Exploring immune therapy for renal cancer


A paradigm shift in cancer immunotherapy

inline image

It is now clear that the immune system (both innate and adaptive) not only protects the host from tumor development, but also selects for the formation of tumor cell variants more resistant to immune attack.1 This implies that clinically-detectable malignancies derive from cancer cells previously “edited” by the host's immunity. It follows that immunotherapy regimens must take into account the fact that the tumor has already found a way to circumvent immune recognition/elimination, including the creation of immune suppressive local tumor environments. It is therefore crucial to develop strategies aimed at overcoming such immunosuppressive mechanisms, as well as enhancing effector T cell responses.

After many disappointments, years of effort in tumor immunology have finally resulted in two important developments in cancer immunotherapy. One is the US Food and Drug Administration (FDA) approval of sipuleucel-T, and the other is the application of anti-cytotoxic T lymphocyte antigen 4 (CTLA-4) antibody therapy. These two innovations, together with the approval of tyrosine kinase inhibitors, will have a huge impact on cancer immunotherapy for the treatment of renal cell carcinoma (RCC) and are likely to cause a paradigm shift in treatment rationales. The emphasis will be increasingly on (i) overcoming immunosuppressive environments rather than mere activation of immune responses; and (ii) a focus on the individual “patient response” rather than “tumor response”. I hope readers will get a sense of this paradigm shift in cancer immunotherapy. In this editorial, we have brought together a series of reviews from experts to explore immune therapy for renal cancer.

RCC, like malignant melanoma, appears to be one of the most immune-sensitive cancers occasionally undergoing dramatic spontaneous regressions. This has encouraged a strategy of using immunomodulating therapies more frequently for these cancers than many others. Nearly two decades ago, cytokine-based therapy using high dose interleukin-2 (HD IL-2) was approved by the US FDA, in 1992, and results of its use to treat advanced RCC were already reported in 1995.2 HD IL-2 treatment has resulted in durable tumor remission in a minority of patients, but with severe adverse effects. Cytokine-based therapy using low-dose IL-2 and/orinterferon-α (IFN-α), with reduced side-effects, was standard therapy for RCC for a long time. However, this approach has now been more or less superseded after the development of targeted therapies including vascular endothelial growth factor (VEGF) inhibitors, such as bevacizumab, or multikinase inhibitors, such as sunitinib and sorafenib.3 In the next chapter, Tomita reviews cytokine therapies in the era of targeted therapy. He points out the difficulties and importance of choosing treatments according to the tumors and characteristics of patients rather than merely following guidelines. He proposes that the combination therapy of cytokines and targeted drugs is the future direction.

After the first identification of a human tumor antigen recognized by CD8+ T cells from a melanoma patient in 1991,4 many other such targets have been identified in different cancers. However, compared with melanoma, only a few promising tumor antigens have been identified in RCC. This has limited the development of tumor antigen-specific immunotherapies.5 It is to be expected that if more tumor-specific antigens were to be identified in RCC, tumor antigen-based approaches would become more feasible, as in malignant melanoma, given that RCC can efficiently induce immune responses. To overcome this limitation, Kobayashi developed adoptive immunotherapy using γδ T cells that can recognize tumors in a major histocompatibility antigen (MHC)-independent manner. His group has successfully expanded γδ T cells from RCC patients' peripheral blood mononuclear cells (PBMC) using 2-methyl-3-butenyl-1-pyrophosphate (2M3B1PP). The results of a phase I clinical trial include one complete response, five stable disease and five progressive disease.

Tumor antigen-specific immunotherapeutic agents, including peptide-based vaccines, DNA vaccines consisting of genes encoding the target antigen or dendritic cell (DC) vaccines pulsed with tumor antigen peptides and so on, have been developed and are now being tested for treating several different cancers. In a multicenter phase III trial, significantly greater tumor regression after radical nephrectomy was observed in patients who received autologous tumor lysate vaccination than with surgery alone.6 In a 10-year survival analysis of 1267 RCC patients undergoing radical nephrectomy subsequently treated with autologous tumor cell vaccines, it was shown that this adjuvant treatment resulted in a significantly improved overall survival in pT3-stage RCC patients.7 Controlled trials using the recent TNM classification and incorporating known risk factors for prognosis are warranted. In this issue, Tatsugami and Naito review the principles of DC-based immunotherapy. Based on their careful observation that regulatory T cells were decreased by IFN-α therapy, but were increased by IL-2, they used IFN-α as an adjuvant for DC therapy. Suekane et al. review peptide-based cancer vaccine. They carried out a phase I trial of so-called personalized peptide vaccine for cytokine-refractory metastatic renal cancer patients. Though the results of clinical trials are not satisfactory, combination therapies with personalized peptide vaccine and molecular target drugs or cytokines are anticipated to achieve a breakthrough in the treatment of RCC.

