The fifteenth International Conference on Progress in Vaccination against Cancer (PIVAC-15), 6–8 October 2015, Tübingen, Germany: looking back on 15 years of progress

Introduction

PIVAC-15, held 6–8 October 2015 in Tübingen, Germany, marked the 15th anniversary of the PIVAC conference series (http://www.tati-group.de/pivac), held every year since PIVAC-1 in 2001 in Cambridge, UK. Prominent themes covered during the 2015 PIVAC meeting included therapies which can be used to modulate the immune system or reduce tumour-induced immune suppression, the selection of more appropriate immunotherapy targets, and methods that can be used to improve antigen presentation. In the context of 15 years of “Progress in Vaccination against Cancer”, speakers were asked to consider vaccination in the era of immunomodulatory antibody therapies that have so revolutionised the treatment of several types of cancer over the last few years.

Therapies for immune modulation and targeting tumour-induced immune suppression

Immune suppression or evasion by tumour cells is now a widely appreciated factor that limits the effectiveness of immunotherapies. Developing specialised therapies that can address this is likely to lead to improved efficacy of immunotherapies, and a number of diverse approaches were presented during the meeting. For example, Cornelis Melief (Leiden University Medical Center, Leiden, Netherlands, and ISA Pharmaceuticals) presented data suggesting that chemotherapy can be used to dampen suppressive myeloid cell activity and results in enhanced efficacy of cancer vaccines; both low- and high-dose platinum-based chemotherapy were equally effective when combined with peptide vaccination, and this was superior to either given alone. However, when administered as monotherapy, high-dose chemotherapy was more effective, suggesting that the therapeutic benefit when chemotherapy was combined with peptide vaccination was not due to the direct killing of tumour cells. Melief showed that chemotherapy can result in reduced levels of suppressive myeloid cells in late-stage cervical cancer patients. This was accompanied by boosted tumour-specific T cell numbers and function following vaccination and in a mouse tumour model resulted in higher levels of intra-tumoural T cells and tumour regression, demonstrating how standard therapies may be used to improve the efficacy of immunotherapies. One such approach for more effective cancer treatment may be to develop treatments that combine long peptide vaccination with immunogenic chemotherapy in addition to the use of immunomodulatory antibodies which block CTLA-4, PD-1, PD-L1, IL-6, IL-10 or TGFβ. Other approaches to ablate suppressive activity in the tumour microenvironment were outlined by Rolf Kiessling (Karolinska Institute, Stockholm, Sweden) who proposed a variety of methods such as targeting tumour stroma which has the advantage of being more genetically stable and more accessible to therapeutic agents. Angiomotin is a tumour stromal target in breast cancer that is not expressed on normal breast tissue or tumour tissue. Pre-clinical models show that the combination of an angiomotin DNA vaccine with a HER2 vaccine can provide tumour protection, which is associated with the inhibition of tumour angiogenesis and with the generation of angiomotin-specific antibodies. Kiessling remarked more broadly that in order to increase the efficacy of cancer vaccines, they should be combined with methods to reduce immune suppression such as tumour angiogenesis or checkpoint inhibitors. Data were also presented which showed that ipilimumab treatment results in lower levels of suppressive cells, both myeloid-derived suppressor cells (MDSCs) and regulatory T cells (Tregs). This provides evidence that the combination of immunomodulatory antibodies like ipilimumab with cancer vaccines may result in improved efficacy. Concerning the large body of knowledge about MDSCs now rapidly accumulating, Suzanne Ostrand-Rosenberg (University of Maryland, Baltimore, USA) presented an historical view of the current appreciation that tumour-induced immune suppression occurs at least partly through the release of inflammatory factors that promote the accumulation of suppressive myeloid populations. These cells seem to represent one of the major obstacles to the successful implementation of immunotherapies. Ostrand-Rosenberg showed how early studies performed in vitro and with smaller tumours in vivo using genetic modification of tumour cells to induce expression of MHC II and CD80 tended to be effective, but that this form of therapy was less successful for larger tumours. This was primarily due to more potent immune suppression exerted by larger tumours. A potential target to combat this tumour-induced immune suppression is the Nrf2 (Nuclear factor (erythroid-derived 2)-like 2) transcription factor, which regulates the expression of antioxidant proteins that protect against oxidative damage. Nrf2 increases the survival of MDSCs by reducing intracellular reactive oxygen species (ROS) and apoptotic sensitivity as well as increasing the suppressive and inflammatory activity of MDSCs. Although the quantity and suppressive potency of tumour-infiltrating MDSC can be reduced by targeting Nrf2, MDSC appear to be homeostatically regulated so that this decrease is accompanied by an increase in the rate of MDSC generation. This remains an area of intensive investigation.

