Rare Event Analysis by Flow Cytometry

In the performance of rare event analysis by flow cytometry it is often useful to know the minimum number of events to acquire to be confident of not missing the population of interest. Pertinent clinical areas include: 1. detection of residual disease, 2. detection of small PNH clones, and 3. detection of fetal cells in the maternal circulation. Since all events may be categorized as either "cells of interest" or "others", one approach is to use the cumulative distribution function of the binomial distribution. This calculation only deals with false negatives due to: 1. too few cells acquired ("total cells"), 2. too low a detection limit ("detection limit"), or 3. too high a requirement for the number of cells of interest to be observed ("cells of interest"). In contrast, this calculation has no bearing on false positives due to poor separation of the cells of interest from other cells or spurious "cells" of interest ("cosmic dust"). Roederer provides an excellent discussion of false positives [1]. Similar calculations are utilized in the area of quality control.