Abstract
Background/Aim: Since androgens affect urothelial bladder cancer (UBC), we examined whether 5α-reductases (5-AR) have genomic alterations in UBC and whether 5α-reductase inhibitors (5-ARI) affect UBC. Materials and Methods: The cBioPortal was used to analyze genomic alternations of 5-ARs in UBC cancer genomic datasets. Next, we used the Taiwan National Health Insurance Research database to conduct a population-based case-control study to investigate the effect of a 5-ARI, finasteride on UBC incidence. We also performed an XTT assay to examine the direct effect of finasteride on UBC cells. Results: We found that 5-AR genomic alternations were observed in 29% of UBC patients and patients with alternations had shorter disease-free survival. Also, the use of finasteride with >180 cDDDs reduced the risk of UBC. Finasteride could directly inhibit UBC cell growth. Conclusion: Based on our findings, we concluded that 5-AR could be explored as a therapeutic target for UBC with 5-ARIs.
Urothelial Bladder cancer (UBC) is the most common cancer of the urinary tract with an estimated 79,030 new cases and 16,870 deaths projected to occur in the United States in 2017 (1). Most UBCs present as non-muscle invasive bladder cancer, but an approximate 25% of UBC patients are diagnosed with muscle-invasive disease, while only a small proportion of UBC patients have distant metastases detectable at their initial diagnosis (2). Low-grade non-muscle-invasive tumors that recur frequently rarely progress to muscle invasion, but muscle-invasive tumors that are usually diagnosed de novo frequently metastasize (3, 4). Almost half the muscle-invasive UBC patients eventually develop obvious distant metastases (2), thus making UBC a deadly disease. Comprehensive genomic characterization of UBC reveal a complex landscape of molecular alterations including multiple genes involved in cell cycle regulation, chromatin regulation, and kinase signaling pathways (3). These complicated molecular features cause various pathogenesis and heterogeneity of UBC, resulting in the therapeutic failures to obtain the effective molecularly targeted agents to treat this disease (3, 4). In the era of precision medicine, it is proposed to match the most accurate and effective treatment to each individual cancer patient based on their genomic alternations to find the therapeutic agents that target altered genes or pathways (5). Therefore, with this precision medicine approach, it is possible to also find a potential therapeutic option in UBC based on its genomic alterations.
An epidemiological study has shown that UBC occurs more often in men than in women (male:female ratio is ~3:1) (1). In cell based experiments and animal models, male sex hormones, androgens and androgen receptors have been demonstrated to be implicated in the development and progression of UBC (6-9). Using UPII-SV40T transgenic mice that spontaneously develop UBC, Johnson et al. reported that UBC progression was repressed by castration and restored by the administration of dihydrotestosterone (DHT), a more potent androgen than testosterone (10). Furthermore, Miyamoto showed that in ARKO mice models, DHT alone was able to promote bladder cancer development under chemical carcinogen N-butyl-N-(4-hydroxybutyl) nitrosamine (BBN) induction (7), suggesting DHT is involved in UBC development and progression. Furthermore, in a retrospective study, androgen deprivation therapy (ADT) using luteinizing hormone-releasing hormone (LHRH), a standard treatment to treat advanced prostate cancer (PC) patients has been shown to significantly reduce the risk of UBC recurrence in PC patients who had coincidental UBC (11). Based on these evidence, we hypothesized that DHT could play a critical role in contributing to male dominance in UBC and AR-mediated gender difference in UBC development and progression. Therefore, the genomic alternations of 5α-Reductases that convert testosterone into the more potent DHT (12) could influence UBC development and progression. Thus, it is possible to use 5-ARI as a therapeutic agent for preventing and treating UBC development and progression.
There are three isoenzymes of 5-AR: steroid 5α-reductase 1, 2, and 3 (SRD5A1, SRD5A2 and SRD5A3), which are expressed in many tissues: type 1 5α-steroid reductase (SRD5A1) is predominantly expressed in the skin, scalp, sebaceous gland, liver, and brain; type 2 5α-steroid reductase (SRD5A2) is found predominantly in androgen target organs such as the prostate, genital skin, and seminal vesicles; whereas type 3 5α-steroid reductase (SRD5A3) is overexpressed in castration-resistant prostate cancer (13, 14). 5α-reductase inhibitor drugs are used to treat benign prostatic hyperplasia and prostate cancer (15, 16). Finasteride, a 5-ARI, mainly inhibits type II and III 5-AR, but is less effective on type I 5-AR and was shown to decrease the mean serum level of DHT by 71% after 6 months of use (17, 18).
