Key points

Introduction

Radical cystectomy (RC) with pelvic lymph node dissection (PLND) remains the gold standard for local and regional therapy for muscle-invasive bladder cancer (MIBC), with more contemporary standards including perioperative cisplatin-based chemotherapy. For highly selected patients, surgical bladder preservation approaches including radical transurethral resection (TUR) and partial cystectomy (PC) with or without bilateral PLND may be reasonable alternatives to RC/PLND. A limited number of series have suggested that in appropriately selected patients, who represent a small fraction of all patients with MIBC, these approaches may offer reasonable oncologic outcomes while avoiding the morbidity associated with radical surgery and body image issues associated with urinary diversion while simultaneously preserving baseline urinary and sexual function. Furthermore, in the era of increasing perioperative systemic chemotherapy for MIBC, bladder-sparing approaches including radical TUR and PC have been investigated in patients demonstrating clinical response to neoadjuvant chemotherapy.

The multifocal nature of bladder cancer precludes many patients with MIBC from bladder-sparing surgical approaches. Two concepts have been proposed to account for bladder tumor multifocality. The first, known as the “field defect” hypothesis, is that the entire urothelium is exposed to a variety of carcinogenic insults that result in independent tumors. The second is that multiple tumors arise from the spread of a single clone through intraepithelial migration or implantation. Regardless of the cause, the multifocal nature of bladder tumors and recurrences needs to be taken into account when evaluating and considering bladder-sparing approaches to MIBC.

However, pathologic findings of no residual tumor at RC (pT0) suggest that there is a subset of patients who would be well-suited for bladder-sparing surgical approaches. The reported proportion of patients with a history of MIBC who achieve a pT0 response through TUR alone (without systemic neoadjuvant chemotherapy) ranges from 5% to 20% in the literature. Although this identifies a subset of patients who might do well with surgical bladder-sparing therapies, approximately 7.5% of patients who achieve pT0 through TUR alone have LN-positive disease detected at time of RC/PLND, and a larger proportion of pT0 patients experience disease recurrence and ultimately death from bladder cancer. In sum, although pT0 rates achieved through TUR alone suggest that there may be a subset of patients appropriate for surgical bladder-sparing approaches, the subset does not necessarily include all pT0 patients, nor can such patients be reliably identified. Clinical understaging remains a crucial obstacle to the successful implementation of bladder-sparing surgery for MIBC.

Regardless of the surgical technique used to manage invasive bladder cancer, regional pelvic LN spread is a known pathway of progression. Patients with non–muscle-invasive and pT2 disease at time of RC/PLND have a risk of LN involvement ranging from 5% to approximately 20%. Although LN involvement represents a very strong negative prognostic finding, some patients with low-volume regional LN disease at time of RC/PLND experience long-term disease-free survival without further therapy. Strategies for the management of regional LN management for patients selected for bladder-sparing approaches have not been clearly established; most series of radical TUR or PC as definitive therapy for MIBC have eliminated or limited the extent of PLND. On the other end of the surgical spectrum, another arena for risk-adapted treatment of MIBC is the identification of patients for whom expanding the extent of surgery may be beneficial. Expanding the extent of lymphadenectomy in patients with MIBC being treated with RC/PLND improves the accuracy of staging and has a potential therapeutic benefit. Studies are ongoing to define the relationship between the extent of lymphadenectomy and disease progression, disease-specific survival, and overall survival.

Introduction

Radical cystectomy (RC) with pelvic lymph node dissection (PLND) remains the gold standard for local and regional therapy for muscle-invasive bladder cancer (MIBC), with more contemporary standards including perioperative cisplatin-based chemotherapy. For highly selected patients, surgical bladder preservation approaches including radical transurethral resection (TUR) and partial cystectomy (PC) with or without bilateral PLND may be reasonable alternatives to RC/PLND. A limited number of series have suggested that in appropriately selected patients, who represent a small fraction of all patients with MIBC, these approaches may offer reasonable oncologic outcomes while avoiding the morbidity associated with radical surgery and body image issues associated with urinary diversion while simultaneously preserving baseline urinary and sexual function. Furthermore, in the era of increasing perioperative systemic chemotherapy for MIBC, bladder-sparing approaches including radical TUR and PC have been investigated in patients demonstrating clinical response to neoadjuvant chemotherapy.

The multifocal nature of bladder cancer precludes many patients with MIBC from bladder-sparing surgical approaches. Two concepts have been proposed to account for bladder tumor multifocality. The first, known as the “field defect” hypothesis, is that the entire urothelium is exposed to a variety of carcinogenic insults that result in independent tumors. The second is that multiple tumors arise from the spread of a single clone through intraepithelial migration or implantation. Regardless of the cause, the multifocal nature of bladder tumors and recurrences needs to be taken into account when evaluating and considering bladder-sparing approaches to MIBC.

