Abstract
Background/Aim: Cross-talk between forkhead box O1 (FOXO1), a transcriptional factor known to function as a tumor suppressor via the PI3K/AKT pathway, and glucocorticoid receptor (GR) has been implied in non-urothelial cells. The present study aimed to investigate the association of FOXO1 and GR expression in bladder cancer and its prognostic significance.
Materials and Methods: Immunohistochemical staining for GR, FOXO1, and p-FOXO1 (a phosphorylated/inactivated form) was performed in a set of bladder cancer tissue microarray comprising 50 low-grade non-invasive tumors, 28 high-grade non-muscle-invasive tumors, and 51 high-grade muscle-invasive tumors. Western blotting for FOXO1 and p-FOXO1 was also conducted in human bladder cancer cells.
Results: GR expression was detected in 109 [84.5%; 39 (30.2%) weakly positive (1+), 39 (30.2%) moderately positive (2+), 31 (24.0%) strongly positive (3+)] tumors, whereas FOXO1 and p-FOXO1 were immunoreactive in 17 [13.2%; 16 (12.4%) 1+, 1 (0.8%) 2+] and 71 [55.0%; 57 (44.2%) 1+, 14 (10.9%) 2+] tumors, respectively. The expression levels of GR were positively and negatively correlated with those of FOXO1 (p=0.003) and p-FOXO1 (p=0.009), respectively. GR(0/1+)/FOXO1(0) and GR(0/1+)/p-FOXO1(1+/2+) were significantly more often observed in high-grade (vs. low-grade) or muscle-invasive (vs. non-muscle-invasive) tumors. Both univariate and multivariate analyses revealed that GR(0/1+)/FOXO1(0) and GR(0/1+)/p-FOXO1(1+/2+) were associated with a significantly higher risk for the recurrence of non-invasive disease or progression of muscle-invasive disease. In 2 GR-positive bladder cancer lines, glucocorticoids (i.e., dexamethasone, prednisone) and a GR antagonist (i.e., RU486) induced the levels of FOXO1 and p-FOXO1 expression, respectively.
Conclusion: The expression levels of GR and FOXO1 or p-FOXO1 were strongly correlated in bladder cancer. Specific GR/FOXO1 and GR/p-FOXO1 expression profiles served as independent predictors of disease recurrence or progression.
Introduction
Urinary bladder cancer, which histologically shows urothelial carcinoma in most cases, remains one of the most frequently diagnosed malignancies, with 613,791 new cases and 220,349 deaths estimated globally in 2022 (1). Moreover, the worldwide number of disease-related mortality appears to be considerably increasing [e.g., estimated 165,100 deaths in 2012 (2)]. Approximately 70% of patients are initially diagnosed with non-muscle-invasive bladder tumor but disease recurrence is relatively common after transurethral surgery and intravesical pharmacotherapy (3-7). Notably, a subset of patients presenting with pT1 disease and developing tumor progression even after intravesical BCG immunotherapy ultimately succumb to disease despite undergoing radical cystectomy (8-10). Meanwhile, patients with muscle-invasive bladder cancer face a high risk of progression even after receiving aggressive treatment including cystectomy combined with systemic cisplatin-based chemotherapy or targeted therapy administered in neoadjuvant and/or adjuvant setting(s) (5, 11-15). These challenges underscore the urgent need for new biomarkers to better predict disease recurrence and/or progression in patients with bladder cancer.
