Objective Inverse associations between systemic inflammation and cholesterol (‘the lipid paradox’) have been reported in rheumatoid arthritis (RA) and, in established axial spondyloarthritis (axSpA), but little is known about this relationship in early axSpA, which is the focus of the present study. Method In the Swedish part of the SPondyloArthritis Caught Early (SPACE) cohort (patients with chronic back pain for ≥3 months, ≤2 years; age at onset <45 years), serum levels of total cholesterol (TC) and apolipoproteins ApoA1 and ApoB were measured at inclusion, together with parameters reflecting inflammatory disease activity [C-reactive protein (CRP), Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), and sacroiliitis by magnetic resonance imaging (MRI) following Assessment of SpondyloArthritis international Society (ASAS) criteria]. All patients included in the analysis either had axSpA based on a high physician’s level of confidence or fulfilled the ASAS criteria for axSpA. Associations between lipids/lipoproteins and inflammation were assessed using multivariable linear regression models. Results In the 64 patients included, there were inverse associations for CRP with TC, ApoA1, and ApoB in age–sex-adjusted models. The negative associations with CRP remained significant for TC and ApoB in multivariable models adjusted for age, sex, BASDAI, and current smoking (p = 0.048). There were no significant associations for the lipid parameters with BASDAI or inflammation on MRI of the sacroiliac joints. Conclusion Inverse associations between systemic inflammation and lipids, particularly TC and ApoB, are present in early axSpA, similar to those shown for other inflammatory joint diseases. These patterns must be considered when including lipids in the evaluation of cardiovascular disease risk.