Bacillus Calmette-Guérin immunotherapy for genitourinary cancer

Authors


Correspondence: Nilay M. Gandhi, James Buchanan Brady Urological Institute, John Hopkins Medical Institutions, Baltimore, MD 21201, USA.

e-mail: nilaymgandhi@gmail.com

Abstract

What's known on the subject? and What does the study add?

  • The administration of Bacillus Calmette-Guérin (BCG) immunotherapy has become the standard of care for high-grade non-muscle invasive bladder cancer (NMIBC) and carcinoma in-situ (CIS) in terms of prevention of recurrence and progression. While most agree on a 6 week induction cycle, various maintenance schedules (if any at all) have been implemented without a unifying consensus.
  • This review assesses the historical emergence of BCG immunotherapy, beginning with its discovery as a vaccinatin for tuberculosis to its effect on the host immune system and potential therapeutic benefits for various oncologic conditions. The data establishing BCG immunotherapy as the standard of care for high-grade NMIBC and CIS over other bladder instillation modalities is presented in addition to the effect maintenance BCG therapy has on sustaining the immuno-protective effect.
  • Bacillus Calmette-Guérin (BCG) immunotherapy is currently the most effective treatment of non-muscle invasive bladder cancer and one of the most successful applications of immunotherapy to the treatment of cancer.
  • This review summarises the history and development of BCG as a modern cancer treatment, appraises current optimal application of BCG immunotherapy in bladder cancer, discusses promising new therapies closely related to BCG, and briefly explores the possibility that BCG or related treatments may have an application in other urological malignancies.
  • BCG is a nonspecific stimulant to the reticuloendothelial system and induces a local inflammatory response with the infiltration of granulocytes followed by macrophages and lymphocytes, particularly helper T cells.
  • The initial BCG controlled trial showed a statistically significant reduction in tumour recurrence and found the advantage increased with duration of follow-up. Similar results were reported in much higher risk patients in an independent concurrent study. Follow-up suggested that a single 6-week course of intravesical BCG provided long-term protection (up to 10 years) from tumour recurrence and even reduced disease progression.
  • While induction BCG (six weekly instillations) reduced recurrence, progression and mortality at 10 years, this advantage was lost by 15 years, and patients remained at high risk for progression without the use of maintenance BCG.
  • In a meta-analysis by the Cochrane group, induction BCG was found to be markedly superior to mitomycin C in high-risk patients but not in low-risk patients. Additionally, the National Comprehensive Cancer Network guidelines lists the use of intravesical BCG as preferred therapy, citing Category 1 data for high-grade Ta, all T1, and any Tis tumours.
  • Maintenance BCG therapy may be the most important advance in BCG treatment of bladder cancer since the initial introduction. The risk of tumour recurrence and disease progression is life-long in most patients, but the immune stimulation induced by BCG wanes with time.
  • Logarithmic dose reduction of BCG in patients with increasing side-effects will typically prevent escalation of toxicity.
  • Simple dose reduction, appropriate antibiotics, and understanding treatment contraindications have greatly increased the safety of BCG. The 3-week maintenance schedule for 3 years has been evaluated in randomised clinical trials and appears to be the current optimal treatment.
  • With the success achieved in bladder cancer and the relative safety and economy of BCG, consideration should be given to further research for its effectiveness in other genitourinary malignancies.
Abbreviations
CIS

carcinoma in situ

IFN

interferon

IL

interleukin

MMC

mitomycin C

MSK

Memorial Sloan-Kettering Cancer Center

NMIBC

non-muscle invasive bladder cancer

SWOG

Southwest Oncology Group

TRAIL

TNF-related apoptosis-inducing ligand

Introduction

BCG has been steeped in controversy since it was first introduced in 1921 as a vaccine against the scourge of tuberculosis. Touted as a promising new cancer treatment in the 1970s, it subsequently was discarded as being ineffective once controlled trials in advanced bladder disease failed to confirm its benefit [1]. Taking advantage of animal and immunological studies that defined optimal conditions for BCG response, urologists have developed BCG into the most effective therapy currently available in the treatment of aggressive non-muscle invasive bladder cancer (NMIBC). This review summarises the history and development of BCG as a modern cancer treatment, appraises current optimal application of BCG immunotherapy in bladder cancer, discusses promising new therapies closely related to BCG, and briefly explores the possibility that BCG or related treatments may have an application in other urological malignancies.

