Poor prognostic value of lymphovascular invasion for pT1 urothelial carcinoma with squamous differentiation in bladder cancer

Lymphovascular invasion (LVI) is the primary and essential step in the systemic dissemination of cancer cells. The aim of our study was to assess the independent prognostic role of LVI for pT1 urothelial carcinoma with squamous differentiation in bladder cancer. We retrospectively analyzed the clinical and pathological information of 206 patients diagnosed pT1 urothelial carcinoma with squamous differentiation. Of the 206 patients, LVI was detected in 57 (27.6%) patients. The 5 year cancer specific survival (CSS) rates were 87.2% in LVI (−) and 52.4% in LVI (+) (p < 0.001). According to univariate analysis, tumor multiplicity, tumor size, recurrence and LVI were the prognostic factors associated with CSS. Additionally, tumor size and LVI significantly influenced the CSS in multivariate analysis. TURBT had shorter median CSS than RC in recurred patients with LVI (+). Our study suggested that LVI is an important predictor for survival of pT1 urothelial carcinoma with squamous differentiation. LVI positive status and tumor size ≥3 cm led to a higher risk of death. RC should be routinely performed in recurred LVI (+) bladder cancer patients of pT1 urothelial carcinoma with squamous differentiation.

guidelines 13 . The prognostic factors were assessed including age, gender, tumor grade, tumor multiplicity, tumor size, recurrence, and LVI. In order to eliminate interference of influencing factor of bladder tumor grading we redo the analysis in subgroup of low-grade and high-grade tumor with or without LVI. The prognostic implications of these factors on cancer specific survival (CSS) rates were analyzed. All demographic and pathological variables were queried. Variables were evaluated for inconsistencies and data integrity. The pathologic stage was based on the 2009 Union for International Cancer Control (UICC) TNM staging system 4 . Grade was based on the 2004 World Health Organization (WHO) grading system for non-invasive urothelial neoplasia 2 .
Pathology. All surgical specimens were submitted en bloc for pathological evaluation. Sectioning was performed on a case by case basis to provide adequate evaluation of grade and stage. Independent pathologic re-review of three representative slides from each patient was performed by two pathologists on all specimens to confirm reported pathologic findings and to confirm LVI status. The presence of intercellular bridges or keratinization was indicative of squamous differentiation. The presence of LVI in TURBT specimens was assessed using conventional hematoxylin and eosin (H&E) staining and immunohistochemical staining (IHC) markers against the lymphatic (D2-40) and vascular endothelium (CD 31) 14,15 . IHC assessment of LVI was performed on TURBT specimens of primary diagnosis. LVI was defined as the presence of the invasion of cancer cells into blood vessels or the lymphatic system or both and neoplastic cells within an endothelium-lined space.The criteria for diagnosing LVI did not change over the study period.
Statistical methods. Cancer specific survival was considered from the day of surgery to the day of bladder cancer specific death. The chi-squared test and Student's t -test were used to evaluate the association between categorical and continuous variables, respectively. The Kaplan-Meier method was used to calculate overall survival trends, and differences were assessed using the log-rank statistic. Univariate and multivariate Cox regression models were used to analyze overall survival after operation. All reported P values were two-sided, and a P value of ≤ 0.05 was considered to indicate statistical significance. Statistical analysis was performed with SPSS software (Version 22).

Pathology and immunohistochemistry. Squamous differentiation was observed and confirmed by
pathologists in all 206 cases. The component of tumor was considered to be squamous when intercellular bridges and/or keratinization were evident (Fig. 6). The tumors showed strands or nests of infiltrating tumor cells with large and medium sized nuclei, often with nucleolus, and a not clearly separated amphophilic or eosinophilic cytoplasmic background. Stained sections in H&E were used to evaluate the presence of LVI (Fig. 7), IHC staining of CD31, CD24, and D34 was then performed. IHC stain in these cases were positive for CD31, CD24 and CD34 (Fig. 8).

