Chronic Myeloid Leukemia Patients on Tyrosine Kinase Inhibitor Have Normal T Cell Responses to Vaccination but An Impaired IgM Humoral Response Associated with Loss of Discrete Memory B Cell Subsets,

Abstract 3753

The tyrosine kinase inhibitors (TKIs) imatinib (IM), nilotinib (NIL) and dasatinib (DAS) are remarkably effective as single-agent therapy for chronic myeloid leukemia (CML) in chronic phase (CP). However little is known of their potential impact on the immune response. No human in vivo studies to assess how these molecular-targeted drugs affect immune function in patients are available and data from in vitro and animal studies with imatinib have been contradictory, ranging from impaired antigen-specific T-cell response to enhanced stimulation of tolerant T cells. Furthermore, although the immunomodulatory effects of TKIs on T cells, NK cells and dendritic cells have been explored in vitro, little is known of their potential impact on B cells. To characterize the in vivo immunomodulatory effects of TKIs, 51 patients with CP-CML in complete cytogenetic response (CCyR) on standard dose IM (n=26), DAS (n=14) or NIL (n=12) and 28 adult controls were recruited during two influenza seasons (2008 and 2009). Patients and controls were concomitantly immunized with an influenza vaccine (Ph. Eur. 2008/2009 or Ph. Eur. 2009/2010, CSL Biotherapies) and with the 23-valent polysaccharide pneumococcal vaccine (Pneumovax II; Sanofi Pasteur MSD). Peripheral blood mononuclear cells (PBMCs) and serum samples were collected from patients and donors prior to vaccination and T and B responses to vaccination were assessed at 4 weeks and at 2–3 months post-immunization. T-cell responses to influenza vaccine were analyzed quantitatively and qualitatively using flow cytometry and intracellular cytokine assay for TNF-α, IFN-γ, IL-2 and the cytotoxicity marker CD107a. Serum titers of IgM and IgG pneumococcal antibodies were determined by ELISA. Analysis of B cell subsets was performed using flow cytometry and correlated with the pneumococcal IgM and IgG humoral response.

Following vaccination, Flu-specific T cells were detected in 24/51 (47.0%) patients on TKI and 15/24 (62.5%) healthy controls (p=0.16). Polyfunctional T-cell responses (defined as the production of 2 or more cytokines or one cytokine and the cytotoxic marker CD107a) were induced in 6/10 evaluable patients and 4/8 normal controls (p=1.0).

T-cell independent humoral responses to vaccination were assessed in 45 patients and 12 healthy controls by measuring pneumococcal IgM titers. Four weeks postimmunization, 11/12 (92%) controls achieved IgM pneumococcal Ab titers >80 U/ml compared to only 23/45 (53%) CML patients on TKI (p=0.010). The pneumococcal IgM titers were significantly lower in patients with CML on TKI compared to healthy controls (median, 89.0 U/ml, range 5–200 vs 200 U/ml, range 58–200, p=0.0006), suggesting that CML patients on TKI have impaired IgM responses to vaccination. To further characterize the humoral immune response to Pneumovax, we stratified CML patients based on their pneumococcal IgM titers. We found a significantly lower percentage of IgM memory B cell subset in CML patients who failed to mount a significant pneumococcal IgM response compared to patients who achieved a pneumococcal IgM response (median, 6.25% vs 16.4%, p=0.0059) and healthy controls (median, 6.25% vs 14.3%, p=0.0086). Furthermore, we found a significant correlation between anti-pneumococcal IgM titers and IgM memory B cell percentage (Spearman rank correlation test, r=0.61, p<.0001).

To investigate a putative role of TKIs for the loss of IgM memory B cell subsets in CML patients, we determined the frequencies of IgM memory B cells in paired samples collected from 15 CML-CP patients at diagnosis (i.e. prior to initiating IM) and once CCyR was achieved. We found a significant decrease in the percentage of IgM memory B cells in CML-CP patients treated with IM compared to the pre-treatment sample (median 9.4%, vs. 15.2% respectively, p=0.0023).

In summary, patients with CML on TKIs can mount effective T-cell immune responses to influenza vaccination. Our data suggest that TKIs (IM, DAS and NIL) impair T-cell independent humoral immune responses, namely IgM responses to vaccination. This is associated with a loss of IgM memory B cell subsets. Further investigations to understand the mechanisms by which TKIs may impact B-cell subsets are underway. These results are of particular interest in terms of the long-term effects of TKI on tumor immune surveillance and susceptibility to infections and may have implication for vaccination strategies in CML patients.