Urologists are very familiar with BCG, but many in other specialties are unaware of the advantages it provides in the treatment of bladder cancer. Indeed, BCG remains largely ignored while immunotherapies using newer molecular or immunological techniques that have response rates far lower than BCG are enthusiastically endorsed. The story of the development of BCG as a vaccine against tuberculosis and its subsequent transformation into a cancer treatment provides important lessons in perseverance, observation and application of the scientific method.

Tuberculosis

It is estimated that the most adults in the 19th century were infected with tuberculosis and the mortality was as high as 25% [2]. About one in seven deaths during that period were attributed to tuberculosis. As evidenced by the existence of a BCG response gene, the prevalence and mortality of tuberculosis make it highly probable that the infecting organism has played a significant role in natural selection [3]. This in fact underscores the expectation of BCG as a potent immune stimulant. Inspired by Pasteur's successes in obtaining protective vaccines with attenuated microorganisms and by Koch's demonstration in 1884 that Mycobacterium Tuberculosis was the causative organism in tuberculosis, scientists of the day raced to develop an effective vaccine. Nocard isolated Mycobacterium bovis from the milk of an infected heifer. This strain was transferred to Albert Calmette and Camille Guérin in 1904 while working in the Pasteur Institute in Lille, where tuberculosis mortality was >43%. The initial strain was extremely virulent and was subsequently diminished by successive passages over a bile-potato medium every 3 months. Between 1908 and 1921, the strain underwent 231 passages and demonstrated avirulence without reversion [4]. This finding was presented at the Academie des Sciences as a new ‘race of biliated tuberculosis bacilli’ and named ‘bacilli tuberculeux bilie’, later simplified to bacillus Calmette-Guérin [5].

In July 1921, the first BCG vaccination was given as an oral suspension (2 mg orally at the third, fifth, and seventh day of life) to a newborn girl whose mother died with tuberculosis at birth [6]. Close follow-up of the first 30 vaccinated babies demonstrated safety and protection from tuberculosis, therefore rapid spread of the vaccination programme began over France and the rest of Europe. A presentation at the Conference of the League of Nations in 1928 showed a decrease in mortality to 1.8% in those vaccinated compared with 25–36.2% in the non-vaccinated in Paris [7]. Vaccination proceeded uneventfully until 1930 when BCG was implicated in the deaths of 70 infants in Lübeck, Germany. The true cause was later determined to be contamination of the locally prepared vaccine with live Mycobacterium Tuberculosis grown in the same laboratory, but Calmette, who died in 1933, was never to recover from the incident [8]. Fear of BCG persisted and it took several decades to rekindle interest in the vaccine as a possible anti-neoplastic agent.

BCG and Cancer

William Coley in the 1890s had successfully used bacterial products (‘Coley's Toxins’) to treat malignancy [9], and the concept that immune stimulation from infectious disease reduces the risk of cancer has been confirmed as recently as 1998 [10]. The notion that BCG might have a role in cancer treatment dates back to 1929, when Pearl [11] correlated a reduced incidence of cancer among patients with tuberculosis at autopsy. BCG was first used to treat stomach cancer by Holmgren [12] in 1935 when successes were reported in a series of 28 patients. In 1936, Rosenthal [13] described the profound stimulation BCG produces on the reticuloendothelial system, which at the time was recognised to be important in the defence against cancer. However, due to the widely publicised events in Lübeck, enthusiasm for the use of BCG as cancer therapy was quelled. Landmark studies in the 1950s led by Old and Clarke [14] were able to show, in animal models, the inhibition of cancer by BCG. Neonates immunised with BCG were found to have a significantly lower incidence of leukaemia [15]. Multiple uncontrolled clinical studies in the 1970s claimed that BCG was effective in the treatment of cancer. In melanoma, Morton [16] reported a 92% regression with intralesional injection of BCG into lesions limited to the skin. Unfortunately, controlled medical oncology trials in the 1970s failed to show significant benefit of BCG, although most of these studies were in patients with advanced disease who would not be expected to respond to BCG. In addition, advances in radiotherapy and chemotherapy led to disinterest in the use of BCG for cancer therapy in all realms except one.

