Bladder Cancer: Imperatives for Personalized Medicine
Bladder Cancer: Imperatives for Personalized Medicine
ABSTRACT: Although the age-adjusted incidence of urothelial carcinoma has stabilized or declined in developed nations as a result of tobacco and environmental regulations, the rising numbers of the elderly and the shift in the tobacco epidemic to underdeveloped and rapidly industrializing nations with less stringent environmental controls augur a major growth in the worldwide burden of this disease. Current understanding of the molecular pedigree of urothelial carcinoma indicates that the disease follows a two-pathway model. The first of these, the common non–muscle-invasive papillary disease (Ta) defined by fibroblast growth factor receptor 3 (FGFR3) mutations and Ras pathway signaling, is characterized by a very low (< 5%) incidence of progression to invasive disease and very low disease-specific mortality. The second, or more lethal form is characterized by carcinoma in situ and invasive (lamina propria or deeper) tumors featuring p53 and Rb defects with a high risk of disease-specific mortality. For high-risk non–muscle-invasive disease, optimized intravesical therapeutics, including adequate transurethral resection, peri-operative intravesical chemotherapy, adjuvant intravesical bacille Calmette-Gurin and/or timely cystectomy, are needed to minimize disease-specific mortality and maximize quality of life. In muscle-invasive organ-confined disease, surgery remains the standard of care, with neoadjuvant chemotherapy providing a survival benefit in a subset of patients. Research strategies that identify disease subsets of muscle-invasive bladder cancer that benefit or do not benefit from adjunctive chemotherapy are required to reduce the relatively high number-needed-to-treat associated with this approach. To facilitate major therapeutic progress in the disease, accelerated study of experimental therapeutics connected to a fuller portrait of the heterogeneous molecular pathophysiology of bladder cancer is needed. Effective multidisciplinary collaboration is imperative in order to implement existing knowledge, enable priority research, reduce costs, and improve on the clinically relevant endpoints of survival and quality of life.
It has been estimated that 2.7 million people worldwide have a history of bladder cancer and that approximately 145,000 die every year from the disease. In the United States, urothelial carcinoma of the bladder (UC) accounts for 95% of bladder cancers. Cigarette smoking accounts for at least 50% of UC in men and 35% in women.[2,3] It is less certain that UC in nonsmokers is related to environmental tobacco smoke exposures. Aromatic amines, polycyclic aromatic and chlorinated hydrocarbons, arsenic-laced drinking water, aristolochic acid, cyclophosphamide exposure, and a range of industrial chemicals have been implicated in urothelial carcinogenesis. Marked variation in individual susceptibility despite seemingly equal carcinogen exposures has been explained by genetic polymorphisms regulating varied detoxification mechanisms.[3,4] Familial bladder cancer is rare, and viral pathogenesis remains unproven. Although chronic inflammation is strongly implicated in the pathogenesis of squamous carcinomas of the bladder, its role in the pathogenesis of UC is unproven. Prior use of nonsteroidal anti-inflammatory drugs, but not aspirin, is associated with a lowered risk of UC among nonsmokers alone. Ninety percent of UC occurs in persons older than 55 years, with the highest risk in those aged 75 to 85 years.
The multipronged approach to control of Schistosoma hematobium infections in Egypt that was followed by steep declines in mortality from associated squamous carcinomas of the bladder is a remarkable public health success story. However, aging among rising world populations and a shift in the tobacco epidemic to rapidly industrializing nations with less stringent environmental protections portend a major increase in the global incidence of and mortality from UC.
The periodic cystoscopies employed in the treatment and monitoring of bladder cancer place a significant burden on existing healthcare resources; in a 1995 survey of five solid tumors in the elderly, Medicare payments from diagnosis to death were highest for bladder cancer. Cost-effective paradigms for controlling the global scale and impact of the disease are needed.
A Survey of Existing Management Paradigms and Challenges in Bladder Cancer
Transparently, the most effective single method for preventing and reducing deaths from bladder cancer would be the eradication of cigarette smoking. A more vigorous role for the oncology community in efforts to achieve this goal is strongly recommended.[3,10] Given that no more than a quarter of bladder tumors have lethal potential, it would be extremely difficult for a population-based screening test to demonstrate reduction in bladder cancer–specific mortality. For the diagnosis of UC, a highly sensitive and specific biomarker for malignancy that performs better than urine cytology and radiographic and endoscopic evaluation has not been identified.
Education of patients and physicians with regard to the importance of prompt evaluation of macroscopic hematuria can make a difference in preventing deaths from the disease. Microscopic hematuria carries far less specificity for cancer. There is insufficient evidence for an evidence-based algorithm for the investigation of hematuria. Cystoscopy and upper tract evaluation are recommended in all patients with microscopic or macroscopic hematuria, particularly in the absence of infection or stones. Irritative voiding symptoms may imply carcinoma in situ and should also be evaluated with cystoscopy. Urine cytology has high specificity and positive predictive value for high-grade disease and carcinoma in situ but low sensitivity for low-grade lesions. Other studies that are available—eg, the Nuclear Matrix Protein 22 (NMP22) test, fluorescence in situ hybridization (FISH), while often used, have yet to make an impact on the diagnosis and management of UC.
