Inhibition of antigen-dependent B-cell receptor (BCR) signaling is considered a promising therapeutic approach in chronic lymphocytic leukemia (CLL), but experimental in vivo evidence to support this view is still lacking. We have now investigated whether inhibition of BCR signaling with the selective Syk inhibitor fostamatinib disodium (R788) will affect the growth of the leukemias that develop in the E-TCL1 transgenic mouse model of CLL. Similarly to human CLL, these leukemias express stereotyped BCRs that react with autoantigens exposed on the surface of senescent or apoptotic cells, suggesting that they are antigen driven. We show that R788 effectively inhibits BCR signaling in vivo, resulting in reduced proliferation and survival of the malignant B cells and significantly prolonged survival of the treated animals. The growth-inhibitory effect of R788 occurs despite the relatively modest cytotoxic effect in vitro and is independent of basal Syk activity, suggesting that R788 functions primarily by inhibiting antigendependent BCR signals. Importantly, the effect of R788 was found to be selective for the malignant clones, as no disturbance in the production of normal B lymphocytes was observed. Collectively, these data provide further rationale for clinical trials with R788 in CLL and establish the BCR-signaling pathway as an important therapeutic target in this disease. (Blood. 2010;116(23):4894-4905)

Key Points• Cell autonomous BCR interactions and interactions with low-affinity autoantigens drive leukemia development in an in vivo model of CLL.• BCR signals induced by binding to external antigen can increase the aggressiveness of CLL.Chronic lymphocytic leukemia (CLL) is a common B-cell malignancy characterized by a highly variable course and outcome. The disease is believed to be driven by B-cell receptor (BCR) signals generated by external antigens and/or cell-autonomous BCR interactions, but direct in vivo evidence for this is still lacking. To further define the role of the BCR pathway in the development and progression of CLL, we evaluated the capacity of different types of antigen/BCR interactions to induce leukemia in the Em-TCL1 transgenic mouse model. We show that cell autonomous signaling capacity is a uniform characteristic of the leukemia-derived BCRs and represents a prerequisite for CLL development. Low-affinity BCR interactions with autoantigens generated during apoptosis are also positively selected, suggesting that they contribute to the pathogenesis of the disease. In contrast, high-affinity BCR interactions are not selected, regardless of antigen form or presentation. We also show that the capacity of the leukemic cells to respond to cognate antigen correlates inversely with time to leukemia development, suggesting that signals induced by external antigen increase the aggressiveness of the disease. Collectively, these findings provide in vivo evidence that the BCR pathway drives the development and can influence the clinical course of CLL. (Blood. 2015;125(10):1578-1588

The PI3K/Akt pathway is activated in response to various microenvironmental stimuli that regulate the survival and proliferation of chronic lymphocytic leukemia (CLL) B-cells, including triggering of the B-cell receptor (BCR). Although this pathway is frequently targeted in cancer, no significant alterations have yet been identified in CLL. We now show that the phosphatase PH domain leucin-rich repeat protein phosphatase (PHLPP1), a recently identified tumor suppressor and negative regulator of the Akt kinase, is absent or expressed at substantially reduced levels in CLL B-cells. To determine what the consequences of PHLPP1 loss on BCR signaling are, we downregulated or re-expressed PHLPP1 in lymphoma cell lines and primary CLL B-cells, respectively. Downregulation of PHLPP1 increased BCR-induced phosphorylation and activation of the Akt, GSK3 and ERK kinases, whereas re-expression had the opposite effect. Importantly, re-expression of PHLPP1 in primary CLL cells prevented upregulation of Mcl-1 and inhibited the increase in leukemic cell viability induced by sustained BCR engagement. Enforced expression of PHLPP1 also affected the response to other microenvironmental stimuli, particularly in terms of ERK phosphorylation. Collectively, these data show that CLL cells lack an important negative regulator of the Akt and ERK pathways, which could confer them a growth advantage by facilitating the propagation of crucial microenvironment-derived stimuli.

