Flow Cytometric Detection of ZAP-70 in Chronic Lymphocytic Leukemia: Addition of a Novel Second Monoclonal Antibody Improves Correlation with IgV_H Mutation Status.

Background: Expression of ZAP-70 is associated with unmutated immunoglobulin heavy chain variable region genes and poor prognosis in B-cell CLL. Reliable clinical flow cytometric assays have been difficult to establish, as many anti-ZAP-70 antibodies show a suboptimal signal-to-noise ratio, and an imperfect correlation with IgV_H mutation status.

Methods: We studied fresh peripheral blood or bone marrow specimens from forty-three patients with CLL/SLL. Samples were stained for intracytoplasmic ZAP-70 using a standard monoclonal antibody, 1E7.2/PE (Caltag, Burlingame, CA), and the novel monoclonal antibody J13–1164/PE (kindly provided by BD Biosciences, San Diego, CA). Threshholds were set using isotype-matched negative control antibodies. Mutational status of the IgV_H genes was assessed by RT-PCR, followed by standard Sanger DNA sequencing. Divergence from germline IgV_H segments (IMGT database) was calculated using VBASE, with 2% or less changes over codons 1–94 of IgV_H regarded as unmutated.

Results: Staining with J13–1164 revealed a clear apparent cutpoint in levels of positivity, with 26/43 (60%) of the CLL cases showing >49% of cells above the isotype threshhold. Most of the remaining cases (15/43, 35%) showed <26% of cells positive. The 1E7.2/PE reagent yielded a similar number of positive cases, with 28/43 (65%) showing >20% positive cells (the cutpoint suggested by Rassenti et al., N Engl J Med 2004, 351:893). However, staining with 1E7.2 was more difficult to interpret, as many cases had levels of positivity that were close to the cutpoint, with 15/43 (35%) showing 10–30% of cells positive. Both antibodies showed similar degrees of correlation with IgV_H mutation status, with 72–75% of unmutated cases showing the expected positivity for ZAP-70, and 57–64% of mutated cases negative for ZAP-70 (see Table 1). The two antibodies yielded concordant results in about half of cases (21/43), and the concordant cases showed a strong correlation with IgV_H mutation status: 15/16 cases positive with both antibodies had unmutated genes, and 4/5 cases negative with both antibodies had mutated genes (see Table 2). The cases with discordant staining between the two antibodies were a mixture of cases with mutated or unmutated IgV_H genes.

Conclusions: The J13–1164 anti-ZAP-70 antibody provides more easily interpretable results in routine clinical use than the 1E7.2 clone, with brighter staining in many cases and a more distinct difference between positive and negative cases in our series. Compared to the 1E7.2 clone, the J13–1164 clone shows a similar degree of correlation with the IgV_H mutation status. A combination of both antibodies provides the strongest correlation with IgV_H mutation status, with 19/21 concordant cases showing the expected staining pattern.