Androgen receptor signaling induces hemorrhage and angiogenesis in the irradiated bladder

Abstract

The underlying mechanism of radiation cystitis remains unknown, however, angiogenesis induced by hypoxia seems to be important because hyperbaric oxygen therapy which suppresses HIF-1 is clinically effective and significantly associated with androgen receptor signaling. We herein assessed the impact of androgen deprivation therapy on radiotherapy-induced bladder hemorrhage in men with prostate cancer and that of androgen receptor signaling on angiogenesis in irradiated bladder cell lines and a mouse model of radiation cystitis. In 507 patients with prostate cancer undergoing external beam radiation therapy, univariate (hazard ratio 0.61, p = 0.039) and multivariate (hazard ratio 0.50, p = 0.006) analyses revealed that the use of androgen deprivation therapy was associated with a significantly lower risk of gross hematuria. In irradiated bladder cells, the levels of FLT1 and KDR expression were significantly elevated when pretreated with dihydrotestosterone, which was abolished by an anti-androgen hydroxyflutamide. In male mice with radiation cystitis, castration significantly reduced the incidence of hematuria. Correspondingly, microvessel density and VEGFR expression in the bladders in the castration group were significantly lower than those in the sham surgery group. Our results suggest that androgen receptor activation contributes to inducing angiogenesis in irradiated bladder cells. Androgen deprivation therapy thus has a potential for preventing radiation cystitis.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-15245-9.

Introduction

Radiotherapy is commonly used to treat pelvic malignancies including bladder and prostate cancers. However, radiotherapy can affect not only the targeted organs but also surrounding areas. In fact, adverse events in the gastrointestinal and genitourinary systems are common in patients with bladder and prostate cancer following radiotherapy. Injection of a hydrogel spacer between the prostate and rectum has been shown to reduce a risk of intestinal complications¹. Meanwhile, the incidence of radiation cystitis (23–80%) and severe hematuria (5–8%) remains high after radiotherapy for prostate or cervical cancer^2–6. Remarkably, there has been no effective way to prevent radiation cystitis yet.

The clinical course of radiation cystitis can be divided into acute and chronic phases. In the acute phase (within 6 months after radiotherapy), the damage to the urothelium and microvasculature is observed. Microvascular damage induces the development of edema, accompanied by acute inflammation⁷. Vascular endothelial cells are considered the primary targets for bladder injury after radiotherapy, along with subsequent bladder fibrosis. In the chronic phase (6 or more months later), radiation-induced vascular injury develops fibrosis of the submucosal arterioles and capillaries, leading to tissue hypoxia^8,9. Hyperbaric oxygen therapy has been proposed as a treatment option for radiation cystitis, and with success rate ranging from 60–92%^10–15. Moreover, hyperbaric oxygen therapy suppressed hypoxia-inducible HIF-1 and its downstream signaling¹⁶ which stimulated angiogenesis by increased transcription activity of VEGF¹⁷. Given these facts, angiogenesis induced by hypoxia is most likely associated with the development of radiation cystitis.

We have demonstrated preclinical evidence indicating a strong association between androgen receptor (AR) signaling and radio-sensitivity in bladder cancer¹⁸. Specifically, AR expression was significantly elevated in radio-resistant cells, whereas irradiation was significantly more effective in AR-negative or AR-positive cells pretreated with the anti-androgen hydroxyflutamide (HF). Meanwhile, it has been documented that most of benign/non-neoplastic urothelial cells express AR^19,20.

It has never been tested if the blockade of AR signaling during peri-radiotherapy period prevents radiation cystitis. Therefore, the aim of the present study was to determine the impact of ADT during peri-radiation period on the risk of gross hematuria in patients with prostate cancer. Moreover, we assessed the association of AR signaling with angiogenesis in the irradiated bladder using cell lines and mouse models.

Results

Patient characteristics

We divided our cohort into two groups: those with ADT (n = 249) and those without ADT (n = 258). Table 1 shows the distribution of patients according to ADT status during the peri-radiation period. The two groups differed significantly in the age, radiation dose, PSA, clinical T stage and Gleason score.

