Long-term longitudinal changes in bone matrix and turnover markers in rheumatoid arthritis: A retrospective cohort...

The increasing age of patients with rheumatoid arthritis (RA) has heightened concern regarding osteoporosis and bone fragility. While bone loss in RA has been well studied, less is known about long-term changes in bone quality-related markers under different disease activity states and antirheumatic treatments. In this retrospective observational study, 25 patients with RA aged ≥ 50 years were followed for up to 2 years, during which serum bone matrix and turnover markers, including pentosidine, tartrate-resistant acid phosphatase-5b, and total procollagen type 1 N-terminal propeptide, were measured every 6 months. Patients were classified into a Remission group (Disease Activity Score in 28 joints based on C-reactive protein < 2.3) and a High/Moderate/Low Disease Activity (H/M/LDA) group (Disease Activity Score in 28 joints based on C-reactive protein ≥ 2.3). Longitudinal changes were evaluated using linear mixed-effects models with group-by-time interaction terms. Additional treatment-based comparisons were performed within the Remission group between patients treated without biological/targeted synthetic disease-modifying antirheumatic drugs (non-b/tsDMARDs) and those treated with b/tsDMARDs, with additional adjustment for baseline age and estimated glomerular filtration rate. At baseline, serum pentosidine levels were higher and serum total procollagen type 1 N-terminal propeptide levels were lower in the H/M/LDA group than in the Remission group, whereas tartrate-resistant acid phosphatase-5b levels did not differ. Linear mixed-effects analyses showed no significant group-by-time interactions between the Remission and H/M/LDA groups. Within the Remission group, serum pentosidine increased at later time points in patients treated with non-b/tsDMARDs, whereas no clear increase was observed in those treated with b/tsDMARDs. However, after adjustment for baseline age and estimated glomerular filtration rate, the overall group-by-time interaction for pentosidine was no longer statistically significant. These findings suggest that bone quality-related deterioration may remain a concern even in RA patients in clinical remission. However, because the adjusted overall interaction was not statistically significant, the observed association between b/tsDMARD use and longitudinal pentosidine changes should be considered exploratory and requires further validation in larger studies.