OPTIMAL PROCESSING OF HUMAN UMBILICAL-CORD BLOOD FOR CLINICAL BANKING

Human umbilical cord blood (UCB) has been successfully used as an alte rnative source of allogeneic hematopoietic stem cells for pediatric tr ansplantation, Clinical banking of UCB requires volume reduction and r ed cell depletion for cost-effective storage. We have compared process ing UCB by Ficoll, Percoll, methylcellulose, gelatin, starch, and red cell lysis. As individual UCB collections vary widely in colony formin g cell (CFC) and CD34(+) cell content, each UCB (n=26) was processed b y three or more techniques in parallel with Ficoll as the ''standard'' method. Gelatin gave a consistently high recovery of CFC (92%) and CD 34(+) cells (86%). Between 0.10-2.50% of the leukocytes in gelatin-tre ated UCB were Cn34(+) with an intra-assay variation of 2.1%. Combining data from individual experiments, tile correlation bi tween CD34(+) a nd CFC content was excellent (r=0.77). Lysis rated second in terms of CD34(+) and CFC recoveries but is not as practical because of the larg e volumes involved. Ficoll and Percoll came third but are more expensi ve and more involved techniques. Starch sedimentation proved to be slo w, while methylcellulose processing lost over 60% of CFC and CD34(+) c ells. After gelatin processing, we calculated 70-mL donations of UCB w ould contain a mean +/- SD of 9 +/- 2 x 10(8) nucleated cells, 32 +/- 18 x 10(5) CD34(+) cells, and 20 +/- 12 x 10(5) CFC with greater than 95% red cell depletion. Recent published computer studies suggest that as few as 2 x 10(5) CD34(+) cells may be needed for sustained engraft ment of allogeneic marrow in adult transplant recipients. We conclude that a average 70-mL UCB donations contain sufficient marrow repopulat ing cells for adult recipients,