Real-world drug discontinuation of acitretin (ACI), ciclosporin (CsA), fumaric acid esters (FAE) and methotrexate (MTX) in patients with moderate-severe psoriasis

Real-world discontinuation of ACI, CsA, FAE and MTX in patients with moderate-severe psoriasis is poorly characterised. The aim of this study was to determine whether treatment history affects the discontinuation rates of these therapies.
The British Association of Dermatologists Biologics and Immunomodulators Register (BADBIR) is a pharmacovigilance register of systemic therapies for psoriasis. In total, 4115 patients registering to BADBIR from 2007-2017 on ACI, CsA, FAE or MTX with at least 6 months of follow-up were analysed. Exposure time was calculated from initiation to: discontinuation date; censor at the latest follow-up; or death. Patients were categorised as incident (first systemic), prevalent (previously prescribed registration therapy), or previous systemic users (previously prescribed another systemic). Cumulative incidence was used to determine the mean proportion (%, 95% confidence interval [CI]) of patients discontinuing therapy in the first two years of follow-up.
In total, 767 (19%), 1022 (25%), 335 (8%) and 1991 (48%) patients registered to ACI, CsA, FAE and MTX, respectively. The proportions of incident, prevalent and previous systemic users, respectively, were similar for ACI (42%; 16%; 42%), CsA (38%; 16%; 46%) and MTX (41%; 18%; 41%) with 66% previous systemic users registering on FAE (19% incident; 15% prevalent). The overall mean discontinuation rates at two years were: 31% ACI (27-34), 41% CsA (38-45), 27% FAE (19-43) and 26% MTX (23-28) for incident users; 33% ACI (27-39), 40% CsA (35-44), 34% FAE (25-46) and 26% MTX (23-29) for prevalent users; and 36% ACI (33-40), 39% CsA (36-42), 35% FAE (31-39) and 31% MTX (29-33) for previous systemic users, respectively.
In conclusion, patients with previous systemic use prior to their registration therapy had higher discontinuation rates than incident or prevalent users. These findings require further exploration to determine whether predictors of treatment discontinuation or stop reasons differ between previous exposure categories.