Background: The disease of chronic lymphocytic leukemia (CLL) has been shown to exhibit varying clinical outcomes based on reported laboratory parameters. One of these parameters involves the measurement of the protein levels of zeta-associated protein (ZAP-70) in CLL cells. A standardized assay has not yet reached consensus in the clinical cytometry community.Methods: We developed a system using the 8-peak Rainbow beads as our fluorescence calibrator along with a fixed cell control. Using a panel of CD19-PE, CD5-FITC, and ZAP-70-Alexa 647, we stained normal whole blood, and blood and bone marrow from patients with CLL to determine the level of ZAP-70 expression in T-cell, B-cell, and CLL-cell populations. ZAP-70 expression was reported in molecules equivalent fluorescence (MEFL) based on the calibration of the flow cytometer with the 8-peak Rainbow beads.Results: Daily assay performance as well as operating MEFL defined ranges for ZAP-70 detection in CLL were developed. A rank-order, nonparametric approach to reference ranges was used to assign a cutoff for ''negative'' as well as ranges for ''intermediate'' and ''positive'' staining using T and B cells from a pool of 50 normal subjects, and CLL cells from 395 patients. The assay was validated in a multi-institutional study and has demonstrated correlation with published techniques. Since its initial development, the assay has been implemented at two additional laboratory sites and has been shown to produce reproducible, correlated data at all sites.Conclusions: Strict adherence to standardization can yield an assay that is predictable, reliable, and reproducible as well as capable of multisite implementation. The Rainbow beads provide a common platform for system calibration. The fixed cell culture controls provide a common target to test antibody. The final level of control tests the sensitivity of ZAP-70 detection in a normal peripheral blood sample stained along with the submitted CLL patients. The acceptance/rejection of test results must meet all three levels of control before patient results are reported. q