Search

TPS7098 Background: Acquired mutations in oncogenic kinases remains an obstacle between valid therapeutic hypotheses and meaningful patient benefit. The inhibition of FLT3 can be efficacious in AML...

TPS8604 Background: Immunotherapy is a promising treatment modality for low grade non-Hodgkin’s lymphomas. SD-101 (Dynavax Technologies) is an immunostimulatory synthetic CpG molecule that activate...

The therapeutic efficacy of some anti-tumor monoclonal antibodies (mAbs) depends on the capacity of the mAb to recognize the tumor-associated antigen and induce cytotoxicity via a network of immune effector cells. This process of antibody-dependent cell-mediated cytotoxicity (ADCC) against tumor cells is triggered by the interaction of the fragment crystallizable (Fc) portion of the mAb with the Fc receptors on effector cells like natural killer cells, macrophages, γδ T cells, and dendritic cells. By augmenting ADCC, the antitumor activity of mAbs can be significantly increased. Currently, identifying and developing therapeutic agents that enhance ADCC is a growing area of research. Combining existing tumor-targeting mAbs and ADCC-promoting agents that stimulate effector cells will translate to greater clinical responses. In this review, we discuss strategies for enhancing ADCC and emphasize the potential of combination treatments that include US Food and Drug Administration-approved mAbs and immunostimulatory therapeutics., Video abstract

BackgroundWhile lymph node metastasis is among the strongest predictors of disease-free and overall survival for patients with breast cancer, the immunological nature of tumor-draining lymph nodes is often ignored, and may provide additional prognostic information on clinical outcome.Methods and findingsWe performed immunohistochemical analysis of 47 sentinel and 104 axillary (nonsentinel) nodes from 77 breast cancer patients with 5 y of follow-up to determine if alterations in CD4, CD8, and CD1a cell populations predict nodal metastasis or disease-free survival. Sentinel and axillary node CD4 and CD8 T cells were decreased in breast cancer patients compared to control nodes. CD1a dendritic cells were also diminished in sentinel and tumor-involved axillary nodes, but increased in tumor-free axillary nodes. Axillary node, but not sentinel node, CD4 T cell and dendritic cell populations were highly correlated with disease-free survival, independent of axillary metastasis. Immune profiling of ALN from a test set of 48 patients, applying CD4 T cell and CD1a dendritic cell population thresholds of CD4 > or = 7.0% and CD1a > or = 0.6%, determined from analysis of a learning set of 29 patients, provided significant risk stratification into favorable and unfavorable prognostic groups superior to clinicopathologic characteristics including tumor size, extent or size of nodal metastasis (CD4, p < 0.001 and CD1a, p < 0.001). Moreover, axillary node CD4 T cell and CD1a dendritic cell populations allowed more significant stratification of disease-free survival of patients with T1 (primary tumor size 2 cm or less) and T2 (5 cm or larger) tumors than all other patient characteristics. Finally, sentinel node immune profiles correlated primarily with the presence of infiltrating tumor cells, while axillary node immune profiles appeared largely independent of nodal metastases, raising the possibility that, within axillary lymph nodes, immune profile changes and nodal metastases represent independent processes.ConclusionThese findings demonstrate that the immune profile of tumor-draining lymph nodes is of novel biologic and clinical importance for patients with early stage breast cancer.

This study discusses quality of life in post-socialist Mongolia. Yadargaa, a fatigue-related illness in traditional Mongolian medicine, results from lifestyle imbalance. We examine the distribution of yadargaa and its association to socioeconomic changes under capitalism. Ethnographic interviews concerning yadargaa were conducted with health professionals, yadargaa patients, and laypersons. Epidemiological methods were used to identify risk groups, to estimate the point prevalence, and to assess the distribution of meanings and interpretations of yadargaa. The epidemiological sample included 194 individuals, half urban and half rural. Nearly half of the epidemiological sample suffered from yadargaa (49%). These yadargaa sufferers felt that they benefited less than non-yadargaa subjects from the current socioeconomic changes. Among these, perceived change in employment opportunities was one of the best predictors of yadargaa. Additionally, yadargaa sufferers were predominantly women, the elderly, and urban residents. Yadargaa varies greatly in presentation; Western psychiatric categories are only able to explain half of yadargaa cases. In conclusion, yadargaa strongly associates with disenfranchised groups in the capitalist economy. As a culturally constructed indicator of quality of life, yadargaa is a window into the lives of women and men in post-socialist Mongolia.

Monoclonal antibodies (mAb) induce tumor regression through antibody-dependant cellular cytotoxicity (ADCC). We recently showed that an agonistic anti-CD137 mAb stimulates natural killer (NK) cells which have been activated by a tumor-specific mAb, resulting in increased ADCC against cancer cells.

Meeting abstracts The naturally occurring oncolytic virus, reovirus, exhibits cytotoxic effects on cancer cells. Reovirus is safe, and currently is in multiple clinical trials testing for the treatment of different cancers. NK cells are innate immune effectors that mediate antibody dependent

Abscopal effects, systemic tumor regression following localized therapy, are induced by radiation therapy and augmented with intratumoral immunostimulation. Based on a preclinical lymphoma model, we previously investigated low-dose immunostimulation with a Toll-like receptor 9 (TLR9) agonist in combination with fractionated, low-dose radiation therapy in relapsed/refractory NHL (NCT00185965) and Mycosis Fungoides (NCT00226993). In an attempt to improve the potency of the immune responses and the rate of clinical responses, the dose of CpG was increased 3-fold and enrollment broadened to include treatment-naive as well as relapsed/refractory low-grade lymphoma (NCT00880581). We treated 15 treatment-naive patients and 15 relapsed/refractory patients with follicular lymphoma using low-dose radiotherapy to a single tumor site followed by 18mg of the C-G enriched, synthetic oligodeoxynucleotide (CpG) TLR9 agonist, PF-3512676 injected at the same site, with injections repeated 10 times weekly. Clinical responses were assessed at distant, untreated tumor sites. Immune responses were evaluated by measuring T-cell activation after in vitro re-stimulation with autologous tumor cells. The in situ vaccination with escalated-dose CpG was well tolerated with 16 cases of grade 1 to 2 local or systemic reactions including 2 cases of possibly-related autoimmune disease and no treatment-limiting adverse events. Among treatment-naive and relapsed/refractory patients, four and three patients, respectively, had partial responses with median duration of response of 29 and 12 weeks, respectively. Two and four patients, respectively, had stable disease of duration greater than one year with median time to best clinical benefit among patients with a response or stable disease of 31 and 12 weeks. The range of time to best response was broad, from 10 to 184 weeks (see Figure). Median progression-free survival was similar among treatment-naive and relapsed/refractory patients, at 41 and 35 weeks, respectively, and median overall survival was not reached in either cohort with median follow-up of 2.6 and 3.5 years. Importantly, in response to in situ vaccination, all patients made tumor-specific immune responses within 2 to 4 weeks post-vaccination with the most informative markers being the activation marker CD278 (ICOS) for CD4 T cell response among the CD45RO+ memory subset, and perforin and granzyme B for CD8 T cell responses. Based on the anti-lymphoma activity observed we have recently initiated two Phase I/II dose-escalation trials of a second-generation TLR9 agonist and radiation therapy in relapsed/refractory low-grade NHL and relapsed NHL post-allogeneic transplant (NCT01745354). Figure 1 Figure 1. Disclosures Advani: Seattle Genetics, Inc.: Other, Research Funding; Genentech: Research Funding; Janssen Pharmaceuticals: Research Funding; Pharmacyclics: Research Funding; Celgene: Research Funding; Takeda International Pharmaceuticals Co.: Research Funding.

Introduction: The BTK inhibitor, ibrutinib is FDA-approved in MCL and CLL, with activity in the majority of CD20+ B-cell malignancies. As rituximab-combination chemotherapy is today's standard of care in CD20+ B-cell malignancies, we previously investigated and determined ibrutinib antagonizes rituximab-dependent NK-cell mediated cytotoxicity (ADCC) due to ibrutinib's secondary irreversible binding to interleukin-2 inducible tyrosine kinase (ITK) which is required for FcR-stimulated NK cell function including calcium mobilization, granule release, and overall ADCC. We hypothesized that BTK inhibitors, BTK-InhA (ACP-196), BTK-InhB (BGB-3111) and BTK-InhC (undisclosed), each with lower ITK binding, may preserve NK cell function and therefore synergize rather than antagonize rituximab. Methods: Rituximab and trastuzumab-dependent NK-cell mediated cytotoxicity was assessed using lymphoma and HER2+ breast cancer cell lines as well as autologous CLL tumor cells. In vitro NK cell cytokine secretion, degranulation and cytotoxicity were assessed by IFN-g release, CD107a mobilization and chromium release. Results: FcR-stimulated NK cells following exposure to rituximab-coated lymphoma cells or trastuzumab-coated HER2+ breast cancer cells express high and moderate levels of ITK and BTK, respectively. Ibrutinib, in a dose-dependent manner (0.1 and 1uM), and not BTK-InhA, BTK-InhB or BTK-InhC (each at 1uM), inhibited both rituximab- and trastuzumab-induced NK cell cytokine secretion in vitro (Fig A *p=.018, **p=.002, ***p Conclusions: Ibrutinib is clinically effective as monotherapy and in combination with rituximab, despite inhibition of ADCC in vitro and in vivo murine models due to ibrutinib's secondary irreversible binding to ITK. Preclinically, the efficacy of therapeutics which do not inhibit NK cell function, including three novel BTK inhibitors, is superior to ibrutinib. Clinical investigation is needed to determine the impact of this finding on patients with lymphoma receiving rituximab. Figure 1 Figure 1. Figure 2 Figure 2. Figure 3 Figure 3. Disclosures Wang: BeiGene: Employment. Lannutti:Acerta Pharma: Employment. Johnson:Acerta Pharma: Employment.

Background: Previous national data demonstrate racial/ethnic survival disparities among hematopoietic stem cell transplant recipients. It is unclear, however, if these disparities persist in large academic settings. Therefore, we sought to investigate racial/ethnic transplant-related mortality and morbidity disparities after autologous and allogeneic transplant in our single-institution major transplant center, adjusting for the influence of patient factors, clinical factors, insurance type, and neighborhood socioeconomic status. Methods: We conducted a retrospective chart review of all patients who received a hematopoietic stem cell transplant (autologous and allogeneic) at Stanford University during the years of 1998-2012. Kaplan Meier method predicted survival by race/ethnicity stratified by transplant type. Cox proportional hazard models estimated mortality and transplant-related morbidity (graft-versus host disease for allogeneic transplant recipients) adjusted by demographic (age, gender, race/ethnicity), clinical (diagnosis, year of transplant, treatment regimen), and socioeconomic (insurance, neighborhood socioeconomic status) factors. Results: There were a total of 3,407 patients included in the analysis who received a hematopoietic stem cell transplant during the years of study. Unadjusted Kaplan Meier Curves showed a statistically significant improved survival for Hispanics and Asian/Pacific Islander (API) patients compared with NWH allogeneic transplant recipients (p = 0.01), however, there were no differences in survival for allogeneic transplant recipients by race/ethnicity (p=0.77). Cox proportional hazard models, among autologous transplant recipients, after adjustment for demographic, clinical, and socioeconomic status factors, demonstrated an increased mortality for API patients compared with non-Hispanic white (NHW) patients (HR 1.25 95% CI (1.01-1.57). Non-Hispanic black (NHB) had an equivalent mortality and Hispanics had a non-statistically significant mortality compared with NHW populations (HR 0.97 95% CI (0.69-1.35); HR 0.82 95% CI (0.61-1.10), respectively). Among allogeneic transplant recipients, we found no mortality disparities by race/ethnicity in fully adjusted models. However, we found transplant-related morbidity disparities with an increased odds of graft-versus-host disease among API recipients compared with NHW recipients. (OR 1.57; 95% CI (1.06-2.31)). Conclusions: After adjustment for individual, clinical, socioeconomic and insurance status factors, we found increased mortality after autologous transplant and a higher odds of treatment-related morbidity after allogeneic transplant among API patients compared with other racial/ethnic groups. The worse outcomes among these groups despite equal access to post-transplant related care in our institution requires further investigation. Disclosures No relevant conflicts of interest to declare.