Although the importance of patient selection and personalized treatment for RCC are shown by other authors, Eto et al. review personalized treatment in the immunotherapy for metastatic RCC. They showed that the single nucleotide polymorphism (SNP) in signal transducer and activator 3 (STAT3) were associated with better response to IFN-α. It is expected that the improved patient selection will result in better clinical response.

Although it is not a RCC vaccine at this point, we would like to mention the first such agent to gain FDA approval: sipuleucel-T (Provenge), developed for treating advanced prostate cancers. Sipuleucel-T is based on the individual patient's monocytes, which are incubated with a fusion protein consisting of granulocyte-macrophage colony-stimulating factor (GM-CSF) and prostatic acid phosphatase (PAP). Antigen-specific vaccination with this product induced marked infiltration of effector T cells specific for PAP into the prostate gland, and yielded a statistically significant difference in overall survival between certain immunotherapy groups and the placebo group.8 After completion of three phase III trials, sipuleucel-T was approved in April 2010 as the first antigen-specific immunotherapy, thus becoming a landmark for the field of cancer immunology and immunotherapy.

Another recent breakthrough in this field is the development of cancer immunotherapy using CTLA-4 antibodies (ipilimumab and tremelimumab). These antibodies inhibit an immunological checkpoint and take the brakes off T cell responses, amplifying the activation of CD4+ and CD8+ effector cells. CTLA-4 blockade has been evaluated in many malignancies, but again, the most mature data are available from melanoma patients. In August 2010, the results of phase III trials of ipilimumab for melanoma were reported.9 This new treatment boosted immune responses against melanoma and yielded significant survival advantages for treated patients. Ipilimumb has also been tested in several other cancers including RCC and prostate, and objective clinical responses have been reported.10,11 However, use of this agent was also associated with clinically important immune-related toxicity. Critical issues related to autoimmunity as side-effects still remain; nonetheless, CTLA-4 blockade is likely to be the next promising cancer immunotherapy; FDA announced its approval of ipilimumab for the treatment of advanced metastatic melanoma on 25 March 2011.

For many years, we have focused on how to activate tumor-specific immune responses by inducing and expanding CTL and improving the recognition of tumor antigens to develop cancer immunotherapy. Now we understand more about the immunosuppressive mechanisms acting at the tumor site and the crucial importance of the tumor microenvironment. The phase II clinical trials with ipilimumab proved the concept that overcoming immunosuppressive conditions and breaking immune tolerance are important for developing effective therapies. Many attempts are now being made to counteract the commonly high circulating levels of immunosuppressive factors in cancer patients, including TGF-β, IL-10 and VEGF, regulatory T cells and myeloid-derived suppressor cells (MDSC), as well as the immunological checkpoints mediated by cell surface molecules, such as CTLA-4, PD-1 and others.

In a placebo-controlled randomized phase III trial in which sipuleucel-T was given to patients with metastatic asymptomatic castration-resistant prostate cancer (CRPC), the primary end-point was progression-free survival.12 Although the primary end-point did not achieve statistical significance (P = 0.052), a difference in overall survival was statistically significant (P = 0.01, HR = 1.70). In a randomized phase II trial of a poxvirus-based vaccine approach targeting prostate-specific antigen in metastatic CRPC patients, the primary end-point, progression-free survival, was also not met, but again, an overall survival advantage favoring the investigational agent was observed.13 In phase II and III trials of ipilimumab for the treatment of metastatic melanoma, there was a significant improvement in overall survival among patients.9 In all these studies, response patterns different from those seen with other therapies, for example, chemotherapy, are beginning to be recognized. Thus, there might be a tumor burden increase at first, even with development of new lesions, but nevertheless, it is later followed by a meaningful clinical response. Therefore, new response criteria, which could be termed “immune-related response criteria”, or irRC, have been proposed.14 Because immunotherapy must induce, facilitate and/or amplify cellular immune responses before it can affect tumor burden or patient survival, adjustments of established end-points to address the different kinetics of immunotherapy compared with cytotoxic agents are required for appropriate investigation of future immunotherapies in clinical trials. We can now feel confident that this paradigm shift in cancer immunotherapy is ushering in an era of targeted biological therapy, which will result in much improvement of the cancer patient's lot.

