Redefining cancer immunotherapy—optimization, personalization, and new predictive biomarkers: 4th Cancer Immunotherapy and Immunomonitoring (CITIM) meeting, April 27–30, 2015, Ljubljana, Slovenia

Introduction

It has become increasingly accepted that the immune system plays a key role in controlling malignant growth and dissemination, and in many cases, it is able to eradicate cancer. Immunotherapy aimed at increasing natural anti-tumor immunity eliminating immune escape variants and overcoming tumor-induced immunosuppression has been the focus of significant research efforts over the past two decades and remains at the center of attention of basic and clinical immunologists and medical oncologists. Different strategies to boost anti-tumor immunity in many types of cancer have been developed with overall marginal clinical improvement. However, the recent advances in the T cell-based therapy and in cancer treatment with antibodies blocking immune checkpoints proved that immunotherapy is experiencing a major breakthrough and should be re-evaluated as an important treatment modality for cancer.

Redefining cancer immunotherapy, including investigation of molecular pathways of tumor–host interaction, search for biomarkers to monitor treatment and predict responses, and development of combinatorial approaches for cancer therapy, was the main focus of the 4th International Conference “Cancer Immunotherapy and Immunomonitoring (CITIM)”, which was held in Ljubljana, Slovenia, in April 2015. More than 180 participants attended CITIM-2015 conference from about 30 countries with more than 60 regular and short presentations given by expert tumor immunologists and clinical oncologists.

This report provides a short overview of the novel pre-clinical models, therapeutic and immunomonitoring strategies, and clinical achievements discussed at a Keynote lecture, three Keynote presentations and 13 plenary sessions.

The conference was opened by a Keynote lecture delivered by O. Finn (University of Pittsburgh, Pittsburgh, PA, USA), in which prophylactic cancer vaccines were proposed as an alternative approach to traditional cancer vaccination. Immune responses to various common tumor antigens could also be found in healthy individuals who have never experienced cancer. Cancer-associated antigens, including MUC1 and cyclin B1, are differentially expressed on tumor and normal cells, with predominant tumor forms of these antigens expressed in inflammatory and pre-malignant lesions. Prophylactic MUC1 vaccine was used in a clinical trial that tested the feasibility of administering a colon cancer prevention vaccine in patients with a history of advanced colonic polyps. Half of the vaccinated patients responded to the vaccine determined by measuring anti-MUC1 IgG levels and long-term T cell memory tested with a booster injection at 1 year. Compared to responders, non-responders had increased numbers of circulating myeloid-derived suppressor cells (MDSCs) and a myeloid inflammatory gene signature in pre-vaccination peripheral blood mononuclear cell (PBMC) transcriptomes. This trial provided the basis for an ongoing multi-center, randomized, placebo-controlled efficacy trial that will test the ability of the vaccine to reduce polyp recurrence and thus reduce colon cancer risk. O. Finn showed examples of premalignant lesions of many different tumor types that could be considered appropriate settings for preventive cancer vaccines.

A Keynote presentation by W. Zou (Ann Arbor, MI, USA) addressed the role of epigenetic events in the control of the interaction between tumor cells and immune system in the cancer microenvironment with a focus on the mechanisms which epigenetically affect cancer “stemness” and effector T cell tumor trafficking. His work explains why some tumors have rich T cell infiltration (“inflamed”) and others are not (non-inflamed). Epigenetic silencing of TH1-type chemokines determines effector T cell trafficking to the tumor microenvironment. His studies provide new insights into the regulatory mechanisms of the tumor–host interaction and suggest novel combinatorial therapies to target the immune suppressive mechanisms, including Tregs and inhibitory PD-L1 and PD-1 pathway.

In his Keynote presentation, P. thor Straten (University Hospital Herlev, Herlev, Denmark) discussed the recent advances in an adoptive cell transfer using tumor-infiltrating lymphocytes (TIL) in melanoma patients. The results of the phase I/II study and preliminary data from a European phase III study demonstrated promising clinical results, although in some cases the clinical effect was weak and not durable. Suggested ways to improve this therapeutic approach include a better cellular source with increased capacity to home to tumor sites and an improved monitoring of the patients with better characterization of predictive markers for clinical responses.

In his Keynote talk, S. van der Burg (Leiden University Medical Centre, Leiden, The Netherlands) presented new data on the synergy between chemotherapy and vaccination in cervical cancer, which indicated that standard chemotherapy may have an immune stimulatory effect by deletion of suppressive myeloid cells in a mouse tumor model and in patients with metastatic cervical cancer. Comprehensive patient immune monitoring in a clinical trial with standard chemotherapy in combination with HPV16 SLP vaccination (overlapping synthetic long peptides of the oncogenes E6 and E7) confirmed the beneficial effect of myeloid cell depletion associated with a robust induction of sustained HPV16-specific T cell responses.

