Developing, Planning and Conducting an Interim Analysis: Lessons From the DEVOTE Cardiovascular Outcomes Trial


BACKGROUND: In 2013, a randomized, double-blind, active comparator-controlled, event-driven cardiovascular outcomes trial (DEVOTE) was initiated to compare the cardiovascular safety of insulin degludec (degludec) versus insulin glargine 100 units/mL (glargine U100) in patients with type 2 diabetes at high risk of cardiovascular events. The FDA agreed that an interim analysis could form the basis for an early regulatory approval. We report here the operational model developed to support the DEVOTE interim analysis and the results.

METHODS: The interim analysis model was designed to reduce the risk of any confidentiality breaches. The Data Access Management Plan comprehensively described the interim analysis operational processes and procedures to maintain the integrity of the ongoing trial while the interim analysis was conducted, submitted, and acted upon by the FDA, and also until completion of the full trial. Most importantly, those who were unblinded to the interim results were limited to a team of 14 members.

RESULTS: A total of 150 first major adverse cardiovascular events were recorded at cut-off for the interim analysis. The estimated hazard ratio was 0.92 (95% CI 0.67, 1.27) and non-inferiority to glargine U100 was confirmed as the upper bound of the confidence interval was below 1.8, as prespecified. Based on these results, the FDA approved the use of degludec and degludec/insulin aspart (IDegAsp) in the United States in 2015 before trial completion.

CONCLUSIONS: The DEVOTE interim analysis succeeded as a model by which to conduct an interim analysis and submit confidential data for regulatory review and action while continuing the trial to address a primary hypothesis.

Read More

Theilgaard H, Mollerup I, Helmark IC, et al. Ther Innov Regul Sci 2019;53:279–86.