Chemotherapy should be immediate after TUR in patients with non-muscle invasive disease
Bladder cancer is projected to be the fourth most common cancer diagnosed in 2007 and the eighth most common cause of cancer-specific death in men, accord-ing to the American Cancer Society. Approximately 70% of bladder tumors present as non-muscle invasive (NMI) disease and 10%-20% will progress to muscle invasion.
Non-muscle-invasive disease is a diverse spectrum which includes Ta, Tis, and T1 lesions. It is well established that grade is an indepen-dent predictor of progression. Low- grade tumors tend to be NMI, with a recurrence rate of 50%-70%; however, progression to muscle invasion is low (less than 5% of cases). High-grade tumors (including Tis) have a high recurrence rate of approximately 80%, and up to 50% may progress in stage, which places these patients at higher risk for metastasis and death. With high-grade NMI disease, the benefits of bladder preservation with intravesical therapy must not outweigh the potential cure by radical cystectomy if necessary. In this review, we will examine the latest data available regarding intravesical therapy for this disease.
Transurethral resection
When discussing institution of intravesical therapy, accurate patient selection is paramount. It has been well established in multiple studies that bladder cancer is often understaged, especially with regards to invasion of muscle (Cancer. 1999;86:1035-1043; J Urol. 1999; 162:74-76; J. Urol. 2002;60:822-824; J Urol. 2005;174:2134-2137). For this reason, a repeat transurethral resection (TUR) is advocated, either if there is no muscle present in the original resection or if the primary resection reveals T1 disease, even if muscle was present. We wait for four to six weeks because this time interval al-lows an insight into the biology of the disease; for example, a tumor that recurs as T1 in four weeks is inherently more aggressive than one that does not recur or recurs as a Ta lesion only. The importance of re-TUR is emphasized by a recent report by Herr et al. They reported on early re-staging TUR of a cohort of 352 patients with T1 bladder cancer, finding that 58% had residual tumor (26% NMI), of which 66% had recurred and 35% progressed in stage during five years of follow-up (BJU Int. 2006;97:1194-1198; J Urol. 2007;177:75-79). Of the 26% with residual T1 tumors at repeat-TUR, 82% progressed to muscle invasion in five years compared with 19% progression with P0 disease on re-staging TUR. These findings emphasize the need to appropriately stage patients with TUR and possibly re-stage TUR to identify patients who would be better managed by early cystectomy.
Bacillus Calmette-Guérin
Bacillus Calmette-Guérin (BCG), an attenuated form of Mycobacterium tuberculosus, is currently our most effective agent against NMI bladder cancer. The exact mechanism of action remains to be elucidated, but it is thought that BCG works with the assistance of helper T-cells and cytokines (including TRAIL), causing an immune reaction. Many years ago, Morales and colleagues established its use as immunotherapy for bladder cancer (J Urol. 1976;116:180-183). Analysis of early trials revealed the unequivocal benefits of decreased recurrence and increased time to recurrence with maintenance prophylactic therapy (Urol.1991;38:507-513).
There are many commercial producers of BCG, and while efficacy data for different strains can vary slightly (with no clear “winner”), it has been established that 10 million organisms must be delivered per instillation. BCG is usually provided in a lyophilized format and, after reconstitution in 50 mL of saline for intravesical instillation, is introduced into the bladder via a catheter. Patients are then asked to ‘hold’ BCG in their bladder for 60-120 minutes (although some data suggest that as little as 30 minutes might be sufficient).
Symptoms with therapy commonly include dysuria, urgency, frequency, and occasionally hematuria. Systemic symptoms are common but are self limiting (e.g., low grade fevers, flu-like symptoms). Serious adverse events, however, may be caused by an intravascular inoculation most commonly caused by traumatic catheterization resulting in severe BCGosis (systemic tuberculosis). For this reason therapy should be delayed after TUR for a minimum of two to three weeks and traumatic catheterizations should preclude any further instillations at that setting. BCG is most efficacious against carcinoma in situ, with initial treatments responses of 70%-80% (Urol. 1991;38:507-513; J Urol. 2000;163:1124-1129; J Urol. 1995:153:564-572; J Urol. 1982;128:27-30).
Controversy surrounds BCG and its ability to prevent long-term progression of bladder cancer. Evidence is lacking that a single induction dose of BCG slows progression, but more data suggest that maintenance BCG prevents progression. In a meta-analysis sponsored by the European Organization for Research and Treatment of Cancer (EORTC), Sylvester et al found a reduction of 27% in the odds of progression in patients who received maintenance BCG compared with controls (J Urol. 2002;168:1964-1970).
There are various maintenance schedules, but we have adopted the SWOG 6+3 schedule, which entails six weeks induction with additional three weekly instillations every three months for the first year, then every six months for the second year, and a final instillation at year 3. This schedule (also known as the “Lamm protocol”) was reported by Lamm and demonstrated in complete responders that BCG using three weekly treatments at six-month intervals increases long-term response rates from 65% to nearly 90%.
