ZAP-70 expression is a stage independent prognostic marker in CLL. However, inter-laboratory variation is large and there is neither a consensus nor regulatory approved methodology.
Two anti-ZAP70 clones (1E7.2 and SBZAP) were compared in 45 untreated CLL patients. Nine different methods for ZAP-70 expression analysis were evaluated: M1, isotype control to determine negative; M2, internal residual T-cell to determine positive; M3, normal donor (ND) T-cell to determine positive; M4, internal T-cell/clone ratio; M5, ND residual T-cell/clone ratio; M6, clone/normal remaining B-cell ratio; M7, clone/ND B- cell ratio; M8, CLL- Z score; M9, modified CLL-Z score. A scoring system was designed integrating both 1E7.2 and SBZAP clones to assign ZAP-70 expression.
The correlation coefficients for the four selected highest statistically significant methods were as follows (M1=0.71, M3=0.72, M7= 0.67, and M9=0.64). These four methods were used to generate a combined score. The two reagents showed agreement using the designed scoring system for 37/45 samples (82%) and 8/45 (18%) showed equivocal result with one of the two clones. Seven of the eight equivocal samples were resolved using the scoring system.
Four of the nine methods of analysis were compared for each reagent. The use of two independent ZAP-70 reagents increases analytical certitude and the scoring method aids in the resolution of equivocal results. The combined use of two reagents, four methods of analysis and a scoring method allowed for assignment of ZAP-70 expression in 44/45 samples (98%) tested and improved performance of this important prognostic assay.