# Toward Personalized Withdrawal of TNF-α Inhibitors in Non-Systemic Juvenile Idiopathic Arthritis: Predictors of Biologic-Free Remission and Flare

## TL;DR

This study identifies factors that predict successful withdrawal of TNF-α inhibitors in children with non-systemic juvenile idiopathic arthritis, aiming to help personalize treatment strategies.

## Contribution

The study provides novel predictors of biologic-free remission and flare during TNF-α inhibitor withdrawal in pediatric JIA patients.

## Key findings

- Higher baseline CHAQ scores and elevated calprotectin and hsCRP levels at withdrawal were linked to increased flare risk.
- Subclinical synovitis and a history of uveitis were significant predictors of disease relapse.
- Early clinical response and low-dose methotrexate use were associated with sustained drug-free remission.

## Abstract

Background: Tumor necrosis factor-α (TNFα) inhibitors have significantly improved outcomes in children with non-systemic juvenile idiopathic arthritis (JIA), achieving long-term clinical remission for many patients. However, the optimal strategy for TNF-α inhibitor withdrawal remains unknown, whether through abrupt discontinuation, gradual dose reduction, or interval extension. Objective: We aim to identify patient-, disease-, and treatment-related predictors of successful TNF-α inhibitor withdrawal in children with non-systemic JIA. Methods: In this prospective, randomized, open-label, single-center study, 76 children with non-systemic JIA in stable remission for ≥24 months on etanercept or adalimumab were enrolled. At the time of TNF-α inhibitor discontinuation, all patients underwent a comprehensive evaluation, including a clinical examination, laboratory tests (serum calprotectin [S100 proteins] and high-sensitivity C-reactive protein [hsCRP]), and advanced joint imaging (musculoskeletal ultrasound and magnetic resonance imaging [MRI]) to assess subclinical disease activity. Patients were randomized (1:1:1, sealed-envelope allocation) to one of three predefined tapering strategies: (I) abrupt discontinuation; (II) extension of dosing intervals (etanercept 0.8 mg/kg every 2 weeks; adalimumab 24 mg/m2 every 4 weeks); or (III) gradual dose reduction (etanercept 0.4 mg/kg weekly; adalimumab 12 mg/m2 every 2 weeks). Follow-up visits were scheduled at 3, 6, 9, 12, and 18 months to monitor for disease relapse. Results: Higher baseline Childhood Health Assessment Questionnaire (CHAQ) scores (≥2), elevated serum calprotectin [S100 proteins] and hsCRP levels at withdrawal, imaging evidence of subclinical synovitis, and a history of uveitis were all significantly associated with increased risk of flare. No significant associations were found for other clinical or demographic characteristics. Conclusions: Early significant clinical response, absence of subclinical disease activity, and concomitant low-dose methotrexate therapy were key predictors of sustained drug-free remission. These findings may inform personalized strategies for biologic tapering in pediatric JIA.