Translational Mini-Review Series on Vaccines: Dendritic cell-based vaccines in renal cancer
Clinical Pathology, Department of Biomedical Sciences, University of Foggia, Italy,
Department of Dermatology and Immunology, University of Pittsburgh School of Medicine, PA, USA, and
Elena Ranieri PhD, Chair of Clinical Pathology, Department of Biomedical Sciences, University of Foggia, Ospedali Riuniti di Foggia, Viale Luigi Pinto, 1, 71100, Foggia, Italy. E-mail: email@example.com
Elena Ranieri PhD, Chair of Clinical Pathology, Department of Biomedical Sciences, University of Foggia, Ospedali Riuniti di Foggia, Viale Luigi Pinto, 1, 71100, Foggia, Italy. E-mail: firstname.lastname@example.org
Renal cancer is a relatively uncommon solid tumor, accounting for about 3% of all adult malignancies, however this rate incidence is rising. The most common histological renal cell carcinoma (RCC) subtype is clear cell carcinoma that makes up approximately 70–80% of all renal neoplasms and appears to be the only histological subtype that is responsive to immunotherapeutic approaches with any consistency. Therefore, it has been hypothesized that immune-mediated mechanisms play important roles in limiting tumor growth and that dendritic cells (DC), the most potent APC in the body, and T cells are the dominant effector cells that regulate tumor progression in situ. In this context, the development of clinically effective DC-based vaccines is a major focus for active specific immunotherapy in renal cancer.
In the current review we have not focused on the results of recently published RCC clinical trials, as several excellent reviews have already performed this function. Instead, we turned our attention to how the perception and practical application of DC-based vaccinations are evolving.
Renal cell carcinoma (RCC) has the third highest mortality rate among genitourinary cancers, with an estimated 12 000 deaths/year reflecting a continued trending increase in the incidence of this cancer by 2–3% per year [1,2] and 8130 estimated deaths in 2006 . RCC subtypes include clear cell RCC, papillary RCC and chromophobe RCC. Although more rare entities such as sarcomatoid RCC are pathological descriptions still in use, they probably represent variants of the dominant form, clear cell RCC, with divergent genetic features [4,5]. RCC clear cell type makes up approximately 70–80% of all renal neoplasms and appears to be the only histological form that is responsive to immunotherapeutic approaches with any consistency. In addition, it has been observed that metastatic lesions regress spontaneously in up to 1% of cases, suggesting that RCC-specific anti-tumour T lymphocytes are capable of rejecting tumour cells in certain circumstances .
Several immunotherapeutic approaches have been used previously in the adjuvant setting of RCC, such as regional/systemic administration of a range of biological response modifiers such as bacille Calmette–Guérin (BCG), interleukin (IL)-2, IL-12 and interferons (IFN), either as single agents, in combination with each other or in combination with substances such as 13-cis-retinoic acid and/or 5-fluorouracil . However, to date, regardless of the types of biological response modifiers or the forms of tumour preparations applied clinically, objective response rates have been disappointing, with no vaccines demonstrating a significant impact on the survival of RCC patients in randomized phase III studies. Hence, the development of an effective vaccine remains a major focus in the active specific immunotherapy.
In this context, dendritic cells (DC), the most potent antigen-presenting cells (APC), represent attractive vectors for immunotherapy because of their unique properties that allow efficient capture of microenvironmental antigens, effective trafficking to lymph nodes and major histocompatibility complex (MHC) class I- or II-antigen presentation for the induction and maintenance-specific CD8+ and CD4+ T cell immune responses. The development of clinically effective DC-based vaccines is the major focus for active specific immunotherapy in RCC . In this context, the growing understanding of DC physiology, along with accepted standard procedures that allow for DC isolation and ex vivo conditioning from different human tissue sources, has fuelled the integration of DC into novel cancer immunotherapies.
DC originate from CD34+ bone marrow stem cells and give rise to circulating precursors that home to the tissue where they reside as immature cells exhibiting a high degree of phagocytic capacity. They are distributed widely over peripheral tissues where, upon tissue damage, they capture locoregional antigen, process it into the peptides presented on major histocompatibility complex (MHC) molecules and migrate to the lymphoid organs, where they induce primary T cell immune responses . During migration and within the secondary lymphoid organs, DC undergo maturation from antigen-capturing cells to APC.