To test our hypothesis, we explored the UBC cancer genomics dataset from The Cancer Genome Atlas (TCGA) projects to analyze the genomic alternations of 5-ARs in UBC. We also conducted a population-based survey to investigate whether the use of finasteride could affect the risk of UBC using Taiwan Insurance Research Database (NHIRD) from Taiwan National Health Insurance (NHI) program which is a mandatory general health insurance program, providing complete medical care coverage to all Taiwanese residents. The cell-based experiments were also performed to directly determine the effect of finasteride on bladder cancer cells. The clinical observations and in vitro experiments indicated that 5-AR genomic alternations were present and associated with a worse outcome in UBC patients and there was a significantly decreased risk of UBC among patients who received finasteride compared with the general population. Furthermore, finasteride could directly suppress the growth of bladder cancer cells. All suggest that 5-ARI could be beneficial to UBC patients as a therapeutic option to prevent and treat UBC patients who have genomic alternations in 5-AR, which contribute to the progression of UBC.
Materials and Methods
The cBioPortal database analysis. We analyze the genomic alternations of three 5-AR genes using the cBioPortal tool, an open platform that provides access to data of cancer genomic studies from the TCGA projects (19, 20). Our selected genomic profiles of all three 5-ARs included alterations (amplification, deep deletion, missense mutations), CNA from GISTIC and RNA seq data with the default setting to perform query on all completed tumors in Bladder Urothelial Carcinoma (TCGA) dataset.
Cell culture and XTT assay. Human TCC-SUP bladder cancer cells were obtained from the American Type Culture Collection and maintained in DMEM supplemented with 10% fetal bovine serum. To test the effects of finasteride on cell growth, the cells were seeded at 5×103 cells/well onto 96-well plates and incubated overnight. At time 0, media were replaced with fresh complete DMEM media supplemented with 10% fetal bovine serum (FBS) or 10% dextran-coated charcoal stripped FBS (DC-FBS) and then cells were treated with finasteride (AdooQ BioScience, Irvine, CA, USA) with concentrations as indicated. After 72 h, XTT assay kit (Sigma-Aldrich, St. Louis, MO, USA) was used to measure cell viability, which was expressed as a percentage of the absorbance measured in the vehicle-treated cells. The mean of three parallel samples was calculated. Experiments were performed in triplicate and standard deviations were calculated based on the mean of three experiments.
Clinical observations. This study used the cancer dataset and LHID 2010 of the NHIRD from 1 January, 2002 to 31 December, 2013. The NHIRD is maintained by Taiwan's National Health Research Institutes, and is made available to researchers (http://nhird.nhri.org.tw/date_01_en.html). For cases, we first identified 16,784 patients who had received a first-time bladder cancer diagnosis (ICD-9-CM code 188.X) between January 1, 2003 and December 31, 2012 in cancer dataset. The date of receiving their first-time bladder cancer diagnosis was defined as the index date. We then selected controls from the remaining enrollees without bladder cancer diagnosis in LHID 2010. The controls included all patients without bladder cancer. We attempted to reduce selection bias by bundling many confounding covariates that may be present in an observational study with this number of variables; we then implemented a systematic random sampling design to select a matching sample from the controls frequency matched by age and the year of the index date. The matching comparison to case ratio was 1 to 1. In this study, the year of the index date for cases was defined as the year in which the cases received their first bladder cancer diagnosis. However, for controls, the index year was simply a matched year in which the controls had a healthcare utilization. Furthermore, we defined the first healthcare utilization occurring in the index year as the index date for controls.
The defined daily dose (DDD) is a unit for measuring a prescribed amount of a drug; it is the assumed average daily maintenance dose of a drug consumed for its main indication in adults. The cumulative DDD (cDDD), which indicates the duration of exposure, was estimated as the sum of dispensed DDD of 5-alpha-reductase inhibitor. To examine the dose-effect relationship, we categorized the use of 5ARI into three groups in each cohort by cDDD (0, 1 to 179, and more than 180).
The independent variables were co-morbidity (diabetes mellitus, cerebrovascular disease, chronic kidney disease, hypertension, and hyperlipidemia), geographical area of residence, and socioeconomic status (SES).