Previous studies have highlighted the oncogenic potential of certain steroid hormone receptors, such as androgen receptor and estrogen receptor-β, in promoting urothelial carcinogenesis and cancer growth (16-18). Conversely, treatment with a glucocorticoid, prednisone (PRED), or a synthetic glucocorticoid receptor (GR) ligand, compound A (CpdA), has been shown to strongly suppress the development of chemical carcinogen-induced bladder cancer in mice (19, 20), suggesting a potential tumor-suppressive role for GR. However, GR functions in bladder cancer progression appear to be complicated and context-dependent (21). Specifically, while CpdA reduces cell proliferation, dexamethasone (DEX) and other glucocorticoids (excluding corticosterone and PRED that show no significant effects) have been reported to rather enhance it (22-24). DEX has also been shown to promote tumor growth in mouse xenograft models (22, 25). Interestingly, despite these differential effects on cell proliferation, DEX, corticosterone, PRED, and CpdA were found to inhibit cell invasion (22-24). One of these studies suggested that the divergent effects on cell proliferation were dependent on the balance between GR-mediated transactivation and transrepression induced by each ligand (24). In addition, an isoform of GR to which glucocorticoids do not bind, GRβ, has been shown to promote the migration of bladder cancer cells (26).
Forkhead box O1 (FOXO1), a member of the forkhead transcription factor family, is known to involve various cellular functions including regulation of cell cycle progression and apoptosis (27-29). In cancer cells, FOXO1 activity is suppressed through phosphorylation via kinase pathways, such as PI3K/AKT, which leads to functional inactivation and subsequent induction of cell survival (28-31). FOXO1 has thus been regarded as a tumor suppressor. Indeed, we previously demonstrated that FOXO1 inhibition or knockdown promoted urothelial tumorigenesis and tumor growth (32), as well as resistance to cisplatin therapy in bladder cancer cells (33). Subsequent studies have suggested the involvement of the FOXO1 pathway in modulating bladder cancer progression via other signals, such as HnRNP-F (34), HYOU1 (35), miR-1247-3p (36), and SOX2 (37). Furthermore, down-regulated expression of FOXO1 or up-regulated expression of an inactivated form of FOXO1, phospho-FOXO1 (p-FOXO1), has been observed in bladder cancer specimens (32, 38, 39).
A previous study using skeletal muscle cells demonstrated that DEX and a GR antagonist mifepristone (RU486) induced and reduced, respectively, the levels of FOXO1 gene expression (40). This suggests a potential cross-talk between GR and FOXO1 signals. However, to the best of our knowledge, this interaction has not been explored in the context of bladder cancer. The present study was therefore designed to investigate the association between GR and FOXO1/p-FOXO1 expression in bladder cancer tissues and its clinical implications in relation to oncologic outcomes.
Materials and Methods
Tissue specimens. A set of previously constructed tissue microarray (TMA) composed of retrieved bladder tumor specimens obtained from transurethral surgery performed at The Johns Hopkins Hospital was utilized in the present study, with appropriate approval from the Institutional Review Board (41). These cases included 11 papillary urothelial neoplasms of low malignant potential, 39 non-invasive low-grade papillary urothelial carcinomas, 25 non-invasive high-grade papillary urothelial carcinomas, 3 high-grade urothelial carcinomas invading the lamina propria (pT1), and 51 muscle-invasive high-grade urothelial carcinomas. All 54 patients with invasive tumor, including 3 pT1 cases, eventually underwent radical cystectomy. None of these patients received anti-cancer therapy prior to tissue collection.
Immunohistochemistry. Immunohistochemical staining was performed on 5-μm TMA sections using a primary antibody to GR (clone H-300; dilution 1:200; Santa Cruz Biotechnology, Dallas, TX, USA), FOXO1 (clone C29H4; dilution 1:200; Cell Signaling Technology, Danvers, MA, USA), and p-FOXO1 (Ser 256; dilution 1:200; Sigma-Aldrich, St. Louis, MO, USA), and a broad-spectrum secondary antibody with Histostain®-SP kit (Invitrogen/Thermo Fisher Scientific, Waltham, MA, USA). Immunoreactivity was evaluated semi-quantitatively by multiplying the percentage of immunoreactive cells (1-10%=1; 11-50%=2; 51-80%=3; 81-100%=4) by the staining intensity (weak=1; moderate=2; strong=3). The final scores were categorized as negative (0; score 0-1), weakly positive (1+; score 2-4), moderately positive (2+; score 6-8), or strongly positive (3+; score 9-12).