The PI3K/AKT pathway plays a central role in regulating cellular growth and survival. This pathway is activated by signals derived from various receptors and is tightly regulated through the action of several phosphatases, including SHIP and PTEN, which hydrolyze the PI3K product PIP3, and the recently identified PHLPP, which directly dephosphorylates AKT. Hyperactivation of the PI3K/AKT pathway has been implicated in the pathogenesis of many types of cancer, including chronic lymphocytic leukemia (CLL) and B-cell lymphoma. In addition, gene expression profiling and real-time RT/PCR analysis have recently shown differential expression of PHLPP mRNA in CLL subsets classified according to the presence of the 13q14 abnormality, with many CLL cases demonstrating absent PHLPP expression altogether. These findings prompted us to compare the levels of PHLPP expression in primary CLL B-cells (n=17) with normal tonsillar B-cells (n=4) and various lymphoma cell lines, including the diffuse large B-cell lymphomas (DLBCL) DHL-4, DHL-6, DHL-8, DHL-10, WSU, Toledo, Ly1, Ly3, Ly7 and Ly18, the Burkitt’s lymphoma BJAB and the prolymphocytic leukemia MEC1. Immunoblotting analysis revealed abundant and uniform expression of PHLPP in normal B-cells and in 7 out of 12 investigated lymphoma cell lines. Higher levels were observed in the BJAB, Ly1 and Ly18 cell lines, whereas PHLPP was undetectable in the DLBCL cell lines WSU and Toledo. Remarkably, PHLPP was either not expressed or was expressed at markedly reduced levels in all of the investigated CLL samples, with levels of expression ranging from 0 to 10% of the levels in normal B-cells. In contrast, the levels of expression of the phosphatase SHIP were relatively similar between CLL and normal B-cells. To determine what are the consequences of reduced PHLPP expression on signaling through AKT in malignant B-lymphocytes, we downregulated PHLPP in BJAB and DHL-4 cells by RNA interference. A significant reduction in the levels of PHLPP was achieved in both cell lines, which amounted to 20–40% of the levels in cells transfected with the control siRNA. Immunoblotting analysis of protein extracts from cells transfected with PHLPP and control siRNA did not show a difference in AKT phosphorylation on Ser473 and Thr308, indicating that a reduction in PHLPP expression is not sufficient to augment basal AKT activity. To determine the effects of PHLPP downregulation on agonist-induced AKT activation, we investigated phosphorylation on Ser473 and Thr308 in BJAB and DHL-4 cells stimulated through the B-cell receptor. In both cell lines downregulation of PHLPP resulted in more than a 50% increase in BCR-induced AKT phosphorylation. In contrast, phosphorylation of other signaling molecules that are also activated by BCR crosslinking, such as PLCγ2 and ERK, appeared unaffected by PHLPP downregulation. These data confirm the functional relevance of PHLPP in AKT regulation in B-lymphoid cells and implicate reduced or absent PHLPP expression in CLL B-cells as a potential determinant of BCR-induced AKT signaling in CLL.

887 CLL B-cells depend on various signals from the microenvironment for survival and proliferation. Among these, antigenic stimuli that are propagated through the B-cell receptor (BCR) are considered particularly important for the development and progression of CLL, suggesting that the BCR signaling pathway could be an important target for therapeutic intervention. We have previously characterized some of the critical components of the signaling pathway downstream of the BCR in CLL B cells and identified the protein tyrosine kinase Syk as a promising therapeutic target. In a recent study we showed that CLL B-cells frequently have increased basal/constitutive Syk activity and are moderately sensitive to the cytotoxic effect of the selective Syk inhibitor R406 [Gobessi et al, Leukemia 2009]. More importantly, the survival signal induced by sustained BCR engagement was completely abolished by R406, suggesting that this compound may exert an even greater effect in vivo by inhibiting antigen-dependent Syk activation. We have now tested this possibility in the Eμ-TCL1 transgenic mouse model of CLL. Aged Eμ-TCL1 mice develop CD5+ B-cell leukemias that, similar to aggressive human CLL, show features of an antigen-driven process, including expression of stereotyped BCRs and reactivity with common autoantigens and microbial agents [Yan et al, Proc Natl Acad Sci USA 2006]. For our experiments we used a TCL1 leukemia (TCL1-002) that does not grow in vitro, but can be propagated in syngeneic recipients in vivo. TCL1-002 cells express an unmutated stereotyped BCR encoded by the VH12/VK4 combination, which reacts with phosphatidylcholine, an autoantigen exposed on the surface of senescent erythrocytes. In vitro experiments showed that R406 is not cytotoxic for TCL1-002 cells, although it completely inhibited both the basal and BCR-induced activation of signaling pathways downstream of Syk. The absence of a direct cytotoxic effect provided a unique opportunity to investigate whether inhibition of BCR signaling will affect leukemia growth in vivo. For this purpose, 1×107 TCL1-002 cells were injected intraperitoneally in 18 syngeneic mouse recipients. Three days later treatment was started in 8 mice with R788, which is the water-soluble prodrug of R406, at a daily dose of 80mg/kg during 18 consecutive days. Because of the rapid clearance of the drug (serum half-life <2 hours) R788 was administered in 3 divided doses at 4 hour intervals. Two weeks after the end of treatment leukemia developed in all mice from the control group (median WBC counts 131×106/ml, range 12-300×106/ml), whereas all R788-treated mice showed normal WBC numbers (median 6×106/ml, range 3-8×106/ml, P<0.001). Three weeks later all mice in the control group had died (median survival 46 days), whereas all mice in the R788 group were still alive and only two of them had detectable leukemic cells. R788 also showed some efficacy in the treatment of mice with overt TCL1-002 leukemias (WBC >50×106/ml). Whereas all mice from the control group (n=9) died between 6 and 18 days from the beginning of therapy, 4 out of 9 mice from the R788 group survived for more than 33 days. The mechanism of R788 activity was primarily related to inhibition of leukemic cell proliferation, as evidenced by a substantial decrease in the percentage of Ki67-positive cells after 7 days of treatment (30% before, 5% after therapy, P<0.001). To investigate whether R788 will also be effective against other TCL1 tumors we treated five TCL1 mice with preleukemic mono- or oligoclonal B-cell expansions during a four week period. R788 reduced the percentage of CD5+/B220+ cells in 2 cases, whereas in 2 other cases the percentage increased. Interestingly, the pattern of clonal Ig gene rearrangements changed during therapy, suggesting that only certain TCL1 clones are sensitive to R788 treatment. In summary, this study shows that R788 can effectively inhibit the growth of certain TCL1 tumors and provides the first in vivo experimental evidence suggesting that inhibition of antigen-dependent BCR signaling could be an effective therapeutic approach in CLL. Disclosures: No relevant conflicts of interest to declare.

