Background/aim: This study aimed to investigate the effectiveness of a supervised group exercise therapy based on the biopsychosocial model introduced simultaneously with antitumor necrosis factor (TNF) therapy in anti-TNF-naive patients with active ankylosing spondylitis (AS). Materials and methods: Forty-eight patients were divided into two groups: the control group (CG; n = 36) received only anti-TNF therapy, and the study group (SG; n = 12) received the supervised exercise therapy based on the biopsychosocial model in addition to anti-TNF therapy. The measurements of disease activity and functionality were evaluated by The Bath AS Disease Activity Index (BASDAI) and The Bath AS Functional Index (BASFI) respectively. Other outcome measures evaluated biopsychosocial status, emotional state, spinal mobility, pain, fatigue, sleep, and quality of life. All measurements were applied to both groups at baseline and repeated 12 weeks later. Results: BASDAI and BASFI analyses revealed significant differences between groups in favor of the SG (p < 0.05). At the end of the 12 weeks, the results showed that there were additional improvements in all outcome measurement parameters in the SG compared to the CG. Conclusion: The supervised group exercise therapy based on the biopsychosocial model introduced simultaneously with anti-TNF therapy is more effective than only anti-TNF therapy in anti-TNF-naive patients with active AS.