Quite different tumour escape mechanisms are also clearly documented in many cancer types. Federico Garrido (Hospital Universitario Virgen Nieves; Universidad de Granada, Granada, Spain) suggested that loss of MHC proteins on tumour cells is a major mechanism of immune evasion and that treatment with immunotherapy may even contribute to the selection of MHC-negative tumour cells and the emergence of metastatic lesions more resistant to immune attack. He classified MHC loss as either “soft” (reversible) or “hard” (irreversible), with induction of MHC expression possible when treating soft tumour lesions with immunotherapy, radiotherapy, low-dose chemotherapy or combinations thereof. Irreversible MHC loss may require more involved approaches such as genetic transfection. The suggestion that MHC loss is closely linked to the therapeutic efficacy of immunotherapy implies that tumour MHC expression should be monitored in consultation with physicians when treating patients with immunotherapies.

Taking an historical perspective, Theresa Whiteside (University of Pittsburgh, Pittsburgh, USA) presented an overview of the enduring earlier reluctance of the medical and scientific community to accept the notion of tumour-induced immune suppression and the concept of suppressor cells. This is now a universally accepted phenomenon, with subtle differences between different types of suppressor cells now better understood, for example differences between natural Tregs derived from the thymus which prevent autoimmunity and inducible Tregs which are programmed by tumour microenvironmental factors such as IDO, tumour exosomes or adenosine. Tumour microenvironmental programming of these cells induces them to become long-lived, resistant to Treg-depleting strategies and strongly immunosuppressive which mediate suppression through up-regulated expression of inhibitory receptors such as PD-1, T cell immunoglobulin and mucin-domain containing-3 (TIM-3), Fas, or lymphocyte-activation gene 3 (LAG-3) or through other suppressive molecules such as adenosine or TGFβ. Whiteside presented data showing that cetuximab therapy increased the level of Tregs and that this correlated with patient survival. Furthermore, contrary to expectations that immunomodulatory antibodies will reduce or eliminate Tregs, they can instead function to stimulate these cells. Therefore, immunotherapy with checkpoint inhibitors may have to be combined with agents blocking immunosuppressive factors such as adenosine, IDO, TGFβ or IL-10. Continuing the theme of recognising differences in cell types with important roles in immunity, and focussing on the recently-recognised important impact of the microbiota on immune responses, Francine Jotereau (INSERM U892 and University of Nantes, Nantes, France) presented her work on double-positive CD4CD8αα (DP8α) regulatory T cells induced by Faecalibacterium prausnitzii, the most abundant bacterium in the human intestinal microbiota of healthy adults. Uncovering the role of novel cell types and the microbiota is likely to be important for the future of immunotherapy to treat cancer and other diseases.

Pierre Coulie (de Duve Institute, Brussels, Belgium) emphasised points touched on by other speakers when saying that the future of cancer immunotherapy will consist of combined therapies. Simultaneously dampening immune suppression as well as providing immune stimulants will be required. This may be accomplished by combining therapies to reduce tumour suppression with immunomodulatory antibodies and with the immunogenic delivery of tumour-specific antigens to collectively result in high numbers of antigen-specific T cells. Coulie pointed out that each of these aspects will likely need to be personalised, and for maximum benefit they should be administered to patients early in the course of treatment before immune selection can occur. Rather unexpected and intriguing ways of modulating immunity to reduce suppression and enhance anti-tumour activity were presented by Per thor Straten (Center for Cancer Immune Therapy (CCIT), Copenhagen University Hospital Herlev, Denmark). He reported that mice taking exercise in the form of voluntary wheel running show enhanced activity in several immunological pathways and resulted in an influx of immune cells into tumours. This was associated with reduced tumour incidence and retarded growth in various tumour models. Depletion of NK cells and studies in athymic nude mice which lack functional T cells but retain NK cells show that the exercise induced anti-tumour effect in this case was mediated by NK cells.