Hirokazu Matsushita M.D., Ph.D. and Kazuhiro Kakimi M.D., Ph.D.

Department of Immunotherapeutics (Medinet), The University of Tokyo Hospital, Tokyo, Japan

Immunotherapy in targeted therapy era

Reality in the cytokine era

inline image

Until 2005, drugs of systemic therapy to metastatic renal cell carcinoma (mRCC), which were shown to be beneficial, were cytokines. Interferon-α (IFN-α) had been proven to have longer survival compared with methyl-progesterone or vinblastine in two randomized trials. Interleukin-2 (IL-2) showed a low, but durable, complete response despite severe toxicity when used in a higher dose. It had to be referred to as a “modest response”, because their response rates were 10–15%. Once patients had progression under cytokine therapy, they succumbed to disease without exception. However, it is true that a small number of patients survived longer than 5, even 10 years, with cytokine therapy.15

Targeted drugs: Their promising results and limitations

The first début of a targeted drug to mRCC was in 2005. Sorafenib, a tyrosin kinase inhibitor (TKI) to inhibit angiogenesis, showed obvious superiority to the placebo in prolongation of progression-free survival (PFS) in patients having regrowth of tumors after cytokine therapy.16 After targeted drugs consecutively represented promising results including longer PFS, higher response rates were found in different clinical settings. Among them, clinical evidence suggests that sunitinib is a first-line drug to favorable and intermediate Memorial Sloan-Kettering Cancer Center (MSKCC) risk RCC patients.17 Temsirolimus is considered as a first line drug for RCC patients with poor risk background (more than two).18 Everolimus is the first choice for patients with disease progression after treatment with TKI.19 The combination of bevacizumab and IFN-α,20 and pazopanib is also a treatment option to naive and cytokine-refractory RCC. Thus, building a new road for the cytokine-refractory “impasse” has surely succeeded with targeted drugs, meaning longer survival is promised after initial cytokine therapy. Also, another initial pathway, “sunitinib” leads to longer survival than that of cytokine, and it is also true with “temsirolimus” in poor risk patients.

One of biggest issues that we have is these choice of drugs do not necessarily give the same survival in each patient who has different a background in terms of tumor and patient character. For instance, no tumor character other than clear cell carcinoma is considered in the well-known RCC treatment algorism. Therefore, it cannot be necessarily applied to non-clear histology. In addition, when we talk about the character as its quality, one can ask where the consideration for tumor quantity is in the algorism. It is absolute that patients bearing a higher tumor volume have poorer prognosis. I cannot help but frown when a resident says, irrelevantly to a patient's and tumor character, and simply selected a drug according to the guidelines.

Cytokines revisited

Thus, it is becoming more evident that “mighty” targeted drugs might be an illusion. One of the new movements of cytokines was to explore the possibility of its combination therapy. A promising regime is bevacizumab plus IFN-α, showing longer PFS comparable with sunitinib. In European countries, this combination therapy is one of the major options of first-line treatment for mRCC. We can learn another reality from a recent high-dose IL-2 treatment study, SELECT (ASCO 2011 # 4514), which showed a remarkable response rate and higher complete response (CR) rate than targeted drugs. As for durable CR, immunotherapy might be dominant to obtain a longer cancer-free status than targeted drugs. Presumably, memorized immune cells to attack RCC cells might render a longer clinical CR by continuing the destruction of newly bearded RCC cells from remnant mother cells. Furthermore, the higher response rate shows that the selection of treatment modality based on the experienced physicians' instinct might be justified. However, it is a medical science to clarify practical evidence for prevailing fruit of “the brilliant instinct based on experience”.

What is the most beneficial treatment in each patient?

It seems that some patient groups might have vulnerable characteristics to cytokine therapy, but it has not been fully investigated. Gene polymorphism21 or apotosis-related molecule expression22 were proven to have a significant correlation to susceptibility to immunotherapy. Clinical demographics that have an implication to response should be accumulated, such as metastasis site, tumor volume, patient age and laboratory findings, because it is easier to apply them to a treatment selection.

A recent important key word has been “order made treatment”, meaning properly individualized treatment. To achieve a more sophisticated made-to-order treatment, more intensive and detailed investigation is warranted in a larger number of mRCC patients. It will give us a “breakthrough” for the present situation, and authentic collaboration is definitely needed.

