World J Urol DOI 10.1007/s00345-014-1472-5

ORIGINAL ARTICLE

Urothelial carcinoma in bladder diverticula: outcomes after radical cystectomy Brian Hu · Raj Satkunasivam · Anne Schuckman · Gus Miranda · Jie Cai · Siamak Daneshmand 

Received: 23 November 2014 / Accepted: 22 December 2014 © Springer-Verlag Berlin Heidelberg 2014

Abstract  Purpose  To characterize the clinical and pathologic staging of patients with urothelial carcinoma (UC) in bladder diverticula (BD) after undergoing radical cystectomy (RC) and determine the impact of UCBD on recurrence and survival. Methods  We reviewed our institutional database of patients who underwent RC for UC (1971–2009). Outcomes were compared between patients with and without UCBD. Kaplan–Meier curves estimated recurrence-free survival (RFS) and overall survival (OS). Multivariable Cox regression evaluated associations between UCBD and survival. Results  Inclusion criteria were met in 1991 patients. UCBD was seen in 77 (4 %) patients and occurred exclusively in men (mean age 68 ± 8.5 years). The highest pathologic stage tumor was found in the BD in 44 (57 %) of these patients. Pathologic upstaging was more common with UCBD compared with UC not in BD (48 vs. 39 %, p = 0.031). On univariate analysis, no differences in RFS or OS were observed comparing patients with or without UCBD stratified by clinically organ-confined (≤T2) and extravesical (>T2) disease. On multivariable analysis, the presence of UCBD was not associated with differences in RFS (HR 0.92, 95 % CI 0.59–1.42, p = 0.69) or OS (HR 0.98, 95 % CI 0.74–1.31, p = 0.92). Conclusion  Upstaging was common in patients with UCBD, observed in almost half of the patients. There were

B. Hu · R. Satkunasivam · A. Schuckman · G. Miranda · J. Cai · S. Daneshmand (*)  USC Institute of Urology, Norris Comprehensive Cancer Center, Keck School of Medicine University of Southern California, 1441 Eastlake Ave, Suite 7416, Los Angeles, CA 90033, USA e-mail: [email protected]

no differences in RFS or OS after RC when comparing patients with or without UC in a diverticulum stratified by stage. Keywords  Diverticulum · Urothelial carcinoma of the bladder · Radical cystectomy · Survival

Introduction Urothelial carcinoma in BD has an incidence of 0.8–10 % and can present a clinical challenge [1]. An outpouching in the bladder urothelium through the muscularis propria can complicate the delivery of local therapies, and lack of muscle-invasive disease in BD can make decisions on extirpation more difficult. Poor outcomes were initially reported in UCBD, leading many to advocate for early partial or RC in this population [1–4]. The reason for the worse survival seen in these studies is not known. Some feel the lack of a muscular layer leads to earlier cancer invasion and dissemination. Others hypothesize that biologic differences exist in UCBD as BD are associated with inflammatory conditions such as urinary stasis, infection, and calculi. Clinical factors, such as delays in diagnosis or inaccuracies in staging, could also contribute. Complicating the management of UCBD is the recent literature demonstrating the feasibility of conservative treatment strategies. Two studies utilized transurethral resections with or without intravesical therapy for patients with UC contained within BD [2, 5]. Though outcomes were favorable, a subset of patients in each study still progressed and died of disease. This underscore the need to better understand UCBD as potential differences in biology and staging could contribute to adverse outcomes, even in organ-confined disease.

13



World J Urol

One fundamental question persists, which is whether UCBD, in and of itself, represents a negative prognostic factor. Thus far, studies have been small, making it difficult to adjust for other clinical variables. Additionally, most studies included patients undergoing various treatments, which further confounds comparisons. For these reasons, we analyzed our institutional database for outcomes in patients with UCBD treated with RC, representing the largest study to date in this population.

RC stratified by clinical stage: organ-confined (≤T2) or extravesical (>T2). Log-rank tests were used to compare the differences in RFS and OS in subgroups. Through stepwise selection, Cox proportional hazard models were used to evaluate the independent prognostic factors in the multivariable setting. A subset univariate and multivariable analysis was performed for patients with bladder cancer that had the highest pathologic stage tumor in the BD. All analyses were performed with SAS software, version 9.3 (SAS Institute Inc., Cary, NC).

Materials and methods Results Our institution prospectively maintains demographic, clinical, and pathologic data for all patients undergoing RC for bladder cancer under Institutional Review Board approval. Inclusion criteria included patients who underwent RC for curative intent for primary UC between 1971 and 2009. Clinical and pathologic variables were obtained. Clinical staging was based upon transurethral resection, bimanual examination, and radiographic studies. Data regarding resection of tumor within the diverticulum were not available as a separate variable. The presence of BD was determined by pathologic reporting from experienced genitourinary pathologists. Tumor upstaging was defined as an increase from the highest clinical stage to pathologic stage. Ta and Tis were considered the equivalent, and substaging (e.g., cT2a, cT2b) was not utilized. Lymph node upstaging was defined as clinically negative but pathologically positive nodes. The distribution of categorical demographic and clinical characteristics between patients with and without UCBD was compared with the Pearson’s Chi-square or Fisher’s exact test. The Kruskal–Wallis test was used to test differences in not normally distributed continuous variables between groups. Kaplan–Meier analyses were used to estimate the probabilities of RFS and OS every year after

A total of 2,642 patients underwent RC, of which 1991 (75 %) met inclusion criteria. The median follow-up for the UCBD group (10.3 years) was comparable to that of patients without UCBD (12.9 years, p  = 0.91). A total of 77 patients (4 %) had UCBD. Of these, 44 (57 %) had the highest pathologic stage tumor within the BD. The remainder of UCBD (n = 33) were found in association with separate, more pathologically advanced tumors. Clinical and pathologic characteristics are shown in Table 1. Clinical (Fig. 1a) and pathologic (Fig. 1b) stage distributions were significantly different between the groups (p  = 0.029 and p  = 0.0053, respectively). There were 32 cases of clinical T2 disease in patients with UCBD. Of these, 15 (47 %) were in patients with the UCBD, representing the worst pathologic stage tumor. There were 16 cases of pathologic T2 disease in patients with UCBD. Of these, only two were in patients with the UCBD, representing the worst pathologic stage tumor. Figure 1c demonstrates the incidence of upstaging with stratification by clinical T stage and node status. Upstaging of the primary tumor was seen in 48 % of cases of UCBD compared with 39 % in cases without UCBD (p  = 0.031). Lymph node upstaging was 21 % for both groups (p = 0.67).

Table 1  Clinical and pathologic characteristics

Bold values indicate statistical significance (p T2 2.34 (1.96–2.81)

Urothelial carcinoma in bladder diverticula: outcomes after radical cystectomy.

To characterize the clinical and pathologic staging of patients with urothelial carcinoma (UC) in bladder diverticula (BD) after undergoing radical cy...
414KB Sizes 1 Downloads 8 Views