Prognostic Factors in Bladder Cancer

abstract & commentary

Source: Zeiger K, et al. Br J Urology 1998;82:667-672.

The clinical course of superficial bladder tumors is variable and has been difficult to predict. Some have proposed that there are at least two distinct variants of this disease, a low-grade, benign form and a malignant, invasive form.1,2 However, there are examples of those that appear to be low grade, but are invasive, and there are high-grade lesions that do not seem to invade. To complicate things further, both low- and high-grade lesions have been found in the same bladder. In this report, factors that were ascertained at the time of endoscopic examination and microscopic review were related to clinical course and survival.

The study included review of 584 consecutive patients with newly diagnosed superficial bladder tumors that presented to Hvidovre Hospital (Copenhagen) from 1976 through 1984. Of these, 419 were men and the mean age was 68.3 years. The diagnosis comprised cystoscopy with biopsy, bimanual palpation on relaxation, voided urine cytology, and pre-selected site biopsies. Known risk factors, e.g., tumor size, histological grade, multiplicity, positive urine cytology, and dysplasia (as assessed by random or pre-selected site biopsies) were evaluated as predetermining factors for new occurrences and survival.

Of the 584 cases, 95% were transitional cell carcinomas (TCC), 42% of tumors were confined to the epithelium at diagnosis, and 62% were considered superficial. Of the TCC, 53% were well or moderately differentiated (grade 0-2) and 47% were undifferentiated or anaplastic (grade 3 or 4). Of the 32 tumors that were not TCC, all but four were invasive.

Treatments varied, as would be predicted by stage. Those with T0 or Ta lesions were treated by transurethral resection (TUR). T1 lesions were also, for the most part (71%), treated by TUR. However, 22% received radiation therapy (some of these after attempted TUR). Those with T2 tumors were either treated by TUR (41%) or radiation therapy (54%) and only three of the 61 patients had open bladder resections. Of the 157 patients with advanced cancers (T3, T4), radiation therapy was the primary treatment in 68%, nine (6%) underwent open resection, and 25% received palliative treatment only.

Invasion of the lamina propria was the most significant prognostic factor detected in multivariate analysis. While 14% of patients with Ta tumors had died from cancer after 15 years, 63% of the T1 tumors were eventually fatal, reaching the mortality of T2 tumors. Other significant prognostic factors included tumor size and, to a lesser extent, histologic grade. Multiplicity and concomitant epithelial changes, as assayed by voided urine cytology and pre-selected site biopsies, were relevant prognostic factors for Ta but not T1 tumors.


The goal of this study was to provide an assessment of the commonly used prognostic factors for bladder cancer as applied to low-grade or superficial lesions. Examining a large series of patients from one academic center treated fairly consistently for almost a decade and followed for up to 20 years, this retrospective review adds an important perspective. In such a series of patients with chronic disease, determining cause of death and attribution of the impact of underlying disease to that death can be problematic. Thus, Zeiger and colleagues were challenged in this analysis but did a credible job of eliminating confounding events in their survey. The findings that invasion of the lamina propria had greater predictive value than other features such as histologic grade, tumor multiplicity, and concomitant epithelial changes (as assayed by pre-selected site biopsies and voided urine cytology) may provide rationale for future investigators to develop more aggressive treatment strategies for patients discovered to have such lesions.


1. Anderstrom C, et al. J Urol 1980;124:23.

2. Spruck CH. Cancer Res 1994;54:784.