Evaluating the osteoclastogenic potential of peripheral blood mononuclear cells in children with chronic nonbacterial osteomyelitis

Chronic nonbacterial osteomyelitis (CNO) is an autoinflammatory bone disease in which osteoclastogenesis may play a critical role. Cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) has been shown to inhibit osteoclastogenesis derived from PBMCs. Our objectives were to determine whether osteoclastogenesis is altered (1) in patient PBMCs, (2) by patient serum, and (3) by a CTLA-4 analog, abatacept. Patients with active or inactive CNO, inactive juvenile idiopathic arthritis (JIA), and healthy controls were enrolled. CD14+ monocytes isolated from healthy donors were incubated with serum samples, macrophage colony-stimulating factor (MCSF), and RANKL for 7 d. Tartrate-resistant acid phosphatase positive cells with 3 or more nuclei were counted and compared across samples. In another study, PBMCs from patients were incubated with MCSF and RANKL with or without abatacept for 7 d. A log-linear mixed model considering patients nested within block random effects was applied to compare the osteoclast counts, RANKL, and OPG levels across groups. Dunn's multiple comparisons test was performed using the Benjamini-Hochberg method. No significant difference was seen across 4 groups within the serum study. The highest serum concentration was associated with more osteoclasts compared to the lowest concentration (p = .007). Inactive JIA patients' PBMCs had greater osteoclastogenic potential compared to control (p = .03). Abatacept increased osteoclastogenesis (p = .02). Serum RANKL concentrations were similar across 4 groups, while the CNO inactive group OPG concentration was higher than that in JIA inactive group (p < .05). Neither PBMCs nor the serum from patients with active CNO consistently altered osteoclastogenesis.