However, pathologic findings of no residual tumor at RC (pT0) suggest that there is a subset of patients who would be well-suited for bladder-sparing surgical approaches. The reported proportion of patients with a history of MIBC who achieve a pT0 response through TUR alone (without systemic neoadjuvant chemotherapy) ranges from 5% to 20% in the literature. Although this identifies a subset of patients who might do well with surgical bladder-sparing therapies, approximately 7.5% of patients who achieve pT0 through TUR alone have LN-positive disease detected at time of RC/PLND, and a larger proportion of pT0 patients experience disease recurrence and ultimately death from bladder cancer. In sum, although pT0 rates achieved through TUR alone suggest that there may be a subset of patients appropriate for surgical bladder-sparing approaches, the subset does not necessarily include all pT0 patients, nor can such patients be reliably identified. Clinical understaging remains a crucial obstacle to the successful implementation of bladder-sparing surgery for MIBC.

Regardless of the surgical technique used to manage invasive bladder cancer, regional pelvic LN spread is a known pathway of progression. Patients with non–muscle-invasive and pT2 disease at time of RC/PLND have a risk of LN involvement ranging from 5% to approximately 20%. Although LN involvement represents a very strong negative prognostic finding, some patients with low-volume regional LN disease at time of RC/PLND experience long-term disease-free survival without further therapy. Strategies for the management of regional LN management for patients selected for bladder-sparing approaches have not been clearly established; most series of radical TUR or PC as definitive therapy for MIBC have eliminated or limited the extent of PLND. On the other end of the surgical spectrum, another arena for risk-adapted treatment of MIBC is the identification of patients for whom expanding the extent of surgery may be beneficial. Expanding the extent of lymphadenectomy in patients with MIBC being treated with RC/PLND improves the accuracy of staging and has a potential therapeutic benefit. Studies are ongoing to define the relationship between the extent of lymphadenectomy and disease progression, disease-specific survival, and overall survival.

Bladder-sparing approaches for muscle-invasive bladder cancer: radical transurethral resection

TUR, a diagnostic and therapeutic procedure in patients with non-MIBC, has been studied as a single-modality approach in patients with MIBC. Although the recurrence-free survival and long-term survival of unselected patients with muscle-invasive disease treated with TUR alone are inferior to those achieved by RC/PLND, in carefully selected patients, bladder-sparing approaches using radical TUR may be appropriate.

Historical series have reported relatively poor survival of patients with MIBC treated with TUR alone. For example, in 1977, Barnes and colleagues reported a 5-year survival rate of 31% for 75 patients exclusively undergoing TUR for the treatment of stage T2 bladder cancer. In contrast, Herr reported that patients with muscle-invasive disease treated with radical TUR achieved a 5-year survival rate of 68%. However, only 45 of 217 patients were eligible for bladder-sparing treatment with radical TUR and 15 of these 45 patients (33%) were subsequently treated with RC or developed metastatic disease.

Henry and colleagues reported a retrospective series comparing the outcomes of patients with stage B bladder cancer treated from 1974 to 1983 with TUR alone (N = 43), preoperative radiation and RC (N = 40), RC alone (N = 15), and definitive radiation therapy alone (N = 16). The 5-year survival rates for 43 patients with stages B1 and B2 disease treated with TUR were 63% and 38%, respectively; however, this was reported as comparable with survival rates in patients treated with radical surgery. Although the distribution of stage, grade, and number of tumors was not significantly different among the treatment groups, the sample size for each group was small and the groups were not randomized. For example, patients in the TUR groups were older and had more comorbidities than the other patients while having smaller tumors.

Selection of patients for bladder preservation with radical TUR is critical. A second or restaging TUR (reTUR) provides an opportunity for more accurate staging in patients initially diagnosed with non-MIBC, allows for assessment of the completeness of the initial resection in patients with muscle-invasive disease, and provides important information regarding patient selection. In a study by Herr, reTUR in 96 patients with non–muscle-invasive tumors revealed residual non–muscle-invasive disease in 55% and upstaging to T2 disease in 20%. Furthermore, in 54 patients with muscle-invasive disease at first TUR, 12 (22%) had no residual tumor and 30 (56%) had residual T2 disease.

