Clinical Outcomes of Patients With Variant Histology of Urothelial Carcinoma After Radical Cystectomy

Article information

J Urol Oncol. 2023;21(3):249-256

Publication date (electronic) : 2023 November 30

doi : https://doi.org/10.22465/juo.234600400020

Dan Bee Lee, Jae Yeon Kim, Yun Ha Lee, Won Hoon Song, Seung Soo Lee, Sung Woo Park, Jong Kil Nam

Department of Urology, Pusan National University Yangsan Hospital, Pusan National University School of Medicine, Yangsan, Korea

Corresponding author: Jong Kil Nam Department of Urology, Pusan National University Yangsan Hospital, Pusan National University School of Medicine, 20, Geumo-ro, Mulgeum-eup, Yangsan-si, Gyeongsangnam-do 50612, Korea Email: tuff-kil@hanmail.net

Received 2023 August 4; Revised 2023 August 25; Accepted 2023 September 5.

Abstract

INTRODUCTION

Bladder cancer (BC) ranks as the 10th most prevalent cancer type globally [1]. The most common pathological type is pure urothelial carcinoma (PUC), but this form of carcinoma is known for its tendency towards divergent differentiation. As a result, these tumors often show variant histology (VH) in conjunction with urothelial histology [2]. The 2016 World Health Organization’s classification of the urothelial tract includes several histologic variants: micropapillary, plasmacytoid, sarcomatoid variants of urothelial carcinoma (UC), squamous cell neoplasms, glandular neoplasms, neuroendocrine tumors, and others [3]. These are divided into urothelial and nonurothelial variants. Urothelial variants demonstrate urothelial differentiation alongside other morphologies, while nonurothelial variants display distinct characteristics.

VH is associated with important prognostic and therapeutic implications. Several reports have described the presence of VH as a poor prognostic factor. Compared to PUC, VH has a worse prognosis due to more locally advanced disease and occult regional lymph node metastasis at the time of diagnosis [4]. Since VH often presents at a more advanced stage, it is important to estimate the prognosis after adjusting for stage [5]. Unfortunately, the clinical staging of BC is inadequate due to the relative inaccuracy of transurethral bladder tumor resection results. As such, reported findings are often inconsistent, and definitive data on the effect of the histological type on survival are currently lacking [6].

Although there are defined criteria for managing VH in BC, they are largely based on subgroup analyses, relatively small studies, and expert consensus [7]. In some cases, the risk associated with VH necessitates aggressive treatment, such as early radical cystectomy (RC). Histological types such as micropapillary, plasmacytoid, and sarcomatoid are considered for early RC due to their significant risk of progressing to muscle-invasive and potentially metastatic diseases [8-10]. At present, the early administration of neoadjuvant chemotherapy followed by local treatment (either cystectomy or radiotherapy) is recommended for patients with small cell carcinoma of the bladder [11]. However, the survival outcomes for squamous or glandular differentiation, nested variants, and other rare variants are comparable to those of PUC. Consequently, these are managed in the same manner as UC of the same stage [7].

Given the considerable impact of RC on postoperative complications and patient quality of life, it is essential to determine whether the presence of higher-risk VH justifies aggressive treatment with early RC. Our study evaluated the clinical outcomes and survival expectancies of post-cystectomy patients with VH based on their stratified risk profile.

MATERIALS AND METHODS

1. Data Collection

We identified 327 BC patients who had undergone RC at our institution from February 2010 through June 2021 (Fig. 1). This study retrospectively analyzed patients who were eligible for follow-up observation for at least 2 years after surgery. Patients with nonurothelial variants, such as pure squamous cell carcinoma, pure adenocarcinoma, and lymphoma, and any metastasis were excluded, resulting in a total of 299 patients. Patients were categorized by histology into PUC (n=244) and VH (n=55) groups using the 2016 World Health Organization classification. The differentiation patterns included in the study were squamous differentiation, glandular differentiation, lipoid variant, nested variant, tubular type differentiation, and giant cell differentiation variants of urothelial cancer, small cell carcinoma, micropapillary, plasmacytoid, and sarcomatoid variants of urothelial cancer. This study was approved by the Institutional Review Board of Pusan National University Yangsan Hospital (approval No. 05-2023-173). Anonymized and deidentified information was used for analyses; therefore, informed consent was not obtained.

Fig. 1.

Study design flow chart.

3. Statistical Analysis

We retrospectively evaluated the clinical data of 299 BC patients who were treated with RC. Patients were analyzed according to age, sex, stage, grade, administration of chemotherapy, and type urinary diversion (continent vs. incontinent). Patients with VH were stratified by the researchers into aggressive (n=35) and more aggressive (n=20) groups (Fig. 2) with reference to the 2022 NCCN (National Comprehensive Cancer Network) Guidelines for Bladder Cancer. The more aggressive histologic types included micropapillary, plasmacytoid, and sarcomatoid.