Choosing more appropriate targets for immunotherapies

Both cancer vaccines and immunomodulatory antibody therapies aim primarily to stimulate T cell responses against tumours of sufficient vigour to overcome the suppressive cancer microenvironment and to destroy or control neoplastic cells. Pioneering work on the nature of these tumour-associated antigens (TAAs) was described by Hans-Georg Rammensee (University of Tübingen, Tübingen, Germany) who gave a brief overview of the history of the development of the current understanding of MHC peptide presentation before detailing his work in pursuit of tumour-specific peptides. In the mid 1990s, Rammensee began to analyse the HLA ligandome of cancers with intermediate mutational loads such as hepatocellular carcinomas, renal cell carcinomas, ovarian carcinomas and several leukaemia types. Using peptide prediction based on relevant HLA class I molecules, particular attention was given to HLA ligands which may contain mutations. Despite these attempts, mutated peptides presented by HLA molecules were never found. However, analysing the entire spectrum of detectable HLA ligands in tumour samples in comparison with adjacent autologous healthy tissue revealed large numbers of germline peptides with apparently tumour-specific expression. Thus, these results do not support the hypothesis that mutated antigens will be highly immunogenic, instead, T cell responses to germline TAAs correlate with patient survival or clinical benefit. In light of these findings, the “Tübingen approach” proposes that there are large numbers of apparently tumour-specific peptides that are different from tumour to tumour and which can be identified by HLA ligand analysis in tumour and adjacent healthy tissue. Thus, treating patients according to this approach may begin with an initial round of vaccination with non-tailored “off the shelf” peptides, followed by a cocktail of tumour-specific peptides for each individual patient. This would need to be combined with effective immune modulation, and in line with this Rammensee remarked there is a current need to identify better adjuvants to improve the efficacy of cancer vaccination.

Along these lines, Per thor Straten also reported his laboratory’s use of reverse immunology to characterise tumour antigens, with a special emphasis on proteins that play crucial roles in the functioning of cancer cells such as survivin and B cell lymphoma 2 (Bcl-2). He has shown that these antigens can be recognised by T cells from TIL and PBMC, and that peptides from some have been shown to induce T cell reactivity and potential clinical activity in small scale trials. His laboratory is currently expanding on targets and target types in this approach, as well as increasing the use of adjuvants. A vaccination trial using a long B cell lymphoma-extra large (Bcl-xl) peptide and a novel poly-IC-based nano-adjuvant in prostate cancer will begin in 2016. An ongoing issue is that despite the large number of target antigens that can be used for vaccination, there are relatively few good adjuvants. An alternative approach of thor Straten’s laboratory is to use in vitro expanded TILs for adoptive cell therapy. In melanoma, this has been shown to induce complete and lasting clinical responses in approximately 20 % of patients. TIL therapy has shown that when it works, it is capable of eliminating large tumour masses (also in combination with IL-2). By conducting comprehensive monitoring, thor Straten has been able to determine characteristics of TIL cultures and biological responses which are associated with clinical response to this therapy. Accumulating data suggest that TIL may serve as both a prognostic and predictive marker. Kiessling also described the advantages and disadvantages of using proteins which play indispensible roles in tumour cells as targets for cancer vaccines. Because these proteins are generally widely expressed on cancer cells, there may be less risk of immune escape associated with targeting them, but many of these are self proteins which are subject to central tolerance. Kiessling pointed out an inverse correlation between MHC I and HER2 expression, suggesting that these tumours would be better treated with an integrated approach including NK cells and antibodies. This is particularly relevant for HER2 which has a cell surface domain, and because HER2/3 signalling regulates MHC class I polypeptide-related sequence A/B (MICA/B) expression and results in enhanced NK cell recognition via MICA/B natural-killer group 2 member D (NKG2D).

