Citations

Abstract Background The role of intratumoral PD-1+ T cells and their contribution to patient outcome in follicular lymphoma (FL) has been controversial. While some studies have found that increased numbers of PD-1+ T cells correlate with an inferior prognosis or have no impact on patient outcomes, other studies observed that the numbers of intratumoral PD-1+ T cells are an indicator of a favorable outcome in FL. In previous work, we observed that PD-1 can be expressed by different subsets of intratumoral T cells in FL including T follicular helper (TFH) cells and exhausted T effector cells. However, the function, prevalence, distribution and clinical importance of the respective intratumoral PD-1+ T subsets in FL patients are largely unknown. Goal To characterize PD-1+ T cell subsets and to determine the biological and clinical relevance of these T cell subsets in FL. Results In an analysis of biopsy specimens from 33 previously untreated FL patients, we found that PD-1 is abundantly expressed on intratumoral T cells. A median of 54.9% (range: 17.2-92.6, n=33) of CD4+ or 45.8% (range: 12.8-81.7, n=33) of CD8+ T cells express PD-1 on the cell surface. We found by flow cytometry that PD-1 is expressed on intratumoral CD4+ T cells with bright or dim intensity that represents two different subpopulations in FL. By immunohistochemistry, we identified that CD4+PD-1bright T cells predominantly reside in the lymph node follicles while PD-1dim T cells are mainly located in an inter-follicular pattern. Intratumoral CD4+PD-1bright T cells have a TFH cell phenotype and express CXCR5, secret IL-21 and are BCL-6 positive with no TIM-3 expression. In contrast, CD4+PD-1dim T cells have an exhausted phenotype, express TIM-3 and do not express BCL-6 and CXCR5. Functionally, CD4+PD-1dim T cells displayed reduced proliferation, cytokine production and cell signal transduction. When the CD8+ T cells were analyzed, we found that intratumoral CD8+ T cells, primarily reside in the inter-follicular regions, express PD-1 with low intensity and exhibit similar phenotypic and functional changes to inter-follicular CD4+PD-1dim T cells. These cells express TIM-3 but lack CXCR5 and BCL-6. They also displayed reduced proliferation and cytokine production. Clinically, we observed that both the numbers of CD4+PD-1dim and CD8+PD-1dim exhausted T cells significantly correlate with a reduced progression-free survival in FL patients treated with single agent rituximab (p=0.007 and p=0.04 respectively; n=31). Although it did not reach statistical significance, CD4+PD-1bright TFH cells appeared to correlate with a favorable survival, suggesting that TFH and exhausted T cells differentially impact patient outcome in FL. Conclusion Taken together, these results indicated that PD-1 expression defines two subpopulations with distinct functions that differentially impact patient outcome in FL. This finding provides not only an explanation for the previously discrepant clinical observations, but is also important in the design of future anti-PD-1 antibody-based therapy in FL. Disclosures: No relevant conflicts of interest to declare.