Table 1.

Characteristics of prostate cancer patients receiving radiotherapy combined with and without ADT.

		ADT	No ADT	P value
Age
< 70 years	(N = 210)	90 (36.1%)	120 (46.5%)	0.023
≥70 years	(N = 297)	159 (63.9%)	138 (53.5%)
Radiation dose
< 76 Gy	(N = 159)	65 (26.1%)	94 (36.4%)	0.016
≥76 Gy	(N = 348)	184 (73.9%)	164 (63.6%)
IMRT
No	(N = 135)	62 (24.9%)	73 (28.3%)	0.445
Yes	(N = 372)	187 (75.1%)	185 (71.7%)
3D
No	(N = 65)	32 (12.9%)	33 (12.8%)	1.000
Yes	(N = 442)	217 (87.1%)	225 (87.2%)
PSA
< 10ng/ml	(N = 248)	81 (32.5%)	167 (64.7%)	< 0.001
≥10ng/ml	(N = 259)	168 (67.5%)	91 (35.3%)
T stage
T1 or 2	(N = 330)	121 (48.6%)	209 (81.0%)	< 0.001
T3	(N = 177)	128 (51.4%)	49 (19.0%)
Gleason score
< 8	(N = 344)	129 (51.8%)	215 (83.3%)	< 0.001
≥8	(N = 163)	120 (48.2%)	43 (16.7%)
Anticoagulant use
No	(N = 422)	208 (83.5%)	214 (82.9%)	0.953
Yes	(N = 85)	41 (16.5%)	44 (17.1%)

ADT, androgen deprivation therapy; IMRT, intensity modulated radiation therapy.

Gross hematuria rates stratified by ADT status

The 10-year gross hematuria rate was 15.2% in patients who underwent ADT vs. 24.1% in those who did not and the risk of hematuria was significantly higher in patients who did not undergo ADT (p = 0.039, Fig. 1). Univariate Cox regression analysis showed that ADT (p = 0.039) and Gleason score (p = 0.069) were strong predictors of gross hematuria (Table 2). In the multivariate Cox regression analysis, ADT (hazard ratio [HR] 0.50, p = 0.006) and Gleason score (HR 1.90, p = 0.012) were independently associated with the gross hematuria (Table 2).

Fig. 1.

Cumulative incidence of gross hematuria in prostate cancer patients receiving RT with or without ADT.

Table 2.

The relationship between clinico-pathological factors and the incidence of gross hematuria in prostate cancer patients receiving radiotherapy.

		Univariate		Multivariate
		HR (95%CI)	p value	HR (95%CI)	p value
Age
< 70 years	(N = 210)	Ref	0.722
≥70 years	(N = 297)	1.09 (0.69–1.71)
Radiation dose
< 76 Gy	(N = 159)	Ref	0.164
≥76 Gy	(N = 348)	0.72 (0.45–1.15)
IMRT
No	(N = 135)	Ref	0.354
Yes	(N = 372)	1.29(0.75–2.20)
3D
No	(N = 65)	Ref	0.102
Yes	(N = 442)	1.93 (0.88–4.24)
PSA
< 10ng/ml	(N = 248)	Ref	0.429
≥10ng/ml	(N = 259)	0.83(0.53–1.31)
T stage
T1 or 2	(N = 330)	Ref	0.96
T3	(N = 177)	0.99(0.62–1.58)
Gleason score
< 8	(N = 344)	Ref	0.069	Ref	0.012
≥8	(N = 163)	1.54(0.97–2.46)		1.90 (1.15–3.13)
ADT during peri-radiation period
No	(N = 249)	Ref	0.039	Ref	0.006
Yes	(N = 258)	0.61(0.38–0.97)		0.50 (0.30–0.82)
Anticoagulant use
No	(N = 422)	Ref	0.282
Yes	(N = 85)	1.36 (0.78–2.35)

IMRT, intensity modulated radiation therapy; ADT, androgen deprivation therapy.

