Imatinib mesylate (IM) treatment results in marked reduction in burden of leukemia cells in chronic myelogenous leukemia (CML) patients, as indicated by achievement of complete cytogenetic remission and major reduction in Bcr-Abl transcript levels on Q-PCR analysis. However patients treated with IM alone without prior interferon treatment appear to invariably relapse on discontinuation of IM treatment. In addition we and others have shown that residual Bcr-Abl+ progenitors persist in IM-treated CML patients following achievement of CCR. These observations suggest that despite its remarkable activity in CML, IM fails to eliminate all malignant stem and progenitor cells in CML patients. However our previous studies were conducted on patients within the first year or two of IM treatment, whereas recent studies have indicated that Bcr-Abl levels continue to decline on Q-PCR analysis with continued IM treatment. This together with the decreasing rate of disease relapse observed after 3 years of IM treatment raises the possibility that prolonged IM treatment may cause depletion of residual CML stem cells. In this study we investigated whether prolonged IM treatment was associated with a reduction in Bcr- Abl+ stem and progenitor cells. We evaluated 14 CML patients followed at our center who were in CCR, had been treated with IM for at least 4 years, and from whom multiple cryopreserved bone marrow samples were available for study. Bone marrow mononuclear cells (MNC) were thawed, CD34+ cells were selected by immunomagnetic columns, and CD34+CD38+ (38+) committed progenitors and CD34+CD38− (38−) stem/primitive progenitor cells were isolated by flow cytometry sorting. Q-PCR analysis of Bcr-Abl and Bcr transcript levels was performed on RNA isolated from MNC, 38+ and 38− cells and Bcr-Abl levels were reported as the ratio of Bcr-Abl to Bcr. Bcr-Abl levels in MNC were 0.010±0.005, 0.011± 0.005 and 0.013±0.005 at 3, 4 and 5 years. We observed that Bcr-Abl levels were higher in both 38+ and 38− cells in comparison with levels in MNC. A gradual decline in Bcr-Abl levels in 38+ cells was seen (0.285±0.185 at 3 years, 0.121±0.056 at 4 years, and 0.071±0.028 at 5 years). In contrast high Bcr-Abl levels were maintained in the 38− fraction despite continued IM treatment (0.162±0.086 at 3 years, 0.116±0.041 at 4 years, and 0.361±0.107 at 5 years). In contrast to IM-treated patients, Bcr-Abl transcripts were not detected in MNC and CD34+ cells from BM of CML patients who had received allogeneic hematopoietic cell transplants (n=5). To further investigate whether malignant stem cells persisted after prolonged IM treatment, MNC from 5 of the patients described above were transplanted by tail vein injection into sublethally irradiated NOD/SCID-IL2Rγ-chain knockout (NSG) mice. High levels of human cell engraftment were observed 4–5 weeks after injection, and Q-PCR analysis revealed high levels of Bcr-Abl expression in engrafted cells from 4 of 5 patients, confirming the presence of Bcr-Abl+ cells with NOD/SCID mouse repopulating capacity. In conclusion, our results clearly demonstrate the persistence of Bcr-Abl+ stem cells in the BM of CML patients in prolonged remission after 5 years of IM treatment. The observed persistence of leukemia stem cells raises the concern that patients remain at risk for relapse on drug discontinuation or through acquisition of IM resistance. The assays described here may have considerable utility for evaluating and monitoring the effects of experimental treatment strategies directed against residual CML stem cells.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author