BR density centrifugation enriches blasts without altering immunophenotype data of blasts.

(A-D) SSC versus FSC display of cell fractions of PB from an MDS patient. PB containing 1.5% blasts (A) was subjected to BR density centrifugation (B-C) or Ficoll-Paque density centrifugation (D). Blasts were markedly enriched in the BR interface (B) but not recovered in the BR precipitate (C). Blast enrichment by Ficoll-Paque density centrifugation was minimal (D). Blast percentages determined in Wright-Giemsa–stained cytospin preparations by examining 100 cells were 63%, 0%, and 5% for panels B, C, and D, respectively. Circles marked as Neutro, Lymph, and Blasts in panels A and B indicate neutrophil, lymphocyte, and blast populations, respectively. (E-F) CD45 versus SSC display of normal BM cells before (E) and after BR density centrifugation (F, the BR interface). Immunophenotyping showed that the predominant cells in R1 were immature myeloid cells (CD34 57%, CD13 49%, CD33 50%) with minor contamination by immature B cells (CD10 22%, CD19 20%), the cells in R2 were stage I immature B cells (CD34 57%, CD10 75%, CD19 87%, CD20 27%, CD13 and CD33 < 15%), and the cells in R3 were stage II immature B cells (CD34 8%, CD10 67%, CD19 71%, CD20 56%, CD13 and CD33 < 15%). (G-J) Samples from 2 de novo AML cases (cases I and II) were subjected to blast immunophenotyping before (G,I) and after BR density centrifugation (H,J). The circles on the CD45 versus SSC display indicate gates for blast immunophenotyping. Immunophenotype data remained unchanged after BR density centrifugation.