Androgen inactivation occurs mainly through the glucuronidation conjugative reaction mediated by UDP-glucuronosyltransferases (UGTs). This metabolic process is involved in the control of systemic and local androgen bioavailability.
To examine the relationship among expression of the androgen-inactivating UGT2B28 enzyme, circulating steroid hormone levels, and clinical phenotype in prostate cancer (PCa).
Design, setting, and participants
We conducted an analysis of a high-density prostate tumor tissue microarray consisting of 239 localized PCa cases. The study of 51 additional PCa patients with no copies of UDP glucuronosyltransferase 2B subfamily, polypeptide B28 (UGT2B28) in their genomes was performed to confirm the importance of the enzyme on circulating hormone levels.
Outcome measurements and statistical analysis
Steroid hormones were measured by mass spectrometry. Multivariate Cox proportional hazard models assessed the influence of UGT2B28 on progression, and general linear model regression evaluated variations in hormone levels.
Results and limitations
Tumor overexpression of UGT2B28 was associated with lower prostate-specific antigen levels at diagnosis, higher Gleason scores, margin and nodal invasion status, and it was shown to be an independent prognostic factor associated with progression. Enzyme overexpression correlated with 30% higher circulating levels of testosterone (T) and dihydrotestosterone (DHT). Patients with no copies of UGT2B28 in their genomes have lower levels of T (19%), DHT (17%), its glucuronide metabolites (18–38%), and enhanced levels of the adrenal precursor androstenedione (36%).
The UGT2B28 steroid-inactivating pathway modifies circulating T and DHT levels, and UGT2B28 overexpression is associated with high-grade PCa. Our work has uncovered the role of UGT2B28 as a regulator of steroidogenesis and underscores the interconnectivity among the steroid-inactivation capacity of cancer cells, hormone levels, disease characteristics, and the risk of cancer progression.
The androgen-inactivating UGT2B28 enzyme influences hormone levels, clinical and pathologic factors, and the risk of cancer progression.
Keywords: UDP-glucuronosyltransferase, Prostate cancer prognosis, Disease aggressiveness, Steroid metabolism, Circulating hormone levels.
Prostate cancer (PCa) is the most common cancer in men and the second leading cause of cancer death in North American men . PCa is a heterogeneous and complex genetic disease that relies primarily on steroid hormones for development, growth, and cell survival until the later stage of the disease. This is supported by the clinical activity and improved survival associated with abiraterone and enzalutamide treatments in phase 3 clinical randomized studies conducted in chemotherapy-naive  and  and chemotherapy-refractory  and  castration-resistant patients. The clinical relevance of steroid biotransformation locally is thus well acknowledged , even in the advanced disease setting. Local de novo androgen dihydrotestosterone (DHT) synthesis (to activate the androgen receptor) was shown to arise from testosterone (T) secreted by the testis, cholesterol, and increased conversion of precursors through the conventional androgen-biosynthetic and/or the backdoor pathways . The roles of multiple steroidogenic enzymes located in peripheral endocrine tissues of PCa cells that convert precursors into potent steroids are key in determining local exposure to hormones  and may consequently influence disease biology.
To maintain adequate hormone levels, the glucuronidation reaction participates in maintaining intracellular steroid hormone homeostasis within the prostate by inactivating steroids through the addition of a sugar moiety to a hydroxyl position within the steroid molecule (Fig. 1) . Once conjugated with glucuronic acid, androgens are more hydrophilic, released into circulation, and readily excreted through bile or urine, preventing binding to the androgen receptor . In humans, this reaction is mostly catalyzed by UDP-glucuronosyltransferases (UGT) of the UGT2B subfamily of enzymes, namely by UGT2B15, UGT2B17, and UGT2B28, which are all expressed in normal and neoplastic PCa cells  and . However, compared with UGT2B15  and UGT2B17 that are mostly active against DHT and its metabolites , UGT2B28 has distinctive androgen specificity, being active against T and DHT metabolites (Fig. 1) . UDP glucuronosyltransferase 2B subfamily, polypeptide B28 (UGT2B28) is located on chromosome 4q13, along with genes encoding other UGT2B family members, and it is one of the most commonly deleted genes in the human genome . Germline copy-number variation in UGT2B28 has been associated with Addison disease, an autoimmune disease of the adrenal cortex associated with adrenal insufficiency , with biochemical recurrence (BCR) in PCa patients after prostatectomy , and with female hyperandrogenism in a single case report . These findings support an important role for UGT2B28 in regulating hormonal exposure of steroid target cells.
