Immunotherapy for superficial bladder cancer

Abstract

The treatment of superficial bladder cancer requires adjuvant therapies besides transurethral resection because of a high recurrence rate after this standard treatment alone. Current adjuvant therapies involve intravesical chemotherapy for patients at low and intermediate risk for recurrence and progression, and intravesical bacillus Calmette-Guérin for patients at intermediate and high risk. However, these adjuvant therapies fail in a significant number of patients, dictating the need for new and improved adjuvant treatment modalities for superficial bladder cancer. Immunotherapy aiming at the modulation of the immune system of the patient is a promising alternative adjuvant. This review discusses the current status of the clinical development of various immunotherapy approaches for superficial bladder cancer, including passive immunotherapy, immune stimulants, immunogene therapy and cancer vaccination.

Bladder cancer

Bladder cancer is the ninth most common cancer worldwide and the fourth most frequent cancer among men in the USA. Worldwide, about 336,000 new cases were registered in 2000, two-thirds of them in developed countries, and the mortality is estimated at more than 130,000 patients per year [1]. The incidence of bladder cancer is about three times higher in men than in women.

More than 90% of bladder cancers are transitional cell carcinoma (TCC), while the remainder are accounted for by adenocarcinoma and squamous cell carcinoma. The transitional epithelial cells are located in the inner mucosal layer of the bladder wall. The mucosal layer is separated from the outer muscle layer by the lamina propria. About 70% of bladder tumors are limited to the mucosa and do not invade the muscle layer. These tumors are therefore also called superficial bladder cancer. Treatment of bladder cancer depends on the stage and pathological grade of the tumor, and often multimodality therapy is required. An overview of the current treatments for bladder cancer is presented in Table 1. The therapeutic regimen for superficial bladder cancer is dictated by the risk of recurrence and progression, which can be predicted by clinical and pathological data (see Table 1).

Table 1.

Current standard treatments for bladder cancer

Classification	Description	Treatment
Tis	Carcinoma in situ restricted to mucosa	Intravesical BCG, if needed followed by radical cystectomy
Ta	Noninvasive papillary carcinoma	TUR + adjuvant intravesical chemotherapy for low- and intermediate-risk tumors or adjuvant BCG for intermediate- and high-risk tumors
T1	Tumor invades subepithelial connective tissue	TUR + adjuvant intravesical chemotherapy for low- and intermediate-risk tumors or adjuvant BCG for intermediate- and high-risk tumors
T2	Tumor invades muscle	Radical cystectomy
	T2a superficial muscle
	T2b deep muscle
T3	Tumor invades perivesical tissue	Radical cystectomy
	T3a microscopically
	T3b macroscopically
T4	Tumor invades neighboring organs	Radical cystectomy
	T4a prostate, uterus, vagina
	T4b pelvic or abdominal wall
N+ and metastases	Metastases in regional lymph nodes, distant lymph nodes and/or distant organs	Radical cystectomy + systemic chemotherapy

Classification is according to the TNM classification of malignant tumors (International Union Against Cancer, Sixth Edition, 2002). Tumors with Tis, Ta and T1 classification are considered superficial bladder cancer. Three risk groups are discriminated for these tumors (based on the number of tumors at diagnosis, recurrence rate in the previous period, tumor size and anaplasia grade), namely low-risk tumors (single, Ta, grade 1, ≤3 cm diameter), high-risk tumors (T1, grade 3, multifocal or highly recurrent, Tis) and intermediate-risk tumors (all other tumors, Ta-1, grade 1–2, multifocal, >3 cm diameter). For more details on treatment options, see review by Oosterlinck et al. [2]

