4- IFN treatment. In the late 1970’s

  4-     Immunotherapeutic approaches Nowadays, type I IFNs are approved for the treatment of a number of cancers ranging from hematological to solid, such as chronic myeloid leukemia, myeloproliferative neoplasms, melanoma, renal cell carcinoma and Kaposi’s sarcoma 83-85 but still, dose-limiting toxicity and the pleiotropic nature of these cytokines oftentimes compromise the success of treatment and their applicability on the clinic. So, in order to avoid systemic toxicity and safely deliver the cytokines to their targets, immunotherapeutic approaches are highly demanded. Recently, the activation of the innate compartment through PRR signaling and the consequent induction of type I IFNs has gained much attention within the field of tumor immunotherapy, with satisfactory tolerability and, in general, no need for specific tumor markers 86. It has been postulated that properly activated innate mechanisms in the tumor microenvironment could determine the success or failure of some forms of immunotherapy, through blocking of immune evasion and activation of adaptive immunity 86. TLRs, RIG-I-like receptors (RLRs) and the stimulator of interferon gene (STING) are prominent candidates and are currently under investigation, so the following sections are a round-up of some of the most recent publications reporting pre-clinical data and available clinical trials on the subject.    The issue of cancer patient tolerability to treatment with IFN-?/? is still not quite solved 85. The biggest challenge today in the therapeutic use of type I IFNs, as well as other cytokines, are the toxic side effects, including fatigue, fever, nausea, depression, leukopenia and others, compromising the efficacy of the treatment and reducing patient’s quality of life 85. As early as the 1970’s, there have been reports documenting IFN-mediated toxicity, initially attributed to the low purity of the IFN preparations 87. However, even more purified preparations still induced the same symptoms, proving to be the main dose-limiting factor 88. The route of administration is also a focal point and intravenous infusion was shown to allow the administration of higher doses in comparison to intramuscular route 89.   Hematological malignancies were the first group to benefit from type I IFN treatment. In the late 1970’s and 1980’s, pioneer clinical studies were conducted to verify the feasibility of use of type I IFNs in the clinic. Gutterman and coleagues reported a favorable response in 2 myeloma patients, one who had been resistant to cytotoxic treatment and the other who had relapsed after cytotoxic treatment 90. Solid tumors, such as renal cell and breast carcinomas, melanoma and lung cancers, were also evaluated as targets for IFN treatment, however, both for hematological and solid tumors, results were timid and not encouraging, with no response in late-stage patients and moderate responses for tumors of viral origin 31. The low rate of success of anti-tumor IFN as a monotherapy drove researchers to seek other strategies to apply this cytokine in the clinic, such as in combination with cytotoxic drugs, or as an adjuvant to radiotherapy or surgery in earlier stages of disease 31. Recombinant DNA technology eventually led to large scale production of pure preparations of IFNs, which were subsequently the first cytokines to be approved as an anti-cancer treatment 39. However, the issue of toxicity still remains and several strategies are still under investigation to circumvent this problem and benefit from IFN signaling in a more “physiological” fashion. An important finding was that the conjugation of polyethylene glycol (PEG) with IFNs reduce both their clearance rate and their immunogenicity, leading to less frequent administration and consequently less adverse side effects 91. Pegylated IFN-?2b (PEG-IFN-?2b) is currently the main choice for the long-term treatment of viral hepatitis, presenting with less toxicity than non-pegylated form 92.     4.1 – Toll-like Receptors (TLRs) agonists There are 10 different TLRs characterized in humans, each specialized in the recognition of different pathogen-associated molecular patterns (PAMPs). TLRs 1, 2, 4, 5 and 6 are found in the cell surface, while TLRs 3, 7, 8 and 9 are expressed in the cytosol, on endosomal membranes 86. Bacillus Calmette-Guérin (BCG), monophosphoryl lipid A (MPL) and imiquimod, which signal through TLRs 2/4, 4 and 7, respectively, have been approved for treatment of bladder, breast and other types of cancer 93-95.   