Objectives: To evaluate the characteristics of patients (pts) with PsA treated by ustekinumab (UST) or secukinumab (SEK) and to compare real-world persistence of UST and SEK in PsA. Methods: In this retrospective, national, multicentre cohort study, pts with PsA (CASPAR criteria or diagnosis confirmed by the rheumatologist) initiating UST or SEK with a follow-up ≥6 months were included from January 2011 to April 2019. The persistence between SEK and UST was assessed after considering the potential confounding factors by using pre-specified propensity-score methods. Causes of discontinuation and tolerance were also collected. Results: A total of 406 pts were included: 245 with UST and 161 with SEK. The persistence rate was lower in the UST group compared with the SEK group [median persistence 9.4 vs 14.7 months; 26.4% vs 38.0% at 2 years; weighted hazard ratio (HR) = 1.42; 95% CI: 1.07, 1.92; P =0.015]. In subgroup analysis, the persistence rate of SEK associated with MTX was significantly higher than that of UST associated with MTX: HR = 2.20; 95% CI: 1.30, 3.51; P =0.001, in contrast to SEK vs UST monotherapy: HR = 1.06; 95% CI: 0.74, 1.53; P =0.75. Discontinuation due to inefficacy was reported in 91.7% (SEK) and 82.4% (UST) of pts. Discontinuation due to an adverse event was reported in 12.2% (SEK) and 7.7% (UST) of pts. Conclusion: In this first study comparing UST and SEK, the persistence of SEK was higher than that of UST in PsA. In subgroup analysis, this difference was only found in association with MTX.