In the present study, we sought to examine the expression of UGT2B28 in normal and tumoral prostatic tissues. We assessed its relationship with clinicopathologic characteristics of PCa tumors, circulating steroid hormone levels, and risk of cancer progression. We also established a circulating steroid profile associated with a complete deficiency of the UGT2B28 pathway in cancer patients.
2. Patients and methods
2.1. Patients and clinical data
Table 1 lists the characteristics and descriptive statistics of the study cohort. Patients who received neoadjuvant hormone therapy were excluded. All participants provided written informed consent for genetic analysis, and the local research ethics committee approved the research protocol. PCa progression was defined as BCR, metastasis, and/or death  and . An additional cohort of 1742 patients with localized PCa was studied specifically to identify those carrying no copies of UGT2B28 (homozygous for the UGT2B28 deletion [–/–]) (Supplementary Table 1). These PCa patients underwent radical prostatectomy and are from the PROCURE Prostate Cancer Biobank, launched in 2007, involving four university hospitals in Quebec .
n = 239
|Age at diagnosis, yr|
|Follow-up median, mo||87.2|
|Biochemical recurrence, n (%)||56 (23)|
|PSA level at diagnosis, n (%)|
|≤10 ng/ml||163 (69)|
|>10–20 ng/ml||58 (24)|
|>20 ng/ml||17 (7)|
|Pathologic Gleason score, n (%)|
|Pathologic T stage, n (%)|
|Nodal invasion, n (%)|
|Hormone therapy, n (%)|
|Margin status, n (%)|
PCa = prostate cancer; PSA = prostate-specific antigen; SD = standard deviation.
2.2. Development of a polyclonal affinity-purified antibody directed against UGT2B28
Affinity-purified polyclonal antibody against UGT2B28 was raised in rabbits by GenScript (Piscataway Township, NJ, USA) using the PolyExpress protocol and the immunogenic peptide comprising the UGT2B28 amino acids 113–124. Its specificity is depicted in Supplementary Figure 1.
2.3. Tissue microarrays
For the preparation of tissue microarrays (TMAs), one to three representative paraffin blocks from each tumor was selected. Six representative 0.6-mm tumor cores were taken and placed on a recipient paraffin block with a tissue arrayer (Beecher Instruments, Sun Prairie, WI, USA). All primary tumor slices were examined and graded by a pathologist (B.T.). Analysis of UGT2B28 expression was carried out by immunohistochemistry analysis using the UGT2B28 (Ab2321) polyclonal antibody (dilution 1:500) of 5-μm-thick sections from the paraffin-embedded tumor samples in the TMAs (n = 239). TMA staining was scored by two experienced lab members (H.B. and H.H.) and the pathologist of our team (B.T.), all three of whom were blind with respect to the tumor characteristics. The intensity of staining was scored as 0 or 1 when staining was absent or negligible, respectively, and as 2+ or 3+ when moderate or strong staining was observed, respectively. The percentages of cells with 2+/3+ staining were considered positive for expression for both nuclear and cytoplasmic compartments. Data are presented as the mean plus or minus the standard error of the mean (SEM).
2.4. Genotyping of the UGT2B28 copy-number variation
Genomic DNA was prepared from peripheral blood mononuclear cells collected from patients described in Table 1 and Supplementary Table 1 on a preoperative ambulatory clinical visit. Genomic deletion of UGT2B28 was detected as described  and .