The bone marrow (BM) is a complex microsystem to support lifelong blood production. At steady-state most hematopoietic stem cells (HSC) are quiescent. However, in situations of increased demand, their activation is triggered by an array of signals, such as cytokines. Following hematopoietic cell transplantation (HCT) in the phase of hematopoietic reconstitution maximal blood production is needed. In an HCT donor HSC are given together with immune cells in the belief that T cells support HSC engraftment and regeneration of the blood system. Yet, states of immune mediated BM insufficiency, hypoplasia, and cytopenias are often observed. Moreover, with increased use of reduced intensity conditioning (RIC) engraftment failure has reemerged as a serious problem. Here, we studied the effects of distinct T cell subsets on hematopoietic reconstitution following HCT; specifically, we examined how conventional CD4+CD25- T cells (CD4conv) vs regulatory T cells (CD4+CD25+, Treg) modify the BM environment and influence donor-HSC activity and engraftment. We used minor-mismatched mouse models, non-myeloablative total body irradiation (TBI) conditioning and transplantations of purified HSC (KTLS; cKit+Thy1.1loLin–Sca-1+) plus selected T cell subsets. Recipients of HSC, HSC+Treg or HSC+CD8+ - but not HSC+CD4conv- demonstrated prompt donor engraftment with mixed chimerism in all lineages. Transplantation of HSC+Treg resulted in significantly faster lymphocyte recovery and higher levels of donor chimerism compared with recipients of HSC, HSC+CD8+ or HSC+CD4conv. Particularly B-cell regeneration was markedly higher in HSC+Treg-recipients compared with all other groups. In contrast, (B-) lymphopoiesis was severely impaired in recipients of HSC+CD4conv; when lymphocytes recovered eventually they were of host type. Moreover, in BM and spleens of HSC+CD4conv recipients pronounced hypocellularity was observed. This suppression of hematopoiesis was due to IFNg secretion of donor CD4conv cells, which were activated by dendritic cells via IL-12. High cytokine levels (of both IL-12 and IFNg) were only detectable in the BM (and not the spleen) of HSC+CD4conv recipients, where they resulted in an arrest of early hematopoiesis at the stage of short-term HSC and decreased cell-cycle activity within the progenitor compartment. As a consequence more mature multipotent progenitors were lacking. The key role of IFNg in halting hematopoietic maturation was confirmed by using CD4conv cells from IFNg-/- mice, which had no suppressive effects on BM cellularity and maturation of blood cells; rather, recipients of HSC+IFNg-/-CD4conv cells had equivalent cell numbers and subset distributions as mice given HSC alone. We hypothesized that differences of hematopoietic regeneration and donor engraftment relate to cell cycle activity of HSC in presence of CD4conv vs Tregs. To study the influence of these CD4-subsets on HSC cycling in more detail FACS-purified Treg vs CD4conv cells were infused into congenic mice following low-dose TBI-stimulation. In fact, on d+8 post-infusion HSC in the BM of Treg-recipients had increased cell-cycling activity in long-term-HSC and multipotent-progenitor fractions compared with mice given CD4conv cells or radiation only. These data lead us to speculate that Tregs promote, directly or indirectly, HSC proliferation; in the context of an allogeneic HCT this increased cycling of host HSC my open-up the niche space required to allow donor HSC to engraft. In contrast, CD4convinhibited HSC cycling activity, resulting in BM hypoplasia and cytopenias; at the same time donor HSC engraftment was impaired due to HSC-niche occupation by quiescent host HSC. Our findings underscore the critical role of T cells in regulating hematopoiesis under physiologic conditions and even more following allogeneic HCT. While donor T cells are generally believed to improve regeneration of the blood post-HCT and to be required to overcome host barriers, our data suggest Treg facilitate engraftment and hematopoiesis by increasing HSC cycling-activity and thereby making marrow sites available. CD4conv appear to have the opposite effect, resulting in decreased HSC proliferation and maturation - thus occupation of HSC niches. Our studies are of particular relevance to allogeneic HCT settings using RIC, where host HSC persist and grafts can be rejected by residual host immune cells. Disclosures No relevant conflicts of interest to declare.

Oncolytic virotherapy is safe and clinically active in solid tumors, however its efficacy in hematologic malignancies as well as in combination with checkpoint inhibitors and radiation is unexplored.To simulate advanced lymphoma, A20 lymphoma cells were injected subcutaneously on bilateral flanks of BALB/c mice and treatment initiated on day 17 to only the right flank tumor with local irradiation (Irr), intratumoral (i.t.) vaccinia virus (VACV) and i.t. anti-CTLA-4 mAb (Irr-VACV-CTLA4, Figure 1). The Irr-VACV-CTLA4 regimen was the most effective in eradicating or shrinking both treated and untreated tumors and extending survival, followed by the Irr-VACV regimen. Treatment with Irr-VACV-CTLA4, led to initially a mature, activated NK cell (KLRG1+CD27+) infiltrate (day 6 post-treatment) followed by a CD8+T cells infiltrate (day+13) in treated tumors. In contrast, treatment with VACV-CTLA4 led to activated NK-cell accumulation (day +6) followed by a CD8+T-cell infiltrate (day+13) in non-treated tumors. Importantly, CD8+CD44hi T cells isolated from the blood and spleens of the treated mice showed functional specificity to A20 lymphoma cells, but not to MHC-matched tumor cells (CT26) in intra-cellular stains for IFN-g. Splenocyte-derived A20-specific CD8+CD44hi T cells were induced most efficiently in the Irr-VACV-CTLA4 regimen-treated mice, while blood-derived A20-specific CD8+CD44hi T cells were induced most efficiently in the Irr-VACV regimen-treated mice. Viral plaque assays (VPA) showed lack of live viral particles in both treated and untreated tumors upon sacrificing mice 4 to 10 weeks after treatment initiation. Surprisingly, VPA assays identified live virus in the livers of Irr-VACV-CTLA4 regimen-treated mice, which paralleled a reduced metastatic load. Our findings are the first to demonstrate the potential of combination immunotherapy with oncolytic viruses and checkpoint inhibitors in hematologic malignancies. The antitumor activity is attributed to the induction of an effective and specific immune response. This finding is corroborated by the significant infiltration with mature activated NK cells, followed by CD8+T cells, in both treated and untreated tumors. Importantly, the tumor-specific CD8+T cells showing a memory phenotype (CD44hi) suggest the effective induction of a potent immune memory response. Effective targeting of distant metastases after intratumoral administration is also an important finding with significant clinical implications. This novel combination immunotherapy with oncolytic viruses and checkpoint inhibitors following local tumor irradiation is now being translated to a Phase 1 proof-of-concept clinical trial in non-Hodgkin's lymphoma at our institution. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare.

Introduction: Kohrt et al., Blood, 2014 demonstrated that ibrutinib antagonizes ADCC function of rituximab in vitro in ADCC assays and in vivo in the DHL-4 xenograft model through inhibition of FcgammaR signaling in immune effector cells, possibly mediated by inhibition of ITK. Obinutuzumab (GA101) is a glycoengineered type II CD20 antibody that mediates higher direct cell death induction than rituximab, and by being glycoengineered mediates enhanced induction of ADCC and ADCP. Here we aimed to investigate the impact of ibrutinib on the immune effector function of obinutuzumab as compared to rituximab. Experimental methods: The impact of ibrutinib (dose range 30, 100, 300 ng/ml to cover Cmax and Ctrough in patients) on NK cell mediated ADCC induction by obinutuzumab and rituximab was investigated using SU-DHL4 and Z138 cells as targets in LDH and chromium release assays or measuring CD16 downmodulation and the degranulation marker CD107a. IFNg release as a surrogate for NK cell activation was investigated using DHL-4 target cells or an autologous in vitro system using leukemic cells derived from CLL/NHL patients. Depletion of CD19 positive B-cells was determined in whole blood from healthy volunteers in flow cytometry-based whole blood assay. In vivo the combination of obinutuzumab or rituximab (10 mg/kg once weekly for 3 weeks) with ibrutinib (25mg/kg BID days 14-28) was investigated in the DHL-4 xenograft model. Results: In ADCC assays, ibrutinib (dose range 30, 100, 300 ng/ml) resulted in a reduction of the ADCC potency of obinutuzumab and rituximab. However, at saturating antibody concentrations of 10 ug/ml, ADCC mediated by obinutuzumab was retained while ADCC mediated by rituximab was strongly reduced as measured by chromium release (Figure 1A). Interestingly, in the whole blood B cell depletion assay only little impact of ibrutinib on obinutuzumab-mediated B cell depletion in terms of EC50 and maximal killing was observed at clinically meaningful concentrations of ibrutinib (30, 100, 300 ng/ml), while the activity of rituximab could be completely abolished with 300 ng/ml ibrutinib (Figure 1B). Notably, control experiments using an effector dead version of obinutuzmab that cannot any longer mediate ADCC or ADCP demonstrate that the retained B cell depletion by obinutuzumab in presence of ibrutinib is not due to direct cell death induction, but also due to immune effector cell mediated function (ADCC and ADCP). In the DHL-4 xenograft model where ibrutinib as a single agent has no anti-tumoral efficacy, the combination resulted in a reduced anti-tumoral efficacy of rituximab, whereas efficacy of obinutuzumab was not affected (Figure 1C). Conclusions: Surprisingly, we found that the inhibitory effect of ibrutinib on the immune effector mediated activity of obinutuzumab is not observed when compared to rituximab. Most notably, ADCC at saturating antibody doses, whole blood B cell depletion and in vivo efficacy of obinutuzumab were retained in presence of clinically relevant concentrations of ibrutinib covering Cmax and Ctrough levels, whereas the activity of rituximab was almost completely abolished under these conditions. We hypothesize that the differential behavior of obinutuzumab and rituximab may be related to the enhanced FcgRIII affinity and stronger FcgRIII signaling activation mediated by obinutuzumab as a consequence of glycoengineering that may subsequently overwrite inhibitory effects of ibrutinib. While the clinical relevance of the observed preclinical antagonism for the combination of rituximab with ibrutinib still needs further clinical investigation, these preclinical data strongly support the clinical investigation of ibrutinib in combination with the glycoengineered Type II CD20 antibody obinutuzumab for the treatment of chronic lymphocytic leukemia and other B-cell malignancies. Figure 1 Figure 1. Disclosures Herter: Roche: Employment. Bacac:Roche: Employment. Umana:Roche: Employment. Klein:Roche: Employment, Equity Ownership, Patents & Royalties.

