Cancer immunotherapy achieves breakthrough status: 12th annual meeting of the association for cancer immunotherapy (CIMT), Mainz, Germany, May 6–8, 2014
KeywordsCIMT 2014 Cancer immunotherapy Cancer immunology Tumor vaccination
Adoptive cellular transfer
Antibody-dependent cellular cytotoxicity
Chimeric antigen receptor
Association for Cancer Immunotherapy
CIMT Immunoguiding Program
Cytotoxic T-lymphocyte-associated protein 4
Cytotoxic T lymphocytes
DNAX Accessory Molecule-1
Incomplete Freud’s adjuvant
Programmed cell death 1
Programmed death-ligand 1
CIMT Regulatory Research group
T effector cells
TCR-engineered reference samples
T cell immunoreceptor with Ig and ITIM domains
T regulatory cells
Vascular endothelial growth factor A
In December 2013, the Science magazine has designated cancer immunotherapy (CIMT) as the breakthrough of the year mainly due to clinical success stories in the area of checkpoint inhibitory antibodies and cellular gene therapy. This year’s CIMT 2014 meeting covered these recent success stories as well as additional technologies and therapeutic concepts that may form the base for the next wave of innovation to reach patients in the near future. Reflecting the increasing attention the field is attracting, the annual meeting of the Association for CIMT could observe yet another rise in the number of international participants (n = 950). A total of 85 used the opportunity to give lectures and 321 presented their cutting-edge innovations in poster sessions. CIMT offered a sizable stage for the dissemination and intensive discussion of research and development efforts in the areas of tumor vaccination, combination therapy, tumor microenvironment, immunoguiding, immunoinformatics and genomics, cellular therapies and multiple measures to improve immunity, thereby reflecting current trends and the state of the art in the field of cancer immunotherapy. The following is a review of the highlights of the CIMT 2014 meeting.
Immune checkpoint inhibitors and combinations thereof
The potency of checkpoint inhibitors has led to a dramatic change in the field of cancer immunotherapy. Everyone knows about the success story of ipilimumab (Yervoy®) and the excitement induced by novel checkpoint inhibitory antibodies that are now being tested in a series of different cancer entities. Various checkpoint inhibitory antibodies have now delivered further promising data suggesting unprecedented anti-tumoral activity when applied as monotherapy or in combinations. Insights into combinations of various checkpoint inhibitors and other therapies were given by Michael Curran (MD Anderson Cancer Center, Houston, USA). The inhibitory signals of CTLA-4 and PD-1 are not redundant, and blocking one receptor has been shown to lead to up-regulation of the other. Accordingly, blocking both pathways at the same time combined with irradiated B16 melanoma cells expressing Flt3-ligand (Fvax) led to strongly enhanced survival in B16 melanoma models. Notably, adding anti-PD-L1 to the combination of anti-CTLA-4 and anti-PD–1 was favorable with respect to tumor rejection and intra-tumoral effector T cell versus T regulatory (Treg) cells ratio in mice. Blocking of all the three pathways within the PD-1/PD-L1 system (PD-L1:PD-1, PD–L2:PD–1, PD-L1:B7-1) bears the potential for synergistic effects in patients. In a phase I clinical trial on anti-PD-1 and anti-CTLA-4 treatment of melanoma, concurrent treatment was superior to sequential treatment with respect to induction of CD4+ and CD8+ T cell proliferation. CTLA-4/PD–1 double-positive T cells were increased in patients’ blood after treatment, in line with mouse experiments showing that this combination leads to enhanced CTLA-4/PD-1-positive tumor-infiltrating lymphocytes, associated with tumor rejection. Clinical data from cohort of patients indicated that an increased frequency of CD4+ ICOS+ T cells in tumors and/or blood correlates with increased likelihood of overall survival. Multi-dimensional analysis allowing to include even more markers such as the proliferation of effector T cells (granzyme B+, Ki67+) and the highly suppressive Tim3+ Treg (Tim3+, Ki67+) needs to be further tested, but may in the future help identifying the right patient for a given combination of checkpoint inhibitory antibodies. A further combination, anti-CTLA-4 and agonistic anti-4-1BB (CD137), was also synergistic when applied with Fvax in mice. In mouse models, the treatment up-regulated the highly cytotoxic population of KLRG1+ T cells. The master regulator of this T cell phenotype was eomesodermin, a T-box transcription crucial for embryonic development of mesoderm. In line with these non-clinical findings, in a patient treated with anti-CD137, eomesodermin and granzymes were upregulated in CD8+ T cells and NK cells. Notably, combined with anti-CTLA-4, the liver inflammation induced by anti-4-1BB was ameliorated in mice. Given therapeutic synergy and mutual amelioration of adverse events, a clinical trial of this combination appears to be scientifically justified. The positive synergistic effects of known checkpoint inhibitors and the fact that there are many more less well-investigated checkpoint interactions, justify the search for new targets for this class of drugs. This was the focus of the presentation by Mark Smyth (QIMR Berghofer Medical Research Institute, Brisbane, Australia), who introduced CD96 as a novel target in cancer