Primitive Quiescent CD34+ Cells in Chronic Myeloid Leukemia Are Targeted by In Vitro Expanded Allogeneic Natural Killer Cells, Which Are Functionally Enhanced by Bortezomib Treatment.

Primitive quiescent CD34+ cells in chronic myeloid leukemia (CML) are relatively resistant to the tyrosine kinase inhibitors imatinib and dasatinib, which may explain the persistence of detectable BCR-ABL transcripts following treatment with these agents. Conversely, allogeneic stem cell transplantation (SCT) can eradicate residual CML, suggesting that quiescent stem cells are eliminated by graft-versus-leukemia (GVL) effects. We studied the progeny of CD34+ cells after 4 days culture in serum-free media supplemented with interleukin-3, interleukin-6, stem cell factor, granulocyte-colony stimulating factor and Flt-3 ligand in 14 CML patients (8 chronic phase, 6 advanced phase) who subsequently received T cell depleted SCT from their HLA-identical sibling donors. Cycling CD34-negative and CD34+, and non-cycling quiescent CD34+ CML cells were isolated by fluorescence activated cell sorting. Fluorescence in situ hybridization in 4 representative CML patients revealed over 80% BCR-ABL positivity in both quiescent and cycling CD34+ and CD34-negative populations. Using real-time quantitative polymerase chain reaction, we found the expression of BCR-ABL, and leukemia-associated antigens (LAA), WT1, PR3 and ELA2, were the same in both cycling and quiescent CD34+ cell populations in CML. LAA expression was not significantly different when compared with similarly cultured CD34+ cells from healthy donors. Pre-SCT quiescent CD34+ cells from CML patients were lysed by natural killer (NK) cells from their donors but were less susceptible than their cycling CD34+ and CD34-negative counterparts. Purified donor NK cells (n=7) expanded after 11–13 days culture with interleukin-2 and irradiated EBV-LCL lysed quiescent CD34+ CML cells as well as their cycling CD34+ and CD34-negative progeny. Previous studies have demonstrated that bortezomib can sensitize malignant cells to NK-cell tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) (Lundqvist et al Cancer Res. 2006 Jul 15;66(14):7317–25). Addition of bortezomib 10nM to CD34+ cell cultures enhanced cytotoxic effects of expanded donor NK cells on quiescent CD34+ CML cells. As observed with other malignancies, this enhanced sensitivity to NK-cytotoxicity correlated with increased expression of TRAIL receptors DR4 and DR5 on the surface of CD34+ quiescent cells, compared with cycling CD34+ or CD34-negative cells. Bortezomib treatment did not significantly affect the expression of MHC Class I, MIC A/B or Fas (CD95) on CD34+ quiescent or cycling cells. These results suggest that adoptive transfer of in vitro expanded donor NK cells with concomitant administration of bortezomib to the recipient may enhance cytotoxicity to quiescent CD34+ cells and may improve NK-mediated GVL effects. This may be particularly applicable to CML patients who are increasingly transplanted in more advanced stage disease, and so are at a greater risk of relapse post-SCT.