Objectives: The validity of endothelial progenitor cells as biomarkers and their therapeutic potential depend on the accuracy of techniques used for enumeration. This study assessed the agreement between 6 flow cytometric methods and a CFU assay used for EPC quantification.
Methods: Two blood samples were obtained from 30 healthy volunteers (60 samples). CD34+/VEGFR2+ cells were analyzed with flow cytometry, starting from whole blood (A-C) or PBMC (D-F), using different gating strategies: A: lymphocyte gating; B and D: exclusion of autofluorescent cells (CD3 negative selection); C and E: exclusion of autofluorescence and cell aggregates (pulse shape analysis by FSCarea/FSCpeak); F: exclusion of autofluorescence, cell aggregates and non-nucleated cells (Draq 5). PBMC were cultured under endothelial cell conditions to assess CFU numbers.
Results: Moderate agreement was found between methods B-C and D-E (ICC 0.647 and 0.530). Comparison of methods B-D and C-E showed poor agreement (ICC 0.178 and 0.249). This was also the case for techniques that considerably differed with regard to gating strategies (A-B, A-F, B-F). CFU numbers did not correlate with flow cytometric quantification (all p>0.05).
Conclusions: Agreement between methods for EPC quantification is moderate to poor, which may explain apparent controversies in literature. Although each protocol is highly reproducible, this study cautions against comparing study results gathered with different enumeration techniques.