Systematic review of the use of CRP in clinical trials for psoriatic arthritis: a concern for clinical practice?

BACKGROUND: C reactive protein (CRP) levels are suggested as serum biomarkers in the diagnosis and prognosis of psoriatic arthritis (PsA). However, increased CRP levels are found in less than 50% of PsA patients even in the presence of active disease. OBJECTIVES: To evaluate the role of CRP levels in interventional clinical trials in PsA patients to better understand the trial generalisability, relationship with disease activity and predictive value for treatment response and decision making. METHODS: A systematic review was conducted via PubMed, Cochrane and Embase. We focused on phase III trials in PsA. RESULTS: Eight of 22 studies applied minimum baseline CRP levels for inclusion. Baseline CRP levels were wide-ranging (0.1–238 mg/L) and lower in studies without CRP in the enrolment criteria. All 22 studies used the American College of Rheumatology (ACR20) response and other endpoints that integrated CRP levels. One of seven studies that evaluated individual ACR-score components revealed a decrease in CRP levels along with improvement of other endpoints. Subanalyses show conflicting evidence on CRP levels as predictor of disease course. CONCLUSION: CRP levels were inconsistently used as inclusion criterion in clinical trials, often limiting generalisability of the data. The use of composite scores such as ACR20 or Disease Activity Score-28-CRP is also limited since baseline levels of CRP affects their sensitivity to change. High CRP levels may be an individual predictor for disease progression and response to treatment, but the current conflicting findings and selective patient trial inclusions, do not allow CRP to play a very prominent role in treatment decision making.