Investigation of androgen receptor expression in urothelial carcinoma of the bladder Authors: André Luís Nunes Avelino, Felipe Dias Gonçalves, Jhonas Nathanael Menezes Ferreira, Gislane Rocha Vasconcelos, Conceição Aparecida Dornelas Published: 2025 ABSTRACT BackgroundStudies have demonstrated the importance of the androgen receptor (AR) in the development of bladder cancer and have proposed it as a potential biomarker or therapeutic hypothesis. However, the relationship between AR expression and both clinical features of patients and intrinsic tumor characteristics remains uncertain.AimsTo evaluate the association between AR immunohistochemical expression and clinical as well as pathological characteristics of bladder cancer.MethodsWe conducted a retrospective cohort including patients diagnosed with bladder cancer through transurethral resection of the bladder (TURB) between 2016 and 2021. For each case, AR immunoexpression was tested and analyzed in relation to sex, age, smoking history, number of tumor foci, tumor size, vascular invasion, histological grade, carcinoma in situ, and pathological stage.ResultsLow AR expression was observed in 97.5% of cases, while high expression was identified in 1.4% and indeterminate cases accounted for 1.1% of the cohort. AR immunoexpression was not related to clinical or pathological variables.ConclusionNo significant associations were identified between AR expression and clinical or pathological variables, suggesting that its potential role as a biomarker in bladder cancer warrants further investigation.Clinical trial numberNot applicable. Background Bladder cancer is a common malignancy, and both its incidence and mortality rates are expected to continue increasing in the coming years. Notably, the number of new cases is approximately three times higher in men compared with women [1]. The development of this neoplasm is influenced by a combination of social, environmental, and genetic factors. Notable risk factors include smoking (the main risk factor), family history [2], diets rich in processed foods [3], infection with schistosomiasis, use of medications such as cyclophosphamide and angiotensin II receptor blockers [2], occupational exposure to carcinogens such as arsenic and aromatic amines [2, 4], as well as advanced age [5]. There is also evidence that alterations in the urinary microbiota - triggered by hormonal, inflammatory, or dietary factors - may promote a pro-inflammatory environment conducive to the formation of carcinogenic substances [2]. Figure 1 illustrates the main risk factors involved in bladder carcinogenesis. Fig. 1Bladder cancer risk factor. Illustration of the main risk factors for bladder cancer. Source: Created with BioRender [6] Bladder cancer risk factor. Illustration of the main risk factors for bladder cancer. Source: Created with BioRender [6] Smoking and occupational exposure to carcinogens, both more frequent among men, partially explain the higher incidence of bladder cancer in males [7]. Nevertheless, hormonal influence, particularly that of androgens, cannot be disregarded. The androgen receptor (AR) acts as a transcription factor that modulates the biological effects of androgens across several physiological and pathological processes [8]. The relationship between male sex hormones, AR signaling, and bladder cancer has been supported by studies in rats, for example, exposed to the bladder carcinogen N-butyl-N-(4-hydroxybutyl) nitrosamine (BBN). In these cases, reduced incidence of bladder cancer was observed in male animals subjected to anti-androgenic therapy (surgical castration, flutamide, or LH-RH agonists) [9], as well as in AR knockout mice [10]. Miyamoto et al. also demonstrated that supplementation with the androgen dihydrotestosterone (DHT) increased the proliferation of human bladder cancer cell lines in vitro [10]. Several mechanisms have been proposed to explain the role of AR in bladder carcinogenesis, including activation of proliferative pathways such as EGFR, ERBB2, and MAPK; suppression of cell cycle inhibitory genes such as p21, p27, and PTEN; upregulation of anti-apoptotic proteins such as BCL-2, promoting cell survival; and downregulation of UDP-glucuronosyltransferases, enzymes responsible for metabolizing and detoxifying carcinogens, leading to accumulation of mutagenic compounds in the bladder. In addition, AR signaling has been associated with