immunotherapy. CD96 occurs on T cells, NK-T cells and NK cells, like its related molecules DNAM-1 (DNAX Accessory Molecule-1) and TIGIT (T cell immunoreceptor with Ig and ITIM domains) and is up-regulated upon IL-2 cultivation of NK cells just like DNAM-1. All three receptors share the ligand CD155, a stress-induced nectin-like surface molecule, whereas DNAM-1 and TIGIT are known to have opposed functions, leading to co-stimulation and inhibition of cellular activation, respectively. To investigate the physiological function of CD96, Mark Smyth and his team generated CD96-deficient mice. While these mice showed apparently normal immune homeostasis and NK cell repertoire and function, they developed a hyper-inflammatory response after LPS challenge. NK cells showed increased IFN-gamma secretion upon stimulation, which could be partly reduced by simultaneous loss of the activating molecule DNAM-1, pointing toward an antagonism between DNAM-1 and CD96. Using a MCA-induced fibrosarcoma model and models of lung metastasis (B16 melanoma and E0771 breast cancer), they showed that tumorigenesis and metastasis formation were inhibited by blocking CD96 function. These effects required functional NK cells, DNAM-1 and IFN-gamma. In the AT3 (breast cancer) model of primary tumor growth, mainly T cells were important for the anti-tumor effect of CD96 blockade. Combination of anti-CD96 with inhibition of CTLA-4, PD-L1 and especially PD-1, was synergistic with respect to the reduction in primary tumor growth and metastasis. Also, combining anti-CD96 with trastuzumab, a HER2-binding monoclonal antibody was favorable in a HER2+ breast cancer model. Altogether, the CD96–CD155 and opposing DNAM-1-CD155 axes are novel promising targets for immune checkpoint control, and it will be very exciting to see upcoming further mechanistic insights and especially the translation into the human system.
While everybody in the field seems to be excited about the opportunities to reach enhanced efficacy of immunotherapies by combining them with each other or with other treatments, the next speaker also raised some concerns. Yutaka Kawakami (Keio University, Tokyo, Japan) pointed out that with the increasing cost of combination therapies the need for rational use according to individual features of a given patient or patient’s tumor becomes a mandatory task. Predictive biomarkers mirroring the patient’s cancer characteristics or immune status can be helpful in this regard, as exemplified by the immunoscore assessing the type and spatial distribution of tumor-infiltrating immune cells or by plasma cytokine patterns. Oncogene/signal activation in cancer cells trigger multiple immunosuppressive cascades in the tumor microenvironment, resulting in a reduction in CD8+ T cell responses. The balance of anti-tumor T cell induction and suppression pathways regulated by cancer cell and host characteristics may define individual immune-conditions. For example, while immunogenic passenger mutations may induce anti-tumoral CD8+ T cells, oncogene activation and driver mutations rather lead to a suppression of anti-tumoral CD8+ T cells. If confirmed, this notion may enable the development of stratifying biomarkers based on the genome-analysis of tumor material. More translational research leading to a better understanding why an individual patient responds to a given treatment based on its molecular signature and enables a rational use of cancer immunotherapies and combinations thereof are highly desirable.
Cellular gene therapy
The adoptive transfer of re-directed T cells that are genetically engineered to recognize tumor-associated molecules either by expression of chimeric or classical T cell receptors is emerging as an effective modality for cancer therapy. Carl June (Abramson Family Cancer Research Institute, Philadelphia, USA), a true pioneer in the field of chimeric antigen receptor (CAR) T cells, reported on recent developments and ongoing work in his group. Clinical trials in adult and pediatric patients with chronic as well as acute lymphocytic leukemia showed unprecedented effects in a high fraction of treated patients as reported in recent milestone publications. In contrast to TCR-engineered lymphocytes, CAR T cells provide the advantage that they recognize cell surface molecules independent of MHC-restrictions. This makes CARs particularly suitable for commercialization in affected patient populations. Adoptive transfer of T cells expressing a CAR-recognizing CD19 on B cell-associated hematologic malignancies (CART19 or CTL019) allows to re-program not only specificity, but also differentiation of the genetically modified T cells through the equipment with a 4-1BB signaling domain (to promote proliferation and survival) in addition to a CD3 zeta chain. Accordingly, it could be shown in clinical trials that transferred CTL019 cells (1.5 × 105–1.2 × 107 cell/kg) can proliferate in the recipient patient 2–4 log reaching very high frequencies of total mononuclear blood cells. Proliferation of the cells in vivo is thereby most likely mediated by healthy B cells as well CD19-positive tumor cells. Interestingly, the frequency of CTL019 reached in the blood seems to distinguish therapy responders from non-responders and may, therefore, become an early indicator of subsequent clinical efficacy. It could be deduced from clinical studies that each CTL019 cell must have killed approximately 1,000 tumor cells.