Abstract Gene expression profiling has shown that diffuse large B-cell lymphoma (DLBCL) clusters into three major subtypes based on similarity in expression patterns to their cell of origin (COO): germinal center B-cell-like (GCB-DLBCL), activated B-cell-like DLBCL (ABC-DBLCL) and primary mediastinal B-cell lymphoma (PMBCL). These subtypes of DLBCLs are associated with distinctly different overall survival rates after standard immunochemotherapy. However, clinical and prognostic heterogeneity remains within COO subsets and strategies are needed to further stratify patients to identify and target high-risk subsets. A comprehensive genomic analysis of COO on a clinically defined set of DLBCL cases has not been performed and the aim of this study was to use whole-exome sequencing (WES) data from 58 paired tumor-normal DLBCL samples to assess association of known DLBCL genomic alterations with cell COO as well as for identification of novel and relevant genetic biomarkers. To investigate genomic alterations associated with DLBCL subsets, we analyzed WES and genome wide copy number data from 58 paired tumor-normal DLBCL tumors. Gene expression profiling or Hans classification was performed to determine DLBCL COO subtype; 31 patients were classified as GCB and 27 as non-GCB. The WES data were used to 1) assess the association of known DLBCL genomic alterations with COO, and 2) identify novel alternations associated with COO. Statistical analysis was performed and the data were ranked by significance (p≤ 0.05) within each DLBCL subtype. In total, 45 genomic abnormalities were analyzed for their association with either GCB, non-GCB or both. Mutations in CREBBP, EZH2, MEF2B, FOXO1 and REL have also been reported as GCB driver mutations and we observed GCB patients with these mutations, but the mutation clustering was not wholly associated with GCB. MYC double-hits were exclusively found in the GCB-subtype group. For the non-GCB cases we found that mutations in MAP2K3 and MYD88 were significantly associated with this subtype(p< 0.05). In addition to mutational patterns, we identified several copy number alterations (CNA) across both groups. Chromosomal losses in GCB patients were found at chromosomes 10q11.21-10q24.23, 4q12-4q35.2, 3q12.1-3q29, 4p12-4p16.3, 10p11.21-10p15.3, and 14q11.2-14q24.3 whereas gains were localized to 7q11.1-7q36.3, 7p11.2-7p22.3, and 1q21-1q32.1 (p < 0.05). No CNA was observed to directly associate with non-GCB patients, however, a loss at 9p21 and gains at 9q24.1 and 18q21.33 trended with the non-GCB subtype, supporting previous reports. Loss at 10q23.31 or a gain in 2p13-2p12 have been reported as being specific for GCB and our data confirmed the association of 10q23.31 with GCB while a gain at 2p13-2p12 (REL)was found in both subtypes. To further understand genomic differences between DLBCL subtypes, we evaluated the relative percentage of each genomic feature. 18/45 (40%) were only observed in GCB patients whereas 2/45 (5%) were specific to the non-GCB subtype. The majority (25/45, 55%) overlapped between the two subtypes. Throughout our analysis we noted that 7 non-GCB cases lacked any of the driver mutations analyzed in the study. While all cases carried mutations, they consisted of low frequency mutations that were not specifically associated with COO. 2/7 cases had a gain at 9p24.1 that included CD274 and JAK2. Because 9p24.1 gains have not been fully defined in DLBCL, we reviewed all cases and identified 4 (7%) with a 9p24.1 gain, 3 of which were non-GCB and 1 GCB. One non-GCB case was EBV+ and none of the cases showed evidence of PMBCL. Of the 9p24.1 cases, three had RNAseq data available and we found that PDL1 and JAK2 expression was elevated (12 fold p< 0.01 and 7 fold p< 0.01, respectively) when compared to the 9p24.1 normal cases (n=32). While outcome was not the focus of the study, we did note that 6/7 cases that lacked driver mutations achieved event free survival at 24 months (EFS24). Taken together, this analysis has further characterized the genetic profile of each COO subtype and has identified novel GCB CNAs which require independent replication. Additionally, we identify a subgroup of non-GCB DLBCL patients that do not harbor known driver mutations and require further genomic study to better resolve the biology of these tumors. Together, these data provide insight on the genetic heterogeneity of DLBCL and identify genetic variants that may inform subtype specific therapy. Disclosures Nowakowski: Morphosys: Research Funding; Celgene: Research Funding; Bayer: Consultancy, Research Funding. Rimsza:NCI/NIH: Patents & Royalties: L.M. Rimsza is a co-inventor on a provisional patent, owned by the NCI of the NIH, using Nanostring technology for determining cell of origin in DLBCL..