Phase I/II clinical trials evaluating intratumoral administration of oligodeoxynucleotides containing unmethylated cytosineguanosine motifs (CpG-ODN), a TLR9 agonist, for the treatment of neurological malignancies have been conducted, with reasonable tolerability but modest results 96, 97. Intratecal and subcutaneous injection of this compound were well tolerated by patients with neoplastic meningitis, with lymphopenia and inflammatory reactions being the most significant symptoms 98. Association of oligodeoxynucleotides with bevacizumab improves median survival, highlighting the advantages of combining different immunotherapeutic approaches. Both intradermal and intramuscular injection of a 9-polipeptide vaccine derived from breast carcinoma, along with a helper tetanus peptide and TLR3 ligand Poly-ICLC has minimal toxicity to patients but very low immune responses to 2 out of 9 vaccinated peptides 99. Sato-Kaneko et al demonstrated, in a preclinical model of cutaneous squamous cell  carcinoma, that combination of checkpoint inhibition with anti-PD-1 antibody and TLR7 and 9 agonists enhanced the anti-tumor properties of either agent alone, both at injection and distant sites. This effect correlates with differentiation of M1 macrophages and infiltration of IFN-? producing CD8+ T cells in the tumor and spleen 100. Biweekly injections of a TLR7 agonist called 852A in heavily pre-treated, high risk chronic lymphocytic leukemia patients was well tolerated, induced the production of inflammatory cytokines and IgM 101. Interestingly, in a single patient, exposure to the TLR7 agonist seemed to render drug-resistant tumor cells more sensitive to a vincristine-based chemotherapeutic regimen. Indeed, there have been numerous reports of synergy between IFNs and cytotoxic drugs (adressed in this book by Malvicini et al on Chapter 11), as well as of a chemosensitizing property of type I IFNs 102-105. The emerging field of nanotechnology/nanomedicine is investing in TLR-based immunotherapies, in order to precisely deliver agonists to their cellular targets. For instance, encapsulation of resiquimod, a TLR7 ligand, into pegylated polymer-based nanoparticles are  successfully uptaken by APCs, including DCs, which home to draining lymph nodes 106. Such an approach could be a tool to enhance specific anti-tumor T cell responses especially when in combination with other immunotherapies like antigen vaccination, for example. However, TLRs signaling in immune cells and also in cancer cells is a complex network that has not been fully elucidated. In fact, activation of TLRs in malignant cells can lead to tumor promoting effects such as immune evasion, chronic inflammation and metastatic dissemination 107, 108.     4.2 – RIG–like Receptors (RLRs) agonists      RIG-like receptors are a family of PRR specialized in the sensing of cytoplasmic viral RNA. They are members of the DExD/H box RNA helicases and are divided into 3 subgroups: retinoic acid–inducible gene I (RIG-I), melanoma differentiation–associated factor 5 (MDA5), and laboratory of genetics and physiology 2 (LGP2), which lacks N-terminal caspase-recruitment domains (CARDs) and was originally identified as a negative regulator of RLR signaling, but has been shown to synergize with MDA5 109-111. Currently, the use of RLR ligands in the treatment of cancer has not been approved by the FDA, but growing evidence suggests that their activation and consequent signaling can trigger beneficial effects, like the preferential induction of cell death in malignant cells via IFN-dependent and independent mechanisms and immunostimulatory effects over APCs and NK cells 112, 113, 86.   As reviewed by Wu et al, RIG-I and MDA5 activation is able to induce tumor cell apoptosis in melanoma, prostatic, breast, neurological, gastric and hepatic cancers 109. Since RIG-I preferentially binds to RNA sequences with 5′ triphosphorylated (5′ ppp) ends, many silencing RNA molecules have been investigated in order to achieve gene silencing as well as RIG-I activation. For example, murine models show that the pancreatic cell line Panc02 treated with silencing RNA show increased expression of IFN-? mRNA, IL-6 and IP-10 (an IFN regulated chemokine that attracts CD8+ T cell via binding of CXCR3) 114 as well as  induction of cell death with immunogenic features 115. This last effect can be abrogated after RIG-I or MDA5 