Yoshihiko Tomita M.D., Ph.D.

Department of Urology, Yamagata University School of Medicine, Yamagata, Japan

Immunotherapy with dendritic cells for renal cell carcinoma

inline image

Immunotherapy plays a significant role in the management of renal cell carcinoma (RCC) patients with metastatic disease and various treatments for RCC, such as cytokine-, antigen- or dendritic cell (DC)-based immunotherapy, have been carried out in multiple clinical trials. Although antitumor immune responses and clinically significant outcomes have been achieved in these trials, the response rate is still low and very few patients show long-term clinical improvement. This review summarizes the principles of antigen-specific immunity and DC-based immunotherapy for RCC.

Antigen-specific immunity and dendritic cells

Tumor antigens are processed by the proteasome within tumor cells and are presented at the surface as peptides on major histocompatibility antigen (MHC) class I molecules. Recognition of these complexes by antigen-specific cytotoxic T lymphocytes (CTL) triggers CTL-mediated cytotoxicity. CTL cannot be activated by direct recognition of antigen on tumor cells; they require activation by antigen-presenting cells (APC), such as dendritic cells (DC), and CD4 helper T cells. The generation of CTL from CD8 T cells requires not only the binding of the T cell receptor on the CD8 T cell with a peptide antigen presented on MHC class I molecules of the APC, but also a costimulatory signal provided by molecules including CD80 and CD86 (Fig. 1). DC act as APC and are able to present foreign antigens to T cells and play a central role in regulating immune responses. They exist in a variety of tissues including lymphoid tissue, non-lymphoid organs and blood. They are able to take up particulate and soluble antigens, and migrate into the lymph nodes. Once in the lymph nodes, DC present antigen to T cells and induce specific immune responses, including the induction of CTL.

Figure 1.

Induction of antigen-specific tumor immunity via dendritic cells.

Since it was first reported that DC can be generated from peripheral blood cells in vitro, immunotherapy with DC has been used to treat cancer patients.23 DC are generated from both myeloid and lymphoid progenitors derived from the bone marrow. Myeloid progenitors induce mainly cellular immunity. Immature DC can be induced from CD34+ cells, CD14+ mononuclear cells and adherent-mononuclear cells in peripheral blood by culturing them in a medium containing granulocyte-macrophage colony-stimulating factor (GM-CSF) and interleukin (IL)-4. These immature DC are phagocytic, but they do not possess a high ability to present antigen. After exposure to cytokines, such as TNF-α, IL-1, IL6 prostaglandin E2 and Flt-3 ligand, immature DC differentiate into mature DC and show a high degree of antigen-presenting ability, the ability to induce and express costimulatory molecules, and the ability to augment MHC class I and class II molecule expression.

Immunotherapy with DC for renal cell carcinoma

Immunotherapy with DC has relied on a variety of methods for generating DC, different types of antigen and different adjuvants.24 DC and antigens for vaccination have been derived from autologous (auto) and allogeneic (allo) cells. Although immunotherapy with allo tumor cells tends to induce immune responses against common antigens of the tumor, immune responses against non-self-antigens can also be induced. Auto-DC might have an advantage for host immune systems compared with allo-DC, because the allo-DC might be recognized and attacked as non-self cells by the immune system.

The serological identification of antigen by recombinant cDNA expression cloning (SEREX) method is a serological approach that combines antigen cloning techniques to identify tumor antigens based on IgG antibodies in patient serum and subsequent identification of the tumor antigens from cDNA libraries. Immunotherapy with DC and tumor peptides that were identified using the SEREX procedure was carried out in some experiments. Because most tumor antigens are only weakly immunogenic, adjuvant approaches using cytokines, keyhole limpet hemocyanin (KLH) and MHC class II peptide have been used.24

Immunotherapy with DC in combination with cytokines

IL-2 can stimulate the maintenance and proliferation of T cells activated by DC in vitro, but the combination of IL-2 with DC therapy was shown to have no advantage in a clinical trial.25 Recent reports show that IL-2 contributes to the maintenance of regulatory T (Treg) cells, which play a key role in regulating immune responses, such as those responsible for autoimmune diseases, graft rejection and antitumor immunity. The suppression of DC-activity by Treg cells included their phenotypic maturation, pro-inflammatory cytokines secretion and the resulting ability to present antigens. In our research on RCC patients, Treg cells were decreased after initiation of IFN-α therapy, but were increased after IL-2 therapy.26

From the aforementioned results, we used IFN-α as an adjunctive agent for DC therapy, because IFN-α enhances the antitumor effect and activates DC.27 We showed the safety and efficacy of combination therapy with IFN-α and DC in patients with progressive renal carcinoma after IFN-α and IL-2 therapy.28 Further examination is required to determine what cytokine will enhance antitumor immunity caused by DC therapy.


