Webinars and Videos

Looking at RAR with 3 aims: 1) A surprising amount of the literature considers just the two arm setting, when arguably the most interesting setting for this technique is the multi-arm one. 2) There have been a number of papers over the years critical of RAR and I’d like to add present some results in favour of RAR. 3)

Designing a phase 1 dose escalation using Bayesian Logistic Regression Model (BLRM), with open enrolment (rather than cohort) using FACTS N-CRM.

Using FACTS to design a two-arm study where differences in adherence between the arms drive the effect size. Uncertainty in the effect size prompted spacing the interims to correspond to different sample sizes resulting in irregular interim analyses. Using a uses a finite mixture model from previous data on the analysis endpoint to justify the irregular interims.

Is this the most complex and most successful trial ever designed and run? Looking at the AWARD 5 trial for what became Eli Lilly’s Trulicity drug. The drug used response adaptive randomization, longitudinal modelling, dose response modelling and a utility function combining 4 endpoints. The trial selected a successful dose and switched early from phase 2 to a successful phase 3.

In this September 6, 2024 episode by SciPod Health and Medicine, SciPod summarizes the recent paper “Lecanemab for Patients with Early Alzheimer Disease: Bayesian Analysis of a Phase 2b Dose-Finding Randomized Clinical Trial” by Dr. Donald Berry (Founder of Berry Consultants) and Scott Berry (President of Berry Consultants).

This Podcast series by JAMA provides valuable interviews about the statistical and methodological foundations of clinical research.