Risk-updating nomogram for refractory systemic JIA in children: multicentre development and external validation

Area of Science:

• Pediatrics
• Rheumatology
• Clinical Prediction Models

Background:

• Systemic juvenile idiopathic arthritis (SJIA) is a severe inflammatory condition.
• Identifying children at risk of refractory SJIA (rSJIA) is crucial for timely intervention.
• Current models may not fully capture early treatment response.

Purpose of the Study:

• To develop and validate a pragmatic prognostic model for rSJIA risk.
• To integrate baseline disease activity with early treatment response.
• To improve risk stratification for personalized SJIA management.

Main Methods:

• Retrospective multicenter cohort study of newly diagnosed SJIA patients.
• Developed an update model using sJADAS27, refractory rash, and time to sustained clinical inactive disease (tCID).
• Assessed model discrimination and calibration in training, internal testing, and external validation cohorts.

Main Results:

• The update model demonstrated excellent discrimination in internal testing (AUC 0.919) and external validation (AUC 0.907).
• Baseline predictors (sJADAS27, refractory rash) and early response (tCID) were evaluated.
• The update model outperformed baseline-only models and sJADAS27 alone.

Conclusions:

• An integrated nomogram-based approach effectively updates rSJIA risk post-initial treatment.
• This model supports personalized treatment strategies for SJIA.
• Early integration of disease activity and treatment response enhances prognostic accuracy.