Comparison of single and dual-platform assay formats for CD34(+) haematopoietic progenitor cell enumeration

Most techniques for CD34(+) cell enumeration are dual platform assays. That is, they derive absolute numbers of CD34(+) cells from either the flow cyt ometrically assessed per cent (%) CD34(+) cells within the nucleated cells and/or the white blood cell count from a haematology cell analyser. Recentl y, so-called single-platform assays have been developed, in which the absol ute number of CD34(+) cells is directly derived from a single flow cytometr ic measurement. The present study aims to compare the variation between eig ht laboratories in CD34(+) cell counts from paired assays of 15 samples usi ng a common single (ProCOUNT(TM)) and the local dual-platform method. Six l aboratories used the 'SIHON' and two the 'ISHAGE' protocol for CD34(+) cell enumeration.Use of the single-platform method reduced the inter-laboratory variation in per cent and absolute numbers of CD34(+) cells, as measured b y interquartile ranges, by half but did not lead to an appreciable reductio n of the inter-laboratory variation in white blood cell counts. Thus, part of the reduced inter-laboratory variation obtained with ProCOUNT(TM) may ha ve been a result of the use of standardized procedures and reagents to dete ct CD34(+) cells. In order to eliminate any variation arising from the use of different local protocols for percentage of CD34(+) cell assessments, a comparison was made of the ProCOUNT(TM)-derived absolute CD34(+) cell numbe rs (i.e. single platform) with the dual-platform absolute CD34(+) cell numb ers calculated by multiplying ProCOUNT(TM)-derived percentage of CD34(+) ce lls and with the corresponding haematology analyser-derived white blood cel l count. Regardless, the interquartile ranges of absolute CD34(+) cell numb ers remained almost a factor of two smaller with the use of the single plat form method. Thus, these results suggest that single-platform methodology c an reduce the variation in absolute CD34(+) cell numbers between laboratori es.