Sophisticated Bayesian Methods are gaining a lot of traction as they bring flexibility and speed to clinical trial design and analysis. These methods are transforming clinical research as they are being applied in a wide range of therapeutic areas as well as medical device trials. For example, a cluster-randomized trial designed to assess the effectiveness of a machine-learning-based clinical decision support system for physicians treating patients with depression, two Bayesian adaptive designs for cluster-randomized trials are proposed to allow for early stopping for efficacy at pre-planned interim analyses.^[1] The Pfizer/BioNTech vaccine is also based on a Bayesian trial.

What are Bayesian Methods

Bayesian statistics allow us to make probabilistic inferences on the parameter of interest, which is missing in a traditional frequentist approach. The Bayesian approach allows every new piece of data to serve as evidence to update a hypothesis. Bayes’ rule consists of a “prior” either based on evidence already collected, or scientific findings in the case of early-phase trials. A rule then explains how to update these priors in order to make sense of newly collected evidence. A “posterior” is then the result of the prior being updated in light of this new evidence. As the trials evolves with new in-trial insights, these Bayesian methods enable statisticians and sponsors to create flexible trial designs and accelerate learning.

Wide Application of Bayesian Statistics

Innovations in Bayesian methods are occurring at every phase of clinical research from early phase dose-escalation trials to the creation of sound market access strategy. Their wide range of usage includes:

• The design flexibility provided to meet the multiple objectives of early phase studies
• The historical and external data and information borrowing mechanisms that strengthen late-phase regulatory dossiers
• Complex trial designs as they enable greater flexibility and better ability to respond to the needs of the master protocol designs

According to Professor Yuan Ji of the University of Chicago, who spoke at a recent Cytel webinar, several recent trends in Bayesian statistics are beneficial to understand.

MUCE

One of the more innovative elements of Bayesian designs are Multiple Cohort Expansion (MuCE). MUCE is an advanced Bayesian approach superior to the Simon’s 2-stage design for expansion cohorts trials and master protocols. Such methods are particularly important for areas like oncology where several doses and several indications must be tested for successful completion of early phase trials, and optimal choice of dose and population to move on from early phase to a reasonable dosage for Phase 3.

Bayesian for Exploratory Studies

In the Cytel webinar, Professor Ji provides a Bayesian perspective on statistical modeling and decision making for clinical trials. In the early phase drug development, more flexible modeling approaches are typically allowed due to their exploratory nature. However, errors made in early phase have grave impact to drug development since these errors lead to major waste of resources. During the webinar Prof Ji discusses:

1. How Bayesian modeling can help mitigate the potential errors in decision making in early-phase expansion cohort trials. In late phase, due to current regulatory constraints, statistical errors are expressed in frequentist type I and type II error rates.
2. How Bayesian and frequentist sequential decision making are different, especially through interim analyses. Various Bayesian approaches are presented during the webinar.

Reference

[1] New designs for Bayesian adaptive cluster-randomized trials

Share Via