We have designed a novel approach for treating tumors utilizing a combination of active immunotherapy and targeted kinase inhibition. We injected tumors with unmethylated CG-enriched oligodeoxynucleotides (CpG), an agonist for the toll like receptor 9 (TLR9). This in situ CpG injection resulted in local tumor eradication by NK cells and macrophages but on its own was not able to induce a systemic anti-tumor immune response. Ibrutinib is an irreversible inhibitor of Bruton's tyrosine kinase (Btk), a key enzyme in the signaling pathway downstream of B Cell receptor (BCR) is now known to also inhibit ITK, a key enzyme in the survival of Th2 T cells. By doing so, it can shift the balance to the more effective Th1 immune response. In our murine model, the combination of in situ injection of CpG and systemic treatment with Ibrutinib resulted in eradication of tumor cells not only in the local site but also at distant sites. This anti-tumor immune response required both CD4 and CD8 T cells as the effect was lost in nude, scid and T cell depleted mice. Additionally, transferred T cells from animals previously treated with CpG and Ibrutinib protected na•ve mice from tumor challenge. This immune-enhancing effect of Ibrutinib is unexpected. It reflects on the mechanism of action of the drug against B cell malignancies and it provides a novel way of enhancing anti-tumor immune responses against other types of cancer. Figure 1 Figure 1. A. six to eight weeks BALB/C mice were inoculated with 1x106 H11 cells s.c. into the right and left sides of their abdomen, tumor growth was monitored with a digital caliper. Therapy was started when tumors reached a size of 0.7-1cm in the largest diameter, CpG (100µg/injection) was given intratumorally to the left tumor every day in days 1-5, Ibrutinib (6mg/kg) was given daily, intraperitonealy (IP) on days 1-8. B. Growth curve of the right (non-treated) tumor C. Mice survival D. Mice at day 4 after treatment. Disclosures No relevant conflicts of interest to declare.

Meeting abstracts IL-15 activates and induces the proliferation of CD8+ T cells and NK cells. The Cancer Immunotherapy Trials Network (CITN) is conducting a Phase I, open-label, dose-escalation study of subcutaneous (SQ) rhIL-15 in advanced melanoma, renal cell, non-small cell lung and squamous

Objective: Inhibition of indoleamine-2,3-dioxygenase-1 (IDO-1), indoleamine-2,3- dioxygenase-2 (IDO-2), and/or tryptophan-2,3-dioxygenase (TDO) represent novel opportunities for reversing the immunosuppressed microenvironment found in cancer patients. However, little is known about the relative expression of these enzymes in tumors and associated lymph nodes of mouse tumor models, including those previously utilized in the preclinical characterization of tool compounds or even clinical candidates that inhibit one of these enzymes. This report describes the systematic characterization of expression of these enzymes in several of the most commonly used mouse tumor models. Study Design: Tumor models evaluated included B16F10 melanoma, B16-GMCSF melanoma and PAN02 pancreatic in C57BL/6 mice, and 4T1 breast and CT26 colon in Balb/c mice. Cells were injected s.c. into the flanks of 8-week old female mice. In each model, tumors and draining lymph nodes were collected from separate cohorts of mice when the mean tumor volumes reached 100 and 1,000 mm3, respectively. Expression of the 3 enzymes of interest was assessed using a variety of techniques, including qRT-PCR, Western blot and FACS. Results: A highly complex picture arises from our study. Expression levels of each of the enzymes studied here were highly dependent on the choice of tumor cell, the tissue analyzed (tumor vs. lymph node), as well as the stage of growth of the tumor. In general, the most commonly expressed of the 3 enzymes was IDO-1, primarily on tumor cells and dendritic cells in the lymph nodes. Conclusion: It is critically important, in the preclinical evaluation of potential new drugs, to select mouse models that recapitulate specific and well-characterized elements of the signaling pathway targeted by those drugs. The results presented in this communication should be of particular interest to those investigators pursuing a therapeutic strategy of immunosuppression reversal. Citation Format: Rajkumar Noubade, Holbrook Kohrt, Lisa Marshall, Idit Sagiv-Barfi, Jonathan Hebb, Cariad Chester, Amanda Rajapaksa, Erin Waller, Steve Young, Jay Powers, Juan Jaen. Expression of tolerogenic enzymes IDO-1, IDO-2 and TDO in commonly used mouse tumor models and impact on model selection for evaluation of immunosuppression reversal by novel therapeutics. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 3648. doi:10.1158/1538-7445.AM2014-3648

Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CA Introduction: Tumor irradiation induces innate and adaptive immune responses which, rarely, lead to tumor regression at distant sites, the abscopal effect. We have previously demonstrated that immunotherapy including Toll-like-receptor agonists (CpG) and checkpoint inhibitors (anti-CTLA4) both preclinically and clinically ([NCT00185965][1] & [NCT01769222][2]) can significantly increase the rate of systemic, abscopal responses (Kim, Blood 2012 & Brody, JCO 2010). Here we provide the first report of a preclinical murine model and patient case series following local radiation and systemic anti-PD-L1 ([NCT01375842][3]). Methods: Preclinical modeling was performed in a two-tumor, syngenic, A20, lymphoma BALB/c model combining fractionated single tumor radiation and systemic (i.p.) anti-PD-L1. Patients receiving MPDL3280A, a human mAb containing an engineered Fc-domain, as part of the phase 1 clinical trial with mixed responses or asymptomatic progression of disease were eligible for the addition of local radiation therapy. Murine and human immune responses including cell phenotype and function, specifically assessing expression of PD-L1 and production of IFNγ were determined by standard flow cytometry and time of flight mass cytometry (CyTOF). Results: Fractionated radiation delayed tumor growth at the treated site only, and systemic anti-PD-L1 reduced tumor growth rate at both sites, however despite prolonged survival all mice died by day 38 following either monotherapy (radiation or anti-PD-L1). In contrast, combination local fractionated radiation and systemic anti-PD-L1 flattened tumor growth at both the irradiated and un-irradiated site, and prolonged survival with 50% survival at day 48 post-tumor inoculation. Modulation of PD-L1 expression post-radiation and tumor-specific augmentation of IFNγ secretion correlated with the enhanced anti-tumor activity. Five patients including four with solid tumors received fractionated, non-definitive dose radiation with at least stabilization of systemic progression in all patients and a RECIST partial response at systemic sites in 1 patient, notably with a synovial sarcoma. Transient, grade 1-2 inflammatory adverse events (fevers, flu-like symptoms) occurred with no DLTs or serious immune-related toxicities. Modulation of PD-L1 expression, T cell phenotype and IFNγ secretion was observed and updated clinical and immune response will be presented. Conclusion: We provide the first report of combination local radiotherapy with anti-PD-L1 demonstrating synergy in a preclinical model and clinical activity in a limited case series. The magnitude of the immune response and abscopal response rate in mice and humans provides proof-of-concept that anti-PD-L1 may be an equally if not more potent combination immunotherapy with radiation compared to our experience with CpG and/or anti-CTLA4. Citation Format: Idit Sagiv-Barfi, Amanda Rajapaksa, Debra Czerwinski, Serena Chang, Jonathan Hebb, Cariad Chester, Erin Waller, Gregg Fine, Daniel Chen, Marcin Kowanetz, Bryan Irving, Ronald Levy, Holbrook Kohrt. Local tumor irradiation combined with α-PDL-1 immune checkpoint inhibition results in local and systemic anti-tumor responses: Successful translation of a mouse model to a human case series. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2941. doi:10.1158/1538-7445.AM2014-2941 [1]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT00185965&atom=%2Fcanres%2F74%2F19_Supplement%2F2941.atom [2]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT01769222&atom=%2Fcanres%2F74%2F19_Supplement%2F2941.atom [3]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT01375842&atom=%2Fcanres%2F74%2F19_Supplement%2F2941.atom

Preclinical models have been developed and applied to predict the clinical efficacy of immune checkpoint antibodies. Now these models can be used to dissect the mechanisms by which such immunotherapeutic antibodies work and to build the rationale for combining immune checkpoint-targeting antibodies with potential synergistic activity in cancer patients.

Some evidence suggests that monoclonal antibodies (mAb) can induce an adaptive immune response against tumor cells ("vaccinal effect"). Recently, we have shown that an anti-CD137 mAb can enhance the "vaccinal effect" of an anti-tumor mAb (cetuximab), thereby transforming a passive, monoclonal, short-term immunotherapy into an active, polyclonal, long-lasting immune response.

7606 Background: Thymomas (T) contain lymphocytic infiltrates, and as they arise from an immune organ confer a unique tumor microenvironment. Tumor expression of the immune checkpoint ligand, PD-L1...

3007 Background: 4-1BB agonists markedly enhance cytotoxic T-cell responses, resulting in anti-tumor activity in several models. PF-05082566 is a fully humanized IgG2 agonist monoclonal antibody ta...

3017 Background: Anti-CD137 antibody was shown in both murine cancer models and in a first-in-human, phase I trial (Sznol et al., 2008) to increase peripheral activated CD8 T cells and IFN-inducibl...

Background Sitagliptin is a new oral glucose-lowering medication that acts via the incretin hormone system. The most common side-effects are headache and pharyngitis, and few serious adverse events were observed during clinical trials. Dose adjustment is recommended in renal insufficiency, but long-term safety experience is limited. Case report We present a patient with chronic renal insufficiency who developed leg pain, weakness and tenderness after starting treatment with high-dose sitagliptin while on simvastatin. The patient had acute renal failure and rhabdomyolysis that resolved with cessation of sitagliptin, simvastatin, ezetimibe, diuretics and olmesartan. All drugs except sitagliptin, ezetimibe and simvastatin were resumed, and the patient was subsequently started on lovastatin without recurrence of rhabdomyolysis. Conclusions High doses of sitagliptin may have worsened this patient's renal failure and precipitated rhabdomyolysis by increasing circulating levels of simvastatin. Given the high likelihood that sitagliptin will be co-administered with statins and renally active medications, further study of long-term safety of sitagliptin in renal sufficiency may be warranted.

Introduction Ibrutinib is an irreversible inhibitor of Bruton’s tyrosine kinase (BTK) with promising clinical activity in phase I/II clinical trials in CD20+ B-cell malignancies for which rituximab-combination chemotherapy is today’s standard of care. Given homology between BTK and interleukin-2 inducible tyrosine kinase (ITK), we recently confirmed based on molecular and phenotypic analysis that ibrutinib irreversibly binds ITK (Dubovsky J, et al: Blood 2013). ITK expression in FcR-stimulated NK cells leads to increased calcium mobilization, granule release, and cytotoxicity (Khurana D et al: J Immunol 2007). As FcR stimulation is requisite for antibody-dependent NK-cell mediated cytotoxicity (ADCC), we investigated if ibrutinib influenced rituximab’s anti-lymphoma activity as well as trastuzumab’s anti-HER2+ breast cancer activity. Methods Rituximab and trastuzumab-dependent NK-cell mediated cytotoxicity was assessed using lymphoma and HER2+ breast cancer cell lines. In vitro NK cell cytokine secretion, degranulation and cytotoxicity were assessed by IFN-g release, CD107a mobilization and chromium release. Xenotransplant lymphoma and breast cancer models in athymic, nu/nu mice were used to assess efficacy of ibrutinib and rituximab or trastuzumab combination regimens. Using an autologous system from patients with chronic lymphocytic leukemia (CLL), inhibition of NK cell function was compared between ibrutinib and CGI-1746 (Di Paolo JA, et al: Nature Chem Biol 2011), a previously reported selective BTK inhibitor lacking ITK inhibitory function. Results FcR-stimulated NK cells following exposure to rituximab-coated lymphoma cells express high and moderate levels of ITK and BTK, respectively. Ibrutinib inhibited both rituximab- and trastuzumab-induced NK cell cytokine secretion in a dose-dependent manner at 0.1 and 1mM of ibrutinib in vitro (Fig 1A *p=.009, **p=.001, ***p Conclusions Preclinical investigation of a combination therapy currently in Phase II and III clinical trials, adding a promising novel agent, ibrutinib, to a current standard of care, rituximab, for the treatment of B-cell lymphomas, surprisingly results in antagonistic effects. We demonstrate that the abrogation of both rituximab’s and trastuzumab’s anti-tumor efficacy is a result of ibrutinib’s inhibition of FcR-stimulated NK cell function, specifically ADCC. Selective BTK inhibitors or alternative ibrutinib dosing schedules, sequential versus concurrent, may preserve the anti-lymphoma efficacy of both agents. Disclosures: Buggy: Pharmacyclics: Employment, Equity Ownership.

