Assay method, cord blood bank, nucleated cell count, and neonatal factors influence the CD34+ cell count in cord blood: The coblt experience

CD34+ cells in umbilical cord blood are important in determining the hematopoietic stem cell dose and its potential influence on neutrophil recovery following cord blood transplantation. In the NHLBI sponsored Cord Blood Transplantation (COBLT) Study, standardized flow cytometry procedures to assess CD34+ cell counts were developed by three COBLT cord blood banks (CBBs). Two CBBs (Banks A and B) used a single platform method (standard ProCOUNT analysis) and one CBB (Bank C) used ISHAGE gating strategies. CD34+ cell counts from 4962 banked CBUs (ProCOUNT 81.9%, ISHAGE 18.1%) were reviewed. The mean total CD34+ cell count was 2.6 x 10' (SD=2.6, median=2.0), the mean total viable nucleated cell count (TNC) was 9.0 x 10s (SD=3.9, median=8.1 ), and the mean total CFU-GM count was 0.9 x 10'(SD=0.9, median=0.7). In univariate analyses, TNC, time from collection to initiation of processing, birth weight, gestational age, gender, race, collection volume and type of delivery were significantly associated with total CD34+ cell count (p< 0.05). In a multivariate analysis, TNC, birth weight, gestational age, gender, and race remained significant (R2=.22). The model shows that greater CD34+ cell counts are associated with greater TNC, greater birth weight, earlier gestational age, males and Caucasians. In addition, when bank and a bank by race interaction were included in the model, bank was significantly associated with total CD34+ cell count (rx 0.001) despite significant efforts at standardization. The average CD34+ cell count for CBUs from Caucasian donors at Banks B and C were 1.0 x 10'cells and 1.9 x 106 cells greater (respectively) than for similar donors at Bank A. A significant bank effect was also found when comparing only the two banks which are using ProCOUNT (p< 0.001). Although several factors were found to be significantly associated with total CD34+ cell counts, unexplained variability i n total CD34+ cell counts remains. This likely reflects the substantial natural variation of total CD34+ cells in cord blood and the variation in assay technique. In the controlled setting of the COBLT study, the finding that bank still influences total CD34+ cell count demonstrates a continued need to develop and validate CD34+ enumeration among CBBs.

Duke Scholars

Author Joanne Kurtzberg Pediatrics, Transplant and Cellular Therapy