Fig. 2.

Histologic grouping of 299 patients who met the inclusion criteria.

Finally, multivariable Cox proportional hazards models were fitted to predict the clinicopathologic variables influencing overall survival (OS) and progression-free survival (PFS) rates. OS and PFS were estimated with the Kaplan-Meier method. Categorical data were compared using the chi-square test, and quantitative variables were compared using the t-test. All statistical analysis was performed using IBM SPSS Statistics ver. 27.0 (IBM Co., Armonk, NY, USA). A p-value <0.05 was considered statistically significant.

RESULTS

From 2010 to 2021, 327 patients with surgically resected BC were identified. Of these, 244 (74.61%) were found to have PUC, and 55 (16.82%) had VH. Squamous differentiation was the most common histological type observed (6.73%), followed by micropapillary variant (3.34%) and glandular differentiation (1.83%) (Table 1).

Table 1.

Patient distribution according to histologic type

Table 2 summarizes the clinicopathologic features of 299 patients. Our cohort included 266 men and 33 women. The median age at diagnosis was 69 years (range, 63–75 years). Age, sex, grade, and type of urinary diversion did not show statistical significance. Overall, 29.8%, 18.4%, 29.4%, and 22.4% of patients presented with pathologic T stages pT1, pT2, pT3,4 and any pT with a positive lymph node, respectively. The UC with VH group was more likely to show advanced T stage (pT3–4: 23.4% vs. 56.3%) and to be greater than or equal to pN1 or cM1 (21.3% vs. 27.3%). In conclusion, the presence of VH was significantly associated with an advanced pathologic tumor stage (p<0.001).

Table 2.

Patient characteristics

The rate of pathologic upstaging or downstaging relative to the clinical stage after RC was measured to evaluate the likelihood of VH being upstaged. Any increase from the initial cT stage and/or cN stage was defined as pathologic upstaging. Upstaging was recorded in 25.1% of patients. More aggressive VH showed the highest incidence of upstaging (58%), followed by less aggressive VH (39%) and PUC (22%) (Fig. 3).

Fig. 3.

The rate of upstaging and downstaging relative to clinical stage after radical cystectomy in patients with pure urothelial carcinoma (A), more aggressive variant (B), and less aggressive variant (C).

During a median follow-up of 35 months, 75 patients experienced disease progression (22.9%), and 49 (21.4%) died. The median time to progression was 18 months (range, 1–142 months). The 5-year PFS rates of PUC and UC with VH were 56.6% and 3.6%, respectively. Kaplan-Meier analysis was used to estimate OS and PFS according to the following categories: pure UC versus UC with VH (Fig. 4A, B) and pure UC versus UC with aggressive VH versus. UC with more aggressive VH (Fig. 4C, D). No statistically significant differences were identified between patients with PUC and those with UC with VH in terms of OS and PFS (Fig. 4A, B). Although the UC with VH group had a lower PFS rate than the PUC group, no statistical significance was found between the survival curves (Fig. 4B). However, UC with more aggressive VH demonstrated significant differences in OS (p=0.013) and PFS (p=0.002) (Fig. 4C, D).

Fig. 4.

Overall survival (A) and progression free survival (B) in patients with pure urothelial carcinoma vs. urothelial carcinoma with variant histology. Overall survival (C) and progression free survival (D) in patients with pure urothelial carcinoma versus urothelial carcinoma with less aggressive variant versus urothelial carcinoma with more aggressive variant.

Univariate and multivariate analyses identified clinicopathologic parameters associated with disease progression and mortality after RC. On multivariate analysis, the presence of more aggressive VH was significantly associated with both OS (odds ratio [OR], 2.07; 95% confidence interval [CI], 1.76–3.98; p=0.030) and PFS (OR, 2.79; 95% CI, 1.33–5.91; p=0.070) when compared to PUC (Table 3). However, the survival outcomes of UC with less aggressive VH were comparable to PUC: OS (OR, 1.17; 95% CI, 0.68–2.03; p=0.560) and PFS (OR, 0.90; 95% CI, 0.41–1.99; p=0.800). The odds of mortality were 1.94 times higher for patients with pathologic T stage higher than 2 than for patients with pathologic T stage lower than 2 (OR 1.94; 95% CI, 1.26–2.88; p=0.002). The odds of progression were 1.87 times higher for patients with stage>pT2 (OR, 1.87; 95% CI, 1.03–3.61; p<0.016). Incontinent urinary diversion was associated with OS (OR, 3.32; 95% CI, 2.04–5.44; p=0.013) and PFS (OR, 2.14; 95% CI, 1.11–4.12; p<0.009).

Table 3.