Antitumor immunity and cancer progression: reject or protect

I. Witz (Tel Aviv University, Tel Aviv, Israel) presented studies indicating that cancer metastases are site-specific and some tissue-specific features may help to create a “niche” for dormant micrometastases. In addition, according to the data obtained from an experimental cancer model, patterns of tumor cell gene expression are different depending on the success of metastasis. In mouse neuroblastoma tumors, which form slowly developing dormant micro-metastases, upregulation of genes associated with cell death and growth arrest predominated, while in larger, macro-metastatic tumors expressed genes were associated with invasion (matrix metalloproteinases, MMP) and survival (hexokinase; mitogen-activated protein kinases, MAPK).

Metastatic dissemination and resistance to immunotherapy are associated with different mechanisms of cancer immune escape. Tumors frequently lose MHC class I molecules on their surface and evade the recognition by cytotoxic T cells. F. Garrido (University of Granada, Granada, Spain) described molecular mechanisms of tumor MHC I loss and possible ways to restore expression either with cytokines, if alterations are regulatory, or by gene transfer in case of permanent genetic defects. He emphasized that the efficacy of cancer immunotherapies, including the novel immune checkpoint inhibitors, is reliant on proper presentation of tumor-associated antigenic peptides to cytotoxic T lymphocytes (CTLs) via MHC I molecules. In a mouse model of methylcholanthrene-induced fibrosarcoma, several isolated tumor cell clones with different MHC I expression patterns demonstrated distinct local tumor growth rates, metastatic capacity, and response to treatment. The importance of defining the status of tumor MHC class I expression in patients before immunotherapy and during the course of treatment was emphasized.

W. Murphy (University of California, Sacramento, CA, USA) presented new evidence suggesting that cytokine therapy induces primary CD4 T cell paralysis. Immunotherapy induced profound cytokine release causing T cell activation, thereby increasing the potential for auto reactivity and needs for control. W. Murphy reported a profound suppression of primary CD4 T cell-mediated immune responses in preclinical models and in patients undergoing high-dose IL-2 therapy. This suppression targeted naive CD4 but not CD8 T cells and was mediated through transient inhibition of the signal transducer and activator of transcription (STAT)-5b transcription factor signaling pathway via suppressor of cytokine signaling-3 (SOCS-3). These events resulted in complete paralysis of primary CD4 T cell activation affecting memory generation.

Importantly, different approaches to induce tumor rejection have been proposed by CITIM speakers. A. Jewett (UCLA School of Dentistry and Medicine, Los Angeles, CA, USA) presented data indicating that split anergy in NK cells, defined as decreased cytotoxicity along with increased secretion of cytokines, leads to generation of regulatory NK cells. She demonstrated that regulatory/split anergized NK cells increase tumor differentiation and prevent tumor metastasis in humanized BLT mice. Generation and use of newly established super NK cells with potent cytotoxic and regulatory functions in BLT humanized mice demonstrated significant lysis, differentiation and control of oral, pancreatic, glioblastoma and lung cancer stem-like tumors. Differentiation of stem-like tumors by the NK cells was mediated by the membrane bound and secreted forms of IFN-γ and TNF-α. In addition, differentiated tumors were unable to proliferate and were resistant to NK cell-mediated cytotoxicity. Depending on the tumor type, NK-differentiated tumors either did not induce or increased cytokine release by the NK cells, leading to the control or increase in inflammation, respectively.

T. Lah (University of Ljubljana, Ljubljana, Slovenia) discussed the role of stem-like cells in the resistance of glioblastoma multiforme (GBM) to chemo/radiotherapies. The group demonstrated that mesenchymal stem cells (MSCs) in the tumor microenvironment may exert anti-tumor effects. Additional studies confirmed that bone marrow- and umbilical cord-derived MSCs possess an intrinsic ability to secrete immunomodulatory cytokines that inhibit cell cycle and induce senescence and differentiation of GBM and GBM stem-like cells, which can be exploited in the design of future cellular therapies targeting glioma.

Adoptive cell therapy utilizing ex vivo differentiated CTLs (Tc1 cells) has shown significant clinical promises for the treatment of established tumors. However, complete responses remain infrequent in most patients, and improvements in this approach are needed. Q. Yi (Lerner Research Institute, Cleveland Clinic, Cleveland, OH, USA) presented novel data demonstrating that IL-9-secreting T cells may have an important role in anti-tumor immunity. Th9 cell-derived IL-9 not only inhibited tumor progression but also promoted greater tumor clearance than Th1 cells. Tumor-reactive Tc9 cells differentiated from CD8 T cells under Th9-polarizing conditions demonstrated greater anti-tumor responses against large established tumors in tumor-bearing mice, suggesting that in patients Tc9 cells may exert superior anti-tumor efficacy upon adoptive transfer.

P. Shrikant (Mayo Clinic-Arizona, Scottsdale, AZ, USA) suggested a novel approach to activate anti-tumor immunity by modulating metabolic programs in antigen-induced CD8 T cells. The activation and differentiation of CD8 T cells in anti-tumor responses was shown to be regulated by uncoupling protein 2 (UCP2), a mitochondrial carrier protein expressed by T cells upon antigen stimulation, which induces a metabolic shift from mitochondrial respiration to glycolysis required for growth of CD8 T cells. Inhibition of UCP2 functions by the drug Genipin changed metabolic profile of CD8 T cells leading to reduced spare respiratory capacity, increased lipid synthesis, decreased fatty acid oxidation, and facilitated glycolysis.

