Composite outcome measures that successfully differentiate active treatments from placebo in psoriatic arthritis trials: a GRAPPA-OMERACT systematic review and network meta-analysis

Objectives

To identify which composite outcome measure most effectively distinguishes active treatments from placebo in randomised controlled trials (RCTs) of biologic or targeted synthetic disease-modifying antirheumatic drugs (b- or tsDMARDs) for psoriatic arthritis (PsA).

Methods

A systematic literature review (PROSPERO ID: CRD42024578203) of Cochrane Central Register of Controlled Trials and PubMed identified RCTs comparing b- or tsDMARDs with placebo reporting ≥2 of 7 composite outcome measures shortlisted for evaluation by the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis–Outcome Measures in Rheumatology (GRAPPA-OMERACT) working group (American College of Rheumatology Response Criteria [ACR], Composite Psoriatic Disease Activity Index [CPDAI], Disease Activity index for PSoriatic Arthritis [DAPSA], Minimal Disease Activity [MDA], Psoriatic Arthritis Disease Activity Score [PASDAS], and 3- and 4-Visual Analogue Scale [VAS]). Discriminant capacities were assessed using odds ratio (OR) for binary outcomes, with continuous outcomes converted from standardised mean differences and analysed through network and multivariate meta-analysis techniques.

Results

Of 2483 references, 24 trials were included (43 randomised comparisons). CPDAI was rarely reported, and no RCTs reported 3-VAS and 4-VAS. The main network meta-analysis showed differences in discriminant properties for DAPSA vs ACR20 (OR: 0.80; 95% CI: 0.66-0.97; favouring ACR20), DAPSA vs MDA (OR: 0.71; 0.57-0.87; favouring MDA), and PASDAS vs DAPSA (OR: 1.35; 1.10-1.66; favouring PASDAS). Supported by multivariate meta-analysis: MDA (OR: 5.06; 4.20-6.09), ACR20 (OR: 4.01; 3.41-4.73), PASDAS (OR:3.70; 3.00-4.56), DAPSA (OR: 3.02; 2.44-3.75), and CPDAI (OR: 1.86; 0.95-3.64). Sensitivity analyses confirmed a pattern of consistent numerical advantages for MDA and PASDAS. Exploratory analyses showed ACR70 had more discriminant capacity than ACR20 and DAPSA but not ACR50, MDA, and PASDAS.

Conclusions

ACR20, MDA, and PASDAS demonstrated greater discriminant capacity than DAPSA. Exploratory analyses suggested greater discriminant capacity for ACR70 compared with ACR20 and DAPSA but not with ACR50, MDA or PASDAS.