BCG in Bladder Cancer

It was with this background that studies of BCG in bladder cancer began. Studies by Coe and Feldman [17] in 1966 showed that the bladder responded to BCG with a delayed type hypersensitivity reaction seen with the skin, and in 1974, Silverstein et al. [18] reported response of melanoma metastatic to the bladder treated with intralesional BCG. This success in melanoma led, at the suggestion of David McCullough, to the evaluation of BCG in an animal model of bladder cancer [19]. At the same time, Morales independently began clinical studies and in 1976, published results of the first successful clinical trial of intravesical BCG administration in nine patients with recurrent bladder cancer.

The original BCG protocol was initially conceived by Morales in 1972 [20]. A Canadian urologist, Morales obtained the BCG strain manufactured by the Montreal Institute Armand. Morales knew that, to establish a delayed hypersensitivity reaction, a minimum of 3 weeks of treatment was necessary and would be gauged by an intradermal BCG injection. Additionally, he had determined that adverse effects resolved within 1 week, prompting a weekly schedule. The Armand Frappier vaccine was packaged as 120 mg each dispensed in six separate vials, therefore establishing a six weekly bladder instillation protocol. The results of that study found six weekly intravesical plus percutaneous administrations to result in a 12-fold reduction in bladder tumour recurrence [21].

This success prompted the National Cancer Institute to request proposals for controlled human clinical trials using Morales' technique and contracts were given to Lamm at the University of Texas in San Antonio and Pinsky at Memorial Sloan-Kettering Cancer Center (MSK) in New York. The initial BCG controlled trial, published in 1980, showed statistically significant reduction in tumour recurrence in 54 evaluable patients [22] and showed an advantage that increased with time [23]. Similar results were reported in much higher risk patients in the MSK study [24]. Subsequent follow-up of these patients suggested that the benefit of a single 6-week course of intravesical plus percutaneous BCG provided long-term protection from tumour recurrence and even reduced disease progression. However, the 15-year follow-up of the MSK series showed the limitations of BCG protection without initiation of maintenance therapy. While induction BCG (six weekly instillations) reduced recurrence, progression and mortality at 10 years, this advantage was lost by 15 years, and patients remained at high risk for progression and the development of disease in the prostatic urethra (24%) and upper tracts (25%). A mortality rate of 32% was seen in patients with disease progression into the upper tracts and in 44% of those with prostatic urothelial disease [25].

BCG Immunotherapy Technique in Bladder Cancer

Immune Response to BCG

It is often said that the true mechanism of action for BCG is unknown. In fact, a great deal is actually known about the immune response to BCG. As a complex living organism, the responses induced to BCG infection are broad and highly varied. BCG attaches to bladder tumour cells as well as urothelial cells by means of specific fibronectin and integrin receptors [26, 27]. Internalization of BCG is correlated with immune response and sensitivity to BCG. In vitro studies suggest that poorly differentiated cell lines internalise BCG and are sensitive to it [28]. Clinical studies similarly suggest that low-grade tumours are relatively less sensitive to BCG [29]. Antigens specific for BCG are expressed on the surface of tumour cells and MHC class II antigen expression is upregulated [30-32].

BCG is a nonspecific stimulant to the reticuloendothelial system and induces a local inflammatory response with the infiltration of granulocytes followed by macrophages and lymphocytes, particularly helper T cells. Phagocytosis and the ratio of helper/suppressor cells are increased [33]. A wide range of cytokines are induced, including interleukin (IL)-1, IL-2, IL-6, IL-8, IL-10, IL-12, TNF-α, interferon (IFN) γ, granulocyte/macrophage colony stimulating factor and soluble intercellular adhesion molecule I [34]. The cellular immune (Th1 response) predominates and is correlated with tumour destruction.