Another potential therapeutic target for improving anticancer cell therapy was proposed by J. Kos (University of Ljubljana, Ljubljana, Slovenia). Cathepsins C and H, expressed in CTLs and NK cells, are involved in processing pro-granzymes into proteolytically active forms, which trigger cell death in their target cells. The activity of cysteine cathepsins in immune cells is controlled by endogenous cystatins, cysteine protease inhibitors, such as cystatin F. It was demonstrated that the level of active cystatin F in NK cells is related to lower expression and activity of cathepsins C and H and impaired NK cytotoxic function. Thus, cathepsins C and H and cystatin F could be therapeutic targets for improving anticancer cell therapy.

Diagnostic and prognostic value of tumor-infiltrating immune cells

The recent findings described by W. Fridman (Cordeliers Research Centre, Paris, France) suggested that the type and phenotypic characteristics of tumor cells were critical for shaping the immune contexture of a given tumor. By comparing the immune infiltrates of pulmonary metastases from colorectal cancer (CRC) and renal cell carcinoma (RCC), the group discovered that a high density of CD8 T cells correlated with good prognosis for CRC metastasis and less favorable outcome for RCC metastasis. In subgroups of RCC patients with high risk of death, tumors showed a higher expression of genes related to inflammation, immunosuppression, and angiogenesis; similar expression pattern was found in cultured RCC cell lines but not in CRC cell lines.

The role of tumor immune infiltrates in different types of malignancy was addressed in greater detail by C. Sautes-Fridman (Cordeliers Research Centre, Paris, France) in her presentation about the status of dendritic cells (DC) and intra-tumoral tertiary lymphoid structures (TLS) as conductors of the antitumor response. TLS-associated mature DCs were linked to a strong Th1 and cytotoxic immune phenotype with the lowest risk of death. Follicular B cell density also correlated with long-term survival of patients with non-small cell lung carcinoma (NSCLC). Hence, TLS have an important role in shaping an effector and memory immune response mediated by T and B cells against this type of cancer.

D. Gutkin (University of Pittsburgh School of Medicine, Pittsburgh, PA, USA) demonstrated that while the clinical and immunological significance of tumor-infiltrating lymphocytes is well recognized, the role and subtypes of tumor-infiltrating myeloid cells (TIMCs) is poorly characterized. However, with improved immunohistochemical techniques, it became possible to identify TIMC subpopulations and their intra-tumoral localization. Together with clinical data, a “Myeloid Immunoscore” with potential diagnostic and prognostic value was proposed.

A. Porgador (Ben-Gurion University of the Negev, Beer Sheva, Israel) discussed the importance of optimizing ways to define specific cancer biomarkers predictive of disease progression and efficacy of individualized therapy. Targeting both the malignant cells and their microenvironment is critically important to achieve effective tumor control and to restrain recurrent cancer and micrometastases. The group developed a new method for normalizing the expression of immune-associated genes (immFocus), employing a whole transcriptome sequencing and showed that it can be used for interrogation of a cancer immunome, for advancing immune-based prognosis and for characterizing immune checkpoints.

Z. Fridlender (Hadassah-Hebrew University Medical Center, Jerusalem, Israel) presented novel data on the role of tumor-related neutrophils in cancer immunology. Circulating neutrophils in tumor-bearing mice and cancer patients can prevent the metastatic process by direct tumor cytotoxicity. It was demonstrated that tumor-associated neutrophils (TANs) polarized by the tumor microenvironment can have either anti-tumorigenic (‘N1’) or pro-tumorigenic (‘N2’) functions. Using a transcriptomics approach in mice, the group confirmed that N1/N2 TANs differ from each other, from naïve neutrophils (NNs) and from the granulocytic fraction of MDSC (G-MDSC).

Tumor-induced immunosuppression: mechanisms and strategies to overcome

D. Gabrilovich (The Wistar Institute, Philadelphia, PA, USA) discussed mechanisms regulating activity of MDSCs with a focus on a novel role of STAT3 in the regulation of these cells in tumor sites. STAT3 signaling plays a critical role in expansion of MDSC. After migration to tumor site, monocytic MDSC (M-MDSC) rapidly differentiate to tumor-associated macrophages (TAMs). However, selective inhibitors of STAT3 activity did not affect the presence of MDSCs and TAMs in tumors. Moreover, STAT3 activity in M-MDSCs in tumors was lower than in M-MDSCs in spleens or peripheral blood, which was caused by hypoxia in a HIF-1α-independent manner but directly controlled by activation of CD45 protein tyrosine phosphatase.

Y. Nefedova (The Wistar Institute, Philadelphia, PA, USA) reported that tumor cell chemoresistance is regulated by myeloid cells in the bone marrow microenvironment. Using an in vivo model of multiple myeloma (MM), she demonstrated that the depletion of CD11b cells resulted in delayed MM growth and increased chemosensitivity, while GM-CSF-driven expansion of myeloid cells led to a reduced survival of MM-bearing mice and diminished efficacy of chemotherapy.

