Keywords

20.1 Introduction

Urothelial carcinoma of the bladder leads to significant morbidity and mortality worldwide, accounting for about 5 % of all cancer deaths in humans [1]. Immunotherapy with Bacillus Calmette-Guérin (BCG) has become a first-line treatment after performing transurethral resection (TUR) on high-grade, nonmuscle, invasive bladder cancer (NMIBC) [2]. In addition, BCG reduces the risk of progression in high-risk NMIBC [3]. Moreover, immunotherapeutic approaches involving cytokines are being used to treat renal cancer.

Nonetheless, the efficacy of immunotherapy in the treatment of both renal and bladder cancer is variable. In this chapter, after briefly discussing the subtypes and staging of bladder cancers in addition to renal cancer, attention is given to the role of immunotherapy in the treatment of bladder and renal cancer and the challenges involved are discussed.

20.2 Histological subtypes and staging

Approximately 90 % of bladder tumors are transitional cell carcinoma (TCC), 5–10 % are squamous, and less than 2 % are adenocarcinoma. Bladder carcinomas are heterogeneous, ranging from superficial papillary tumors to invasive carcinomas.

20.2.1 Nonmuscle Invasive Bladder Tumor

Nonmuscle invasive bladder tumors are carcinomas that do not infiltrate the bladder muscle and represent 70–80 % of bladder tumors. They consist of stage Ta or T1 papillary tumors and carcinoma in situ (CIS). Stage Ta tumors reach the epithelial layer of the bladder, while T1 stage tumors slightly infiltrate the lamina propria. T1 stage tumors are of greater concern than Ta stage, especially stage T1 grade 3 tumors, which are likely to recur quickly and move to a higher stage at recurrence. The CIS shows a high recurrence rate and is often a sign of rapid progress toward infiltration. CIS is in fact known as the superficial type, the most damaging of all types of non-muscle, invasive bladder cancer (NMIBC). However, it is rare and accounts for less than 10 % of NMIBCs diagnosed.

20.2.2 Invasive Bladder Tumor

These tumors invade the lamina propria and reach at least the bladder muscle (stage T2). They can extend to the perirenal fat bladder (stage T3), or even invade nearby organs such as the prostate (stage T4). They require more aggressive therapeutic measures, such as radical cystectomy with or without systemic chemotherapy. Immunotherapy has no valid effect on this type of tumor.

20.3 Clinical Use of BCG Immunotherapy for NMIBC

Bacillus Calmette–Guerin (BCG) is the most common intravesical therapy for treating NMIBC. BCG is a live attenuated strain of Mycobacterium bovis developed in 1921 as a vaccine for tuberculosis, and BCG has since been given to people as vaccination against tuberculosis. Since its first use in 1976, Mycobacterium bovis BCG has been established as the most effective adjuvant treatment for preventing local recurrence and tumor progression, following transurethral resection (TUR) of NMIBC. However, its effectiveness has been both variable and unpredictable. Despite nearly 40 years of clinical use, the mechanism(s) by which the intravesical administration of BCG results in elimination of bladder tumors remains undefined. Although BCG is currently regarded as the most effective treatment available for the management of NMIBC, up to 40 % of patients do not respond to treatment and are at risk of disease recurrence and progression. Unfortunately, predictive markers for recurrence and progression are lacking. In patients with intermediate to high-risk bladder cancer, such as those with high-grade Ta/T1 or CIS, BCG is often given due to the higher risk of disease recurrence and progression. The clinical efficacy of BCG has been demonstrated in a number of randomized trials and meta-analyses. Adjuvant administration of BCG after TUR has been shown to prevent both recurrence and progression compared with TUR alone or TUR with intravesical chemotherapy.

20.3.1 History

Bacillus Calmette–Guérin was used for the first time in 1921 as a vaccine against tuberculosis, and its mycobacterial antitumor effect was observed in 1929 by Pearl [4]. In fact, TB patients developed fewer malignancies than the general population. Coe and Feldman in 1966 [5] showed that the bladder might be the site of delayed hypersensitivity reactions in addition to the skin.

In 1970, BCG was used to treat cases of metastatic melanoma [6]; then, in 1974, Hanna et al. [7] demonstrated the antitumor effect of BCG on the hepatocarcinoma. They defined the foundations of local immunotherapy: a sufficient quantity of live bacilli; direct and prolonged contact between BCG and the tumor; the tumor volume after resection should be as small as possible.

In 1975, by using local BCG therapy, deKernion reported the first treatment of a bladder tumor, a metastatic malignant melanoma. This result led Morales et al. [8] and Martinez-Pineiro et al. [9] in 1976 to test the prophylactic effect of BCG on superficial bladder tumors.

These encouraging results were confirmed by a study by a group of doctors from Laval University [10]. The first controlled study confirming the efficacy of BCG was reported by Lamm et al. [11]. Since then, treatment with intravesical instillation of BCG has proved to be the most effective therapeutic agent in treating superficial bladder tumors, especially carcinoma in situ (CIS) [12]. This efficiency has led its approval by the US Food and Drug Administration agency in 1990.