After transurethral resection (TUR) of superficial bladder cancer, the standard treatment for this type of tumors, about 60–70% of the tumors will recur of which 25% will progress to a higher stage or grade. The risk of recurrence can be reduced by the use of adjuvant therapies. Intravesical chemotherapy as an adjuvant has been shown to reduce the recurrence rate and to increase the recurrence-free interval ([3, 4], for review see [5]). However, this adjuvant has no effect on the time to progression to muscle-invasive disease [6, 7]. In addition, maintenance chemotherapy fails to improve the beneficial effects of intravesical chemotherapy. Intravesical bacillus Calmette-Guérin (BCG) immunotherapy is another effective form of adjuvant for the treatment of superficial bladder cancer (for review see [8]). The recurrence rate after TUR has been demonstrated to be significantly reduced by intravesical BCG [9]. In contrast to intravesical chemotherapy, BCG also decreases the risk of progression [10], and the beneficial effects of BCG are enforced by maintenance treatment [11]. Although superiority of intravesical BCG over chemotherapy for the prevention of tumor recurrence has been demonstrated [12, 13], it is generally accepted that only patients with bladder cancer at intermediate and high risk for recurrence and progression should be treated with intravesical BCG due to its significant toxicity.

BCG is now accepted to be the gold standard for intravesical adjuvant treatment of superficial bladder cancer with an intermediate or high risk of recurrence and progression. There is currently no other drug or therapy available for the adjuvant treatment of superficial bladder cancer that has proven to be as efficacious as BCG. Unfortunately, BCG is not the adjuvant remedy for all patients. About one-third of patients fail to respond to BCG [14], while about one-third of the responders will develop recurrent tumors with a poor prognosis [15, 16]. In addition, with long-term follow-up, the tumors of a substantial proportion of patients will eventually progress [17]. As a result, there is an apparent need for further improvement of adjuvant therapies for superficial bladder cancer. The mechanism of action of BCG is based on the induction of an immune response, as discussed later. Taking the efficacy of BCG into consideration, superficial bladder cancer is obviously sensitive to immunomodulation. Thus, immunotherapy aiming at the modulation of the patients’ immune system offers a promising tool for new and optimized adjuvant treatment for this type of cancer. This review presents the current status of the clinical development of immunotherapy for superficial bladder cancer.

Immunotherapy for cancer

It is well known that cancer can evoke an immune response by both the innate and the cellular adaptive immune system, directed at the removal of the malignant cells from the body. In the initial phase of this response, natural killer (NK) cells and macrophages, the most important effector cells of the innate immune system, recognize tumor cells as non-self through recognition of expression patterns of activating and inhibitory receptors after cell–cell contact. Subsequently, these effector cells kill the tumor cells through Fas–Fas ligand interaction, or through the production of cytokines and lytic granules. Tumor-derived polypeptides or proteins released during this process are endocytosed and intracellularly processed into tumor-associated antigen peptides by dendritic cells (DCs), which are professional antigen-presenting cells (APCs). These cells form a bridge between the innate and adaptive immune system by presenting the tumor-associated antigen peptides in complex with major histocompatibility complex (MHC) molecules both to T helper type 1 (Th1) cells and to cytotoxic T cells. It is crucial that Th1 cells become activated in this initial phase of the adaptive immune response, since these cells direct the cascade of reactions towards a cellular immune response by stimulating and assisting cytotoxic T cells, the effector cells of the cellular adaptive immune system. In contrast, activation of Th2 cells triggers a humoral immune response. The subpopulation of cytotoxic T cells recognizing the tumor-associated antigen peptides presented will expand and subsequently migrate to the tumor in order to specifically kill the tumor cells bearing the recognized antigens. At the end, the cellular adaptive immune response will also generate antigen-specific memory effector T cells that can protect against recurrent tumor cells or metastases, thus foreseeing immunosurveillance. The innate and cellular adaptive immune systems are tightly regulated and linked by cytokines and chemokines produced by the various immune cells. These proteins also modulate processes like tumor growth and angiogenesis.