Recent studies show that BCG, like other mycobacteria, contains high amounts of CpG oligodeoxynucleotide motifs, known to induce TNF-related apoptosis-inducing ligand (TRAIL) by IFN production. After BCG administration, the cytokine response induces TRAIL up-regulation. Urine IFN and TRAIL levels were initially undetectable in BCG therapy patients but subsequent treatments led to identification of TRAIL and TRAIL-expressing neutrophils in the urine and this expression was correlated with response to BCG therapy [35].

BCG in Bladder Carcinoma in situ (CIS)

Animal studies have shown that effective BCG immunotherapy requires an immune competent host, an adequate dose of BCG, juxtaposition of BCG and tumour cells, and a limited number of cancer cells. These conditions are optimally met in bladder cancer, especially CIS. In 1980, Morales [36] reported a 71% complete response rate in patients with CIS of the bladder using a 6-week BCG induction. BCG (Theracys) was initially approved by the USA Food and Drug Administration for bladder cancer in 1990 based upon its demonstrated superiority (71% complete response, 48% disease free at 5 years) over doxorubicin (47% complete response, 18% disease free at 5 years) in the treatment of CIS [37]. The consistency of the efficacy of BCG in CIS is noteworthy, regardless of strain utilised (Table 1). Complete responses were seen with every strain of BCG despite variations in dose and schedule, and the 70% response rate is remarkably consistent [38]. One controlled study provides important clinical information in the management of CIS. The Southwest Oncology Group (SWOG) study of 6-week induction followed by 3-week maintenance BCG vs standard 6-week induction without maintenance included 269 randomised patients with CIS. An increase in complete response (58% at 3 months to 68% at 6 months) was seen in the induction-only arm, suggesting that while residual CIS is a poor prognostic feature, it may take up to 6 months for the full complete response to occur. With three additional weekly BCG maintenance instillations beginning at 3 months, 64% of those with residual CIS at 3 months had resolution at 6 months, yielding an overall complete response rate of 84% [39]. With continued 3-week maintenance BCG after 3 months, >70% of patients with an initial complete response remained tumour-free for >5 years; a remarkable achievement considering that series historically reported progression in 54% of patients with CIS within that same timeframe [40].

Table 1. Comparison of BCG strains in the treatment of CIS of the bladder [38].
Strain/seriesTotal number of patientsComplete responses, nComplete response rate, %Range of response rates, %
Connaught/84503577970–92
Tokyo/4111867763–84
Pasteur/72301717440–80
Tice/62771977156–82
Evans/61801176553–88
A. Frappier/6145876039–100
S. African/113969
Danish/1422867
Romanian/1332164
RIVM/115960
Total149610827239–100

Maintenance BCG, Disease Progression, and Mortality

It is generally accepted that induction BCG is highly effective in reducing disease recurrence, but controversy persisted about the role of maintenance BCG and its ability to reduce disease progression and mortality. Lamm [41] initially reported a four-fold reduction in the rate of tumour recurrence in patients treated with quarterly single BCG instillations. A controlled evaluation in 1987 of this regimen in 42 patients failed to show a significant reduction in tumour recurrence [42]. In the same year, Badalament et al. [43] reported a randomised study of 93 patients comparing 6 week induction with monthly maintenance BCG and found no significant benefit for maintenance therapy. These studies could be criticised for their relatively short follow-up and lack of power, but nevertheless, they had a major impact on clinical practice. As it stands today, the benefit of maintenance BCG and the optimal schedule continue to be questioned [44]. Many urologists have adopted an initial 6-week BCG induction course followed by a repeat 6-week cycle at the time of tumour recurrence, despite absence of randomised clinical trials supporting this approach. Palou et al. [45] found that 6-week maintenance BCG every 6 months for 2 years was not significantly better than induction alone in a study of 126 patients with CIS followed for an average of 79 months.

