Vascular endothelial growth factor (VEGF) inhibits differentiation and maturation of dendritic cells (DC), suggesting a potential immunosuppressive role for this proangiogenic factor. Bevacizumab, sorafenib and sunitinib target VEGF-mediated angiogenesis and are active against several types of cancer, but their effects on the immune system are poorly understood. In this study, VEGF and supernatants of renal carcinoma cell lines cultured under hypoxia were found to alter the differentiation of human monocytes to DC. Resulting DC showed impaired activity, as assessed by the alloreactive mixed T-lymphocyte reaction. Bevacizumab and sorafenib, but not sunitinib, reversed the inhibitory effects of VEGF, but not of those mediated by tumour supernatants. Dendritic cells matured under the influence of VEGF expressed less human leukocyte antigen-DR (HLA-DR) and CD86, and this effect was restored by bevacizumab and sorafenib. Finally, tumour-cell supernatants decreased interleukin-12 (IL-12) production by mature DC, and such inhibition was not restored by any of the tested drugs, delivered either as single agents or in combination. The deleterious effects of tumour-cell supernatants were mainly mediated by thermostable molecules distinct from VEGF. These results indicate that inhibition of the differentiation of monocytes to DC is a multifactorial effect, and that they support the development of combinations of angiogenesis inhibitors with immunological modulators.

Migration of DC into lymphatic vessels ferries antigenic cargo and pro-inflammatory stimuli into the draining LN. Given that tissues under the influence of viral infections produce type I IFN, it is conceivable that these cytokines enhance DC migration in order to facilitate an antiviral immune response. Cultured lymphatic endothelium monolayers pretreated with TNF-a were used to model this phenomenon under inflammatory conditions. DC differentiated in the presence of either IFN-a2b or IFN-a5 showed enhanced adhesion to cultured lymphatic endothelial cells. These pro-adhesive effects were mediated by DC, not the lymphatic endothelium, and correlated with increased DC transmigration across lymphatic endothelial cell monolayers. Transmigration was guided by chemokines acting on DC, and blocking experiments with mAb indicated a role for LFA-1. Furthermore, incubation of DC with IFN-a led to the appearance of active conformation epitopes on the CD11a integrin chains expressed by DC. Differentiation of mouse DC in the presence of IFN-a also increased DC migration from inflammed footpads toward popliteal LN. Collectively, these results indicate a role for type I IFN in directing DC toward LN under inflammatory conditions.

Twenty-four patients with metastatic cancer received two cycles of four daily immunizations with monocyte-derived dendritic cells (DC). DC were incubated with preheated autologous tumor lysate and subsequently with IFN-α, TNF-α, and polyinosinic:polycytidylic acid to attain type 1 maturation. One DC dose was delivered intranodally, under ultrasound control, and the rest intradermally in the opposite thigh. Cyclophosphamide (day -7), GM-CSF (days 1-4), and pegIFN alpha-2a (days 1 and 8) completed each treatment cycle. Pretreatment with cyclophosphamide decreased regulatory T cells to levels observed in healthy subjects both in terms of percentage and in absolute counts in peripheral blood. Treatment induced sustained elevations of IL-12 in serum that correlated with the output of IL-12p70 from cultured DC from each individual. NK activity in peripheral blood was increased and also correlated with the serum concentration of IL-12p70 in each patient. Circulating endothelial cells decreased in 17 of 18 patients, and circulating tumor cells markedly dropped in 6 of 19 cases. IFN-γ-ELISPOT responses to DC plus tumor lysate were observed in 4 of 11 evaluated cases. Tracing DC migration with [(111)In] scintigraphy showed that intranodal injections reached deeper lymphatic chains in 61% of patients, whereas with intradermal injections a small fraction of injected DC was almost constantly shown to reach draining inguinal lymph nodes. Five patients experienced disease stabilization, but no objective responses were documented. This combinatorial immunotherapy strategy is safe and feasible, and its immunobiological effects suggest potential activity in patients with minimal residual disease. A randomized trial exploring this hypothesis is currently ongoing.