DC are a heterogeneous group of cells. They are comprised of multiple cell subsets that display differences in their functionality and tissue localization. Two major subsets of human blood DC have been defined in the literature . CD11c+/CD123dim/BDCA-2– DC, defined classically as ‘myeloid’ DC, are capable of phagocytosing antigens via their Fc receptors (CD32, CD64, FcεRI) and they tend to be potent stimulators of Th1-type polarized T cell responses. In contrast, CD11c–/CD123hi/BDCA-2+‘plasmacytoid’ DC that fail to express Fc receptors exhibit poor phagocytic potential compared to myeloid DC, and are notably weaker stimulators of allogenic mixed lymphocyte responses (MLR). However, they are impressive producers of IFN-α both during the course of viral infections and in steady state, allowing them the potential to support Th1-type immunity that is believed to be crucial to effective anti-RCC responses.
The infiltration of DC into primary tumour lesions is associated with significantly improved patient survival and a reduced incidence of recurrent disease in patients with a broad range of malignancies . This suggests an important immune-regulatory role for DC in the local tumour environment. The presence of cancer has been described to lead to numerical and functional abnormalities of DC in vivo and in vitro. Reduced DC counts may be due to cancer therapy or factors related to the presence of the cancer itself . Furthermore, studies on breast, prostate, renal cell and transitional cell carcinoma have shown that DC are poorly recruited and/or activated within tumour tissues .
In light of the extensive range of defects that have been elucidated in anti-cancer immune responses in patients, the central role of DC in inducing potent T cell responses to many different cancers has targeted their integration or manipulation in cancer immunotherapies.
Vaccine therapy in RCC
Because current therapies for patients with RCC are limited and are associated commonly with significant toxicities and side effects, the prospective development of alternative safe and effective approaches is imperative. Immunotherapy has become clinically important in the treatment on RCC, in large part, as this disease has proved refractory to conventional treatment modalities, such as chemo- and radiotherapy (Fig. 1). The increased knowledge of tumour immunobiology and recent progress in the understanding of several pathways and processes underlying carcinogenesis has led to the development of new immunotherapeutic strategies. To date, RCC has been treated principally with cytokine-based immunotherapies. The collective data derived from clinical trials using IL-2 and/or IFN-α have delineated clearly a subset of patients who react favourably to immunotherapy, independent of the regimen or combination given , and have demonstrated that these immunological manipulations can result in the durable objective responses in the setting of metastatic disease . More recently, neutralization of the biological activity of immunosuppressive cytokines produced by RCC, such as IL-6 [16,17] and transforming growth factor (TGF)-α, with highly specific monoclonal antibodies is currently under investigation .
Numerous alternative approaches based on cellular vaccines for RCC patients have also been developed for assessment in phase I/II clinical trials, including the use of attenuated or gene-modified tumour cell-based vaccines [19,20] and the adoptive transfer of lymphokine-activated killer cells (LAK) and tumour-infiltrating lymphocytes (TIL) in combination with systemic rhIL-2 administration [21,22]. The results from these trials have been met with a variable degree of enthusiasm, as in many cases tumour-specific immune responses can be identified in a significant frequency of treated patients; however, this translates into only rare occurrences of objective clinical responsiveness [based on Response Evaluated Criteria In Solid Tumours (RECIST) criteria, the evaluation criteria that replaced the WHO criteria for response evaluation].
Most recently, emphasis has shifted to the use of DC for the development of cell-based vaccines for RCC patients. This kind of active immunotherapy represents a promising investigational approach for stimulating RCC-specific T cell responses in cancer patients without inducing toxic side-effects . While early preclinical modelling of DC-based vaccines generated profound enthusiasm, the efficacy of therapeutic DC-vaccination against cancer in clinical trials has been modest and this approach has recently been criticized .
Overall, we remain optimistic that improved cancer vaccines will ultimately yield favourable clinical results, particularly after these approaches have been modified in a manner that integrates recent progress related to the physiology of DC and our improved understanding of how tumours and the host immune system interact with each other. In the current review we have not focused on the results of recently published RCC clinical trials, as several excellent reviews have already performed this function [25,26]. Instead, we will turn our attention to how the perception and practical application of DC-based vaccinations are evolving.