DC-based immunotherapy has been evaluated in phase I/II studies that were not randomized and were based on different trial designs. Although these studies showed tumor-specific immune responses, such as delayed-type hypersensitivity reactions (DTH), IFN-γ production or lymphocyte proliferation in response to tumor cells, those responses were surrogate end-points. Significant clinical responses, such as tumor regression, have been seen at a low frequency. The requirements for specialized culture facilities and expertise in DC therapy make it difficult to treat only a small number of cases at a time. Further progress in this field will require larger comparative trials of patients in earlier stages of disease in order to determine the efficacy of this approach.

Katsunori Tatsugami M.D., Ph.D. and Seiji Naito M.D., Ph.D.

Department of Urology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan

Personalized peptide vaccination for renal cancer patients

Cancer vaccine for renal cancer patients

inline image

Of the recent advances in the treatment of metastatic renal cell carcinoma (RCC), immunotherapy remains an important field of investigation. Because RCC is one of the most immunoresponsive cancers in humans, immunotherapy remains a basis of promising treatment strategies. Non-specific stimulations through cytokines, passive specific immunotherapy with antibodies and active specific immunotherapy seem to be suitable options for RCC. The goal of developing curative RCC vaccines is to stimulate the immune system to recognize and to destroy existing tumor cells. RCC vaccines are explored in the metastatic and adjuvant setting.

Therapeutic cancer vaccines are currently under active clinical investigation worldwide.24 Such therapeutic vaccines can be divided into autologous tumor cell-based vaccines, genetically modified tumor cell-based and dendritic cell (DC)-based vaccines, and peptide-based vaccines. We have been investigating peptide vaccination for metastatic RCC patients, because peptides are non-biological chemicals that can be synthesized on an industrial scale under the current standards of good manufacturing practice. In contrast, cell-based vaccines, such as tumor-cell vaccines or peptide-pulsed DC therapies, have several disadvantages, including limited cell sources for each patient, difficulties in maintaining uniform vaccine quality, labor intensity and high production costs.

Identification of target antigens in specific immunotherapy for renal cancer patients

Several new RCC-associated antigens and derived HLA-class I ligands were recently identified. We previously reported that most target antigens encoding peptides used for vaccination were expressed in cell lines from renal cancer cells.29 They were SART1, SART2, SART3, MRP3, EZH2, HER2/neu and PTHrP. In those studies, however, prostate-specific antigen (PSA) or prostatic acid phosphatase (PAP) antigen were undetectable in the cell lines. Therefore, we further investigated expression of PSA and PAP antigens in the primary culture of both RCC cells and non-tumorous kidney cells by the reverse transcription polymerase chain reaction method, and found the expression of PAP, but not PSA, in both types of cells. Subsequently, we investigated PSA protein expression in metastatic RCC cells by the immunochemical staining method, and found that PSA antigens were expressed in RCC cells from two of four samples that were surgically harvested from lung metastases. Furthermore, we found that carcinoembryonic antigen, ubiquitin-conjugated enzyme variant Kua and Lck antigens were also expressed in both types of the cells.

What is a personalized vaccination?

We showed that each cancer patient has different sets of activated T cells against cancer antigens, and that these activated T cells are detectable before vaccination. Our approach to developing a cancer vaccine for advanced cancer patients is to carry out six weekly injections of four peptides selected from the patients' own peripheral blood mononuclear cells (PBMC) based on the strength of pre-existing immunity toward the cancer antigen; we call this a personalized vaccination. We started a phase I clinical study of peptide vaccination with a regimen of conventional prophylactic vaccination; however, the predesigned vaccination induced a weak primary immune response and no clinical response. Personalized vaccinations have induced both a strong secondary immune response and a clinical response.30


There are as yet no definitive biomarkers to predict clinical responses, which hamper the development of cancer vaccines. In a total of 500 advanced cancer patients who received a personalized peptide vaccination, we have shown that the IgG response is superior to cytotoxic T lymphocyte (CTL) response as an immunological biomarker that is predictive of overall survival.31 These results might provide new insights to better understand biomarkers of cancer vaccine for advanced cancer patients. Application of these results to other types of cancer vaccine using common proteins or common peptides in a non-personalized manner could be worth considering.

