Search

Lack of allelic exclusion in the T-cell receptor (TCR) α locus gives rise to 2 different TCRs in 10% to 30% of all mature T cells, but the significance of such dual specificity remains controversial. Here we show that human CD4+CD25+regulatory T (Treg) cells express 2 distinct Vα chains and thus 2 TCRs at least 3 times as often as other T cells. Extrapolating from flow cytometric analysis using Vα2-, Vα12-, and Vα24-specific monoclonal antibodies (mAbs), we estimated that between 50% and 99% of the CD25+Treg cells were dual specific, as compared with about 20% of their CD25–counterparts. Moreover, both TCRs were equally capable of transmitting signals upon ligation. Cells with 2 TCRs also expressed more FOXP3, the Treg-cell lineage specification factor, than cells with a single TCR. Our findings suggest that expression of 2 TCRs favors differentiation to the Treg-cell lineage in humans and raise the question of the potential functional consequences of dual specificity.

Lack of allelic exclusion in the T-cell receptor (TCR) α locus gives rise to 2 different TCRs in 10% to 30% of all mature T cells, but the significance of such dual specificity remains controversial. Here we show that human CD4+CD25+ regulatory T (Treg) cells express 2 distinct Vα chains and thus 2 TCRs at least 3 times as often as other T cells. Extrapolating from flow cytometric analysis using Vα2-, Vα12-, and Vα24-specific monoclonal antibodies (mAbs), we estimated that between 50% and 99% of the CD25+ Treg cells were dual specific, as compared with about 20% of their CD25– counterparts. Moreover, both TCRs were equally capable of transmitting signals upon ligation. Cells with 2 TCRs also expressed more FOXP3, the Treg-cell lineage specification factor, than cells with a single TCR. Our findings suggest that expression of 2 TCRs favors differentiation to the Treg-cell lineage in humans and raise the question of the potential functional consequences of dual specificity.

Before the era of tyrosine kinase inhibitors (TKIs), interferon alpha (IFN-α) was the treatment of choice in CML and prolonged survival of responding patients. Recent studies suggest that combination of IFN-α with TKI improves therapy outcome. Significantly, a proportion of IFN-α treated patients in prolonged complete cytogenetic remission (CCyR) have been able to discontinue treatment without disease relapse (Mahon et al. JCO 2002). The mechanism of action of IFN-α therapy is incompletely understood; the drug exerts both direct cytotoxic and immunomodulatory effects on leukemic cells. The aim of this project was to study the immunomodulatory effects of IFN-α in CML patients in prolonged remission and isolate biological markers predicting response.The study population consisted of CML patients treated with IFN-α monotherapy (n = 10, median therapy time 146 months, range 63–231 months) and CML patients who had discontinued IFN-α therapy but remained in remission for >2 years (n = 9, median therapy time 120 months, range 80–184; median time without therapy 53 months, range 24–96). None of the patients were previously treated with TKI therapy. In addition, non-CML patients (3 patients with essential trombocythemia and one patient with polycythemia vera) treated with IFN-α and healthy volunteers (n = 43) were included as controls. Lymphocyte populations in all four groups were characterized with comprehensive immunophenotyping panels. The clonality of T-cells was analyzed by a TCR γ/δ rearrangement assay by PCR. Lymphocytes were further sorted into CD3+ TCR αβ+ and CD3+ TCR γδ+ populations (n = 8). Additionally, plasma levels of 25 cytokines were measured with a multiplex bead-based cytokine assay (Luminex®).The proportion of NK-cells from lymphocytes was significantly increased in IFN-α discontinued patients (median 26%, range 18–51%) compared to healthy volunteers (11%, 5–21%) or patients on IFN-α therapy (12%, 6–31%)(P=0.0005). Similarly the proportion of CD8+ cells from T-cells was significantly increased in both CML IFN-α groups (55% in IFN-α discontinued patients, 44% in IFN-α treated patients vs. 31% in healthy volunteers; P<0.05 for both groups). Also a larger proportion of T-cells expressed the long-term memory antigen CD45RO in IFN-α patients (74%, 58% vs. 44% in healthy controls, P<0.01). The proportion