Hiroshi Shiku (Mie University, Tsu, Japan) presented preclinical and clinical data for an adoptive cellular transfer (ACT) therapy using TCR gene-transduced T cells recognizing a HLA-A*2404-restricted MAGE-A4 epitope. CD8+ T cells showed anti-tumor efficacy in mouse models when combined with either GITR stimulation or peptide vaccination. Based on these data, ACT was combined with vaccination using the TCR’s cognate peptide in a phase I clinical trial in esophageal cancer patients (n = 10) without preconditioning. The safety profile was good, adverse events being mostly skin reactions due to the peptide vaccination. The TCR-engineered lymphocytes peaked in peripheral blood 5–6 days after infusion, persisted for a long time, and T cells from peripheral blood were shown to recognize MAGE-A4+ HLA-A*24+ tumors. Survival in this trial was better than in another trial in esophageal cancer using a NY–ESO–1 protein cancer vaccine, for which an association of higher dose with better survival had been shown. However, randomized trials with a control arm will need to be conducted to capture the effect size of the treatment. Several other ACT trials are planned or ongoing, also employing various preconditioning regimens and targeting further cancer types. With respect to the vaccination administered together with ACT, the Shiku group considers using long peptides (LPs) as the vaccine format. This is based on mouse studies indicating that short peptides (SPs) can lead to apoptosis in CD8+ T cells and induce T cells with low capacity for memory formation. After vaccination of mice with DNA or proteins, the signaling molecule STAP2 (signal transducing adaptor family member 2) was up-regulated in sorted antigen-specific T cells, which was not the case after short peptide vaccination. STAP2 was associated with IL-7R expression in vitro and in vivo. SiRNA targeting STAP2 led to weakened memory T cell responses and failure of T cells to reduce tumor volume. Similar studies will enable the rational selection of the best vaccine format and vaccine scheduling to enhance the number and function adoptively transferred TCR-engineered lymphocytes in vivo and thus increase clinical efficacy of ACT. Finally, Hiroshi Shiku presented data on a novel retroviral vector for TCR transfer, which harbors an siRNA to suppress the endogenous TCR expression. Using this vector led to higher expression of the transduced TCR, and better tumor killing, at low virus copy numbers per cell, pointing toward better safety. The siTCR-engineered T cells showed reduced alloreactivity, providing an opportunity to potentially use allogeneic cells for ACT in the future.
Jürgen Kuball (University Medical Center, Utrecht, The Netherlands) reported on the potential applications of γδT cells in allogeneic stem cell transplantation. Jürgen Kuball and his group found that Vδ2-negative γδT cells were increased in case of CMV reactivation after stem cell transplantation, and that those cells cross-recognized both CMV antigens and leukemic blasts. This explained the earlier finding that CMV reactivation was associated with a decreased relapse rate.
This finding could be used in the clinic applying the concept of innate allogeneic stem cell transplantation. In such a regimen, αβT cells are depleted from the transplant, thereby getting rid of a large source of variability which would contribute a lot to unpredictability of outcomes as only innate immune cells are administered together with the stem cells. Another factor that impacts on the success for transfer of engineered lymphocytes is the conditioning regime applied to prepare the patient for the therapy. One example is the finding that the correct timing of anti-thymocyte globulin (ATG) in a conditioning regimen based on fludarabine, ATG and busulfan played a huge role in the success of engraftment of the transplants. With the appropriate regimen, the Utrecht group achieved both profound host immune cell depletion and rapid reconstitution of donor immune cells without substantial GvHD in patients after allogeneic stem cell transplantation. The immune reconstitution was dominated by Vδ2-negative γδT cells, and no severe CMV reactivation was observed, indicating the profound innate control was sufficient to avoid this.
