Dasatinib is a potent tyrosine kinase inhibitor (TKI) used in the treatment of Philadelphia-chromosome positive (Ph+) leukemia. We and others have recently observed that in a proportion of patients, dasatinib induces an absolute, oscillating blood lymphocytosis characterized by an expansion of pre-existing clonal cytotoxic memory cells (Kreutzman et al. Blood 2010). Lymphocytosis has been correlated with autoimmune adverse effects (pleural effusions, colitis) and with exceptionally good therapy responses in patients with advanced leukemia. The factors causing lymphocytosis are unknown.
To study the cellular and molecular mechanisms of lymphocytosis in dasatinib treated patients and to assess the relation of drug intake, plasma level and lymphocyte counts.
Blood samples from 23 consecutive Ph+ leukemia (CML, Ph+ALL; first and second line therapies) patients treated either with dasatinib (n=17, daily doses 50–100 mg), imatinib (n=2, 400 mg), nilotinib (n=2, 300 mg) or bosutinib (n=2, 500 mg) were collected before drug intake and after 1, 2, 4 and 6 hours after a single oral dose. Similar data was collected from 1 healthy control after 100 mg of dasatinib. Immunophenotype and T-cell receptor (TCR) V beta expression were determined with flow cytometry and a multiplex bead assay was used for the measurement of plasma cytokines. A differential gene expression analysis was done with an Agilent 44k oligonucleotide array. Plasma dasatinib concentrations were measured by HPLC/MS.
In all patients and in one healthy control dasatinib induced a rapid and marked mobilization of blood leukocytes with peak values after 1–2 hours (h) of oral drug intake. Most substantial mobilization was observed for lymphocytes (median fold change [mFC] 2.15, range 1.05–5.12, 0h vs. 1h, P<0.0001) with peak levels at 1 hour up to 10.5 × 109/L and for monocytes (mFC 1.64, range 0.95–3.25, P=0.0026). There was a small increase in neutrophil count peaking at 2 h (mFC 1.32, P=0.008), a trend for lower platelet counts at 2–4 h post dasatinib intake, but no changes in hemoglobin. At 12 hours the leukocyte and lymphocyte counts returned to normal levels.
In other TKI treated patients (imatinib, nilotinib, bosutinib), no similar changes were observed (Fig a).
A close correlation between plasma dasatinib concentration and lymphocyte count was observed in all patients studied (Fig b). In a dose escalation experiment (2 patients, 20–100 mg), an intra-patient correlation between dose and peak lymphocyte count for doses ≥40 mg was evident.
By lymphocyte immunophenotyping, a preferential mobilization of NK-, NKT- and γδ+ T-cells was observed in all patients, albeit the bulk of lymphocytes consisted of αβ+ T-cells. Two patients with advanced leukemia and an exceptionally good response to dasatinib were studied in more detail. Their dominant lymphocyte population consisted of CD8+CD45RO+HLA-DR+CD62Lneg activated effector memory T-cells with a two-fold and four-fold increase in their relative and absolute frequency, respectively, at 1 h post dasatinib. Mobilization of single TCR Vbeta clones accounted for the observed changes in the lymphocyte population. A differential gene expression analysis from CD8+ T-cells at 0 h vs. 1 h showed upregulation of IFN-γ inducible genes such as CXCL11 and CXCL9 and complement protein C1qB previously correlated with CD8+ T cell viral immune responses. Downregulated genes included ß-catenin and inhibin known to play a role in lymphocyte maturation. The chemotactic CXCL10 (IP-10) and CXCL-9 (MIG) plasma cytokine levels were significantly increased consistent with an IFN-γ type immune activation. Both these cytokines are ligands for CXCR3 expressed on T- and NK-cells, which is a key inducer of cellular responses in leukocyte trafficking.
In all patients and in one healthy control, dasatinib induced a rapid mobilization of lymphocytes in blood which followed closely drug plasma concentration, and was not observed with other TKIs. The hypothesis of temporal inhibition of downstream signaling of e.g. adhesion molecules on lymphocytes is under investigation. Rapid cytotoxic lymphocyte mobilization is likely a causative factor behind complete and prolonged therapy responses and autoimmune phenomena associated with dasatinib. This drug effect is accurately dose-controllable and not specific for Ph-positive leukemia.
Mustjoki:BMS, Novartis: Honoraria. Rousselot:BMS, Novartis: Research Funding; BMS: Honoraria. Molimard:Novartis: Research Funding. Smykla:BMS: Employment, Equity Ownership. Lee:BMS: Employment, Equity Ownership. Porkka:BMS, Novartis: Consultancy, Honoraria, Research Funding.
Asterisk with author names denotes non-ASH members.