Estimators of a Dose-Reduction Factor: A Comparison Study

• Activity Number: 522
• Type: Contributed
• Date/Time: Wednesday, August 1, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Reid Landes*+ and Shelly Y. Lensing and Ralph L Kodell and Martin Hauer-Jensen
• Companies: University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences
• Keywords: relative potency ; ED50

Abstract

The FDA requires animal lethality studies in research for effective countermeasures to radiation exposure. Typically, radiation researchers report the estimated dose expected to kill 50% of the animals (LD50). When comparing an irradiated group treated with a countermeasure to an irradiated control, an important parameter is the dose reduction factor (DRF) - a ratio of two LD50s. Here, we provide point and interval estimators of the DRF arising from Fieller's theorem, an inverted likelihood ratio test, a Bayesian method, and a bootstrap method. We compare statistical properties of these estimators with a Monte Carlo study covering scenarios typically found in radiation research.