Effective cancer immunotherapy is more likely to be achieved through combined approaches. For example, through the use of methods which reduce tumour-induced immune suppression as discussed above, in conjunction with therapies that effectively prime the immune system against TAAs. A considerable amount of attention was given to this topic during the meeting, with approaches involving both mutated and germline antigens presented. For example, Gustav Gaudernack (Institute for Cancer Research, Oslo University Hospital, Norwegian Radium Hospital, Oslo, Norway) presented the work of his laboratory which is vaccinating cancer patients using molecularly defined targets that derive from proteins which play a central role in carcinogenesis. This approach was spurred on from observations dating back 25 years that some patients with tumours which had mutations in the Ras oncogenes had circulating memory T cells specific for Ras mutations. The approach of Gaudernack’s laboratory employs long synthetic peptides which allow each patient to select the epitope most fitting their own HLA make-up and which also involve CD4+ T cells which may present to antigen-presenting cells (APCs) and in turn prime cytotoxic CD8+ T cells. These clinical trials showed that patient survival always correlated with immune responses against the vaccine, and that up to 50–80 % of patients (depending on the cohort) were able to mount an immune response. These responses were diverse in nature but specific for certain Ras mutations, resulting in poly-functional CD4+ and CD8+ T cells from a broad range of HLA molecules. One of these clinical trials investigated patients with Ras mutated tumours and involved a 6-dose vaccination schedule with a synthetic Ras peptide. The vaccinated group had a median survival of 26.8 months compared with 14.6 months without vaccination, with some patients showing memory responses 8 years following the last vaccine dose. Collectively, this approach has shown that even late-stage patients can respond to peptide vaccines, but prolonged vaccination schedules may be required to develop immune responses. These results also document that vaccination with mutant peptides can result in very long-term immunological memory and that immune responses against single peptides in an individual patient can be complex, with multiple CD4+ and CD8+ clones differing in HLA specificity. The complementary interactions between the two major subsets of T cells in anti-tumour immunity have been the subject of studies by many investigators over the years. Responding T cells can sometimes show unconventional phenotypes and are often poly-functional without following T helper cell type 1/2 (Th1/Th2) convention; however, this complexity and diversity of T cell responses may underlie their therapeutic benefit. In this context, Costas Baxevanis (Cancer Immunology and Immunotherapy Center, St Savas Cancer Hospital, Athens, Greece) discussed the efforts of his laboratory to understand the role of CD4+ T helper cells in the function of cytotoxic CD8+ T cells in humans. This work showed that CD8+ T cells can be effectively activated to mediate cytotoxicity against autologous tumour via T helper-dependent activation of the autologous tumour antigen-loaded dendritic cells (DCs). Specifically, interactions between CD40 ligand and CD40 on CD4+ cells and DCs, respectively, seemed to be essential for the activation the DCs to present antigen and co-stimulate the priming of CD8+ T cells to lyse autologous tumour targets. These findings inspired the use of epitopes that stimulate CD4 helper T cells in vaccines. Using the HER2/neu antigen as a model, Baxevanis investigated whether peptide-activated T helper cells could assist in breaking tolerance against CD8+ epitopes which are present in TAAs. In vivo studies showed that this approach resulted in reduced tumour growth or prolonged survival. This work was continued in clinical trials with the use of AE37—a HER2 peptide vaccine that has been modified to stimulate CD4+ T cells, and combined with GM-CSF as an adjuvant. Responses to this vaccine were seen up to 3 years after treatment, and patients who responded or who had pre-existing immunity to the vaccine showed low levels of Tregs and had T cells which produced low levels of TGFβ but high levels of IFNγ. Similarly, patients who responded or who had pre-existing immunity showed superior survival or reduced rates of disease recurrence.