CD8 + T-cell exhaustion and reduced natural killer (NK) cell activity, which weakens the immune response against tumor cells [11]. Nevertheless, AR may also counteract neoplastic development by inhibiting the expression of CD44, a gene associated with disease progression [12]. In addition, AR can suppress PD-L1 expression in bladder cancer cells, thereby enhancing the CD8 + T-cell–mediated antitumor immune response [13]. Consequently, low AR expression has been associated with high PD-L1 levels and an increased risk of tumor progression and recurrence [14]. The purpose of this study was to investigate the relationship between AR expression and the clinical as well as pathological characteristics of patients with urothelial carcinoma of the bladder. Methods Study design This was a retrospective, observacional study conducted at Haroldo Juaçaba Hospital/Ceará Cancer Institute, approved by the Institutional Research Ethics Committee under protocol number 73234723.1.0000.5528. Between January 2016 and December 2021, patients with urothelial carcinoma of the bladder who underwent transurethral resection (TURB) with complete macroscopic excision and documented clinical and pathological data were included. Cases with non-neoplastic diagnoses or tumors at other sites were excluded. AR immunohistochemistry was then performed and correlated with clinical and histological variables, including age, sex, smoking history, tumor multifocality, tumor size, vascular invasion, histological grade, pathological T stage, and presence of carcinoma in situ. Most patients with non–muscle-invasive tumors (pTis, pTa, or pT1) received intravesical BCG therapy. In cases of intolerance or BCG unavailability, gemcitabine was administered as an alternative. Patients with muscle-invasive disease (≥ pT2) were primarily treated with radical cystectomy. Treatment regimens were not analyzed in detail, as therapeutic information was beyond the scope of this study. No experimental intervention or additional molecular analysis was performed, as the study relied on institutional resources available at the time. All data were anonymized before statistical analysis. Immunohistochemistry A pathologist blinded to patient clinical data reviewed hematoxylin and eosin (H&E)-stained slides to identify and mark tumor areas in the corresponding paraffin blocks. For tissue microarray (TMA) construction, two circular cores of 1.0 mm in diameter were extracted from each case. Immunohistochemistrye. AR expression was assessed using immunohistochemistry with a prediluted antibody (AR, clone SP107, Cell Marque, Sigma-Aldrich®) on the Ventana BenchmarkXT automated system (Ventana Medical Systems®). The incubation time was 32 minutes, and antigen retrieval was carried out at pH 8.0. Slides were counterstained with hematoxylin, washed, dehydrated, and mounted. Immunohistochemical scoring The score was calculated by multiplying the percentage of tumor cells showing nuclear staining (0% = 0; 1–10% = 1; 11–50% = 2; 51–80% = 3; 81–100% = 4) by the staining intensity (0 = negative; 1 = weak; 2 = moderate; 3 = strong). Final scores were classified as: 0–1 = negative; 2–4 = weak positivity; 6–8 = moderate positivity; 9–12 = strong positivity [15]. Data classification Low AR expression was defined as negative or weak positivity, while high AR expression included moderate or strong positivity. Cases with fewer than 100 viable cells were considered indeterminate [16]. Slides were independently reviewed by two pathologists blinded to patient data. Discrepant cases were reviewed. Positive and negative internal controls were used to validate staining. Statistical analysis Comparisons between groups were performed using the chi-square and Fisher’s exact tests for categorical variables and the Kruskal-Wallis test for nonparametric continuous variables. Bonferroni correction was applied for multiple testing. Analyses were conducted using R® software and Microsoft Excel 2016®, with statistical significance set at 5%. Results After applying exclusion criteria, 281 patients were included, of whom 67% were men and 33% women, with a mean age of 68 years. Table 1 presents the clinical and pathological characteristics of the patients included in the study. Table 1– Clinical and pathological data of patients subjected to TURBCharacteristicsN = 2811Age > 65181 (64%) ≤ 65100 (36%)Sex F92 (33%) M189 (67%) Smoking history180 (64%)Foci > 161 (22%) 1207 (74%) Not evaluated13 (4.6%)Size < 3.0 cm97 (35%) ≥ 3.0 cm135 (48%) Not evaluated49 (17%) Vascular invasion10 (3.6%)Grade High171 (61%) Low110 (39%)T stage in situ3 (1.1%) pTa91 (32%) pT1131 (47%) pT256 (20%) Concomitant in situ carcinoma5 (1.8%)Legend: 1n (%); Mean ± Standard Deviation (Median) Table 1– Clinical and pathological data of patients subjected to TURBCharacteristicsN = 2811Age > 65181 (64%) ≤ 65100 (36%)Sex F92 (33%) M189 (67%) Smoking history180 (64%)Foci > 161 (22%) 1207 (74%) Not evaluated13 (4.6%)Size < 3.0 cm97 (35%) ≥ 3.0 cm135 (48%) Not evaluated49 (17%) Vascular invasion10 (3.6%)Grade High171 (61%) Low110 (39%)T stage in situ3 (1.1%) pTa91 (32%) pT1131 (47%) pT256 (20%) Concomitant in situ carcinoma5 (1.8%)Legend: 1n (%); Mean ± Standard Deviation (Median) – Clinical and pathological data of patients subjected to TURB Legend: 1n (%); Mean ± Standard Deviation (Median) Low AR expression was observed in the majority of cases, 274 individuals (97.5%), while high expression was identified in 4 (1.4%) and indeterminate cases accounted for 3 (1.1%) of the cohort. Figure 2 shows the range of androgen receptor (AR) immunohistochemical staining patterns observed in bladder cancer. Fig. 2Photomicrograph of AR Immunostaining. Immunohistochemical expression of urothelial carcinoma of the bladder, illustrating different levels of AR expression. A and B: low AR expression (negative and weak, respectively). C and D: high AR expression (moderate and strong, respectively) (Immunohistochemistry, 400× magnification) Photomicrograph of AR Immunostaining. Immunohistochemical expression of urothelial carcinoma of the bladder, illustrating different levels of AR expression. A and B: low AR expression (negative and weak, respectively). C and D: high AR expression (moderate and strong, respectively) (Immunohistochemistry, 400× magnification) Smoking history was associated with high AR expression (p = 0.035); however, after Bonferroni correction, the association lost statistical significance (q = 0.953). Table 2 details the relationship between all clinicopathological variables analyzed and AR immunohistochemical expression. Table 2Association between AR expression and clinical-pathological featuresAR scoreVariablesTotal1High1Low1Indeterminate1p 2valueq3valueTotal28142743Age0.332>0.999 >65181 (64%)4 (100%)175 (64%)2 (67%) ≤65100 (36%)0 (0%)99 (36%)1 (33%)Sex0.690>0.999 F92 (33%)1 (25%)91 (33%)0 (0%) M189 (67%)3 (75%)183 (67%)3 (100%)History of smoking0.0350.953 Yes180 (64%)2 (50%)178 (65%)0 (0%) No60 (21%)2 (50%)57 (21%)1 (33%) Not evaluable41 (15%)0 (0%)39 (14%)2 (67%)Focus0.737>0.999 >161 (22%)0 (0%)61 (22%)0 (0%) 1207 (74%)4 (100%)200 (73%)3 (100%) Not evaluable13 (4.6%)0 (0%)13 (4.7%)0 (0%)Size0.854>0.999 <3.0 cm97 (35%)2 (50%)94 (34%)1 (33%) ≥3.0 cm135 (48%)2 (50%)132 (48%)1 (33%) Not evaluable49 (17%)0 (0%)48 (18%)1 (33%) Vascular invasion10 (3.6%)0 (0%)10 (3.6%)0 (0%)>0.999>0.999Degree0.281>0.999 High171 (61%)4 (100%)165 (60%)2 (67%) Low110 (39%)0 (0%)109 (40%)1 (33%)Stage T0.801>0.999 in situ3 (1.1%)0 (0%)3 (1.1%)0 (0%) pTa91 (32%)1 (25%)89 (32%)1 (33%) pT1131 (47%)3 (75%)127 (46%)1 (33%) pT256 (20%)0 (0%)55 (20%)1 (33%) Carcinoma in situ associated5 (1.8%)0 (0%)5 (1.8%)0 (0%)>0.999>0.9991n (%);2For the p-value,Fisher's exact test and the Kruskal-Wallis test were used3Bonferroni correction for multiple testing Table 2Association between AR expression and clinical-pathological featuresAR scoreVariablesTotal1High1Low1Indeterminate1p 2valueq3valueTotal28142743Age0.332>0.999 >65181 (64%)4 (100%)175 (64%)2 (67%) ≤65100 (36%)0 (0%)99 (36%)1 (33%)Sex0.690>0.999 F92 (33%)1 (25%)91 (33%)0 (0%) M189 (67%)3 (75%)183 (67%)3 (100%)History of smoking0.0350.953 Yes180 (64%)2 (50%)178 (65%)0 (0%) No60 (21%)2 (50%)57 (21%)1 (33%) Not evaluable41 (15%)0 (0%)39 (14%)2 (67%)Focus0.737>0.999 >161 (22%)0 (0%)61 (22%)0 (0%) 1207 (74%)4 (100%)200 (73%)3 (100%) Not evaluable13 (4.6%)0 (0%)13 (4.7%)0 (0%)Size0.854>0.999 <3.0 cm97 (35%)2 (50%)94 (34%)1 (33%) ≥3.0 cm135 (48%)2 (50%)132 (48%)1 (33%) Not evaluable49 (17%)0 (0%)48 (18%)1 (33%) Vascular invasion10 (3.6%)0 (0%)10 (3.6%)0 (0%)>0.999>0.999Degree0.281>0.999 High171 (61%)4 (100%)165 (60%)2 (67%) Low110 (39%)0 (0%)109 (40%)1 (33%)Stage T0.801>0.999 in situ3 (1.1%)0 (0%)3 (1.1%)0 (0%) pTa91 (32%)1 (25%)89 (32%)1 (33%) pT1131 (47%)3 (75%)127 (46%)1 (33%) pT256 (20%)0 (0%)55 (20%)1 (33%) Carcinoma in