Serum levels of interleukin-34 and RANKL as multivariable predictors of bone erosion seen by ultrasonography in patients with ankylosing spondylitis

Abstract

Abstract Background Ankylosing spondylitis (AS) is a chronic inflammatory arthritic disease, and sacroiliitis, enthesitis, and propensity for sacroiliac and spinal fusion are characteristic pathological features. Interleukin-34 (IL-34) plays a role in the induction and differentiation of osteoclasts. Other inflammatory factors are not directly involved in the induction and differentiation, but play an indirect role by modulating the level of receptor activator of nuclear factor-κB (RANKL) and other molecules during the process of inflammatory bone destruction in AS. However, to our knowledge, the relationship between enthesitis and bone erosion, and IL-34 and RANKL in AS has not yet been elucidated. Objective To determine the correlation between serum IL-34, RANKL, and disease severity including enthesitis and bone erosion in patients with AS and develop multivariable predictive model. Methods We conducted a cross-sectional study of 40 patients with AS, compared with 40 patients with osteoarthritis, and 40 healthy volunteers. Their serum levels of IL-34 and RANKL were measured using enzyme-linked immunosorbent assays (ELISAs). Enthesitis and bone erosion were assessed with real-time ultrasonography. Spearman rank correlation coefficients were determined to analyze the relationship between the variables. Multiple logistic regression was used to determine associations and receiver operating characteristic (ROC) curve analyses were conducted to determine the diagnostic performance of cytokine levels. Results In patients with AS, serum levels of IL-34 (878.9 ± 116.4 pg/mL) and RANKL (155.6 ± 13.8 pg/mL) were significantly (P < 0.01) higher than those in patients with osteoarthritis (626.6 ± 79.0 and 138.1 ± 15.3 pg/mL, respectively) or a healthy group (612.9 ± 61.1 and 104.9 ± 15.4 pg/mL, respectively). Serum levels of IL-34 were not significantly correlated with the levels of RANKL. In patients with AS, serum levels of IL-34 and RANKL adjusted for age and weight were significantly correlated with enthesitis (0.798, P < 0.01; 0.347, P < 0.05, respectively) and bone erosion (0.822, P < 0.01; 0.368, P < 0.05, respectively). The area under the ROC curve (AUC) for the serum levels of IL-34 was 0.995 between patients with AS and healthy individuals. When serum level of IL-34 was >697.1 pg/mL, the sensitivity (SE) was >99% and specificity (SP) was 95.0%. The AUC for IL-34 was 0.982 between patients with AS and patients with osteoarthritis. When serum IL-34 was >688.4 pg/mL, the SE was >99% and SP 85.0%. IL-34 correlation with the number of bone erosions of enthesis was rs = 0.795, P < 0.01. The AUC for serum RANKL was 0.993 between patients with AS and healthy individuals. When serum RANKL was >126.2 pg/mL, the SE was 97.5% and SP 97.5%. The AUC for serum RANKL was 0.798 between patients with AS and patients with osteoarthritis. When serum RANKL was >149.3 pg/mL, the SE was 70% and SP was 80.0%. Conclusions In patients with AS, serum levels of IL-34 and RANKL may be useful indicators of enthesitis, especially for bone erosions. IL-34 is associated with AS-associated enthesis damage and is a potential biomarker for predicting subsequent progression in patients with AS.