OBJECTIVES: To compare treatment persistence and adherence over one year in biologic naïve psoriatic arthritis (PsA) patients initiating apremilast or biologics using a U.S. administrative claims database.

METHODS: This retrospective study selected adult PsA patients initiating apremilast or biologics within the January 2013- June 2016 Truven Health MarketScan claims databases. Patients were required to be apremilast/biologic naïve on the index agent in the 12-month pre-index period and have continuous enrollment in the 12-month pre- and post-index periods (index date = initial apremilast or biologic claim). Biologic users were matched 2:1 to apremilast users. Treatment persistence at 12 months was defined as continuous treatment without (1) a >60-day gap in therapy or (2) a switch to a different PsA treatment and measured during the 12-month post-index period. Patients were adherent if their medication possession ratio (MPR) was ≥80% while persistent on the index agent.

RESULTS: A total of 381 patients initiating apremilast were matched to 761 patients initiating biologics. Baseline characteristics were similar in both groups (mean age 51 years, 60% female, mean Charlson score 0.6). Treatment persistence at 12 months for apremilast users was similar to biologic users (43% vs 48%; p=0.082). Average time to non-persistence was 124 days for apremilast users compared to 132 days for biologics users (p=0.194). Among non-persistent users, apremilast users and biologic users had similar switch rates (7% vs 9%; p=0.389) and discontinuation rates (93% vs 91%; p=0.389). Average persistence-based MPR was 0.870 for apremilast users compared to 0.854 for biologic users (p=0.056), and the adherence rates were 77% and 73%, respectively (p=0.175).

CONCLUSIONS: At 12 months, persistence and adherence as well as drivers of non-persistence were similar for biologic naïve patients initiating apremilast or biologics for the treatment of PsA in the U.S.