Abstract 1223

Sonic hedgehog (Shh) pathway activators, particularly smoothened (SMO), are implicated in driving mouse leukemia progression by promoting leukemia stem cell maintenance. Thus, we investigated the role of the Shh signaling pathway in a humanized mouse model of blast crisis (BC) CML as well as the effects of a novel SMO antagonist on leukemia stem cell (LSC) quiescence, survival and self-renewal both in vitro and in vivo.


Human BC CML progenitor cells (n=3 patients) were treated for 7 days in vitro with vehicle (DMSO) or a clinical grade SMO antagonist (Pfizer; PF-04449913) and harvested for Shh target gene Q-PCR analysis. FACS-purified BC CML LSCs (CD34+CD38+Lin-PI-) were transplanted intrahepatically into newborn RAG2-/-gc-/- mice. When human progenitors were detectable in peripheral blood (8 weeks), mice were treated for 14 days daily by oral gavage with 200μl of drug or vehicle (50% PEG in HBSS). The treatment included dasatinib alone (50 mg/kg), PF-04449913 (100 mg/kg) with or without dasatinib (50mg/kg). After treatment the percentage of LSC (CD34+CD38+CD45RA+CD123+Lin-PI-) was determined by FACS analysis of the bone marrow, peripheral blood, spleen and liver and the number of myeloid sarcomas was evaluated. In addition, cell cycle analysis was also performed on the same hematopoietic tissues of vehicle compared with treated mice. To determine if self-renewal potential was abrogated by treatment, human leukemic progenitors were serially transplanted and myeloid sarcoma formation was assessed in secondary recipients after 8 weeks.


BC CML progenitors (n=3 patient samples) harbored increased expression of Gli and downregulation of Ptch1 transcripts compared to normal cord blood. Notably, Gli expression decreased following in vitro treatment with PF-04449913. Treatment with PF-04449913 alone in human BC CML LSC transplanted mice (n=12 exp., n=46 mice) significantly reduced spleen size (p ≤ 0.05). However, SMO inhibition alone did not significantly inhibit bone marrow engraftment compared with vehicle (n=12 exp., n= 47 mice). Although dasatinib alone decreases myeloid sarcoma formation (p ≤ 0.05), like the SMO antagonist alone it did not eradicate the LSC in the liver or bone marrow (n=12 exp., n=36 mice). Combination treatment (n=12 exp., n=34 mice) with PF-04449913 and dasatinib revealed a significant decrease in LSC hepatic engraftment (P≤0.05) compared with the PF-04449913 or dasatinib alone. Moreover, combination treatment completely eradicated the capacity of LSC to form myeloid sarcomas formation (0.0) in secondary transplant recipients (n=5) compared with dasatinib that had an average of 1.8 tumors (n=6), PF-04449913 that had an average of 4.9 (n=7) and vehicle group that had 3.8 tumors (n=16).

Hematopoietic progenitor assays performed with normal cord blood CD34+ progenitors (n=3 samples) showed no reduction in colony survival following treatment with PF-04449913 (1μM) compared with vehicle. Furthermore, there was no significant inhibition of normal CD34+ colony replating following PF-04449913 compared with vehicle treatment. Cell cycle analysis demonstrated a reduction in quiescent human leukemic cells in the peripheral blood and marrow following SMO inhibitor treatment.


Combination therapy with a potent BCR-ABL inhibitor, dasatinib, and a selective SMO inhibitor abrogates the self-renewal capacity of human imatinib resistant blast crisis leukemia stem cells. A phase I clinical trial with PF-04449913 is currently underway and a Phase 1b trial involving dasatinib and PF-04449913 for patients with advanced phase, imatinib resistant CML will begin in the near future.


Jamieson:Pfizer: Research Funding; CIRM: Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.