Another γδT cell-based treatment approach which was presented involves the second subset of γδT cells, which is Vδ2-positive and was also shown to have potent antileukemia activity. Following up on the question how γ9δ2TCR is able to specifically recognize tumors, the group generated first evidence for the role of a small GTPase of the Rho family. Its expression in tumors correlated with the reactivity of γδT cells against primary tumor cells. In the presence of pamidronate, which induces the pyrophosphate antigens recognized by γ9δ2T cells, the GTPase colocalized with butyrophilin 3A1 (BTN3A1), which binds pyrophosphonate antigens and allows then direct interaction of BTN3A1 with the γ9δ2TCR. In conclusion, δ2-positive γδT cells are promising for leukemia therapy as they have the potential to target the leukemic cells, but it may be necessary to enrich for cancer-reactive cells. In a related approach to exploit the capacity of γδT cells to confer anti-tumoral activity, a clinical gene therapy trial with Vδ2 T cell receptors transferred into αβT cells is planned for 2015 in AML patients.
Despite recent setbacks of tumor vaccines that were not able to reach their endpoints in clinical phase III testing, there is still room for developing potent cancer vaccines using novel vaccine formats, drug combinations or personalized approaches. Many different clinical concepts are being investigated, and many clinical trials are ongoing. Therapeutic vaccination with peptides in incomplete Freud’s adjuvant (IFA) is a widely used approach with currently 39 open clinical cancer trials in the USA alone. Willem Overwijk (MD Anderson Cancer Center, Houston, USA) presented his work focusing on the improvement of such IFA-based peptide vaccines. Investigating why many vaccinated cancer patients experience only minor tumor regression despite increased levels of cancer antigen-specific CD8+ T cells using an IFA-based gp100 peptide vaccination in the B16 melanoma model, his work revealed that a significant amount of newly induced T cells are distracted from the tumor to the vaccination site due to high concentration of antigen and the inflammatory environment at the site of injection. Consequently, persisting antigen depots at the vaccination site eventually induce antigen-driven T cell apoptosis and may cause lesions at the injection site. This effect could be reversed using a short-lived saline-based adjuvant, permitting T cell accumulation at the tumor site with minimal T cell activity at the injection site and identifying the non-biodegradable IFA as being responsible for creating the antigen depot at the injection site and the subsequent sequestration of antigen-specific T cells from the tumor site. Addition of an anti-CD40 mAb, the TLR7 agonist imiquimod, and IL-2 to the gp100/saline vaccine resulted in reduced T cell apoptosis and strong anti-tumor activity. Showing that IFA-based vaccination sites also sequester T cells with other tumor antigen specificity that are activated by anti-CTLA-4 therapy, he concluded that synergism with checkpoint blockade depends on the quality and nature of the vaccine and emphasized the implications of these findings for the future development of nonpersisting therapeutic anti-cancer vaccines. Yasuharu Nishimura (Kumamoto University, Kumamoto, Japan) presented his recent work on the identification of cancer–testis (CT) antigen-derived promiscuous CD4+ helper T cell epitopes activating both tumor-specific Th1 cells and cytotoxic T lymphocytes (CTLs) in order to further improve TAA peptide-based cancer immunotherapy. Upon initial identification of three TAA-derived CTL epitope SPs and their assessment for clinical and immunological efficacy in a phase II study in patients with head and neck squamous cell carcinoma, tumor antigen-derived LPs encompassing both HLA class II restricted Th1 cell and CTL epitopes were identified using known CTL epitope sequences and computer algorithm generated data on predicted HLA class II binding peptides. Such a CT-derived Th1 cell epitope LP including a CTL epitope SP induced cross-priming of the respective SP-specific CTLs both in vitro in humans and in vivo in HLA-A24 transgenic mice. The activation of LP-specific Th cells with cognate peptides enhanced the induction of SP-specific CTLs in vitro. Furthermore, LP-specific memory Th cell responses were increased in patients with head and neck squamous cell carcinoma vaccinated with the corresponding three SPs underlining the potential use of TAA-derived LPs containing epitopes that can activate both Th1 cells and CTLs for clinical application. Ulrike Gnad-Vogt (CureVac GmbH, Tübingen, Germany) gave an update on the recent clinical developments for the self-adjuvanted mRNA vaccines (RNActive®) targeting multiple cancer antigens. RNActive®, a mix of free RNA optimized for antigen expression and protamine-complexed RNA to induce an adjuvant effect, is injected intradermally without any transfection agent or further adjuvant. In preclinical tumor challenge experiments, the vaccine-induced CD8+ and CD4+ T cell responses as well as antibody responses and acted synergistically with anti-CTLA4 and anti-PD-1 antibodies. The latter combination was able to achieve tumor eradication and resulted in a protection after rechallenge of mice with tumor cells, even lacking target antigen. The platform technology is currently being evaluated in different clinical studies. While a first phase I/IIa study to treat castration resistant prostate cancer demonstrated that the vaccine CV9103 coding for four tumor-associated antigens has been safe and induces strong T cell responses, a phaseI/IIb trial with the follower vaccine CV9104 was initiated to test the clinical efficacy. As shown in the first trial, immune responses against multiple antigens were associated