Arbitrarily, it was decided to perform this treatment as six weekly intravesical instillations. The mode of administration of the treatment has been constantly optimized; nonetheless, the ideal mode has not yet been determined. In addition, it was found that the peak of the immune stimulation was located after the fourth week of instillation, also suggesting that cycles of four instillations might be sufficient instead of six [13]. Maintenance treatment, adding cycles of three instillations every 6 months after the first normal cycle of six instillations showed good results on both progression and recurrence [14, 15].

20.3.2 Effectiveness

Bacillus Calmette–Guérin immunotherapy has particularly improved disease-free survival time and reduced tumor progression, but a subset of patients remains refractory. In these patients, treatment with BCG has no effect on relapse or progression and may instead be a source of lost time in the indication of cystectomy. BCG is a complex organism whose introduction into the body leads to a significant and nonspecific stimulation of the immune system. Viability, the instilled dose, and the schedule of instillations all have an impact on the immune response after BCG administration.

Six independent studies involving a total of 585 patients showed that the recurrence was less frequent among those who had undergone resection and treatment with BCG compared with those who have had a resection (a recurrence rate of 29 % instead of 67 %) [16].

A study by Lamm [17] comparing the response to treatment according to BCG strain, comprising 1,496 patients with CIS, showed a complete response to 1,082 of these (72 %). A controlled study by the South West Oncology Group (SWOG), comparing the use of doxorubicin and BCG, showed that progression was reduced by the latter from 37 to 15 % [18].

A long-term study lasting 10 years [17] showed that the increase went from 63 % for the group treated only by TUR to 38 % for the group treated with the TUR combined with BCG (p = 0.0063) group. Mortality among these same groups decreased from 45 to 25 % (p = 0.03) [19].

20.3.3 Side Effects

Bacillus Calmette–Guérin consists of a pathogenic strain, which, although attenuated, causes an infection that has mainly mild or moderate side effects such as cystitis (67 %), hematuria (23 %), moderate fevers (25 %), and high and increased urinary frequency (71 %) [16].

A study by Lamm [20] on 1,278 patients treated with BCG, reported a frequency of acute cystitis of 91 %; side effects can be serious and in rare cases cause a systemic infection, especially if instillation is administered in the presence of residual sores from the TUR that have not completely healed. Granulomatosis, multivisceral, lung, liver, kidney, or otherwise, may occur on rare occasions. It is not known whether this reflects a real bacillary sepsis infection or if it is the reflection of an immunological reaction of delayed hypersensitivity.

A clinical phase III study comparing treatment response depending on the dose of BCG (75 or 150 mg) showed that a lower dose led to a similar antitumor effect, while minimizing side effects [21, 22]. In the same vein, Lamm et al. [23] showed that increasing the dose of BCG was accompanied by a decrease in the antitumor effect in a murine model; therefore, the relationship between the dose of BCG and the antitumor effect was bell-shaped.

In summary, the current consensus indicates that BCG is administered as primary treatment for CIS, which are the most aggressive superficial bladder tumors, but against which treatment with BCG gives the best results. For tumors with a low risk of recurrence and progression, including low tumor grade and stage, TUR is the best treatment, with or without intravesical instillation of mitomycin C or adriamycin in the hours after TUR.

In the event of recurrence, BCG treatment is recommended, especially when accompanied by a maintenance cycle of three instillations every 3 months. For potentially recurrent tumors, treatment with BCG still yields better results than intravesical chemotherapy, but the side effects are more pronounced.

20.3.4 Mechanism of the Antitumor Effect

The BCG infection causes the sloughing of superficial cells off both the normal and the cancerous bladder. It has now been agreed to take into consideration that the antitumor activity of BCG is run by the local nonspecific immune response of immunocompetent cells [24]. Several immunogenic aspects have been studied after BCG instillations, such as infiltration of the cell wall by the effector cells [25], the involvement of Cytotoxic T lymphocytes [26], the major histocompatibility complex or expression of adhesion molecules on the urothelial cells [27], and the secretion of cytokines. None has been clearly implicated in the antitumor activity of BCG.

Instillation of BCG results in an increase in the granulocytes in the bladder wall followed by the T cells, mainly CD4+ T cells. The proinflammatory phenotype of Thl (IL-2, IL-12, IFN) dominates after stimulation with BCG [27] and it is assumed that this phenotype accompanies a favorable response [28]. Macrophages and other antigen-presenting cells are stimulated after treatment with BCG; it appeared that urothelial cells also internalize mycobacteria and were involved in antigen presentation and cytokine secretion [29].

Another study evaluating the lymphocyte response after instillation of a high or low dose of BCG, or low, but accompanied by IFN-α-2b, showed no difference in the quality or quantity of the immune response driven by these three types of treatment [30]. Forty percent of people who do not respond to BCG have been successfully treated with instillation of IFN-α-2b [31] and more side effects brought about by this type of treatment are less pronounced with the addition of BCG [32].

