ALBERT

Description: NOX-A12 is a novel mirror-image oligonucleotide (Spiegelmer®) that specifically antagonizes the chemokine stromal cell-derived factor 1 (SDF-1, CXCL12). SDF-1 is highly expressed by stromal and endothelial cells, and binding to its classical receptor CXCR4 leads to the activation of multiple signaling pathways regulating cell migration, retention and survival. SDF-1-induced inside-out activation of the integrin VLA-4 (a4ß1, CD49d/CD29) via CXCR4 is essential for homing of various normal and malignant hematopoietic cells to bone marrow (BM). We previously found that VLA-4 is indispensable for BM homing of chronic lymphocytic leukemia (CLL) cells. Hence, CXCR4 binding to SDF-1 may be a prerequisite for VLA-4 mediated trafficking of CLL cells into the protective BM microenvironment and for their retention in lymphoid organs, favoring their survival and chemotherapy-resistance. In this project we compared the efficacy of NOX-A12 and the CXCR4 antagonist AMD3100 to mobilize CLL cells from their protective niches to the peripheral blood, thereby sensitizing them to cytotoxic agents such as fludarabine, and to prevent their recirculation into the protective lymphoid microenvironment. To assess the contribution of SDF-1/CXCR4 signals to homing of CLL cells, we performed in vivo short term adoptive transfers of human VLA-4+ CLL cells into immune deficient mice. CLL cells (5-10 x 106) were injected into the tail vein of NOD/SCID mice that were treated either with NOX-A12 (20 mg/kg) or AMD3100 (5 mg/kg). After a homing period of 3 hours lymphatic organs were analyzed for human CLL cell count. We found that for the majority of investigated CLL samples (except patient samples bearing the chromosomal aberration trisomy 12), CXCR4/SDF-1 inhibition reduced CLL cell recirculation into BM, thereby redirecting them into the spleen. The mean (± SD) relative homing rate (homed human CLL cells relative to injected cells per 1 x 106 acquired murine cells) of CLL cells to BM decreased upon NOX-A12 treatment from 137 (± 87) to 30 (± 17) (p = 0.047), and upon AMD3100 treatment from 137 (± 71) to 38 (± 24) (p = ns). Further, we determined CXCR4 surface levels of CLL cells before injection and after cells had homed to BM. We noticed that homed CLL cells expressed elevated CXCR4 surface levels as mean MFIR (± SD) rose from 25 (± 13) to 93 (± 11) and that CXCR4 levels were further increased upon CXCR4/SDF-1 inhibition with NOX-A12 to MFIR 225 (± 2) and with AMD3100 to MFIR 213 (± 7). Next, we used the Tcl1 transgenic transplantation model that mimics human CLL to test NOX-A12 and AMD3100-induced CLL cell mobilization. C57BL/6J wild type mice engrafted with splenocytes (1 x 107) from Tcl1tg mice with established disease were treated with separate doses of NOX-A12 (0, 10, 20 and 50 mg/kg) by single intravenous injections. Separate doses of AMD3100 (0, 3, 5, and 10 mg/kg) were administered subcutaneously. Blood was collected from the tail vein before treatment, 1 hr, 3 hrs, 6 hrs, and 24 hrs post treatment and analyzed for absolute leukocyte counts and numbers of CD5+/CD19+cells. NOX-A12 demonstrated a considerable potential to mobilize CLL cells and other lymphocytes as the relative CLL cell and absolute lymphocyte blood count raised within 1 hour by 50% when using mice with high tumor load and by 200% when using mice with lower tumor load. Mobilization reached a peak after 6 hours with 100% and 220% relative increase to 0 hrs value. Compared to untreated control mice, CLL cell blood count raised from 80% to 120% within 6 hours. Even low doses (10 mg/kg) of NOX-A12 showed a high degree of mobilization of CLL cells and total lymphocytes. Higher doses lead to a prolonged mobilization duration. AMD3100 treatment did not lead to a significant increase in CLL cell or total lymphocyte blood count in peripheral blood. In summary our data demonstrate a high efficiency of NOX-A12 to mobilize CLL cells and to prevent their recirculation into protective lymphatic organs. Abrogating SDF-1/CXCR4 signaling in combination with chemotherapy may provide an attractive approach for treatment of CLL. Currently, the effect of combined NOX-A12 and fludarabine treatment on progression and overall survival is evaluated using the Tcl1 transgenic transplantation model. Disclosures: Kruschinski: Noxxon: Employment. Greil:NOXXON Pharma AG: Research Funding. Hartmann:NOXXON Pharma AG: Research Funding.