Despite these ‘negative’ studies, maintenance therapy may be the most important advancement in BCG treatment of bladder cancer since the introduction of Morales' original 6-week induction regimen. Maintenance immunotherapy is vital, all while balancing the administration of excessive BCG to prevent causing immunosuppression. The risk of tumour recurrence and disease progression is life-long in most patients, but the immune stimulation induced by BCG wanes with time. A controlled trial of 384 patients using an immunologically sound maintenance schedule (intravesical and percutaneous BCG each week for 3 weeks given at 3, 6, 12, 18, 24, 30, and 36 months) clearly showed that 3-week maintenance BCG markedly reduced long-term recurrence and disease worsening [39]. While other, presumably inferior, maintenance schedules have not achieved statistical significance in randomised controlled trials, a meta-analyses showed that induction + maintenance BCG is superior to induction BCG alone and may be required for the most important benefit of BCG – the reduction of tumour progression [46].

A 6-week induction course induces a prolonged protection from tumour recurrence and may help explain the failure to show a benefit of maintenance studies with short follow-up. In animal studies, no benefit of BCG maintenance was seen at 9 months, but a highly significant benefit was documented at 15 months [47]. In Herr's [25] series, a significant reduction in tumour recurrence, disease progression, and mortality persisted for 10 years, but was no longer significant at 15 years. This is in agreement with the initial finding of 10 year conferred protection from the MSK series [24]. The SWOG maintenance study of 384 evaluable patients (up to three weekly instillations at 3, 6, 12, 18, 24, 30 and 36 months) found a highly significant reduction in tumour recurrence (P < 0.001), as well as a significant reduction in disease worsening (P < 0.04), defined as stage progression or the requirement for cystectomy, radiation therapy, or systemic chemotherapy [39].

In the meta-analysis by Bohle et al. [48], all six comparison studies using maintenance BCG schedules found BCG to be clearly superior to mitomycin C (MMC), while only one study reported statistical significance in groups not using maintenance BCG. However, the most convincing data is in the meta-analysis published by Sylvester et al. [46]. In that analysis of 24 studies comprised of 4863 patients, BCG was compared with surgery, chemotherapy, or other immunotherapy with endpoints of disease progression and cancer-specific mortality being investigated. Disease progression was found to be significantly reduced by 37% (P < 0.001); however, this benefit was only seen in patients receiving maintenance BCG. Bladder cancer mortality was reduced by 19%, but statistical significance was not achieved (P = 0.20) due to the short median follow-up of only 2.5 months. Malmström et al. [49] analysed nine randomised trials comparing the long-term efficacy of intravesical MMC against BCG (induction alone and with maintenance) with time to first recurrence as the primary endpoint. BCG maintenance was found to confer a 32% reduction in the risk of recurrence for BCG compared with MMC, producing a statistically significant 4% difference against trials without BCG maintenance (P < 0.001). It was determined that for prophylaxis against recurrence, BCG maintenance was required to achieve the demonstrable superiority over MMC.

Subsequent randomised clinical trials have shown that the 3-week, 3-year maintenance schedule used by the SWOG has achieved success that dwarfs that of other intravesical treatments. In a European study of 957 patients with intermediate-to-high risk urothelial carcinoma without CIS, the 3-week maintenance schedule was compared with intravesical epirubicin chemotherapy using the same schedule. The 3-week maintenance BCG regimen significantly decreased recurrence (P < 0.001), metastasis (P < 0.046), and improved overall and disease-specific survival (P < 0.023) [50]. In a subsequent study assessing 3 years of BCG maintenance against 1 year maintenance therapy while maintaining a 3-week instillation schedule, full dose 3-week maintenance for 3 years significantly reduced recurrence compared with one-third dose BCG maintenance for 1 year [51].