with longer survival, six antigens are targeted in the CV9104 trial. CureVac also developed RNActive® vaccines for treatment of non-small cell lung cancer. (A phase I study with CV9201 targeting five antigens showed safety and immunogenicity. Currently, the successor product CV9202, targeting six antigens is investigated in a phase Ib trial in combination with radiation. Recently, CureVac has entered a partnership with the Cancer Research Institute and Ludwig Institute for Cancer Research to test RNActive® vaccines in combination with checkpoint inhibitory antibodies. Given the different product candidates that CureVac has brought into early and advanced clinical development, we will soon learn if and how this interesting new technology will translate into benefit for patients. Ugur Sahin (TRON gGmbH, Mainz, Germany) introduced the Ribological® RNA vaccination platform and its application in non-personalized and personalized CIMT approaches. Ribological® RNAs contain several modifications such as a modified 5′-cap structure, additional regulatory sequences and HLA-targeting sequences that result in increased stability and translation as well as improved antigen presentation in dendritic cells generating strongly improved prophylactic and therapeutic anti-tumoral activity in animal models as compared to unmodified mRNA. Strong vaccine-induced T cell responses have been observed in an ongoing clinical phase I trial with a lymph node DC targeting RNA-based cancer vaccine encoding two melanoma-associated antigens. In the ongoing study, patients with advanced melanoma have been treated safely with repeated doses of up to 600 µg RNA injected into inguinal lymph nodes. To further improve RNA-based vaccination and advance from local to systemic DC targeting, a novel liposomal formulation has been developed that allows RNA protection from degradation by plasma RNAses after intravenous application and highly selective RNA vaccine targeting APCs predominantly residing in the spleen. Such RNA lipoplexes induce an IFN-α mediated maturation of targeted DCs and activation of effector cells which led to an enhanced vaccination effect. Preclinical experiments to assess the pharmacodynamic activity of the novel RNA formulation showed enhanced induction of antigen-specific CD8+ and CD4+ T cell responses as compared to naked RNA that was administered into the lymph node. Vaccination with RNA lipoplexes also led to rejection of highly aggressive CT26 lung metastasis in Balb/c mice. The first in human testing of the novel liposomal RNA formulation in a phase I clinical trial in patients with advanced melanoma will be initiated in 2015. With regard to ideal, patient-specific target antigens for therapeutic tumor vaccination in particular and the future of personalized medicine in general, Ugur Sahin and co-workers are currently studying a personalized vaccination approach with IVAC® Mutanome vaccines in a phase I trial in patients with advanced melanoma. Mutanome vaccines are poly-neo-epitopic RNA vaccines targeting the unique tumor mutation signature of an individual patient. Such vaccines are tailored and manufactured on demand for a single patient and administered as an individual treatment. Acknowledging clonal heterogeneity and targeting multiple immunogenic tumor mutations unique to a given patient’s tumor, treatment with such individualized vaccines may lead to highly efficient immune responses as has been shown in preclinical murine models. A first in human clinical trial has been initiated, and five patients have been recruited. Safety, immunogenicity and clinical activity data are expected in 2015.
Immunoinformatics and genomics
More and more, cancer treatments for patients are based on the genetic fingerprint of the individual tumor disease. The genetic information of the cancer is used for stratification, prognosis and target identification. The research and development of cancer immunologists is more and more an interplay of immunology and informatics. Therefore, CIMT hosted a plenary session on immunoinformatics and genomics for the first time. John Castle (TRON, Mainz, Germany) presented how TRON provides the infrastructure and bioinformatic tools to support the clinical translation of individualized immunotherapies. He recapitulated that the landscape of mutations can be separated in reoccurring and patient-specific tumor mutations. The latter can be potential targets for individualized cancer vaccines. The bioinformatics unit at TRON is able to identify these somatic, non-synonymous potential neo-epitopic mutations and has developed statistical algorithms for NGS mutation detection and immunogenicity prediction. After the preclinical proof of concept has shown that TRON’s identification and immunogenicity prediction of mutations is functional, the algorithms are currently used in several clinical trials. The most important factor for the selection of specific peptide sequences for the use in cancer immunotherapies is the prediction of the immunogenicity potential of these peptides. This is mainly influenced by the quality of the peptide-MHC interactions and the recognition by T cell epitopes. Morten Nielsen (Technical University of Denmark, Lyngby, Denmark and Universidad Nacional de San Martin, Argentina) presented the recent advances in immunoinformatic prediction methods. He stated that for MHC-I, peptide binding can be accurately predicted using state-of-the-art prediction methods resulting in stable prediction with a very low false-positive rate of <1 %. For MHC-II, nonetheless, affinity is a relative poor correlate to peptide immunogenicity resulting in binding predictions with relative high false-positive epitope discovery rates. It remains unclear which factors other than affinity determine what makes an HLA-II binder a T cell epitope.
