Background: This study assesses the feasibility of a new volumetric cytometry system for the enumeration of CD34+ cells in apheresis components, peripheral blood, and cord blood samples in routine laboratory work. This system is compared with the following flow cytometry protocols: Milan, ISHAGE, ISHAGE with 7-AAD, and flow-count fluorospheres.
Study design and methods: Correlation, linearity, and reproducibility studies were performed for the various methods. Clonogenic cultures were performed, as an external control, to assess the correlation between the number of CD34+ cells per microL and the number of colony-forming units per microL.
Results: The linear regression analysis demonstrated that the five methods were comparable (R2 ranged from 0.86 to 0.96 and slopes were close to 1). The CD34+ assay and the flow-count methods showed poor linearity for CD34+ cell counts below 10 cells per microL (R2 = 0.46 and 0.47). The reproducibility assay for a CD34+ count of 10 cells per microL showed a CV of 12 percent and 25 percent for the Milan and CD34+ assay methods, respectively. The mean CV among all five methods for the 46 evaluated samples was 20 percent. There was a strong correlation between the number of CD34+ cells per microL and colony-forming units per microL in cord blood and apheresis samples (r = 0.71-0.81).
Conclusion: The CD34+ assay is useful in CD34 enumeration in cord blood, leukapheresis samples, and peripheral blood samples and provides comparable results to the Milan, ISHAGE, ISHAGE with 7-AAD, and flow-count methods. Nevertheless, peripheral blood samples with low CD34 absolute counts (below 10 cells/microL) should be analyzed by alternative flow cytometry protocols. Even though the same operator performed the study in a single laboratory, the high inter-method CV suggests that differences in sample preparation and gating strategy are factors that increase variability. Protocols with fewer intermediate steps or fully automated protocols such as the CD34+ assay are expected to reduced intra- and inter-laboratory variability.