O18 Risk of incident psoriatic arthritis in psoriasis patients treated with immunomodulators: a prospective cohort study
Ali Al-Janabi, Oras Alabas, Brian Kirby, Nick Reynolds, Weiyu Ye, Anne Barton, Catherine Smith, Richard WarrenAbstract
Introduction and aims
Psoriatic arthritis (PsA) affects 20% of individuals with psoriasis. It remains unclear whether systemic immunomodulators used for psoriasis differentially influence the risk of new-onset PsA. This study compared incident PsA risk across therapeutics used for psoriasis.
Methods
Data from the British Association of Dermatologists Biologics and Immunomodulators Register (BADBIR) were analysed. Included participants were aged ≥16 years, had no PsA diagnosis at enrolment and were treated for plaque psoriasis with methotrexate (MTX), tumour necrosis factor inhibitors (TNFi), interleukin (IL)-12/23i, IL-17i or IL-23i monotherapy. Exposure time was counted until PsA onset, treatment discontinuation, last follow-up or death. Risk of incident PsA by drug class was estimated using propensity-score weighted flexible parametric survival models.
Results
The primary analyses included 22 661 treatment exposures among 15 872 participants, with 86 909 person-years exposure time (MTX 15%, TNFi 37%, IL-12/23i 27%, IL-17i 13% and IL-23i 8%). There were 1120 incident PsA events. Relative to MTX, risk of incident PsA was lower with grouped biologics [hazard ratio (HR) 0.82, 95% confidence interval (CI) 0.71–0.95], TNFi (HR 0.82, 95% CI 0.69–0.96) or IL-23i (HR 0.50, 95% CI 0.37–0.68). The effect of TNFi, but not IL-23i, was dependent on mean Psoriasis Area and Severity Index (PASI) during the exposure. In previously biologic-naïve individuals, all biologic classes conferred a lower risk than MTX, independently of mean PASI. Relative to TNFi, risk of incident PsA was lower with IL-23i (HR 0.56, 95% CI 0.39–0.80), but not IL-12/23i (HR 0.85, 95% CI 0.72–1.00) or IL-17i (HR 1.08, 95% CI 0.88–1.34). When exposures were restricted to those starting pre-IL-23i availability in BADBIR, the risk was lower with IL-12/23i than TNFi (HR 0.57, 95% CI 0.45–0.71).
Conclusions
Biologics were associated with reduced incident PsA risk compared with MTX, particularly in biologic-naïve individuals. IL-23 inhibition appears to be associated with a lower risk of incident PsA than TNFi, but this could be due to prescribing bias. Further studies are needed to address this issue.