Effects of Plerixafor (AMD3100) In Combination with Tyrosine Kinase Inhibition In a Murine Model of CML.

Most patients with chronic myeloid leukemia (CML) treated with the tyrosine kinase inhibitor (TKI) imatinib attain durable complete cytogenetic responses. However, residual leukemia cells persist even in the best responders and disease recurrence usually occurs when imatinib is discontinued. The bone marrow niche is thought to protect leukemic stem cells (LSC) upon drug challenge by providing survival signals. The interaction between SDF1 and its receptor CXCR4 is an important mechanism that facilitates HSC homing. In CML progenitors CXCR4 is downregulated, resulting in impaired SDF1 mediated migration and adhesion. Recent studies have shown that TKI inhibition of BCR-ABL upregulates CXCR4 expression, which restores the SDF1/CXCR4 axis and may allow CML LSC to retreat to the bone marrow niche, protecting them from BCR-ABL inhibition. We hypothesize that this effect of TKI therapy may provide a survival signal to LSC and contribute to disease persistence. Conversely, interrupting the SDF1/CXCR4 interaction with plerixafor, a small molecule antagonist of CXCR4, may restore TKI vulnerability and result in elimination of residual leukemia cells.

Treatment of mice with plerixafor (5 mg/kg daily SC) disrupted the CXCR4/SDF1 interaction as indicated by mobilization of CXCR4+ cells into the blood. We then tested if the addition of plerixafor to TKI (imatinib or dasatinib) reduces disease burden in a murine CML retroviral transduction/transplantation model. Mice (10/group) were treated with plerixafor in combination with imatinib (100 mg/kg BID orally), imatinib alone, plerixafor alone, or carrier. All plerixafor alone and mock treated mice died of leukemia by day 16 post transplant. There were no significant differences in median survival between combination and imatinib only groups (30 days). Although both groups achieved normalization of white blood counts (WBC) there was continued evidence of disease with persistence of 60–80% GFP+ leukemic cells in the peripheral blood and 4 fold larger spleens and livers as compared to normal mice. In a second experiment we tested dasatinib as a more potent TKI in an attempt to obtain a more faithful model of residual disease. Dasatinib treatment (10 mg/kg BID orally, N=36) was initiated upon engraftment (day 12 post transplant). Once the WBC normalized the cohort was divided into 2 groups: dasatinib only (N=19) and plerixafor+dasatinib (N=17). Kaplan Meier analysis showed no significant difference in the survival between combination vs dasatinib only groups (median 70 vs 71 days). WBC of all treated mice remained in the normal range with persistence of 60–80% GFP+ leukemic cells. Although spleen (0.04± 0.01g) and liver weights (0.7± 0.1g) were normal in both groups histopathological analysis showed increased infiltration of myeloid cells in spleens, livers and lungs of the combination group as compared to the dasatinib only group. Immunoblotting for pCrkL from spleen and marrow cells confirmed that dasatinib was inhibiting BCR-ABL kinase activity suggesting that disease persistence is not due to the lack of BCR-ABL kinase inhibition. Leukemic mice from both groups developed neurological symptoms including rear limb paralysis, leaning to one side, head tilting and twisting, and eventually complete loss of motor control. This was significantly more frequent in plerixafor +dasatinib group (92%) as compared to the dasatinib only group (21%) (p<0.001) and symptoms occurred earlier in the combination group (40 vs 58 days). Preliminary histopathological analysis of brain from a subset of animals suggests more extensive leukemic infiltrates in the combination group as compared to the dasatinib group. Control mice transplanted with untransduced marrow and treated with plerixafor and/or dasatinib for up to 80 days remained asymptomatic.

1) The addition of plerixafor to TKIs does not reduce leukemia burden in this murine CML model 2) Plerixafor+dasatinib appears to promote extra medullary leukemic infiltrates in liver, lung, spleen and brain 3) A state of minimal residual disease was not achieved, despite efficient inhibition of BCR-ABL kinase activity and the lack of efficacy and potential detrimental effects of plerixafor may reflect the fact that in this model active leukemia persists. Caution will be warranted in clinical combination studies of patients with highly active CML.

***First three authors have equal contribution.

Deininger:Genzyme: Research Funding; Novartis: Consultancy; BMS: Consultancy; Ariad: Consultancy.