Theoretically, the innate and cellular adaptive immune systems should be capable of clearing malignant cells from the body. However, the majority of tumor cells often develop several immunosuppressive mechanisms, including antigen masking and downregulation of the expression of MHC molecules and tumor antigenic peptide processing, processes that are essential for the recognition of a tumor cell by T cells. Furthermore, tumor cells are capable of inhibiting immune trafficking. As a result, the innate and cellular adaptive immune systems fail to respond, leading to uncontrolled tumor progression and metastasis. Immunotherapy aims at the modulation of the immune system of the patient in order to overcome the tumor immunotolerance. Two main forms of immunotherapy can be distinguished, namely passive immunotherapy and active immunotherapy. Passive immunotherapy uses immunologic therapeutics that after transfer to the patient directly target the tumor cells and do not activate the immune system. In contrast, therapeutics used in active immunotherapy do not interfere directly with the tumor cells but trigger the immune system of the patient to do so. Both forms of immunotherapy have been or are currently being studied extensively in a preclinical and clinical setting for a variety of cancers, including superficial bladder cancer, with encouraging results.

Passive immunotherapy for superficial bladder cancer

For the treatment of superficial bladder cancer, the number of clinical reports on passive immunotherapy is limited. An overview of the results obtained with clinical testing of monoclonal antibodies or immune cells is presented in this section.

Monoclonal antibodies

For a broad range of malignancies, monoclonal antibodies directed against tumor-selective antigens with and without complexion with a drug, isotope or toxin have been studied in preclinical and clinical studies. A successful example is trastuzumab (commercial name Herceptin, supplied by Genentech, Inc. San Francisco, CA, USA), an FDA-approved monoclonal antibody directed against human epidermal growth factor receptor 2 (HER2) and indicated for the treatment of HER2-positive metastatic breast cancer. This monoclonal antibody is also being studied currently in combination with chemotherapy in a phase II trial in patients with HER2-positive locally recurrent or metastatic urinary tract cancer, including bladder cancer, at the University of Michigan Comprehensive Cancer Center, USA [18, 19].

For the treatment of bladder cancer, no clinical phase I or II trial using monoclonal antibodies has yet been published. It has been demonstrated in a clinical setting that intravesical administration of an anti-MUC1 mucin monoclonal antibody labeled with a radioisotope leads to efficient uptake of the agent by tumor cells in patients with superficial bladder cancer [20]. Nevertheless, it was concluded from this study that increased retention of the therapeutics was required for further development of an effective therapy. No data on efficacy of this treatment have been published.

Immune cells

With respect to non-specific cell-mediated passive immunotherapy, one clinical trial has demonstrated the safety and efficacy of intravesical instillation with macrophage-activated killer (MAK) cells in 17 patients with superficial bladder cancer after TUR [21]. MAK cells are generated by in vitro activation of autologous mononuclear cells by interferon-gamma (IFN-γ). Upon transfer to the patient, these cells are believed to kill tumor cells in a non-specific way. It was found that the number of recurrences during the 1-year follow-up (eight recurrences) was decreased compared to the year before treatment (34 recurrences). It should be noted that the rather long interval between TUR and the first infusion of MAK cells, about 48 days, may have negatively influenced the efficacy of the adjuvant treatment.

For specific cell-mediated passive immunotherapy, T cells are isolated from the patient and stimulated in vitro by tumor-associated antigens. In this way, activated tumor-specific T cells can be returned to the patient for specific destruction of tumor cells. This approach is also called adoptive immunotherapy. No reports on the clinical evaluation of this therapeutic modality for superficial bladder cancer are available. Nevertheless, a preclinical study on the efficacy of T cells specific to over-expressed p53, obtained from mice immunized with an immunodominant peptide of human p53, in a murine bladder cancer model indicates that adoptive immunotherapy might also be feasible for this type of malignancy [22]. A major hurdle for the clinical development of adoptive immunotherapy may be the requirement of suitable tumor-associated antigens that are specific for bladder cancer and therefore, this approach requires a substantial amount of preclinical research. Tumor antigens derived from the MAGE family of cancer-testis antigens may be suitable in this respect [23–26]. Adoptive immunotherapy has proven to be successful for the treatment of melanoma [27, 28], and this methodology may be successful as well for the treatment of superficial bladder cancer.

