Determination of Bioassay Cut Point Using Confidence Limit of Percentile
*Meiyu Shen, FDA 


Cut point of the assay is the level of response of the assay at or above which a sample is defined to be positive and below which it is defined to be negative. Food and Drug Administration Guidance for Industry on Assay Development for Immunogenicity Testing of Therapeutic recommends the cut point to have an upper negative limit of approximately 95 percent. The sample percentile is a point estimate with variability that decreases with larger sample sizes. Therefore it does not assure to have at least 5 percent false positive rate when the sample size is not sufficiently enough. Considering this, a confidence limit needs to be used instead. We conducted an extensive literature review on the estimation of the statistical cut point and performed comparison of the selected methods for bias, the coverage percentage and false positive rate of the extreme values. We evaluate sample percentile, and a few confidence limit approaches including exact confidence limit of normal percentile (Charkraborti and Li, 2007), approximate Z confidence limit of normal percentile. It is shown that the coverage percentage of sample percentile is much larger than (1 - percentage) with the number of assays conducted in practice. We recommend using the lower confidence limit of the percentile for cut point determination.