Difficult-to-manage and treatment-refractory psoriatic arthritis in patients treated with biological and targeted synthetic DMARDs: prevalence and characteristics in 5 Nordic registries.

Objectives

The European Alliance of Associations for Rheumatology (EULAR) recently proposed a framework for 'difficult-to-manage' (D2M) and 'treatment-refractory' disease in psoriatic arthritis (PsA). We explored real-world prevalence and characteristics of patients fulfilling registry-based components inspired by EULAR's framework among patients with PsA initiating biologic or targeted synthetic disease-modifying antirheumatic drugs (b/tsDMARD).

Methods

Cohort study from 5 Nordic biologics registries with follow-up of patients initiating first b/tsDMARD 1999 to 2020. First (step 1a), 3 cohorts were defined: discontinuation of ≥2, ≥3, or ≥4 b/tsDMARDs, respectively. Additional components were added sequentially to each cohort: (1b) b/tsDMARD-discontinuation reason (inefficacy/side effects); (1c) discontinuation of ≥2 different mechanisms of action (with 1a+1b+1c providing nuanced understanding of switching patterns); (2) problematic signs/symptoms (patient global score ≥ 30 mm); (3) persistent disease activity (tender/swollen joint count ≥1, C-reactive protein [CRP] > 10 mg/L, or Disease Activity index in PSoriatic Arthritis 28-joint counts >14). Treatment-refractory disease required primary/secondary b/tsDMARD inefficacy (not side-effects/intolerability/contraindications), and objective inflammation (tender/swollen joint count ≥ 1, or CRP > 10 mg/L).

Results

Among 14,362 patients, discontinuation of ≥2/≥3/≥4 b/tsDMARDs occurred in 36%/18%/10%, respectively, during a median(IQR) follow-up of 6.3 years (2.4-10.7). Applying all components (1a-3), D2M prevalences were 2%/3%/3%, respectively. Corresponding treatment-refractory prevalences were 1.2%/1.2%/0.8%, respectively. Female sex, depression, and opioid use increased with the number of b/tsDMARD discontinuations.

Conclusions

In routine care, multiple b/tsDMARD discontinuations were common. Two percent were D2M, and 1.2% were treatment-refractory, both characterised by a higher prevalence of women, depression, and opioid use. Although our retrospective design did not allow testing of the EULAR definitions, our findings highlight the complexity of D2M and treatment-refractory PsA and underline the need for prospective studies.