Active immunotherapy for superficial bladder cancer

Therapeutics used in active immunotherapy directly stimulate the innate and cellular adaptive immune system of the patient to attack the tumor cells. There are various strategies for active immunotherapy that have been studied for various types of cancer. An overview of the strategies clinically tested for superficial bladder cancer is presented in this section.

Immune stimulants of non-human origin

Various non-human immune stimulants including BCG, keyhole limpet hemocyanin derived from a murine mollusk [29], rubratin derived from Nocardia rubra [30], Corynebacterium parvum [31] and a mycobacterial cell wall extract from Mycobacterium phlei [32] have been evaluated in preclinical and clinical studies as adjuvant therapy for superficial bladder cancer. Of these agents, BCG is believed to be the most effective and has become the gold standard for adjuvant immunotherapy in superficial bladder cancer. Extensive research performed on BCG immunotherapy has shown that this type of treatment induces a local inflammatory response. The exact mode of action of BCG is still unknown. Since athymic mice were reported as not developing an anti-tumor response upon treatment with BCG [33], an intact immune system is a prerequisite. Evidence for the involvement of both the cellular adaptive and the innate immune system comes from both preclinical and clinical studies. T-cell depletion in immunocompetent mice led to a decrease of the BCG-mediated anti-tumor response [34]. In addition, the ineffective response to BCG therapy in mice deficient in NK cells or in mice depleted of NK cells by treatment with a monoclonal antibody provides evidence for the importance of NK cells [35]. In the bladder wall of patients with bladder carcinoma treated with intravesical BCG, infiltration by predominantly T cells and macrophages was observed [36]. Furthermore, the production of various cytokines in the bladder wall has been found [37], as well as urinary cytokine secretion [38] and pyuria [39].

The current hypothesis for the mechanism of action of BCG is that as a first step BCG binds to the bladder wall through the extracellular matrix protein fibronectin [40]. Subsequently, BCG is internalized by the urothelial cells via endocytosis, followed by the recruitment and activation of macrophages and Th1 cells most likely through chemokines secreted by urothelial cells. Activation of Th1 cells may be established by DC-mediated presentation of endocytosed tumor antigens that have been released by cells damaged by the BCG infection. Th1 cells induce the proliferation and activation of cytotoxic T cells and lymphokine-activated killer (LAK) cells through the secretion of cytokines like IL-2 and IFN-γ. In addition, tumor necrosis factor-α (TNF-α) produced by activated macrophages has a direct cytotoxic action on tumor cells and inhibits tumor cell proliferation as well as angiogenesis. Besides the activation of leukocytes, BCG may modulate the phenotype of the tumor cells by the induction of adhesion molecules like MHC molecules and ICAM-1 [41, 42]. In this way, the interaction with effector cells is facilitated, thus making the tumor cells a better target for elimination. Eventually, the tumor cells are destroyed directly by a number of effector cells including cytotoxic T cells, BCG-activated (BAK) killer cells, and LAK cells, as well as indirectly by the anti-proliferative and anti-angiogenic effects of the cytokines released.

Preclinical and clinical research is currently concentrating on the improvement of the efficacy of BCG using various strategies. These include combined treatment with BCG and a cytokine [43, 44], treatment with recombinant BCG expressing a cytokine [45], treatment with a BCG DNA vaccine [46], and combined treatment with a BCG DNA vaccine and a cytokine DNA vaccine [47].

Cytokines

Cytokines are naturally occurring mediators of the immune system encompassing proteins with a variety of biological functions ranging from immune cell activation (e.g. interleukins, colony stimulating factors and interferons) to direct tumor cytotoxicity (e.g. TNF-α). Various clinical trials have studied the anti-cancer effects of recombinant cytokines like IFN-α, IFN-γ, IL-2, and IL-12. Unfortunately, intravenous delivery of cytokines in patients appeared to result in severe and sometimes lethal toxicity. This problem is overcome by administration via intravesical instillation in patients with superficial bladder cancer. Indeed, various clinical trials have demonstrated the safety and efficacy with respect to tumor size, progression or recurrence rate of adjuvant instillation of a number of cytokines, including IFN-α [48], IFN-γ [49], IL-2 [50], and TNF [51]. A disadvantage of cytokine-based immunotherapy is that recombinant cytokines produced by bacteria lack post-translational modifications. This often results in a decrease in efficacy due to instability and increased clearance, leading to the need for high dosages and repetitive administration and as a result to high production costs. In addition, to date BCG immunotherapy has appeared superior to cytokine immunotherapy. A combination of BCG and cytokine immunotherapy is currently being evaluated for its value to improve the efficacy of BCG.

