OP0059 Golimumab versus tocilizumab for severe and refractory juvenile idiopathic arthritis-uveitis. multicenter study of 33 patients

Background Uveitis is a severe manifestation of Juvenil Idiopathic Arthritis (JIA). Anti-TNFa are recommended in refractory cases, mainly infliximab (IFX) or adalimumab (ADA) (Levy-Clarke et al. Ophthalmology 2014; 121: 785–796). However, sometimes they are ineffective, contraindicated or not tolerated. The next therapeutic step is not defined. Objectives To compare the efficacy of Golimumab (GLM) and Tocilizumab (TCZ) in related AIJ uveitis refractory to conventional immunosuppressive drugs and anti-TNFα. Methods Multicenter study of 33 patients with uveitis associated-JIA. They were refractory to conventional treatment with high dose of corticosteroids and at least a) one conventional immunosuppressive drug and b) one anti-TNFa. For this reason it was decided to iniciate TCZ or GLM. TCZ was used in 25 patients: 8 mg/kg/4 w iv (n=21), 8 mg/kg/2 w (n=2); 8 mg/kg/8 w (n=1) and 2.9 mg/kg sc/w (n=1). GLM was used in 8 patients (50 mg/sc/month). We assessed visual acuity (VA), degree of intraocular inflammation, vitreous inflammation and macular thickening (with OCT). Quantitative variables were expressed with mean±SD or median [IQR], according to its distribution. They were compared with the Student t or the Mann-Whitney U test, respectively. Dichotomous variables were expressed as percentages and compared by the chi-square test. Results We studied 33 patients/61 affected eyes. There were no significant differences between TCZ and GLM at baseline in sex (♂/♀;4/21 vs 3/5; p=0.19), mean age (18.5±8.3 vs 19.9±8.7; p=0.55), positive ANA (95% vs 100%; p=0.7), uveitis duration before TCZ or GLM onset (116.4±93.6 vs 142.3±74.7 p=0.46), number of previous biological treatments (1.9±1.1 vs 2±1.4; p=0.84), VA (0.57±0.35 vs 0.5±0.37; p=0.42), combined immunosuppressive therapy (88% vs 75%; p=0.37), presence of cells in the anterior chamber (median, 1 [0–1] vs 1 [0.25–1.5]; p=0.6), vitritis (0 [0–0] vs 0 [0–1]; p=0.7), macular thickening (358.7±92.2 vs 313.6±77.1; p=0.32). There were no significant difference in the efficacy between TCZ and GLM (TABLE). After a mean follow-up of 20.48±11.7 months with TCZ and 24.25±17 months with GLM the following side effects were observed: TCZ: viral conjunctivitis plus bullous impetigo (n=1), severe thrombocytopenia and pneumonia. This last patient showed hemolytic anemia, thrombocytopenia and splenomegaly, for this reason treatment with TCZ was discontinued. With GLM cutaneous reaction was observed in 2 patients. Conclusions TCZ and GLM seem to be equally effective and safe for refractory uveitis associated-JIA. The superiority of one or the other should be established with prospective randomized studies “Head to Head” Disclosure of Interest None declared