Cell lines and chemicals. Human bladder urothelial carcinoma cell lines, 5637 and 647V, originally obtained from the American Type Culture Collection (Manassas, VA, USA) were cultured in Dulbecco’s Modified Eagle Medium (Thermo Fisher Scientific) supplemented with 10% fetal bovine serum and penicillin/streptomycin (100 units/ml each). All GR ligands, DEX, PRED, and RU486, were purchased from Sigma-Aldrich.
Western blotting. Total protein lysates (30 μg) from cultured cells were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membrane (Thermo Fisher Scientific) electronically. The membranes were blocked and incubated with a specific antibody [i.e., FOXO1 (dilution 1:500), p-FOXO1 (dilution 1:500), GAPDH (clone 6c5; dilution 1:5,000; Santa Cruz Biotechnology)] overnight at 4°C and a secondary antibody (i.e., anti-mouse or anti-rabbit IgG HRP-linked antibody; Cell Signaling Technology) for 1 h at room temperature. Protein bands were visualized using the Clarity Western ECL Substrate (Bio-Rad, Hercules, CA, USA) and imaged with the ChemiDoc™ MP system (Bio-Rad).
Statistical analyses. Statistical comparisons employed Fisher’s exact test for categorized variables and Student’s t-test for continuous variables. Correlations between variables were determined by the Spearman’s correlation coefficient (CC). Recurrence-free survival in patients with non-invasive tumor, as well as progression-free survival (progression defined as the development of local recurrence or metastasis after cystectomy) in patients with muscle-invasive tumor, was calculated by the Kaplan-Meier method, with group comparisons by the log-rank test. A Cox regression model was applied for multivariate analysis of prognostic factors. All statistical analyses were performed using SPSS (version 27; IBM Corp., Armonk, NY, USA) or Prism (version 10.5.0; GraphPad Software, Boston, MA, USA). A p-value of less than 0.05 was considered to indicate a statistically significant difference.
Results
Associations between GR and FOXO1/p-FOXO1 expression in surgical specimens. Figure 1 shows representative images of GR, FOXO1, and p-FOXO1 expression in bladder tumor specimens. Overall, GR was immunoreactive in 109 tumors [84.5%; 1+ in 39 (30.2%) cases, 2+ in 39 (30.2%) cases, and 3+ in 31 (24.0%) cases], whereas FOXO1 and its phosphorylated/inactivated form, p-FOXO1, were immunoreactive in 17 tumors [13.2%; 16 (12.4%) with 1+ and 1 (0.8%) with 2+] and 71 tumors [55.0%; 57 (44.2%) with 1+ and 14 (10.9%) with 2+], respectively (Table I). Statistical analysis revealed a significant positive correlation between the levels of GR and FOXO1 expression (CC=0.281, p=0.003) and a significant negative correlation between those of GR and p-FOXO1 expression (CC=−0.179, p=0.009). Moreover, high GR expression (i.e., 2+/3+) was strongly associated with positive (1+/2+) FOXO1 (p=0.003) or negative/weak (0/1+) p-FOXO1 (p=0.011).
Immunohistochemistry of GR, FOXO1, and p-FOXO1 in bladder cancer TMA. Original magnificatioń200. GR: Glucocorticoid receptor; FOXO1: forkhead box O1; p-FOXO1: phospho-FOXO1.
Correlations between the levels of GR and FOXO1/p-FOXO1 expression in bladder tumors.
