Dasatinib is a Src family kinase inhibitor and is widely used in the management of CML. It has been shown to have immunosuppressive effects in vitro and clinical use is associated with the development of CD8 T cell and Natural Killer cell expansions that can lead to lymphocytosis. Interestingly, increased lymphocyte counts have been shown to be associated with improved responses to dasatinib.

We measured lymphocyte subsets in 94 patients with CML who were treated with either imatinib or dasatinib and were in at least a complete cytogenetic remission. 55 were treated with imatinib and 39 with dasatinib.

The overall lymphocyte count was increased in dasatinib-treated patients, at 2.0 * 109/L versus 1.5 * 109/L. Furthermore, within the dasatinib –treated group, CMV seropositivity was associated with a marked increase in overall lymphocyte count within the dasatinib-treated cohort, at 2.3 * 109/L versus 1.7 * 109/L.

CD8+ T cells were also increased within the dasatinib-treated cohort, and CMV seropositivity was associated with a marked further increase in cell count. Indeed, CMV seropositive dasatinib-treated patients had 60% higher CD8 T cell counts compared with imatinib treated CMV-seropositive patients, at 0.5 * 109/L versus 0.3*109/L (p = 0.006). A general linear model showed that both CMV seropositivity (p < 0.001) and dasatinib treatment (p = 0.001) were both associated with increases in CD8 T cell numbers. The most marked changes were seen within the CD45RA+ CD27+ EM group and the CD45RA+ CD27- EM group of CD8 T cells. Here there were double the number of CD45RA+ CD27- CD8 T cell responses in dasatinib-treated CMV seropositive patients compared with those on imatinib (0.24 * 109/L versus 0.11 *109/L, p = 0.01), and a threefold increase in EM CD45RA- CD27- CD8 T cells (0.09 * 109/L versus 0.034 * 109/L, p = 0.05). Dasatinib-treated patients also showed increased functional CMV-specific T cell responses, as measured by IFNg release to CMV peptide pools, at 5.1% of CD8 T cells versus 1.4% (p = 0.019)

gd T cell numbers were also increased in patients on dasatinib compared with imatinib. The median gd T cell count was 0.05 * 109/L versus 0.02 * 109/L (p = 0.005) and here CMV seropositivity was also associated with the highest gd T cell count. Further analysis of the gd T cell subset revealed that the majority of the gd T cells in the CMV-seropositive group expressed the atypical Vd1 T cell receptor, as opposed to the Vd2 TCR most commonly found in peripheral blood. The median Vd1 gd T cell count in dasatinib-treated CMV seropositive patients was six times that of those on imatinib, at 0.06 * 109/L versus 0.0097 * 109/L (p = 0.02). A general linear model showed that both dasatinib treatment (p = 0.013) and CMV seropositivity (p = 0.002) were associated with significant increases in the Vd1 gd T cell count. The predominant memory subset in those patients that had expansions was of the CD45RA+ CD27- TEMRA subset.

The largest expansions of Vd1 gd T cells were seen in patients who were CMV IgM seropositive and here the Vd1 gd T cell count was threefold higher at 0.16 * 109/L compared with 0.05 * 109/L (p = 0.014). Sorted Vd1 gd T cells from these patients showed specificity for CMV-infected fibroblast monolayers as measured by IFNg release. RACE-PCR was then used to interrogate the T cell repertoire of the Vd1 gd T cell population which was found to be strikingly oligoclonal in all cases. Indeed, in one patient a single clone of Vd1 gd T cells made up 99% of the total Vd1 population.

In addition to cellular immune response, the humoral CMV-specific immune response was also increased in patients on dasatinib. CMV seropositive patients on dasatinib had higher CMV-specific IgG titre than those on imatinib, at 20.1 IU/ml versus 16.1 IU/ml (p = 0.048). Furthermore, within the CMV IgG seropositive patients, 4 of 23 dasatinib-treated patients were CMV IgM seropositive at various timepoints on treatment, compared with only 1 of 38 of those on imatinib.

CMV infection is known to lead to a marked increase in the number of cytotoxic lymphocytes in the blood of healthy individuals. These data reveal that this effect is augmented by the action of dasatinib, which may potentially act by increasing the level of subclinical viral reactivation. Interestingly, these cytotoxic populations include cells has been shown to have anti-leukaemia properties and may have a role in immune anti-tumour responses.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

Sign in via your Institution