Physician Global Assessment (PGA) and Psoriasis Area and Severity Index (PASI): why do both? A systematic analysis of randomized controlled trials of biologic agents for moderate to severe plaque psoriasis

Abstract

Background: Although there are many psoriasis assessment tools currently published, one of the unmet needs in psoriasis research remains consensus about the single best validated and reproducible assessment tool.

Objectives: In this systematic review we sought to determine the degree of correlation between two commonly used psoriasis assessment tools, the Psoriasis Area and Severity Index (PASI) and Physician Global Assessment (PGA).

Methods: Randomized controlled systemic trials in moderate to severe psoriasis were reviewed using the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. We recorded and compared the percent of patients achieving both 75% reduction in PASI score (PASI 75) and PGA 0 or 1 (clear or almost clear) at 8 to 16 weeks, 17 to 24 weeks, and greater than 24 weeks of treatment with the investigational drug.

Results: Our literature review yielded 30 randomized controlled trials using biologic agents in moderate to severe psoriasis. We found that the two assessment tools correlate very tightly except at the lower bounds of therapeutic efficacy. The r(2) values for the correlation between PASI 75 and a score of clear or almost clear on the PGA were 0.9157 at 8 to 16 weeks and 0.892 at 17 to 24 weeks.

Limitations: Limitations of our study include the small number of randomized controlled trials publishing the percent of patients achieving 75% reduction in PASI score and a score of clear or almost clear on the PGA after 24 weeks of therapy. In addition, our results are not generalizable beyond the patients with moderate to severe plaque psoriasis.

Conclusion: The two assessment tools are substantially redundant and either alone is a sufficient tool for assessing psoriasis severity in patients with moderate to severe disease. Because the PASI is better validated and more detailed, it remains the score of choice for clinical trials, but the simpler PGA may be well suited for community-based outcomes projects.