Little is known about the function and phenotype of leukemic stem cells (LSCs) in chronic myeloid leukemia (CML) or about specific markers that discriminate LSCs from normal hematopoietic stem cells (HSCs). CD26 has recently been described as a specific marker of CML LSCs. In the current study, we investigated this marker in a cohort of 31 unselected CML patients. BCR/ABL1 positivity was analyzed in highly enriched stem cell fractions using fluorescence in situ hybridization (FISH) and reverse transcription PCR (RT-PCR). The proportion of CD26+ LSCs and CD26− HSCs varied considerably among the patients analyzed, and the percentage of CD26+ cells correlated with leukocyte count. The CD26 expression robustly discriminated LSCs from HSCs. This required a strict gating of the stem cell compartment. Thus, in patients with very low LSC or HSC numbers, only the highly sensitive RT-PCR method discriminated between clonal and non-clonal cells, while a robust FISH analysis required larger numbers of cells in both compartments. Finally, our data show that the numbers of CD26+ CML LSCs correlate with responses to treatment with BCR-ABL1 inhibitors.

No abstract

BackgroundAnti-CD19 chimeric antigen receptor T cells (CART-19) frequently induce remissions in hemato-oncological patients with recurred and/or refractory B-cell tumors. However, malignant cells sometimes escape the immunotherapeutic targeting by CD19 gene mutations, alternative splicing or lineage switch, commonly causing lack of CD19 expression on the surface of neoplastic cells. We assumed that, in addition to the known mechanisms, other means could act on CD19 to drive antigen-negative relapse.MethodsHerein, we studied the mechanism of antigen loss in an in vivo CD19-negative recurrence model of chronic lymphocytic leukemia (CLL) to CART-19, established using NOD-scid IL2Rgnull mice and HG3 cell line. We validated our findings in vitro in immortalized B-cell lines and primary CLL cells.ResultsIn our in vivo CLL recurrence model, up to 70% of CART-19-treated mice eventually recurred with CD19-negative disease weeks after initial positive response. We found that the lack of CD19 expression was caused by promoter DNA hypermethylation. Importantly, the expression loss was partially reversible by treatment with a demethylating agent. Moreover, this escape mechanism was common for 3 B-cell immortalized lines as well as primary CLL cells, as assessed by in vitro coculture experiments.ConclusionsEpigenetically driven antigen escape could represent a novel, yet at least partially reversible, means of CD19 loss to CART-19 in B-cell tumors.