DC-based vaccine in patients with RCC
In contrast to the melanoma setting, where peptide-specific vaccination approaches have been conducted liberally since the mid-1990s which, in part, reflects the prototype nature of this tumour type with regards to its cellular and molecular immunological characterization, far less is known about tumour-associated antigens (TAA) in RCC. This has served to limit modern vaccine development for this disease. For this reason, initial vaccines for RCC focused primarily on the use of tumour cells themselves that provide a source of as-yet unknown TAA as immunizing agents and DC pulsed with whole tumour lysates. Most of these trials followed a relatively similar format of protocol, with a typical clinical trial schema shown in Fig. 2. Prior to the start of vaccine treatment, a radical nephrectomy is performed in patients exhibiting good performance status, serving to simultaneously de-bulk disease and provide a source of tumour antigen for vaccines. Two phase I trials have been performed using DC pulsed with tumour lysates, prepared from either cultured autologous tumours or derived from established allogenic RCC lines as a source of TAA [27,28]. In both trials, objective responses have been observed in a small percentage of patients, with some durable responses noted. However, DC-based vaccinations have not yet been shown to result in the general clinical improvement of treated patients.
Another controversial approach has involved the immunization of RCC patients with hybrids of DC fused to autologous RCC tumour cells. Early phase I trial results, reported by Kugler et al. , described tumour regressions in seven of 17 RCC patients using allogenic monocyte-derived mature DC fused in an electric field to autologous tumour cells. However, a follow-up phase II study failed to confirm these promising early results .
Considerable progress in the characterization of TAA has allowed for the development of new immunotherapeutic interventions against RCC cancer. G250, also known as carbonic anhydrase IX (CA IX), is the first widely expressed, RCC-associated antigen identified and may constitute a novel target for specific immunotherapy in RCC patients. G250 is a cell membrane protein overexpressed on 95% of all RCC types and there is no detectable expression on normal kidney tissue . This enzyme may play a role in the regulation of cell proliferation in response to hypoxic conditions and could be involved in oncogenesis and tumour progression . G250-derived peptides have been identified that are capable of being recognized by specific human leucocyte antigen (HLA)-A24-restricted cytotoxic T lymphocytes (CTL) , which has led to the performance of a recent phase I trial assessing the efficacy of peptide-based vaccination of HLA-A24-positive patients with progressive, cytokine-refractory RCC . Overall, the vaccinations were well tolerated by treated patients, with no adverse events reported. Interestingly, most patients displayed a vaccine-associated expansion in peptide-specific CTL in their peripheral blood. These promising clinical results warrant the continued exploration of G250-derived peptides in future phase II studies.
While GMP-grade, peptide-based vaccine approaches are technically simple and clinically attractive, in many cases this form of vaccine can be provided only to patients with certain HLA alleles, such as HLA-A2, found in approximately half of European Caucasians. To overcome this limitation, a recent phase I clinical trial used DC loaded with autologous tumour-derived RNA . This protocol has the theoretical advantage of providing a comprehensive source of tumour-expressed antigens (including those containing idiotypic mutations, etc.) to patient DC, thereby defining a novel ‘patient-specific’ immunogen. However, a more recent study has also generated a DC-based vaccine using RNA prepared from allogenic tumour cells (a well characterized and highly immunogenic tumour cell line, RCC-26), based on the supposition that a given RCC line expresses ‘shared’ antigens that are common to all/most RCC . This allogenic approach has the advantage of generating vaccines using an ‘off-the-shelf’ tumour cell line that can be applied to any patient, regardless of their HLA type. In this trial, RNA-loaded DC were shown to be capable of activating effector-memory CTL specific for TAA expressed by the RCC-26 cell line.
The immunostimulatory properties of (RNA-transfected) DC vaccines might then be enhanced further by disrupting regulatory pathways that suppress the activation and function of tumour-specific T effector cells in cancer patients. Indeed, current efforts to improve vaccine efficacy are focused on eliminating negative forces and enforcing positive signals. One strategy for augmenting DC function is the elimination of regulatory T cells (Treg) that play an essential role in down-modulating immune responses against self-(auto)antigens  and can be identified by co-expression of CD4 and CD25 and the transcriptional regulator forkhead box P3 (FOXP3) . Previous studies have shown that high Treg cell frequency can be found in advanced cancer patients  in association with reduced patient survival . In a recent study, potent tumour antigen-specific immune responses could be activated effectively by tumour RNA-transfected DC vaccines, particularly when recombinant fusion denileukin diftitox (DAB389IL-2) was administrated prior to the vaccine in order to deplete Treg cells systemically . Significantly higher frequencies of tumour-specific CD8+ T cells were identified in patients treated with combined DAB389IL-2 and vaccine than in those subjects receiving the vaccine alone.