Abstract Abstract 1586 Transformation growth factor (TGF-β) is a highly pleiotropic cytokine critical to a variety of cellular events such as cell differentiation and apoptosis. TGF-β is synthesized as a prepro-TGF-β precursor and secreted after being processed in Golgi apparatus as a latent form that non-covalently combines both TGF-β and latency-associated protein (LAP). Our previous work in B-cell NHL has shown that the intratumoral T cell composition results in the establishment of a profoundly inhibitory tumor microenvironment. However, the underlying mechanism is only partially understood. In this study, using patient specimens and lymphoma cell lines, we evaluated the role of TGF-β in the tumor microenvironment and determined the effect of TGF-β on the generation of intratumoral TH1, TH17 and Treg cells in B-cell NHL. First, we determined expression of TGF-β and found that a latent form of TGF-β was specifically expressed on the surface of CD19+ B cells, but not on other types of cells from B-cell lymphoma biopsy specimens. By screening cell lines, we found that latent TGF-β was also expressed on the surface of lymphoma cell lines, confirming the finding. Second, we tested whether surface expression by lymphoma cells led to the secretion of TGF-β in culture medium. Using an ELISA assay, we detected variable levels of latent TGF-β in the culture medium of primary malignant B cells (median 100 pg/ml per million cells, range: undetectable −229 pg/ml, n=7). Similarly, lymphoma cell lines secreted variable amounts of TGF-β from undetectable to 200 pg/ml per million cells. Next, we determined the effect of TGF-b on intratumoral T cell proliferation and differentiation. As expected, exogenous addition of TGF-β inhibited the proliferation of T cells. Notably, the proliferation of intratumoral T cells was significantly reduced when co-cultured with lymphoma cells bearing an active form of TGF-β compared to that with lymphoma cells without TGF-β. Using flow cytometry, we showed that the addition of exogenous TGF-β enhanced Foxp3 expression in activated CD4+, CD4+CD45RA+ or CD4+CD45RO+ intratumoral T cells, suggesting that TGF-β promotes the generation of Treg cells in tumor microenvironment. In contrast, TGF-β suppressed the expression of IFN-γ in activated CD4+ T cells and inhibited the up-regulation of IL-12 and IL-23-induced IFN-γ expression in CD4+ cells, indicating that TGF-β suppresses the generation of TH1 cells. TGF-β alone slightly inhibited IL-17 expression in CD4+ T cells; however, TGF-β, in the presence of IL-6 and IL-23, up-regulated IL-17 expression in CD4+ T cells, suggesting proinflammatory cytokines are able to reverse the suppression induced by TGF-β. These results suggest that TGF-β controls the generation of TH1, TH17 and Treg cells contributing to the imbalance of effector TH cells and inhibitory Treg cells in the tumor microenvironment of B-cell NHL. Since malignant B-cells produce TGF-β, these results further support the important role of malignant B cells in the regulation of intratumoral T cell differentiation and the host immune response. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 2700 Background: Follicular lymphoma is a heterogeneous disease in terms of disease course and prognosis. Clinical risk factors, including age, presence of B symptoms and bulky disease have been shown to have prognostic value. More recently, elements of the tumor microenvironment, including CD68 positive macrophages as well as programmed death-1 (PD-1) expressing lymphocytes, have been shown to be associated with a worse prognosis. However, there is a paucity of data in terms of microenvironment changes that take place in the setting of transformation into diffuse large B cell lymphoma (DLBCL). CD14 is a monocytic marker, and increased numbers of CD14+ cells in the peripheral blood have been associated with a poor prognosis in non-Hodgkin's lymphoma and CLL. It has also been shown that CD14+ cells are present in biopsy specimens in DLBCL and that these cells may support malignant cell growth. This study was designed to examine the presence of CD14+ cells in follicular lymphoma that transformed to DLBCL and to determine the prognostic value of CD14 expression in terms of time to transformation. Methods: Patients with follicular lymphoma who underwent transformation into DLBCL were identified from the Mayo Clinic Lymphoma Database. Controls were follicular lymphoma patients who did not undergo transformation, and were matched 2:1 (control:transformed) on age at diagnosis, year of diagnosis, sex, stage, and length of follow-up time. Tissue specimens at the time of diagnosis were stained for CD14 and CD68, and characterized semiquantitatively based on the presence and distribution of positive cells. Time to transformation (TTT) was assessed using the Kaplan-Meier method. Results: 159 patients were included in the study: 58 patients transformed and 101 did not transform into DLBCL. Median age at diagnosis was 64 years. 150 specimens were sufficient for immunohistochemical analysis. When patients with transformed disease were compared to those who did not transform, those who transformed had greater numbers of CD14+ cells present in their initial diagnostic biopsy (p=0.0415). When only patients who transformed were considered, CD14+ cells in a follicular pattern was associated with a median time to transformation of 3.9 years, CD14+ cells in an interfollicular pattern had a median TTT of 4.9 years and CD14- specimens were associated with a median TTT of 6.2 years (p = 0.0322). The staining pattern for CD68 differed from that of CD14 and was not significantly associated with time to transformation. Conclusions: The presence of intratumoral CD14+ cells at diagnosis in follicular lymphoma is associated with subsequent disease transformation. Our results suggest that when CD14+ cells organize in the follicle, follicular lymphoma patients are likely to transform into DLBCL sooner. Further studies are therefore warranted to define the role of these cells in the tumor microenvironment during transformation to DLBCL. Disclosures: No relevant conflicts of interest to declare.

