Since the emergence of tyrosine kinase inhibitors (TKIs), long-term survival of patients with chronic myelogenous leukemia (CML) has been improved. However, those TKIs have not fully succeeded in curing CML, mainly due to TKI-resistant CML stem cells. CML stem cells are often difficult to analyze because they represent an extremely minor population of CML cells. To overcome this obstacle, we established integration-free induced pluripotent stem cells (iPSCs) from bone marrow (BM) cells of two patients with CML in chronic phase (CML-CP) and obtained CML pre-hematopoietic progenitor cells (CML-pre-HPCs), immature hematopoietic cells phenotypically defined by CD34+/CD45-/CD43+ cells.

In semisolid and liquid cultures, CML-pre-HPCs recapitulated the principal features of CML stem cells, multi-potency and the resistance against imatinib. Gene expression enrichment analysis for CML-pre-HPCs demonstrated that several gene sets, including those related to the maintenance of hematopoietic stem cells were enriched. In addition, we found that a disintegrin and metalloprotease 8 (ADAM8), also known as CD156, was highly enhanced in CML-pre-HPCs and the expression level of ADAM8 was even increased after the treatment of imatinib in vitro. To address the significance of ADAM8 in CMP-CP patient, we evaluated purified ADAM8+ cells by fluorescence-activated cell sorting (FACS) in primary samples.

First, FACS analysis found that ADAM8+ cells were enriched more in BM samples of patient with newly diagnosed CML-CP than normal or other types of leukemias among CD34+ fraction. ADAM8+ cells were enriched in CD34+/CD38- fraction compaered to CD34+/38+ fraction in BM of CML-CP patients, indicating that ADMA8+ cells represent immature hematopietic cells. In cell viability assays, ADAM8+/CD38+ CML cells in newly diagnosed CML-CP patient enhibited imatinib-resistance and imatinib-induced apoptosis in vitro was strongly suppressed in ADAM8+ CML cells compared to ADAM8- cells. Even in CD34+/CD38+ fraction, which was previously known as TKI-sensitive fraction, ADAM8+ cells exhibited TKI-resistance in both cell viability and apoptosis assay, indicating that ADAM8 would be a useful marker of TKI-resistant CML cells.

Finally, to evaluate the significance of ADAM8 as a marker of TKI-resistant CML cells in vivo, we measured the frequency of CML cells in BM samples of CML-CP patients who had achieved major or complete molecular response (MMR; n = 2 or CMR; n = 1) after the administration of TKIs by limiting dilution analysis. In CML patients with MMR, CML cells remained in ADAM8+ cells at higher frequency in spite of steep decline of CML cells in ADAM8- cells The frequency of CML cells was as high in CD34+/CD38+/ADAM8+ fraction as in CD34+/38- CML stem cell fraction. Even in a patient with CMR, residual CML cells were detected in ADAM8+ population among CD34+/CD38+ fraction, whereas CML cells were undetectable in ADAM8- population.

In conclusion, we have established a powerful platform with CML-iPSCs to investigate the pathophysiology of TKI-resistant CML stem cells. Using this platform, we have identified ADAM8 as a novel marker of TKI-resistant CML cells. CD34+/CD38+/ADAM8+ fraction, as well as CD34+/CD38- fraction, was an important population that defines residual CML cells even in CML-CP patients with deep molecular response after the treatment of TKI. ADAM8 would become an attractive candidate of novel therapeutic targets against TKI-resistant CML cells.


Kataoka:Yakult: Honoraria; Boehringer Ingelheim: Honoraria; Kyowa Hakko Kirin: Honoraria.

Author notes


Asterisk with author names denotes non-ASH members.