BackgroundInterleukin-8 (IL-8, CXCL8) is readily produced by human malignant cells. Dendritic cells (DC) both produce IL-8 and express the IL-8 functional receptors CXCR1 and CXCR2. Most human colon carcinomas produce IL-8. IL-8 importance in malignancies has been ascribed to angiogeneis promotion.Principal FindingsIL-8 effects on human monocyte-derived DC biology were explored upon DC exposure to recombinant IL-8 and with the help of an IL-8 neutralizing mAb. In vivo experiments were performed in immunodeficient mice xenografted with IL-8-producing human colon carcinomas and comparatively with cell lines that do not produce IL-8. Allogenic T lymphocyte stimulation by DC was explored under the influence of IL-8. DC and neutrophil chemotaxis were measured by transwell-migration assays. Sera from tumor-xenografted mice contained increasing concentrations of IL-8 as the tumors progress. IL-8 production by carcinoma cells can be modulated by low doses of cyclophosphamide at the transcription level. If human DC are injected into HT29 or CaCo2 xenografted tumors, DC are retained intratumorally in an IL-8-dependent fashion. However, IL-8 did not modify the ability of DC to stimulate T cells. Interestingly, pre-exposure of DC to IL-8 desensitizes such cells for IL-8-mediated in vitro or in vivo chemoattraction. Thereby DC become disoriented to subsequently follow IL-8 chemotactic gradients towards malignant or inflamed tissue.ConclusionsIL-8 as produced by carcinoma cells changes DC migration cues, without directly interfering with DC-mediated T-cell stimulation.

Dendritic cells (DC) are endowed with the ability to cross-present antigens from other cell types to cognate T cells. DC are poised to meet polymorphonuclear leukocytes (PMNs) as a result of being co-attracted by interleukin-8 (IL-8), for instance as produced by tumor cells or infected tissue. Human monocyte-derived and mouse bone marrow-derived DC can readily internalize viable or UV-irradiated PMNs. Such internalization was abrogated at 4°C and partly inhibited by anti-CD18 mAb. In mice, DC which had internalized PMNs containing electroporated ovalbumin (OVA) protein, were able to cross-present the antigen to CD8 (OT-1) and CD4 (OT-2) TCR-transgenic T cells. Moreover, in humans, tumor cell debris is internalized by PMNs and the tumor-cell material can be subsequently taken up from the immunomagnetically re-isolated PMNs by DC. Importantly, if human neutrophils had endocytosed bacteria, they were able to trigger the maturation program of the DC. Moreover, when mouse PMNs with E. coli in their interior are co-injected in the foot pad with DC, many DC loaded with fluorescent material from the PMNs reach draining lymph nodes. Using CT26 (H-2d) mouse tumor cells, it was observed that if tumor cells are intracellularly loaded with OVA protein and UV-irradiated, they become phagocytic prey of H-2d PMNs. If such PMNs, that cannot present antigens to OT-1 T cells, are immunomagnetically re-isolated and phagocytosed by H-2b DC, such DC productively cross-present OVA antigen determinants to OT-1 T cells. Cross-presentation to adoptively transferred OT-1 lymphocytes at draining lymph nodes also take place when OVA-loaded PMNs (H-2d) are coinjected in the footpad of mice with autologous DC (H-2b). In summary, our results indicate that antigens phagocytosed by short-lived PMNs can be in turn internalized and productively cross-presented by DC.

Various monoclonal antibodies (mAb) target immune system molecules to enhance immunity by costimulating T cells (i.e., CD137, OX40, CD40, GITR) or interfering in coinhibitory signals (i.e., CTLA-4, PD-1). These powerful agents can be guided by cancer vaccines to enhance immunity against tumor but not self tissues. Clinically powerful therapeutic synergies are at hand.