Statistical analysis. Disease-free survival curve was generated by the cBioPortal and displayed with values calculated using the log-rank test and Kaplan-Meier plots. Statistical significance of the p data (p-values) was provided by the program. The conditional logistic regression was used to measure the odds ratios (ORs) and 95% confidence intervals (95% CIs) for the effect of finasteride on UBC risk. In cell-based experiments, student's t-test was performed to determine whether there was a statistically significant difference between the control and treatment. All statistical analyses were performed using SPSS statistical software. Data are present as mean±standard deviation (SD). A p-value <0.05 was considered to be statistically significant.
Results
To prove that the altered androgen pathway could be present in UBC, we analyzed the mutations, copy number alterations (CNA) as well as mRNA expression of 5-ARs that include three isoenzymes encoded by 3 different genes: 3-oxo-5α-steroid 4-dehydrogenase 1 (SRD5A1), SRD5A2, and SRD5A3 by the cBioPortal tool on the Bladder Urothelial Carcinoma (TCGA, Provisional) data on all 125 complete tumor samples that have mRNA, CNA and sequencing data (19, 20). The TCGA network analyzed UBC samples consisted of chemotherapy naive, muscle-invasive, high-grade urothelial tumors (T2-T4a, Nx, Mx), as well as peripheral blood and/or tumor-adjacent, histologically normal-appearing bladder tissue to define the molecular landscape of UBC (3). The results showed that there were total genomic alternations in 37 (29%) of 126 sequenced cases/patients (Figure 1A). For the SRD5A1 gene, there were 15 cases (11.9%) of amplification and 1 case of mutation in 126 cases. For the SRD5A2 gene, there were 2 cases (1.6%) of amplification and 1 case of deletion in 126 cases. And for the SRD5A3 gene, there was 1 case of amplification in 126 cases. Furthermore, mRNA upregulation was found to be present in 21 (17%) of 126 sequenced cases/patients with 11%, 0.8% and 5% on SRD5A1, SRD5A2, and SRD5A3 respectively (Figure 1B).
Disease-free survival Kaplan–Meier plot analysis showed that patients with genomic alterations on SRD5A1, SRD5A2, and SRD5A3 genes had lower median survival (17.35 vs. 25.23 months) (Figure 2A). The network view of 5-AR genes shows that their biologically interacted genes also have genomic alterations in UBC and there are FDA approved drugs (ex. finasteride) targeting genes in the network (Figure 2B). These genes are involved in hormone regulation, have “catalysis-precedes” interactions with 5-ARs genes and their genomic alterations are mostly gene amplifications and mRNA up-regulation as 5-AR genes do, suggesting that hormone imbalance could play a critical role in UBC development and progression.
Since finasteride is listed as a drug which could target 5-ARs in UBC, we wanted to know the relationship between finasteride and UBC. We then performed a case-control study on the effect of finasteride on the risk of UBC in the Taiwan National Health Insurance Research Database. We identified UBC patients as study cases and non-UBC patients, matched with gender and age as controls from 2002 to 2013. The demographic data and comorbidities of the cohort are shown in Table I. We used the cumulative defined daily doses (cDDDs), which indicates the duration of exposure to estimate the sum of dispensed DDD of 5-alpha-reductase inhibitor, finasteride. We then used conditional logistic regression measured odds ratio to analyze the correlation between the cDDDs of finasteride and the risk of UBC. And we found that patients with high cDDDs (> 180) have a lower risk of UBC (Table II).
To validate the clinical observations that 5-ARs are involved in the development and progression of UBC and 5-ARI could have an effect on the prevention and intervention of UBC development and progression, we examined whether finasteride could have an effect directly on UBC cells. We then treated TCC-SUP cells, a bladder cancer cell line derived from an undifferentiated, Grade IV transitional cell carcinoma (21) with different concentrations of finasteride and then cells were incubated for 72 h. The cells were harvested for XTT viability assay. The result showed that in the presence of 10% FBS which contains a castrate level of testosterone (22.0±6.1 pg/mL), which still could be converted to produce a physiologic (i.e. 10 nM) level of intracellular DHT by prostate cancer cells (22), all concentrations of finasteride suppressed cell growth (Figure 3A). We also examined the effect of finasteride in the presence of 10% of dextran-coated charcoal stripped FBS (DC-FBS), which removes most steroids, and finasteride was able to repress cell growth at higher does (Figure 3B).