Currently there is much interest in TAAs which represent patient and tumour-specific neoantigens, and which are the result of genetic mutations in the cancer cells. Else Marit Inderberg-Suso (Oslo University Hospital, Oslo, Norway) described an animal model of adoptive cell therapy using CD4+ and CD8+ T cells targeting a frameshift mutation in the TGFβ gene. These T cells were transfected with a TCR allowing them to recognise tumour cells carrying TGFβ mutations. Transfected TGFβ-specific T cells produced IFNγ and TNF and resulted in reduced tumour load and prolonged survival in mice; such approaches are currently of great topical interest in human clinical trials, with further HLA class II restricted TCRs under development. A different category of TAA may be more amenable to immune targeting, namely viral epitopes in virus-induced tumours. Melief suggested that due to lack of central tolerance, viral epitopes (and other non-self antigens arising in the tumour by mutation) are perhaps more suitable as therapeutic vaccine targets, but that concentrated delivery with effective adjuvants is crucial. This is in contrast to most antigens so far investigated in cancer vaccination, which are differentiation or testis antigens for which the repertoire of available T cells is usually blunted due to central tolerance. Delivering long peptides covering MHC I and MHC II epitopes in a concentrated form is important to avoid antigenic competition, to allow the collaboration of CD4+ and CD8+ T cells and to result in effective T cell responses. This may be used alongside adjuvants such as TLR ligands and the combination of therapies to reduce immune suppression or otherwise boost the immune system such as checkpoint inhibitors, immunogenic chemotherapy or the neutralisation of suppressive cytokines.

Targets other than TAAs will also need to be considered in designing successful immunotherapies. Thomas Sayers (Leidos Biomedical Research Inc., Frederick, USA) described his work which has focused on discerning the relative contribution of the two T cell cytolytic pathways to tumour regression (release of lytic granules vs secretion of death ligands). His work has shown that T cell function by the release of lytic granules and the secretion of death ligands are not functionally redundant in vivo, which led to testing the therapeutic benefit of death receptor agonists. TRAIL receptor agonists showed anti-tumour activity in animal models, but clinical benefit was not widely observed. Therefore, agents that could sensitise cancer cells to the effects of TRAIL-induced apoptosis were sought. The proteasome inhibitor bortezomib was found to sensitise some cancer cells to TRAIL in vitro and in vivo, but it is immunosuppressive at higher doses, leaving a narrow therapeutic window. A high-throughput screening approach of more than 50,000 compounds identified the natural product, withanolide E, as an agent capable of sensitising renal cancer cells to TRAIL-induced cell death with low levels of toxicity. Renal cancer has thus far responded poorly to immunotherapy, and few target antigens have been identified; therefore, an alternative method of immunotherapy such as this may hold promise as a more efficacious form of treatment. The Sayers laboratory is currently determining whether more potent analogues of withanolide E improve the therapeutic effect of combing them with TRAIL agonists, as well as investigating the molecular mechanisms that determine how these agents sensitise cancer cells to death ligand-mediated apoptosis.

Methods for improving antigen presentation

Alternative methods for potentiating anti-cancer immunity may reside at the level of the antigen-presenting cell. Esteban Celis (Georgia Regents University Cancer Center, Augusta, USA) presented his view on how peptide vaccination can be optimised to achieve better clinical results. He pointed out that although numerous CD8+ epitopes have been identified, peptide vaccinations have largely suffered from being weakly immunogenic. This is likely due to sub-optimal immune responses (quantity and quality of T cells) and suppressive features of the tumour microenvironment. The immune responses induced by anti-cancer peptide vaccination stand in contrast to the strong immune responses that commonly result due to viral or bacterial infection. This observation was the inspiration for the strategy to optimise peptide vaccines by mimicking infections. To achieve this, APCs must present antigen along with danger signals derived from cytoplastic pattern recognition receptors such as RNA helicases or stimulator of interferon genes (STING). For example, including the bacterial secondary messenger c-di-GMP to the TriVax vaccination approach (TriVax consists of a synthetic peptide corresponding to the minimal T cell epitope, poly-IC adjuvant and anti-CD40 antibodies) increased immune responsiveness and slowed tumour progression compared with using TriVax alone. Another method of optimising vaccines is to ensure that professional rather than non-professional APCs present the peptide in question to T cells, which can be achieved by the use of long peptides. Data from Celis’s laboratory suggest that peptide attributes (for example hydrophobicity or amphipathicity), route of administration and use of adjuvants are more important than peptide length for eliciting strong immune responses. To achieve large numbers of cytotoxic T cells by peptide vaccination, two separate events are required: 1) Peptide priming by professional APCs, along with CD40 activation and TLR signalling, 2) T cell expansion (in this case either by professional or non-professional APCs), in which type I interferon induced by retinoic acid-inducible (RIG-I)-like receptor stimulation by poly-IC plays an important role. The efficacy of this approach can be further boosted with the use of anti PD-1 antibodies.