Phase I trial of personalized vaccination for renal cancer patients

Only a limited number of clinical studies with peptide-based RCC vaccines have been reported to date. We carried out a phase I trial of personalized peptide vaccination for cytokine-refractory metastatic renal cancer patients.32 Among 10 patients, there were no major adverse events, although most of the patients developed grade 1 or 2 local dermatological reactions at the vaccine injection sites. There were no hematological, hepatic or neurological toxicities, and performance status remained stable during the vaccine treatment. There was only a slight increase in peptide-specific interferon-γ production in the postvaccination PBMC, despite the higher levels of CTL activity in prevaccination PBMC. In contrast, an increase in peptide-specific IgG of postvaccination (sixth) plasma was observed in the majority of patients. These findings show that humoral responses, but not cellular responses, were markedly boosted by personalized peptide vaccination in cytokine-refractory metastatic RCC patients. These findings might encourage further clinical trials of personalized peptide vaccination. Combination therapy with personalized peptide vaccination and a molecular target drug or cytokines might provide a breakthrough in the treatment of advanced renal cancer.


Despite the limited clinical efficacy of most of the therapeutic vaccines in RCC studies to date and their often small number of patients and non-standardized methodology, there is still interest in the use of vaccines that have much less toxicity than other current therapies for RCC. The discovery of new tumor-associated antigens or immunostimulatory peptides, and increasing insight in basic immunology and molecular biology will certainly lead to the development of more powerful RCC vaccines.

Shigetaka Suekane M.D., Ph.D.,1 Masanori Noguchi M.D., Ph.D.,2 Fukuko Moriya M.D., Ph.D.,1 Kei Matsuoka M.D., Ph.D.1 and Kyogo Itoh M.D., Ph.D.3

1Department of Urology,
2Clinical Research Division of the Research Center for Innovative Cancer Therapy,
3Department of Immunology and Immunotherapy, Kurume University School of Medicine, Fukuoka, Japan

New strategy of adoptive-immunotherapy using γδ T cells

inline image

More than a half a century has passed since Foley reported that mice could reject the sarcomas re-inoculated in skin after resection of the same strain of sarcomas. This research showed that the immune system is involved in tumor surveillance. Morgan discovered T cell growth factor in 1976 and Taniguchi identified the involvement of the interleukin (IL)-2 gene in 1983. In the 1980s, Rosenberg developed the use of adoptive immunotherapy of lymphokine-activated killer (LAK) and systemic administration of recombinant IL-2. Contrary to expectations, although initially showing promise, the attempt did not work well in the long term. LAK cells are composed of antigen non-specific T cells and natural killer cells, and have the ability to kill various types of human tumor cells in vitro. LAK cells are also efficient at killing such cells in vivo and even in mouse models. The cause of eventual failure was not fully understood at that time. Boon reported in 1991 the identification of the human gene MAGE-1, which directs the expression of antigen recognized on a melanoma by autologous cytolytic T lymphocytes (CTL). Many researchers worked energetically to discover tumor-associated and/or tumor-rejected antigens, and various approaches were developed involving immunotherapy, using mainly antigen specific αβ T cells as CTL. Active immunization with tumor antigens and also tumor antigen-pulsed dendritic cells (DC) has the same objective of inducing antigen-specific αβ T cells.33 In the immune system, αβ T cells are so sophisticated that they can not adapt to mutation, loss or downregulation of tumor antigens on the tumor cells, which are themajor mechanisms of escape from immunotherapy based on αβ T cells.

The T cell receptor (TCR)-γ chain gene was discovered by chance during identification of TCR αβ chain genes, and it was established in around 1986 that γδ T cells were one of the subpopulations of T cells. The behavior of natural ligands of γδ TCR was unclear for quite a while until Tanaka reported in 1995 that γδ T cells recognize isopentenyl-pyrophosphate (IPP) by the TCR.

