Estrogen Receptor Expression in Papillary Urothelial Carcinoma of the Bladder and Ovarian Transitional Cell Carcinoma

Archives of Pathology & Laboratory Medicine, Feb 2005 by Croft, Philip R, Lathrop, Sarah L, Feddersen, Richard M, Joste, Nancy E

Context.-Relatively little is known about estrogen receptor (ER) expression in papillary urothelial carcinoma (PUC) of the bladder. Greater understanding of this feature of PUCs could aid with the treatment and identification of the origin of metastases, particularly with relation to the morphologically similar entity of ovarian transitional cell carcinoma (TCC).

Objective.-To assess the presence of ERs in PUC of the bladder, its metastases, and ovarian TCC.

Design.-Formalin-fixed, paraffin-embedded archival tissue from 92 primary bladder PUCs, 11 PUC metastases, and 11 primary or metastatic ovarian TCCs was immunostained with a monoclonal antibody against the human ER β-molecule. The ER-positive and ER-negative tumors were compared by the patients' sex and age, tumor grade, and the presence or absence of invasion. Statistical analysis was performed on the PUC results, first defining a positive result as staining of at least 10% of nuclei and then repeated using any percentage of staining as a positive result.

Results.-By the 10% criterion, 11% of PUCs of the bladder were ER positive. Invasive PUCs were more likely to be ER positive (P = .10). Women with ER-positive PUCs were older than their male counterparts (P = .03). By the second criterion, 22% of all PUCs were ER positive, and both higher grade and the presence of invasion were significantly associated with ER expression (P = .004 and .01, respectively). All 11 PUC metastases were totally ER negative. Ten of the 11 ovarian TCC cases exhibited strong and diffuse ER expression.

Conclusion.-Depending on the criterion used, up to 22% of bladder PUCs were ER positive. Higher grade and the presence of invasion were significantly associated with ER expression in these bladder carcinomas. In contrast, almost all of the ovarian TCCs marked strongly for ERs, a characteristic that may help differentiate these lesions from PUCs metastatic to the ovary.

(Arch Pathol Lab Med. 2005;129:194-199)

Driven in part by questions regarding the pathophysiology of urinary incontinence, several studies performed during the past few decades have attempted to elucidate the distribution of steroid hormone receptors within the human lower urinary tract.1-8 Although the findings have not always been consistent,7 most investigators have identified at least small concentrations of estrogen receptors (ERs) in the trigone, posterior bladder neck, and proximal urethra in women1,3-5; although less numerous, studies of men have demonstrated ERs in the prostatic urethra but not in the bladder.6

In contrast, relatively little is known about the presence of ERs in carcinomas that arise in the lower urinary tract, in particular papillary urothelial carcinoma (PUC) of the bladder. A few studies9,10 have addressed this issue with frozen tissue; another study11 used paraffin-embedded tissue and immunohistochemical analysis. In the current study formalin-fixed, paraffin-embedded tissue sections of bladder PUCs and their metastases were evaluated for ER expression using immunohistochemical analysis. The results were analyzed for significant associations with the sex and age of the patients, tumor grade, and presence or absence of invasion. The ER status of primary and metastatic ovarian transitional cell carcinomas (TCCs) was also assessed to compare the characteristics of these 2 morphologically similar lesions.

MATERIALS AND METHODS

Patients and Specimens

Ninety-two patients with PUCs of the bladder treated from 1991 to 2001 at the University of New Mexico (UNM) Health Sciences Center were identified through a search of the UNM Department of Pathology surgical pathology computer database. In an effort to minimize the effect that any prior treatment for the carcinoma may have on hormone receptor status, only the initial patient specimen identified in the selected timeframe was used. Eleven specimens of metastatic PUC from 9 patients were identified in the same timeframe. Eleven cases of primary or metastatic ovarian TCCs from 5 patients (2 primary ovarian TCCs and the following sites of metastasis: colon [3], omentum [2], peritoneum, small intestine, uterus, and uterine cervix) were included for comparison with the results of the bladder cases. No cases of PUC metastatic to the ovaries were identified in the database in the same timeframe. The hematoxylin-eosin-stained slides of all 114 cases were reviewed, and a slide and paraffin block representative of the original diagnosis were selected for each. Where patients had multiple concurrent bladder specimens of differing histologic grades, the tissue that demonstrated the highest-grade lesion was selected.

All primary PUC cases were then organized by sex, grade, and presence or absence of invasion. Of the 92 primary bladder PUC specimens reviewed, 32 were from women and 60 were from men. The ages of the patients ranged from 30 to 93 years (mean, 65 years; median, 64 years). Most cases were originally graded using the World Health Organization (WHO) 1973 grading system; for the few cases that were evaluated using either the Bergkvist or the 1998 WHO/International Society of Urological Pathology classification schemes, the grades were translated into the WHO 1973 grading system. Twenty-one cases were grade I, 29 were grade II, and 42 were grade III. Invasive PUCs accounted for 49 of the 92 primary cases.