The interpretation of long-term trials of biologic treatments for psoriasis: trial designs and the choices of statistical analyses affect ability to compare outcomes across trials.

Psoriasis is a chronic disease requiring long-term therapy, which makes finding treatments with favourable long-term safety and efficacy profiles crucial. The goal of this review is to provide the background needed to evaluate properly long-term studies of biologic treatments for psoriasis. Firstly, important elements of design and analysis strategies are described. Secondly, data from published trials of biologic therapies for psoriasis are reviewed in light of the design and analysis choices implemented in the studies. Published reports of clinical trials of biologic treatments (adalimumab, alefacept, etanercept, infliximab or ustekinumab) that lasted 33 weeks or longer and included efficacy results and statistical analysis were reviewed. Study designs and statistical analyses were evaluated and summarized, emphasizing patient follow-up methods and handling of missing data. Various trial designs and data handling methods are used in long-term studies of biologic psoriasis treatments. Responder analyses in long-term trials can be conducted in responder enrichment, re-treated nonresponder or intent-to-treat trials. Missing data can be handled in four ways, including, from most to least conservative, nonresponder imputation, last-observation-carried-forward, as-observed analysis and anytime analysis. Long-term clinical trials have shown that adalimumab, alefacept, etanercept, infliximab and ustekinumab are efficacious for psoriasis treatment; however, without common standards for these trials, direct comparisons of these agents are difficult. Understanding differences in trial design and data handling is essential to make informed treatment decisions.