George Coukos (Ludwig Cancer Research Center, Lausanne, Switzerland) introduced recent work on the tumor microenvironment and T cell infiltration in ovarian cancer. Intra-tumoral T cell levels are associated with long-term survival or cure of patients. Half of the patients with a high T cell infiltration in their tumor have no recurrence of their ovarian cancer after 96 months. To open therapeutic opportunities also for those patients whose tumors do not show strong T cell infiltration, George Coukos and his group were investigating how T cells can be prevented from tumor entry. Their work revealed a new mechanism involving the tumor endothelial barrier. The death mediator Fas ligand (FasL) was found to be selectively expressed in the vasculature of tumors, but not in normal vasculature. FasL expression was associated with low CD8+ T cell infiltration and a predominance of immune suppressive FoxP3+ Treg cells within a tumor. Vascular endothelial growth factor A (VEGF-A), interleukin 10 (IL-10) and prostaglandin E2 (PGE2) cooperatively promoted the induction of FasL expression in endothelial cells, which then acquired the ability to kill effector CD8+ T cells. However, Treg were not killed because they harbor increased levels of c-FLIP. In a mouse model, pharmacologic inhibition of VEGF-A and PGE2 led to an increased influx of tumor-rejecting CD8+ T cells over Treg, which was dependent on attenuation of FasL expression and resulted in CD8 T cell-dependent suppression of tumor growth. To translate these findings into clinical application, a pilot clinical trial to determine the feasibility and safety as well as immunogenicity of an autologous dendritic cell vaccine administered intra-nodally alone or in combination with bevacizumab (anti-VEGF) and cyclophosphamide (CTX) for patients with recurrent ovarian, fallopian tube or primary peritoneal cancer has been started. Preliminary data show promising effects on progression-free survival and increased tumor-infiltrating lymphocytes after vaccination and bevacizumab plus CTX application. Further combination of the approach with PGE2-downmodulation via cyclooxygenase-2 inhibition (e.g. by aspirin) remains to be clinically explored.
Sergio A. Quezada (UCL Cancer Institute, London, UK) presented his work on mechanisms underlying the anti-tumor activity of anti-CTLA-4. The initial hypothesis about the mechanism of action of anti-CTLA-4 antibodies was that by blocking inhibitory signals into effector T cells, they are able to break their tolerance. Quezada’s group recently demonstrated that to elicit full tumor protection, anti-CTLA-4 needs to bind not only T effector (Teff) cells but also Treg cells expressing elevated levels of CTLA-4. Blocking CTLA-4 increased the Teff/Treg ratio within the tumor relative to lymph nodes, which was caused by elevated numbers of CD4+ Teff cells as well as Treg cells in the lymph nodes on the one hand, while on the other hand, CD4+ Teff cells were increased, and Treg cells were selectively reduced within the tumor. The reduction in Treg cells involved a FcγRIV-dependent mechanism associated with the presence of FcγR-expressing macrophages within the tumor and elevated CTLA-4 expression by tumor-infiltrating Treg cells. Thus, these data can explain the paradoxical effects of anti-CTLA-4 on Teff- and Treg cell accumulation in the lymph nodes versus tumor by an antibody-dependent cellular cytotoxicity (ADCC) mechanism, which preferentially takes place in the tumor microenvironment. Therefore, monoclonal antibodies (mAB) targeting immune checkpoints should be evaluated for their potential to make use of this mechanism. To achieve this, mAB target expression on CD4+ and CD8+ Teff versus Treg cells should be investigated. For instance, PD-1 expression was found to be similar on CD8+ Teff cells and on Treg cells, thus ADCC would not be expected to be beneficial. Taken together, the findings highlight the significant role of the tumor microenvironment in determining the outcome of antibody-based immunotherapies. In conclusion, approaches making use of the capacity of the tumor microenvironment to deplete antibody-associated Treg cells could be utilized to further enhance the anti-tumor activity of immunotherapeutic mABs. Laurence Zitvogel (Institut Gustave Roussy, Villejuif, France) presented recent work which suggests that the gut microbiota shapes the anti-cancer immune response. CTX, a cytotoxic drug used against hematologic, and solid malignancies disrupts the mucosal integrity and induces a translocation of Gram+ bacteria into secondary lymphoid organs (e.g. spleen), which subsequently leads to the priming of “pathogenic” TH17 (pTH17) cells in the spleen. They share the characteristics of TH1 cells (expression of IFN-gamma and surface exposure of the chemokine receptor CXCR3) and TH17 cells (expression of interleukin IL-17 and CCR6). The pTH17 cells can transfer anti-tumor efficacy. By showing that germ-free and antibiotic-treated-specific pathogen-free mice had a reduced anti-cancer response which could be reversed by an adoptive transfer of polyclonal pTH17 cells, they could solve the mode of action how CTX is able to stimulate anti-tumor immune responses. These findings have been extended to other treatment modalities including platinum salts and different kinds of immunotherapy protocols (such as TLR9L, anti-IL–10R, anti-CTLA4 Ab).