Elevated expression of AR and VEGFR genes in irradiated bladder cancer cells

To investigate the changes in AR expression following irradiation, we performed RT-PCR on three AR positive bladder cancer cell lines, UMUC3, 5637-AR, and 647 V-AR. RT-PCR showed that AR gene expression levels in irradiated bladder cancer cell lines were significantly higher than that in the respective control lines (Fig. 2). Correspondingly, irradiation induced the expression of VEGFR genes, FLT1 (Fig. 3A) and KDR (Fig. 3B) in all three cell lines.

Fig. 2.

Changes in AR expression in bladder cancer cells after irradiation. Quantitative RT-PCR of AR in UMUC3, 5637-AR, and 647 V-AR cultured for 24 h after 2 Gy irradiation. Expression of AR was normalized to that of GAPDH. Transcription amount is presented relative to that of control cells. Each value represents the mean (+ SD) from three independent experiments. *p < 0.05 (vs. control).

Fig. 3.

Changes in the expression of VEGFR genes after irradiation. Quantitative RT-PCR of FLT-1 (A) and KDR (B) in UMUC3, 5637-AR, and 647 V-AR cultured for 24 h after 2 Gy irradiation. Expression of each gene was normalized to that of GAPDH. Transcription amount is presented relative to that of control cells. Each value represents the mean (+ SD) from three independent experiments. *p < 0.05 (vs. control).

Elevated expression of VEGFR genes in androgen-treated bladder cancer cells

The effects of androgen on VEGFR genes were next examined. The expression of FLT1 (Fig. 4A) and KDR (Fig. 4B) were significantly up-regulated by DHT treatment in AR positive bladder cancer cells, which was at least partially abolished by HF. However, FLT1 (Fig. 4C) and KDR (Fig. 4D) expression did not differ significantly in AR negative bladder cancer cells after treatment of DHT/HF.

Fig. 4.

Effects of androgen on VEGFR genes in AR-positive (A,B) and AR-negative (C,D) bladder cancer cells. Quantitative RT-PCR of FLT1 (A,C) and KDR (B,D) in UMUC3/UMUC3-AR-shRNA, 5637-vector/5637-AR, and 647 V-vector/647V-AR cultured with ethanol (mock), 10 nmol/L DHT, and/or 1 µmol/L HF for 48 h and subsequently irradiated (2 Gy). Expression of each gene was normalized to that of GAPDH. Transcription amount is presented relative to that of mock treated cells. Each value represents the mean (+ SD) from three independent experiments. *p < 0.05 (vs. mock treatment)　*p < 0.05 (vs. DHT treatment).

Incidence of hematuria in radiation cystitis model mice with castration

The incidence of microscopic hematuria was compared between male mice with radiation cystitis that underwent orchiectomy and those that underwent sham surgery. As shown in Fig. 5, the cumulative incidence of hematuria was significantly lower in castrated animals than in controls (p = 0.003). Correspondingly, immunohistochemical analysis of the bladders resected from the radiation cystitis model showed that VEGFR2 expression was considerably lower in urothelial cells from castrated mice (H-score 45.0 ± 1.6) than those in controls (95.6 ± 3.6; p < 0.001) (Fig. 6A). Similarly, MVD in the bladders from castrated mice (55,496 ± 17,153) was significantly lower, compared with controls (207,790 ± 83,879; p = 0.007) (Fig. 6B).

Fig. 5.

Effects of androgen deprivation on hematuria in radiation cystitis model mice. Seven days after surgery (orchiectomy or sham), all male C3H/Neu mice received irradiation (20 Gy). Cumulative incidence for the detection of hematuria via three consecutive positive urine tests was compared.

Fig. 6.

