Maintenance Therapy for Superficial Bladder Cancer

January 1, 2001

Transurethral resection remains the standard for first-line treatment of transitional cell carcinoma of the bladder. This technique clearly defines the pathologic grade and is essential in determining the clinical stage of

The role of maintenance intravesicalchemotherapy or immunotherapy in possibly preventing tumor recurrence orprogression in the urinary bladder is a recurring question that has continued toelude resolution. Issues of controversy in each case appear to involve studydesigns that examine (1) whether maintenance therapy is truly effective,and (2) the choice of the best regimen. However, other, probably morefundamental, issues also need to be considered. These issues include theintrinsic biological potential of the various disease forms being treated,mechanism of action of various agents and regimens upon which rationale fortheir use may be based, and end points used to determine treatment efficacy.

Of the various trials that have assessed the efficacy ofintravesical chemotherapy or immunotherapy in preventing recurrence (andpossibly also progression), most have demonstrated at least short-term efficacywith intravesical chemotherapy but little long-term benefit for either.

In fact, several multi-institutional studies have reported anoverall 7% to 15% benefit with a number of intravesical chemotherapeutic agentsin preventing recurrence (but possibly with less efficacy in high-grade disease)and little, if any, benefit in preventing progression.

Scrutinizing Results With Intravesical BCG

Risk of progression has pertained largely to high-grade"superficial" disease (or disease of highly malignant potential). Amajority of studies have therefore focused on the use of intravesical bacillusCalmette-Guérin (BCG) because of its apparent efficacy in carcinoma in situ.Although intravesical BCG has become the standard of treatment for flatcarcinoma in situ, and more recently, for high-grade transitional cell cancerthat invades the lamina propria, closer scrutiny, particularly in longer-termstudies, indicates that the efficacy of BCG in preventing both recurrence andprogression may be less impressive than initial reports suggested.

For example, original reports described a 50% to 70% therapeuticefficacy with a single 6-week induction course of BCG in the elimination of flatcarcinoma in situ; an additional 50% of initial failures could apparently besalvaged by a second 6-week course. However, one-third of patients who failed asecond course were found to have metastatic disease at the time of surgicalexploration. Moreover, a long-term review of the same patients who wereinitially thought to be therapeutic "successes" found development ofmetastatic disease in one-third of patients and the need for cystectomy in anadditional one-third.

These results suggest several possible interpretations. For one,the variable outcomes may reflect the likelihood that a heterogeneous group oftumors comprised what was initially described as homogeneous for stage andgrade. Thus, some tumors appeared to be responsive to BCG and others did not—thelatter possibly expressing their potential for more aggressive behavior.

Alternatively, methods that were initially used to detectdisease may not have been sufficiently sensitive, or regimens that were used maynot have been sufficiently effective to maintain an initial response. Thus,patients whose tumors responded initially might have maintained their responsewith continued treatment if patients who were early failures (as expressed byprogressive disease) were excluded.

Moreover, several studies have demonstrated that certainmolecular aspects of a tumor may indicate a greater intrinsic aggressiveness.Some of these tumors may also be intrinsically more resistant to intravesicaltreatments and may require earlier cystectomy if they fail to respond tointravesical therapy both promptly and completely.

Selecting the ‘Best’ Regimen

The specific therapeutic regimen may also determine efficacy.When considering intravesical chemotherapy, for example, increased dosage,prolongation and/or repetitiveness of exposure, and precise delivery of anactive agent to chemosensitive cells would be important parameters to consider.

Several studies have begun to address these issues bycharacterizing conditions that may be necessary to optimize clinical efficacy.These conditions are likely to be quite different for biological responsemodifiers (eg, BCG, interferon) as immunotherapeutic agents. With these agents,increased dosage may actually compromise further recruitment of an effectiveimmune response and may be suppressive once an optimum dose/efficacy has beenexceeded. On the other hand, the rationale for a "booster" treatmentat the appropriate interval may effectively enhance leukokine production andthereby further therapeutic benefit.

Questions remain as to (1) why certain tumors do not respond toan induced immune response, and (2) whether this reflects characteristics of thetumor itself or the actual ability to invoke and then boost an effective immuneresponse. Recent studies purporting to demonstrate the efficacy of a booster BCGregimen—the so-called 6 + 3 protocol, as referenced in the article by Baselliand Greenberg[1,2]—do not definitively answer these questions.