of regulatory T-cells (CD4+CD25+FoxP3+) was increased in IFN-α groups (6.1%, 5.2% vs. 3.8% in healthy volunteers, P=0.01). Similar changes in immunoprofile were not observed in IFN-treated non-CML patients.Clonal TCR γ/δ rearrangements were observed in 18 of 19 (95%) IFN-treated CML patients as compared to 3 of 22 (14%) in healthy volunteers (P<0.01). In both IFN-α CML patient groups a unique rearrangement pattern was observed: 14/19 (79%) of patients had the Vγ9 gene clonally rearranged. This clonal rearrangement resided in CD3+ γδ+ T-cell population, as assessed by cell sorting. Two of four non-CML patients treated with IFN-α had the same clonal rearrangement, as well as one healthy control (1/22; 5%). Similar clonality patterns have not been observed in dasatinib or imatinib treated CML patients (Kreutzman et al. Blood 2010).IFN-α treatment was associated with a distinct plasma cytokine profile in CML patients. IP-10, IL-6, IL-12, eotaxin, MCP-1, and IFN-γ levels were significantly increased in IFN-α treated CML patients. In particular, eotaxin and MCP-1 levels differed significantly between healthy controls and IFN-α patients who had successfully discontinued IFN-α therapy (428 vs. 1173 pg/ml, P<0.0001 and 107 vs. 459 pg/ml, P=0.0003, respectively). In IFN-α treated non-CML patients, eotaxin or MCP-1 levels were not increased.Our results show that IFN-α treatment induces distinct changes in the immunoprofile of CML patients. Patients who had successfully discontinued IFN-α therapy differed markedly from healthy controls. IFN-α therapy was associated with increased numbers of NK-cells and clonal γδ+ T-cells. These cells possess potent anti-leukemic activity and may contribute to the prolonged therapy responses in this group of patients. Furthermore, plasma cytokine profile could be a helpful biomarker when considering which patients can discontinue the IFN-α treatment without imminent disease relapse.Faber: BMS, Novartis: Consultancy, Honoraria. Porkka:BMS, Novartis: Consultancy, Honoraria, Research Funding. Mustjoki:BMS, Novartis: Honoraria.

Abstract 1201

Tyrosine kinase inhibitors (TKIs) targeting the BCR-ABL fusion protein are an effective therapy for Philadelphia chromosome positive (Ph+) leukemia. Dasatinib, a 2nd generation pan-TKI, also inhibits wild-type kinases, which may result in unexpected clinical responses. Recent data suggest that dasatinib has a potent immunosuppressive effect on T- and NK-cells in vitro. In contrast, we have noticed a marked expansion of lymphocytes in blood in a subset of dasatinib patients, but its clinical significance and molecular mechanisms are unclear. In this multicenter study, 19 Ph+ leukemia patients with marked lymphoproliferation in blood while on dasatinib (9 CML CP, 2 CML AP, 2 CML BC, 6 Ph+ALL) were identified. Clonality, immunophenotype, and intracellular signaling was analyzed and related clinical information was collected. In addition, prevalence and prognostic significance of the phenomenon was retrospectively assessed in a Phase II clinical study on 46 Ph+ ALL patients. An abrupt lymphocytosis (peak count range 4–20×109/L) with large granular lymphocyte (LGL) morphology was observed after a median of 3 months from the start of therapy (range 1–8 months). In most patients, LGL lymphocytosis was long-lasting and continued throughout dasatinib therapy, albeit with marked fluctuations in absolute lymphocyte count (Fig. 1a). In all evaluable patients (n=4), lymphocyte counts normalized after drug discontinuation. All patients were previously exposed to imatinib without similar changes in lymphocyte count or morphology. By immunophenotyping, 14 patients had a cytotoxic T-cell and 5 patients a NK-cell phenotype. After initiation of dasatinib therapy, the CD4/CD8 ratio shifted, expression of activation antigens HLA-DR and CD57 increased, expression of homing antigen CD62L decreased and relative numbers of regulatory T-cells were significantly decreased (p<0.001). All patients with T-LGL phenotype had mono/oligoclonal rearrangements in TCR g/d genes. Quantitative allele-specific oligonucleotide PCR analysis from samples before and during dasatinib therapy indicated that LGL lymphocytosis results from expansion of a pre-existing terminal memory cell clone (CD8+CD62Llow). By using single-cell profiling