Serum osteopontin as a predictive marker of responsiveness to methotrexate in juvenile idiopathic arthritis.

OBJECTIVE

To evaluate if serum concentrations of osteopontin (OPN) at baseline in patients with juvenile idiopathic arthritis (JIA) represent a potential predictor of responsiveness to methotrexate (MTX).

METHODS

At diagnosis, 60 children with active JIA received MTX in addition to nonsteroidal antiinflammatory drugs. After 12 months of MTX treatment, 30 patients were defined as responders; the 30 nonresponders received anti-tumor necrosis factor-alpha therapy (etanercept) in addition to MTX; this group was then enrolled for an additional 12-month study period. No patient had received steroids within 6 weeks before entering the study. Fifty healthy children matched for sex and age acted as controls. OPN serum levels were measured at baseline, before MTX, and then at 6 and 12 months. In the nonresponder patients, OPN was evaluated again after 6 and 12 months of etanercept treatment.

RESULTS

At baseline, OPN values were significantly higher (p = 0.0003) in JIA patients than in controls, with no significant differences among the different JIA subtypes. At baseline, OPN levels were lower in responders than in nonresponder patients (14.16 +/- 10.1 microg/ml vs 33.2 +/- 18.1 microg/ml, respectively). After 12 months of MTX treatment, OPN levels were significantly reduced in comparison to baseline in both responder and nonresponder groups (p = 0.0017, p = 0.0048, respectively). In nonresponders, etanercept significantly reduced OPN levels at 6 and 12-month followup in comparison to baseline (p = 0.002, p = 0.008, respectively). No significant differences were found among OPN levels and disease activity.

CONCLUSION

Serum levels of OPN at baseline represent a possible marker to predict the responsiveness to MTX in patients with JIA.