silencing. Moreover, culture of CD8+ DC with treated Panc02 cells improves their maturation, making them more efficient in the cross-presentation of tumor antigens to CD8+ T cells. Finally, in vivo vaccination with 5′-ppp RNA-treated Panc02 cells renders mice immune to a subsequent challenge and therapeutic administration of poly(I:C)(a MDA5 ligand) decrease tumor burden in tumor bearing mice. Very similar results were obtained with transfection of different human pancreatic cancer cell lines with poly(I:C) complexed with lipofectamine (to deliver it to the cytosol and bind to RLRs), or with systemic administration of PEI-conjugated poly(I:C) to mice bearing Panc02 tumors 116. The use of a glutaminase silencing 5’ppp RNA both inhibits this essential enzyme and induces type I IFN production 117. This silencing RNA acts through intrinsic apoptotic mechanisms in tumor cells only, with no cytotoxic effect in non-transformed cells. Additionally, silenced cells produce IFN-? and IP-10 and express more MHC class I and Fas molecules, facilitating CTL-mediated killing. RIG-I signaling also induces production of reactive oxygen species and impairs autophagic degradation of damaged mitochondria, leading to tumor cell death.   4.3 – Stimulators of Interferon Genes (STING) agonists       The STING receptors are located in the membranes of the endoplasmic reticulum and their signaling pathway is triggered via sensing of DNA in the cytosol by cyclic GMP-AMP synthase (cGAS). STING activation can lead to type I IFN production via IRF3 or to secretion of pro-inflammatory cytokines via NF-?B 118. Thus, STING is thought to be involved in the genesis of DNA-mediated inflammatory disorders such as systemic lupus erythematosus or Aicardi-Goutières syndrome 119, 120. In cancer, its importance comes mainly from the fact that in the tumor microenvironment, DNA released by dying tumor cells or DNA containing vesicles can gain access to the cytosol of infiltrating DCs, which in turn initiate an anti-tumor immune response 121. Utilizing Ifnar-/- mice or by administering IFN-blocking antibodies, researchers have observed that radiation-mediated anti-tumor responses are dependent on type I IFN signaling 122, 123. Moreover, given that the myeloid differentiation primary response gene 88 (MyD88) molecule, a downstream effector of TLR signaling, is essential for the efficacy of chemotherapy, researchers have investigated whether that is the case for radiation therapy as well. However, Deng and colleagues found that anti-tumor responses following radiation between MyD88-/- and WT animals were very comparable 123. Importantly, the authors also observed the in tumor-bearing, irradiated mice that had STING signaling knocked out, the anti-tumor effects of radiation were impaired, through abrogation of IFN-? production by tumor infiltrating DCs and ineffective cross-priming activity. Knocking out IRF3, a downstream target of STING activation, had similar effects. Indeed, other studies with gene-targeted mice demonstrated that deletion of STING or of IRF3 rendered mice incapable of rejecting transplanted immunogenic tumors due to inefficient priming of CD8+ T cells by DCs in the tumor tissue, while no such effect was observed through deletion of TLR, MyD88, IRF7 or mitochondrial antiviral signaling protein (MAVS), these last two being downstream targets of RLR signaling  124, 123. Taken together, these results highlight the role of STING as the predominant innate immune pathway of tumor detection and rejection in vivo.   Intratumoral injection of the flavonoid 5,6-dimethylxanthenone-4-acetic acid (DMXAA), a STING agonist compound, promotes potent tumor rejection of B16, TRAMP-C2 and 4T-1 tumors, induces long lasting immunologic memory and increases frequency of IFN-? producing tumor-specific CD8+ T cells in the spleen. Conversely, STING deficient mice are refractory to this agent 125. Tumor infiltrating macrophages and DCs respond to  treatment with STING agonists by producing high concentrations of type I IFNs, and the main effector cells were found to be CD8+ DCs.  