All cytokines produced after treatment are of the IL-6 type, TNF-α, and IL-1-P, a chemokine (IL-8), and a growth factor (granulocyte macrophage colony-stimulating factor [GM-CSF]). On the other hand, no cytokine associated with the Th pathway was constitutively produced after stimulation. This is because the Th cytokines are produced mainly by immune cells after stimulation. BCG and IFN-α-2b seem to have different and independent stimulatory effects on cytokine production by tumor cells in the bladder [33].

In the mouse model, activated macrophages from the strain susceptible to BCG infection, produce more IL-10 than macrophages activated in the strain resistant to infection. The induction of IL-10 by the pathogen is likely to reduce its immunogenic response to the infected host [34]. In addition, a synergistic effect was observed in the antitumor effect when immunotherapy was performed with BCG/IFN-α blending (murine model). While both types of treatment led to increased levels of CD4+ and CD8+ T cells, the mixture resulted in the most significant increase in αβ T cells [30].

In addition, Hara et al. [35] showed that the antitumor effect was canceled in a mouse model deficient in CD4+ T cells, but not in CD8+ T cell-deficient models, implying that CD4+ T cells are absolutely necessary, unlike CD8+ T cells. In addition, BCG increases the production of the chemokines MCP-1 and RANTES, two chemo-attractants that stimulate the cytostatic response of memory T cells against tumor and the release of lysosomal enzymes [36].

20.3.5 Action of BCG

The interaction of BCG with the luminal surface of the urothelium is the first stage of infection. Accumulation of BCG near the bladder wall and its adhesion factors limit an adequate clinical response. Modulation of the adhesion of BCG to the bladder influences the response induced in mice [37].

This membrane adhesion may be due to nonspecific physicochemical interactions between the wall of the urothelium and BCG or involving specific interactions between receptors and their ligands, such as fibronectin and glycosaminoglycans. For example, the attachment of BCG to the bladder involving fibronectin was blocked by antifibronectin antibodies or by the addition of soluble fibronectin [38]. When adhesion is blocked, there are no hypersensitivity reactions and no tumor rejection. From this contact, mycobacteria are either phagocytosed by macrophages or internalized in tumor or urothelial cells [29].

The local immune response is closely related to the interaction of three systems: the host (the patient), BCG (mycobacteria), and tumor. This interaction leads to a cascade of immunological events, some of which are essential for the protective effect of BCG against recurrence and progression. It is currently thought that there are three phases in the immune response to BCG.

20.3.6 The Role of the Host

The immune response to infection by intracellular germs such as BCG varies and is probably related to the host. In mice, a resistance gene to BCG vaccination has been identified. This natural resistance-associated macrophage protein (Nramp 1) gene is involved in T cell response to vaccination by mycobacteria [39]. The product of this gene expressed by the macrophages plays a role in molecular expression, MHC class II, antigen-presenting in humans and potentially in the inflammatory response [39, 40]. In addition, Nramp1 could control the replication of intracellular bacteria through the cell produced by infected phagosomes [41]. It may also inhibit the development of mycobacteria by promoting the production of nitric oxide (NO), a potent antimycobacterial agent [42].

In humans, the functions encoded by Nramp 1 in the mice would be under the control of many genes, some of which have been isolated on the short arm of chromosome 2 (2q35) [43]. The identification of these genes and their polymorphism may be useful in the future for the prediction of response to BCG. This polymorphism is interesting, as HLA class II antigens are involved in the response to mycobacteria [44].

20.3.7 The Immune Response to Mycobacteria and BCG

The immune response to mycobacteria is related to infection of antigen-presenting cells (monocytes, macrophages, DCs) and is associated with the production of cytokines such as IFN, IL-12, and IL-15. These cytokines are involved in the activation of T helper cells (CD4+) and may be involved in the production of T cell helper 1 (Th1) cells (producing IFN and IL-2) associated with a localized form of TB or T helper 2 (Th2) cells (producing IL-4 and IL-10) and associated with a generalized form of the disease [45]. During intravesical BCG instillation, three phases can be distinguished: initiation, the internalization phase of antigen presentation, and the cytotoxic phase. These last two correspond to the effector phase.

The induction phase is characterized by the contact between BCG and the urothelium. The increased binding to fibronectin may increase the activity of BCG. This mycobacteria adhesion can also be carried out by glycosaminoglycans. From this contact, the bacteria may be phagocytosed by macrophages that belong to the group of antigen-presenting cells or internalized in urothelial cells or tumor cells [46]. This stage involving the adhesion, penetration, and activation of antigen-presenting cells is an important step in the response to mycobacteria and the response to BCG [47].