Immunogene therapy

The problems of recombinant cytokines with respect to instability and toxicity due to the required use of high doses may be overcome by immunogene therapy. This form of immunotherapy uses therapeutic gene transfer to induce the expression of an immunomodulator like a cytokine by tumor cells or surrounding normal cells, resulting in the stimulation of the immune system. An advantage of gene therapy is that targeted, sustained, and controlled expression of an immunomodulator can be achieved at the site of the tumor. A disadvantage is that the current vectors used in gene therapy are not able to penetrate multiple cell layers. However, for the treatment of superficial bladder cancer it will be sufficient if the vector penetrates the urothelium, the inner layer of the bladder wall where the tumor cells reside. The most commonly used and clinically tested gene therapy vector for cancer treatment is the replication-defective variant of adenovirus. This virus enters a target cell via the coxsackie/adenovirus receptor (CAR) on the cell membrane. There is controversy on the level of expression of CAR in bladder cancer. Studies using bladder cancer cell lines indicate that the expression of CAR is decreased on tumor cells and that other viruses like vaccinia virus and canarypox virus are therefore more potent alternatives [52]. In contrast, clinical bladder cancer specimens were found to be positive for CAR as well as for alpha integrins, other membrane proteins involved in adenovirus binding [53]. However, the expression of CAR appeared to decrease with increasing stage and grade of the tumor [54], suggesting that adenovirus is primarily suitable in gene therapy for superficial bladder cancer. An alternative for viruses is provided by non-viral vectors. Preclinical studies in an orthotopic mouse bladder cancer model have demonstrated efficient gene transfer as well as an anti-tumor response using either liposomes bearing DNA encoding for cytokines [55] or a vector composed of polyethylenimine and DNA encoding for p53 [56].

Treatment of superficial bladder cancer by (immuno)gene therapy is still in an early clinical phase of development. According to the National Institutes of Health Office of Biotechnology Activities [57] and the National Cancer Institute [19], there is currently only one ongoing immunogene therapy trial, organized by the Cancer Institute of New Jersey, USA. This trial involves treatment by instillation with a fowlpox vector encoding GM-CSF and/or a fowlpox vector encoding TRICOM (TRIad of CO-stimulatory Molecules) as neoadjuvant therapy for patients with muscle-invasive bladder cancer who are scheduled for surgery. TRICOM, a combination of the co-stimulatory molecules ICAM-1, B7.1 and LFA-3, is a product of Therion Biologics Corporation (Cambridge, MA, USA). Until now, details of only one clinical (non-immuno)gene therapy trial have been published for bladder cancer. In this phase I trial by Pagliaro et al. [58], 13 patients with locally advanced bladder cancer were treated by adenovirus-mediated transfer of the human p53 gene. No serious toxicity was observed up to a dose of 10¹² virus particles, and evidence for an anti-tumor response was noted in one patient, which was most likely due to an aspecific inflammatory response induced by the viral vector. This first trial indicates that gene therapy can be safely applied for bladder cancer, but also that issues like improved penetration of the mucosal layer, more efficient gene delivery and improved transgene expression should be addressed in future studies, including immunogene therapy trials. For the improvement of the vector passage through the mucosal layer, a polyamide called Syn-3 is a promising transduction enhancer [59].

