological drugs for inflammatory bowel disease, these therapies are costly and do carry some risks, providing incentive for exploring strategies to discontinue therapy in patients with prolonged remission. We presently review multiple cohort studies indicating the overall risk of relapse after stopping an anti-tumor necrosis factor (TNF) in inflammatory bowel disease patients is roughly 40% at 12 months after therapy cessation. Despite methodological differences across studies, it appears that patients without deep remission (ie, patients with endoscopic or biomarker evidence of inflammation) are at increased risk of relapse after stopping anti-TNF, as are those with high-adequate levels of anti-TNF before stopping. In patients who relapse after anti-TNF cessation, retreatment with the same biological seems to reinduce clinical response in most patients. Immunological reasons responsible for this high success rate for retreatment are elucidated, but resorting to retreatment also implies a small but finite risk of a severe flare leading to surgery, which should be borne in mind. Thus, stopping attempts should probably be reserved for patients with low risk for severe outcome should a relapse occur. Proactive endoscopic monitoring after drug cessation is imperative to reduce these risks. The recently introduced concept of treatment-cycles is discussed, along with a pragmatic algorithm of decision tree for therapy discontinuation in the selected appropriate patients.