Efficacy and safety of tocilizumab in adult-onset Still's disease: Real-life experience from the international AIDA registry

Long-term efficacy and safety of tocilizumab (TCZ) in adult-onset Still's disease (AOSD) mostly derive from small case series. Herein we report a registry-based study investigating TCZ efficacy and safety in a cohort of patients with AOSD evaluated by clinical and serum inflammatory markers as well as drug retention rate analysis. This is an international multicentre study analyzing data from patients with AOSD regularly enrolled in the AIDA registry. TCZ efficacy was evaluated between baseline and last follow-up assessment in terms of changes in the Pouchot score and laboratory findings. Drug-retention rate was estimated by the Kaplan-Meier method, while Cox-regression analysis was employed to detect potential predictive factors of treatment withdrawal. Data from 31 patients (15 men, 16 women) refractory to the conventional therapies and treated with TCZ were extracted from the AIDA registry. Mean ± SD time of treatment duration with TCZ was 24.32 ± 20.57 months. Median (IRQ) Pouchot score significantly decreased throughout the study period (p=0.001) with a significant difference between baseline [2.00 (4.00)] and 6 month-follow-up [0.00 (0.00)] (p=0.003) and between baseline and last follow-up assessment [0.00 (0.00)] (p=0.032), while no differences were observed between 6 month-evaluation and last follow-up assessment (p=0.823). Similarly, laboratory parameters significantly decreased from baseline to the last follow-up visit. With regard to drug survival, cumulative TCZ retention rate at 12-, 24-, and 36-month follow-up visit were 83.1%, 71.7% and 63.7%, respectively, without significant differences between biologic naïve patients and those previously treated with other biologics (p=0.329). Likewise, no significant differences were observed between chronic articular course of AOSD and other types of disease course (p=0.938) or between patients co-administered with conventional immunosuppressants and patients receiving TCZ as monotherapy (p=0.778). Cox-regression analysis identified no variable associated with a higher hazard of treatment withdrawal. Treatment was discontinued in 9 patients due to long-term remission (n=4), adverse events (n=2), loss of efficacy (n=1), non-medical reason (n=1) and unspecified cause (n=1). Mean glucocorticosteroids daily dose significantly decreased from baseline (18.36 ± 24.72 mg) to the last follow-up assessment (4.02 ± 4.99 mg, p=0.003). TCZ allows control of disease activity as well as normalization of serum inflammatory markers in both systemic and chronic articular form of AOSD. Additionally, TCZ displays an excellent drug retention rate while minimizing the risk of long-term exposure to corticosteroids.