Validation of Minimal Residual Disease Flow Cytometry Method for Residual Disease Monitoring in Chronic Lymphocytic Leukemia

Monitoring Minimal Residual Disease (MRD) of Chronic Lymphocytic Leukemia (CLL) patients who achieved complete remission has been difficult and challenging using flow cytometry detection methods in peripheral blood Earlier flow cytometry methods described for detection of MRD relied on the assessment of CD20 expression that maybe compromised during rituximab therapy (CD5/19/20/79B and light chain analysis) or were done using the classic CLL staining panel (CD5/19/23 and light chain analysis) that had a low sensitivity. To facilitate the development of methods that would be suitable for rituximab containing regimens and have better sensitivity, a panel of antibody combinations was identified for use in an internationally standardized flow cytometric approach for MRD detection in CLL-treated patients (Rawstron et al., 2007). This panel consists of 5 combinations of antibodies specific for 1-sIgKappa/sIgLambda/CD19/CD5, 2-CD45/CD14/CD19/CD5, 3-CD43/CD79b/CD19/CD5, 4-CD22/CD81/CD19/CD5 and 5-CD20/CD38/CD19/CD5 antigens. The first two combinations are utilized for confirmation of clonality and assessment of Tcell contamination rate within the B cell gate, whereas combinations 3, 4 and 5 are specific for MRD detection. This standardized 5-combination panel was technically validated by Genoptix Medical Laboratory for use in BiogenIdec’s clinical trials in CLL with Lumiliximab. To validate this panel assay, selected CLL samples were mixed with normal donor blood or “disease-free” bone marrow specimens, to achieve 1%, 0.1%, 0.05% and 0.01% of CLL cells in “non-CLL” leukocytes. Our results show that it is possible to identify up to 1 CLL cell in 10,000 normal cells in some but not all cases, and that on a consistent basis our analysis is able to identify 1 CLL cell in 1000 normal cells. In comparison, the classic screening panel commonly used to diagnose CLL has a limit of detection (LOD) of about 1% (1 leukemic cell in 100 normal cells); thus this new method provides a 10 to 100 fold improvement in sensitivity. The enhanced sensitivity and LOD of the new assay is mainly due to the reduction of the non-CLL cell background in peripheral blood and bone marrow Furthermore, to make analysis guidelines consistent among analysts, an internal semi-quantitative clustering scoring analysis system was developed for this assay. The cluster scoring scheme assigns a negative score to scattered events with no clearly visible clone; well grouped events with tight visible clones are assigned positive scores and those gated events are considered for MRD final call. To reliably define the presence of MRD, at least 2 of the 3 MRD specific antibody combinations must have a positive clustering score. Due to the small number of cells to be analyzed, collection of 500,000 events is recommended. Finally, the results of additional studies showed that the anticoagulant employed may impact the stability of the antigens over a period of 7 days, and for accurate MRD determination blood specimens should be drawn into Heparin vacutainer tubes when long shipping times are expected.