Variance of the Stimulation Index for the Local Lymph Node Assay

Takashi Omori1 and Takashi Sozu2
1Department of Biostatistics, Kyoto University School of Public Health, Kyoto, Japan
2The Center for Advanced Medical Engineering and Informatics, Osaka University, Osaka, Japan

AATEX 12(3):212-217, 2007

The local lymph node assay (LLNA) is a well-established method for assessing skin sensitization; it involves the use of mice and provides an alternative to tests requiring guinea pigs. In this assay, the endpoint of interest is the stimulation index (SI) that is defined as the mean disintegrations per minute (DPM)/mouse for the chemical-treated group divided by that for the vehicle control group. Therefore, the SI is a ratio of 2 variables. However, some researchers often ignore the variation in the vehicle control group while assessing chemicals based on the SI. This approach, which we refer to as the ignorance approach, may underestimate the variation in SI. We developed an alternative approach by the delta method, and we evaluated the performance of both approaches through a simulation and a numerical examination. We found that the ignorance approach is not acceptable for estimating the variance of SI because the value obtained is severely biased toward underestimation. Instead, analysis by using the delta method approach is recommended. However, it should be noted that this approach is also slightly biased in the case of large variations in the DPM/mouse values in the vehicle control group.

Key words: LLNA, stimulation index, statistical analysis, variance, Delta method


(AATEX: Altern. Animal Test. EXperiment.: Alternatives to Animal Testing and EXperimentation)