Associations between vitamin D, disease activity, and clinical response to therapy in rheumatoid arthritis

Objective Vitamin D deficiency is a potential risk factor for autoimmunity. Prior studies of the association between vitamin D levels and rheumatoid arthritis (RA) disease activity have yielded conflicting results. Methods Serum 25(OH)vitamin D levels were measured at baseline in 499 participants with active RA, ages 18-85 years, enrolled in a randomised clinical trial of golimumab (Go-Before Trial). Subjects were methotrexate and biologic therapy naive. Multivariable linear regression was used to assess associations between vitamin D levels and disease activity scores (DAS28), van der Heijde-Sharp (vdHS) erosion scores, and serum inflammatory markers. Generalised estimating equations were used to evaluate the associations between vitamin D status and the response to therapy over 52 weeks, using the DAS28 and ACR response. Results Forty-eight percent of participants were vitamin D deficient, defined as serum 25(OH)vitamin D <20 ng/mL. Deficiency was not associated with greater DAS28 (beta-0.021 [95% CI -0.22, 0.18]), adjusted for age, race, sex, BMI, disease duration and glomerular filtration rate. Vitamin D deficiency was not associated with baseline vdHS scores or inflammatory markers in adjusted or unadjusted models. There was no association between baseline vitamin D deficiency and change in DAS28 (beta = -0.024 [-0.30, 0.25]), proportion meeting ACR response (OR 0.82 [0.56, 1.20]), or radiographic progression at 52 weeks (OR 0.91 [0.59-1.40]). Conclusions Vitamin D levels were not associated with RA disease activity, inflammatory markers, or vdHS scores at baseline. Furthermore, there was no association between baseline vitamin D level and response to therapy or radiographic progression.