Natural killer (NK) cells mediate anti-lymphoma activity by spontaneous cytotoxicity and antibody-dependent cell-mediated cytotoxicity (ADCC) when triggered by rituximab, an anti-CD20 monoclonal antibody (mAb) used to treat patients with B cell lymphomas. The balance of inhibitory and activating signals determines the magnitude of NK cell's efficacy by spontaneous cytoxicity. Using a killer cell immunoglobulin-like receptor (KIR) transgenic murine model, we show that blockade of the interface of inhibitory KIRs with MHC class I antigens on lymphoma by anti-KIR antibodies prevents a tolerogenic interaction and augments NK cell spontaneous cytotoxicity. In combination with anti-CD20 mAbs, anti-KIR treatment induces enhanced NK cell-mediated, rituximab-dependent cytotoxicity against lymphoma in vitro and in vivo in syngeneic and KIR transgenic murine lymphoma models. Specifically targeting murine NK cells in vitro, anti-Ly49C/I F(ab')2 increased anti-CD20 mAb-mediated NK cell degranulation as measured by CD107a mobilization and interferon-γ release, as well as increased cytotoxicity as assessed by chromium release. In the syngeneic EL4-huCD20 lymphoma model, anti-Ly49C/I F(ab')2 enhanced the anti-lymphoma activity of anti-CD20 mAb in vivo (Fig 1A-1B) and was NK cell-dependent with efficacy abrogated by NK cell depletion with anti-Asialo-GM1. To validate these observations and the potential efficacy of a fully human anti-KIR mAb (IPH2101, lirilumab), we demonstrated, in vitro, dose-dependent KIR2DL3 saturation and tumor lysis following blockade of KIR2DL3/HLA-C with lirilumab. In the transgenic KIR murine model, lirilumab therapy improved survival in an NK cell-dependent manner in both a prophylactic and therapeutic HLA+ (221 HLA-Cw3) lymphoma model. In combination, lirilumab therapy synergistically enhanced rituximab's anti-lymphoma efficacy in vivo in an NK cell-dependent manner (Fig 2A-C). These results support a therapeutic strategy of combination, rituximab and KIR blockade through lirilumab, illustrating the potential efficacy of combining a tumor targeting therapy with an NK cell agonist thus stimulating the post-rituximab anti-lymphoma immune response. Disclosures: Thielens: Innate Pharma: Employment, Equity Ownership. Sola:Innate Pharma: Employment, Equity Ownership. Chanuc:Innate Pharma: Employment, Equity Ownership. Fuseri:Innate Pharma: Employment. Bonnafous:Innate Pharma: Employment, Equity Ownership. Vivier:Innate Pharma: Membership on an entity’s Board of Directors or advisory committees. Romagne:Innate Pharma: Employment, Equity Ownership, Membership on an entity’s Board of Directors or advisory committees. Andre:Innate-Pharma: Employment, Equity Ownership. Blery:Innate Pharma: Employment, Equity Ownership.

Background A clinical trial is ongoing at Stanford for MCL patients in first remission that interdigitates an autologous CpG-stimulated tumor cell vaccine with autologous peripheral blood stem cell transplant (PSCT) (NCT00490529). In this trial, blood samples collected before and after vaccination and serially post-transplant are assayed for minimal residual disease (MRD) and for T cell repertoire using the LymphoSIGHT™ sequencing method (Faham et al., Blood 2012). We identified a set of T cell clones that appear to be responding to the vaccine, and therefore we investigated whether the number of these clonotypes was correlated with MRD status. Methods Using universal primer sets, we amplified rearranged IgH variable (V), diversity, and joining (J) gene segments from genomic DNA. Amplified products were sequenced to obtain >1 million reads. The B cell tumor-specific sequence was identified for each patient based on its high frequency in the original tumor biopsy. The presence of the tumor cells was then monitored in serial blood samples with a sensitivity of 1 cell per million leukocytes. The same blood samples were used for amplification, sequencing and analysis of the entire TCRβ repertoire. To facilitate identification of tumor vaccine-induced TCRβ clonotypes, we sequenced the TCRβ repertoire immediately before and after the administration of both the priming vaccination and a booster vaccination. We developed a metric called the vaccine response score (V score). This metric is calculated for each clonotype and reflects the increase in frequency after the initial vaccination AND after the boost. The formula for calculating V score is: V = F1 x F2 x square root [1/ (|F1 – F2| + 1)], where F1 and F2 represent the fold-change of the priming and boost vaccinations, respectively. Clonotypes with a V score >10 were deemed to be vaccination-induced by virtue of these frequency changes. Results In a series of 12 vaccinated patients, the number of clonotypes with V score ≥ 10 ranged between 0 and 262, with a median of 57. We utilized an antigen-specific analysis to validate that clones with high V scores (≥ 10) were in fact tumor-specific. For this analysis, we incubated peripheral blood mononuclear cells (PBMCs) with the tumor and then sequenced the TCRβ repertoire from cells obtained after culture. Clones that were enriched after culture compared to pre-stimulation PBMCs were deemed to be antigen-specific. These clones that are antigen-specific are highly likely to have a high V score compared to a random frequency-matched set of clones (P two tailed = 1.8 x 10-10), providing further evidence that clones with a high V score are tumor-specific. We then analyzed the relationship between V score and clinical outcome. Patients could be stratified into two groups with “high” (> 25) or “low” ( Conclusions T cell repertoire analysis identified clonotypes responding to the vaccination, and the presence of these vaccine-specific clonotypes correlates with MRD positivity at the important landmark of one year post-PSCT. Further analysis of additional patients enrolled on the MCL trial is ongoing. This data underscores the prognostic relevance of the sequencing-based V score metric and provides a novel approach for assessment of cancer immunotherapy responses. Disclosures: Faham: Sequenta: Employment, Equity Ownership, Membership on an entity’s Board of Directors or advisory committees.

Background Mantle cell lymphoma (MCL) is an aggressive non-Hodgkin's lymphoma (NHL) and represents one of the most challenging lymphoma types to treat. The B-cell antigen Receptor (BCR) signaling pathway is increasingly implicated in the pathogenesis of some B-cell malignancies, and represents an attractive target for therapy. Recent clinical trials have shown that MCL is sensitive to small molecule inhibitors targeting BCR signaling, with response rates reported to be 68% for the BTK-inhibitor Ibrutinib, but only 40% for the PI3Kδ-inhibitor CAL-101. By contrast, chronic lymphocytic leukemia (CLL) has high response rates to both drugs. We hypothesized that basal- or BCR-induced phosphorylation of key molecules that transmit the BCR signal might differ across different NHL, and thereby help us understand therapy responses to small molecular inhibitors. Methods Single cell flow cytometry measurements of signaling were acquired for samples of MCL (n=34), all obtained prior to therapy. For comparison, samples from CLL (n=14), follicular lymphoma (FL, n=27), diffuse large B-cell lymphoma (DLBCL, n=12), as well as healthy donor peripheral blood (n=8) and tonsils (n=4) were included. BCR-induced signaling was obtained by stimulation with goat anti-human IgM F(ab')2 and IgG F(ab')2 fragments for 4 or 45 minutes. Phosphorylation of 14 signaling molecules was measured under basal (unstimulated) and activation-induced signaling, and CD20, CD3, CD5 and BCL2 were used for discrimination between malignant B cells and tumor infiltrating T cells. Results Measurements of basal phosphorylation levels of signaling molecules displayed differences among the NHL types tested. MCL together with DLBCL had highest frequency of patients with tumor cells having elevated levels of p-p65 NF-kB. In contrast, CLL and FL did not have higher basal levels of p-p65 NF-kB. Elevated basal levels of p-Syk and p-AKT1 were most frequently found in CLL and DLBCL, and were less frequent in MCL and not found in FL. In addition, elevated basal level of p-SFK and p-STAT5 were found in a group of MCL patients. BCR-induced signaling in tumor cells showed striking differences across NHL types. For key BCR-signaling molecules including SFK, SYK and PLCg2, BCR-induced phosphorylation was significantly higher in MCL as compared to the other NHL, and was strikingly different from the low levels observed in CLL (Figure 1). Similar results were obtained for Src family kinases (SFK), ERK1/2, AKT1 and STAT5. Although BCR-induced signaling was highly potentiated in most MCL cases, there was a large variation within the MCL cohort. Hence, the fold change of BCR-induced p-PLCg2, relative to unstimulated cells, ranged from 0.28 to 3.01 in MCL tumors, as compared to 0.44 to 0.89 for healthy donor B cells. Furthermore, BCR-induced p-SYK in MCL correlated strongly with p-PLCg2 (r=0.96, p Conclusion Differences in basal phosphorylation levels of key signaling molecules were found across NHL. BCR-induced signaling was highly potentiated in MCL as compared to the other NHL, and was associated with increased expression of CD79B and IgM. Differences in phosphorylation levels of signaling molecules between different NHL, as well as heterogeneity within each entity might help us understand therapy responses to small molecular inhibitors, although this needs to be tested in follow-up studies. Disclosures: No relevant conflicts of interest to declare.

Activation of TLR9 by direct injection of unmethylated CpG nucleotides into a tumor can induce a therapeutic immune response; however, Tregs eventually inhibit the antitumor immune response and thereby limit the power of cancer immunotherapies. In tumor-bearing mice, we found that Tregs within the tumor preferentially express the cell surface markers CTLA-4 and OX40. We show that intratumoral coinjection of anti-CTLA-4 and anti-OX40 together with CpG depleted tumor-infiltrating Tregs. This in situ immunomodulation, which was performed with low doses of antibodies in a single tumor, generated a systemic antitumor immune response that eradicated disseminated disease in mice. Further, this treatment modality was effective against established CNS lymphoma with leptomeningeal metastases, sites that are usually considered to be tumor cell sanctuaries in the context of conventional systemic therapy. These results demonstrate that antitumor immune effectors elicited by local immunomodulation can eradicate tumor cells at distant sites. We propose that, rather than using mAbs to target cancer cells systemically, mAbs could be used to target the tumor infiltrative immune cells locally, thereby eliciting a systemic immune response.