Associations of GR/FOXO1 or GR/p-FOXO1 levels with tumor grade and stage. The cohort was divided into 4 groups based on the expression levels of GR and FOXO1: 1) GR(2+/3+)/FOXO1(1+/2+) (n=15); 2) GR(2+/3+)/FOXO1(0) (n=55); 3) GR(0/1+)/FOXO1(1+/2+) (n=2); and 4) GR(0/1+)/FOXO1(0) (n=57). GR(0/1+)/FOXO1(0) was significantly (p<0.001) more often seen in high-grade tumors (59.5%) than in low-grade tumors (20.0%), as well as in muscle-invasive tumors (70.6%) than in non-muscle-invasive tumors (26.9%) (Table II). Similarly, the cohort was divided into 4 groups based on the expression levels of GR and p-FOXO1: 1) GR(2+/3+)/p-FOXO1(0) (n=33); 2) GR(2+/3+)/p-FOXO1(1+/2+) (n=37); 3) GR(0/1+)/FOXO1(0) (n=25); and 4) GR(0/1+)/p-FOXO1(1+/2+) (n=34). GR(0/1+)/FOXO1(1+/2+) was significantly (p<0.001) more often seen in high-grade tumors (39.2%) than in low-grade tumors (6.0%), as well as in muscle-invasive tumors (43.1%) than in non-muscle-invasive tumors (15.4%) (Table III).
Association of GR/FOXO1 expression with tumor grade and stage.
Association of GR/p-FOXO1 expression with tumor grade and stage.
Associations of GR/FOXO1 or GR/p-FOXO1 levels with patient outcomes. Univariate survival analysis was conducted to assess the prognostic impact of GR/FOXO1 and GR/p-FOXO1 expression patterns. In 75 patients with non-invasive bladder tumor, GR(0/1+)/FOXO1(0) was associated with a higher risk of postoperative recurrence, compared with GR(2+/3+)/FOXO1(1+/2+) (p=0.004) or GR(2+/3+)/FPXP1(0) (p=0.075) (Figure 2A). When grouped, patients with GR(0/1+)/FOXO1(0) tumor had significantly worse recurrence-free survival, compared with all others (p=0.011; Figure 2B). Similarly, in these non-invasive patients, the risk of recurrence was significantly higher in GR(0/1+)/p-FOXO1(1+/2+) cases than in GR(2+/3+)/p-FOXO1(0) (p=0.007), GR(2+/3+)/p-FOXO1 (1+/2+) (p<0.001), or GR(0/1+)/p-FOXO1(0) (p=0.039) cases (Figure 2C). Thus, there was a significant difference in recurrence-free survival between GR(0/1+)/p-FOXO1(1+/2+) patients and all other patients grouped (p<0.001; Figure 2D). In 51 patients with muscle-invasive disease, progression-free survival after cystectomy was compared. Only 1 patient had GR(2+/3+)/FOXO1(1+/2+) tumor, while none had GR(0/1+)/FOXO1(1+/2+) tumor. There were significant differences in progression-free survival between GR(0/1+)/FOXO1(0) cases and GR(2+/3+)/FOXO1(0) cases (p=0.044; Figure 3A) or combined GR(2+/3+)/FOXO1(1+/2+) and GR(2+/3+)/FOXO1(0) cases (p=0.025; Figure 3B). A significant difference in progression-free survival was also observed between the GR(2+/3+)/p-FOXO1(0) and GR(0/1+)/p-FOXO1(1+/2+) groups (p=0.008), but not in any other 2 group comparisons (Figure 3C). GR(0/1+)/p-FOXO1 (1+/2+) tumors exhibited a significantly higher risk of progression, compared with all other tumors grouped (p=0.012; Figure 3D).
Recurrence-free survival in patients with non-invasive bladder tumor according to the levels of GR/FOXO1 expression (A, B) or GR/p-FOXO1 expression (C, D). GR: Glucocorticoid receptor; FOXO1: forkhead box O1; p-FOXO1: phospho-FOXO1.
Progression-free survival in patients with muscle-invasive bladder cancer according to the levels of GR/FOXO1 expression (A, B) or GR/p-FOXO1 expression (C, D). GR: Glucocorticoid receptor; FOXO1: forkhead box O1; p-FOXO1: phospho-FOXO1.