In the case of haematological malignancies, some of these tumours themselves have properties of APC. Farzin Farzaneh (King’s College London, London, UK) outlined two methods his laboratory is undertaking to improve immune response to vaccines in leukaemia. The first is based on the observation that acute myeloid leukaemia (AML) blasts share common features with APCs such as expression of co-stimulatory molecules, and that vaccination with modified AML blasts which more closely resemble APCs may allow immune responses against these cells. Pre-clinical studies have shown that tumour cells expressing co-stimulatory molecules and appropriate Th1 cytokines are able to mediate immune rejection of previously established tumours. Current phase I clinical trials in AML patients with poor prognoses are testing safety and efficacy of vaccination with autologous AML cells that have been genetically modified to express CD80 and IL-2, which are usually absent on these cells. Previous work has shown that these modified cells can stimulate T cells and enhance the cytolytic activity of NK cells against autologous unmodified AML cells. In addition to this, a new vaccination approach which consists of using combinations of adjuvants for immune activation is being investigated, with the aim of using it as a platform for the induction of cellular immunity against chronically experienced antigens in cancer or in chronic infections. This method, dubbed “CASAC” (combinations of adjuvants for synergistic activation of cellular immunity), has been tested in pre-clinical studies which have shown that combining adjuvants such as TLR agonists with IFN and anti-CD40 with target peptides can break immune tolerance to chronically experience antigens associated with cancer such as tyrosinase-related protein-2 (TRP2), Wilms tumour protein 1 (WT1) and Glypican-3. CASAC has been shown to result in clonal expansion of T cells and the induction of antigen-specific immunity, including in vivo cytolytic activity and lysis of antigen-positive cells, even in aged mice.