Immunostaining of VEGFR2 and CD34 in bladder resected from radiation cystitis model mice. VEGFR2 (A) and CD34 (B) expression in the bladders resected from radiation cystitis model mice receiving sham surgery (n = 8) or castration (n = 7). Bar graphs show H-score of VEGFR2 expression or MVD. Each value represents the mean (+ SD). *p < 0.05 (vs. sham surgery).

Discussion

The underlying mechanism of radiation cystitis remains poorly understood to date. As mentioned earlier, there are acute and chronic phases in radiation cystitis. In the acute phase, hemorrhagic and inflammatory findings are observed. Early events in vascular endothelial cells contribute to the progression of radiation-induced vascular damage. Stewart et al. showed that irradiation with a single dose of 25 Gy to the mouse bladder induced compensatory proliferation of endothelial cells, with this increase in proliferation activity being dose-dependent up to 22 months after irradiation^21,22. In another study, the expression of VEGFα was increased in urine samples from patients with hemorrhagic cystitis²³. During inflammation, the secretion of various immune-related molecules is induced following leukocytes migration into the irradiated area^24,25. Specifically, the multiplication of macrophages and T cells leads to the production of inflammatory mediators, including NFκB, SMAD2/3, and cytokines²⁶. Interestingly, activation of NFκB was shown to up-regulate AR expression in bladder cancer cells²⁷. In the chronic phase, TGF-β mainly promotes fibrosis and subsequent vascular damage²⁸. Hyperbaric oxygen therapy is effective in the chronic phase, particularly for improving hematuria, although it is not a definitive treatment.

Emerging evidence suggests a strong association between AR signaling and DNA repair genes involved in modulating radio-sensitivity in bladder cancer¹⁸. In the present study, we further demonstrated that irradiation induced AR expression in bladder cancer cells. Meanwhile, significant associations between AR signaling and angiogenesis have been suggested²⁹. Specifically, AR knockout mice exhibited impaired blood flow recovery, and AR-deficient endothelial cells showed reduced angiogenic capability. Also, immunoprecipitation experiments in endothelial cells revealed an interaction between AR and VEGFR, which prompted the recruitment of downstream signaling components²⁹. Moreover, VEGF expression was down-regulated in mouse xenograft models of bladder cancer undergoing ADT³⁰. Additionally, AR signaling has been shown to induce angiogenesis through the VEGF pathway in prostate cancer cells^31,32. Collectively, the results of the present study align with previous data and further suggest that AR activation induces angiogenesis in the bladder.

To the best of our knowledge, little is known about factors influencing radiation cystitis, including the role of ADT. Consistent with our findings, a clinical trial showed that the incidence of severe adverse bladder toxicity was marginally higher in prostate cancer patients treated with radiotherapy plus short-duration (4 months) ADT (4%) than in those with radiotherapy plus longer-duration (9 months) ADT (2%; p = 0.054)³³. Limited studies have demonstrated that smoking and antithrombotic treatment increase the risk of radiation cystitis, whereas IMRT reduces it³⁴. However, they were not significant predictors of gross hematuria in the present study, suggesting that ADT has a stronger impact on predicting the risk of gross hematuria than the other factors. A recent study showed that post-prostatectomy radiotherapy was associated with a significantly increased risk of radiation cystitis, compared with radiotherapy as primary treatment, although the radiation dose of post-prostatectomy setting was lower than that of primary setting³⁴.

There were several limitations in the present study, including a retrospective design, and, thus, unknown sources of bias might exist in the results obtained. In our in vivo study, hematuria was evaluated only during the acute phase. The inhibition of androgen signaling has a potential of preventing the occurrence of chronic radiation cystitis based on the fact that AR signaling regulate the activity of TGF-β which further modulates fibrosis³⁵. Therefore, further investigation is required to clarify whether ADT contributes to preventing the progression from acute to the chronic radiation cystitis. The present study only included male mice and patients. As seen in the incidence and pathogenesis of bladder cancer, in which AR signaling has been suggested to play a critical role³⁰, sex-specific differences in AR-mediated angiogenesis in non-neoplastic urothelial cells may exist. Further studies using female animals should therefore be performed. Moreover, the involvement of estrogen-mediated signals may also need to be assessed.