3622 Background: PD-L1 has been reported to be expressed broadly in human cancer and can lead to inhibition of anti-tumor T-cell responses. MPDL3280A, a human monoclonal antibody containing an engineered Fc-domain designed to optimize efficacy and safety, targets PD-L1, blocking PD-L1 from binding its receptors, including PD-1 and B7.1. Pts with a broad range of tumors were examined in an expansion study with MPDL3280A. Methods: Pts with locally advanced or metastatic tumors received MPDL3280A administered IV q3w. Pts were treated for up to 1 y. Response was assessed by RECIST v1.1. Results: As of Jan 10, 2013, 20 pts with tumor types other than NSCLC, RCC and mM were evaluable for safety and treated at doses of 0.01-20 mg/kg (≤1 [n=3], 3 [n=1], 10 [n=3] and 20 mg/kg [n=13]). Median pt age was 67 y (range 26-80 y), 100% were PS 0-1, 90% had prior surgery and 45% had prior radiotherapy. 95% of pts received prior systemic therapy. Pts received MPDL3280A treatment for a median of 96 days (range 22-330). The incidence of all G3/4 AEs, regardless of attribution, was 50%. No G3-5 pneumonitis or diarrhea was reported. Confirmed RECIST responses were observed in several tumor types, including CRC, GC and SCCHN. Additionally, antitumor activity (tumor shrinkage that has not yet met criteria for a RECIST response) has been observed in sarcoma and lymphoma. Analysis of biomarker data from archival tumors revealed that all pts reported to have RECIST response had baseline tumors that were PD-L1 positive. Updated data will be presented. Conclusions: Treatment with MPDL3280A was well tolerated, with no pneumonitis-related deaths. Rapid responses were observed in pts with CRC, GC and SCCHN. PD-L1 tumor status appears to correlate with responses to MPDL3280A. These data suggest that PD-L1 is a conserved mechanism for mediating tumor immune escape across a range of tumor types. MPDL3280A may be broadly active as anti-cancer therapy in PD-L1–expressing tumors, supporting further investigation. Clinical trial information: NCT01375842.

3015 Background: Cetuximab therapy results in beneficial, yet limited, clinical improvement for patients with KRAS wildtype (WT) colorectal (CRC) and head and neck (HN) cancer. The efficacy of cetuximab, an IgG1 monoclonal antibody against EGFR, is due in part to antibody-dependent cell-mediated cytotoxicity (ADCC) by natural killer (NK) cells. CD137 is a costimulatory molecule expressed following activation on NK and memory, antigen-specific, CD8 T cells. Methods: We investigated the hypothesis that the combination of cetuximab with anti-CD137 mAb will enhance innate and adaptive immunity, thereby improving cetuximab’s anti-tumor efficacy in preclinical models and a prospective trial, NCT01114256. Results: NK cells increased their expression of CD137 by a factor of 30-40 when exposed to cetuximab-coated, EGFR-expressing HN and CRC cell lines. An agonistic anti-CD137 mAb enhanced NK cell degranulation and cytotoxicity 2-fold (~45 to 90% tumor lysis assayed by chromium release). The combination of cetuximab and anti-CD137 mAbs was synergistic in a syngeneic, human-EGFR-transfected murine tumor leading to complete tumor resolution and prolonged survival. NK cell depletion, significantly, and CD8 T cell depletion, partly, abrogated the anti-tumor efficacy of this combination. A series of HN and both KRAS WT and mutant CRC xenotransplant models demonstrated synergy with cetuximab and anti-CD137 mAbs. In our clinical trial, 54 patients with HN cancer receiving cetuximab therapy, circulating and intratumoral NK cells upregulated CD137 with amplitude influenced by duration post-cetuximab and host FcyRIIIa polymorphism. Interestingly, in 10 HLA-A2+ patients, following cetuximab, an increase in EGFR-specific, CD137-expressing, CD8 T cells directly correlated with the percent increase in CD137-expressing NK cells. Conclusions: Our results demonstrate the synergy of combining an agonistic mAb, anti-CD137, augmenting ADCC and T cell memory following a tumor-targeting mAb, cetuximab, in HN and KRAS mutant and WT CRC cancer. These results support a novel, sequential antibody approach by targeting first the tumor and then the host innate and adaptive immune system. Clinical trial information: NCT01114256.

TPS3108 Background: CD137 (4-1BB) is a costimulatory molecule that belongs to the TNF superfamily. It is upregulated on activated lymphocytes, NK cells and dendritic cells and plays an important role in the potentiation of antigen-specific immune responses as well as in antibody-dependent cell-mediated cytotoxicity (ADCC). Urelumab is an agonistic antibody targeting the CD137 receptor. Preclinical evidence has shown that there is modulation of CD137 expression on NK cells after exposure to rituximab. Anti-CD137 agonist monoclonal antibody has been shown to have single-agent anti-lymphoma activity and to potentiate the anti-lymphoma activity of rituximab through enhancing ADCC. We hypothesized that upregulation of CD137 on NK cells by rituximab followed by urelumab could afford a mechanism-based approach to achieve enhanced biologic and/or clinical activity compared to either single agent alone. Here we describe a phase Ib study to investigate the clinical and biologic effects of combined treatment with urelumab and rituximab in patients with relapsed/refractory B-cell malignancies. Methods: This phase I study (n=100) will include dose escalation (Part 1) using a 3+3+3 design and cohort expansion (Part 2). In Part 1, successive cohorts of patients with relapsed/refractory B-NHL will be treated as follows: Cohort 1 (0.1 mg/kg q3weeks) and Cohort 2 (0.3 mg/kg q3weeks) with both cohorts in combination with rituximab 375 mg/m2 given weekly for the first 4 weeks of each 12 week cycle. In Part 2, cohorts of CLL (n=30), follicular lymphoma (FL) (n=30), and diffuse large B-cell lymphoma (DLBCL) (n=20) will be treated at the dose level found to be safe for the urelumab/rituximab combination. The primary objective of the study is to evaluate the safety and define a safe and effective dose of the urelumab/rituximab combination. Secondary objectives include assessment of the antitumor activity, pharmacokinetics, and immunogenicity. Exploratory objectives include investigation of the immunoregulatory activity of this combination in peripheral blood and paired tumor biopsy specimens and the association of these effects with clinical response/toxicity. Clinical trial information: NCT01775631.

Infiltration of specialized immune cells regulates the growth and survival of neoplasia. Here, in a survey of public whole genome expression datasets we found that the gene for chemerin, a widely expressed endogenous chemoattractant protein, is down-regulated in melanoma as well as other human tumors. Moreover, high chemerin messenger RNA expression in tumors correlated with improved outcome in human melanoma. In experiments using the B16 transplantable mouse melanoma, tumor-expressed chemerin inhibited in vivo tumor growth without altering in vitro proliferation. Growth inhibition was associated with an altered profile of tumor-infiltrating cells with an increase in natural killer (NK) cells and a relative reduction in myeloid-derived suppressor cells and putative immune inhibitory plasmacytoid dendritic cells. Tumor inhibition required host expression of CMKLR1 (chemokine- like receptor 1), the chemoattractant receptor for chemerin, and was abrogated by NK cell depletion. Intratumoral injection of chemerin also inhibited tumor growth, suggesting the potential for therapeutic application. Preliminary data in other adenocarcinoma tumor types (eg breast) show similar results, supporting broad applicability of such an approach. These results show that chemerin, whether expressed by tumor cells or within the tumor environment, can recruit host immune defenses that inhibit tumorigenesis and suggest that down-regulation of chemerin may be an important mechanism of tumor immune evasion. Citation Format: Russell Pachynski, Brain Zabel, Justin Monnier, Andrew Gentles, Christina Swanson, Holbrook Kohrt, Husein Hadeiba, Ash Alizadeh, Eugene Butcher. The chemoattractant chemerin suppresses melanoma by recruiting natural killer cell antitumor defenses. [abstract]. In: Proceedings of the AACR Special Conference on Tumor Immunology: Multidisciplinary Science Driving Basic and Clinical Advances; Dec 2-5, 2012; Miami, FL. Philadelphia (PA): AACR; Cancer Res 2013;73(1 Suppl):Abstract nr A74.

TLR-9 ligand (CpG) injected directly into a tumor can induce a therapeutic immune response. However regulatory T-cells (Tregs) eventually inhibit the anti-tumor immune response and thereby limit the power of cancer immunotherapies. We found that Tregs within the tumor preferentially express the cell surface markers CTLA-4 and OX40. We show that intra-tumoral co-injection of anti-CTLA-4 and anti-OX40 together with CpG depletes tumor-infiltrating Tregs. This in situ immunomodulation performed with low doses of antibodies into a single tumor generates a systemic anti-tumor immune response able to eradicate disseminated disease in mice, including established CNS lymphoma with leptomeningeal metastases, sites usually considered as sanctuaries for conventional systemic therapies. These results support a paradigm shift in cancer therapy where instead of using monoclonal antibodies to target cancer cells systemically, we use monoclonal antibodies at low doses to target the tumor infiltrative immune cells locally. The consequences of this strategy are the generation of a more potent anti-tumor immune response, and the expectation of less toxicity. On the basis of our preclinical results we are now opening a Phase I/II clinical trial testing the combination of local low dose radiotherapy with escalating doses of Ibilumimab (anti-CTLA4). Patients with metastatic melanoma, colorectal cancer and lymphoma are eligible. The intention will be to find a safe dose of the antibody when injected locally together with radiotherapy and then to assess the patients for immune responses and for tumor responses at uninjected sites of disease. Citation Format: Aurelien Marabelle, Holbrook Kohrt, Ronald Levy. Local immunomodulation at a single site of tumor generates a therapeutic immune response that cures metastatic disease at distant sites, including the brain. [abstract]. In: Proceedings of the AACR Special Conference on Tumor Immunology: Multidisciplinary Science Driving Basic and Clinical Advances; Dec 2-5, 2012; Miami, FL. Philadelphia (PA): AACR; Cancer Res 2013;73(1 Suppl):Abstract nr IA25.