of nodal phosphoproteins (ERK1/2, STAT1, STAT3, STAT5, STAT6) by multi-color FACS, we showed that during dasatinib therapy LGL patients were more responsive to cytokine or growth factor stimuli (IL-2, IL-4, IL-6, IFNs, GM-CSF), when compared to dasatinib patients without lymphocytosis or to healthy controls. This indicates alternative signaling pathway usage not inhibited by dasatinib and/or cytokine hypersensitivity. Dasatinib-related severe adverse effects were common in patients with LGL lymphocytosis. Pleural effusions and/or pulmonary infiltrates were seen in 13/19 patients, modest CMV reactivation in 9/19 patients, colitis in 11/19 patients and long lasting mild fever in 14/19 patients. Accumulation of pheno- and genotypically identical cytotoxic T-cells was also detected in pleural effusion and colitis biopsy samples. Adverse effects were temporally related to dasatinib therapy and emerged after the appearance of LGL lymphocytosis. All patients with relapsed poor prognosis leukemia (Ph+ ALL n=5, blastic phase CML n=2) achieved durable complete molecular responses (CMR) during dasatinib therapy, CMR still ongoing in 3 patients (follow-up 18+ months). In the confirmatory Phase II dasatinib study in Ph+ ALL (START-L), patients with lymphocytosis had superior survival compared to patients without lymphocytosis (Fig. 1b). In conclusion, we propose that by inhibiting immunoregulatory kinases, dasatinib may induce a reversible state of aberrant immune reactivity in a distinct subset of patients and result in anti-leukemia and anti-host responses driven by cytotoxic T/NK LGL cells. Further studies pinpointing the mechanisms, such as the target antigens on the malignant cells and activation pathways are ongoing. Figure Figure

Tyrosine kinase inhibitors (TKIs) targeting the BCR-ABL fusion protein are an effective therapy for Philadelphia chromosome positive (Ph+) leukemia. Dasatinib, a 2ndgeneration pan-TKI, also inhibits wild-type kinases, which may result in unexpected clinical responses. Recent data suggest that dasatinib has a potent immunosuppressive effect on T- and NK-cells in vitro. In contrast, we have noticed a marked expansion of lymphocytes in blood in a subset of dasatinib patients, but its clinical significance and molecular mechanisms are unclear. In this multicenter study, 19 Ph+ leukemia patients with marked lymphoproliferation in blood while on dasatinib (9 CML CP, 2 CML AP, 2 CML BC, 6 Ph+ALL) were identified. Clonality, immunophenotype, and intracellular signaling was analyzed and related clinical information was collected. In addition, prevalence and prognostic significance of the phenomenon was retrospectively assessed in a Phase II clinical study on 46 Ph+ ALL patients. An abrupt lymphocytosis (peak count range 4–20×109/L) with large granular lymphocyte (LGL) morphology was observed after a median of 3 months from the start of therapy (range 1–8 months). In most patients, LGL lymphocytosis was long-lasting and continued throughout dasatinib therapy, albeit with marked fluctuations in absolute lymphocyte count (Fig. 1a). In all evaluable patients (n=4), lymphocyte counts normalized after drug discontinuation. All patients were previously exposed to imatinib without similar changes in lymphocyte count or morphology. By immunophenotyping, 14 patients had a cytotoxic T-cell and 5 patients a NK-cell phenotype. After initiation of dasatinib therapy, the CD4/CD8 ratio shifted, expression of activation antigens HLA-DR and CD57 increased, expression of homing antigen CD62L decreased and relative numbers of regulatory T-cells were significantly decreased (p<0.001). All patients with T-LGL phenotype had mono/oligoclonal rearrangements in TCR g/d genes. Quantitative allele-specific oligonucleotide PCR analysis from samples before and during dasatinib therapy indicated that LGL lymphocytosis results from expansion of a pre-existing terminal memory cell clone (CD8+CD62Llow). By using single-cell profiling of nodal phosphoproteins (ERK1/2, STAT1, STAT3, STAT5, STAT6) by multi-color FACS, we showed that during dasatinib therapy LGL patients were more responsive to cytokine or growth factor stimuli (IL-2, IL-4, IL-6, IFNs, GM-CSF), when compared to dasatinib patients without lymphocytosis or to healthy controls. This indicates alternative signaling