Optimizing the ongoing search for new treatments for Parkinson disease: using futility designs

Abstract

Many agents are being considered for treatment of Parkinson disease (PD). Given the large number of agents and the limited resources to evaluate new agents, it is essential to reduce the likelihood of advancing ineffective agents into large, long-term Phase III trials. Futility design methodology addresses this goal. The authors describe how a single-arm Phase II futility study uses a short-term outcome to compare a treatment group response to a predetermined hypothesized or historically based control response. The authors present advantages and limitations of futility designs along with examples derived from the data archive of a large Phase III efficacy study of treatments to delay PD progression, the Deprenyl And Tocopherol Antioxidative Therapy Of Parkinsonism (DATATOP) trial. Using the same control progression rate and treatment effect assumptions used to power the original DATATOP trial, the authors calculated the number of subjects needed to conduct two 12-month futility studies. DATATOP was designed to enroll 800 patients. Using data on 124 consecutive subjects randomized into each of the DATATOP treatment groups, the authors identified tocopherol as futile and deprenyl as worthy of further study. Using Phase II information, DATATOP could have been simplified from a 2 x 2 factorial design to a comparison of deprenyl vs placebo. While not testing efficacy, futility designs provide a strategy for discarding treatments unlikely to be effective in Phase III. A limitation is the dependence on historical data or hypothesized outcomes for untreated controls. Futility studies may decrease the time to identify treatments unworthy of further pursuit and reduce subjects' exposure to futile treatments.