The effector phase is characterized by the presentation by antigen-presenting cells (APC) to T helper cells, certain proteins of BCG produced by degradation (immunogenic protein), followed by the activation of cytotoxic cells. After mycobacteria infection, macrophages associated with other antigen-presenting cells, which include the urothelial cells [47, 48], manufacturent antigens and a number cytokines (IL-1, IL-6, IL-8, IL-10, IL-12, TNF, IFN) are activated [49, 50]. These cytokines are critical to the recruitment of immune cells (T lymphocytes, macrophages, and neutrophils), which infiltrate the bladder wall in large numbers during instillation [51] in parallel with the over-expression of adhesion molecules (ICAM-1) and co-stimulatory B7 molecules. These cytokines probably amplify the phenomena of antigen presentation. Soluble forms of these adhesion molecules (ICAM-1) are also found in the urine after instillation of BCG, in addition to the over-expression of molecules of major histocompatibility complex (MHC) class I and II by urothelial cells [52]. The MHC class I and II molecules are involved in the phenomena of antigen presentation, as exogenous antigens are usually presented by MHC molecules and class II antigens are expressed by endogenous molecules of MHC class I. The MHC class II molecules are expressed only by APC (macrophages, monocytes, B lymphocytes, DCs, endothelial cells), while MHC class I molecules are expressed by all other cells (except red blood cells and oocytes). These exogenous antigens are degraded by lysosomes and then present on the surface of antigen-presenting cells, bound to the MHC class II molecules. This antigen peptide and MHC class II molecules are then presented to CD4+ (T helper) [53].

In this system, IFN stimulates the power of phagocytosing macrophages and their production of endotoxins. These phenomena are associated with the over-expression of adhesion molecules ICAM-1, LFA3 (APC) and co-stimulatory B7-1 and B7-2 (APC), CD28 (T cells); they probably amplify the response associated with the phenomena of antigen presentation [54]. The antigens are linked to endogenous MHC class I molecules after being prepared in the endoplasmic reticulum and sent to the cell surface, where they are recognized by CD8 lymphocytes [55]. Cytokines thus promote the cytotoxic action of lymphocytes [56], and, in addition to the cytotoxic activity, they protect lymphocytes against tumor cells [57, 58].

Mycobacteria preferentially induces cytokine responses corresponding to a Th1 (IL-2, IL-12, IFN), which is a favorable response to the development of cellular immunity [59, 60]. The Th1 promotes the expansion and proliferation of cytotoxic cells and is characterized by the expression of certain cytokines such as IL-2 or IFN. This Th1 (IL-2, IFN) was detected in the urine of patients after intravesical BCG instillation and is related to the prognosis of the disease [51]. Similarly, after intravesical BCG instillation, overproduction of the RNA messenger for IL-2 occurs in peripheral cells, a phenomenon which correlates with good response to BCG [61]. However, the response to BCG is probably not linear. In addition, it probably not only correlates with instillation doses, but is also related to the instillation protocol.

Some animal experiments suggest that high doses might be capable of inducing immunosuppression and production of Th2-type cytokines [14]. Experimentally, the increase doses cause an inversion of the response, probably associated with the suppressive response of Th2 cells characterized by the production of IL-4, IL-10, IL-5, and IL-6, an immune response to favorable humoral immunity [61]. Also, the production of IL-4 could promote the growth of B lymphocytes, activate complement (C3a, C5), and reduce the expression of IL-1 and TNF.

Cytokines induced by BCG regulate cellular response to cytotoxic vocation, wish is an integral part of the effector phase. Cytotoxic cells most frequently described in the bladder wall after intravesical instillations are CD8+ T cells. Experimentally, lymphocytes (CD4+ and CD8+) are essential to the development of a response against mycobacteria [62]. CD8+ cells seem to have a cytotoxic effect through adhesion molecules (ICAM-1) and/or via the Fas system present on target tumor cells. The population of CD4+/CD8+ was increased in the bladder after intravesical instillation, with a predominance of CD4+ T cells. CD4+ lymphocytes produce cytokines capable of inducing the maturation of Cytotoxic T lymphocytes. The antitumor activity is related to CD4+ and the interaction between Fas and CD40 ligands. Indeed, the interaction between CD40 (membrane glycoprotein of the TNF receptor family) and its ligand may play a major role in the activation of cytotoxic T lymphocytes and promote the Th1 response [63]. Expression of the CD40 ligand on the surface of T lymphocytes may increase monocyte survival by protecting apoptotic phenomena at sites of inflammation [63]. Moreover, the expression of CD40 at the tumor cell surface could act as a replacement to antigen-presenting cells by promoting apoptosis induced by CD4+ cells expressing CD40 ligand on its surface. Similarly, the expression of Fas ligand on the surface of CD4+ lymphocytes is able to induce tumor apoptosis.

Several other cytotoxic cells have been obvious: neutrophils, NK cells, BAK, LAK cells, and gamma delta (ɤɕ). Neutrophils are the cells that are more abundant in the bladder wall after intravesical instillation of BCG [64]. These cells are capable of producing cytokines or soluble cytokine receptors, such as interleukin 1. Soluble IL-1 receptor is capable of reducing the production of IL-1, IL-8, and TNF-α; therefore, the immune response is decreased [64]. Other cells (BAK, LAK) play a direct cytotoxic role against urothelial tumor cells in vitro [65]. These cells co-expressed CD8+ and CD56+ markers on their surface with the possibility of producing IL-12 and initiating an effective antitumor response. Their mechanism of action may involve the Fas-L/Fas system or perforin/granzyme A and B.

