Predicting Prognosis in Patients With Superficial Bladder Cancer

Introduction

Bladder cancer is the most common malignancy encountered by urologists. It is estimated that, in 1998, 54,400 cases of bladder cancer will be diagnosed in the United States, leading to approximately 12,500 deaths.[1]

More than 90% of bladder tumors are transitional cell carcinomas. Risk factors for the development transitional cell carcinoma include cigarette smoking and exposure to arylamines (such as aniline dyes), certain drugs, and pelvic radiation.[2-5] Cigarette smoking is the cause of 25% to 60% of all bladder cancers, and smokers have two times the risk of developing this cancer than do nonsmokers.[5]

Fortunately, most patients (80%) with newly diagnosed bladder cancer present with “superficial” disease, which can be managed with transurethral resection or intravesical therapy. The term “superficial bladder cancer” refers to tumors confined to the mucosa (Ta), those with lamina propria invasion (T1), and carcinoma in situ (Tis).[6]

The recurrence rate for superficial bladder tumors is high (two-thirds), and, despite advances in management, some patients still develop stage progression. It is well known that such stage progression occurs more often in patients with high-grade lesions and T1 tumors. For this reason, it seems imprecise to group all “superficial” tumors together.

Tumors that pose a high risk for progression and death are multifocal Ta lesions, tumors with associated Tis disease, and T1 lesions. Eventually, one-half of these high-risk patients will require a cystectomy, and one-third are at risk of dying from bladder cancer over 15 to 20 years.[6]

A prognostic marker that could identify individuals at risk for recurrence and progression might dramatically change the clinical management of patients with bladder cancer. The availability of such a marker would permit the selective use of aggressive treatment in patients who are at high risk of recurrence and progression while sparing low-risk patients from un-necessary procedures.[7]

Researchers are investigating several prognostic factors that may provide such a model. Traditional prognostic factors in bladder cancer have included tumor characteristics based on cystoscopic and pathologic findings. Certain biological markers may also play important roles in predicting high-risk tumors. These include DNA ploidy, S-phase, certain monoclonal antibodies, the p53 (alias TP53) tumor-suppressor gene, the retinoblastoma (Rb) gene, cell adhesion molecules, and angiogenesis.

Tumor Characteristics

Cystoscopic Findings

Characteristics of bladder tumors that are visible with cystoscopy can have important prognostic implications. For example, tumor size may play a role in progression to muscle invasive disease. In one series, tumors > 5 cm had a 35% progression rate, as compared with a 9% rate for smaller tumors.[8]

The number of tumors is also important for predicting recurrence but not progression.[8] Recurrence rates in patients with single tumors range from 18% to 60%, as opposed to rates of 40% to 90% in those with multiple tumors.[6]

The timing of first recurrence is another significant cystoscopic finding. Fitzpatrick et al reported that, among patients with Ta lesions who had no recurrences at their first post-resection cystoscopy, 79% had no subsequent recurrences.[9] In contrast, 90% of patients with recurrences at 3 months also had additional recurrences.

Lastly, tumor appearance or morphology can also be prognostic, with sessile (solid) masses having a higher recurrence rate than papillary masses.[9]

Pathologic Findings

Pathologic characteristics of superficial bladder tumors that have the greatest prognostic significance are tumor stage and grade. The most commonly used staging system for transitional cell carcinoma is the tumor, node, metastasis (TNM) classification, which has recently been updated (Table 1).[10] It has been documented that recurrence and progression rates for Tis, Ta, and T1 lesions vary greatly. However, most of the studies reporting on the prognostic value of tumor stage and grade in superficial transitional cell carcinoma have follow-up periods of only 5 years or less.[9-12] Table 2 lists a few of the studies that have followed patients for more than 5 years.[8,13,14]

Clinical Studies With Long-Term Follow-up—Heney and colleagues in the National Bladder Cancer Group studied 207 patients with superficial tumors who were followed for a mean of 39 months.[8] Their findings, which are similar to those of other researchers, confirm that higher-grade and higher-stage lesions recur and progress more often. Patients with grade 1, 2, or 3 disease had progression rates to muscle invasive disease of 2%, 11%, and 45%, respectively.

In terms of stage, patients with Ta lesions progressed only 4% of the time, as compared with 30% of individuals with T1 tumors. Rates of progression were also higher in patients who had dysplasia elsewhere in the bladder at the time of initial resection. Lamina propria invasion was a significant prognostic factor when combined with grade. Progression was observed in only 6% of patients with grade 2, Ta tumors, as compared with 21% of those with grade 2, T1 lesions.

Holmang et al followed 176 patients with primary Ta or T1 tumors for 20 years.[13] All 176 patients were treated initially with transurethral resection (TUR), while 65 patients required either cystectomy or radiation therapy immediately or within a year after TUR. Therefore, the results of the study need to be interpreted with some caution.

Overall, the study showed that there were frequent (52%) tumor recurrences, which were confined to the bladder, with a 14% progression rate for Ta lesions and 39% for T1 lesions. The overall risk of dying from superficial bladder cancer was 22% (11% for Ta lesions and 30% for T1 tumors). Similar progression rates were reported by Kurth et al in the European Organization for Research and Treatment of Cancer genitourinary (EORTC-GU) study.[14]

Long-term studies have also been performed in “high-risk” populations, including patients with T1 lesions, multiple Ta lesions, or associated Tis. Herr et al reported the 10-year outcome of 221 such patients who had undergone multiple TURs and one or more cycles of intravesical bacillus Calmette-Guérin (BCG) therapy.[15] Tumor progression developed in 42% of patients by 10 years, and superficial tumor recurrences occurred in 49%. Most (82%) of these stage progressions or recurrences had occurred by 5 years. Of note, 18% of patients had a disease-free interval of 5 years or more prior to recurrence or invasion.[1]

Proposed Substaging of T1 Lesions—Some researchers have proposed that the current TNM clas-sification system should be modified to include substaging of T1 lesions into T1a and T1b.[16,17] It is felt that this substaging might improve prognostic information. As proposed, T1a lesions would not involve the muscularis mucosa of the lamina propria, whereas T1b lesions may invade into or through this layer.

Holmang et al studied 121 patients with T1 transitional cell carcinomas in whom they subclassified 90% of the lesions as either T1a or T1b.[16] The T1b tumors were more often grade 3 than the T1a tumors (79% vs 40%;

P < .001). The T1b tumors had a significantly higher overall rate of progression than the T1a lesions (53% vs 31%). The death rate from bladder cancer was nearly twice as high for patients with T1b disease than for those with T1a disease (44% vs 23%; P < .05).

Similar results have been reported by Hasui and coworkers.[17] Although this substaging system appears to provide prognostic information using TUR chips, it may be an impractical to use due to technical difficulties, such as the cautery effect, orientation, and lack of observer agreement.[2]