We have synthesized more than 60 kinds of monoethyl-pyrophosphate derivatives and 2-methyl-3-butenyl-1-pyrophosphate (2M3B1PP), which show potent stimulation of γδ T cells more than 100-fold that of IPP. We developed a bulk-culture system of γδ T cells using 2M3B1PP and IL-2.34γδ T cells induced by 2M3B1PP with potent cytotoxic activity against not only various types of tumor cell lines, but also against autologous renal cell carcinoma (RCC) cells, and we attempted to apply γδ T cells as a new adoptive immunotherapy.35

Cytotoxic activity of γδ T cells is induced by recognition of stress-inducible major histocompatibility antigen (MHC) class I chain-related A (MICA) proteins on target tumor cells by natural killer group 2D (NKG2D) of γδ T cells. MHC class I on target tumor cells induced an inhibitory signal through NKG2A/CD94 to γδ T cells at the same time, but cells lacking in MHC class I can be killed by γδ T cells. In contrast, many tumor cells upregulate the mevalonate-pathway. IPP, which is one of the intermediate metabolisms of the pathway, accumulates in tumor cells and γδ T cells, and recognizes IPP of the TCR and shows cytotoxity by release of perfolin, granzyme and TNF-related apoptosis-inducing ligand, resulting in apoptosis of the tumor cells. Also, activated γδ T cells secrete various types of cytokines, such as interferon-γ (IFN-γ), IL-2 and TNF-α, and promote Th-1 type immune reaction in other immune cells. Nitrogen-containing bisphosphonates, such as zoledronic acid (Zol), inhibit farnesyl-pyrophosphate (FPP) synthetase, which is one of the important enzymes in the mevalonate pathway. Inhibition of FPP synthetase resulted in an accumulation of IPP in the tumor cells, which γδ T cells recognizes and then kills the tumor cells with ease (Fig. 2). Assembling these strands of evidence and applying this knowledge of γδ T cells as effector cells has advantages for immunotherapy compared with αβ T cells. γδ T cells make good use of the universal characteristics of tumor cells in surveillance.

Figure 2.

Activation and inhibition mechanisms of γδ T cells in tumor stroma. γδ T cell recognizes isopentenyl-pyrophosphate accumulated in tumor cell by T cell receptor and major histocompatibility antigen (MHC) class I chain-related A (MICA) on tumor cell by NKG2D, and then γδ T cells show potent cytotoxicity against tumor cells. Tumor stroma contains various kinds of cells to inhibit γδ T cell functions and also tumor cells inhibits γδ T cell functions by interaction of the inhibitory molecules.

We set up a phase I/IIa clinical trial of this new approach of adoptive immunotherapy using autologous γδ T cells induced by 2M3B1PP followed by administration of low-dose IL-2 and Zol. Patients who underwent nephrectomy because of RCC and received IFN-α therapy for recurrent and/or distant metastasis of RCC as a first-line therapy, which resulted in failure, were enrolled. The protocol was reviewed by our Institutional Review Board and registered to the National Cancer Institute (NCI) as a clinical trial (http://www.cancer.gov/clinicaltrials/TRIC-CTR-GU-05-01). After we had obtained written informed consent, the patients underwent leukopheresis to harvest of peripheral blood mononuclear cells, which were then stimulated with 2M3B1PP and IL-2 for 11 days. All these procedures were carried out in the cell-processing center (CPC) at Tokyo Women's Medical University Hospital. Each patient then received 1.4 million units of IL-2 and 4 mg of Zol. Activated γδ T cells were injected intravenously. IL-2 at 1.4 million units was given for four consecutive days. Each patient received these treatments once a month for a period of 6 months. A total of 11 advanced RCC patients were enrolled in the clinical trial and we obtained the results of one complete response36, five stable disease and five progressive disease. We found that this approach was safe and well tolerated, and we also observed some clinical responses. Based on these results, we have improved this approach and planned a phase II clinical trial now approved by the Japanese Ministry of Health, Labor and Welfare.

γδ T cells were still enigmatic and their mechanisms of tumor surveillance were not fully understood. Recent reports show the pitfalls of γδ T cells-based immunotherapies.37 One of the mechanisms of the pitfalls involves the inhibitory factors produced by tumor cells, such as TGF-β, prostaglandin E2, adenosine, soluble NKG2D ligands, galectin-3, HLA-G and indoleamine 2,3 dioxygenase, which weaken potent cytotoxic activity and proliferation of γδ T cells. Other aspects of the mechanisms involve suppressive cells from the tumor microenviroments, such as regulatory T cells (Treg), immature DC (iDC), myeloid-derived suppressive cells (MDSC), mesenchymal stem cells (MSC) and negative costimulators, such as PD-1 (Fig. 2). Almost all these mechanisms are related to tumor microenvironments, an understanding of which is very important in developing not only γδ T cell-based immunotherapy, but also other immunotherapies.