Abstract T-cell exhaustion plays an important role in attenuating the function of immune cells in B-cell non-Hodgkin's lymphoma (NHL) and PD-1 expression is typically used to identify exhausted T-cells. We have however previously shown that not all PD-1+ cells are exhausted and that PD-1 is differentially expressed on two distinct T-cell subpopulations, with high expression on T follicular helper cells and dim expression on exhausted T cells. Other markers are therefore needed to more clearly identify exhausted intratumoral T cells. To further define exhaustion of intratumoral T cells, we determined the co-expression, regulation and function of PD-1, TIM-3 and LAG-3 on CD4+ or CD8+ T cells by flow cytometry. Using biopsy specimens from follicular B-cell NHL, we found that the percentages of PD-1+ and TIM-3+ T cells were 53.1% (range: 17.2-81.2%, n=32) and 34.5% (range: 14.9-62.6%, n=34) in CD4+ T cells and 46.8% (range: 12.8-81.7%, n=32) and 40.4% (range: 15.0-78.4%, n=34) in CD8+ T cells, respectively. We observed that TIM-3 was predominantly expressed on PD-1dim T cells and TIM-3+ cells accounted for 40% of CD4+ PD-1dim or 45% of CD8+ PD-1dim T cells. Similarly, LAG-3 was variably expressed on intratumoral T cells from B-cell NHL. A median of 9.54% (range: 3.01-15.46, n=6) of CD4+ or 20.48% (7.93-33.9, n=8) of CD8+ T cells express LAG-3. We found that LAG-3+ T cells almost exclusively came from PD-1+ TIM-3+ cells, forming a defined population of intratumoral PD-1+ TIM-3+ LAG-3+ CD4+ or CD8+ T cells. While the majority of LAG-3+ T cells were effector memory T cells (CD45RA- CCR7-), some LAG-3-expressing T cells displayed a phenotype of terminally-differentiated T cells (CD45RA+ CCR7-). Functionally, the intratumoral TIM-3+ LAG-3+ T cells exhibited reduced capacity to produce cytokines (IL-2, IFN-γ) and granules (perforin, granzyme B). Similar to TIM-3, LAG-3 expression was strongly up-regulated on CD4+ or CD8+ T cells by IL-12, a cytokine that has been shown to induce T-cell exhaustion. Interestingly, we observed that while expression of TIM-3 on CD8+ T cells was upregulated by IL-12 at an early time point (day 1), LAG-3 was only induced after TIM-3 up-regulation (day 3) and almost exclusively on TIM-3+ T cells. Furthermore, we found that blockade of both TIM-3 and LAG-3 signaling was able to reverse the exhausted phenotype of CD8+ T cells resulting in increased IFN-γ and IL-2 production. This effect was further enhanced when CD8+ T cells were treated with both anti-TIM-3 and anti-LAG-3 Abs. Taken together, these results suggest that PD-1, TIM-3 and LAG-3 were involved in the induction of exhaustion of T cells in B-cell NHL. We find that PD-1, TIM-3 and LAG-3 are expressed on the same T cells and that blocking TIM-3 and LAG-3 can reverse T-cell exhaustion signaling. These results suggest that PD-1, TIM-3 and LAG-3 play a synergistic role in the development of T cell exhaustion in NHL. Disclosures No relevant conflicts of interest to declare.