Interpretation of Outcomes

In these studies, patients with superficial bladder cancer (butpredominantly with carcinoma in situ) treated with a 6-week intravesical BCGinduction course were randomized to receive no additional treatment or anadditional 3-week booster instillation course after 6 weeks. Control patientsexperienced a median recurrence-free survival of 35.7 months, compared to 76.8months for those who did receive the maintenance or booster treatment. Estimatedmedian time to progression in the untreated group was 111.5 months, whilenumbers in the maintenance group were insufficient to permit comparableestimations.

In these control and maintenance groups (comprised of patientswith carcinoma in situ and/or rapidly recurring stages Ta or T1 disease), thecomplete response rates to the initial 6-week induction course of BCG were 87%and 83%, respectively. This seems inordinately high and could imply that somepatients did not have particularly aggressive tumors, either in terms ofrecurrence or progression. This possibility is also suggested by the lengthyrecurrence-free survival (median: 35.7 months) and the nearly 10-year mediantime to progression in the nonbooster group.

Thus, the intrinsic biology of disease—whether indolent oraggressive—could have played a decisive role in outcomes. Correspondingly,this same biological potential could explain the eventual recurrence of diseaseeven in the booster-treated group, as suggested by the ultimate medianrecurrence in these patients of 76.8 months.

Also somewhat problematic in the interpretation of outcomes isthe issue of treatment compliance. A majority of patients in the maintenance(booster) group actually failed to comply with the full regimen of plannedtreatment, demonstrating only a 16% completion rate (39 of 243 patients). Thepurported advantage of maintenance therapy in prolonging disease-free recurrenceand preventing progression seems difficult to assess if the majority of patientsdid not actually receive the prescribed treatment regimen. Conceivably, lengthymaintenance therapy may not ultimately be necessary; only initial boosting maybe needed for treatment efficacy. However, patient selection might also haveplayed a role in the obtained results.

Findings from studies conducted by the Swedish-Norwegian BladderCancer Study Group, which compared maintenance mitomycin (Mutamycin) withmaintenance BCG, may be of importance in this regard.[3] Following an initialweekly treatment for 6 weeks with either agent, maintenance treatment consistedof monthly instillations for 1 year and then every 3 months the following year.No difference in progression rates was detected between the two groups(approximately 19% each)—rates of progression were highest for stage T1disease and grade 3 disease. Tumor recurrence decreased in the BCG-treatedgroup, and this translated into an improved disease-free survival.

Effect of End Point Selection

In such studies, the selection of end points for analysis may beimportant in considering the interpretation of ultimate treatment efficacy.Disease-free survival only indicates the absence of detectablerecurrence. Conceivably, this may reflect sensitivity of the methodology used todetect recurrence—rather than the ultimate efficacy of treatment.

Attempts to visualize disease endoscopically may not detectchanges in various "fields" of the urothelium if they have not yetbeen phenotypically expressed. Moreover, cells that have undergone neoplastictransformation may not be detectable by urinary cytology, either because offailure to recognize cells as cytologically malignant or failure to obtain cellsthat are sufficient for visual analysis. On the other hand, absence ofdetectable disease may be an advantage—but only if it can be safely assumedthat any undetectable abnormal cells will not express aggressive behavior whilethey remain either unrecognized or unresponsive to maintenance treatments.

BCG has become the standard in treating high-risk superficialurothelial cancer. However, observations that substantial numbers of tumors maynot fully respond to such therapy indicate the need to further delineateeffective regimens. In addition, individual neoplasms need to be characterizedin ways that may determine both their responsiveness to particular treatmentsand the likelihood of their recurrence and progression.

The role of BCG in treating low-risk disease should beconsidered with caution, since its potential toxicity may outweigh potentialbenefit. If recruitment of an immune response is indeed responsible for BCGefficacy, finding other means to recruit this response—with reduced toxicity—seemslike a reasonable path to pursue.


Taken together, numerous advances have been made in our approachto the treatment of different forms of superficial urothelial cancers. We nowhave earlier diagnoses of disease, the recognition of potentially aggressivetumor diatheses on the basis of their molecular characteristics and phenotypicexpression, and an understanding that different pathways of tumor developmentmay reflect or represent varying biologic potentials. Each of these factors hascontributed to an improved assessment of the type of tumor diathesis that may bepresent and the requirement for a particular type of treatment approach.

