Use of 5-Fold Staining for Multiparameter Flow Cytometry-Based Quantification of Minimal Residual Disease in Acute Myeloid Leukemia Results in Improved Sensitivity.

The quantification of minimal residual disease (MRD) by multiparameter flow cytometry (MFC) using triple staining has been shown to yield prognostic information independent of other parameters in patients with acute myeloid leukemia (AML). Due to the immunophenotypic heterogeneity of AML the application of 5-fold staining may result in a better characterization of the leukemia-associated aberrant immunophenotype (LAIP) and thus in an improved sensitivity of the method as compared to triple staining. We analyzed bone marrow samples from 114 patients with newly diagnosed and untreated AML by MFC using a comprehensive antibody panel with 5-fold combinations. Sensitivity was estimated by quantification of LAIP-positive cells for each LAIP in 18 normal bone marrow samples. In each patient at least one LAIP was identified (total, 203 LAIPs). The LAIPs were present on a median of 15.88% of the bone marrow cells at diagnosis (range, 2.11% to 79.64%). The median number of normal bone marrow cells displaying the LAIPs ranged from 0.001% to 0.065% (median, 0.010%). As a result, the logarithmic difference (LD) in LAIP-positive cells between leukemic and normal bone marrow amounted to a median of 3.33 (range, 1.96 to 4.88). Similarly, if only the most sensitive LAIP was considered for each patient the median frequencies of LAIP-positive cells were 14.07% (range, 2.11% to 77.57%) in leukemic bone marrow and 0.010% (range, 0.001% to 0.065%) in normal bone marrow. Importantly, however, in this setting the resulting LD amounted to a median of 3.45 (range, 1.96 to 4.88). In order to estimate the impact of applying 5-fold staining on the sensitivity the information of each of the applied colors was skipped once while the results of the other four colors, respectively, were used. Skipping one color resulted in an increase of LAIP-positive normal bone marrow cells (median, 0.050%; range, 0.001% to 3.6%) while the percentages of LAIP-positive leukemic cells changed only marginally (median, 22.65%; range, 2.25% to 90.06%). The gain in LD by applying 5-fold staining in comparison to 4-fold staining amounted to a median of 0.58 (maximum gain, 3.14). In 32 patients a total of 120 follow-up samples have been analyzed appyling the combination of antibodies that allowed the best LAIP definition. The LD from diagnosis to follow-up amounted to a median of 2.82 (range, 0.77 to 4.82). Clinical follow-up data is available in 26 of these 32 patients. MRD assessment after completion of consolidation therapy has been performed in 15 patients. The median LD between diagnosis and follow-up assessment is 2.84 (range, 1.07 to 4.33). Separating patients according to this median LD identified a group of patients with no relapses yet (LD >2.84) while patients with an LD <2.84 had an event-free survival of only 50% at one year (p=0.075). These data confirm that flow cytometrically-based assessment of MRD is feasible in AML and results in prognostic information. It is suggested that the application of 5-fold staining significantly improves the sensitivity and thereby the overall accuracy of the method.