V. Umansky (German Cancer Research Center, Heidelberg, Germany) reported that chronic inflammation in tumor microenvironment is responsible for melanoma progression and resistance to immunotherapy. In the ret transgenic mouse melanoma model, they observed an accumulation of numerous chronic inflammatory factors and highly immunosuppressive MDSCs as well as an upregulation of CD39 and CD73 expression on T cells with effector/memory phenotype and properties at the tumor site. In advanced melanoma patients, the frequency of monocytic MDSCs and serum levels of IL-1β, IFN-γ and CXCL10 were increased. Thus, inflammatory factors and immunosuppressive cells have predictive value for clinical outcome and disease progression.

M. Baniyash (The Hebrew University, Jerusalem, Israel) emphasized the importance of controlling chronic inflammation mediated by MDSCs. Under physiological conditions, bone marrow-derived MDSC-like cells have low suppressive activity, while under exacerbated inflammatory conditions, they undergo differentiation arrest and polarization toward an activated immunosuppressive cells, providing a mechanism of immunological homeostasis. However, chronic inflammation makes MDSC-mediated immunosuppression harmful, since it reduces anti-tumor responses and stimulates carcinogenesis. A prolonged survival of tumor-bearing mice can be achieved by MDSC elimination or induction of their differentiation into a more immune-stimulatory phenotype.

DCs play a fundamental role in antitumor immunity and their activity is regulated by various soluble immunomodulatory factors. M. Shurin (University of Pittsburgh Medical Center, Pittsburgh, PA, USA) demonstrated that different members of the TGF-β superfamily affect DCs in a different mode. TGF-β, Nodal and Activin A exerted different effects on the ability of DCs to activate and polarize T cells in vitro and to induce anti-tumor immunity in vivo. Nodal-treated DCs were weak inducers of anti-tumor immunity in vivo, whereas Activin A increased the anti-tumor potential of DC vaccines. Activin A-treated DCs up-regulate proliferation and survival of effector T cells in vitro via the TNF family cytokines BAFF and APRIL.

S. Ostrand-Rosenberg (University of Maryland Baltimore County, Baltimore, MD, USA) presented new data demonstrating that a soluble form of CD80 (CD80-Fc) can inhibit the programmed death (PD)-1 pathway of immune suppression by binding to PD-L1 on tumor cells and consequently preventing the interaction with its receptor PD-1 on activated T cells. CD80-Fc was demonstrated to be more effective than antibodies to either PD-1 or PD-L1 in sustaining T cell production of IFN-γ in vitro and in extending survival time of tumor-bearing mice. Studies with human T cells demonstrated that CD80-Fc inhibited PD-L1/PD-1-mediated immune suppression and simultaneously delivered co-stimulatory signals to activated T cells thereby amplifying T cell activation.

An alternative strategy of targeting immunosuppressive mechanism using self-reactive T cells was proposed by M. H. Andersen (Copenhagen University Hospital, Herlev, Denmark). They discovered circulating T cells, which lack tolerance to important self-proteins expressed in regulatory immune cells such as indoleamine 2,3-dioxygenase (IDO), PD-L1, and FoxP3. Anti-regulatory T cells (Treg) are able to recognize and eliminate regulatory immune cells and influence general immune reactions. Naturally existing PD-L1-specific cytotoxic T lymphocytes are able to recognize and kill both malignant lymphoma cells as well as normal PD-L1-expressing immune cells.

T. Whiteside (University of Pittsburgh Cancer Institute, Pittsburgh, PA, USA) presented data suggesting that tumor-derived exosomes (TEXs) play a key role in potentiating Treg-mediated suppression. TEXs were previously shown by this group to induce apoptosis of activated CD8 T cells, inhibit NK cell activity, and promote accumulation and expansion of Treg cells. New data demonstrated that TEX, which carry immunosuppressive cargo such as CD39 and CD73 ectonucleotidases (which regulate adenosine production) or TGF-β1, pro-peptide latency-associated peptide (LAP), and active TGF-β1, exert profound inhibitory effects on immune cell functions. TEXs co-incubated with Treg alter the mRNA profile and protein expression in these cells, enhancing their suppressive activity.

An approach to elicit anti-tumor efficacy through antibody-mediated selective depletion of intra-tumoral Treg was presented by N. Wilson (Agenus Bio, Lexington, MA, USA). Rapid and selective elimination of GITR^highOX40^high intra-tumoral Treg, which was correlated with tumor regression, was achieved by agonistic antibodies targeting GITR and OX40 that co-engaged activating FcγRs expressed by effector cells within the tumor.

Regulatory pathways in the tumor immunoenvironment

N. Karin (Technion, Haifa, Israel) gave an overview of the role of chemokines in shaping the tumor microenvironment. Chemokines regulate cell trafficking and are involved in cancer progression. Previous results demonstrated that the CCR2–CCL2 interaction is essential for supporting tumor-associated macrophage mobilization from the bone marrow to the peripheral blood and later recruitment at the tumor site. New observations originating from experimental models of prostate cancer and melanoma showed the key role of CCR5 and its ligands in directing the mobilization of MDSCs from the bone marrow to the peripheral blood, and later their accumulation at the tumor site.