To determine whether the patterns of GR/FOXO1 or GR/p-FOXO1 expression could independently predict oncologic outcomes, multivariate Cox analysis of clinicopathologic factors was performed. In patients with non-invasive tumor, GR(0/1+)/FOXO1(0) [hazard ratio (HR)=2.355, 95% confidence interval (CI)=1.100-5.041, p=0.027] and GR(0/1+)/p-FOXO1(1+/2+) (HR=3.669, 95% CI=1.547-8.702, p=0.003) showed a significantly increased risk of recurrence (Table IV). In muscle-invasive cases, GR(0/1+)/FOXO1(0) (HR=2.984, 95% CI=1.130-7.880, p=0.027) and GR(0/1+)/p-FOXO1(1+/2+) (HR=2.522, 95% CI=1.195-5.322, p=0.015) showed a significantly increased risk of postoperative progression (Table V).
Univariate and multivariate analyses of the factors including GR/FOXO1 or GR/p-FOXO1 expression in cases with non-invasive bladder tumor.
Univariate and multivariate analyses of the factors including GR/FOXO1 or GR/p-FOXO1 expression in cases with muscle-invasive bladder tumor.
Effects of GR ligands on FOXO1/p-FOXO1 expression in bladder cancer lines. We further examined the regulatory relationship between GR activity and FOXO1/p-FOXO1 expression in human bladder cancer cell lines. Western blot analysis in 2 cell lines, which were known to express the GR (24), demonstrated that glucocorticoids, DEX and PRED, increased the levels of FOXO1 expression in both lines and a GR antagonist RU486 reduced those at least in 5637 cells (Figure 4A). Moreover, RU486 at least partially blocked the DEX- or PRED-mediated induction of FOXO1 expression. RU486 treatment also led to increased levels of p-FOXO1 expression in these lines (Figure 4B).
Associations between GR and FOXO1 signals in bladder cancer cells. Representative western blotting of FOXO1 (A) or p-FOXO1 (B) in 5637 and 647V cells treated with ethanol (mock), 100 nM DEX, 10 nM PRED, and/or 1 μM RU486, as indicated. GAPDH served as a loading control. The densitometry values for FOXO1 and p-FOXO1, normalized by GAPDH and expressed relative to mock treatment, represent the means±standard deviations from 3 blots. *p<0.05 (vs. mock treatment). #p<0.05 (vs. DEX or PRED treatment only). GR: Glucocorticoid receptor; FOXO1: forkhead box O1; p-FOXO1: phospho-FOXO1; DEX: dexamethasone; PRED: prednisone; RU486: mifepristone; GAPDH: glyceraldehyde-3-phosphate dehydrogenase.
Discussion
In previous studies, the expression status of GR (25, 42, 43), as well as FOXO1 and/or p-FOXO1 (32, 38, 39, 44), has been separately determined in urothelial cancer tissues. Moreover, prior observations in non-urothelial cells have suggested that GR signaling may affect FOXO1 expression (40). Based on these insights, we investigated the relationship between the levels of GR and FOXO1/p-FOXO1 expression in bladder cancer, along with the prognostic significance of these expression patterns.
Prior immunohistochemical analyses demonstrated a significant down-regulation of GR (42) or FOXO1 (32) expression in bladder cancer tissues, compared with non-neoplastic urothelial cells, as well as a significant down-regulation of GR (42) or FOXO1 (32, 38, 39) expression in high-grade (vs. low-grade) or muscle-invasive (vs. non-muscle-invasive) bladder cancer specimens. In contrast, p-FOXO1 expression was reported to be significantly up-regulated in bladder cancer (vs. non-cancer) (32), high-grade bladder cancer (vs. low-grade tumor) (32), and muscle-invasive upper urinary tract urothelial carcinoma (vs. non-muscle-invasive cancer) (44). We confirmed these findings and further demonstrated that GR(0/1+)/FOXO1(0) was significantly more common in high-grade (59.5%) or muscle-invasive (70.6%) tumors than in low-grade (20.0%) or non-muscle-invasive (26.9%) tumors. Similarly, GR(0/1+)/p-FOXO1(1+/2+) was significantly more frequently observed in high-grade (39.2%) or muscle-invasive (43.1%) tumors than in low-grade (6.0%) or non-muscle-invasive (15.4%) tumors.