Non–muscle-invasive bladder cancer (NMIBC)
Tumors staged as Ta, T1, or carcinoma in situ (CIS) are grouped as NMIBC and account for 75% to 85% of bladder cancers. The terms "superficial" or "NMIBC' may serve to conceal the lethal biology contained within these entities. Guidelines for the management of NMIBC have been developed by the American Urological Association (AUA) and the European Association of Urology (EAU); these are continually updated at www.uroweb.org. The guidelines emphasize the importance of careful cystoscopic evaluation and a complete and correct transurethral resection (TUR) to making the correct diagnosis and facilitating removal of all visible tumors. Pathology reports should specify lesion grade, depth of invasion into the bladder wall, and whether lamina propria and muscle cells are present in the specimen.
A risk-adapted approach to perioperative or postoperative adjunctive intravesical chemotherapy or immunotherapy with bacille Calmette-Gurin (BCG) is presented in the EAU guidelines. Risk stratification is determined by assigning weights to the number of tumors, tumor diameter, prior recurrence history, Ta vs T1 disease, presence or absence of CIS, and grade (according to the 1973 World Health Organization classification); these weighted factors are used to generate total scores for recurrence (0 to 17) and progression (0 to 23). Patients are then stratified on the basis of their scores as low-, intermediate-, or high-risk for recurrence and progression at 5 years; each of these risk groups has its own correlative treatment and surveillance recommendations.
For example, in patients at low risk for recurrence and progression, a single postoperative instillation of chemotherapy (eg, mitomycin C) is recommended, with follow-up cystoscopy at 3 months—and if cystoscopy results are negative, again at 9 months and then yearly for 5 years. By contrast, for patients with a high risk of progression, the recommendations call for induction BCG (weekly × 6) followed by maintenance therapy as tolerated for at least 12 months—or immediate cystectomy. The US guidelines are similar: a single perioperative instillation of chemotherapy (mitomycin) is recommended for all patients undergoing transurethral resection of bladder tumor (TURBT), with the addition of adjuvant intravesical BCG induction and maintenance (3 weekly instillations at 3 and 6 months and then every 6 months) for high-grade tumors.
Although the risk of disease progression is reduced with intravesical BCG, because of the inclusion of low-risk NMIBC lesions in randomized trials, as well as the short follow-up time when the results of these trials are reported, it has been difficult to demonstrate an overall survival benefit. Furthermore, the chronicity of UC mandates lifelong follow-up of patients; thus, studies likely underestimate the true impact of disease recurrence and progression after bladder-sparing strategies. It is unlikely that a randomized trial will be performed comparing survival and quality of life outcomes of immediate cystectomy vs intravesical BCG therapy with or without salvage cystectomy for high-risk NMIBC. There are no intravesical therapies for high-risk disease that have clearly improved outcomes over those achievable with BCG induction and maintenance therapy.
While maintenance BCG reduces recurrence and progression rates, there is no consensus on the optimal dose, schedule, and duration of therapy (beyond 1 year). Early identification of BCG failure and subsequent transition to cystectomy is critically necessary to prevent deaths from progression to metastatic disease.[16,17] It has been estimated that between 30% and 45% of bladder cancer deaths could be avoided by earlier implementation of cystectomy in surgically fit patients with NMIBC. This remarkable estimate, if accurate, is far in excess of the absolute survival benefit of 5% that neoadjuvant chemotherapy provides for muscle-invasive disease (MIBC).
BCG failure is inconsistently defined in the literature; a consensus definition has been proposed. Persistent disease at 3 months following induction BCG therapy in high-risk NMIBC or failure to achieve disease-free status at 6 months following initial BCG therapy with either maintenance or retreatment at 3 months should prompt consideration of immediate cystectomy. To facilitate better treatment choices, molecular predictors of BCG outcome require elucidation and incorporation into clinically useful predictive tools. Preliminary data suggest that the urinary cytokine response to BCG may predict response to BCG.
Among surgically unfit patients, definitive studies that establish the efficacy of intravesical chemotherapy or other modalities as standard of care are awaited. Patients with micropapillary histology, evidence of lymphovascular invasion, or prostatic urethral involvement may be better served by immediate cystectomy. The presence of multiple or high-risk tumors is associated with a higher risk of upper tract recurrences. Persistent urine cytology positivity in association with negative cystoscopic findings warrants a search in upper tracts, with prostatic, urethral, and random bladder biopsies. There is limited evidence to suggest that smoking cessation following a diagnosis of NMIBC reduces the risk of recurrence and progression of disease. However, neither the AUA nor the EAU guidelines cite smoking cessation as a management goal in NMIBC.