E. Yefenof (Hebrew University-Hadassah Medical School, Jerusalem, Israel) presented the results of a study aimed at identifying factors responsible for resistance of leukemia to the pro-apoptotic activity of glucocorticoid (GC) hormones, which act by binding to the GC receptor (GR), and regulate microRNA expression. Hematopoietic tumors, which respond to glucocorticoids, up-regulate expression of miR-103 that promotes apoptosis, upregulates Bim, a pro-apoptotic protein crucial for GC-induced death, and abrogates c-Myc expression. MiR-103, together with Bim and GR, acts as a “tumor suppressor” that leads to reduced proliferation, cell cycle arrest, and tumor cell death.

R. Wiltrout (National Cancer Institute, Frederick, MD, USA) discussed the role of TNF-family members in the oncogene-driven liver cancer. They linked LTβR and oncogenic AKT signaling in the development of liver cancer. AKT/β-catenin-transfected livers displayed increased expression of LTβ and LTβR. Administration of LTβR antagonist (LTβR-Fc) reduced tumor progression and enhanced survival, while LTβR activation induced proliferation and progression of hepatic tumor phenotypes in an AKT-dependent manner. The transcriptome analysis of samples from patients with intra-hepatic cholangiocarcinoma linked increased LTβR network expression with poor patient survival.

A. Keskinov (University of Pittsburgh, Pittsburgh, PA, USA) discussed the role that the peripheral nervous system plays in the tumor microenvironment. Results revealed that co-interaction between dorsal root ganglion cells and melanoma cells resulted in neuronal and neuroglial activation, the release of myeloid cell-attracting chemokines and the formation of inflammatory-like immunosuppressive microenvironment that supports survival and proliferation of malignant cells in vitro and in vivo. These data introduce a new important player in the tumor microenvironment that can be considered as a new additional target of anticancer therapy.

Chronic inflammation and cancer

R. Apte (Ben-Gurion University of the Negev, Beer Sheva, Israel) presented data on the effects of IL-1β in the tumor microenvironment on the balance between inflammation and anti-tumor immunity. Injection of 4T1 breast cancer cells, which induce spontaneous metastases in the lungs, into IL-1β-deficient mice produced tumors, which regressed without metastatic dissemination into lungs. This effect was associated with CD8 T cell infiltration and affects monocyte differentiation. Less macrophages and more dendritic cells were observed in the tumor microenvironment. These findings suggest that the therapy based on blocking IL-1β could modify the inflammatory responses and lead to a reduced invasiveness of breast cancer cells.

E. Voronov (Ben-Gurion University of the Negev, Beer Sheva, Israel) presented data demonstrating that IL-1α from colon epithelial cells and myeloid cells has dual effects in acute colon inflammation and carcinogenesis. Using mice with a complete deficiency of IL-1α (IL-1α KO mice) or mice with a specific deletion of IL-1α in intestinal epithelial cells (IL-1α^int/Δ) or myeloid cells (IL-1α^Lyz/Δ), they found that the development and prognosis of acute colon inflammation and disease depend on the type of cells secreting IL-1α. IL-1α KO mice and IL-1α^int/Δ mice showed less mortality, while mice with a conditional deletion of IL-1α in myeloid cells showed a higher mortality rate compared to control mice.

L. Eisenbach (Weizmann Institute, Rehovot, Israel) presented results indicating that interferon-inducible transmembrane protein family (IFITM) members play an essential role in ameliorating colitis and preventing colon cancer initiation. By using mouse strains deficient in mIFITM3 or the entire IFITM locus, they showed that the absence of a single IFITM3 allele results in stronger inflammation, while in IFITM3 depleted mice, the pathogenic Th17 pathway is dominating. They further revealed that in agreement with the assumption that inflammation may induce carcinogenesis, IFITM3 inhibit tumorigenesis in a colitis-associated colon cancer model.

The role of microbiota in colon inflammatory microenvironmental damage and carcinogenesis were addressed by L. Vannucci (Institute of Microbiology, Prague, Czech Republic). It was reported that healthy microbiota-containing mucosa of rats expressed more pro-inflammatory cytokines (IFN-γ, IL-1, TNF-α) than the mucosa of germ-free rats, with significant differences in collagen architecture of mucosa between the groups. Significant differences in IL-6, IL-1α and TGF-β1 expression were found during early periods of chronic colitis and bowel carcinogenesis. L. Vannucci emphasized the significance of the evaluation of colon mucosa/stroma for early detection of cancer risk related to colonic microenvironment damage.

A. Ben-Baruch (Tel Aviv University, Tel Aviv, Israel) addressed inflammation-driven tumor-stroma interactions in breast cancer. The group studied the impact of “tumor microenvironment (TME) stimulation” (TNF-α, estrogen and epithelial growth factor) on the metastatic spread of luminal-A breast tumor cells, focusing on the CCR7–CCL21 axis that mediates lymph node infiltration. Human breast tumors of the luminal-A subtype expressed lower levels of CCR7 compared to more aggressive subtypes. While CCR7 over-expression in luminal-A breast tumor cells increased chemotaxis in response to the corresponding chemokine, CCL21, this effect was cancelled by “TME stimulation.” The mechanisms involved were described and the therapeutic implications were discussed.

