METHODOLOGY FOR JUSTIFYING THE REDUCTION IN THE USE OF LABORATORY ANIMALS NEEDED FOR RELEASE AND STABILITY TESTING PURPOSES

Abstract

United States Food and Drug Administration (FDA) regulatory guidance documents state that valid scientific data should be obtained by using the minimum number of animals possible; however, guidance documents do not establish that minimum threshold. Instead, they suggest the use of computer simulations in order to determine it. This project uses toxin neutralization assay (TNA) data and employs three different types of resampling simulations: bootstrap, jackknife, and monte carlo, in order to investigate reducing the number of animals to a minimum which are required to achieve a relative potency (RP) that is comparable to a benchmark dataset from a four dose level (DL), twenty animals per DL, study. In these studies, the results showed that computer simulations which used benchmark TNA values, neutralization factors which protects 50% of the cells (NF₅₀), proved it is possible to reduce the total number of animals required to estimate a four DL RP by 25%. Thus, improving the return on investment and efficiency as well as enhancing humane animal use.