Muscle-invasive bladder cancer (MIBC)
There are two large randomized prospective phase III trials that demonstrate the benefit of neoadjuvant chemotherapy in MIBC. The European Organisation for the Treatment and Cure of Cancer (EORTC) study of treatment of MIBC with cisplatin, methotrexate, and vinblastine and the Southwest Oncology Group (SWOG) study of methotrexate, vinblastine, Adriamycin (doxorubicin), and cisplatin (MVAC) demonstrated that neoadjuvant therapy resulted in a 10-year absolute survival benefit of approximately 5% (number needed to treat [NNT] = 20). Pathological complete remissions with neoadjuvant therapy are associated with 85% long-term survival; conversely, pathological node involvement following neoadjuvant chemotherapy portends a high frequency of lethal disease progression.
In contrast to neoadjuvant studies, the randomized adjuvant chemotherapy trials have been repeatedly criticized as underpowered and flawed, and two recent trials were closed due to poor accrual. A randomized comparison of 5 cycles of post-cystectomy MVAC chemotherapy (n = 70) vs 2 cycles of neoadjuvant and 3 cycles of adjuvant MVAC (n = 70) estimated that there were no major differences in survival outcomes between the two arms despite better tolerance of neoadjuvant MVAC. Similar drop-out rates (10%) were observed when patients proceeded from neoadjuvant chemotherapy to surgery or proceeded from surgery to chemotherapy.
By contrast, in a large retrospective study, one-third of patients with MIBC were judged unfit to receive adjuvant chemotherapy following cystectomy as a result of the high rate of postoperative complications. Patients with extravesical or pN+ MIBC who have recovered quickly from cystectomy should be considered for adjuvant chemotherapy, preferably on a clinical trial. While the optimum number of adjunctive cycles of chemotherapy is undefined, a minimum of 3 cycles should be targeted for completion. Currently, the use of gemcitabine (Gemzar)-cisplatin or dose-intense MVAC regimens has been transposed to the adjunctive therapy of MIBC, given the therapeutic equivalence and lesser toxicity of these regimens compared with traditional MVAC in metastatic disease.[26,27]
Despite level 1 evidence to support its use, it has been estimated that ≤ 12% of patients with MIBC receive neoadjuvant chemotherapy; it is likely that physician, patient, and institutional barriers contribute to this statistic. The attitude, skill, and knowledge variables that account for these barriers require better understanding.
Organ-confined micropapillary UC, squamous carcinomas, and adenocarcinomas of the bladder are best managed with early cystectomy, while neoadjuvant etoposide-cisplatin–based chemotherapy is preferred in small-cell carcinomas. In patients with ≥ T3 or N+ small-cell carcinoma, prophylactic cranial irradiation should be considered, as the brain is often the exclusive site of relapse. In UC with mixed glandular and squamous features, there is evidence of a persistent survival benefit with neoadjuvant chemotherapy.
Given the quality of life consequences of removal of the bladder and the prevalence of MIBC among the surgically unfit, there is abiding interest in advancing organ-conservation through the integration of TUR, chemotherapy, and radiation therapy. There are no direct comparisons of surgery and radiotherapy in the management of MIBC. Patient selection and the availability of salvage surgery contribute to disease-specific and overall survival outcomes. Following chemotherapy, a third of patients will harbor residual MIBC despite negative cystoscopies and biopsies, and organ preservation is thus not recommended. Current guidelines for MIBC recommend that radiation-based bladder conservation attempts be reserved for the medically unfit.
Metastatic bladder cancer (MBC)
Chemotherapy remains the mainstay of management in MBC, and while advances in this area have long since reached a plateau, the development of less-toxic regimens, including gemcitabine-cisplatin and dose-dense MVAC,[26,27] has facilitated the design of novel combination strategies. Among subgroups of patients with metastatic disease who have a favorable risk profile (minimally symptomatic, no visceral metastases), long-term survival with cisplatin-based chemotherapy may be seen in up to 25% of patients, whereas the median survival of patients with poor risk profiles (symptomatic, visceral metastases) is 9 months. These data demonstrate the importance of risk stratification for the design and interpretation of clinical trials in MBC. The integration of biomarkers into prognostic and predictive models has the potential to further enhance the accuracy of clinical trial design and therapy planning. In the second-line setting, although response rates are low, vinflunine has been shown to modestly improve survival over best supportive care and is available in Europe.
Metastatic bladder cancer remains the major testing ground for novel therapeutics, and while to date novel agents, including angiogenesis inhibitors and signal transduction inhibitors, have not yielded significant single-agent activity in MBC, a range of combination strategies are being pursued in single-arm and randomized studies. The challenge of linking accurate biological information with drug design and/or treatment strategy is key to the success of these efforts.