Selective Persistence of Distinct Stem/B Leukaemic Stem Cells In Childhood Acute Lymphoblastic Leukaemia In Clinical Remission.

Abstract 1585

Recent studies utilising surrogate leukaemic stem cell (LSC) assays have suggested that LSCs in acute lymphoblastic leukaemias (ALLs) might be neither rare, nor phenotypically or functionally distinct. However, studies of candidate LSCs in surrogate assays might not recapitulate the full leukaemic potential of candidate LSCs in patients, and in particular their responsiveness and resistance to therapeutic targeting. Therefore, we have investigated the identity, molecular and functional properties, and persistence of different subsets of candidate LSCs in childhood ALL, at diagnosis and during the course of clinical and molecular remissions in response to chemotherapy, and their relationship to subsequent relapses. First, we investigated 6 patients diagnosed with “good prognosis” TEL-AML1+ ALL, and at diagnosis we found TEL-AML1+ leukaemic cells within the immature B cell progenitor compartment (proB: 34+38+19+), mature B-cells (34-19+), as well as in a population expressing an aberrant combination of stem cell (34+38-/lo) and B-cell (19+) cell surface markers. These stem/B (34+38-/lo19+) cells were all TEL-AML1+ and not present in age-matched normal bone marrow controls. In contrast, haematopoietic stem cells (HSC: 34+38-19-) were not part of the TEL-AML1+ leukaemic clone in any of the patients. 15 days into chemotherapy, all TEL-AML1+ mature B-cells were eliminated in all patients, and this was followed by a clearance of leukaemic proB cells by day 28 of treatment. In striking contrast, leukaemic stem/B cells were still detectable at day 28, but in all TEL-AML1 patients, at later stages all leukaemic cells including the stem/B cells were undetectable, and at the same time these patients went into complete remission with less than 1 leukaemic cell in 10e4 cells detectable. A similar pattern was observed in a case of “high risk” BCR-ABL+ ALL: BCR-ABL+ proB and B-cells were efficiently eliminated by day 90 of the course of chemotherapy, and up to 180 days into the treatment only 34+38-/lo19+ stem/B cells remained part of the BCR-ABL+ clone. In agreement with the persistence of BCR-ABL+ 34+38-/lo19+ stem/B cells, this patient relapsed 17 months after the initiation of chemotherapy. In order to understand the underlying mechanisms of the observed functional and therapeutic heterogeneity seen in leukaemic subpopulations, we performed comparative gene-expression analysis of diagnostic leukaemic stem/B and proB cells of TEL-AML1+ patients. This analysis revealed a differential gene expression pattern between leukaemic stem/B and proB cells, with positive regulators of cell cycle being the most distinctly up regulated genes in leukaemic proB cells. In agreement with this, cell cycle analysis of 3 diagnostic TEL-AML1+ cases also showed proB cells to be more actively cycling compared to the more quiescent state of the leukaemic stem/B compartment (proB: G0 42%; G1 40%; S,G2,M 18% vs. stem/B: G0 81%; G1 18%; S,G2,M 1%), providing a potential mechanistic basis for the relative therapy resistance of ALL stem/B cells.

Taken together the present studies suggest that quiescent 34+38-/lo19+ stem/B cells are selectively resistant to chemotherapy, and most likely the origin of relapses when these occur in childhood ALL.