Discussion
UBC is three to four times more common in men than in women, and represents the fourth most common cancer and the eighth most common cause of cancer death in males in the U.S.A. (1). This gender disparity is proposed to be explained by the roles of androgens and androgen receptor (AR) in the development and progression of UBC, which have been examined in clinical observational and animal models (23). Furthermore, androgen deprivation therapy was shown to prevent bladder cancer recurrence (11), suggesting that androgen modulation could be a therapeutic option to prevent or treat UBC. In the era of precision medicine, it is proposed to obtain molecular profiling of cancer samples and find altered genes or pathways as targets for specific treatments. In the current study, by analyzing TCGA UBC dataset, we found that 5-AR genes are found to be altered in DNA and RNA levels. The main genomic alternations are gene amplification and mRNA up-regulations. We then further demonstrated that a 5-ARI, finasteride reduces the risk of UBC in our population-based study and in cell-based experiments, finasteride was also shown to suppress bladder cancer cell growth. Therefore, our study provides a novel insight for a possible therapeutic intervention with 5-ARIs for UBC patients whose genomic profiles of 5-AR genes are altered since it is proposed to have clinical trials based on patients with relevant druggable genomic alterations (3).
Corroboratively, Morales et al. reported that finasteride was associated with a reduced incidence of bladder cancer (hazard ratio: 0.634; 95% confidence interval=0.493-0.816; p=0.0004) as observed by self-report in the Prostate, Lung, Colorectal, and Ovarian cancer screening trial, large prospective screening study (24). A recent study examined the intravesical recurrence rate among men with non-muscle invasive bladder cancer, who have or have not received androgen suppression therapy (AST) by androgen deprivation therapy for prostate cancer or another 5-ARI, dutasteride, for benign prostatic hyperplasia and found lower intravesical recurrence in patients with AST (with, 12.5% vs. without, 30.1%) and 3.1% patients without AST, but no patients with AST progressed to muscle invasive bladder cancer (25). In consistent with our results, these studies also suggest the potential therapeutic option using 5-ARI to prevent and treat UBC.
The majority of patients with UBC are treated with surgery and chemotherapy and treatment for muscle-invasive bladder cancer has not advanced beyond cisplatin based combination chemotherapy and surgery in the past 30 years with no FDA approved targeted agents (3). Tumor heterogeneity and acquired resistance because of somatically acquired genetic, epigenetic, transcriptomic, and proteomic alterations in cancer cell cones may contribute to chemotherapy's failure and resistance to targeted therapy in UBC. However, the molecular profiling of cancers would provide valuable information about genomic alterations in cancer, which may grant cancer cells growth and survival advantages, but become Achilles' heel of the malignancy, becoming possible actionable targets (3). A single-patient case reported that with whole-genome sequencing, the tumor genetic profile of a patient with metastatic bladder cancer who achieved a durable (>2 years) and ongoing complete response to everolimus, a drug targeting the mTORC1 (mammalian target of rapamycin) was obtained and revealed a two-base-pair deletion in the TSC1 (tuberous sclerosis complex 1) gene and a nonsense mutation in the NF2 (neurofibromatosis type 2), both associating with mTORC1 dependence, which may explain the patient's drug response (26). This finding demonstrates the feasibility of precision medicine that the genomic alternations of patients may aid in the identification of targeted anticancer drugs that UBC patients most likely would respond to. 5-ARIs, investigated in our present study, may hold the promise in the treatment of UBC with this precision medicine approach.
In conclusion, our study found that there are genomic alternations of 5-AR in UBC patients and finasteride could reduce the risk of UBC and inhibit cancer cell growth. These results suggest that the efficacy and feasibility of 5-AR inhibitors, such as finasteride, to prevent or treat UBC patients with altered 5-AR genes.
Acknowledgements
This work was supported by the MOST grants (MOST 103-2633-B-039-003- and MOST 106-2314-B-303-017-), CMU grant (CMU105-S-51), CMUH grant (DMR-CELL-17020), Buddhist Tzu Chi General Hospital grant (TTCRD 104-12) and in part by Taiwan Ministry of Health and Welfare Clinical Trial Center (MOHW105-TDU-B-212-133019).
Footnotes
↵* These Authors contributed equally to this study.
Conflicts of Interest
The Authors disclosed no conflicts of interest.
- Received August 31, 2017.
- Revision received September 25, 2017.
- Accepted September 26, 2017.
- Copyright© 2017, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved