Quantification of ZAP-70 expression in chronic lymphocytic leukemia: T/B-cell ratio of mean fluorescence intensity provides stronger prognostic value than percentage of positive cells

Abstract

Expression of ZAP-70 measured by flow cytometry belongs to the most powerful prognostic parameters in chronic lymphocytic leukemia (CLL). However, many technical factors such as setting of the positivity threshold may significantly influence results.. Quantification using mean fluorescent intensity (MFI) may eliminate the subjective error which is inevitable in the isotype control method. The aim of the present project was therefore to assess the prognostic significance of ZAP-70 using three different methods. Evaluation was performed by 1) percentage of positive cells compared to isotype control (cut-off 20%), 2) MFI ratio of Tcells/CLL cells (cut-off 3.0); 3) MFI ratio of ZAP-70/isotype control on CLL cells (cut-off 2.5). MFI method with T-cells/CLL cells ratio was the best in the identification of patients with unfavourable outcome: ZAP-70 positive patients had significantly shorter time to treatment (TTT, median 24 vs. 55 months, p=0.0001) and overall survival (OS, median 97 vs 174 months, p=0.0074). The differences in TTT a OS were not significant with the use of isotype percentage and MFI isotype methods. Combined analysis of ZAP-70 with CD38 expression or IgVH mutation status lead to identification of a subgroup with the longest TTT and OS (ZAP-70 and CD38 negative, p<0.0001 and p=0.012; ZAP-70 negative and mutated IgVH genes, p<0.0001 and p=0.0019).In conclusion, our results suggest that measurement of ZAP-70 expression in CLL by MFI using T-cells/CLL cells ratio might be the optimal method for accurate prediction of clinical course. Combined analysis of ZAP-70 with CD38 or IgVH mutation status further refined individual patient´s prognosis.

T/B ratio does not reflect levels of ZAP70 expression in clonal CLL B‐cells due to ZAP70 overexpression in patient T‐cells

“…1). Isoclonic ratios and T/B ratios were calculated as described (12,(15)(16)(17)(18)(19)29 and see Materials and Methods).…”

“…The main problem was to find the correct internal control. Among proposed methods, calculating the ZAP70 T/B ratio is one of the earliest and most common approaches (12,(15)(16)(17)(18). Other methods based on ratios calculated on normal B-cell MFIs have drawbacks such as the very low percentage of residual B-cells or the difficulty of separating them from CLL cells by flow cytometry in the case of weak CD5 expression (11,13,32).…”

“…Regarding the antibody and fluorochrome, various authors agree that the phycoerythrin-conjugated (PE-conjugated) SBZAP monoclonal antibody (mAb) is one of the best reagents for this method (11)(12)(13). Currently most groups express FCM results of ZAP70 expression in CLL B-cells as a ratio between mean fluorescence intensities (MFI) of ZAP70 in CLL B-cells and in normal residual or externally added B-cells (11,13,14) or, more frequently, as a ratio between ZAP70 MFIs in residual T-cells and CLL B-cells (ZAP70 T/ B ratio) (12,(15)(16)(17)(18)(19). The main disadvantage of the former method is that residual normal B-cells may be virtually absent in peripheral blood of patients with CLL and external addition of normal B-cells to the blood sample prior to staining supposes that ZAP70 levels in normal subjects are biologically stable.…”

Prognostic value of ZAP-70 expression in chronic lymphocytic leukemia as assessed by quantitative polymerase chain reaction and flow cytometry

“…). The resultant ZAP‐70 MFI values for T‐ and B‐cells were then expressed as a ratio of intensity of expression on B‐cells as compared to T‐cells , with a ratio of ≥0.2 used as a cut‐off for positive results. …”

European canine lymphoma network consensus recommendations for reporting flow cytometry in canine hematopoietic neoplasms

“…Some antigens (such as ZAP-70 protein in human CLL) are reported comparing MFI of neoplastic cells to those of T cells in the same sample. This method is reported as easy to perform and optimal for accurately predicting outcome in CLL (20).…”