Absolute leukocyte count as an internal quality control for the CD34 cell enumeration assay

Abstract

In hematopoietic stem cell transplantation, the peripheral blood CD34+ cell count determines when to start the stem cell collection, and the CD34+ cells/ kg calculated in the apheresis products establishes when it has reached the desirable number of cells that guarantees engraftment. Due to these clinical implications, the International Society for CellularTherapy (ISCT) has described several recommendations on how to perform the flow cytometry (FC) assay for the enumeration of CD34+ cells [1]. Key issues in their latest update [2] are a single-platform assay with a dual combination of anti-CD45/anti-CD34 conjugates, a cell viability dye and fluorescent counting beads.The participation in international quality assessment programs contributes to standardize the procedure and improves the experience of the operators, and the use of controls with known CD34+ values helps to assess the quality of the reagents and the FC acquisition procedure. However, none of them can guarantee that the number of CD34+ cells obtained in every single sample is an accurate result. In this study, we explored whether the absolute white blood cell (WBC) count could be used as an internal quality control for monitoring the whole CD34+ cell enumeration assay.TheWBC count was calculated with FC and with a hematologic analyser (Beckman Coulter LH750), and Pearson’s correlation coefficient was used for comparison of the data. Between July 2013 and April 2015, 100 samples (62 mobilized ethylenediaminetetraacetic acid [EDTA] anti-coagulated blood samples and 38 leukapheresis products) were processed for enumeration of CD34+ cells within 12 h after extraction. Samples belonged to patients diagnosed with malignant lymphoma (n = 14), multiple myeloma (n = 10), Hodgkin disease (n = 2) and acute myeloid leukemia (n = 2). During the period of the study, 3 different lots of the BD Biosciences stem cell enumeration kit were used for both CD34+ cell enumeration and WBC count. Following the recommendations of the manufacturer, samples containing >40000 WBCs/μL were diluted. In all samples, a minimum number of 100,000 leukocytes, 1000 beads and 100 CD34+ events were acquired on FACSCanto II (BD Biosciences) flow cytometer. For the analysis, the Infinicyt software program (Cytognos) was used, and WBC count was calculated on the same file of data recorded for the CD34+ cell enumeration. Selection of WBC was done using a modified version of the ISCT sequential gating strategy for CD34+ cells (Supplementary Figure S1). Briefly, leukocytes were defined as events expressing CD45 and lacking the viability dye 7-Amino Actinomycin D (7-AAD). Low events and doublets were excluded using the forward (FSC)/sideward (SSC) light scatter dot-plot and the FSC area/FSC height dotplot, respectively. Finally, beads were selected in the FL1/FL2 dot-plot for making the calculation of the WBC count. According to the automatic analyser, the median WBC count was 29890/μL (range, 1100– 104,200) in peripheral blood samples, and 173,850/ μL (range, 19200–357,400) in the leukapheresis products. A good correlation between theWBC count calculated with FC and the hematologic analyser was found for the full cohort (r = 0.99; Figure 1A), and when peripheral blood (r = 0.99) and leukapheresis products (r = 0.98) were analysed separately. No