Persistence, effectiveness and safety of ustekinumab compared to TNF inhibitors in psoriatic arthritis within the Italian PsABio cohort

Objective To compare real-world persistence, effectiveness and tolerability of ustekinumab versus TNF inhibitors (TNFi) in psoriatic arthritis (PsA). Methods One-year data from Italian subjects enrolled in the PsABio study (PsA patients receiving 1st- to 3rd-line treatment with ustekinumab or TNFi) were evaluated. Treatment persistence was analysed using Kaplan-Meier curves; hazard ratios (HR) of stopping treatment, and the corresponding 95% confidence intervals (CI), were computed through Cox regression models. Proportions of patients reaching clinical effectiveness endpoints were analysed using logistic regression, including propensity score (PS) adjustment for imbalanced baseline covariates, and non- response imputation if treatment was stopped/switched. Results Among 222 participants with follow-up data (effectiveness set), 101 received ustekinumab and 121 TNFi. In the ustekinumab group, 74.3% continued treatment up to 12 +/- 3 months compared to 63.6% in the TNFi group. Ustekinumab showed better persistence than TNFi, overall and in specific subgroups (females, monotherapy without methotrexate, BMI <25 or >30 kg/m(2), patients receiving ustekinumab as 2nd-line treatment instead of a second TNFi). Overall, the PS-adjusted HR of treatment discontinuation was 0.46 (95% CI: 0.26-0.82) for ustekinumab vs. TNFi. cDAPSA LDA/remission was achieved in 43.5% of ustekinumab and 43.6% of TNFi-treated patients, while MDA was achieved in 24.2% and 28.0% of patients, respectively. After PS adjustment, odds ratios of clinical effectiveness did not differ significantly. Both treatments showed an acceptable safety profile. Conclusion This prospective, real-life study found a better persistence of ustekinumab than TNFi in PsA patients. At 1 year, both treatments showed similar effectiveness.