Cancer immunotherapy: lessons from clinical trials and perspectives

A correlation between antigen-specific T cells, MDSCs, and survival among melanoma and breast cancer patients was discussed in a presentation by G. Pawelec (University of Tübingen Medical School, Tübingen, Germany). CD4 T cells reactive to Her-2 were detected in most patients, whereas half of the patients lacked Her-2-reactive CD8 T cell responses and tended to have higher frequencies of MDSCs. Importantly, the 5-year survival rate of patients who had mounted a CD8 T cell response and had a lower frequency of MDSCs at baseline was 100 % compared to only 38 % in patients without Her-2-reactive CD8 T cells and with higher frequencies of MDSCs.

S. Neelapu (University of Texas, Houston, TX, USA) focused on modulating immune responses in lymphoma. They discovered that in lymphoma, both peripheral and tumor-infiltrating CD4 T cells recognize mutated neoantigens. They are exploring strategies to enhance the activation of these endogenous anti-tumor immune responses using various immune stimulatory and immune checkpoint inhibitor therapies.

G. Chatta (Virginia Mason Medical Center, Seattle, WA, USA) presented an update of existing prostate cancer vaccination strategies and discussed the latest results on improving the efficacy of the existing immunotherapy protocols. Various approaches are being explored to overcome cancer-mediated immunosuppression and to target prostate-specific proteins.

Y. Thanavala (Roswell Park Cancer Institute, Buffalo, NY, USA) discussed ways to overcome immunosuppressive pathways in patients with hepatocellular carcinoma (HCC). Selective in vitro depletion of the immunosuppressive cells, such as Treg and MDSCs, resulted in moderate improvement in T effector cell function in HCC patients. Sorafenib targets multiple kinase receptors including VEGF, c-kit and Flt-3, and these are abundantly expressed on Treg and MDSC. Therefore, it was proposed that this drug may reduce the immunosuppressive burden in HCC patients. This study suggested that sorafenib has immunomodulatory effect from both immunological and clinical perspectives in patients with advanced HCC.

G. Koski (Kent State University, Kent, OH, USA) reported a progressive loss of anti-HER-2 Th1 responses in breast tumorigenesis, which could not be restored by standard therapy (chemotherapy + Herceptin), but could be enhanced by vaccination. Since clinical responses to standard therapy appeared to correlate with a degree of retention of anti-HER-2 Th1 responses, G. Koski proposed that monitoring HER-2 Th1 activity via ELISPOT screening could be used as a predictive tool for the response to standard therapy.

L. Butterfield (University of Pittsburgh Cancer Institute, Pittsburgh, PA, USA) demonstrated that a combination of tumor-associated antigen (TAA)-encoding adenoviral vectors with plasmid DNA or DC improved the efficacy of vaccination and induced activation of polyclonal CD8 and CD4 T cell responses in patients with HCC and malignant melanoma. DCs exposed to alpha fetoprotein (AFP) (commonly secreted by HCC tumors) have a variety of gene expression changes and suppressed function, which may adversely affect immunity and anti-tumor responses in HCC patients.

Optimization of cancer immunotherapy: new combinations, targets, models, and mechanisms

It has been previously described that different types of cancer immunotherapy are able to promote T cell infiltration into malignant lesion, but fail to induce tumor rejection. A. Hurwitz (Agenus, Inc, Lexington, MA, USA) addressed the importance of T cell avidity for target peptides in optimizing T cell responsiveness for adoptive T cell therapy. By comparing three mouse transgenic lines bearing T cell receptor (TCR) transgene specific for a tyrosine related protein-2 melanoma peptide displaying high (“TCRhi”), low (“TCRlo”), or very low avidity (“TCRvlo”), they demonstrated that “TCRlo” T cells displayed reactivity to the antigen but failed to reduce tumor growth, despite infiltration into the tumor. TCRvlo T cells matured with a memory or antigen-experienced phenotype and responded to pro-inflammatory cytokines in the absence of antigenic stimulation. On the other hand, TCRhi T cells delayed B16 melanoma growth but were more susceptible to tolerization, but were also more responsive to CD4 T cell help.

M. Goldberg (Harvard Medical School, Boston, MA, USA) presented evidence indicating that drug-resistant ovarian cancer may benefit from a combinatorial therapy of antibodies against cytotoxic T lymphocyte-associated (CTLA) protein-4 together with decitabine, a DNA methyl transferase inhibitor. The anti-tumor effects of decitabine therapy were demonstrated in a paclitaxel-resistant syngeneic murine ovarian cancer model, in which it extended the survival of mice bearing subcutaneous or orthotopic tumors. This treatment upregulated the expression of chemokines that recruit NK cells and T cells, increased the percentage of NK and CD8 T cells, and decreased the percentage of MDSC in the peritoneal cavity.