Discussion
McDonald and Thompson reported the value of LVI as a criterion to assess the severity of urothelial bladder tumors for the first time 16 . Attention on the clinical significance of LVI in bladder cancer is growing, and a number of recent evidences have enhanced the significance of LVI for urothelial carcinoma of the bladder. Some papers indicated that LVI was an independent and significant prognostic factor for disease-specific survival 17 . Canter et al. analyzed the data from 356 patients treated with radical cystectomy by univariate analysis which found that the presence of LVI was a risk for overall, cancer-specific and recurrence-free survival (p < 0.0001) 7 . Cho et al., who conducted retrospective analyses of 118 patients reported that LVI, as an independent prognostic factor of progression and metastasis in pT1 bladder cancer, was significantly associated with disease recurrence 18 . The result   was consistent with the study performed by Lopez and Angulo, in which multivariate analysis revealed that LVI was an independent prognostic factor in TURBT surgical specimens of T1 bladder cancer 19 . However, there are short of data on the significance of LVI in patients of TURBT with urothelial carcinoma with squamous differentiation in bladder cancer, especially pT1 urothelial carcinoma of bladder. Our study showed further evidence suggesting that LVI was a pathological variable that might play an important role as a prognostic indicator in patients with pT1 urothelial carcinoma with squamous differentiation in bladder cancer. In our study, the presence of LVI was an independent prognostic factor related with disease survival (P < 0.001). In addition, there are hardly any data on the survival of the TURBT or RC in recurred patients in LVI for pT1 urothelial carcinoma with squamous differentiation in bladder cancer. Our finding indicated that recurred  patients operated by TURBT had shorter median CSS duration than those operated by RC in LVI (+) (P = 0.025). However, no significant difference was observed between the two groups in LVI (−) (P = 0.466). According to the results, we suggest surgeons should operate RC routinely in recurred patients for pT1 urothelial carcinoma with squamous differentiation with LVI (+) in bladder cancer.   Shariat et al. conducted a retrospective review of 4257 radical cystectomy specimens and stratified the patients by LVI status and pathological stage 20 . Akdogan et al. demonstrated that ureteral UC had a higher recurrence rate and poorer survival rate than renal pelvic UC 21 . Our series found that LVI (HR 3.774, 95% CI 2.167-6.571, P < 0.001) was the independent prognostic factors associated with CSS in both univariate analysis and multivariate analyses. In addition, tumor multiplicity (HR 1.778, 95% CI 1.020-3.099, P = 0.042), tumor size (HR 1.936, 95% CI 1.103-3.399, P = 0.021), and recurrence (HR 1.988, 95% CI 1.130-3.496, P = 0.017) were the prognostic factors associated with CSS, too. However, in multivariate Cox proportional hazard analysis, only LVI (HR 4.806, 95% CI 2.550-9.055, P < 0.001) and tumor size (HR 2.942, 95% CI 1.557-5.562, P = 0.001) significantly influenced the CSS.
The presence of LVI in patients with newly diagnosed T1 urothelial carcinoma in bladder cancer is associated with decreased recurrence-free survival 18 . T1 urothelial carcinoma in bladder cancer accounts for about 30% of non-muscle-invasive bladder tumors, with varying degrees of aggressiveness and progression rates of up to 50% 22,23 . The lamina propria lying just beneath the epithelial lining is rich in lymphatic and blood vessels that allows for early lymphatic and hematogenous tumor spread 24 . In our study, the tumors showed strands or nests of infiltrating tumor cells with large and medium sized nuclei, often with nucleolus, and a not clearly separated amphophilic or eosinophilic cytoplasmic background. Stained sections in H&E were used to evaluate the presence of LVI. IHC stain in these cases were positive for CD31, CD24 and CD34.

Conclusions
LVI is an important predictor for survival for pT1 urothelial carcinoma with squamous differentiation in bladder cancer. LVI positive status and tumor size ≥ 3 cm led to a higher risk of death which led to a higher mortality rate. Surgeons should operate RC routinely in recurred patients for pT1 urothelial carcinoma with squamous differentiation with LVI (+) in bladder cancer.