In vitro study of the urothelium after BCG treatment has highlighted the role of NK cells and has led to the hypothesis that cytotoxic effector cells are probably of a different nature [66]. Thus, lymphocytes, which are specifically activated by mycobacteria and have cytolytic activity against tumor cells in vitro, could play this role [67]. These lymphocytes do not usually express the CD4+ or CD8+ phenotype and their ability to recognize the antigen is not restricted by MHC. This cytolytic activity seems to be reactivated in a second contact with BCG and may be involved in the quality of response to BCG by a memory-associated phenomenon. In addition, these cells have the ability to stimulate other lymphocyte populations (CD4+, CD8+) in response to antigenic stimulation [68]. Some authors have shown that BCG was able to induce the maturation of DCs from circulating mononuclear cells and modulate the expression of the CD40 molecule on the cell surface of urothelial tumors [69]. The expression of CD40 participates in the activation of T helper cells and sensitizes tumor cells to apoptosis via mechanisms involving the Fas-L/Fas system; over-expression of CD40 and activation of the CD40/CD40-L system may also contribute to the activation of B lymphocytes and NK cells.

20.3.8 The Role of Tumors

There are probably resistance mechanisms developed by the tumor, allowing it to escape immune host surveillance, and also treatment with BCG. In a functional immune system, Cytotoxic T lymphocytes (NK, CTL, CD8 LAK, CD4) are capable of inducing tumor apoptosis through the perforin and granzyme A or B systems via the Fas-L/Fas system. These systems are sometimes nonfunctional. Thus, Cytotoxic T lymphocytes may have perforin, granzyme A or B, and Fas-L/Fas system deficiency and may not be active against the tumor [70]. The tumor can also escape from this system by reducing the co-stimulation molecules (B7) or accession (ICAM-1) on its surface molecules required for tumor antigen presentation to Cytotoxic T lymphocytes. It can also escape the lower antigen MHC class I system at its surface (abnormal transport proteins TAP-1), depriving cells of their cytotoxic potency. The loss of normal function of P53 involved in cell apoptosis and DNA damage repair [71] can also prevent the natural phenomena of apoptosis initiated by Cytotoxic T lymphocytes and interfere with the activity of BCG. Nonetheless, the tumor may also attack the immune system. The production of immunosuppressive cytokines such as IL-10 or TGF-β1 or molecules such as Fas ligand, which are capable of inducing apoptosis of activated T cells could promote tumor growth [70]. This is a better understanding of the principles and mechanisms involved in the antitumor response of BCG, which can help to guide the clinician toward local immunotherapy.

20.3.9 Scheme of Optimal Therapy

The optimal treatment BCG scheme remains to be defined. Indeed, many protocols have been proposed (six instillations, eight instillations, 6 + 6, and finally the use of maintenance therapy). There are more variations owing to the additional routes of administration (intradermal) or related to the different BCG strains used. Initial studies seeking to validate the interest in maintenance therapy did not show any significant difference. They were involved low numbers or had a period of poor monitoring. The results reported by Dr Lamm have reignited the debate on maintenance therapy and confirm that treatment with six weekly instillations is not the optimal scheme. The use of maintenance therapy with three additional weekly instillations added at 3, 6, 12, 18, 24, 30, and 36 months may improve the outcome in terms of survival without recurrence. This effect is mainly discernible in the mean time until recurrence (36 months without maintenance therapy and 77 months with maintenance therapy) [72].

20.3.10 Predictors of the Outcome of Nonmuscle Invasive Bladder Cancer

Nonmuscle invasive bladder cancer carries a high risk of recurrence and a 10–30 % disease progression rate. Multiplicity, tumor size, and prior recurrence rates are the most important predictors of recurrence, while tumor grade, stage, and CIS are the most important predictors of progression. Although BCG is currently regarded as the most effective treatment available for the management of NMIBCs, up to 40 % of patients do not respond to treatment and are at risk of disease progression. Unfortunately, predictive markers for recurrence and progression are lacking. Prediction of recurrence or progression would be of great clinical benefit. It is the combination of clinical, pathological, molecular, and immunological markers that will allow us to more accurately predict the risk of BCG failure before commencing treatment.

20.3.11 The Clinicopathological Factors Predicting Recurrence and Progression

Although tumor grade and stage are the most accurate prognostic factors in the evaluation of NMIBCs, they cannot always predict the true biological potential of the tumor, as superficial tumors of the same stage and grade may have completely different clinical courses. The European Organisation for Research and Treatment of Cancer (EORTC) scoring system and the risk tables were adopted by the European Association of Urology (EAU) guidelines to better stratify the patients at risk of recurrence and progression and to aid future treatment options by using factors that can be easily applied clinically. The risk calculator is available at the EORTC website at www.eortc.be/tools/bladdercalculator. The EORTC scoring system gives scores to factors such as the number of tumors, tumor size, the prior recurrence rate, stage, the presence of CIS, and grade, and then totals the scores. Efforts have been made to identify other potential prognostic markers that may better stratify and identify the true malignant potential of bladder cancer.