T. Sayers (Frederick National Laboratory for Cancer Research, Frederick, MD, USA) proposed an alternative approach to modulate immune-mediated cytotoxicity. Withanolide E, a steroidal lactone from Physalis peruviana, was found to be highly active for sensitizing human cancer cells to TNF-related apoptosis-inducing ligand (TRAIL)-mediated apoptosis by increasing c-FLIP degradation linked to the impairment of chaperone proteins. As c-FLIP has been reported to be an HSP90 client, this provides a potentially novel mechanism for sensitizing cells to TRAIL. Sensitization of human renal carcinoma cells to TRAIL-induced apoptosis by withanolide E and its lack of toxicity were confirmed in animal studies.

Y. Keisari (Tel Aviv University, Tel Aviv, Israel) described novel experimental method for cancer treatment, which simultaneously destroys small tumor lesions by in situ ablation and stimulates anti-tumor immunity by releasing tumor-derived antigens. He described two approaches: pulsed electric current tumor ablation (PECTA) using intra-tumoral electrodes and diffusing alpha-emitters radiation therapy (DaRT) based on insertion of 224Ra-loaded wires into the tumor. Combination of tumor ablation with immunostimulation using CpG or with the inhibition of Treg or MDSC resulted in a strong anti-tumor response, which inhibited metastasis, and extended the survival of tumor-bearing mice.

Gene and drug delivery by electroporation, a method that had shown a great potential in experimental cancer models, was discussed by G. Sersa (Institute of Oncology, Ljubljana, Slovenia). This group evaluated experimental tumor rejection after intra-muscular and intra-tumoral electro-transfer of the IL-12 gene. Intra-tumoral IL-12 gene electro-transfer induced rejection of both treated and distant (untreated) malignant lesions, with up to 100 % tumor regression. Intra-muscular IL-12 gene electro-transfer was also effective. It was proposed that electro-chemotherapy combined with the peritumoral IL-12 gene delivery might be effective in promoting tumor rejection.

R. Heller (Frank Reidy Research Center for Bioelectronics, Old Dominion University, Norfolk, VA, USA) presented novel results demonstrating that the IL-12 gene delivery generates a local and systemic anti-melanoma effect in both preclinical and clinical studies. In B16 tumor-bearing mice, they showed that disease-free long-term survival and protection following subsequent re-challenge was directly related to the expression levels and kinetics of the delivered transgene. In phase 2 clinical trial, 9 of 28 patients (32.1 %) displayed a partial response and 3 of 28 (10.7 %) showed complete response. In addition, it was observed that 59.1 % of patients had a measured regression of at least one non-treated lesion.

A. Mackiewicz (University of Medical Sciences, Poznan, Poland) reported that a combination of hyper-IL-6 gene-modified cell vaccine with breaking tumor-related immunosuppression by surgical removal of metastases led to long-term survival in a significant fraction of melanoma patients enrolled into phase I/II and II clinical trials. This vaccine represents an allogeneic genetic whole cell vaccine composed of two lines modified with Hyper-IL-6 (H6) cDNA. H6 is comprised of IL-6 and its soluble alpha-receptor, and it directly binds to the beta-receptor and activates JAK–STAT pathway. This vaccine induces a strong anti-tumor immunity.

The presentation by U. Gaipl (University Hospital Erlangen, Erlangen, Germany) was focused on the dose-dependent effect of ionizing radiation on inflammatory events in benign inflammatory and in malignant diseases. At low and intermediate doses, the ionizing radiation exerts anti-inflammatory effects in a preexisting inflammatory environment, while at higher doses it promotes inflammation with an impact on carcinogenesis. It was shown that low-dose radiotherapy changes the phenotype of inflammatory cells to an anti-inflammatory response, without affecting the basic functionality of macrophages, e.g., phagocytosis. Importantly, radiotherapy and immune checkpoint inhibitors complement each other in induction of anti-tumor immunity.

F. Mattei (Istituto Superiore di Sanità, Rome, Italy) demonstrated that a combined administration of IFN and 5-Aza-2′-deoxycytidine significantly inhibits the growth of human and murine melanoma cells, both in vitro and in vivo. This effect correlated with a significant increase in the frequency of tumor-infiltrating CD8 T cells and a decrease in MDSCs. When compared to a single treatment, IFN/Aza-treated melanoma cells exhibit a higher apoptotic rate. In addition, using a matrigel-based microfluidic device, it was shown that human immune cells preferentially migrate toward the IFN/Aza-treated melanoma cells rather than tumor cells left untreated.

Importance of immunomonitoring in clinical trials

The conference provided a great opportunity for attendees to learn about novel techniques and immunomonitoring strategies employed in clinical trials and preclinical studies. C. Gouttefangeas (Eberhard-Karls University, Tübingen, Germany) provided an update on the activities of CIMT Immunoguiding Program (CIP) in assay harmonization during immunomonitoring the response to cancer immunotherapy. The harmonization of the main T cell assays, IFN-γ-Elispot, and HLA-multimer staining is well advanced, and CIP is now investigating immunosuppressive cells; the results of a proficiency panel dedicated to the phenotyping of MDSC subsets were presented and some parameters responsible for inter-laboratory variabilities were identified.