Encouraging preclinical results have been published on immunogene therapy for bladder cancer directed at CD40. CD40, a member of the TNF receptor family, is expressed on the cell membranes of all APCs, including DCs and B cells. For DCs, ligation of CD40 leads to the activation of these cells, resulting in enhanced antigen presentation, increased expression of co-stimulatory molecules and the release of various cytokines. The natural ligand for CD40 is CD154, also known as CD40L. This membrane glycoprotein is transiently expressed on activated leukocytes like T-helper cells, cytotoxic T cells and NK cells. The interaction between DCs and T cells via CD40 and CD154 can be considered critical for the initiation and maintenance of the cellular adaptive immune response. Thus, the ligation of CD40 by therapeutic CD154 provides an attractive tool for active immunotherapy. Besides anti-tumor activity through the stimulation of the cellular immune system, therapeutic ligation of CD40 can also directly combat bladder tumor cells expressing CD40 by inhibition of tumor growth via CD40-mediated apoptosis [59–62]. Clinical trials studying the efficacy of CD154-based cancer therapy using either recombinant CD154 or autologous CD154-transfected tumor cells have shown encouraging results for a number of solid cancers and leukemia (for review see [62]). For epithelial cancers, CD154-based cancer therapy remains to be clinically tested. Preclinical data support the potential value of CD154 immunogene therapy for the treatment of superficial bladder cancer. In experimental mouse bladder cancer models, the expression of CD154 by bladder tumor cells was established through adenoviral- or retroviral-mediated transduction [63–65]. Inoculation of CD154-expressing tumor cells in mice did not result in tumor outgrowth, in contrast to parental tumor cells. Vaccination of mice with CD154-expressing tumor cells could protect against challenge with parental tumor. These anti-tumor effects were found to be mediated by T-helper cells and cytotoxic T cells. In addition, it was observed that CD154-transduced human bladder cell lines had a decreased growth rate, increased apoptosis and modulated expression of costimulatory molecules [65]. Finally, it was reported that autologous tumor cells from patients with renal cell cancer transduced with CD154 could induce the maturation and activation of DCs and T cells [65].

Vaccination

Cancer vaccination involves the administration of tumor antigens to a patient with cancer in order to evoke a tumor-specific cellular adaptive immune response and is extensively being studied for a number of cancers. The principle of vaccination as active immunotherapy for cancer is based on the prerequisite of proper antigen presentation for efficient cellular adaptive immunity against cancer. Antigens should be presented to T cells in complex with MHC molecules expressed on the cell membrane, type I molecules for cytotoxic T cells and type II molecules for T-helper cells. Since tumor cells usually lack immunogenic antigens and downregulate expression of MHC molecules on their cell membrane, they will not easily activate the adaptive immune system. By vaccinating patients with tumor antigens, this escape mechanism may be circumvented. A cancer vaccine can be composed of whole tumor cells or antigenic peptides derived from defined cancer-associated antigen(s), if known for the type of cancer to be treated. Examples of such cancer-associated antigens are melanocyte differentation antigens (Melan A/MART-1, tyrosinase, gp100) and cancer-testis antigens (MAGE-3 and NY-ESO-1). A cancer vaccine can be introduced to the patients as whole tumor cells or as a tumor cell lysate (either stand-alone or loaded on APCs), peptides derived from tumor antigens (either stand-alone or loaded on APCs) or encoding DNA packaged in a plasmid or viral vector, and is often administered in combination with adjuvant immunomodulators like BCG, keyhole limpet hemocyanin or cytokines. Various vaccination approaches, which were recently discussed in an excellent review by Ribas et al. [66], have been tested in a clinical setting for their safety and efficacy. No severe toxicity of vaccination has been reported so far, and clinical efficacy results are promising for a number of vaccination strategies. An overview of vaccination strategies is presented in Table 2.

Table 2.