BCG Toxicity

Most patients tolerate BCG well, but as with most cancer treatments, serious and potentially fatal toxicity can occur [52]. Irritative bladder symptoms (frequency, dysuria, low-grade fever) typically can begin after the second or third instillation and last for 1–2 days as a result of immune stimulation and cytokine production. These symptoms may be associated with a more favourable anti-tumour response to BCG [53]. Severe reactions or infection with BCG result from i.v. absorption of the organism, most commonly from traumatic catheterisation. While BCG sometimes is given in the presence of haematuria, blood arising from a difficultly placed catheter is an absolute contraindication to instillation of BCG. Patients with continuing symptoms from earlier BCG administrations are best treated with antibiotics rather than forging ahead with more BCG. Logarithmic dose reduction of BCG in patients with increasing side-effects will typically prevent escalation of toxicity. While the use of fluoroquinolone prophylaxis has been shown to decrease the incidence of moderate-to-severe BCG-related adverse side-effects [54], there were no differences in local or systemic adverse reactions after prophylactic administration of isoniazid anti-tubercular medication [55]. Patients with BCG sepsis require steroids in addition to gram negative and anti-tubercular antibiotic therapy to reduce the severe hypersensitivity component that can otherwise prove fatal. As the risk for bacterial sepsis and possibly death is high, BCG should never be used as an immediate postoperative intravesical instillation.

BCG Comparisons with Intravesical Chemotherapy

The successful demonstration that BCG reduced tumour recurrence compared with surgery alone was followed by many controlled comparisons of BCG vs intravesical chemotherapy. As shown in Table 2, BCG was uniformly found to be better than chemotherapy with thiotepa, doxorubicin, and MMC [56]. However, comparisons with MMC have shown an inconsistent advantage for BCG. The AUA management guidelines for NMIBC, recently validated in 2010, listed immunomodulator agents (BCG, IFN) and chemotherapeutic agents (thiotepa, MMC, doxorubicin, gemcitabine) as beneficial intravesical therapies. The guidelines state BCG has become a first-line treatment for CIS, and that both BCG and MMC are effective in prophylaxis against tumour recurrence [57]. A SWOG study comparing six weekly instillations of either BCG or MMC followed by monthly instillations for 1 year showed the 3-year benefit in tumour reduction with use of BCG over MMC, prompting early termination of the study (Fig. 1) [58].

Figure 1.

Study comparison between BCG and MMC for bladder tumour recurrence [58].

Table 2. Overview of controlled trials of BCG vs chemotherapy for superficial bladder cancer [56].
 Median rate of recurrence, %No. of positive trials
BCG armChemotherapy arm
BCG vs thiotepa7423/3
BCG vs doxorubicin38642/2
BCG vs MMC30433/6

Further work has clearly shown that BCG is superior to MMC in these high-grade patients. As mentioned earlier, the meta-analysis of Bohle et al. [48] found maintenance BCG to be significantly better than MMC alone or induction BCG alone.48 In a subsequent meta-analysis by the Cochrane group, BCG was found to be markedly superior to MMC in high-risk patients but not in low-risk patients [59]. Additionally, the National Comprehensive Cancer Network guidelines lists the use of intravesical BCG as preferred therapy, citing Category 1 data for high-grade Ta, all T1, and any Tis tumors [60].

Despite comparisons of BCG and MMC in determining superiority for maintenance therapy, Di Stasi et al. [61] assessed the combination of the two treatment methods in evaluating disease-free intervals. In looking at 212 patients (105 receiving sole BCG induction + maintenance and 107 receiving MMC + BCG), those treated with both MMC (40 mg electromotive MMC once a month for 2 months) and BCG (81 mg once a month as one cycle for three cycles) as maintenance therapy showed a higher disease-free interval (difference of 48 months, P = 0.001), lower recurrence (16% difference, P = 0.001), progression (12.6% difference, P = 0.004), overall mortality (10.9% difference, P = 0.01), and disease-specific mortality (10.6% difference, P = 0.01). There were no significant differences in side-effect profiles, and the thought is the electromotive delivery of MMC allows for the destruction of malignant cells that may have escaped the immunological activity induced by BCG.

