Multiple-laboratory comparison of in vitro assays utilized to characterize hematopoietic cells in cord blood.

Background: Understanding the variability in results obtained by multiple laboratories is important because cord blood units are distributed worldwide for transplantation.
Study Design and Methods: Four exercises were conducted by multiple laboratories to assess assay variability on nucleated cell (NC), mononuclear cell (MNC) by hematology analyzers [HAs], and CD34+ cell (flow cytometry) measurements. Exercise 1 was an intralaboratory exercise in which the reproducibility of cell measurements was determined. Exercises 2 and 3 involved the shipment of identical processed cord blood samples. In Exercise 2, laboratory-specific methods were utilized. In Exercise 3, two commercial CD34+ cell methods (Stem-Kit and TruCOUNT) were used. In Exercise 4, CD34+ cell levels were determined on repetitive regating of identical list-mode files.
Results: Intralaboratory reproducibility was highest for NC measurements and lowest for CD34+ cell measurements. In Exercise 2, all laboratories except one utilized HA with an impedance technology and determined comparable results for NC and MNC levels, whereas the other laboratory utilized a HA with an optical counting method. Substantial variation was observed on measuring CD34+ cells with ranges of 32 to 141, 32 to 66, and 25 to 116 CD34+ cells per microL for the three identical samples. In Exercise 3, on the use of one specific commercial assay, the ranges of CD34+ levels were 214 to 411 and 62 to 178 cells per microL for the two identical samples. Nearly all participating laboratories determined comparable CD34+ levels on the use of identical list-mode files.
Conclusion: These studies indicate that substantial variability in CD34+ cell levels were determined with flow cytometry. The variability in NC and MNC levels was minimal with HA methodology.