Background:T cell Ig and ITIM domain (TIGIT) is an immune checkpoint molecule and a novel member of CD28 family. TIGIT is expressed on NK cells, effector and memory T cells, and Treg cells. Upon ligation with CD155, TIGIT delivers an inhibitory signal and negatively regulates anti-tumor responses. While important in normal T-cell biology, the expression and function of TIGIT in the blood and tumor microenvironment of patients with B-cell non-Hodgkin lymphoma (NHL) is completely unknown. Goal: To phenotypically and functionally characterize TIGIT+T cell subsets in B-cell NHL and compare expression of TIGIT on intratumoral T cells to normal controls. Results: In peripheral blood, TIGIT expression was not detected on resting T cells. In contrast, when we analyzed biopsy specimens from B-cell NHL patients, we observed that TIGIT is variably but highly expressed on a subset of intratumoral T cells. A median of 41.1% (range: 24.9-56.1, n=5) of CD4+ or 22.2% (range: 13.4-33.3, n=5) of CD8+ T cells from tumor specimens expressed TIGIT on the cell surface. In normal tonsil tissue, expression level of TIGIT was modest or negligible. A median of 12.0% (range: 9.1-15.8, n=5) of CD4+ or 3.9% (range: 1.3-5.7, n=5) of CD8+ T cells express TIGIT on the cell surface, which is significantly lower than that in B-cell NHL (p=0.005 and p=0.003 for CD4+ and CD8+, respectively). We observed that both CD45RA+ and CD45RA- T cells express TIGIT with the greatest expression on the CD45RA- population in B-cell NHL specimens. Furthermore, we found that TIGIT+ T cells coexpressed other immune checkpoint molecules including PD-1 and TIM-3. Functionally, TIGIT+ T cells displayed reduced cytokine production, as the number of IFN-γ- and IL-2-producing cells was lower in the TIGIT+ population than in TIGIT-T cells. Unlike TIM-3 that is coexpressed with PD-1 and whose expression is upregulated by IL-12, TIGIT expression is not upregulated by IL-12, suggesting that a different mechanism is involved in TIGIT upregulation in B-cell NHL. Conclusion: Taken together, these results indicated that TIGIT is highly expressed on intratumoral T cells in B-cell NHL and is expressed on a population of T-cells with suppressed immune function. TIGIT, while co-expressed with PD-1 and TIM-3, is not upregulated by cytokines such as IL-12, suggesting that it is regulated by an alternate pathway. Inhibition of TIGIT signaling may be an additional mechanism to prevent T-cell suppression and exhaustion in B-cell NHL. Disclosures No relevant conflicts of interest to declare.

Abstract Abstract 2688 Members of the signal transducers and activators of transcription (STAT) family of proteins function as secondary messengers mediating cellular responses to various cytokines. Aberrant activation of STAT5 protein has been implicated in the pathogenesis of hematologic malignancies due to the ability of these transcription factors to regulate genes involved in cellular proliferation and survival. The two highly homologous transcription factors collectively known as STAT5, namely STAT5A and STAT5B, display both redundant and distinct physiologic functions in non-malignant B cells. Yet, despite evidence suggesting independent functions for STAT5A and STAT5B in the pathophysiology of solid tumors, the precise roles of these isoforms in hematologic malignancies such as Waldenstrom's macroglobulinemia (WM) are not known. Initial immunohistochemical staining of WM bone marrow biopsies (n=6) revealed higher levels of STAT5 phosphorylation compared to bone marrows from normal controls (n=6). Additionally, phosphorylated STAT5 was highly expressed in CD19/CD138+ sorted cells obtained from the bone marrow of patients with WM (average ΔMFI compared to isotype control, 4.64 +/− 1.7, n=6) as determined by FACS analysis. Furthermore, western blotting demonstrated baseline STAT5 phosphorylation in both WM and IgM-secreting cell lines (MWCL-1 and BCWM-1, respectfully). The expression of other phosphorylated STAT proteins (1, 3, 4, and 6) was not detected. To better understand the overall role of STAT5 biologically in WM, MWCL-1 and BCWM.1 cells were treated with a specific inhibitor of