Abstract 3703 Introduction Local intratumoral injection of CpG ologonucleotides in conjunction with local low dose radiotherapy can induce regression of uninjected distant sites of disease in patients with follicular lymphoma (FL) (Brody et. al JCO). This maneuver activates the Toll-like receptor 9 (TLR9) within the tumor cells and in adjacent antigen presenting cells in the local tumor environment, such as dendritic cells and macrophages. A systemic anti-tumor T cell immune response is thereby induced. Our original phase I/II trial was conducted in patients who had been previously treated with standard therapies and had recurred with multiple sites of disease. We observed significant clinical responses as well as anti-tumor T cell immune responses as measured in vitro using peripheral blood T cells or T cells obtained from a pleural effusion adjacent to a responding tumor site. Methods We now extended the trial of in situ vaccination to newly diagnosed patients prior to any other therapy, a group that may have a more intact immune system. Patients were eligible if they had FL grades I-IIIa, stage III or IV and were not in need of immediate treatment. One involved lymph node was biopsied and viable suspensions of tumor cells were prepared and cryopreserved for use as stimulators in immune assays. A second site received low dose XRT (2Gy on each of two successive days) together with 10 weekly injections of 18mg of CpG ologonucleotide (PF-3512676, Pfizer) all into the same tumor site, beginning on the second day of XRT. Peripheral blood lymphocytes (PBL) were obtained prior to each injection and two weeks after the last injection. These were cryopreserved and used as responder cells for assays of T cell immune responses. We sought to determine the kinetics of anti-tumor T cell immune responses as well as which type of T cell response that was the most informative. Tumor cells were thawed and activated for 3 days with CpG and soluble CD40L. They were then incubated for 5 days with autologous PBL T cells. Fresh stimulator cells were then added for a final overnight culture. Responding T cells were stained for subtype (CD4, CD8, CD56, CD45RO), for their expression of activation markers (CD25, CD137 and CD278), for their expression of intracellular cytokines (IFN-g, TNF and IL-2), and for their expression of cytotoxic enzymes (perforin and granzyme B). The cells were then analyzed by multiparamer flow cytometry (BD LSRII). For measurement of perforin and granzymeB, the cells were gated on CD8+/CD3+/CD56- to exclude NK cells. Results In response to in situ vaccination, all patients made anti-tumor immune responses. Some generated only CD4 responses, some only CD8 responses and others made both CD4 and CD8 T cell responses. Representative data are shown in the figures below. We found that for CD4 responses the activation marker CD278 (ICOS) was particularly informative, and usually restricted to the CD45RO+ memory subset. For CD8 responses, the most robust readout was intracellular expression of the combination of Perforin and Granzyme B. CD8 Immune responses became positive as early as two weeks after the start of vaccination. CD4 responses became positive by four weeks of vaccination. As in our previous trial, we observed clinical responses, with regression of tumors at uninjected/untreated sites of disease. An evaluation of the magnitude and duration of these clinical responses and their relation to immune parameters awaits a more extended time of followup. Conclusions In situ vaccination efficiently induced immune responses in previously untreated patients with FL. These responses occurred within two weeks after initiation of vaccination. The immune responses was heterogeneous and included both CD4 and CD8 T cells. The most robust measures of T cell response will be correlated with clinical outcome. The identification of a limited panel of correlative immune response markers and an understanding of the response kinetics will allow a focused approach to immuno-monitoring in future clinical trials. Disclosures: Advani: Pharmacyclics: Research Funding.

Abstract 3160 Introduction: Acute leukemias (AL) have a distinct incidence pattern in Hispanics versus Caucasians, with higher rates of B-cell acute lymphoblastic leukemia (ALL) and acute promyelocytic leukemia (APL), and lower rates of non-APL acute myeloid leukemia (AML). Despite the fact that incidence differentials between foreign- and native-born populations can inform the relative contributions of environmental and genetic factors in cancer etiology, no such investigation has been undertaken in Hispanics with AL. Methods: To better understand heritable genetic versus environmental contributors to occurrence patterns, we assessed AL incidence in a large population of California Hispanics according to birthplace. Cancer incidence data were obtained from the California Cancer Registry, part of the National Cancer Institute's Surveillance, Epidemiology, and End Results Program. Patients were included if they were living in California and diagnosed between 2000–2009 with AL. Population estimates were obtained from the 2000 Census Summary Files 3, and estimates by nativity were derived from the 20% Integrated Public-Use Microdata Sample of the censuses. Results: Compared to Caucasians, Hispanics had an increased incidence of B-cell ALL and APL, with incidence rate ratios (IRR) of 2.13 (95% confidence interval [CI] 1.93–2.35) and 1.33 (95% CI 1.12–1.57), respectively. There was no difference in B-cell ALL incidence by nativity, with an IRR for foreign- to native-born Hispanics of 1.07 (95% CI 0.85–1.34). However, foreign-born Hispanics had a greater incidence of APL than native-born Hispanics (IRR 1.79, 95% CI 1.11–2.94). Conclusions: Our findings suggest the increased B-cell ALL incidence for foreign-born Hispanics is influenced by heritable genetic factors. Furthermore, the increased incidence of APL for foreign-born Hispanics may result from environmental exposures currently not described and which require further epidemiologic investigation. Disclosures: No relevant conflicts of interest to declare.

Abstract 2366 The bone marrow (BM) is a complex microsystem that supports lifelong blood production and dynamically responds to injury, blood cell loss and senescence. At steady-state most hematopoietic stem cells (HSC) are quiescent. However, in situations of increased demand, their activation is triggered by an array of signals, such as cytokines released during infections. The susceptibility of BM to injury by T cells (TC) is long recognized, e.g. in aplastic anemia, and BM failure syndromes, and the inhibitory effect of IFNg on the proliferative activity of (HSC) is established. How donor TC given as part of an allogeneic hematopoietic cell transplantation (HCT) affect the recipient BM microenvironment and influence HSC activity has not been elucidated. Yet, graft TC are believed to be essential to overcome host barriers, especially following lower intensity (non-myeloablative) conditioning that spares host HSC and residual host immunity. We examined in an MHC-matched, minor-antigen-mismatched mouse model of non-myeloablative HCT the impact of graft TC subsets on HSC and hematopoietic reconstitution. BALB.K (H2k) and BALB.B (H2b) mice were prepared with sublethal radiation at a dose that allows donor cells (AKR/J [H2k] and C57BL/6 [H2b]) to engraft but also permits survival of host HSC and immune populations. Grafts consisted of purified HSC (KTLS; cKit+Thy1.1loLin−Sca-1+) given alone or supplemented with donor TC or TC subsets. Rather than improve engraftment, donor TC, specifically CD4+ cells, resulted in detainment of host HSC in their niches and consequent failure of donor HSC to engraft. These events appeared to be initiated specifically by conventional alloreactive donor CD4+CD25− cells. These CD4+CD25− cells interacted with immune-competent host cells, stimulating CD11c+ dendritic cells to upregulate the activation markers CD40 and MHC II. Following their activation dendritic cells produced excessive amounts of IL-12. IL-12, in turn, activated donor CD4+ cells to produce IFNg. This Th1-response was prominent in BM but absent in the spleen. In this Th1-cytokine environment, high levels of quiescent host HSC were observed, arrested at the short-term HSC stage, while more mature, multipotent progenitors were absent. Moreover, HSC arrest correlated with marrow aplasia, analogous to that seen in BM failure syndromes. This hypocellularity was only observed in recipients of CD4+ containing grafts, while recipients given HSC alone or HSC in combination with CD8+ TC had prompt blood regeneration. When resting HSC were FACS-purified from the BM of recipients of HSC+CD4+ and infused into secondary recipients they promptly expanded and gave rise to all blood lineages. Although recipients of HSC+CD4+ cells ultimately recovered blood formation, B lymphopoiesis was markedly delayed. No clinical signs of graft-versus-host disease were noted in these mice. Lymphocytes that recovered in these recipients of HSC+CD4+ cells were of host type only. In contrast, recipients of pure HSC demonstrated prompt donor cell engraftment, and mixed chimerism in all lymphoid lineages, including T cells. Our model corroborates the importance of the cytokine microenvironment in hematopoietic reconstitution and marrow failure post-transplant. Moreover, our findings underscore the importance of IFNg in BM aplasia and support the idea that IFNg can, under physiologic conditions, exert regulatory effects on hematopoiesis by controlling the exit of ST-HSC out of the marrow niche. Our studies shed new light on understanding the physiology of engraftment and marrow failure after HCT using nonmyeloablative conditioning, and indicate that the composition of the graft is of critical importance due to its effects on the marrow microenvironment We postulate that in some naturally occurring settings of BM inflammation (e.g., infections) detainment of resting HSC in their niche may provide protection of hematopoiesis and that comparable events are relevant for engraftment post-HCT. Disclosures: No relevant conflicts of interest to declare.

Abstract 3713 Background: The complementarity determining region, or idiotype, of the surface immunoglobulin receptor is a tumor-specific marker on B-cell lymphomas that is unique to each patient. Antibodies against idiotype can induce complete regression of lymphoma in patients, but since this therapeutic approach requires the generation of a custom monoclonal antibody for each patient, it has not been practical. Objective: Here we describe a method for targeting the idiotype on the surface of a B-cell lymphoma by using synthetic idiotype-ligands covalently linked to a recombinant IgG Fc domain (Figure 1A). These peptide idiotype-ligands can be identified through oligopeptide library screens and produced inexpensively by automated solid-phase synthesis. Linkage of idiotype-ligands to the Fc domain serves two purposes: to enhance their pharmacokinetics and to augment their anti-tumor effect by activating immune effector functions. Since each patient-specific peptide can be chemically linked to a common IgG Fc domain, this modular construct design yields a patient-specific therapeutic that does not require the production of a custom biologic macromolecule for each patient. Results: Idiotype peptide-ligands were produced by solid-phase synthesis and covalently linked to the amino-terminus of a recombinant mouse IgG2a Fc domain by native chemical ligation, a method for site-selective polypeptide ligation. The resulting peptibody demonstrated targeted killing of a human lymphoma cell line in vitro by crosslinking surface immunoglobulin and triggering activation-induced death. Additionally, the peptibody triggered complement-mediated cytotoxicity of opsonized lymphoma cells in vitro and activated natural killer cells co-cultured with opsonized lymphoma cells. The peptibody exhibited a favorable pharmacokinetic profile and peptibody treatment was sufficient to clear tumor in SCID mice challenged intravenously with a luciferase-labeled human lymphoma cell line (p = 0.0018; Figure 1B-D). Conclusions: Idiotype-specific peptibodies demonstrate multimodal activity against lymphoma cells in vitro and clear human lymphoma in a disseminated xenograft model. The modular design of this therapeutic may enable a personalized and targeted therapy that is feasible to produce for patients with B-cell lymphoma. Disclosures: No relevant conflicts of interest to declare.

