Abstract
Zap-70 expression by flow cytometry has been shown to be a surrogate marker for un-mutated/germline IgH status in chronic lymphocytic leukemia (CLL).1 Reproducibility and confidence in the accuracy of results can make this assay problematic. We compared our current single-parameter/one-color, immunohistochemistry (IHC) Zap-70 assay with our flow cytometry Zap-70 assay. Twenty-one (21) patient samples submitted to our lab for Zap-70 testing were assayed by IHC and flow cytometry (FC). In addition, two FC analysis methods were used. One analysis method uses isotype controls to establish negative regions and a 20% cut-off for positivity1, with BD CellQuestTM software (CQ). The other analysis method uses internal negative and positive mean channel fluorescence (MCF) differences using BD Paint-A-GateTM software (PAG). The IHC assay resulted in nine positive, six negative and six equivocal results for Zap-70. The equivocal results were due to very low proportions of CLL cells (three cases) or other problems identified on review of stained slides (three cases). The FC assay resulted in six positive, thirteen negative and two equivocal results for both analysis methods. When comparing IHC Zap-70 and FC Zap-70, there was 29% (6/21) discordance using the CQ analysis and 24% discordance (5/21) using the PAG analysis (equivocal cases included in the denominator only). Comparison of the CQ and PAG analysis demonstrated 16% (3/19) discordance (equivocal cases excluded from the denominator). The FC Zap-70 PAG analysis method correlated somewhat better with the IHC Zap-70 results than did the FC Zap-70 CQ analysis method. This may be due to the variability introduced by user-defined/isotype cursor settings. We conclude that Zap-70 expression can be tested for using either IHC or FC, and that IHC testing may have higher sensitivity. Correlation between the results utilizing these different methods is not ideal, and improvement in both assays, likely including flow analysis methods, is needed. We plan additional, comparative evaluation of more cases.
CellQuest™ and Paint-A-Gate™ are trademarks of Becton, Dickinson and Company.
Comparative evaluation of gene-set analysis methods