BCG failures
Many different definitions for BCG failures exist in the literature. This has impaired adequate comparison of trials simply on the basis of improper risk assessment. While several authors have proposed variations in definitions of “failure,” what is practically important is to differentiate “refractory disease” from “recurrent disease.” Patients have BCG refractory disease if they fail to achieve a disease-free state within six months (after induction with or without maintenance). BCG recurrent (or relapsing) disease implies recurrence of disease after being rendered disease-free at the six-month evaluation. Patients who develop a recurrence in the first year (or even first two years) are different from those who develop a late recurrence and are important for trial design. Lastly, O’Donnell recommends that we consider an additional group of patients as “BCG intolerant,” i.e., recurrence in the setting of inadequate BCG dosing due to patient intolerance (World J Urol. 2006;24:481-487).
Possible management of BCG relapses and early failures include repeat courses of BCG; however, this must be undertaken with extreme caution because there are few studies examining the long-term effects on progression (Hinyokika Kiyo. 2005;51:539-543; Br J Urol. 1995;75:180-184; J Urol. 1990;143:710-712). A second course of BCG in this population may achieve a response rate of up to 34%. Nonetheless, especially for high-grade disease, radical cystectomy should be offered as definitive therapy in BCG failures; especially in those receiving adequate doses of BCG with induction and with persistent disease at three to six months. For patients who recur “late” (two or more years after response), re-induction with BCG can be considered, but close follow-up is required to monitor for progression.
BCG + interferon alpha 2b
Interferon alpha 2b (IFN·-2b) has been intensely studied in its application to bladder cancer, and has not been shown to decrease progression as a monotherapy (Br J Urol. 1998;82:829-834; Urol. 1997;49:187-190). IFN-2b combined with BCG has been tried for BCG failures. (IFN-2b is typically given between 50-100 million units along with dose reductions of BCG.) A phase II multicenter trial evaluated 1,007 patients, and of BCG failures treated with IFN-2b + BCG, 45% were disease-free at a median follow-up of 24 months (Urol Oncol. 2006;24:344-348). This trial found overall no advantage to the addition of IFN-2b to BCG versus what one would expect in historical controls of BCG alone in BCG-naïve patients. For this reason, BCG should be given alone as first-line therapy. It is noteworthy that this trial failed to show any benefit for combination therapy in patients who had failed two prior courses of BCG or those who had recurrence of disease within two years of induction therapy. Thus, the initial enthusiasm for combination therapy with IFN-2b must be tempered with the realization that radical cystectomy remains the mainstay of therapy for patients failing BCG therapy (provided they have received at least nine instillations, i.e., one induction and at least one maintenance course).
Intravesical chemotherapy
In 1996, a pooled meta-analysis of 2,535 patients by the European group EORTC and the Medical Research Council (MRC) found a significant prolongation of the disease-free interval, but no effect on progression, metastasis, or survival with intravesical chemotherapy after TUR (J Urol. 1996;156:1934-1940). There are no data in the literature that demonstrate in-travesical chemotherapy prevents progression. However, not all patients will be able to receive BCG (i.e., patients that have received the BCG vaccine or immunosuppression secondary to transplantation) and not all BCG-failing patients are willing to undergo radical cystectomy. Thus, in select patients, intravesical chemotherapy is useful.
Mitomycin C
Mitomycin C (MMC) is an alkylating agent that inhibits DNA synthesis in G1 and S phases of the cell cycle (molecular weight [MW] 334 kDA). Studies that have compared MMC to BCG have shown that BCG with maintenance is superior to MMC (Urology. 2004;63:682-686; Urol. 1998;52:403-410; Eur J Cancer. 1993;29A:1672-1676). Gardmark et al recently published 10-year data on patients who received BCG with maintenance versus MMC for high-risk bladder cancer. They found no significant difference in recurrence or progression (BJU Int. 2007 Jan 22 [published online ahead of print]. Intravesical use of MMC after TUR is growing. A meta-analysis of 1,476 patients who received intravesical chemo-therapy immediately after TUR showed a 39% reduced odds of recurrence (J Urol. 2004;171[6 Pt 1]:2186-2190). For high-risk bladder tumors, 40 mg of MMC in 20 mL of normal saline should be instilled in the bladder for 60 minutes. Common adverse effects are chemical cystitis and severe contact dermatitis if in contact with the skin.
BCG + electromotive MMC
In a novel approach, Di Stasi et al randomized 212 patients with pT1 bladder cancer to receive BCG alone or BCG for two weeks followed by electromotive MMC (in-travesical current 20 mA for 30 min). After a median follow-up of 88 months, they found that BCG + electromotive MMC group had a higher disease-free interval versus BCG alone (69 vs. 21 months) (Lancet Oncol. 2006;7:43-51). Additionally, the BCG + electromotive MMC group had lower recurrence rates, decreased progression, and decreased overall and disease-specific mortality. The same authors showed in a previous study using electromotive MMC that serum levels of MMC were six times that of normal intravesical MMC (J Urol. 2003;170:777-782). These higher, possibly “chemotherapeutic,” systemic levels in theory may be the reason there are better overall outcomes in this study. Importantly, the value of this approach has not been evaluated in BCG failures.