Description: Abstract 90 Introduction: The p53 gene is non-functional in 〉50% of human tumors. In mice deletion of p53 leads to a high incidence of tumors and to a significant acceleration of tumorigenesis induced by repeated gamma-irradiation. While a large number of effects have been described for p53, current concepts of p53-mediated tumor suppression discuss the roles of p53 in regulation of cell cycle and apoptosis as being essential. Two main targets have been identified in this respect: p21Waf1 as an essential regulator of cell cycle arrest downstream of p53 and Puma as the largest single contribution towards p53 induced cell death. Methods: We have generated p21Waf1/Puma doubly deficient (i.e. double-knockout – DKO) mice on a pure C57BL/6 background to investigate the effects on tumorigenesis. Results: In ex vivo irradiation studies DKO thymocytes expectedly showed reduced cell death and loss of a G1/S arrest upon irradiation. When following a cohort of mice for spontaneous tumor development, the DKO mice did not differ from wild-type (WT) controls. Since this may be explained by additional p53 down-stream effectors essential for tumor suppression, we set out to challenge the mice with an established repeated irradiation protocol (4 × 1.75 Gy over 4 weeks) in order to increase the likelihood of uncovering a defect in tumor suppression not apparent in unchallenged mice. While irradiated WT mice developed thymic lymphomas at an expected rate and p53 deficiency accelerated the lymphoma formation as published, irradiated DKO mice did not develop any thymic lymphoma at all. During the irradiation protocol WT mice followed a series of depletion and regrowth cycles in thymic cellularity with a high rate of cell death early post irradiations in TUNEL assays and a surge of proliferation on day 5 after irradiations detected by in vivo BrdU labeling. By contrast in DKO mice thymic cellularity dropped only slightly during the first irradiation cycle. This was followed by a slow and steady decline in cellularity over the following 3 cycles of irradiation. No late apoptotic wave or loss of proliferative capacity of remaining thymocytes could explain the loss of cellularity, nor could senescence of thymocytes be detected by SA-β-Gal staining in situ, suggesting that thymic influx was defective. It had previously been reported for the repeat-irradiation lymphomagenesis model, that the irradiation of hemopoietic precursor cells was essential for tumorigenesis. In contrast to thymic cellularity, DKO LSK numbers stayed relatively stable over the course of the 4 irradiations. By comparison WT LSK numbers dropped to about 50% by the time 4 irradiations were completed. Indeed, short-term repopulating (ST) cells dropped significantly, while long-term repopulating (LT) and multipotent progenitor (MPP) cell populations stayed more stable. In DKO marrows the relative content of LT, ST and MPP cells proved very stable across the irradiation schedule. In vivo BrdU labelling showed that WT LSK had a higher fraction of labelled cells at baseline and a 〉100% increase in the proliferative fraction during irradiation, while in DKO LSK the proliferation index was lower and stayed stably low over time, compatible with the replenishment defect observed in the thymus. DKO stem cells were only slightly more efficient (1.6-fold) than WT in bone marrow reconstitution experiments without challenge. However, when mixed chimeras were then subjected to the irradiation protocol with 4 × 1.75 Gy a clear advantage of the DKO cells became apparent (28-fold). Moreover, when reconstituting lethally irradiated mice with a mixture of WT and DKO marrow taken from repeatedly irradiated donors the efficacy ratio was 1:152. Conclusion: Our data contrast observations made in cell lines, where loss of Puma and p21Waf1 led to a p53-resistant outgrowth of cells. We present in an animal model that loss of Puma and p21Waf1 is not tumorigenic and in fact protects mice from irradiation carcinogenesis. Together with our recently published findings in irradiated Puma singly-deficient mice (Labi G&D 2010), our data suggest that tumorigenesis in irradiated DKO mice is inhibited by effects on hemopoietic stem cell reactivity to DNA damage. A combination of lack of generation of free niche space through protection of hemopoietic stem cells from cell death and a stem cell quiescence state retained in DKO stem cells after irradiation seems responsible for the phenotype. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 803 Introduction: The development of B cell chronic lymphocytic leukaemia (CLL) involves antigenic selection of the B cell clone, as evidenced by a skewed BCR repertoire found in CLL cells, as well as continuing BCR signaling detected in subgroups of CLL. T cell involvement in the maintenance of the malignant clone is also suggested by a number of evidential lines, such as the severe skewing of T cell subsets in CLL and the presence of relevant numbers of CD4 cells in the so-called “proliferation centers” in lymph nodes. However, neither clonal nor immunologic identities of these T cells have been sufficiently determined. Recently, we have shown in Tcl1 transgenic mice that CLL clones were able to directly drive changes in the T cell repertoire, resulting in relatively fast skewing in subset distribution (similar to that observed in human CLL) as well as a clonal selection of T cells, both in spontaneously developing CLL in the model, as well as in immunocompetent congenic recipient mice in experiments transplanting established murine CLL. Methods: To establish evidence for similar interactions in human CLL, we collected a descriptive database determining patterns of T cell diversity in CLL patient blood. We analyzed T cells from 53 previously-untreated CLL patients established TCR V beta clonality and frequency based on CDR3 length polymorphism in sorted CD4 cells and, in a confirmatory subset, based on TCR V gene-specific flow cytometry. CLL samples were also investigated regarding the BCR VH gene usage, mutation status and clinical and prognostic parameters. Results: BCR analysis confirmed antigenic selection in our patient set with 6 IgVH genes accounting for 〉 50% of the cases. Stereotyped CDR3 regions were common and one third of the patients showed unmutatetd IgVH gene sequence. Analysing 20 TCR genes by PCR spectratyping, the TCR CDR3 size distribution pattern revealed a relevant frequency of oligoclonal/monoclonal CD4 T cells in CLL samples. While a number of patients showed completely polyclonal patterns in all their TCR CDR3 regions, others showed either single or multiple clonal TCR families. In corresponding flow cytometric analyses clonal T cells could make up to 48% of total CD4 cells in a given patient. In patients with longitudinal samples we found relevant stability of these TCR patterns over time. By comparing the data from the TCR clonality database with our BCR dataset we established that these clonal patterns significantly, but not exclusively, clustered in unmutated patient samples. Overall we found no strong association with any specific IgVH gene, but intriguingly, in some patients a specific clonal TCR corresponded to a stereotyped BCR receptor. In fact in two patient pairs with identical IgVH rearrangements we found corresponding TCR clones that showed sequence identity between the CD4 T cell clones derived from the other patient. In addition in both pairs we found a shared HLA DR and DQ haplotype. These data strongly suggests that there may be an important link between the antigenic selection on the B-CLL clone and the selection of certain TCR clones, thus for the first time postulating an antigentic identity of the CLL-associated T cells. This proposed identity, however, currently remains unclear. Finally, we tested for an influence of T cell clonality on clinical behaviour of CLL disease in the patients. We could intentify the presence of more than one clonal TCR family as a significant predictor of a short treatment-free interval (p=0.03). This was true for both, patients with mutated and unmutated IgVH receptors, although it remained a trend in the latter. Conclusion: Our results imply that a restricted CD4 T cell diversity may be important for CLL progression and that an as of yet still undefined antigenic drive for T cells may be important for this. This may help to define specific monoclonal CD4 T cells as a promising novel target for future therapeutic intervention. Disclosures: No relevant conflicts of interest to declare.