20.3.12 Grade

Although high tumor grade has always been associated with worse outcome after BCG immunotherapy, in most reports this factor did not correlate with the time to tumor recurrence or progression in either univariate or multivariate analysis [73, 74]. On the other hand, Ajili et al. [75] failed to find a significant association between grade and response to BCG immunotherapy.

20.3.13 Stage

Several studies, both univariate and multivariate analyses, have shown that high tumor stage is associated with a poor BCG immunotherapy response. Some authors showed a correlation with time to recurrence [76]. Other studies showed a correlation with time to tumor progression (TTP) [77]. However, other large studies failed to find any association between tumor stage and the BCG immunotherapy response [73, 78]. Ajili et al. [79] showed that high tumor stage is significantly associated with a worse BCG immunotherapy response in univariate analysis (p = 0.009). In addition, Kaplan–Meier survival curves show reduced relapse-free survival (RFS) for patients with a pT1 tumor (log-rank test p = 0.004).

20.3.14 Multiplicity

Multiplicity is a classical prognostic factor for the recurrence of NMIBCs, but its predictive value for BCG immunotherapy response is controversial. Some studies, including Ajili et al. [75, 80, 81], have suggested that multiplicity might be an independent factor for recurrence after BCG treatment. However, most studies that showed no correlation were probably underpowered [82].

20.3.15 Gender

Several studies with larger cohorts have suggested an association between male gender and a more favorable response to BCG immunotherapy, but the gender difference was never statistically significant [81]. However, Fernandez-Gomez et al. [83] found an association between gender and BCG response in multivariate analysis, in such a way that male patients had a significantly longer time before recurrence than female patients.

20.3.16 Age

Although age has been the patient characteristic most frequently associated with BCG immunotherapy response [84], several studies have not shown any influence of age [80, 85]. Joudi et al. [86] reported that aging appears to be associated with a decreased response to BCG immunotherapy and is particularly apparent in patients older than 80 years. Cho et al. [87] have maintained that younger patients appear to have a more favorable prognosis.

20.4 Markers Predicting Response to BCG

20.4.1 Cell Cycle Regulators

20.4.1.1 P53 Protein

The most frequent molecular events in human NMIBC are mutations of the p53 gene. The p53 protein plays a vital role in the regulation of the cell cycle and is important for genetic stability, cell proliferation, apoptosis, and angiogenesis [88]. A defect in p53 leads to the loss of p53-dependent apoptosis and gives a proliferative advantage. Some studies found that altered p53 gradually increased from normal urothelium to NMIBC to CIS to MIBC [89, 90]. However, there are some contradictory reports regarding the prognostic value of p53 in bladder cancer [91, 92]. Moreover, p53 expression was significantly associated with tumor stage, grade, and disease recurrence.

20.4.1.2 P27 Protein

The cyclin-dependent kinase (Cdk) inhibitor p27 (also known as KIP1) regulates cell proliferation, cell motility, and apoptosis. In cancers, p27 is inactivated through impaired synthesis, accelerated degradation, and by mislocalization. Moreover, studies on several tumor types have indicated that p27 expression level has both prognostic and therapeutic implications. In patients with NMIBCs, p27 has limited predictive value [93, 94].

20.4.1.3 The Retinoblastoma

The retinoblastoma protein (pRB) is a tumor suppressor involved in cell cycle control. The predictive power of pRB may be inferior to other cell cycle regulators in NMIBCs [94]. Esuvaranathan et al. [95] found that the low expression of pRB in patients treated with BCG and IFN-α is associated with a high recurrence rate. In a homogeneous population of T1G3 patients, Cormio et al. [96] demonstrated that an altered RB was associated with decreased progression-free and disease-free survival. Overall, altered RB expression may serve as a predictive maker of BCG treatment outcome.

20.4.2 Apoptotic Markers

Apoptosis is the distinctive form of programmed cell death that complements cell proliferation in maintaining normal tissue homeostasis. Several proteins are involved in the regulation of apoptosis and their abnormal expression is associated with an altered balance between cell growth and cell death. The significance of constitutive apoptosis in the recurrence of NMIBCs has yet to be investigated. So far, little attention has been paid to the potential role of apoptotic protein expression in superficial bladder tumors treated by BCG immunotherapy.

20.4.2.1 Bcl-2 Protein

Bcl-2 is an apoptotic marker that controls ion channels, caspase status, and cytochrome c location. Bcl-2, caspase-3, p53, and survivin have a cooperative effect on the progression of bladder cancer. In published urothelial carcinoma data, the expression of Bcl-2 varies considerably. Its incidence ranges from 69 % to less than 2 % in some studies, including muscle-invasive tumors [97]. Taking into consideration the association between the over-expression of this anti-apoptotic protein and the tumor characteristics, including stage and grade, the reported data are contradictory [25, 26, 30]. Gonzalez-Campora et al. [98] found that Bcl-2 over-expression was associated with low-grade NMIBCs, while none of the high-grade superficial tumors expressed Bcl2. On the other hand, in multivariate Cox regression analysis, Bcl-2 was found to be an independent factor for recurrence [99].