Regarding “conventional” use of APCs as agents for enhancing vaccine efficacy, Andrew Jackson (University of Nottingham, Nottingham, UK) presented data from his laboratory on their work using DCs as a form of immunotherapy for glioblastoma. Better therapies are urgently required for glioblastoma because it is the most aggressive form of brain cancer and it is responsible for more deaths than any other cancer in people under 40 years old. Considering the side effects associated with conventional treatment, immunotherapy holds promise as a less toxic therapy capable of providing long-term protection. DCs are a potential form of therapy because of their function as potent APCs capable of cross-priming and orchestrating immune responses. In vitro-generated DCs have been studied extensively and shown to be safe, but results have been disappointing so far. Circulating DC subsets may be more efficacious than in vitro-generated DCs because they are more physiological and have greater potency, for example owing to their efficient migration to lymph nodes and strong activation of T cells. The difficulty in exploiting these cells as a form of therapy is due to their reduced number in glioblastoma patients and their susceptibility to be suppressed. Enhanced DC function may be achieved by targeting certain pathways, for example tumour lysate and dexamethasone induce the inhibition of mDCs, which can be reversed by inhibiting the p38/MAP kinase-activated protein kinase 2 (MK2) pathway. This may be combined with modulation of other intracellular pathways along with the use of checkpoint inhibitors. Further consideration of the successes and challenges of DC vaccination was discussed by Carl Figdor (Radboud University Medical Centre, Nijmegen, Netherlands). After more than 15 years of DC vaccination and at least 3000 treated patients, DC vaccination has been proven safe and able to induce immune responses. For example in stage IV melanoma, DC vaccination can result in immune responses in 40 % of patients and in increased overall survival, but long-term clinical responses are still limited (fewer than 15 % of patients). Thus far, what we have learned from DC vaccination is that too many different culture protocols and loading procedures were in use, making comparison difficult. A lack of randomised phase II/III trials and regulatory issues are an ongoing obstacle. Finally, only a few specialised centres in Europe are capable of producing these cells. Figdor stressed that improving these aspects will likely lead to improvements in treating patients with DC vaccines. The limited success of inducing long-term clinical responses may be improved by beginning vaccination as early as possible, particularly because this form of therapy has been found to be very safe. Optimisation of vaccine delivery is required, for example the efficacy of monocyte-derived DC vaccines may be improved with the use of RNA, which can extend the expression of whole tumour antigens but also allows manipulation of DC function. Utilisation of different types of DCs may also improve efficacy, for example naturally occurring myeloid dendritic cells (mDCs) or plasmacytoid dendritic cell (pDCs), the development of a fully synthetic DC or in vivo targeting of resident DCs. Modulating immunosuppressive factors and the identification of immune biomarkers by in vivo and ex vivo monitoring before, during and following DC vaccination in order to better guide treatment are additional points that should be addressed. Thus, effective cancer treatment will have to boost antigen-specific immune activation as well as limit immunosuppressive factors. The latter may be challenging because the microenvironment of each metastasis may be different and require a different approach. Current data suggest that the ratio between peri- and intra-tumoural T cells correlates with the survival of metastatic melanoma patients after DC-based immunotherapy. Figdor’s laboratory has recently been performing studies with two types of naturally occurring DCs, mDCs and pDCs, in an effort to improve DC-based immunotherapy. These natural DC subsets express distinct TLRs and have the advantage of not requiring extensive in vitro culture. The first results suggest that these cells are potent immune stimulators that are able to initiate immune responses in cancer patients. The increasing use of checkpoint inhibitors poses the question of how DC vaccination compares to these immunomodulatory antibodies, and whether combinational approaches allow for more effective treatment.

Gosse Adema (Radboud University Medical Centre, Nijmegen, Netherlands) commented more broadly on approaches used to target DCs. Adema pointed out that even if anti-tumour immune responses can be induced by targeting DCs, tumour-induced immune suppression is likely to prevent these responses from resulting in clinical benefit. One method to simultaneously reduce tumour suppression and induce antigen uptake by DCs is tumour ablation. Data from animal models have shown that DC therapy is only effective if accompanied by tumour ablation. This can be combined with therapies to target DCs such as CpG, which has shown to result in DC maturation, the induction of cytotoxic T cells and in tumour protection. Novel approaches designed to ablate tumour cells and stimulate antigen uptake by DCs in vivo are ongoing in Adema’s laboratory and includes methods such as cryo, radio and electric current ablation, diffusing alpha radiation and high-intensity ultrasound.

Conclusion

Progress in cancer immunotherapy over the past 15 years has occurred in many diverse areas. The most notable of these is of course the clinical introduction of immunomodulatory antibodies, which have revolutionised treatment for certain cancers. One of the main points of focus is to now determine which existing therapies can be used in conjunction with these drugs to provide additional patient benefit, and to develop strategies which allow a greater proportion of patients to benefit. The past 15 years have also seen notable progress in the development of cancer vaccines, which may also benefit from combinational approaches with immunomodulatory antibodies and the development of better adjuvants.

Abbreviations

AML

Acute myeloid leukaemia

mDC

Myeloid dendritic cell

MICA/B

MHC class I polypeptide-related sequence A/B

Nrf2

Nuclear factor (erythroid-derived 2)-like 2

pDC

Plasmacytoid dendritic cell

Th1

T helper cell type 1

Treg

Regulatory T cell

Conflict of interest

The authors declare that they have no conflict of interest. Graham Pawelec is chair of the PIVAC scientific organising committee.