Anti-Leukemia Immune Responses in CML Patients Treated with Imatinib Mesylate.

Imatinib mesylate, a selective inhibitor of the bcr/abl tyrosine kinase, has revolutionized the treatment of patients with chronic myelogenous leukemia (CML). Most CML patients in chronic phase achieve hematologic remission with imatinib, while some achieve cytogenetic remission. As imatinib is an oral agent with few side effects, it has rapidly become the first-line therapy for most CML patients. However, this therapy does not represent a cure, as patients who discontinue the drug invariably relapse. Furthermore, imatinib resistance is beginning to emerge in some patients. Hence, the need to find alternate, potentially curative, therapies for CML remains. To date, the only curative treatment for CML is allogeneic bone marrow or stem cell transplantation (ABMT). A major mechanism of the curative potential of ABMT is immunological, as evidenced by the poor clinical outcome with T cell-depleted ABMT, and the efficacy of donor lymphocyte infusions (DLI) upon relapse. We hypothesized that an effective anti-leukemia immune response may emerge in patients entering remission on imatinib which may contribute to its clinical effectiveness. If so, strategies to further enhance this anti-leukemia immune response may lead to a potential cure. To determine if CML patients in remission on imatinib develop anti-leukemia immune responses, blood and bone marrow samples from patients before and after treatment were collected and analyzed. Pre-treatment samples were utilized as sources of autologous leukemic cells to detect anti-leukemia immune responses in post-treatment samples in IFN-g ELISPOT assays. Pre-treatment samples alone, post-treatment samples alone, and when available, serial post-treatment samples mixed together served as controls. In 9 of 14 patients investigated, IFN-g release was detected in pre- and post-treatment samples together with a median response of 22 spots above background (range 10 – 56 dots, p<0.01), whereas serial post-treatment samples together in 8 patients yielded results similar to background (median 5, range 5 – 20). In 6 of these patients in hematologic (or cytogenetic) remission, sufficient cells were available to allow additional analyses via intracellular staining for IFN-g, TNF-a, and IL-2 in autologous leukemia stimulated T cells (CD4 and CD8) and NK cells. In 4 of 6 patients, leukemia-reactive T cells were detected, most prominently in CD4+ T cells expressing TNF-a (1.4 – 37%), followed by IL-2 (0.3 – 12%) and IFN-g (0.1 – 4.6%). NK cells did not show significant expression of these cytokines upon stimulation with autologous leukemia cells. In pre-treatment and post-treatment samples alone, IL-2, TNF-a, and IFN-g expression was not detectable (0 – 0.5%). These results suggest that a significant portion of CML patients in remission with imatinib develop an anti-leukemia immune response, most notably in CD4+ T cells. Mechanisms by which imatinib treatment leads to anti-leukemia immune responses, and the molecular targets to which these cells are directed, will be further investigated. This knowledge will be useful in the development of immunotherapy strategies against CML as well as other leukemias, and raises the hope that immunotherapy may be combined with imatinib to eradicate residual leukemia cells for a durable cure of the disease.