In summary, we retrospectively evaluated the relationship between gross hematuria incidence and clinico-pathological parameters including the use of ADT during the peri-radiation period in patients with prostate cancer to determine whether ADT could play an important role in preventing radiation cystitis. In these patients, simultaneous long-term (≥ 6 months) ADT significantly reduced the incidence of gross hematuria. Additionally, in bladder cancer cell lines, irradiation resulted in significant increases in the expression of AR and VEGFR genes and anti-androgen treatment inhibited their expression in irradiated AR positive cells. Finally, in an animal model for radiation cystitis, orchiectomy significantly reduced the cumulative rate of hematuria as well as VEGFR2 expression and MVD in the bladder. Thus, we demonstrated clinical and preclinical evidence suggesting that AR activation was associated with radiation-induced bladder angiogenesis and hemorrhage in the bladder. These findings further imply that combined ADT is beneficial to patients with prostate or bladder cancer undergoing radiotherapy not only by increasing radio-sensitivity, as we have previously demonstrated, but also preventing the occurrence of radiation cystitis. Given the link between AR signaling and VEGF pathways in bladder cells and the association of VEGF expression levels with bladder cancer prognosis³⁶, inhibition of AR signaling could serve as a novel therapeutic strategy for bladder cancer.

In conclusion, the results of the present study suggest that AR inactivation suppresses angiogenesis in urothelial cells. Thus, ADT during the peri-radiotherapy period has a potential for preventing radiation cystitis.

Methods

Study population

After receiving Institutional Review Board approval at each institution, we retrospectively reviewed the medical records of 520 patients with prostate cancer who received external beam radiation therapy (EBRT), including intensity modulated radiation therapy (n = 382), at Keio University Hospital and Saiseikai Central Hospital between 1995 and 2012. This study was approved by the Ethics Committee of them and was performed in accordance with the guidelines of human research. All subjects involved in the study provided their written informed consent according to the Declaration of Helsinki, as approved by the Ethics Committee of them. EBRT was applied to the prostate only for the low-risk group, to the prostate and base of seminal vesicle for the intermediate-risk group, and to the prostate and seminal vesicle for the high-risk group, according to the D’Amico risk classification³⁷. We excluded patients who was dead (n = 1) or was lost to the follow-up (n = 5) within 2 months after EBRT and those whose detailed data on the information about ADT (n = 3) or urine analysis (n = 4) were unavailable. Thus, we ultimately included 507 patients for analysis in the present study. The mean age of the 507 patients was 70.2 years (range 41–90) and mean follow-up was 102 months (range 3–190). Gross hematuria after EBRT was found in 76 (15.0%) cases. Gross hematuria was defined as grade 2 or higher hematuria in the Common Terminology Criteria for Adverse Events version 5.0. In all patients undergoing ADT, it started before radiotherapy and continued for more than 6 months.

Patients were assessed by urine analysis every 3–6 months for 10 years after EBRT and every 6–12 months thereafter. Ultrasonography, computerized tomography, magnetic resonance imaging, and cystoscopy were performed when urine analysis showed hematuria. Gross hematuria due to other explainable causes, such as infection, genitourinary malignancy, and drug-induced cystitis, was excluded.

Antibodies and chemicals

Anti-VEGFR2 (sc-6251) and anti-CD34 (ab81289) antibodies were purchased from Santa Cruz Biotechnology and Abcam, respectively. We obtained dihydrotestosterone (DHT) and HF from Sigma-Aldrich.