Abstract 4108 A major goal for allogeneic hematopoietic cell transplantation (HCT) is to engineer grafts to contain only cell types required to rescue hematopoiesis, recover anti-infectious immune function, and provide curative graft-versus leukemia (GVL) effects. Such grafts will avoid the life-threatening complication of graft-versus-host disease (GVHD). Before manipulative approaches can be translated into clinical practice, the dynamics of engraftment, expansion, and interactions of isolated donor cell populations must be comprehensively understood. Here, in preclinical mouse models, we dissect the effects and potency of distinct purified donor cell populations to exert GVL. To enhance GVL we vaccinated donors with a tumor-antigen, WT1. AKR/b or C3H/SW mice were prepared with 4 weekly WT1-peptide/incomplete Freund's adjuvant vaccinations. Grafts consisted of FACS-purified hematopoietic stem cells (HSC [cKit+Thy1.1loLin−Sca1+]), augmented with CD4+, CD8+, naïve or memory CD4+ and CD8+T cells (TC), or WT1-tetramer selected (WT1)CD8+TC. HSC +/− TC subsets were infused into lethally irradiated, minor-antigen mismatched C57BL/6 mice. A luciferase-labeled, myeloid lineage C57BL/6 leukemia, FBL3, was inoculated on d0 or 14 post-HCT, modeling tumor persistence or early relapse after HCT, respectively. Graft compositions were tested for their strongest GVL potency measured by bioluminescence imaging and survival. When recipients were given tumor cells at the time of HCT (i) HSC+CD4++CD8+ TC grafts from vaccinated donors rescued most recipients, while those from unvaccinated mice resulted in rapid death from leukemia; (ii) recipients of HSC+CD4++CD8+ from unvaccinated mice died faster than recipients of HSC alone, suggesting GVH-related damage to lymphoid organs diminishes the host's endogenous tumor resistance; (iii) HSC + either CD8+, memory (mem)CD8+, or (WT1)CD8+TC from vaccinated donors provided improved but non-curative protection from leukemia compared to HSC alone; (iv) HSC + CD4+TC grafts (vaccinated or unvaccinated) were associated with rapid death; however, (v) CD4+TC were required to fortify tumor eradication by CD8+TC. To model the clinically relevant scenario of early relapse, we next infused tumor cells on d14 post-HCT. At 1 wk prior to tumor inoculation an in vivo boost WT1-peptide vaccination was applied. (i) Again, HSC+CD4++CD8+ from vaccinated donors achieved better tumor control as compared to HSC +/− unvaccinated CD4++CD8+ TC, but a high proportion of recipients died due to progressive leukemia. We speculate the reason for diminished GVL is a lack of tumor antigens present at the time of TC infusion which resulted in a different donor TC repertoire expanding during lymphopenia. (ii) All subgroups given combinations of naïve/memory CD8+ TC and naïve/memory CD4+ TC had prolonged survival compared with HSC recipients, but no long-term cure. (iii) The only group with improved overall survival received HSC + WT1CD8++ CD4+ TC grafts and a booster vaccine prior to tumor injection. Our findings suggest the benefit was due to an in vivo expansion of tumor-specific TC early post-HCT, driven by the tumor antigen + adjuvant re-exposure. We observed with peptide vaccination that: 1.) ∼2/3 of WT1CD8+TC retained a naïve phenotype, and CD4+TC were required to provide help and promote, suggesting that CD4+TC are required to convert naïve into effector memory CD8+TC. 2.) viral immunizations resulted in robust antigen specific responses compared to more modest development of WT1CD8+TC after the peptide-vaccine. However, the minimal numbers of WT1-specific CD8+TC transferred were highly efficient, whereas recipients of HSC+WT1-depleted TC died rapidly of leukemia, which supports the specificity of the response rather than general alloreactivity. Thus, dynamics of engraftment of donor effector populations are complex, modalities to promote tumor specific CD8+TC expansion early post-HCT are critical to maximize GVL and rely upon the interactions of co-transplanted cell subsets and timing of infusions. Immunogenic maneuvers such as vaccinations are powerful tools to enhance the potency of specific effector populations. Transplant approaches using pure HSC to restore hematopoiesis and manipulated TC populations without deleterious GVHD-inducing effects are the ultimate goal of allogeneic HCT and require informed translation of preclinical models. Disclosures: No relevant conflicts of interest to declare.

Abstract 3087 Background: The Myelodysplastic Syndromes (MDS) and Myeloproliferative Neoplasms (MPN) are hematopoietic malignancies primarily affecting older patients. Myeloablative allogeneic hematopoietic cell transplantation (allo HCT) is potentially curative for MDS and MPN, but has traditionally been limited to younger patients because of the toxicity associated with high-dose chemoradiotherapy. The preparative regimen of Total Lymphoid Irradiation (TLI) and Anti-Thymocyte Globulin (ATG) permits the gradual establishment of donor hematopoiesis that is necessary for the graft versus malignancy effect. However, TLI-ATG is protective against acute Graft versus Host Disease (aGVHD), presumably due to the enrichment of tolerogenic recipient NK-T cells (Lowsky et al., NEJM 2005, Kohrt et al, Blood 2009). Here we report the outcomes of patients with MDS and MPN treated with TLI-ATG and allo HCT. Patients and Methods: Between August 2004 and December 2010, 51 patients were treated on an IRB approved Phase II trial. The median age was 63 years (range 50–73). The initial diagnosis was de novo MDS (n=24), CMML (n=6), MPN (n=8), and therapy-related MDS (t-MDS, n=13). The median time from diagnosis to HCT was 11 months (range 3–160). Among all patients, 41% had progressed to AML at some point preceding HCT. Most patients had received treatment beyond growth factor or transfusion support, including cytotoxic chemotherapy (29%), DNA methyltransferase inhibitors and/or immunomodulatory agents (41%), or cytotoxic chemotherapy and DNA methyltransferase inhibitors (14%). Among the patients with de novo MDS, the IPSS-R score (Greenberg et al., Blood 2012) at the time of HCT was Very High or High (17%), Intermediate (23%), Low or Very Low (53%) or unknown/indeterminate (7%). At the time of HCT, 40% of patients had abnormal marrow cytogenetics. Following provision of informed consent, patients were treated with 10 fractions of TLI (cumulative dose1200 cGy) and ATG (Thymoglobulin, Genzyme, 7.5 mg/kg divided in 5 doses). Primary GVHD prophylaxis consisted of Cyclosporine A and Mycophenolate Mofetil. Patients received G-CSF mobilized peripheral blood progenitor cells from matched related (45%), 10/10 matched unrelated (45%), or mismatched unrelated donors (10%). Results: The median follow up for living patients was 3.2 years (range 1.5–7) and for all patients was 1.4 years (range 0.1–7). The three-year overall survival (OS), non-relapse mortality (NRM), and event-free survival (EFS) were 42% (95% CI 28–56%), 8% (0.3–15%), and 31% (18–44%), respectively. The cumulative incidence of aGVHD Grades II-IV was 14% (95% CI 4–23%) and Grades III-IV 4% (0–9%) and did not differ according to allograft source. The three-year cumulative incidence of Chronic GVHD was 33% (20–47%) with median time of onset of 199 days (range 112–475 days). The three-year cumulative incidence of progression was 61% (47–74%) and the median time to progression was 136 days (range 13–292). Rates of progression were not significantly different for patients with MDS (60%, 95%CI 44–73%), MPN (88%, 71–95%), and t-MDS (45%, 15–70%). The presence of abnormal cytogenetics at the time of HCT did not impact risk of progression (p=0.54). However, patients with IPSS-R Very High and High Risk scores all had progressive disease within the first 100 days. In contrast, patients with Intermediate Risk scores had equivalent progression and progression-free survival rates when compared with those in the Very Low and Low risk groups (Hazard Ratio for progression 1.25, 95% CI 0.36–4.32). CD15 chimerism at day 28 was the most predictive of progression at any subsequent time point. Those with donor CD15>90% donor at day 28 (40% of patients) defined a group with a low likelihood of disease progression (Odds Ratio 0.20, p=.01). Conclusions: For patients with MDS and MPN, nonablative TLI/ATG conditioning prior to allo HCT was safe and associated with low rates of acute GVHD and NRM. Patients whose IPSS-R scores at the time of HCT were Intermediate, Low, and Very Low had significantly better outcomes than those with High or Very High risk scores. Analysis of CD15 chimerism at Day 28 may identify patients at high risk of progression and offers an early time point for intervention. These data suggest that for patients with MDS, earlier transplantation may result in better long-term outcomes due to lower transplant related risks. Disclosures: Off Label Use: ATG (Thymoglobulin, Genzyme) for conditioning prior to allogeneic hematopoietic cell transplantation.

Abstract 1489 Background: A standardized approach to the diagnosis, prognosis and therapy of patients with APL and CNS involvement is limited by the rare nature of the complication. We report the incidence, risk factors, management and outcome of APL with CNS involvement at Stanford University and review the published cases in the literature during the prior three decades. Methods: Between 1983 and Jan 2012, 7 cases of APL with CNS involvement were identified at Stanford University and 65 in the literature during the same interval. All pts in the Stanford series received ATRA and anthracycline containing induction regimens followed by consolidation with anthracycline and cytarabine (7 pts), plus ATRA (5 pts), ATO (1 pt), or IT prophylaxis (1 pt). Five pts received maintenance with ATRA containing regimens in all cases (2 pts relapsed prior to starting maintenance). Median follow-up was 6 years. Results: Seventy-two relapses involving the CNS were seen after 1 to 50 months (median 8 months) and 61 (84.7%) occurred in CR1 with 8 (11.1%) in CR2, 2 (2.8%) in CR3 and 1 (1.4%) in CR4. Forty-seven (66.2%) were accompanied by bone marrow involvement, 14 (29.8%) molecular and 33 (70.2%) hematological. Median age was 35y with 9 pediatric pts (range 1.3 to 20y) and 63 adult pts (range 22 to 69y). Median WBC count was 28,100 cells/mm3 (range 1,100 to 162,000 cells/mm3) with 61.4% >10 and 6.8% >100 cells/mm3. PML/RARα isoform was reported in 29 pts with 37.9% bcr1, 3.4% bcr2, and 58.6% bcr3. Five pts presented with concurrent CNS-localized hemorrhage. Therapy included methotrexate (88.9%), cytarabine (80.6%), arsenic trioxide (ATO, 47.2%), and anthracycline (11.1%) containing regimens with 88.9% receiving intrathecal therapy. Eighty-six percent received cranial radiation with dose ranging 18 to 30 Gy. Autologous or allogeneic transplantation followed salvage therapy in 30.8% and 5.6% of pts, respectively. Fifty-four (75.0%) pts were salvaged successfully, including 88.2% of pts receiving ATO, however 18 (33.3%) pts later relapsed, 4 (22.2%) of which were salvaged with additional treatment. Conclusion: CNS relapse is a rare complications of APL with the majority of cases occurring in pts with WBC >10 cells/mm3, and in first CR. CNS relapse was frequently associated with concurrent bone marrow relapse and approximately one-third occurred with only molecular involvement. A preventative role of IT prophylaxis could not be determined due to only 1 pt receiving prophylaxis. A high percentage of pts salvaged with ATO achieved CR, supporting its therapeutic efficacy and ability to cross the blood brain barrier. Incorporation of ATO in first line therapy for APL may reduce the risk of CNS relapse. Disclosures: No relevant conflicts of interest to declare.

Abstract 582 Background: A clinical trial is being conducted at Stanford University for patients with mantle cell lymphoma (MCL) in first remission that interdigitates an autologous CpG-stimulated tumor cell vaccine with high dose therapy and autologous peripheral blood stem cell transplant (PBST). In this trial, patients receive a combination of PSCT and infustion of vaccine primed autologous T cells. Blood samples are collected before and after the priming vaccination and at regular time points following transplant, and they are assayed for minimal residual disease (MRD) and T cell repertoire by high-throughput sequencing of rearranged immunoglobulin heavy chain (IgH) and T cell receptor (TCR) genes. Methods: We developed a sequencing-based platform, LymphoSIGHT, for MRD quantification in lymphoid malignancies. Using universal primer sets, we amplify rearranged IgH variable (V), diversity, and joining (J) gene segments from genomic DNA. To minimize the chance of somatic hypermutation interfering with detection of a cancer sequence, each IgH sequence is amplified by different sets of PCR primers in the three framework regions of the V segments and a common J segment primer. Amplified products can be sequenced to obtain >1 million reads and are analyzed using algorithms for clonotype determination. Tumor-specific clonotypes are identified for each patient based on their high frequency in the original tumor specimen. Quantitative MRD levels are then determined in serial samples of peripheral blood using spiked-in reference sequences. Our test has a sensitivity of 1 tumor cell per million leukocytes. This technology has also been extended to immunoglobulin light chain and T cell receptors. To quantify T cell responses to tumor vaccinations, the same samples are used for amplification, sequencing and analysis of the entire TCRB repertoire allowing for the assessment of T cell immune responses to the vaccine. Results: To date we have analyzed serial samples from 13 MCL patients who received the protocol of induction chemotherapy, vaccination, and PBST followed by vaccine primed T cell infusions and booster vaccinations. We have a pre-planned landmark analysis of MRD at 12 months post PBST. MRD was detected in blood samples immediately post-transplant in 3 patients, 2 of whom ultimately relapsed (Fig 1). In these two cases, disease was detected by sequencing 14 months and 4.5 months prior to detection by radiologic techniques. Clinical relapse in the second patient was originally restricted to the CNS; however, at a later point lymph node relapse occurred. The third patient has not shown signs of clinical relapse. The remaining 10 patients were MRD negative at 1 year following transplant, and these patients have not shown signs of clinical relapse with a median follow up of >24 months. To identify T cells specific to the vaccination, we searched for clonotypes that were highly enriched (>10X) in an in vitro tumor-stimulated culture. In 2 of 3 patients assayed, we identified such clonotypes. The enrichment of these same clonotypes was also seen in the patients after vaccinations and boosters, adding to the evidence that they are directed against the vaccine. An example for one such patient is shown in Fig. 2. These clonotypes increased in frequency significantly after the boost by comparison to a set of frequency-matched unrelated clonotypes (p Conclusions: Our high-throughput sequencing method for MRD quantification shows promise for predicting clinical relapse in MCL patients after hematopoietic cell transplantation. This method provides detection lead-time that may be superior to clinical methods such as PET and MRI scanning but further follow up and greater numbers of patients will be required. Using the sequencing method, 77% of patients (10/13) were MRD negative at the landmark of 1 year post-transplant. Continued follow-up for molecular and clinical relapse is ongoing. T cell repertoire analysis identified clonotypes responding to the vaccination in some patients and follow up analyses will determine whether the presence of these clonotypes correlates with clinical outcomes in MCL patients. Disclosures: Faham: Sequenta, Inc.: Employment, Equity Ownership, Research Funding. Carlton:Sequenta, Inc.: Employment, Equity Ownership, Research Funding.