pathway usage not inhibited by dasatinib and/or cytokine hypersensitivity. Dasatinib-related severe adverse effects were common in patients with LGL lymphocytosis. Pleural effusions and/or pulmonary infiltrates were seen in 13/19 patients, modest CMV reactivation in 9/19 patients, colitis in 11/19 patients and long lasting mild fever in 14/19 patients. Accumulation of pheno- and genotypically identical cytotoxic T-cells was also detected in pleural effusion and colitis biopsy samples. Adverse effects were temporally related to dasatinib therapy and emerged after the appearance of LGL lymphocytosis. All patients with relapsed poor prognosis leukemia (Ph+ ALL n=5, blastic phase CML n=2) achieved durable complete molecular responses (CMR) during dasatinib therapy, CMR still ongoing in 3 patients (follow-up 18+ months). In the confirmatory Phase II dasatinib study in Ph+ ALL (START-L), patients with lymphocytosis had superior survival compared to patients without lymphocytosis (Fig. 1b). In conclusion, we propose that by inhibiting immunoregulatory kinases, dasatinib may induce a reversible state of aberrant immune reactivity in a distinct subset of patients and result in anti-leukemia and anti-host responses driven by cytotoxic T/NK LGL cells. Further studies pinpointing the mechanisms, such as the target antigens on the malignant cells and activation pathways are ongoing.

In addition to therapeutic efficacy in imatinib-resistant BCR-ABL positive malignancies, dasatinib may have off-target immunomodulatory effects by inhibiting kinases (e.g. SRCs, c-KIT) in immune effector cells. In a proportion of patients, dasatinib treatment is associated with non-infectious pleural effusions and panniculitis, which may reflect aberrant immune reactivity. Here we describe a marked expansion of clonal LGL lymphocytes in blood among 11 BCR-ABL positive patients on dasatinib therapy (5 CML CP, 1 CML AP, 1 CML BC, 4 Ph+ALL). All patients were previously exposed to imatinib without similar changes in lymphocyte count or morphology. The median time to LGL lymphocytosis was 2 months from the start of dasatinib (range 1–8 months). It developed abruptly, with a peak lymphocyte count of 4–20 ×109/L, and was often preceded by low-grade fever. In all evaluable cases, lymphocyte counts normalized after drug discontinuation. By immunophenotyping, 6 patients had a CD3/CD8+ cytotoxic T-cell phenotype and 4 patients had a CD56+ NK-cell phenotype. The LGL lymphocytes were BCR-ABL negative. All CD3-positive cases had a clonal rearrangement of TCR gamma/delta genes by PCR; clonality was not detected in samples prior to dasatinib therapy (n=3). In TCR beta gene repertoire assay, oligoclonal patterns were seen in selected genes. All patients with HLA data had the A2 allele (n=8). Dasatinib-related complications were common: pleural effusions and/or pulmonary infiltrates (n=8), modest CMV reactivation (n=5), severe colitis (n=5), which developed after appearance of LGL lymphocytosis. Immunophenotyping was available from 2 patients with pleural effusions: in both cases the majority of cells were CD3+CD8+ T-cells and no leukemic cells were detected. TCR gamma assay was done from one patient sample and it showed similar clonal pattern as in peripheral blood. However, despite frequent side-effects, response to dasatinib treatment was very good in all cases, including complete molecular responses in 5 advanced phase patients. Transient clonal LGL lymphocytosis has been described in few patient cases with a primary herpesvirus infection. Our patients share striking pheno/genotypic similarities with LGL leukemia, in which an antigen driven expansion of LGLs precedes the occurrence of oncogenic events. However, in our patients LGL lymphocytosis appeared to be a benign phenomenon. We postulate that by virtue of inhibition of key non-BCR-ABL kinases, dasatinib induces a reversible state of autoimmune-like reactivity upon antigen challenge in a proportion of patients with a distinct genetic and/or environmental background. The aberrant immune response may also result in enhanced anti-leukemic control driven by the cytotoxic T/NK LGL cells. Dasatinib could be a valuable tool in uncovering the pathogenesis of LGL lymphoproliferation, putatively related to mutations in kinases targeted by the drug.

