High-Risk Bladder Cancer Management
Bladder cancer caught early is often amenable to treatment, but of course late-stage and aggressive characteristics alter the prognosis for the worse. Understanding high-risk features, including aberrant histologic patterns, empowers the urologist to counsel and treat patients with greater precision.
Bladder cancer staging is intended to reflect the probability of cure with a specific therapy. We advocate that staging should go beyond the anatomic extent of disease and incorporate other factors indicative of the biology of the tumor, including variant histology and biomarkers.
High-risk features include ad-vanced stage (cT3b and higher), the presence of hydronephrosis, lymphovascular invasion, or variant histology such as micropapillary or small-cell patterns.
Lymphovascular invasion (LVI) is an independent predictor of worse outcomes. In a large study led by Quek et al, 702 patients with inva-sive urothelial carcinoma treated with resection alone had significantly reduced recurrence-free survival when LVI was present—42% at 10 years compared with 74% among those without LVI (J Urol. 2005;174:103-106). Overall 10-year survival also was markedly diminished among those with LVI.
The presence of LVI connotes a poor prognosis, equivalent to that of having a palpable tumor. In one study looking at neoadjuvant che-motherapy, patients with stage cT2 or lower with LVI had similar survival rates to those with cT3 stage.
Bladder cancer is a heterogeneous group of malignancies. So-called “variant histology” includes urothelial carcinoma with aberrant differentiation (squamous and/or glandular differentiation, small-cell carcinoma, sarcomatoid carcinoma and micropapillary carcinoma) and non-urothelial carcinoma (squamous cell carcinoma and adenocarcinoma).
Collectively, the presence of variant histology is a poor prognostic indicator in any patient with the disease. Data collected from more than 1,200 bladder cancer patients at M. D. Anderson Cancer Center in Houston revealed a median survival of 2.45 years for those with urothelial carcinoma (UC), compared with 1.76 years for patients with other cell types. Forms associated with the lowest median survival were small-cell (1.53 years), micropapillary (1.30 years), and sarcomatoid (1.29 years) carcinoma.
At M. D. Anderson, UC with squamous and/or glandular differentiation is the second most prevalent histological type after pure UC. It is associated with a higher risk than pure UC, but lower risk than other variant patterns. Compared with pure UC, mixed UC and small cell carcinoma are associated with a 28% and 72% increased mortality risk, respectively. Micropapillary and sarcomatoid cancers are associated with a 61% and 55% increased risk.
Variant histological bladder cancers tend to present at advanced stages. Histological type may be of greatest clinical importance among patients with resectable or organ-confined disease. Once the bladder tumor has bulky extravesical extension or metastases, the difference in outcome between UC and variant histology may not be as marked. In a recent multicenter study, Rogers et al reported on survival following radical cystectomy for urothelial or squamous cell carcinoma compared with other variant histologies (J Urol. 2006; 175:1987-1988).
They found the risk of death was significantly increased among patients with variant histology compared to those with UC or squamous-cell carcinoma of the bladder. Five-year post-surgical cancer-related survival was 18% and 68%, respectively. The effect was seen for both organ-confined and extravesical disease, but this was a cystectomy data set that did not include patients with unresectable disease.
Radical cystectomy is indicated for all resectable bladder cancers, and neoadjuvant therapy is warranted for patients with high-risk features such as variant histology. Preoperative chemotherapy has been shown to improve survival among patients with advanced bladder cancer and may narrow the gap between some variant histology and UC outcomes. At M. D. Anderson, trials are underway investigating perioperative approaches to treating such patients.
In two prospective trials at M.D. Anderson which enrolled a total of 147 patients, five-year disease-
free survival among patients with histologically variant tumors was 50%, compared with 68% among those with pure UC. Furthermore, down-staging was achieved in 45% of tumors with variant histology compared with 55% of UC tumors. For small-cell carcinoma in particular, neoadjuvant chemotherapy resulted in significantly improved cause-specific survival compared with treatment with initial resection (without chemotherapy).
In a study by Brown et al now in press, patients with sarcomatoid carcinoma treated with neoadjuvant chemotherapy demonstrated a higher rate of pT0 disease at cystectomy compared to those undergoing immediate cystectomy, but by multivariate analysis, neoadjuvant chemotherapy did not have an effect on survival.
The optimal treatment of micropapillary bladder carcinoma is unclear, but intravesical therapy should be discouraged. A retrospective study at our institution of patients with this variant revealed high rates of disease progression (67%) and metastasis (22%) among those treated with bacillus Calmette-Guerin and other intravesical therapies (J Urol.2006;175:881-885).
Immediate cystectomy was associ-ated with good survival rates—72% at 10 years. For micropapillary type tumors, neoadjuvant chemotherapy followed by surgery is the treatment of choice. We are currently validating this approach at M.D. Anderson.
Dr. Dinney is professor and chairman of the department of urology and professor in the department of cancer biology at