Abstract 4190 Background: Chronic graft-versus-host disease (cGvHD) and its treatment result in significant morbidity and mortality, limiting the clinical benefit of allogeneic hematopoietic cell transplantation (HCT). The diagnosis of cGvHD can be challenging especially when diagnostic features are absent or when the clinical features are confined to internal organs (i.e., lungs). A noninvasive, diagnostic and prognostic test is needed to determine which pts have or will develop cGvHD. Objective: To identify candidate blood biomarkers using gene expression profiling of the circulating donor immune cell repertoire following allogeneic HCT and develop a classification rule to distinguish cGvHD from non-GvHD pts. Methods: In this cross sectional discovery phase study we identified 63 pts (median age 50 yrs; range 19–64) who were beyond 100 days from transplantation (median 424 days) with complete (100%) donor CD3+ T cell chimerism. Pts were divided into cGvHD and non-cGvHD groups based on conventional criteria. To adjust for the influence of immune suppression (IS) medication a prospective cohort of 7 new onset cGvHD pts who were free of IS drugs at the time of onset of cGvHD and at the time of blood sample acquisition. Peripheral blood mononuclear cells were obtained from the 70 pts on IRB approval protocols. The RNA was processed on the Agilent whole human genome 44k microarray platform. Bioinformatics programs including AILUN, GeneSpring, SAM, and PAM. Results: Three data sets were constructed to address the comparison of cGvHD pts on IS medication to drug free non-GvHD pts. First a ‘screening set’, compared gene expression levels in pts with cGvHD on IS (n=14) to those with cGvHD not yet on IS (n=7) using a two-sample t-test and eliminated more than 27k genes whose t-test statistics were beyond the interval of [-1,1] as differences in expression were possibly related to the presence or absence of these medication. We next applied the Prediction Analysis of Microarray (PAM) method to a ‘training set’ to discriminate 21 GvHD pts from 21 non-GvHD using the 16,103 selected probes from the screening set and found a minimum misclassification error of 6 samples was achieved with a probe set that contained 10 genes. The candidate genes represented a compensatory response to cGvHD rather than a pro-inflammatory gene signature. We next evaluated the performance of the classification rule from the ‘training set’ to differentiate cGvHD pts from non-GvHD pts in the 14 pt ‘test set’ and correctly predicted 3 new cGvHD and all 7 non-GvHD controls. We repeated PAM and adjusted for possible confounding variables, that included length in days of sample acquisition date relative to the transplant date, recipient age, donor gender, disease histology (myeloid or lymphoid malignancy), related or unrelated donor to recipient status, and type of conditioning regimen (full dose or reduced intensity conditioning) and observed a result similar to that from the unadjusted analysis. Specifically, adjusted PAM achieved a minimum cross-validated misclassification error rate of 9 among the entire cohort of 70 samples. Finally, the entire analysis was bootstrapped and cross-validated with similar results obtained; the genes with the top frequencies for discriminating cGvHD were over-expression of IL1R2, ADAMTS2 and TS3, TPST1, SESN1, AREG, IRS2, GPER, BCAT1 and CxCR7. IL1R2 is the decoy receptor for the pro-inflammatory IL1 cytokine, ADAMTS2 and TS3 are pro-collagen modifying enzymes that regulate the clipping of collagen molecules to impart structure to connective tissue, AREG is a key gene that underlies the development of oral lichen planus, and pathway analysis revealed that the combination of the other over-expressed genes converge through IL-4, Il-6 and IL-10 signaling pathways which are considered GvHD blocking signals and controllers of inflammation. Conclusion: In the current discovery phase study we used expression profiling of peripheral blood mononuclear cells combined with biostatical analyses and identified several candidate biomarkers that associated with chronic GvHD. The genes and signaling pathways highlighted in the current analysis suggest that compensatory responses that control inflammation or are involved with profibrotic matrix remodeling represent a dominant GvHD “gene footprint”. The identified genes have biological relevance to chronic GvHD and may further the understanding of its pathogenesis. Disclosures: No relevant conflicts of interest to declare.

2043 Background: The chemokine receptors CXCR4 and CXCR7, and their shared ligand CXCL12/SDF-1, have been implicated in multiple aspects of tumorigenesis, including tumor cell proliferation, angiogenesis, vasculogenesis, and metastasis. Recently, a role for CXCR4 and CXCR7 has been postulated in the control of glioblastoma growth and vasculogenesis. We set out to define the expression patterns of these proteins on primary glioblastoma tumors and associated vasculature. Methods: First, a tumor microarray containing grades 1, 3, and 4 glioma samples were analyzed by immunohistochemical techniques for the expression of CXCR7. Second, two glioblastoma microarrays were analyzed by the same technique for CXCR7, CXCR4, and their shared chemokine ligand CXCL12. The same microarrays were also probed for the expression of the additional CXCR7 ligand, CXCL11/ITAC, as well as the chemokine receptor CXCR3, which is also a receptor for CXCL11. Tissues were scored in a blinded fashion by a certified neuropathologist for both the intensity of signal and location of signal for all proteins tested. Results: CXCR7 expression was largely dependent on tumor grade, as the receptor was usually absent on most grade 1 gliomas but was moderately-to-highly expressed in the majority of anaplastic astrocytomas and glioblastomas. CXCR7 expression on tumor-associated vasculature was seen in all samples and its expression level also increased with glioma grade. In contrast to CXCR7, CXCR4 expression in glioblastoma was usually limited to the glioma cells, with only infrequent expression on the vasculature. Conversely, CXCL12 expression in glioblastoma was usually limited to the vasculature. CXCR3 was detected in a subset of glioblastomas, and only on the glioma cells. The CXCR7 and CXCR3 ligand CXCL11 was, like CXCR7, expressed on both the glioma cells and vasculature. Conclusions: These results, in light of prior evidence for CXCR7 in tumor growth, suggest a positive role for CXCR7 in the development of glioblastoma. Because CXCR7 colocalizes with its ligands, CXCL12 and CXCL11, as well as their other receptors, CXCR4 and CXCR3, the function of CXCR7 in glioblastoma may lie in the regulation of both the CXCR4/CXCL12 and CXCR3/CXCL11 axes.

e21071 Background: Infiltration of specialized immune cells regulates the growth and survival of neoplasia. In a survey of public whole genome expression datasets we found that the gene for chemerin, a widely expressed endogenous chemoattractant protein, is downregulated in melanoma as well as other human cancers. Moreover, high chemerin mRNA expression predicted improved outcomes in human melanoma. Methods: Forced re-expression of chemerin by tumor cells was achieved by transfection. An alternate strategy involved direct intratumoral injection of recombinant chemerin into tumors. Tumor line expression of chemerin was confirmed by ELISA and functionality of expressed, secreted chemerin confirmed by chemotaxis of CMKLR1 (the main receptor for chemerin) positive cells. Tumors were evaluated by FACS and immunofluorescence for presence of recruited leukocytes. Results: In studies of the B16 transplantable mouse melanoma, tumor-expressed chemerin inhibited in vivo tumor growth without altering in vitro proliferation. Growth inhibition was associated with an altered profile of tumor infiltrating cells with an increase in natural killers (NK cells), and a relative reduction in myeloid derived suppressor cells and putative immune inhibitory plasmacytoid DC. Tumor inhibition required host expression of CMKLR1, the chemoattractant receptor for chemerin, and was abrogated by NK cell depletion. Intratumoral injection of chemerin also inhibited tumors, suggesting the potential for therapeutic application. Conclusions: We therefore conclude that chemerin recruits anti-tumor immune effector cells, and that it may act as an endogenous tumor suppressor, whose downregulation in melanoma and potentially other neoplasia contributes to immune evasion and tumor growth.

2514 Background: Antibody-dependent cell-mediated cytotoxicity (ADCC), mediated by natural killer (NK) cells, plays an important role in the efficacy of monoclonal antibodies (mAb)s. CD137 is a costimulatory molecule expressed on immune cells following activation, including NK cells. We hypothesize that as the antitumor efficacy of mAbs is due to ADCC, their activity can be enhanced by stimulation of NK cells with an anti-CD137 agonistic mAb. Methods: Upregulation of CD137 on NK cells was assessed using CD20+lymphoma, HER2+breast, and EGFR+head and neck cell lines and primary patient samples. NK cell degranulation, cytokine release and cytotoxicity were assessed by CD107a mobilization, IFN-γ secretion, and chromium release. Mechanism of synergy was explored by cell depletion in an immune competent mouse model. Xenotransplanted models were used to demonstrate anti-tumor activity and sufficiency of an innate immune response. Results: NK cells in human primary patient samples do not express CD137 at baseline, however CD137 is highly upregulated when encountering mAb-coated tumor cells. MAb-induced NK cell degranulation and cytotoxicity are enhanced by anti-CD137 agonistic mAb. In a murine lymphoma model, anti-CD137 mAb significantly enhances anti-tumor activity of anti-CD20 mAb leading to complete tumor resolution and prolonged survival. NK cell depletion completely abrogates the therapeutic effect. In seven xenotransplant models, sequential administration of rituximab, trastuzumab or cetuximab plus anti-CD137 mAb provided superior reduction in tumor burden and prolonged overall survival. In a phase 0 biomarker study, level of CD137 expression on circulating and intratumoral NK cells was influenced by disease burden, prior treatment, FcγRIII polymorphism, and time since mAb therapy. Conclusions: Our results demonstrate the synergy of anti-CD137 mAb and a tumor-targeting mAb by stimulation of mAb-activated NK cells with anti-CD137 mAb to enhance ADCC. These results support a novel, sequential antibody approach against CD20+B cell, HER2+breast, and EGFR+head and neck malignancies by targeting first the tumor and then the host immune system.