P. Lehmann (Cleveland, OH, USA) critically analyzed the accuracy of tumor-associated peptide predictions and specific CD8 cell responses restricted by well-defined HLA class I alleles. They tested 42 healthy human donors on the accuracy of the prediction of 32 individual peptides that comprise the peptide pool containing well-defined peptide determinants for CMV, Epstein Barr, and Flu virus. Overall, the data showed that reliance on one or a few peptides is likely to miss the majority of the antigen-specific CD8 cells, strongly arguing for the use of peptide pools for immune monitoring.

P. Rodrigues-Santos (University of Coimbra, Coimbra, Portugal) shared the details of the immune monitoring protocol of chronic myeloid leukemia treated with tyrosine kinase inhibitors implemented in their laboratory. Human Immunophenotyping Consortium aims at the standardization of protocols of CML immune monitoring based on multicolor flow cytometry using an extended 10-parameter panel for the analysis of T cells (Th1, Th2, Th17, Tregs), B cells (including plasmablasts), monocytes (classical and non-classical), DC (myeloid and plasmacytoid) and NK cells (CD56^bright and CD56^dim).

D. Maurer (Immatics biotechnologies GmbH, Tübingen, Germany) discussed advantages of a novel personalized approach for cancer immunotherapies that has the potential to improve the efficacy of the current immunotherapeutic treatments. Therapeutic targets are specifically tailored to the antigenic fingerprint of a patient’s tumor by individual selection from large target libraries based on the XPRESIDENT™ platform (also known as the Tuebingen Approach)—a combination of quantitative mass spectrometry, differential transcriptomics, immunology and bioinformatics to identify novel peptide-based antigens for immunotherapy in a high-throughput fashion. Tumor-associated peptides (TUMAPs) are eluted directly from primary tumor tissue and sequenced using a high-throughput tandem mass spectrometry approach resulting in thousands of putative peptide sequences.

P. Romero (Ludwig Cancer Research Center, University of Lausanne, Lausanne, Switzerland) gave an overview of the current status of cancer immunotherapy and of the clinical perspectives of measuring and characterizing T cell responses to tumor antigens. He emphasized that the systematic study of T cell responses to tumors in treated patients may provide valuable clues for the identification of immune biomarkers to guide the optimization of novel cancer immunotherapies. Steady progress in basic research on these topics over the last 25 years has enabled the monitoring of T cell responses at high levels of sensitivity, cell type resolution, and detailed functional parameters.

Concluding remarks

The CITIM-conference provided an opportunity for scientists, clinicians, and students to meet and share latest clinical and experimental achievements in the field of immuno-oncology, as well as to re-evaluate the current status of cancer immunotherapy. The main focus of the conference was related to the optimization of the existing protocols of immunotherapy alone or in combination with other types of cancer treatment. Recent clinical success of antibodies blocking immune checkpoint inhibitors has prompted a new wave of enthusiasm in the developments of novel and efficient therapeutic strategies for cancer treatment. Targeting immune regulatory cell activity by the neutralization of immunosuppressive pathways within the tumor microenvironment simultaneously with inducing anti-tumor immune responses has gained even more attention in the preclinical and clinical settings. Further emphasis was placed on improving the methods of immunomonitoring of cancer patients enrolled in clinical trials for predicting the efficacy of immunotherapies. A special consideration was given to the role of tumor-infiltrating immune cells as biomarkers to predict treatment response and anti-tumor immunity. Personalization of each therapeutic approach is expected to deliver superior clinical efficacy and will depend on the discovery of novel biomarkers of disease progression and response to treatment.

List of abbreviations

AFP

Alpha fetoprotein

CRC

Colorectal cancer

CTL

Cytotoxic T lymphocyte

CTLA

Cytolytic T lymphocyte-associated

DC

Dendritic cell

GBM

Glioblastoma multiformae

GC

Glucocorticoids

GM-CSF

Granulocyte–macrophage CSF

HCC

Hepatocellular carcinoma

IDO

Indoleamine 2,3-dioxygenase

IFITM

Interferon-inducible transmembrane protein family

IFN-γ

Interferon gamma

IL-1

Interleukin 1

MAPK

Mitogen-activated protein kinase

MDSC

Myeloid-derived suppressor cell

MHC

Major histocompatibility complex

MM

Multiple myeloma

MMP

Matrix metalloproteinases

MSC

Mesenchymal stem cell

PBMC

Peripheral blood mononuclear cell

PD-1

Programmed death 1 receptor

PD-L1

Programmed death ligand 1

RCC

Renal cell carcinoma

SOCS3

Suppressor of cytokine signaling-3

STAT

Signal transducer and activator of transcription

TAA

Tumor-associated antigen

TAM

Tumor-associated macrophage

TAN

Tumor-associated neutrophil

TCR

T cell receptor for Ag

TEX

Tumor-derived exosomes

TGF

Transforming growth factor

TIL

Tumor-infiltrating lymphocyte

TIMC

Tumor-infiltrating myeloid cell

TLS

Tertiary lymphoid structures

TME

Tumor microenvironment

TNF

Tumor necrosis factor

TRAIL

TNF-related apoptosis-inducing

Treg

Regulatory T cell

UCP2

Uncoupling protein 2

Compliance with ethical standard

Conflict of interest

The authors state no conflicts of interest.