20.4.2.2 Bax Protein

Bax protein is known to play a pro-apoptotic role. Its expression varies considerably in human tumors and the significance of its role in tumor progression and outcome remains generally unknown. In nonmuscle invasive bladder cancer, it was shown that Bax over-expression was an independent predictor of overall survival [98]. No data are available on the clinical outcome of NMIBCs with regard to the Bax status in tumor urothelial cells treated by local immunotherapy. In univariate analysis, Ajili et al. [99] showed that decreased or absent Bax expression was associated with low-grade tumors and a favorable outcome after treatment. To select the best predictors of recurrence among all the aforementioned variables, multivariate Cox regression analysis was performed. It showed that decreased or absent Bax expression was a significantly favorable independent factor for response to BCG therapy.

20.4.2.3 Survivin

Survivin is an important apoptotic marker. Some authors found that survivin expression analysis might identify patients with NMIBC at high risk of disease recurrence and progression [100]. Moreover, survivin over-expression increased gradually from NMIBC to advanced bladder cancer to metastatic lymph node tissue [100].

20.4.3 Angiogenesis Markers

Angiogenesis is the formation of new capillaries from the existing vascular network and is essential for tumor growth. This process is tightly regulated by angiogenic stimulators, such as fibroblast growth factor, and some angiogenic inhibitors. In various tumor types, angiogenesis is a prognostic factor determining the biological behavior. Indeed, various mechanisms are involved in the angiogenic process, with convergence of these signals permitting transduction and subsequent activation of pathways that promote tumor proliferation, migration, invasion, and ultimately, survival and metastasis.

Several angiogenic markers, including thrombospondin-1 (TSP-1), CD34, and vascular endothelial growth factor (VEGF), are currently thought to be of clinical importance for bladder cancer.

20.4.3.1 Vascular Endothelial Growth Factor

Vascular endothelial growth factor promotes endothelial mitogenesis and migration, extracellular matrix remodeling, increased vascular permeability, and the maintenance of newly formed vasculature. Higher VEGF expression was associated with increasing tumor stage, grade, progression, and recurrence in patients treated with TUR [101]. These findings support the role of VEGF in bladder tumorigenesis and further support it as a potential target for therapy [102].

20.4.3.2 Thrombospondin-1

Thrombospondin-1 is a potent inhibitor of angiogenesis that is independently associated with disease recurrence [101, 102]. Grossfeld et al. [103] previously reported that tumors with p53 alterations are associated with low TSP-1 expression, and that these tumors are more likely to demonstrate high microvessel density (MVD) counts.

20.4.3.3 Platelet/Endothelial Adhesion Molecule

Platelet/endothelial adhesion molecule, also known as CD34, is an endothelial antigen that has been used to highlight the density of intra-tumorous vessels as a direct marker of the degree of MVD of neoangiogenesis. It has been investigated in many other malignancies and is thought to be an important prognostic factor in some locations; however, there are only a few studies dealing with it in urothelial carcinoma. In bladder urothelial carcinomas, most reports are related to muscle-invasive tumors. Indeed, MVD has been extensively investigated in these tumors as a prognostic tool and has been associated with a poor outcome. Ajili et al. [79] showed that MVD was a significantly unfavorable independent factor for response to BCG immunotherapy. Indeed, recurrence was lowest for those with the lowest MVD count and highest for those with the highest MVD. MVD, a surrogate marker for angiogenesis, has been demonstrated to be a prognostic marker associated with the highest risk of recurrence and failure of BCG immunotherapy in patients with NMIBCs.

20.4.4 Inflammatory Markers

20.4.4.1 Dendritic Cells

Dendritic cells have been suggested to play an important role in the response to mycobacteria. Indeed, BCG provokes inflammation involving the contribution of cells associated with the innate immune response. A high number of urinary DCs seem to have a positive effect on the outcome of BCG treatment [104]. On the other hand, the risk of recurrence decreases with high RNA expression of antigen-presenting molecules in normal urothelial cells [105]. In patients with a weak initial immune response, determined by low levels of CD83+ tumor-infiltrating DCs, maintenance BCG proved to be highly effective by activating immune cells [106].

20.4.4.2 Tumor-Infiltrating Macrophages

Extensive infiltration of tumor-infiltrating macrophages has been reported to correlate with a good prognosis in various types of cancer [107] and in nonmuscle invasive bladder cancer. High numbers of TAMs seem to play a negative role in BCG response [106]. Ajili et al. [108] showed that the increased TAM was a significantly unfavorable, independent factor of response to BCG immunotherapy. Indeed, patients with a high TAM count showed a higher risk of recurrence than those with a low TAM count. These data suggest that determination of TAM count might be of value for predicting clinical outcome or prognosis, and that patients with NMIBCs expressing a high TAM count might be less sensitive to BCG immunotherapy. On the other hand, other studies have reported that patients with a high level of infiltration by CD68+ TAMs do not respond to BCG immunotherapy either [106].