In a keynote lecture, Josef Penninger (Institute of Molecular Biotechnology, Vienna, Austria) talked about the immune rejection of metastases. Referring to the seeds and soil theory by Stephen Paget from 1889 who proposed that metastasis depends on cross talk between selected cancer cells (the ‘seeds’) and specific organ microenvironments (the ‘soil’), he gave two examples that this theory still holds true. First, the RANK and RANKL interaction is a critical mechanism in the development of hormone-induced breast cancer and metastatic spread to bone. Here, RANKL can be seen as a candidate “soil factor” that controls tissue-specific metastasis of RANK-expressing tumors “seed”. Inhibition of RANKL results in reduced bone metastasis and protects from progestin-driven breast cancer. As a second example, Josef Penninger reported on the E3 ubiquitin ligase Cbl-b. The inactivation of the E3 ligase activity licenses NK cells to spontaneously reject metastatic tumors. He and his team identified the TAM tyrosine kinase receptors as ubiquitylation substrates for Cbl-b and with that could show that treatment of wild-type NK cells with a TAM kinase inhibitor enhanced the anti-metastatic NK cell activity in vivo and by that reduced murine mammary cancer and melanoma metastases. As Gas6 is interacting with the TAM tyrosine kinase receptors and Warfarin is inhibiting Gas6, they could identify an already approved drug for the treatment of thrombosis and thromboembolism as a possible ‘pill’ that awakens the innate immune system to kill cancer metastases.
CIMT: working groups
The CIMT Immunoguiding Program (CIP) Break-Out session once again displayed the wide range of proficiency panels and workshops currently being performed by the group. Steffen Walter (immatics biotechnologies GmbH, Tübingen, Germany) started off the session by presenting the results of the MDSC proficiency panel. He showed that very high interlaboratory variations were found, which were mainly associated with differences in gating strategies. The NK cell proficiency panel presented by Ruth Challis (University of Southampton, Southampton, United Kingdom) had fewer problems with regard to variability, only differing stimulation protocols were problematic for her harmonization efforts. The dextramer proficiency panel arranged by Henrik Pedersen (Immudex, Copenhagen, Denmark) showed that harmonizing SOPs requiring a standardized number of cells to be stained and acquired, as well as mandating use of DUMP channels and dead cell exclusion, allow inexperienced operators to carry out the stainings with a high degree of competence. In fact, inexperienced operators generally outperformed experienced operators after harmonization. The Treg workshop, introduced by Saskia Santegoets (Leiden University Medical Center, Leiden, The Netherlands), showed that different gating methods of Tregs give very different Treg numbers. She concluded that the traditional CD4+ CD25+ CD127lowFoxP3 gating was best for finding all Tregs, whereas addition of Helios or CD45RA only reduces the number of Tregs found. Nicole Bidmon (University Medical Center Mainz, Mainz, Germany) presented her TCR-engineered reference samples (TERS) that have been previously extensively tested in proficiency panels and now have been transferred into a kit to allow all laboratories with an electroporator to make their own TERS for the control of T cell assays, such as ICS and ELISPOT, in clinical trials. Pia Kvistborg (Netherlands Cancer Institute, Amsterdam, The Netherlands) explained that manual flow cytometry gating was not reproducible and is limited when looking at complex data sets with lots of parameters. Automated gating can solve these issues, but only when more interaction between bioinformaticians and immunologists occurs. Cliburn Chan (Duke University Medical Center, Durham, United States) followed on by introducing his re-flow program, an integrated program that takes flow data, analyses it and is able to report the data in a way that can easily be interpreted. The program still needs some help from humans though, as consistency of labeling and experiment setup is required for the markers to be more easily recognized.