STAT5, N′-((4-Oxo-4H-chromen-3-yl)methylene) nicotinohydrazide. Following 72 hours in culture, this inhibitor significantly decreased IgM secretion at lower doses when compared to controls (10 μM, MWCL-1: 9413 +/− 511 ng/mL vs. 6166 +/− 160 ng/mL, p<0.001; BCWM.1: 13,380 +/− 393 ng/mL +/− 10970 +/− 240 ng/mL, p<0.001) and nearly abolished IgM secretion at higher doses. Significant effects on proliferation and viability were also observed, but only at higher concentrations of drug where the specificity toward STAT5 may have diminished. To characterize the roles of the individual STAT5 isoforms in WM, we have developed doxycycline-inducible STAT5A-shRNA and STAT5B-shRNA and used these to create stable knockdowns in the MWCL-1 cell line. Addition of doxycycline to these cell lines for 72h nearly abolishes STAT5A or STAT5B protein expression, respectively, without affecting expression of the other isoform, while no change in STAT5 expression is detected in cells transduced with the empty vector (EV) control. The functional effects of isoform-specific STAT5 inhibition on WM tumor biology were assessed through proliferation assays and ELISA to determine the utility of therapeutically targeting a single STAT5 isoform. No difference in proliferation was observed in the STAT5A- or STAT5B-shRNA cells following 72h induction with doxycycline as compared to the EV control cells. However, specific knockdown of STAT5A following doxycycline induction significantly decreased IgM secretion as compared to the EV control (EV, 91.8 +/− 10.2% of baseline secretion, STAT5A, 66.7 +/− 15.1% of baseline, p<0.05), whereas IgM secretion was decreased by knockdown of STAT5B but not significantly. Similarly, IL-6 secretion was only diminished in cells lacking STAT5A as compared to EV cells (EV, 94 +/− 14.16% of baseline secretion; STAT5A, 64% +/− 7.6% of baseline, p<0.001). Overall these initial data indicate a role for isoform-specific STAT5 activation in the maintenance of WM tumors, and future studies will further our understanding of the specific roles of STAT5A and STAT5B in WM tumors. Disclosures: Ansell: Seattle Genetics, Inc.: Research Funding.

Abstract Abstract 281 Elevated serum levels of soluble interleukin-2 receptor alpha (sIL-2Ra) have been shown to be a poor prognostic factor in various malignancies and sIL-2Ra levels have generally correlated with tumor bulk and advanced stage of disease. In a clinical trial of single agent rituximab in previously untreated follicular lymphoma (FL) patients, we found that sIL-2Ra levels were elevated compared to controls and increased sIL-2Ra levels prior to treatment were associated with a poor outcome. The mean sIL-2Ra level (+/− standard deviation) in untreated FL patients (n=32) was 2.13 ng/ml (+/− 1.81) and 1.09 ng/ml (+/− 1.53) for normal controls (n=16). The time to progression for previously untreated FL patients after 4 doses of rituximab was 12 months for patients with sIL-2Ra levels above the mean compared to 40 months for patients with low sIL-2Ra levels (p=0.003) confirming that elevated sIL-2Ra is a poor prognostic factor in FL. To explore the mechanism by which sIL-2Ra may contribute to a poor prognosis, we determined the source of sIL-2Ra and whether sIL-2Ra facilitates interleukin-2 (IL-2) signaling. Based on our previous work showing that IL-2 induces Foxp3 expression, we particularly evaluated whether sIL-2Ra may bind IL-2 and contribute to impaired anti-tumor immunity by promoting Treg cell development. We found that sIL-2Ra could be detected by ELISA in the culture medium of IL-2Ra-expressing cell lines such as SuDHL1 and Karpas299 and that activation of these cells with PMA/Ionomycin significantly increased sIL-2Ra levels in the culture medium. Production of sIL-2Ra was associated with an attenuation of surface IL-2Ra expression on SuDHL1 and Karpas299 cells suggesting that surface IL-2Ra is the source of sIL-2Ra. Next, using histo-tagged sIL-2R, we tested whether the complex between sIL-2Ra and IL-2 was able to bind to IL-2R on the malignant cell. We found that sIL-2Ra bound IL-2 and that the complex could be detected on the surface of IL-2R-expressing cells. We then compared the signaling of