Assessing Disease Activity in Pediatric Chronic Nonbacterial Osteomyelitis: A Proposal for Composite Scoring, Including Inactivity Measures.

Abstract

Objective: Chronic nonbacterial osteomyelitis (CNO) is an autoinflammatory bone disease with no validated criteria for assessing disease activity (DA), inactive disease, or remission. To date, DA assessment has relied on subjective judgments from patients, rheumatologists, and/or radiologists. Evaluations based on composite DA measures are emerging. The Pediatric CNO (PedCNO) response score documents relative DA changes during follow-up in analogy to the Pediatric American College of Rheumatology score for juvenile idiopathic arthritis. The international Childhood Arthritis and Rheumatology Research Alliance (CARRA) CNO initiative proposed a numeric composite DA score (CDAS) on the basis of the patient global assessment of DA (PAG), patient assessment of pain (PAP), and clinically active CNO lesions. We aim to propose different composite scores for assessment of DA, including physician assessment of DA and magnetic resonance imaging (MRI) findings; to evaluate the previously published CNO CDAS and the PedCNO response score using established registry data; and to suggest a PedCNO 90% improvement (PedCNO90) category.

Methods: Newly diagnosed patients with CNO were enrolled between 2015 and 2020 and analyzed for clinical course and DA measures (physician global assessment of DA [PGDA]/PAG and pain scores) for up to 4 years of follow-up (YFU) in the National Pediatric Rheumatologic Database.

Results: A total of 400 patients were enrolled. In single numeric scores only, clinical and MRI lesion scores reached significant changes (from baseline to 3 YFU: P = 0.003/P = 0.004). Composite scores, which include MRI DA scores consisting of one clinical patient-based assessment parameter (PAG/PAP of DA), the PGDA and MRI lesion count are less dependent on subjective measures and demonstrate more pronounced changes over time of supposed CNO DA compared with the CNO CDAS. A PedCNO90 response score gradually increased during follow-up, ultimately reaching a PedCNO90 in half of the remaining patients after 4 years.

Conclusion: Composite scores, including MRI lesions and PGDA, seem to be promising tools for describing the activity of CNO and are proposed. The composition of DA scores seems essential for future studies.