The CIMT Regulatory Research group (RRG) focused on the regulatory challenges of cellular gene therapies in their session. Raj Puri (FDA, CBER, Rockville, USA) and Matthias Renner (Paul-Ehrlich-Institut, Langen, Germany) provided an overview of the regulatory landscape for cellular gene therapies in the USA and Europe. Carl June (Abramson Cancer Center, Philadelphia, USA) and John Haanen (NKI, Amsterdam, The Netherlands) presented their personal views on regulatory challenges on this rather new field of medicine that is characterized by large effect sizes in treated patients on the one hand, but also fatal side effects with some treatments on the other hand. This risk benefit profile leads to regulatory challenges in the selection of the right vectors and cell carriers, cell manufacturing issues, clinical trial design issues as well as toxicity management issues. More specific challenges derive from the fact that the treatment is a self-replicating drug, which does not follow the standard rules for pharmacokinetics and pharmacodynamics. Due to the novelty of the field and the fact that each novel product may exert effects that differ from previous treatments, the regulatory requirements for cellular gene therapy are high. During the discussion, some participants expressed that they have the impression that the field of cellular gene therapies is held to higher regulatory standards as compared to other fields of immunotherapy. Cedrik M. Britten (BioNTech RNA Pharmaceuticals, Mainz, Germany) proposed to open a discussion among experts to systematically address regulatory challenges for cellular gene therapy products. It should identify generically applicable principles that may facilitate the definition of safe and efficient non-clinical and clinical testing of novel products in patients, and thus balance the patients’ wish for safety on the one side and access to potentially beneficial innovation on the other side. There was an agreement that the CIMT RRG should initiate this novel research focus in the coming weeks and report back to the CIMT participants in 2015.
After a last year’s successful start, the CIMT Endeavour workshop took place for a second time within the framework of CIMT. The focus in the first workshop was on the challenges that European biotech companies face in early, middle and late-stage cancer immunotherapies and diagnostics development. The conclusion was that many scientists still lack fundamental knowledge regarding which factors drive commercialization, even though they should be taken into account during the early stages of preclinical and clinical research activities. Therefore, the focus of the second workshop was to bring together entrepreneurs and experts to share and discuss the challenges that occur when ideas are translated into business models. Early entrepreneurs in the field of CIMT presented their business idea on which established biotech experts responded by sharing their personal experiences about transforming business ideas into successful founding story. How a university spin-off can be transformed into a Nasdaq-listed biopharmaceutical company has been presented as the Micromet Case Study by Jens Hennecke (Biotech Business and Strategy Advisor) former Senior Vice President for Business Development at Micromet, Inc. He presented the story of how the German company Micromet has been spun-out, financed by different rounds of venture capital to continue the development of BiTE antibodies to achieve a product platform for the treatment of a variety of additional cancer indications and finally, the acquisition of the company by Amgen Inc. CIMT will continue to support interested innovators in academia to reach the right attitude and to gain the right skills for successful commercialization of innovation. The latter is certainly a prerequisite for many novel technologies to reach the market 1 day in the future, and thus to become accessible for patients with medical need.
The concept of modulation of the immune system to treat cancer has claimed its right to exist by proving that many of the therapeutic concepts are not only fictitious, but show strong anti-tumoral effects in treated patients. CIMT is an innovative area of medicine where a large number of novel treatment concepts are emerging, but have yet to be proven. The field has seen a big boost by the latest successes. Nonetheless, there is so much potential in the currently tested concepts that the chance is rapidly increasing that we will soon see novel approved products on the market. The 13th CIMT annual meeting will be held from May 11–13, 2015.
The authors would like to thank Christiane Wellié-Reeve (Mainz, Germany) for carefully proof reading the meeting report. Selected sessions of the CIMT meeting were made possible by research grants from the German Bundesministerium für Bildung and Forschung (BMBF Grant: 031A018). The CIMT working group CIP receives support for educational activities by the Wallace Coulter Foundation (Florida, USA).
Conflict of interest
The authors declare that they have no conflict of interest. Cedrik M. Britten has been the co-organizer of the meeting. Björn-Philipp Kloke has been the co-organizer of the CIMT Endeavour workshop.