Anti-CTLA-4 monoclonal antibodies (mAb) that block the interaction of CTLA-4 with CD80 and CD86 such as tremelimumab and ipilimumab are currently being tested in the clinic for cancer treatment exploiting their properties to de-repress tumor-specific cellular immunity. Addition of the fully human anti-CTLA-4 (tremelimumab) to cultures of human T cells with allogenic dendritic cells (DCs) did not increase proliferation. Magnetic bead-mediated elimination of CD4 1 CD25 1 regulatory T cells (T reg ) before setting up those alloreactive cultures also largely failed to increase primary proliferation. In contrast, predepletion of CD4 1 CD25 1 T reg and culture in the presence of tremelimumab synergistically resulted in increased proliferation and DC:T-cell aggregation. These effects were much more prominent in CD4 than in CD8 T cells. The synergy mechanism can be traced to enhanced CTLA-4 expression in effector cells as a result of T reg elimination, thereby offering more targets to the blocking antibody. Human T cells and allogenic DCs (derived both from healthy donors and advanced cancer patients) were coinjected in the peritoneum of Rag2 2/2 IL-2Rc 2/2 mice. In these conditions, tremelimumab injected intravenously did not significantly enhance alloreactive proliferation unless T reg cells had been predepleted. Synergistic effects in vivo were again largely restricted to the CD4 T-cell compartment. In addition, T reg depletion and CTLA-4 blockade synergistically enhanced specific cytotoxicity raised in culture against autologous EBV-transformed cell lines. Taken together, these experiments indicate that tremelimumab therapy may benefit from previous or concomitant T reg depletion.T cell depends much on CD28 stimulation by CD80 and CD86 at immune synapses formed between T cells and dendritic cells (DC). 1,2 This costimulatory pathway acts in a complementary fashion to the signals that stream from the TCR upon antigen recognition. 2 This costimulatory activity is regulated by a CD28 homolog that receives the name of CTLA-4 (CD152). 1,2 This molecule has several hundred fold more avidity than CD28 for the CD80 and CD86 countereceptors. 1 CTLA-4 has been demonstrated to be an inhibitory (or co-inhibitory) molecule for T cells in culture. 3 It is expressed on the surface of regulatory T cells (T reg ), and becomes intracellularly expressed in activated effector T cells. Indeed, its absence from the genome in mice determines a severe autoimmunity phenotype that

Stimulating the innate and adaptive immunity against cancer necessitates the tricking of a system evolved to fight microbial pathogens and directing its activity towards transformed self-tissue. Efficacious interventions to start and sustain the response will probably require a number of agents to tamper simultaneously or sequentially with several immune mechanisms. Although master switches controlling various functions may exist, the goal of a curative immune response will probably demand the combined actions of several therapeutic components. Synergy occurs when drugs interact in ways that enhance or magnify one or more effects or side effects. In cancer immunotherapy, two agents that have minor or no therapeutic effects as single agents can be powerful when combined. Mouse experimentation provides multiple examples of synergistic combinations. Elements to be combined include chiefly: tumor vaccines, adoptive T-cell therapies, cytokines, costimulatory molecules, molecular deactivation of immunosuppressive or tolerogenic pathways and immunostimulatory monoclonal antibodies. These novel therapies, even as single agents, are extremely complex products to be developed owing to the associated biomolecules, cell therapies or gene therapies. At present, drug-development programs are run individually for each immunotherapeutic agent and combinations are considered only at a later stage in clinical development, even in the absence of formal compulsory regulations to prevent clinical trials with combinations. As a result, instead of the search for maximal efficacy, ease of combination with standard treatments, intellectual property management, regulations and business-based decisions often guide the way. Even though the maximal effort must be made in order to prevent adverse effects in patients, it seems reasonable that combination pilot trials should be performed at an early stage, following safe completion of Phase I trials. These trials should be performed based on evidence for synergy in animal models and be simplified in terms of regulatory requirements. Such 'short-cut' combination immunotherapy trials can bring much needed efficacy earlier to the bedside.