Anthracycline antibiotics
Anthracycline antibiotics work by binding nucleic acid base pairs, inhibiting topoisomerase II (DNA gyrase) and protein synthesis, making its action cell cycle non-specific. Most commonly used anthracycline antibiotics for intravesical therapy are doxorubicin (MW 543.5 kDA), epirubicin (MW 580 kDA), and valrubicin (MW 723.6 kDA). Small studies have shown that doxorubin results in a decreased recurrence rate compared with TUR alone, but is inferior to BCG (J Urol. 1990;143:502-506). Its side effect profile has made it less utilized clinically. Epirubicin has established activity in NMI bladder cancer. A recent EORTC trial compared BCG versus epirubicin for CIS of the bladder and found no differences in terms of complete response rates (BCG 65% and epirubicin 56%), time to progression, or survival (J Urol. 2005;173:405-409). Epirubicin had few adverse effects but BCG appeared to produce more durable responses. Valrubicin is a semi-synthetic derivative of doxorubicin, being lipid-soluble, and it rapidly traverses cell membranes. Steinberg et al studied valrubicin as a second-line therapy in patients with BCG refractory disease with a total of 90 patients and found that 21% achieved a complete response at a median follow-up of 30 months (J Urol. 2000;163:761-767). Valrubicin was voluntarily taken off the market in 2002 secondary to drug stability issues but may be re-released later this year.
Thiotepa
Thiotepa is an alkylating agent that works through cross-linkage of DNA, RNA, and proteins. Thiotepa is cytotoxic but unfortunately has a low molecular weight (189 kDa) which allows for systemic absorption and myelosuppressive side effects in up to 50% of patients (JAMA. 1980;244: 2065-2067).
Gemcitabine
Gemcitabine is a nucleoside analogue that inhibits DNA synthesis and results in apoptosis. Even though the molecular weight is relatively low, five independent phase I/II studies using intravesical gemcitabine demonstrated low systemic levels. Studies giving intravesical doses of 500-2,000 mg have shown a complete response rate between 23%-56% (J Clin Oncol. 2006;24:2729-2734; J Clin Oncol. 2003;21:697-703; Eur Urol. 2004;45:182-186; BJU Int. 2004;93:491-494; J Urol. 2004;172:485-488). While initial reports showed promise, with longer term follow up, Dalbagni et al were unable to demonstrate long-term efficacy with this agent because the one-year recurrence free rates were only 21% in complete responders.
Keyhole-limpet hemocyanin
Keyhole-limpet hemocyanin (KLH), a respiratory pigment of the sea mollusk Megathura crenulata, is a non-specific immunomodulator that stimulates both humoral and T-cell mediated immune responses (Int Arch Allergy Appl Immunol. 1974;47:155-160; Br J Urol. 1995;76:702-707). Lamm et al reported a multicenter phase I/II trial using intravesical KLH and demonstrated complete responses in 58% of patients with CIS without papillary disease, followed by responses in 50% with CIS with papillary disease, and 33% in those with residual papillary disease (Eur Urol. 2000;37 [Suppl 3]:41-44). Data suggested that KLH is more effective than MMC at preventing recurrence. The risk of recurrence with KLH (mean follow-up 20 months) was 14% vs. 39% for MMC (mean follow-up 18 months). With its minimal adverse effects, KLH has great potential for combination therapy with BCG (J Urol. 1988;139:723-726).
Another potentially useful agent is mycobacterial cell-wall-DNA complexes (MCC), which, in theory, maintain the antigenicity of the cell wall. This results in the immunomodulatory impact that yields the “BCG effect” without the pathogenic potential on the intact organism. In an early trial in which MCC was administered to patients with CIS, the combined complete and partial response rates were 56% at 12 months with 31% classified as responders even at four years (J Urol. 2001;166:1633-1637). Large multicenter trials are currently underway evaluating a novel suspension form of MCC in both BCG failures as well as in a phase III fashion, compared with BCG as upfront therapy in BCG naïve patients.
Conclusion
A critical analysis of the literature to date suggests that patients with bladder cancer should undergo a complete TUR. In patients suspected of having NMI disease, TUR should be followed by immediate postoperative instillation of intravesical chemotherapy, preferably within six hours, unless there are contraindications (e.g., bladder perforation). For patients with high-grade T1 disease, a re-look and re-TUR at four to six weeks should be performed before bladder salvage with intravesical therapy or early cystectomy is considered. For pa-tients who decide on bladder-sparing treatment, a six-week induction course of BCG should be followed by maintenance (we use the 6+3 SWOG protocol) because evidence suggests that only those who receive maintenance derive any protection from progression of disease. Patients who fail BCG early (i.e., within 12-24 months) or who never achieve a disease-free state (non-responders) should be counseled on radical cystectomy. No alternate therapy exists that can be considered safe and effective, although several are being studied.
Drs. Gaston and Kamat are affiliated with the University of Texas M.D. Anderson Cancer Center in Houston. Dr.Gaston is a fellow in urologic oncology. Dr. Kamat is assistant professor and director of the fellowship program in urology.