In addition to therapeutic efficacy in imatinib-resistant BCR-ABL positive malignancies, dasatinib may have off-target immunomodulatory effects by inhibiting kinases (e.g. SRCs, c-KIT) in immune effector cells. In a proportion of patients, dasatinib treatment is associated with non-infectious pleural effusions and panniculitis, which may reflect aberrant immune reactivity. Here we describe a marked expansion of clonal LGL lymphocytes in blood among 11 BCR-ABL positive patients on dasatinib therapy (5 CML CP, 1 CML AP, 1 CML BC, 4 Ph+ALL). All patients were previously exposed to imatinib without similar changes in lymphocyte count or morphology. The median time to LGL lymphocytosis was 2 months from the start of dasatinib (range 1–8 months). It developed abruptly, with a peak lymphocyte count of 4–20 ×109/L, and was often preceded by low-grade fever. In all evaluable cases, lymphocyte counts normalized after drug discontinuation. By immunophenotyping, 6 patients had a CD3/CD8+ cytotoxic T-cell phenotype and 4 patients had a CD56+ NK-cell phenotype. The LGL lymphocytes were BCR-ABL negative. All CD3-positive cases had a clonal rearrangement of TCR gamma/delta genes by PCR; clonality was not detected in samples prior to dasatinib therapy (n=3). In TCR beta gene repertoire assay, oligoclonal patterns were seen in selected genes. All patients with HLA data had the A2 allele (n=8). Dasatinib-related complications were common: pleural effusions and/or pulmonary infiltrates (n=8), modest CMV reactivation (n=5), severe colitis (n=5), which developed after appearance of LGL lymphocytosis. Immunophenotyping was available from 2 patients with pleural effusions: in both cases the majority of cells were CD3+CD8+ T-cells and no leukemic cells were detected. TCR gamma assay was done from one patient sample and it showed similar clonal pattern as in peripheral blood. However, despite frequent side-effects, response to dasatinib treatment was very good in all cases, including complete molecular responses in 5 advanced phase patients. Transient clonal LGL lymphocytosis has been described in few patient cases with a primary herpesvirus infection. Our patients share striking pheno/genotypic similarities with LGL leukemia, in which an antigen driven expansion of LGLs precedes the occurrence of oncogenic events. However, in our patients LGL lymphocytosis appeared to be a benign phenomenon. We postulate that by virtue of inhibition of key non-BCR-ABL kinases, dasatinib induces a reversible state of autoimmune-like reactivity upon antigen challenge in a proportion of patients with a distinct genetic and/or environmental background. The aberrant immune response may also result in enhanced anti-leukemic control driven by the cytotoxic T/NK LGL cells. Dasatinib could be a valuable tool in uncovering the pathogenesis of LGL lymphoproliferation, putatively related to mutations in kinases targeted by the drug.