Detection of a CD133⁺/CD38⁻ SL-IC Population in Childhood B Cell Precursor ALL.

Clonality studies of immunoglobulin rearrangements in B cell precursor acute lymphoblastic leukaemia (BCP ALL) has suggested that the disease may arise in cells already committed to the B cell lineage. In contrast, Ph⁺ ALL, which has a less favourable prognosis, is thought to arise in a more primitive haemopoietic cell. This was confirmed recently by studies that demonstrated only the CD34⁺⁺/CD38⁻ subfraction from Ph⁺ cases could engraft NOD/SCID mice. However, more recently there has been an increasing body of evidence to suggest that pre B and common ALL may also arise in a cell with a primitive phenotype. We have previously demonstrated that in childhood BCP ALL, the cells capable of long term proliferation in vitro in suspension culture and in vivo to engraft NOD/SCID mice are CD34⁺/CD10⁻, CD19⁻. We then attempted to further define these ALL progenitor cells by investigating the expression of CD133, the primitive stem cell marker. ALL cells capable of long term proliferation in vitro and NOD/SCID engrafting capacity were derived from the CD133⁺/CD19⁻ subfraction only. These cells were capable of secondary NOD/SCID repopulation, demonstrating they had self-renewal ability. Here, we have attempted to further characterise these ALL progenitor cells to address the question as to whether BCP ALL arises in a common lymphoid progenitor cell or in a more primitive haemopoietic cell. ALL cells from five patients were sorted for CD133⁺/CD38⁺ and CD133⁺/CD38⁻ populations, the sorted subfractions were analysed by cytogenetics and their functional ability was assessed in the NOD/SCID mouse model. Cytogenetic analyses by FISH revealed that both CD133⁺/CD38⁺ and CD133⁺/CD38⁻ subfractions contained the BCR/ABL and ETV6/RUNX1 gene fusions, which had been detected in the patients at diagnosis, and in 1 case with del 17p, this deletion was also noted in the sorted subfractions. These sorted ALL subfractions and unsorted cells were injected intravenously into sublethally irradiated NOD/SCID mice. Bone marrow was harvested at 8–10 weeks post inoculation and analysed for the presence of human cells by flow cytometry. Engraftment was achieved in every case using 2.5x10⁶–10⁷ unsorted cells (0.1–4.5% CD45⁺). There was no evidence of human cell engraftment in recipients of the CD133⁺/CD38⁺ subfraction. However, in each case, engraftment was observed with the CD133⁺/CD38⁻ subfraction, 0.6–3.2% CD45⁺ using as few as 6x10²–4x10⁴ cells. Using this sorting strategy, we were able to enrich NOD/SCID leukaemia engrafting cells by at least 4 logs compared to the bulk ALL population. Cytogenetic analyses demonstrated that the engrafted cells had the same karyotype as the patients at diagnosis, confirming engraftment of leukaemic cells. These findings suggest that the leukaemia has arisen in a cell with a primitive phenotype, similar to that described for normal haemopoietic stem cells and adds further support to the evidence for a primitive cell origin for B cell precursor ALL. Studies are ongoing to determine whether these primitive ALL cells have the same IgH rearrangements that are detected in the bulk ALL population at diagnosis. This primitive ALL population may be resistant to current chemotherapeutic strategies that are targeted against generic properties of the malignant blasts and subsequent relapses may arise from these cells. Hence, identification and characterisation of these putative ALL stem cells is essential for the development of more effective therapeutic strategies.