Factors associated with outcomes of unrelated cord blood transplant: guidelines for donor choice.

ABSTRACT

OBJECTIVE: Optimizing cord blood donor selection based mainly on cell dose and human leukocyte antigen (HLA) disparities may further improve results of unrelated cord blood transplants (UCBT). MATERIALS AND RESULTS: We analyzed 550 UCBTs for hematologic malignancies reported to the Eurocord Registry. Main outcomes and prognostic factors were analyzed in univariable and multivariable analyses incorporating center and period effects and using death and relapse as competitive risks for nonfatal endpoints. Nucleated cell (NC) dose before freezing and number of HLA disparities had a significant influence on outcome. Cumulative incidence (CI) of neutrophil and platelet recovery was associated with the number of HLA mismatches, number of NC before freezing, and use of granulocyte colony-stimulating factor. Coexistence of HLA class I and II disparities and high CD34 cell dose in the graft were associated with graft-vs-host disease grades III-IV. CI of disease relapse was higher in matched transplants showing a graft-vs-leukemia effect increased in HLA-mismatched transplants. Overall 3-year survival was 34.4%. Prognostic factors for survival were recipient age, gender, and disease status. CONCLUSION: Our results provide indications for a better choice of cord blood units according to cord blood cell content and HLA.