Background The efficacy of exercise therapy for ankylosing spondylitis (AS) is well-documented, but dearth of information is for non-radiographic axial spondyloarthritis (nr-axSpA). Biomarkers like serum calprotectin, interleukins IL-6, IL-17 and tumour necrosis factor (TNF)-α may reflect the disease activity of axial spondyloarthritis (axSpA). In this study, we investigated clinical and laboratory parameters of both axSpA subgroups in response to intensive physical exercise. Methods Altogether, 46 patients with axSpA, characterised according to the Assessment of SpondyloArthritis International Society criteria as having nr-axSpA or AS underwent 6-month exercise programme. Clinical outcomes of disease activity, Bath AS Disease Activity Index (BASDAI), AS Disease Activity Index (ASDAS-CRP), mobility, Bath AS Metrology Index (BASMI) and function, Bath AS Functional Index (BASFI) were evaluated at baseline and at the end of the exercise programme. Serum IL-6 and IL-17, TNF-α and calprotectin were measured via ELISA. The clinical and laboratory data of 29 control axSpA patients were used for the evaluation of the results. Results In all axSpA patients, the ASDAS-CRP (2.10 ± 0.12 to 1.84 ± 0.11, p <0.01) and BASMI (1.28 ± 0.14 to 0.66 ± 0.84, p <0.0001) improved after 6 months of exercise therapy. There was a significant improvement in the ASDAS-CRP in the nr-axSpA subgroup (2.01 ± 0.19 to 1.73 ± 0.16, p <0.05) and in the BASMI in both, the nr-axSpA and the AS subgroups (1.09 ± 0.12 to 0.47 ± 0.08, p <0.0001 and 1.43 ± 0.24 to 0.82 ± 0.23, p <0.0001, respectively). Both, ASDAS-CRP and BASDAI, were significantly improved in the exercise axSpA group compared to the control axSpA group (mean -0.26 vs. -0.13 and -0.49 vs. 0.12, respectively, all p <0.05). Only calprotectin was significantly reduced after the exercise programme in nr-axSpA and AS patients (from 2379.0 ± 243.20 to 1779.0 ± 138.30 μg/mL and from 2430.0 ± 269.70 to 1816.0 ± 148.20 μg/mL, respectively, all p <0.01). The change in calprotectin was more profound in the axSpA intervention group (mean -604.56) than in the control axSpA (mean -149.28, p <0.05). Conclusion This study demonstrated similar efficacy for an intensive exercise programme in both nr-axSpA and AS patients. A significant decrease in serum calprotectin levels in both subgroups of axSpA patients after the exercise programme reflected an improvement in the disease activity and spinal mobility.