However, we still need a more precise means of characterizingthe biological potential of an individual tumor diathesis. We also needintravesical agents and regimens that can be employed most effectively inpreventing recurrence and progression. The accuracy of end points ininterpreting results is contingent upon our understanding of a disease, both atits initial presentation and in its ultimate capabilities. Efficacies of varioustreatments need to be studied in this context, and interpretation of outcomesneed to be based on appropriate experimental designs, as well as the acquisitionof meaningful end point data.

Suggested Readings

Catalona WJ, Hudson MA, Gillen DP, et al: Risks and benefits ofrepeated courses of intravesical bacillus Calmette-Guérin therapy forsuperficial bladder cancer. J Urol 137:220-224, 1988.

De Balincourt C, EORTC group: Relationship between efficacy andside effects of BCG instillations in TaT1 bladder cancer. J Clin Oncol6:1450-1455, 1988.

Herr HW: Tumor progression and survival of patients with highgrade, noninvasive papillary (TaG3) bladder tumors: 15-year outcome. J Urol163(1):60-61; discussion 61-62, 2000.

Herr HW, Laudone VP, Badalament RA, et al: BacillusCalmette-Guérin therapy alters the progression of superficial bladder cancer. JClin Oncol 6(9):1450-1455, 1988.

Herr HW, Schwalb DM, Zhang ZF, et al: Intravesical bacillusCalmette-Guérin therapy prevents tumor progression and death from superficialbladder cancer: Ten-year follow-up of a prospective randomized trial. J ClinOncol 13(6):1404-1408, 1995.

Jones PA, Droller MJ: Pathways of development and progression inbladder cancer: New correlations between clinical observations and molecularmechanisms. Semin Urol 11(4):177-192, 1993.

Lamm DL, Blumenstein BA, Crissman JD, et al: Maintenancebacillus Calmette-Guérin immunotherapy for recurrent TA, T1 and carcinoma insitu transitional cell carcinoma of the bladder: A randomized Southwest OncologyGroup Study. J Urol 163(4):1124-1129, 2000.

Lamm DL, Riggs DR, Traynelis CL, et al: Apparent failure ofcurrent intravesical chemotherapy prophylaxis to influence the long-term courseof superficial transitional cell carcinoma of the bladder. J Urol153(5):1444-1450, 1995.

Nadler RB, Catalona WJ, Hudson MA, et al: Durability of thetumor-free response for intravesical bacillus Calmette-Guérin therapy. J Urol152(2 Pt 1):367-373, 1994.

Pawinski A, Sylvester R, Kurth KH, et al: A combined analysis ofEuropean Organization for Research and Treatment of Cancer and Medical ResearchCouncil randomized clinical trials for the prophylactic treatment of stage Ta,T1 bladder cancer. European Organization for Research and Treatment of CancerGenitourinary Tract Cancer Cooperative Group and the Medical Research CouncilWorking Party on Superficial Bladder Cancer. J Urol 156(6):1934-1940, discussion1940-1941, 1996.

Wijlestrom H, Kasinen E, and the Urothelial Cancer Group of theNordic Association of Urology: A Nordic study comparing intravesicalinstillations of alternating mitomycin C and BCG with BCG alone in carcinoma insitu of the urinary bladder. J Urol 161:286-291, 1999.


1. Lamm DL, Blumenstein BA, Crissman JD, et al: Maintenancebacillus Calmette-Guérin immunotherapy for recurrent Ta, T1 and carcinoma insitu transitional cell carcinoma of the bladder: A randomized Southwest OncologyGroup Study. J Urol 163:1124-1129, 2000.

2. Lamm DL, Blumenstein B, Sarosdy MF, et al: Significantlong-term patient benefits with BCG maintenance therapy: A Southwest OncologyGroup Study. J Urol 157:213-216, 1997.

3. Malstrom PU, Wijkstrom H, Lundholm C, et al: 5-year follow-upof a randomized prospective study comparing mitomycin-C and bacillusCalmette-Guérin in patients with superficial bladder carcinoma.Swedish-Norwegian Bladder Cancer Study Group. J Urol 161:1124-1127, 1999.