Another example of a combined therapy involves the pretreatment of RCC patients with low-dose cyclophosphamide (CY) before injection of the DC-based vaccine. Low-dose CY has been described to down-regulate suppressor T cell activity, which may be a prerequisite for the success of tumour immunotherapy [42,43]. In a recent phase I/II study involving 22 RCC patients, 12 patients were treated with allogenic DC-based vaccines alone, while 10 patients received combined treatment with CY + DC-based vaccines. Two mixed responses and one patient with disease stabilization were observed, with all three of these patients treated with combined CY + DC-based vaccines .
Future perspectives of DC-based therapy
During the past two decades, several central questions have served to drive the cancer therapy field forward: can immune modulation promote the regression of established human cancers? What are the antigens involved in the immune recognition of human cancers? Can anti-tumour T cells be generated in patients by active immunization with cancer-associated antigens? What mechanism(s) limit cancer regression despite the in vivo generation of anti-tumour T cells? Are DC vaccines more clinically effective than other types of vaccines?
DC are clearly attractive components for inclusion in therapeutic vaccines designed to increase immune response to tumour antigens. However, there remain a large number of variables that need to be optimized in order to improve the efficacy of DC-based vaccines. These include the source of DC precursors [monocyte-derived DC or naturally circulating blood DC precursor (BDC)], the subtype of DC (myeloid or plasmacytoid) applied, the choice of activation stimuli used to mature DC and the format and amount of antigen used to load DC. Many of these aspects have received expert attention in several recent reviews and have not been discussed in detail here. To date, however, most clinical trials for RCC have used peripheral blood monocytes as a source of DC precursors and it is clear that additional maturational stimuli [such as lipopolysaccharide (LPS), bacterial dinucleotide (CpG); recombinant soluble CD40L; cytokine cocktail including IL-6, IL-1β, TNF-α, prostaglandin E2 (PGE2)] need to be applied to these immature DC in order to obtain a more stimulatory vaccine for clinical application .
The most recent advances in the immunotherapy of cancer suggest that cytokines such as type I IFN, in particular IFN-α, can represent valuable adjuvants for the development of DC vaccines with the potential to mediate superior clinical efficacy. These cytokines potently enhance both T cell and antibody responses to a soluble protein and promote immunological memory by their direct action on DC [46,47]. Because some RCC patients have demonstrated remarkable responsiveness to immunotherapy with IFN-α, it is not unlikely that DC activated by IFN-α may contribute to improved anti-tumour immunity and clinical responsiveness. We are currently performing a study comparing the in vitro immunogenicity of vaccines incorporating DC pretreated with various combinations of cytokines, including IFN-α.
An optimal DC preparation for the vaccination of RCC patients remains problematic and there is a clear need to define additional RCC-associated antigens for inclusion in vaccines and for the effective immunomonitoring of treated patients. While a great deal of attention has been paid to G250 (CAIX) as a potential ingredient in peptide-based DC vaccines, other peptides contributing to a polyepitope-based vaccine would be envisioned to yield a more potent regimen.
Future efforts to define effective RCC therapies should focus probably on combining DC vaccination with other modalities in order to improve the vaccine itself, providing help for the elicited T cells and protecting the immune response against the suppressive tumour microenvironment. Clearly, chemotherapy, which remains a standard of care for most solid cancers, should be considered for inclusion in DC-based combined therapies in future. As most DC vaccines have been applied thus far to RCC patients with advanced disease where the likelihood of response is low, future clinical studies might focus more effectively on treating patients with less advanced disease or in the adjuvant setting. These individuals may not only be more prone to respond, given a more normal state of immunocompetency, but owing to a smaller (i.e. micrometastatic) and/or less aggressive (tumour) target population that requires treatment-induced, immune-mediated eradication in vivo.