Objectives In a 2-year follow-up study of patients with axial spondyloarthritis (axSpA) in clinical remission who tapered TNF inhibitor (TNFi) treatment according to a clinical guideline, we aimed to investigate the proportion who successfully tapered/discontinued therapy and baseline predictors thereof. The proportion regaining clinical remission after flare and the progression on MRI/radiography were also assessed. Methods One-hundred-and-nine patients (78 [72%]/31 [28%] receiving standard and reduced dose, respectively) in clinical remission (BASDAI < 40, physician global score < 40) and no signs of disease activity the previous year tapered TNFi as follows: to two-thirds of standard dose at baseline, half at week 16, one-third at week 32 and discontinuation at week 48. Patients experiencing clinical, BASDAI or MRI flare (predefined criteria) stopped tapering and escalated to previous dose. Prediction analyses were performed by multivariable regression. Results One hundred and six patients (97%) completed 2 years' follow-up; 55 patients (52%) had successfully tapered: 23 (22%) receiving two-thirds, 15 (14%) half, 16 (15%) one-third dose and 1 (1%) discontinued. In patients at standard dose at baseline (n = 78), lower physician global score was the only independent predictor of successful tapering (odds ratio [OR] = 0.79 [95% CI: 0.64, 0.93]; P = 0.003). In the entire patient group lower physician global score (OR = 0.86 [0.75, 0.98]; P = 0.017), lower Spondyloarthritis Research Consortium of Canada (SPARCC) Sacroiliac Joint Erosion score (OR = 0.78 [0.57, 0.98]; P = 0.029) and current smoker (OR = 3.28 [1.15, 10.57]; P = 0.026) were independent predictors of successful tapering. At 2 years, 97% of patients were in clinical remission. Minimal changes in imaging findings were observed. Conclusion After 2 years following a clinical guideline, 52% of patients with axSpA in clinical remission had successfully tapered TNFi, only 1% discontinued. Baseline physician global score was an independent predictor of successful tapering.