Analysis of variation in results of CD34+ hematopoietic progenitor cell enumeration in a multicenter study

A workshop was held in The Netherlands and Belgium with the aim of investigating whether or not the use of a standard protocol vs. local protocols for flow cytometric enumeration of CD34+ hematopoietic progenitor cells would reduce interlaboratory variation. The standard protocol consisted of a three-color, whole-blood staining technique based on fluorescein isothiocyanate (FITC)-labeled CD34, and phycoerythrin (PE)-labeled CD14 and CD66e monoclonal antibodies (reactive with monocytic and myeloid cells, respectively), followed by erythrocyte lysis, washing, fixation, and selection of nucleated cells during data acquisition on the basis of their positivity for LDS-751 (staining DNA and RNA). Data analysis guidelines included the elimination of nonspecific antibody binding by monocytes and myeloid cells by gating on the CD14-, 66e- cells, followed by setting a window on a CD34 vs. sideward light scatter (SSC) plot around the CD34+, SSClow cells. The FITC-labeled isotype control was analyzed with the same gate and window settings, and the false-positive events were subtracted from CD34 result. Four samples (i.e., peripheral blood and apheresis product from two patients) were sent out. Results were received on patient 1 (2) from 36 (38) laboratories. Data obtained by 24 (26) laboratories after correct application of the standard protocol revealed that the median percentage of CD34+ cells of the four samples ranged between 1.1% and 3.7% and the CVs between 18% and 30%. Incorrect performance of the standard protocol by 12 laboratories, mainly resulting from gating errors, yielded a larger variation (CVs ranging between 50% and 82%). CD34 enumeration using local protocols by 29 (34) laboratories yielded median percentage of CD34+ cells ranging between 1.2% and 3.9% and CVs ranging between 34% and 106%. We conclude that correct application of this standard protocol was effective in reducing the interlaboratory variation in percentage of CD34+ cells assessments.