Further analysis of psoriatic arthritis disease activity score (PASDAS) and composite psoriatic disease activity index (CPDAI) using data from a placebo-controlled trial of certolizumab pegol in psoriatic arthritis

Background Several new disease-specific composite disease activity measures have been developed for psoriatic arthritis (PsA). Objectives Using data from a randomized double-blind placebo-controlled trial of certolizumab pegol (CZP) in PsA (RAPID-PsA, NCT01087788) we aimed to compare the performance of the Psoriatic Arthritis Disease Activity Score (PASDAS), the Composite Psoriatic Disease Activity Index (CPDAI), and modifications of the CPDAI with the Disease Activity Score 28 (DAS28), which was developed for use in rheumatoid arthritis. Methods The CPDAI and PASDAS composite measures have been described previously.1,2 In RAPID-PsA not all components of CPDAI were collected, therefore psoriasis body surface area was substituted for Psoriasis Area Severity Index, and the Health Assessment Questionnaire was used as the function modifier for the axial component of this measure (CPDAIm). The CPDAI was further modified by the addition of a patient (pt)-completed outcome domain, based on the pt visual analogue scale scores for arthritis pain and global disease activity (CPDAIpro). The performance of these measures was assessed after 12 weeks of CZP treatment by calculating, for the change from baseline, the standardized response mean, the baseline-adjusted effect size compared to placebo (ESadj, based on analysis of covariance), and the required sample size based on the ESadj. Data are reported for those pts with all outcome components available. Only observed data were used, without imputation. Results The largest ESadj was found for PASDAS, followed by DAS28, CPDAIm and CPDAIpro (Table). These figures translated into differences in sample size estimation for further studies. Conclusions The PASDAS and CPDAI composite measures demonstrated good responsiveness and discriminative ability in this trial, supporting further exploration of their use in PsA clinical trials. Differences in the performance of these new PsA-specific composite measures were seen in this dataset. These results may help inform the selection of appropriate composite measures for PsA in future studies. Addition of a pt-completed outcome domain to the CPDAI did not appear to improve performance of the composite score. (Figure Presented).