Overview of cancer vaccination strategies

Vaccine	Clinical efficacy	References	Potential disadvantage
Autologous or allogeneic whole tumor cells or tumor lysate	Immune responses reported, no clear clinical response observed in most trials	[67–69]	Complexity of personalized vaccine. Low-immunogenicity tumor cells due to immunosuppressive mechanisms.
Tumor-derived peptides	Clinical responses reported for melanoma	[70–73]	Knowledge on tumor-specific antigen and matching HLA haplotype required. Only suitable for HLA-matching patients. Risk of changing antigen pattern on tumor cells.
Gene-modified autologous or allogeneic tumor cellsa	Clinical responses reported for a number of cancers	[74–78]	Complexity of personalized vaccine. Low-immunogenicity tumor cells due to immunosuppressive mechanisms.
Viral vector carrying gene encoding for tumor antigenb	Virus- or antigen-specific immune responses reported, no data on clinical response	[79–83]	Knowledge on tumor-specific antigen required. Low immunogenicity due to virus inactivation by neutralizing antibodies, dominant viral epitopes or humoral response.
Naked DNA plasmid encoding for tumor antigen	Immune responses reported, no clinical response observed	[84–86]	Knowledge on tumor-specific antigen required. Low efficacy of transfection APCs in vivo.
Autologous APCs presenting tumor antigenc	Clinical responses reported for a number of cancers	[87–95]	Complexity of personalized vaccine.

Only vaccination strategies tested in a clinical phase I or II trial are included in this overview

^aGene-modified tumor vaccines generally are autologous or allogeneic tumor cells transfected with a gene encoding for an immunostimulator (e.g. GM-CSF, IL-7 and B7.1)

^bViruses commonly used for vaccination are vaccinia and canarypoxvirus (ALVAC)

^cVarious approaches for constructing APCs (predominantly DCs) that present tumor antigens have been clinically tested, including loading with tumor lysates or peptides, transfection with RNA encoding for tumor antigens, and fusion with whole tumor cells

Bladder cancer cells express several cancer-associated antigens, including a number of cancer-testis antigens like MAGE-3, MAGE-A8, MAGE-A12 and NY-ESO-1 ([25, 26], for review see [24]) and mutated p53 (for review see [96]). As such, targets are available for vaccination for this type of cancer. To date, two clinical studies have reported on the vaccination of patients with bladder cancer. In a study by Nishiyama et al. [97], four patients with MAGE-3-positive advanced bladder cancer were treated with autologous DCs loaded with a MAGE-3 peptide without any severe side effects. Three patients showed a significant decrease in the size of lymph node metastases and/or liver metastasis. In another study by Marchand et al. [98], a partial response of 10 months was observed for 1 out of 3 patients with MAGE-3-positive metastatic bladder cancer who were vaccinated with a recombinant MAGE-3 protein. There is one other phase I clinical trial currently ongoing in the USA studying the effectiveness of vaccination as an adjuvant treatment for bladder surgery [17, 18, 56]. This trial, organized by the Memorial Sloan-Kettering Cancer Center, concerns vaccination by injection with the cancer-testis antigen NY-ESO-1 in combination with BCG and GM-CSF as an adjuvant treatment for patients who have undergone cystectomy for transitional cell cancer.

Conclusion

In a large number of clinical trials for other types of cancer, it was concluded that immunotherapy is probably most efficacious in early-stage cancers. As such, immunotherapy is suitable for the adjuvant treatment of superficial bladder cancer. Indeed, the clinical results of immunotherapy trials are promising for their use in a multimodality treatment strategy required for this malignancy. Theoretically, active immunotherapy aimed at the activation of the cellular adaptive immune system is the most attractive form, since this treatment modality can induce immunosurveillance through the generation of memory effector T cells and may in this way protect the patient from tumor recurrence and metastasis. Ongoing and future clinical trials on active immunotherapy for superficial bladder cancer will demonstrate whether this theory can be supported by clinical-based evidence. There is a great challenge in overcoming a number of major hurdles in the clinical development of active immunotherapy, which are, amongst others, dictated by the heterogeneity of cancer within a patient as well as between patients. For immunogene therapy, efficient and targeted gene transfer as well as sustained, stable therapeutic gene expression for a long period of time will be essential for clinical success. For vaccination, most vaccines are currently restricted to patients matching a defined HLA haplotype or are based on autologous material and therefore complex to produce. Standard vaccines that can be applied in large patient populations are desirable but may not be established in the near future.