Taking account of these factors that inhibit successful immunotherapies, prevention of recurrence and/or surveillance of micrometastasis might be most effective strategies for carrying out γδ T cell-based immunotherapy.

In light of the latest knowledge about the mechanisms of tumor escape from the immune system, we believe it is possible to resolve and overcome these problems by carrying out well-designed translational research and clinical trials.

Hirohito Kobayashi M.D., Ph.D.

Department of Urology, Tokyo Women's Medical University, Tokyo, Japan

Personalized treatment in the immunotherapy for metastatic renal cell carcinoma

inline image

There were an estimated 57 760 new cases and 12 980 deaths expected in 2009 from renal cell carcinoma (RCC) in the USA in 2009.38 Approximately 30–40% of patients with malignant renal cortical tumors will either present with or later develop metastatic disease. RCC has a poor prognosis when diagnosed in advanced stages. Clear cell carcinomas, which account for 75–85% of renal tumors, are characterized by loss of the von Hippel-Lindau (VHL) tumor-suppressor gene, leading to overexpression of proangiogenic vascular endothelial growth factor (VEGF) and platelet-derived growth factor-β (PDGFβ).16,39 As a result, the treatment of metastatic RCC (MRCC) in many countries has recently evolved from being predominantly cytokines-based to now being grounded in the use of drugs that target the dysregulated VEGF and PDGFβ pathways.

However, many Japanese urologists still seem to use cytokine therapy for MRCC, especially in patients whose metastases are limited to the lung or lymph nodes. This tendency is based on the good prognosis of Japanese patients with MRCC in the era of cytokine therapy.15 Considering personalized treatment in the immunotherapy for MRCC, we carried out a retrospective analysis to find out good responders in the interferon-α (IFN-α) treatment for MRCC in Japan. The analysis showed that the single nucleotide polymorphisms (SNP) in signal transducer and activator 3 (STAT3) were most significantly associated with a better response to IFN-α.21 Linkage disequilibrium mapping showed that the SNP in the 5′ region of STAT3, rs4796793, was the most significant predictor of IFN-α response (odds ratio [OR] = 2.73; 95% confidence interval [CI], 1.38–5.78; Fig. 3).21 The highest OR was shown in the CC genotype at rs4796793 compared with the GG + GC genotypes (OR = 8.38, 95% CI, 1.63–42.96; Fig. 3).21 If we would use the CC genotype at rs4796793 as a predictive marker of response to IFN-α therapy in MRCC patients, the positive and negative predictive value would be 52.8% and 88.2%, respectively, on the assumption that response rate of IFN-α is 15% (data not shown). To prove our retrospective SNP result in STAT3, we have been doing a prospective RCC-SNP Ensuring-study for Leading Eligibility of patients in Cytokine Therapy (SELECT) trial since 2007. We hope that we will soon show the final results of the prospective trial on IFN-α.

Figure 3.

Association of rs4796793 with interferon-α (IFN-α) response. The upper figure shows the location of rs4796793 in the STAT3 gene. The lower figure shows the significant association between rs4796793 and IFN-α response. χ2 testing was carried out for the alleles or the genotypes of rs4796793. NON, non-responder; RES, responder (cited from ref.21).

Interestingly, a similar “SELECT” trial using high-dose (HD) interleukin-2 (IL-2) in patients with MRCC was also recently shown.40 Their purpose was to pick up good responders before they began IL-2 therapy. The response rate (28%) for HD IL-2 in their trial was significantly better than the historical experience, likely as a result of improved patient selection (high incidence of prior nephrectomy, low incidence of non-clear cell carcinoma, etc.).40 However, in their trial, analysis of tumor-based predictive markers through central pathology review and staining for carbonic anhydrase 9 (CA-9) was unable to improve the selection criteria for HD IL-2.40 UCLA Survival after Nephrectomy and Immunotherapy (SANI) score41 only showed the possibility to identify patients who were unlikely to respond to HD ID-2.40 These results in IL-2 showed the difficulty in prospectively identifying predictive markers in the immunotherapy for MRCC. However, we believe that our trial to discover good responders in IFN-α treatment will surely lead to personalized treatment in the immunotherapy for MRCC in the era of molecular targeted therapy.

Masatoshi Eto M.D., Ph.D., Wataru Takahashi M.D., Ph.D., Yoshiaki Kawano M.D., Ph.D. and Yoshihiro Wada M.D., Ph.D.

Department of Urology, Faculty of Life Sciences, Kumamoto University, Kumamoto, Japan