Treatment with Tyrosine Kinase Inhibitors May Impair the Potential Curative Effect of Allogeneic Stem Cell Transplantation.

Abstract 857

Tyrosine kinase inhibitors (TKI) like imatinib and dasatinib are the current treatment of choice for patients with chronic myeloid leukemia (CML). Although most patients enter a complete remission during treatment, cure of the disease is usually not achieved since recurrence of the disease is seen in the majority of patients upon discontinuation of the treatment, indicating that the leukemic stem cell is not efficiently targeted. Furthermore, in accelerated phase and blast crisis of CML TKI treatment only results in temporary control of the disease. In these situations allogeneic stem cell transplantation (allo-SCT) and application of donor T cells may be the only curative treatment. Besides the direct anti-leukemic effect of allo-SCT, alloreactive T cells recognizing CML (progenitor) cells, and the formation of immunological memory may lead to effective lifelong immune surveillance. Therefore, we investigated whether the leukemic cells persisting during TKI treatment are susceptible targets for the anti-leukemic effect mediated by donor T cells after allo-SCT and whether continuous TKI treatment may have an additive effect during the immunological intervention. To investigate the anti-leukemic effect of the two strategies, CD34+ positive CML cells were isolated from bone marrow, and labeled with the fluorescent dyes CFSE or PKH to allow monitoring of single cell proliferation. CML cells were exposed to imatinib (1-100μM) or dasatinib (0.01-50nM), and/or to CD8+ alloreactive cytotoxic T lymphocyte (CTL) clones in the presence of proliferation-inducing cytokines. The number, phenotype, and proliferative status of the CML cells persisting after single and combined interventions were measured by quantitative flowcytometric analysis. In the absence of therapeutic interventions the majority of CD34+ CML cells entered proliferation. However, a small population of CD34+ CML stem cells residing in the non-dividing peak could be identified despite the addition of cytokines. Addition of imatinib or dasatinib resulted in efficient dose-dependent induction of cell death of the leukemic cells (99% lysis by 25μM imatinib or 10nM dasatinib). However, the population of quiescent CD34+ CML stem cells was not affected. Moreover, the number of cells present in the non-dividing population increased 2-fold compared to the non-treated controls at the highest TKI concentrations, indicating additional growth arrest of a population of proliferating CML precursor cells. We next tested the capacity of different HLA-A2-restricted CD8+ CTL clones to kill non-treated or imatinib or dasatinib treated CML cells. Whereas the proliferating CD34+ CML precursors were efficiently lysed, the population of quiescent stem cells was capable of withstanding CTL exposure. Detailed phenotypic analysis revealed significant downregulation of HLA-A2 and the adhesion molecules CD49d and CD58 on these quiescent cells, probably resulting in the impaired ability of these target cells to form a high avidity interaction with the T cells. The increased population of non-dividing cells as a result of the TKI pretreatment showed similar resistance to T cell induced cell death, indicating that TKI treatment may even diminish the anti-leukemic effect of allo-SCT. In the absence of therapeutic control, as mimicked by the removal of T cells and TKI from the cultures, outgrowth of the leukemic cells re-occurred, illustrating their capacity to give rise to a relapse of the disease. Next, we analyzed the effect of TKI treatment on T cell survival and functionality. Whereas resting primary T cells were insensitive to TKI treatment, T cells activated by either polyclonal stimulation with anti-CD3/CD28 beads or stimulation with allogeneic stimulator cells died after TKI exposure at similar concentrations as the leukemic cells. In conclusion, TKI treatment results in selection of a population of quiescent leukemic stem cells showing cross-resistance to CTL-induced cell death, most likely due to their inability to form a high avidity interaction. Moreover, T cells actively participating in the anti-leukemic immune response after allo-SCT are suppressed by TKI. These data indicate that continuous TKI treatment may potentially hamper the curative effect of allo-SCT.