Selection of unrelated cord blood grafts

To the editor:

We are concerned about statements in the article by Barker et al,¹  who describe a factor of 0.75 × TNC (total nucleated cell) to “correct” the TNC content of RBC-replete cord blood units. No data are provided to justify such an opinion, and the authors cite only “personal communication with multiple US investigators, 2004-2009.”^(p2334) Implementation of this policy, if incorrect, would underestimate the progenitor cell content of RBC-replete units resulting in the inappropriate rejection of some such units by transplant physicians in favor of RBC-reduced units with an apparently higher TNC dose after application of the “correction factor” for RBC-replete units. This would jeopardize the success of hematopoietic cord blood transplants because it is well known that their success is highly dependent on cell dose.^2,3  The authors attempt to justify their opinion by pointing out that RBC-replete units have a 25% higher TNC than RBC-reduced units, and they imply that the number of hematopoietic progenitor cells is erroneously elevated by that percentage. It is true that the TNC of RBC-replete units is elevated, in part because of the lack of neutrophil depletion. However, the authors ignore the fact that essentially no progenitor cells are lost in plasma depletion-reduction processing (plasma depletion/reduction or PDR) because during processing only a portion of the plasma is removed. In contrast, all processing methods that produce RBC-reduced units lose significant numbers of progenitor cells.^4-7 

To study this issue, we divided 10 cord blood units in half and processed one-half by RBC reduction using the most commonly used hetastarch method and the other half by plasma reduction. As indicated in Table 1, the average TNC of plasma-reduced units after processing was 124% of that in RBC-reduced units (81.56 × 10⁷ vs 65.77 × 10⁷, respectively [P = .002; 2-sample t test]). Of major importance is that plasma reduction caused virtually no loss of CD34⁺ cells, whereas there was a 16% loss of CD34⁺ cells in the RBC-reduced units, so that the postprocessing CD34⁺ cell count in plasma-reduced units was 121% of that in RBC-reduced units (1.415 × 10⁶ vs 1.165 × 10⁶, respectively, [P = .003]). The recoveries of colony-forming units (CFU) were also higher for plasma-reduced products: 225% for CFU-GM (P = .05) and 186% for total CFU (P = .01). This indicates that the higher TNC value of RBC-replete units does not exaggerate the number of progenitor cells available and that a “correction factor” is inappropriate. The cell dose that is ultimately infused can be affected by washing cord blood units^1,6,7  and this may be of significance for the subset of transplant centers that wash plasma-reduced products but do not wash RBC-depleted products.

Because there are no data to support the use of the authors' proposed 0.75 “correction factor,” and our data indicate that the authors' assertion that such correction should be applied to the TNC of RBC-replete units is incorrect, we submit that the authors' important and potentially detrimental statement should not be accepted by the transplantation community at this time.