Flow Cytometric Characterization of Genetically Defined Subgroups of Chronic Lymphocytic Leukemia: Basis for the Comprehensive Definition of Prognostic Parameters.

Chronic lymphocytic leukemia (CLL) is defined based on cytomorphology and immunophenotype. The most important prognostic parameters are cytogenetic aberrations and the expression of CD38 and ZAP70. We analyzed correlations between immunophenotypic patterns and genetic aberrations in 153 peripheral blood samples from patients with newly diagnosed and untreated CLL. Immunophenotyping was performed applying five-fold staining, a comprehensive panel of antibodies, and CD45-SSC-gating. Genetic aberrations were detected by fluorescence in-situ hybridization using probes targeting ATM for the detection of 11q-, D12Z3 for trisomy 12, D13S319 and D13S25 for 13q-, and p53 for 17p-. As compared to cases without trisomy 12, in cases with trisomy 12 the expression levels were higher for CD22 (66% vs. 34%, p=0.001), CD20 (83% vs. 71%, p=0.034), and FMC7 (28% vs. 7%, p=0.000001). This is in accordance with our finding that higher numbers of prolymphocytes have been found in CLL with trisomy 12. As compared to cases with any cytogenetic aberration, cases with normal cytogenetics had lower expression levels of CD5 (82% vs. 89%, p=0.012) and CD19 (79% vs. 85%, p=0.051). Importantly, there were no significant differences in expression levels of CD38 and ZAP70 between cytogenetically defined subgroups of CLL. Thus, in cases with 11q-, trisomy 12, 13q-, homocygous 13q-, 17p-, and normal cytogenetics CD38 was expressed in 51%, 47%, 37%, 33%, 24%, and 38% and ZAP70 together with cyCD79a was expressed in 17%, 11%, 13%, 13%, 22%, and 15%. These data indicates that, with the exception of trisomy 12, the cytogenetic findings are not reflected by specific immunophenotypic features. In particular, this is true for the prognostically and clinically relevant markers CD38 and ZAP70 the expression of which cannot be deduced from cytogenetics. In a second step, correlations between the expression of CD38 and ZAP70 and other immunophenotypic markers have been analyzed. Cases positive for CD38 (more than 30% positive cells) had lower expression levels of CD19 (78% vs. 87%, p=0.005), CD20 (65% vs. 75%, p=0.023), and CD23 (62% vs. 74%, p=0.005). Cases positive for ZAP70 (more than 20% positive cells with coexpression of cyCD79a) had higher expression levels of CD19 (87% vs. 79%, p=0.011), CD22 (40% vs. 26%, p=0.020), and cyCD79a (75% vs. 59%, p=0.009). These findings are in line with both CD38 and ZAP70 being markers that reflect different biologic backgrounds of CLL which also reveal prognostic information. Taken together, these data strongly suggest that studies evaluating different treatment approaches in CLL should provide a detailed characterization of the analyzed patients including both cytogenetic aberrations and immunophenotypic findings as well as novel genetic markers in order to guarantee the detection of subgroup-specific treatment effects.