the sIL-2Ra/IL-2 complex to that of IL-2 alone. As expected, we found that IL-2 stimulated cell proliferation, enhanced STAT5 phosphorylation and upregulated Foxp3 expression in CD4+ T cells in a dose-dependent manner. Administration of an anti-IL-2 antibody attenuated these effects. In contrast, sIL-2Ra alone had little effect on cell proliferation, STAT5 phosphorylation and Foxp3 expression. However, the addition of sIL-2Ra to IL-2 prior to incubation with IL-2R-expressing cells enhanced IL-2-mediated phosphorylation of STAT5 and Foxp3 expression in CD4+ T cells. Compared to IL-2 alone, treatment with sIL-2Ra (25ng/ml) and IL-2 (25ng/ml) increases the number of CD4+ T cells with phosphorylation of STAT5 from 7.9% to 17.4%. Taken together, these results indicate that sIL-2R plays an active biologic role in FL by binding IL-2 and promoting IL-2 signaling rather than depleting IL-2 and blocking its function. Given the findings that IL-2 strongly induces Foxp3 expression and promotes Treg cell development and that Treg cells suppress anti-tumor immunity, we conclude that elevated serum levels of sIL-2Ra facilitate IL-2 signaling and thereby contribute to a poor prognosis in FL. Disclosures: No relevant conflicts of interest to declare.

Abstract We have previously shown that CD4+CD25+Foxp3+ regulatory T cells from NHL tumors suppress the function of infiltrating CD4+ T cells and cytolytic CD8+ T cells. Expression of Foxp3 has been demonstrated to be crucial to the development and function of CD4+CD25+ regulatory T cells. However, the mechanistic details that drive development of Foxp3 expression in T cells, in both the normal and malignant scenario, remains to be fully elucidated. Previous studies suggest that Foxp3 expression in CD4+CD25− T cells can be upregulated by tolerizing stimuli such as activation through TCR, corticosteroids, estrogen, and TGF-beta. Because lymphoma B cells have been shown to induce T-cell tolerance, we postulated that lymphoma B cells may play a role in the generation of regulatory T cells by inducing Foxp3 expression in CD4+CD25− T cells. FoxP3 expression was initially thought to be restricted to CD4+CD25+ regulatory T cell population. However, recent literature suggests that Foxp3 may also be expressed in CD4+CD25− T cells. Using biopsy specimens from patients with B-cell NHL, we found that a subset, 15%, of infiltrating CD4+CD25− T cells express Foxp3 and are capable of suppressing the proliferation and granule production of infiltrating cytotoxic CD8+ T cells. These initial studies suggest that CD4+CD25−Foxp3+ T cells have regulatory function. To explore the underlying mechanism by which Foxp3 expression is regulated, we determined the effect of costimulatory signals on Foxp3 expression in CD4+CD25−Foxp3− T cells. Activation with OKT3/anti-CD28 Ab as well as DC-mediated activation induced Foxp3 expression in a subset of CD4+CD25− T cells. We also found that the presence of lymphoma B cells during activation augmented the induction of Foxp3 expression in CD4+CD25− T cells and that NHL B cell-mediated Foxp3 expression was cell contact-dependent. To better understand the contribution of NHL B cells in Foxp3 expression, we explored the possibility that CD27-CD70 interaction may be involved in Foxp3 expression. Lymphoma B cells express CD70, but not B7-1 and B7-2, which have been shown to be important in protecting tumor cells from lysis and contributing to cancer pathogenesis. Ligation of CD27 by receptor cross-linking enhanced Foxp3 expression in infiltrating CD4+CD25− T cells in B-cell NHL. Taken together these studies reveal a novel role for NHL B cells in development of regulatory T cells. Our data show that lymphoma B cells induce expression of Foxp3 in infiltrating CD4+CD25− T cells and may result in development of T cells with regulatory function within the tumor microenvironment. Our results also suggest a potential role for CD27-CD70 interactions in this process. The ability of malignant B cells to drive development of regulatory T cells may be one mechanism by which lymphoma B cells protect themselves from anti-tumor immunity. (Supported in part by the Iowa/Mayo Lymphoma SPORE CA97274).

