We report 4 cases of CNI uveitis with lack of efficacy of subcutaneous tocilizumab (Table 1), which improved after switching administration route. An improvement of visual acuity with intravenous tocilizumab was observed for all patients. Moreover, corticosteroids could be stopped for two patients. Rapid efficacy of intravenous tocilizumab was observed, between 2 and 3 months. No specific side effect was reported. Interestingly, three out of the four patients had a BMI > 35 kg/m2.

Two prospective trials studied the efficacy of tocilizumab in CNI uveitis. STOP-uveitis included 37 patients with non-anterior uveitis: 18 patients received 4 mg/kg/4 weeks of tocilizumab and 19 patients received 8 mg/kg/4 weeks [4]. Active uveitis was confirmed on vitreous haze assessment. The authors observed a significant improvement in visual acuity, a decrease in vitreous haze and a reduction of central foveolar thickness. There was no significant difference between the two doses of tocilizumab, and no difference between naïve patients and patients previously treated with anti-TNF-α. Tocilizumab, administered subcutaneously, was studied in JIA in the APTITUDE study [8]. This was a multicenter single-arm study including 21 patients with active uveitis, refractory to anti-TNF-α agents. One third of the patients stopped the treatment before the first 3 months because of inefficacy. The study did not meet the prespecified criterion at 12 weeks to justify a phase 3 trial, suggesting less efficacy of subcutaneous tocilizumab in CNI uveitis.

Only half of the patients could stop corticosteroids, the other two patients had to maintain corticosteroid doses greater than or equal to 5 mg/day. In addition, patient n°4 had to bring the tocilizumab injections closer due to a decrease in the effectiveness of the treatment. We still concluded that the treatment was effective. Indeed, the patients we present are multi-refractory patients, with failure of several lines of treatment, with limited therapeutic possibilities.

Abdallah et al. found no difference in terms of efficacy or immunogenicity between intravenous and subcutaneous tocilizumab for 1759 patients with RA [9]. Because subcutaneous tocilizumab was administered at a fixed dose, independent of body weight, the correlation between efficacy and body weight was studied [7, 9]. Tocilizumab concentration was lower for patients with a body weight > 100 kg, i.e. a maximum observed concentration of 69.7 µg/ml for a body weight < 60 kg versus a maximum observed concentration of 26.9 µg/ml for a body weight ≥ 100 kg, and was inversely correlated with body weight. In Arad's study [7], including 100 patients with RA treated with subcutaneous tocilizumab at a dose of 162 mg/week, body weight was inversely correlated with disease improvement. Abdallah et al. reported a target concentration of tocilizumab in RA of 10 µg/ml [9]. There were no anti-drug antibodies in this study. In our study, cases n°2 and n°4 did not have anti-drug antibodies and had a tocilizumab plasma residual level of > 10 µg/ml, with a body weight > 100 kg. No study has focused on the correlation between tocilizumab plasma residual and intraocular inflammation control. The ocular immunological environment may evolve independently of the peripheral circulation, explaining an incomplete correlation. Interestingly, three out of four patients had a BMI > 35 kg/m2, suggesting that, in this population, intravenous tocilizumab might be used in cases of CNI uveitis, for better bioavailability. We did not suspect medication non-adherence in our patients.

Quesada-Masachs et al. [10] also reported the inefficacy of subcutaneous tocilizumab in chronic anterior uveitis in JIA, although all patients were previously controlled with intravenous tocilizumab; switching to subcutaneous tocilizumab was done for patient comfort. Treatment management after relapse was not described. Conversely, in our study, three of the four patients were first treated with subcutaneous tocilizumab.

Interestingly, in CNI uveitis, a similar difference between subcutaneous and intravenous administration route has been described with secukinumab, an anti-IL17 agent. Dick et al. [11] reported the inefficacy of subcutaneous secukinumab in three randomized controlled trials. However, Letko et al. [12] showed a superiority of intravenous versus subcutaneous secukinumab to control intra-ocular inflammation with corticosteroid sparing effect in 33 patients.

Our study has some limitations. This is a retrospective study, with a small population and an inhomogeneous follow-up of all patients. Moreover, the therapeutic strategies chosen may be questionable, particularly the absence of an increase in the dosage of infliximab, explained by the prescription practices.

To summarize, in CNI uveitis, we suggest that before considering tocilizumab failure, therapeutic optimization could be obtained by a switching from a subcutaneous to an intravenous administration route, allowing corticosteroid sparing effect.