Discontinuing Colchicine in Pediatric Familial Mediterranean Fever: Real-Life Experience from a Long-Term Follow-Up Cohort

BACKGROUND: Colchicine is the cornerstone of treatment in familial Mediterranean fever and is generally recommended as lifelong therapy. However, a subset of pediatric patients may experience a milder disease course, raising questions about the feasibility and safety of colchicine discontinuation. Evidence guiding this decision remains limited, and no consensus exists on patient selection criteria or monitoring strategies. OBJECTIVE: The objective of this study was to evaluate long-term outcomes after colchicine discontinuation in a large pediatric familial Mediterranean fever cohort and to identify clinical and genetic factors associated with sustained colchicine-free remission. METHODS: We retrospectively analyzed medical records of 2316 pediatric patients diagnosed with familial Mediterranean fever and followed at the Pediatric Rheumatology Unit of Istanbul University, Istanbul Faculty of Medicine, Istanbul, Türkiye. Patients fulfilling at least one of the Tel-Hashomer or Eurofever/Pediatric Rheumatology International Trials Organization (PRINTO) classification criteria and carrying at least one pathogenic or likely pathogenic MEFV (Mediterranean fever) gene mutation were eligible. Colchicine was discontinued in 127 patients (5.48%), who were classified into two groups: Group 1 (n = 31), patients who required colchicine re-initiation because of recurrent attacks, and Group 2 (n = 96), patients who maintained sustained colchicine-free remission during follow-up. A contemporaneous comparator group of 118 patients who continued colchicine therapy without any attempt at discontinuation (Group 3) was included to provide clinical and genetic context. Demographic, clinical, genetic, and disease severity characteristics were compared across groups. RESULTS: Sustained colchicine-free remission was maintained in 96 of 2316 patients (4.14%; Group 2). Among all patients who discontinued colchicine (Groups 1 and 2 combined), the median duration of colchicine treatment before discontinuation was 3.85 years (interquartile range 1.84-7.10), and the median follow-up after cessation was 4.91 years (interquartile range 1.92-7.23). Baseline clinical, demographic, and genetic characteristics were largely similar between Group 1 and Group 2, and no robust independent predictor of sustained remission was identified. Biallelic pathogenic exon 10 MEFV variants were significantly more prevalent in Group 3 (32.2%) compared with Group 1 (12.9%) and Group 2 (3.1%), reflecting a progressive genetic gradient across the three groups. Patients in Group 3 also exhibited higher frequencies of fever, chest pain, arthritis, arthralgia, erysipelas-like erythema, and exercise-induced leg pain compared with the discontinuation groups. Fulfillment of Eurofever/PRINTO classification criteria was significantly lower in Groups 1 (54.8%) and 2 (54.2%) than in Group 3 (87.3%), consistent with a milder phenotypic profile in patients considered for discontinuation. No patient in the discontinuation cohort developed amyloidosis during follow-up. CONCLUSIONS: Colchicine discontinuation is achievable in a small but consistent subset of patients with pediatric familial Mediterranean fever (4.14% of the total cohort), the majority of whom maintain sustained remission over a long-term follow-up. While patients requiring continuous therapy exhibit a more severe clinical and genetic profile, no robust baseline predictor reliably distinguishes those who will remain in remission from those who will relapse. These findings highlight the need for individualized withdrawal decisions guided by rigorous longitudinal monitoring, and underscore the importance of prospective studies incorporating emerging biomarkers of subclinical inflammation.