situ associated5 (1.8%)0 (0%)5 (1.8%)0 (0%)>0.999>0.9991n (%);2For the p-value,Fisher's exact test and the Kruskal-Wallis test were used3Bonferroni correction for multiple testing Association between AR expression and clinical-pathological features 1n (%); 2For the p-value,Fisher's exact test and the Kruskal-Wallis test were used 3Bonferroni correction for multiple testing Discussion The development of bladder cancer is a complex and multifactorial process driven by multiple determinants [2, 4, 5, 17]. Recent studies have explored the possible role of AR signaling in bladder carcinogenesis. However, the association between AR immunohistochemical expression and clinicopathological features, as well as patient prognosis, remains controversial [17, 18]. This study offers significant insights into this aspect. It was observed that high AR expression was observed in only 1.4% of patients, below the 13–63% range reported in the literature [19, 20]. These findings suggest that AR expression frequency may vary depending on the population studied and the methodology applied. No significant association was found between AR expression and patient age, in line with most previously published studies [16, 20–23]. Likewise, no sex-related differences were observed, consistent with most published data, underscoring the importance of non-hormonal pathways in the development of bladder cancer [16, 21–24]. Nevertheless, a meta-analysis including more than 2,000 participants reported higher AR expression in men [25]. Clinicopathological variables showed no significant association with AR expression, including tumor multifocality, tumor size, vascular invasion, concomitant carcinoma in situ and smoking history, consistent with previous studies [16, 21, 22, 24]. No correlation was found between AR expression and tumor grade or pathological stage, consistent with some studies [24, 26, 27]. However, most studies indicate decreased AR expression with increasing tumor grade and invasiveness [15, 23, 25, 28]. In contrast, some investigations have reported slightly increased AR positivity in high-grade and/or muscle-invasive bladder tumors [21, 22], underscoring the lack of uniformity in this aspect. The variability in results across studies may be attributed to methodological differences, antibody selection, tumor stage variation, or even biological heterogeneity within the same tumor type or across populations. The association between AR and bladder carcinogenesis suggests that the receptor may serve as a potential therapeutic target. In this aspect, the use of AR antagonists such as enzalutamide, flutamide, and bicalutamide has been shown in vitro to reduce the growth of AR-positive tumor cells [29]. There is also preclinical evidence that anti-androgenic drugs, through AR inhibition, can induce PD-L1 expression in tumor cells, thereby enhancing tumor sensitivity to PD-1/PD-L1 blockade immunotherapy [13]. In humans, androgen suppression—whether through androgen deprivation therapy or 5α-reductase inhibitors—has been associated with a reduced risk of developing bladder neoplasia or recurrence [11]. Although studies using anti-androgen therapy or immunotherapy in bladder cancer remain limited and there is no consensus on their routine use, current evidence suggests that AR modulation may represent a promising avenue for the development of novel therapeutic strategies against this malignancy [11]. The retrospective and observational nature of the study imposes inherent limitations. The analysis was restricted to data obtained from clinical and pathological records. Additional molecular validation, functional assays or in vitro drug–target tests could not be performed due to limited laboratory infrastructure and funding availability at the institution. Nevertheless, these constraints do not diminish the relevance of the present clinical–pathological correlations, which provide hypothesis-generating insights for future prospective investigations. Other limitations of this study include incomplete clinical data for some patients (e.g., tumor size), and the small number of cases with high AR expression. Conclusion AR immunohistochemical expression showed no significant association with clinical or pathological variables. Prospective and molecular studies are required to clarify these biological interactions and to determine whether AR modulation could have therapeutic relevance.