Herr further reported on the 10-year outcomes of 151 consecutive patients with MIBC who were offered bladder preservation with radical TUR after reTUR demonstrated no tumor (T0), residual carcinoma in situ (Tis), or non–muscle-invasive (T1) tumor and followed for at least 10 years. Of these 151 patients, 52 elected for immediate RC and 99 elected for bladder preservation. Of the 52 patients who elected immediate RC (all of whom had T0, Tis, or T1 disease on reTUR), 35% were upstaged to pT2, pT3, or pN+ on final pathology, highlighting the problem of clinical understaging. Of the 99 patients treated with TUR alone, 75 (76%) were alive at last follow-up and 24 (24%) died of disease, rates comparable with those treated with RC (71% alive and 29% dead of disease). Analysis revealed that 18% of patients who had T0 on reTUR died of bladder cancer compared with 42% of those who had T1 on reTUR ( P = .003). The author concluded that the finding of residual T1 disease on reTUR be used to exclude patients from consideration of a radical TUR approach.

Solsona and colleagues have reported results from a prospective study of 133 patients with MIBC treated with a macroscopically complete radical TUR who had negative biopsies of the tumor bed. They excluded patients with macroscopically residual tumor after TUR, hydronephrosis, clinical evidence of LN involvement, and distant metastatic disease. Additionally, tumors with a sessile appearance or those larger than 3 cm were excluded from the study. Long-term results from their study included cancer-specific survival rates of 81.9%, 79.5%, and 76.7% and progression-free survival of 75.5%, 64.9%, and 57.8% at 5, 10, and 15 years, respectively. Approximately 30% of the patients in their study experienced disease progression, 22.5% as local bladder progression alone, 2.2% as local bladder progression associated with metastasis, and 5.5% as distant metastasis alone. Disease progression was associated with a high death rate of 67.5%, because only 12 patients were treated with salvage RC. Although 10 of the 12 patients treated with salvage RC (83.3%) were rendered free of disease, 89% of the remaining patients with disease progression who either refused or were not fit for salvage RC died of disease.

As demonstrated by the previously mentioned reports, radical TUR may be an option for highly select patients (excluding those with residual T1 disease on reTUR, sessile lesions, or tumors >3 cm) with MIBC desiring bladder preservation ( Table 1 ). Paramount to the success of radical TUR is the quality of initial TUR and reTUR. Furthermore, the patient and the physician must be dedicated to an intense and life-long regimen of endoscopic and radiographic surveillance. A major obstacle to the success of these approaches remains the inaccuracy of clinical staging, including the limitations of axial and molecular imaging. Given the limited information reported on pelvic progression in these series, it is difficult to assess the need for PLND in these select cohorts. Molecular imaging techniques have not yet matured to allow for accurate detection of primary or recurrent disease. The use of PET with computed tomography (CT) has been limited by urinary excretion, which limits the ability to detect tumors within the bladder; however, newer tracer agents with minimal urinary excretion, such as ¹¹ C-choline, are being tested. Currently, PET-CT has been most effective in staging and the detection of metastatic disease. Kibel and colleagues performed a prospective study using ¹⁸ F-FDG PET-CT in 43 patients undergoing RC and reported a positive predictive value of 78%, negative predictive value of 91%, sensitivity of 70%, and a specificity of 94%, leading the authors to conclude PET-CT may play an important role in planning treatment before cystectomy. MRI is increasingly used in the staging and longitudinal follow-up of patients with bladder cancer; however, the role of newer functional MRI techniques has yet to be definitively determined. As axial and molecular imaging techniques continue to evolve, additional refinements in the selection of patients suitable for surgical bladder preservation strategies will improve outcomes as those with locally advanced or occult regional disease can be identified and excluded. Long-term follow-up of additional prospective studies with clear patient selection criteria is needed for the critical evaluation of radical TUR as an option for select patients with MIBC.

Bladder-sparing approaches for muscle-invasive bladder cancer: partial cystectomy

PC represents an alternative surgical approach to bladder preservation with the advantages of allowing for accurate staging through full-thickness examination of the primary tumor and the ability to perform a concurrent bilateral pelvic lymphadenectomy. As with radical TUR, patient and tumor selection are critical to the success of PC for the treatment of muscle-invasive disease.

Early experience with PC before the 1990s was marked by high recurrence rates, cancer recurrence within surgical wounds, and poor overall cancer control and survival. Historical series described local recurrence rates between 29% and 78%, with 35% being noninvasive recurrences, 50% muscle-invasive recurrences, and 14% presenting with distant metastases. Tumor recurrences within surgical incisions occurred in up to 40% of all patients and up to 54% in those with high-grade tumors. Up to 20% of patients required salvage RC with 5-year overall survival rates of 50%.

