Online Program Home
My Program

Abstract Details

Activity Number: 363
Type: Contributed
Date/Time: Tuesday, August 2, 2016 : 10:30 AM to 12:20 PM
Sponsor: Biopharmaceutical Section
Abstract #320488 View Presentation
Title: The Estimation of Absolute IC50 and Its 95% Confidence Interval
Author(s): Qin Liu* and Xiangfan Yin and Quentin McAfee and Ravi Amaravadi and Jeffery Winkler and Meenhard Herlyn
Companies: The Wistar Institute and The Wistar Institute and University of Pennsylvania and University of Pennsylvania and University of Pennsylvania and The Wistar Institute
Keywords: dose-response ; IC50 ; four-parameter logistic regression model ; bootstrap
Abstract:

Inhibitory concentration 50% (IC50) is commonly used to compare the potency of drugs in pre-clinical anticancer drug screening study. IC50 represents the concentration of a drug that is required to inhibit 50% of cancer cells in the in vitro experiment. The IC50 of a drug is usually estimated using four-parameter logistic (4PL) regression analysis. When comparing IC50s between drugs, it needs to clear if the relative IC50 or the absolute IC50 was estimated. In practice, different drug may reach different maximum and minimum inhibition rates along the tested dose levels. By taking into account such variation in model fitting, the 4PL model estimated IC50 is the relative IC50, which is not comparable among tested drugs. This paper has proposed an approach for the calculation of the comparable IC50 (absolute IC50) and provided a formula for variance estimation with the delta method. After obtaining the absolute IC50 and its variance, the 95% confidence interval of the absolute IC50 is estimated and readily suited to comparisons among drugs.


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

Back to the full JSM 2016 program

 
 
Copyright © American Statistical Association