438 Yield Prediction of Nucleated Cells As Well As CD34+ Progenitors in Allogeneic Bone Marrow Harvests Depends on Differentially Important Biological Donor Variables. a Retrospective Single Center Analysis of 1275 Allogeneic Bone Marrow Collections

Track: Poster Abstracts
Saturday, February 14, 2015, 6:45 PM-7:45 PM
Grand Hall CD (Manchester Grand Hyatt)
Michael Punzel , MediaPark Klinik, Cellex Cologne, Cologne, Germany
Michael Kramer , Universitaetsklinikum Dresden, Dresden, Germany
Helmuth Schmidt , MediaPark Klinik, Cellex Cologne, Cologne, Germany
Rainer Ordemann , Universitaetsklinikum Dresden, Dresden, Germany
Karin Buhrmann , MediaPark Klinik, Cellex Cologne, Cologne, Germany
Annegret Quade , MVZ Labor Quade, Cologne, Germany
Carsten Bartling , MVZ Labor Quade, Cologne, Germany
Gerhard Ehninger , Universitaetsklinikum Dresden, Dresden, Germany
Presentation recording not available for download or distribution as requested by the presenting author.
Cell requests for allogeneic bone marrow grafts are defined by the amount of total nucleated cells (TNC) according to body weight of the recipient. However, outcome of bone marrow transplantation significantly depends on the amount of infused CD34+ stem and progenitor cells representing around 0.5 to 1.5% of TNC. So far, TNC as well as the CD34+ progenitor yield of marrow harvests cannot be predicted. We retrospectively analyzed donor variables, TNC and progenitor yields of 1275 marrow collections (860 male and 415 female) that have been performed in our collection center - all of those with same experienced staff of physicians. Maximum amount of bone marrow that was allowed to harvest was 20ml/kg donor weight but limited to a maximum of 1500 ml. The mean TNC count in the final product was analyzed with 16.6 TNC/nl (5.9-46.6) corresponding to 5.6 ± 4.3 TNC/kg body weight of the recipient. The mean CD34+cell concentration in the graft was measured with 111/µl (11-674) corresponding to 4.3 ± 3.9 CD34+cells/kg. We fitted linear regression model for TNC adjusted for harvest volume and recipient weight to identify influential donor variables. Model selection was performed by backward elimination (BE) with the Bayesian information criterion (BIC). Peripheral blood leucocytes of the donor (WBC) before harvest, constitutional donor variables (weight and height), age, platelet count (PLT), smoking status and mononuclear cell count (MNC) before harvest could be identified as predictive donor parameters for TNC-concentration in the bone marrow product – ordered in descending mode by Wald-test statistic.  Interestingly, the influence of donor parameters for CD34+ progenitor cell concentration in the marrow was different from TNC. Constitutional parameters of the donor were identified to be most important for CD34+ cell concentration in the marrow product followed by WBC, age and PLT, whereas smoking status and MNC did not influence CD34+ cell harvest. With sex specific centered parameters donor sex was only influential - for both TNC as well as CD34+ cells - through the constitutional parameters weight and height.  No significant interactions could be found between harvested volume and the identified donor parameters. Taking also the recipient weight into account we propose a model that allows identification of risk constellations for achieving the designated amount of TNC as well as CD34+ hematopoietic progenitor cells from bone marrow harvests. In addition, this prediction model will enable us to identify additional genetic determinants for collection efficiency in bone marrow.
Disclosures:
Nothing To Disclose