181 Studies conducted over the past decade have revealed a strong association between the mutational status of the immunoglobulin heavy-chain variable region (IGHV) genes and clinical course in patients with chronic lymphocytic leukemia (CLL). In patients with aggressive CLL, the leukemic cells typically express B cell receptors (BCRs) encoded by unmutated IGHV genes, whereas these genes are most often mutated in leukemic cells from patients with indolent disease. The mutational status of the IGHV genes reflects features of the antigen, such as antigen structure, form, presentation and affinity, indicating that the difference in the clinical course between IGHV-unmutated and IGHV-mutated CLL could be due to recognition of different types of antigens. In line with this possibility, recent studies have shown that IGHV-unmutated CLL (U-CLL) cells frequently express polyreactive BCRs that bind with low affinity to both microbial antigens and autoantigens translocated or exposed on apoptotic cells, whereas such reactivity is infrequent in IGHV-mutated CLL (M-CLL). To further explore the possibility that the clinical course in CLL is determined by the availability of particular types of antigenic stimuli, we investigated the impact of different antigen/BCR interactions on leukemia development and behavior in the Eμ-TCL1 transgenic mouse model of CLL. We initially established three cohorts of Eμ-TCL1 transgenic mice that expressed transgenic BCRs with different antigen specificity. Two of these cohorts expressed low-affinity unmutated transgenic BCRs reactive with the antigens phosphatidylcholine (PtC) and Sm (IgPtC and IgSm, respectively), whereas the third cohort expressed a high-affinity mutated transgenic BCR (IgHEL) specific for the antigen hen egg lysozyme (HEL). Of note, Sm is a ribonucleoprotein complex that is translocated to the surface of apoptotic cells and has been shown to be recognized by certain human U-CLL BCRs, whereas PtC is a cell membrane component that is exposed on senescent red blood cells and gut bacteria. Because no data are currently available regarding the reactivity of the M-CLL BCRs, we subdivided the cohort of Eμ-TCL1/IgHEL double transgenic mice into four additional cohorts. These included a cohort without antigen (Eμ-TCL1/IgHEL), a cohort in which HEL was provided as a foreign antigen (Eμ-TCL1/IgHEL double transgenic mice repetitively immunized with particles coated with HEL and CpG oligonucleotides), a cohort in which HEL was provided as a soluble autoantigen (Eμ-TCL1/IgHEL/sHEL triple transgenic mice) and a cohort in which HEL was provided as a membrane-bound autoantigen exposed on apoptotic cells (Eμ-TCL1/IgHEL/mHEL-KK triple transgenic mice). Each cohort consisted of 12–14 animals, of which at least 8 have been followed for >1 year. Animals from all cohorts developed CD5-positive B cell leukemias, but only in Eμ-TCL1/IgSm and Eμ-TCL1/IgPtC mice the leukemic cells expressed a transgenic BCR. In Eμ-TCL1/IgHEL mice the leukemias were always derived from the small percentage of B cells that express an endogenous BCR, whereas B cells that express the transgenic IgHEL BCR were never transformed. Interestingly, leukemia development and progression was more rapid in Eμ-TCL1/IgPtC than Eμ-TCL1/IgSm transgenic mice (7/14 at 6 months of age and 2/10 at 8 months of age, respectively). Since PtC is expressed as both a foreign- (gut flora) and self- (senescent red blood cells) antigen, we investigated whether suppression of gut flora will affect the growth of adoptively transferred Eμ-TCL1/IgPtC leukemias. Pretreatment of syngeneic recipient mice with a three-week course of broad-spectrum antibiotics significantly delayed leukemia growth, suggesting that PtC is more potent in driving the expansion of the leukemic clone when expressed as a foreign than self antigen. To summarize, these data demonstrate that U-CLL can be induced by both microbial antigens and autoantigens exposed on apoptotic cells, including autoantigens that are recognized by human CLL cells, such as Sm. In contrast, M-CLL can not be induced by chronic or repetitive antigen stimulation, regardless whether the antigen is provided as a foreign antigen, as a soluble autoantigen, or as a membrane-bound autoantigen exposed on apoptotic cells. Collectively, these data suggest that the mechanisms that drive U-CLL and M-CLL are different and indicate that only U-CLL is an antigen-driven disease. Disclosures: No relevant conflicts of interest to declare.