Abstract Abstract 1554 T cells in the tumor microenvironment influence the biology of malignant cells in many hematologic malignancies, often through cytokine-mediated interactions. Recent studies involving healthy B cells and CD4+T cells identified an interplay between IL-6 and IL-21, whereby IL-6 increased IL-21 production by T cells, driving the differentiation and IL-6 secretion of nearby B cells. In addition to their known effects on healthy B cell function, IL-6 and IL-21 have also been implicated in the pathology of various lymphomas. In Waldenstrom's macroglobulinemia (WM), IL-6 is elevated in the bone marrow and is associated with increased IgM production. However, the function of IL-21 in the WM tumor microenvironment and its relationship to IL-6 is poorly understood. Our objective in this study was to characterize IL-21 production and function in WM and to examine the role of IL-6 and IL-21 in regulating interactions between malignant B cells and T cells in the tumor microenvironment. Immunohistochemistry revealed significant IL-21 staining in bone marrows of patients with WM (n=5), but the areas of infiltration by WM in the bone marrow sections appeared negative for IL-21 staining. To better understand the origin of IL-21 in in the tumor microenvironment, IL-21 expression was assessed by PCR in the CD19−CD138− fraction of cells remaining in patient bone marrow aspirates after positive selection for malignant B cells (n=5). IL-21 transcript was detected in 4/5 samples. CD19−CD138− cells activated with anti-CD3 and anti-CD28 antibodies expressed higher levels of IL-21 transcript and secreted significantly higher levels of IL-21 protein compared to unstimulated cells, suggesting that IL-21 in the WM bone marrow is derived from activated T cells. Intracellular expression of IL-21 protein was confirmed in CD4+ and CD8+ cells within the CD19−CD138− population using flow cytometry. Furthermore, dual staining of WM bone marrow sections with antibodies against IL-21 and CD3 or CD20 revealed co-staining of IL-21 with CD3+ T cells but not with CD20+ B cells. The response of WM B cells to T-cell derived IL-21 was then assessed in positively selected CD19+CD138+ WM B cells (n=5) and in the MWCL-1 cell line. Using flow cytometry, both the IL-21 receptor and the required common gamma chain subunit were detected on all patient samples as well as on MWCL-1 cells. Treatment of MWCL-1 cells with IL-21 (100 ng/mL) for 72 h increased proliferation by 35% (p<0.05) and IgM secretion by 80% (p<0.005). Similarly, in primary CD19+CD138+ WM cells (n=5), proliferation increased on average by 38% and IgM secretion by 71%. No apoptotic effects were associated with IL-21 in WM. Characterization of STAT activation in response to IL-21 revealed significant phosphorylation of STAT3 in both CD19+CD138+ WM cells and MWCL-1 cells and was associated with increases in BLIMP-1 and XBP-1 protein and decreases in PAX5. As STAT3 activation is known to regulate IL-6, we assessed the effect of IL-21 on B cell-mediated IL-6 secretion using ELISA. IL-21 significantly increased IL-6 secretion by both primary CD19+CD138+ WM cells (n=4) and MWCL-1 cells (87.9 +/− 10.9 ng/mL vs. 297.8 +/− 129.2 ng/mL, p<0.05). Treatment with IL-6 and IL-21 together had no additional effect over IL-21 alone on proliferation or IgM secretion in MWCL-1 cells, but culturing anti-CD3/anti-CD28-activated CD19−CD138−cells from WM bone marrows with IL-6 significantly increased IL-21 secretion (n=3). Overall, these data indicate that T-cell derived IL-21 significantly promotes growth and immunoglobulin production by malignant WM B cells and that subsequent IL-6 secretion by malignant B cells may enhance the secretion of IL-21 by T cells within the bone marrow microenvironment. Disclosures: No relevant conflicts of interest to declare.