20.5 Immunotherapy of Renal Cancer

Immunotherapy has provided the basis for experimental strategies that have introduced new indications for cytoreductive surgery in patients with metastatic renal cell cancer.

20.5.1 General Treatments for Metastatic Renal Cancer

Considering the low efficiency of even aggressive, local treatments, improvement in the prognosis of metastatic renal cancer is inevitably highlighted by the development of treatments with a general aim.

20.5.1.1 Hormonotherapy

The therapeutic option of hormonotherapy is based on experimental findings obtained by using progesterone to block kidney tumors induced in hamsters. Apart from a few isolated results [109], hormonal treatment using progesterone has been proven ineffective as adjunctive therapy for metastatic kidney cancer [110].

20.5.1.2 Chemotherapy

Up to now, no cytotoxic chemotherapy has revealed regular efficiency, either as mono- or as chemotherapy [111]. Current efforts tend to reduce the toxicity of chemotherapies, as well as to increase their effectiveness. The issuance of chemotherapy through implantable pumps may be programmed to limit toxic effects [112]. Based on the principles of chronobiology, this method has been used in the treatment of metastasized kidney cancer with encouraging results: 7.1 % complete response and 12.5 % partial response in a series of 56 patients, confirming the efficacy of treatment with continuous systemic infusion of floxuridine (FUDR), with few side effects.

20.5.1.3 Traditional Immunotherapy

The importance of the relationship between the host and the tumor has long been emphasized in kidney cancer, in particular because of evidence of the spontaneous regression of metastases.

These relationships play prominent roles leading to the development of therapeutic immunotherapy. Traditional immunotherapy is active in that it exogenously stimulates the host immune system. This type of immunotherapy may be specific, directed against a tumor-specific antigen, or nonspecific.

To increase nonspecific host immunity, several molecules have been used on metastasized kidney cancer. The main, active, nonspecific immunotherapy trials have mostly used BCG [113]. Other molecules that have immunomodulatory action were used with polyinosinic-polycytidylic acid (poly I:C), which increases the cytotoxicity of lymphocytes by inducing the production of interferon [10], or with coumarin, which affects mitogen on lymphocytes, and cimetidine, which has an inhibitory effect on T lymphocyte suppressor [114]. The results of this active immunotherapy do not vary depending on the specific molecules used: little effect for some (poly I:C) or few answers for others (BCG).

Good results have been obtained with the coumadin–cimetidine combination, with 3 complete responses and 11 partial responses in a series of 42 patients (33 % overall response), but the results obtained in this preliminary study have not been confirmed by other teams, and seem to have no impact on survival, as in all cases, this tumor regression is transient [114].

Specific active immunotherapy seems very effective. However, conflicting results were obtained. A significant gain in survival was shown in a group of 71 patients who had nephrectomy followed by monitored tumor vaccinations associated with particular dietary supplementation [115], but these results have not been confirmed, and other studies have shown inefficiency in terms of immunotherapy-specific active survival.

Among them, in a study involving 33 patients receiving tumor vaccines in the form of injections of irradiated tumor cells, either autologous or not, no difference in survival was shown between the two groups of patients, i.e., responders (24 % partial response, no response complete), and nonresponders [116]. Thus, whether specific or nonspecific, traditional immunotherapy can be considered an effective treatment for metastasized kidney cancer.

20.5.1.4 New Developments in Immunotherapy

New approaches to the immunotherapy of metastasized kidney cancer apply cytokines, which are molecules involved in the regulation of the immune system. They may be in the form of monokines (such as TNF or IL-1), or lymphokines (ITN and IL-2). These cytokines can be used alone, or in synergistic associations with certain chemotherapeutics: vinblastine [117], 5-fluorouracil and mitomycin [118], or VP16 [119].

Adoptive immunotherapy uses the lymphocytes activated by the patient in in vitro IL-2. It may be circulating lymphocytes (lymphokine activated killer) or intra-tumor lymphocytes (tumor infiltrating lymphocyte).

20.5.2 Prognostic Systems in Metastatic Kidney Cancer

Different prognostic models developed as predictive models are mainly based on response to immunotherapy. One model that has proved its usefulness is that of the French Immunotherapy Group. It is schematically based on circulating neutrophils, the interval between the initial diagnosis and the onset of metastases, the presence of liver metastases, and the number of metastatic sites. The main utility of this model was to classify the first patients for clinical immunotherapy trials and it has clearly. It demonstrated that patients with poor or intermediate prognosis do not benefit from receiving IFN and IL-2. Objectively, this model is mostly considered as a tool for evaluating reply to immunotherapy, rather than a prognostic model.

20.6 Concluding Remarks

Significant advances have been made in introducing novel immunotherapeutic approaches, but future studies are warranted.