Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 508 - Innovative Statistical Methods for Preclinical to Clinical Translatability in Drug Development
Type: Topic Contributed
Date/Time: Thursday, August 11, 2022 : 8:30 AM to 10:20 AM
Sponsor: Biopharmaceutical Section
Abstract #322261
Title: A Bayesian Approach for MRE Back Translation Based Decision Making with the Attenuation Ratio Metric
Author(s): Thomas Bradstreet* and Haiying Tang and Matthew Fronheiser and Thomas Petrone and Lei Zhao
Companies: Bristol-Myers Squibb and CHDI Management, Inc. and Bristol-Myers Squibb and Bristol-Myers Squibb and Bristol-Myers Squibb
Keywords: probability-based decision making; Go/No-Go; risk assessment; posterior distribution; magnetic resonance elastography
Abstract:

Two frequently asked questions in preclinical translational and discovery research are, Given the data we observed: What is the population effect size? How sure are we about the population effect size? Preclinical animal studies often comprise three independent treatment groups: a control; an inducer which induces a disease condition; and a tool compound which attenuates the induced effect. The Attenuation Ratio Metric, R, quantifies the proportional decrease in population mean response from inducer to inducer + tool compound relative to the window of biological opportunity between population mean responses for the control and the inducer. Specifically, R = (MUinducer+tool - MUinducer) / (MUinducer - MUcontrol). The Bayesian strategy conditions on the observed data and evaluates the strength of observed evidence leveraging posterior probabilities of the Attenuation Ratio. We show that the Bayesian strategy provides a direct, more informative, statistical analysis and decision-making tool than the frequentist sampling-theory-based significance testing procedures currently favored by preclinical scientists. Examples source from MRE back translation studies in mice.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program