During the 1990s additional series from centers using strict selection criteria to identify appropriate candidates were reported. These criteria include a solitary tumor in a location suitable for resection with a 2-cm margin of normal bladder, no associated carcinoma in situ, and no history of prior bladder tumors. Adequate bladder capacity and function should be maintained with the 2-cm margin of resection depending on initial bladder characteristics. Ureteral reimplantation, if necessary, is feasible. However, tumors located at the bladder neck or trigone may not be suitable for PC because achieving a suitable margin may be difficult. Using these more restrictive criteria, only 3% to 10% of patients presenting with muscle-invasive tumors remain candidates for PC.

Retrospective analysis of 58 patients from Memorial Sloan Kettering Cancer Center (MSKCC) treated with PC (41 with muscle-invasive disease) identified that tumor multifocality ( P <.001) and the presence of carcinoma in situ ( P = .027) were associated with an increased risk of local recurrence following PC (see Table 1 ). Pathologic confirmation of positive LNs and positive surgical margins were associated with advanced recurrences, defined as the development of recurrent muscle-invasive disease or distant metastases ( P = .012 and P = .022, respectively). A study from MD Anderson reviewed 37 patients with muscle-invasive urothelial bladder cancer and demonstrated that history of prior bladder tumor ( P <.003) was associated with disease recurrence. Adjuvant chemotherapy may be considered for patients with pathologic evidence of extravesical disease or positive LNs after PC. In the series from MD Anderson, high-risk patients with locally advanced disease and positive LNs treated with adjuvant chemotherapy (N = 9) had significantly longer progression-free survival, but overall and cancer-specific survival were not affected. The ability to achieve an adequate negative surgical margin and adherence to these selection criteria cannot be overemphasized when evaluating a potential candidate for PC.

With refinements in patient selection, recurrence rates and overall survival in patients treated with PC have improved. Recurrence rates range from 28% to 48%; with non–muscle-invasive recurrence in 12% to 50%, muscle-invasive recurrence in 17% to 57%, and metastatic recurrence in 14% to 52%. Positive surgical margins are described in 9% to 14% of patients undergoing PC; of the five patients with a positive surgical margin in the MSKCC series, three had muscle-invasive recurrence and one presented with distant metastases. Salvage RC was required in 21% to 28% of patients, with one series describing 75% of patients with muscle-invasive recurrences rendered disease-free after salvage RC. The lack of wound recurrences and tumor implantation in these series is notable and likely reflective of appropriate patient selection and improvements in surgical technique.

These contemporary series report 5-year overall survival rates of 67% to 70%, recurrence-free survival rates of 39% to 62%, and cancer-specific survival rates of 84% to 87%. Between 65% and 74% of patients maintained an intact bladder and 49% to 67% were free of disease with an intact bladder at last follow-up. Although 86% of recurrences were described in the first 2 years after PC, late muscle-invasive recurrences have been reported indicating the need for life-long surveillance in these patients. Furthermore, these series have relatively short follow-up in comparison with mature RC series. As learned from the long-term follow-up from the Solsona series of radical TUR, longer follow-up is crucial to the evaluation of these results because patients remain at risk of recurrence and progression.

Contralateral nodal involvement in bladder cancer is not uncommon and is important when considering PLND for patients undergoing PC. Leissner and coworkers reported on 119 patients whose bladder tumors could be strictly localized to one side of the bladder and found a significant rate of contralateral LN involvement with the risk of contralateral LN metastases only slightly lower than for the ipsilateral side. Furthermore, of 13 patients with unilateral tumors and a solitary LN metastasis, 3 of 13 (23%) of the solitary positive LNs were located on the contralateral side. The aforementioned series of PC did not explicitly describe the extent and laterality of the PLNDs performed. Nevertheless, 9% and 14% of the patients in the MSKCC and MD Anderson series had positive LNs, respectively; this proportion might be higher if all patients were treated with bilateral PLND. Currently, the accuracy of axial and molecular imaging and the ability to predict which patients have occult LN metastases are limited. Patients undergoing a PLND at the time of PC should have a bilateral node dissection performed for accuracy of staging and added potential therapeutic advantage in the setting of limited pelvic nodal involvement.

Minimally invasive approaches are being investigated and adopted for a variety of urologic oncologic surgeries, including RC for bladder cancer. Similarly, robot-assisted PC might represent an alternative to open PC. Surgeons must maintain strict adherence to oncologic principles as the techniques of minimally invasive PC are refined and outcomes should be critically analyzed before widespread adoption of the procedure.