In addition, a synthetic modified cyclic dinucleotide molecule induces IFN-? production in human peripheral blood mononuclear cells, indicating that this pathway could be an effective target for novel immunotherapies. The phenomenon of cellular senescence has been described as an anti-tumor mechanism, given its ability to promote cell cycle arrest, preventing damaged and potentially mutated cells from proliferating. Gluck and colleagues demonstrated that cGAS knockout MEFs do not undergo senescence in the same fashion as their WT counterparts. Gene expression profiling revealed that shutting down the cGAS-STING pathway prevented MEFs from expressing crucial senescence-regulating genes 126. Moreover, cGAS or STING deletion in MEFs or human cell lines prevented them from entering into senescence under conditions of oxidative stress, but treatment with IFN-? in vitro reversed this condition. In an in vivo model, cGAS knockout and WT mice were injected with transposons encoding NrasG12V and markers of senescence were analysed 6 days later. The livers of cGAS knockout mice displayed decreased levels of the molecules p21 and senescence associated ?-galactosidase and these animals had limited capacity to produce the cytokines and chemokines of the senescence-associated secretory phenotype (SASP), with immunostimulatory properties. As a result, clearance of NrasG12V + cells was impaired in these animals.            Recently, a dual role has been ascribed to STING signaling. Liang et al observed that following irradiation of MC38 tumor bearing mice, there is an accumulation of myeloid derived suppressor cells (MDSCs) in the tumor microenvironment that relies heavily in CCR2 induction by STING activation and type I IFN signaling 127. MDSCs impair T cell-mediated anti-tumor responses, promoting radioresistance. So, in order to benefit from the immunostimulatory potential of STING while restraining its regulatory mechanisms, they report that administration of anti-CCR2 antibody combined with radiation and cGAMP (cyclic guanosine monophosphate–adenosine monophosphate, a secondary messenger of the STING pathway) depleted CCR2+Ly6chi population in tumors, enhanced the CD8+ /MDSC as well as CD8+/T regulatory ratios and promoted tumor rejection in 60% of the animals.     5 – Concluding Remarks The process of tumorigenesis and the evolution of cancer have long been disregarded as dependent solely on the actions of proliferating transformed cells.  The concept of tumor microenvironment has brought us the knowledge that non-transformed cells are active participants in the phenomena of tumor growth and metastatic dissemination. The tumor microenvironment allows the interaction of tumor cells and non-malignant host cells, ultimately leading to chronic inflammation, angiogenesis and immunosuppression or tumor elimination. Amongst these host cells, components of the IS strongly interact with tumor cells in a dynamic process termed cancer immunoediting, in which the IS is alerted to the presence of malignant cells and elicits a response. The tumor cells, however, develop several counter-measures to evade the attacks of the IS, which can lead to loss of control and progression of the disease. Thus, a modern understanding of the challenges faced by cancer immunotherapy has to incorporate that not only an immune-mediated anti-tumor response must be generated, but also, the immunosuppressive barriers of the tumor microenvironment have to be breached. Type I IFNs have distinct characteristics that render them important tools in the development of new therapeutic strategies. They have been linked to direct anti-proliferative properties over tumor cells, enhancement of immunogenicity by up-regulation of MHC class I molecules, induction of tumor cell senescence and death with immunogenic features. They also synergize with cytotoxic drugs that are already used in the clinic. Moreover, they have the astounding ability to drive the anti-tumor immune response by modulating the activity of many of its key components, such as NK and B cells. More importantly, they have a tight association with the action of DCs, the main APCs and orchestrators of the IS, which generate highly potent tumor-specific CTLs. In addition, conventional treatments such as radiation and chemotherapy rely on type I IFN signaling to promote the elimination of transformed cells. However, the vast range of biologic effects and complex networks of signaling and feedback loops triggered by type I IFNs are complicating factors that need to be elucidated to circumvent the issues of toxicity and find very specific, effective targets to drive our attention to. Emerging fields such as gene therapy and nanomedicine are promising areas that could effectively harness the potential of type I IFNs and develop more applicable technologies in the future.