Survival analysis in previous immunohistochemical studies described above further showed significantly worse oncologic outcomes in patients with GR-negative/low non-muscle-invasive bladder tumor, GR-negative/low muscle-invasive bladder tumor (42), or FOXO1-negative non-muscle-invasive bladder tumor (32). Marginally worse progression-free survival was also reported in patients with FOXO1-low pTa/pT1/pT2 bladder cancer (38). Additionally, a study indicated that FOXO1 expression independently predicted overall survival in a mixed cohort of 276 bladder cancer patients showing various grades and stages (39). Remarkably, previous studies, except one demonstrating that FOXO1 positivity was independently associated with a reduced risk of recurrence in non-muscle-invasive disease (HR=0.128; p=0.043) (32), did not establish GR, FOXO1, or p-FOXO1 expression as a robust independent prognosticator in patients with urothelial tumor. Meanwhile, p-FOXO1 positivity in muscle-invasive tumors has been linked to cisplatin resistance (33). We herein found that GR(0/1+)/FOXO1(0) and GR(0/1+)/p-FOXO1(1+/2+) were significantly associated with an increased risk for the recurrence of non-muscle-invasive tumors after transurethral resection and the progression of muscle-invasive tumors after radical cystectomy. Notably, in both non-muscle-invasive and muscle-invasive cases, GR(0/1+)/FOXO1(0) and GR(0/1+)/p-FOXO1(1+/2+) were identified as independent predictors of poor prognosis.
As mentioned above, the functional interplay between GR and FOXO1 was suggested in skeletal muscle cells in rat models where sepsis-mediated expression of FOXO1 mRNA was induced and reduced by DEX and RU486, respectively (40). In bladder cancer tissues, we observed positive and negative correlations of GR overexpression with FOXO1 and p-FOXO1 expression, respectively. In bladder cancer cell lines, glucocorticoid treatment induced FOXO1 expression, whereas GR antagonism reduced and increased the expression levels of FOXO1 and p-FOXO1, respectively. These findings suggest GR-mediated inactivation of a tumor suppressor FOXO1 in bladder cancer. Although chromatin immunoprecipitation data previously indicated that another steroid hormone receptor, estrogen receptor-β, could bind directly to the promoter region of FOXO1 in bladder cancer cells (32), we did not detect GR binding to the FOXO1 promoter in our chromatin immunoprecipitation assay (data not shown). Therefore, GR may indirectly regulate FOXO1 expression in bladder cancer cells. Further investigation is thus required to delineate how glucocorticoid-mediated signaling modulates the expression and activity of FOXO1.
Conclusion
Our findings demonstrate strong correlations between the levels of GR and FOXO1/p-FOXO1 expression in bladder cancer. Specifically, GR down-regulation was associated with FOXO1 down-regulation and p-FOXO1 up-regulation, and these patterns were more frequently observed in high-grade muscle-invasive tumors. Furthermore, the expression profiles of GR/FOXO1 and GR/p-FOXO1 were found to independently predict oncologic outcomes in patients with bladder cancer, underscoring their potential utility as prognostic biomarkers. Additional studies are warranted to not only validate our observations but also clarify the molecular mechanisms underlying the functional interplay between GR and FOXO1 signals.
Footnotes
Conflicts of Interest
The Authors declare that they have no conflicts of interest interests or financial ties related to this study.
Authors’ Contributions
H.I. collected and analyzed data and drafted the manuscript. M.A.E.N, T.M., and T.T. collected data. M.O. supervised the study. H. M. conceived, designed, and supervised the study and analyzed data. All Authors read and approved the final manuscript for submission.
Artificial Intelligence (AI) Disclosure
No AI tools, including large language models or machine learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received July 21, 2025.
- Revision received August 15, 2025.
- Accepted August 25, 2025.
- Copyright © 2025 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.