Overall and progression-free survival according to clinicopathologic variables

DISCUSSION

UC is known to manifest in diverse morphological variants. Approximately 25% of BC cases are thought to have VH. Previous studies have reported prevalence rates of 2%–5% for the micropapillary variant, 1%–3% for the plasmacytoid variant, and less than 1% for the sarcomatoid variant [13]. The incidence rates found in this study align with these figures, with rates of 4.35%, 1.22%, and 0.61% for the micropapillary, plasmacytoid, and sarcomatoid variants, respectively. These 3 variants have been suggested in prior studies to be associated with poorer oncologic outcomes than PUC [14,15]. Conversely, the survival rates for squamous differentiation, glandular differentiation, and other rare types have been found to be comparable to those of PUC [13].

However, studies have presented findings that challenge previous research. In particular, the outcomes for patients with T1 micropapillary BC have been contentious. Some studies suggest that upfront cystectomy can provide a survival benefit, while others indicate that bladder sparing is not necessarily inferior. These conflicting results may be attributed to selection bias, and thus are subject to debate [16,17].

As a result, definitive data regarding the impact of histology type on survival have not yet been established [18]. For instance, a meta-analysis conducted by Abufaraj et al. [19] found that the micropapillary variant did not correlate with lower recurrence-free, cancer-specific, or OS rates compared to those observed in individuals with PUC.

The presence of VH has significant implications for disease management. Currently, guidelines on managing BC with VH are primarily based on subgroup analyses, relatively small studies, and expert consensus. The definition of treatment strategies is vague, and recommendations often lean towards a conservative approach to minimize patient harm. Experts agree that VH in non–muscle-invasive bladder cancer (NMIBC) patients should be considered a high-risk feature. Despite the absence of robust evidence, early RC should be considered as an aggressive treatment option [20]. Typically, immediate cystectomy is recommended for micropapillary, sarcomatoid, and plasmacytoid variants [16,21,22]. Other variants, such as squamous differentiation, glandular differentiation, and nested variants, are treated in the same way as conventional UC. For NMIBC patients, several series have shown promising results with bladder preservation therapy in carefully selected patients [13]. Systemic chemotherapy followed by RC or radiotherapy is suggested as a standard treatment for small cell carcinomas [7]. However, there is no evidence supporting the use of chemotherapy for other types of VH [23].

Accurately predicting survivability is crucial for physicians when determining treatment plans. As such, we conducted a retrospective investigation into the clinical outcomes and survival rates of high-risk VH subtypes in RC specimens, aiming to highlight their increased clinical risk compared to PUC and aggressive variants. Generally, patients with VH exhibited higher T stages than those with PUC, indicating a higher prevalence of locally advanced disease. Furthermore, the presence of VH was associated with a high likelihood of pathologic upstaging, particularly in patients who had UC with a more aggressive VH type.

The Kaplan-Meier estimates revealed no significant difference in survival rates between the overall VH group and the PUC group (Fig. 4). In the multivariable analysis, both OS and PFS were similar between PUC and VH types. However, when the variant types were stratified into less aggressive and more aggressive VH, the results aligned with previous research findings. The more aggressive variant demonstrated poorer survival rates for both OS and PFS (Fig. 4). Furthermore, it was linked with a higher risk of progression and death compared to PUC in the multivariate analysis. These findings suggest that our prognostic analysis may be utilized to reinforce the current recommendations for managing UC in cases of BC with VH.

This study has some limitations. Due to its retrospective design, this study is subject to bias. The relatively small sample size and short follow-up period in the VH group compared to the PUC group may render our analysis underpowered, limiting the generalizability of the results on survival. A prospective study starting at transurethral resection with a follow-up of at least 5 years would better meet the objective of this study. Pure squamous cell carcinoma, adenocarcinoma, and small cell carcinoma were excluded from the study because they were not primary BCs and their inclusion could have confounded the results. However, in 2017, two patients with small cell carcinoma were treated with upfront cystectomy due to the lack of established treatment recommendations at that time. Patients in the aggressive VH group had a higher stage at presentation, which could have also contributed to biased results since the baseline demographics were not the same in all groups. Lastly, incontinent urinary diversion was performed on individuals with high morbidity or low-performance scores; this may have reduced their likelihood of survival, which in turn may have contributed to surgery type being identified as a predictor of survival in the multivariate analysis.

CONCLUSIONS

Our study provides a clinical risk stratification of histologic variants in patients with BC postcystectomy. Within the study cohort, 16.82% of the patients analyzed exhibited VH, with 34.5% of these belonging to the more aggressive variant type. Histologic variations such as squamous differentiation, glandular, nested variant, and others, predicted a prognosis similar to PUC. Conversely, the more aggressive variants were linked with pathological upstaging and a poor prognosis in both Kaplan-Meier and multivariate models. As such, it serves as an independent predictor of poor survival and recurrence following RC. Consistent with previous studies, our findings indicate that cystectomy specimens with a more aggressive type tend to have a worse prognosis in MIBC patients. We propose that analyzing histologic characteristics through transurethral bladder tumor resection could aid in therapeutic decision-making between upfront RC and chemotherapy, a topic we plan to explore in further research.

Article information Continued