0389: An International Delphi Survey for the Definition of New Classification Criteria for Juvenile Idiopathic Arthritis

Sunday, October 26, 2025

10:30 AM - 12:30 PM Central Time

Location: Hall F1

Abstract Poster Presenter(s)

NR

Nicolino Ruperto, MD, MPH

Università Milano Bicocca and Fondazione IRCSS S. Gerardo dei Tintori
Monza, Monza and Brianza, Italy

Disclosure(s): Abbvie: Consultant (Terminated, August 22, 2025); Aclaris: Consultant (Terminated, September 26, 2025); AlfaSigma: Consultant (Ongoing); Amgen: Consultant (Ongoing); AstraZeneca: Consultant (Terminated, August 22, 2025); Aurinia: Consultant (Terminated, August 2, 2025); Boehringer-Ingelheim: Consultant (Terminated, August 9, 2025); Bristol-Myers Squibb(BMS): Consultant (Ongoing); Eli Lilly: Consultant (Terminated, August 22, 2025); Galapagos: Consultant (Ongoing); Genentech: Consultant (Ongoing); Guidepoint: Consultant (Terminated, August 8, 2025); Idorsia: Consultant (Terminated, August 9, 2025); Janssen: Consultant (Ongoing); Novartis: Consultant (Terminated, August 2, 2025); Pfizer: Consultant (Terminated, August 2, 2025); Sanofi: Consultant (Terminated, August 1, 2025); Takeda: Consultant (Terminated, August 22, 2025)

Background/Purpose: To improve the current classification criteria for juvenile idiopathic arthritis (JIA), a multi-step initiative coordinated by PRINTO was launched in 2015, starting with a consensus process that culminated in the proposal of the preliminary PRINTO JIA classification criteria. The steps include a large-scale prospective data collection of patients with new-onset JIA and the ongoing validation of evidence-based criteria. To initiate the validation phase, an international Delphi survey was conducted among pediatric rheumatologists to identify the most relevant categories and candidate variables for subsequent data collection.
To identify consensus-based categories and variable sets for the evidence-based revision of JIA classification criteria.

Methods: An invitation to participate in two Delphi rounds was sent to PRINTO/PRCSG membership starting in 2020.

In Round 1, participants evaluated each ILAR category plus the additional category 'Early Onset ANA Positive Arthritis', indicating whether to retain or discard it, and suggesting revised names. They also listed variables used in practice to diagnose each category.

In Round 2, participants re-evaluated each category for retention and ranked the top three preferred names and the ten most useful variables for the category-specific diagnosis among those identified in Round 1. Based on the ranking, a median score for each variable was calculated. Variables falling above the 75th percentile of the score were identified for each category.

Results: 434 physicians responded in Round 1 and 884 in Round 2. Votes to retain the ILAR categories were 236 (73%) for Systemic arthritis, 218 (70%) for Oligoarthritis, 209 (69%) for Polyarthritis Rheumatoid Factor (RF) negative, 276 (92%) for Polyarthritis RF positive, 188 (66%) for Psoriatic arthritis, 225 (78%) for Enthesitis-related arthritis (ERA), 183 (67%) for Early-onset ANA-positive JIA, and 142 (53%) for Other arthritis.

Top ranking labels for each new category were ‘Juvenile-onset Still disease’, ‘Oligoarticular JIA’, ‘Juvenile polyarthritis’, ‘JIA RF positive or anti-CCP positive’, ‘Juvenile psoriatic arthritis’, ‘Early onset spondyloarthritis’, ‘ANA positive juvenile arthritis’ and ‘Unclassified JIA’.

The variables identified as most useful for the diagnosis (in the top quartile of the score) were 15 for Systemic arthritis, 31 for Oligoarthritis, 33 for Polyarthritis RF negative, 33 for Polyarthritis RF positive, 45 for ERA, 26 for Psoriatic arthritis, 27 for Early-onset ANA-positive JIA and 19 for Other arthritis.

Conclusion: The study yielded consensus-based categories to be considered in the new classification criteria and identified candidate variables to be included in the ongoing step of the project, namely the evidence-based validation of new classification criteria for JIA based on real new-onset patients’ data.