Background/Purpose: Methotrexate (MTX) and tumor necrosis factor-α inhibitors (TNFi) are common treatments for children with chronic non-infectious uveitis (NIU). Optimal duration of treatment prior to taper and discontinuation is understudied. Our aim is to identify factors that predict successful medication discontinuation and sustained remission in children with NIU.

Methods: We reviewed records of 120 children with NIU and identified children who tapered and/or discontinued MTX or TNFi (infliximab or adalimumab). We recorded date of medication start, taper, and discontinuation, and time to relapse or remission. Comparisons between children with and without remission were made using Mann-Whitney tests or Chi-square tests.

Results: There were 28 children in whom we attempted to discontinue medication (MTX in 16 and TNFi in 12 (10 infliximab, 2 adalimumab)). They were mostly Caucasian (43%) females (79%), with JIA-associated uveitis (57%) (Table 1). Most common reason for drug discontinuation was remission/inactive disease (61%).

In 16 children, MTX was given for a median of 1.6 years (25^th– 75^th: 1.1 – 3.0) prior to taper. Four (25%) discontinued MTX without taper, and 12 (75%) tapered over a median of 8 months. Only 5 (31%) successfully discontinued medication and sustained remission for a median of 6 months as of last follow-up. The remaining 11 (69%) relapsed in approximately 8 months. Successful discontinuation of MTX was associated with longer duration of therapy (3.3 vs. 1.4 years; p =0.02) and fewer ocular complications (2/5, 40% vs.11/11, 100%, p= 0.02). Gender, race, NIU type, age at MTX start, duration of NIU before MTX, MTX route and dosing, and eye disease at taper/discontinuation were not associated with successful MTX discontinuation. 

In 12 children, TNFi was given for a median of 1.8 years (25^th– 75^th: 1.4 – 2.9) prior to taper or discontinuation. Eight (67%) discontinued TNFi without taper.  All 8 relapsed and needed to restart or switch therapy at a median of 3 months.  Four (33%) attempted to taper medication, but only 1 discontinued but later relapsed after 1.3 years. The remaining 3 children restarted/switched therapy at a median of 10 months after starting taper.

In the 23 patients that relapsed, 9 (39%) relapsed within 3 months, 11 (48%) within 6 months, 17 (74%) within 1 year and 22 (96%) within 1.5 years of medication discontinuation/taper.

Conclusion: Most children were unable to discontinue MTX and TNFi. Our results suggest that longer duration of MTX treatment and fewer ocular complications may be associated with sustained remission. Patients requiring TNFi appear to require medication throughout their disease course.  Factors leading to successful medication discontinuation and remission in children on TNFi need further study.