BCG in other GU Malignancies

BCG in Renal Cancer

In an open label, non-randomised study, Morales et al. [62] compared a group of 20 patients with metastatic RCC receiving intradermal BCG with 36 concurrent and historical controls from the same institution. In the historical control group, all patients (16 patients) were deceased at ≤4 years; however, 35% of the BCG group (seven of 20 patients) were alive up to 5 years after treatment. The ratio of observed to expected deaths in the BCG group was 0.59 compared with 1.6 in the controls (P < 0.01). Complete disappearance of metastasis and long-term survival was noted in two patients (10%) in the BCG group and none of the 36 control patients [63]. In a randomised study of BCG in patients undergoing nephrectomy for RCC in Russia, Mavrichev et al. [64] reported an 8 year survival of 48% in control patients (23 patients) compared with 60.5% in those receiving postoperative BCG immunotherapy (23 patients). The results of these trials appear to be as good as those observed with much more complex, expensive, and toxic therapies [65]. Unfortunately, the initial observations did not receive the attention that BCG enjoyed in bladder cancer and, perhaps because of a controlled study in the USA that was widely reported as negative but not published (Paul Lange, personal communication); further research was not pursued in this country. Metastatic RCC has proven to be most recalcitrant to treatment, but with the advent of effective kinase inhibiters, e.g. Sunitinib and Temsirolimus, further study of adjuvant BCG could prove beneficial.

BCG in Prostate Cancer

Direct injection of BCG into prostatic adenocarcinoma in men induces a granulomatous response associated with tumour necrosis [66, 67], but such approaches have been abandoned due to the risk of fatal septic reactions. Animal model studies have confirmed that BCG inhibits the growth of prostate cancer [68]. In the Dunning R3327 rat model of prostate cancer, BCG plus cryotherapy resulted in cellular and humoral immune responses that prevented growth of transplanted tumour [69]. Using mycobacterial cell wall extracts given via i.p. injection, Morales et al. [70] induced a complete remission in half of rats with peripherally transplanted Dunning R3327H prostate cancer.

Animal studies clearly show that immunotherapy with BCG can induce anti-tumour responses in prostate cancer, but what is the potential clinical application? In a surprising Phase 2 and subsequent Phase 3 clinical trial, Guinan et al. [71] reported remarkable success with what we would now consider is suboptimal BCG immunotherapy. Using percutaneous BCG to the deltoid every 4 months in 28 patients with advanced prostate cancer, BCG immunotherapy improved antibody and delayed type cutaneous hypersensitivity responses, resulting in an 8 month prolongation in survival when compared with historical controls. In a subsequent randomised controlled trial of 33 patients with advanced prostate cancer treated with hormonal therapy, survival was increased from a mean of 5.6 months in controls to 8.1 months with BCG immunotherapy [72]. This modest increase in survival might be maximized if BCG immunotherapy was improved and tumour burden minimised. While sample sizes are small, the published experience with BCG immunotherapy in prostate cancer compare quite favourably with the more complex, expensive and successful Sipuleucel-T immunotherapy for advanced prostate cancer.

Conclusions

Bladder cancer is currently estimated to have ≈75 000 new cases (including CIS) per year in the USA with an estimated yearly mortality of 20% [73]. The lifetime risk is noted to be 1 in 26 males or 1 in 87 females. Recent data show an estimated 80% 5-year cancer-specific survival. More than 70% of patients with bladder cancer are diagnosed with NMIBC, and it is in these patients for whom optimal BCG treatment can help reduce disease progression and improve overall survival. A recent survey of practice patterns revealed that a small minority of patients receive 3-week maintenance BCG, suggesting that bladder cancer management can be greatly improved [74].

BCG immunotherapy has been a major advance in the management of NMIBC. While early experience suggested that BCG was much more toxic than intravesical chemotherapy, simple dose reduction, appropriate antibiotics, and understanding treatment contraindications has greatly increased the safety of BCG. Now the most common intravesical agent used in the USA and Canada, BCG has been shown to reduce the risk of death in patients with NMIBC, but practice pattern studies show that underuse is the rule rather than the exception. The 3-week maintenance schedule for 3 years has been evaluated in randomised clinical trials and appears to be the current optimal treatment. Remarkably, there is data to suggest that BCG may have efficacy in renal and prostate cancer. With the success achieved in bladder cancer and the relative safety and economy of BCG, consideration should be given to further research for its effectiveness in other tumours.

Conflict of Interest

Donald L. Lamm is a Speaker for Endo Pharmaceuticals and Amgen, and provides research support for Endo Pharmaceuticals and Sanofi Pasteur. Alvaro Morales is a Consultant for Bioniche Life Sciences.

Ancillary