Cell culture

Human bladder cancer cell lines, UMUC3,5637 and 647 V were originally obtained from ATCC.　All these lines were authenticated, using GenPrint 10 system (Promega), by the institutional core facility, and routinely tested for mycoplasma contamination using PCR Mycoplasma Detection Kit (Applied Biological Materials). Stable sublines, such as UMUC3-control-short hairpin RNA (shRNA), UMUC3-AR-shRNA, 5637-vector, 5637-AR, 647 V-vector, and 647 V-AR, were established in our previous studies^38,39. These cells were maintained in DMEM (Mediatech) supplemented with 5% fetal bovine serum (FBS) at 37 °C in a humidified atmosphere of 5% CO₂, and cultured in phenol red free-medium with either 5% FBS or 5% charcoal-stripped FBS (for DHT treatment) at least 24 h before experimental treatment.

Reverse transcription (RT) and real-time PCR

The total RNA isolated from cultured cells by TRIzol (Invitrogen) was reverse transcribed, using oligo-dT primers and Omniscript reverse transcriptase (Qiagen). Real-time PCR was then performed, using RT2 SYBR Green Mastermix (Qiagen). The primer sequences are given in Supplementary Table S1.

Radiation cystitis mouse model

The animal protocol in accordance with the ARRIVE guidelines⁴⁰, the National Institutes of Health handbook for the Care and Use of Experimental Animals was approved by the Institutional Animal Care and Use Committee. At 8 weeks of age, male C3H/Hen mice (Sankyo Labo Service Corporation, inc.) received bilateral orchiectomy or sham surgery under anesthesia (2 mg ketamine + 0.2 mg xylazine with saline per a mouse). One week after surgery, these mice were irradiated with 20 Gy using an irradiator (MBR-1520R-3, Hitachi) to induce radiation cystitis, as described previously⁴¹. Urine was assessed for the presence of hematuria twice a week, using Chemostrip 5 OB urine test strips (Roche). After 3 weeks, the animals were euthanized, and their bladders were collected for immunohistochemical analysis.

Immunohistochemical staining

Sections from paraffin-embedded mouse urinary tract tissue were deparaffinized in xylene and rehydrated in a graded ethanol series. After antigen retrieval was performed, the sections were incubated with 0.3% hydrogen peroxide to block endogenous peroxidase, blocking solution for 60 minutes, and an anti-VEGFR2 antibody (1:500 dilution) or an anti-CD34 antibody (1:5000 dilution) at 4 °C overnight. After washing with PBS, tissue sections were incubated with the secondary antibody for 30 min. Color was developed using 3,3′-diaminobenzidine tetrahydrochloride in 50 mmol/L Tris-HCl (pH7.5) containing 0.005% hydrogen peroxide. Finally, the sections were counterstained with hematoxylin.

Evaluation of immunohistochemical staining and microvessel density (MVD)

To assess VEGFR expression, percentage of immunoreactive urothelial cells (0–100) and staining intensity stratified from 0 to 3 (0: no immunoreactivity, 1: low immunoreactivity, 2: moderate immunoreactivity, 3: strong immunoreactivity) were estimated in at least 10 representative fields. The histoscore (H-score) was calculated by applying the following formula: mean percentage × intensity (range 0–300)⁴². In CD34 staining, MVD was evaluated in 5 fields (0.75 mm² surface area) with the most abundant blood vessel aggregations (hot spots) from 10 images randomly captured at 10× magnification by measuring the pixel number in the CD34 positive areas, using ImageJ software (National Institutes of Health).

Statistical analysis

Student t test (two-tailed) was used to analyze differences in variables with a continuous distribution. Clinico-pathological parameters were compared by the chi-squared test. The cumulative gross hematuria rate was estimated, using the Kaplan–Meier method, and compared with the log-rank test. A multivariate analysis was performed using a Cox proportional hazards model with stepwise forward selection. Furthermore, the ideal cut-off values for the variables were calculated by applying a receiver-operating curve analysis. The level of significance was set at p < 0.05. These analyses were performed with the SPSS statistical software package, version 27.0 (SPSS: An IBM Company, Chicago, IL, USA).

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

T.K. and H.I. wrote the draft of the article, did the literature search, and contributed to data collection and analysis. J.F., Y.B., K.M., and H.M. contributed to data collection. M.O. supervised this work.

Data availability

The datasets generated during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.