2.5. Analyses of steroid levels in plasma by mass spectrometry
For the specific purpose of this study, all available TMA-plasma pairs (n = 239) were studied (number of patients as indicated in the tables); patients who received neoadjuvant hormonal treatment were excluded. Steroids were measured in plasma by validated gas chromatography–mass spectrometry or liquid chromatography–tandem mass spectrometry as described  and . Steroids measured and their respective limits of quantification are: dehydroepiandrosterone (DHEA; 0.2 ng/ml), 5-androsten-3β, 17β-diol (5-diol; 0.1 ng/ml), testosterone (T; 0.05 ng/ml)), dihydrotestosterone (DHT; 0.01 ng/ml), androsterone (ADT; 0.05 ng/ml), androstane-3β-17β-diol (3β-diol; 0.01 ng/ml), estrone (E1; 0.005 ng/ml), estradiol (E2; 0.001 ng/ml), 4-androstenedione (4-dione; 0.05 ng/ml), ADT-glucuronide (ADT-G; 1 ng/ml), androstane-3α, 17β-diol 3-glucuronide (3α-diol-3G; 0.25 ng/ml), androstane-3α, 17β-diol-17 glucuronide (3α-diol-17G; 0.25 ng/ml), DHEA-sulfate (DHEA-S; 0.075 μg/ml), and E1-sulfate (E1-S; 0.075 ng/ml); all reference steroids were purchased from Steraloids (Newport, RI, USA).
2.6. Statistical analyses
Quantitative variables are described as mean, SEM, median, and range, and qualitative variables as frequencies and percentages. Parametric (F test or t test) and nonparametric (Kruskal-Wallis, Wilcoxon rank sum) tests were used for continuous data comparisons; chi-square or Fisher exact tests were used for categorical data. In case of multiple comparisons, Tukey-Kramer adjustment was applied. To adjust for differences in the absolute levels of sex-steroid hormones, we estimated residuals of the natural logarithm of the hormone level regressed on age at blood donation and smoking status using general linear models (GLMs), except for UGT2B28-deficient men, for whom smoking status was not available. The association between UGT2B28 expression and variation in hormone level was assessed by performing the GLM regression of hormone residuals on UGT2B28 expression category (0 to ≤10%, >10–50%, and >50%) and in homozygous carriers of UGT2B28 (+/+) compared with UGT2B28-deficient patients carrying no copies of the gene (−/−). To facilitate comparisons between groups, we displayed the hormone levels as untransformed data. The multivariate Cox proportional hazard models were fitted to assess the influence of UGT2B28 on progression. The proportional hazards assumption for Cox models was not rejected. Statistical analyses were performed using SAS Statistical Software v.9.2 (SAS Institute, Cary, NC, USA) and using PASW statistics v.17 (SPSS Inc., Chicago, IL, USA), with a two-sided significance level set at p < 0.05.
3.1. UGT2B28 subcellular localization and expression levels are modified in prostate tumors
To study the distribution of UGT2B28 in prostatic tissues, we used a polyclonal antipeptide antibody that revealed only a faint reactivity with a stably transfected UGT2B10-overexpressing cell line despite high-sequence homology between UGT2B subfamily members (Supplementary Fig. 1). The Ab2321 UGT2B28 antibody showed ∼40 times more reactivity toward UGT2B28 compared with UGT2B10. As shown in Supplementary Table 2, the UGT2B28 staining signal was different between normal (n = 4) and malignant (n = 6) prostatic tissues.
The normal prostate glands presented nuclear staining, particularly in basal cells, with variable staining of secretory cells, whereas stromal fibroblasts also showed staining (Fig. 2A; Supplementary Table 2). For prostatic intraepithelial neoplasia (PIN) lesions, strong nuclear staining of basal cells and cytoplasmic staining of secretory cells was observed (Fig. 2B). The specificity of the Ab2321 antibody for UGT2B28 was also confirmed in the tumor sample of a UGT2B28-deficient patient (Fig. 2C and 2D). Cytoplasmic staining was distinctive of cancer cells, which also presented variable nuclear staining (Fig. 2E and 2F) (Supplementary Table 2). In agreement with cytoplasmic staining in cancer cells, in the lymph node carcinoma of the LNCaP cellular model, UGT2B28 is mostly expressed in the endoplasmic reticulum membranes in the cytoplasm.
3.2. Relationship between UGT2B28 expression in prostate tumors, prognostic factors, and clinical outcomes
Table 1 lists the characteristics of the studied population. The percentage of tumor cells with positive staining was similar between carriers of one copy and carriers of two copies of UGT2B28 (Fig. 3A). For nuclear staining, the number of (+/+) UGT2B28 patients was 26, 62, and 95, and in (+/−) UGT2B28 individuals, the distribution was 6, 23, and 27 in the ≤10, >10–50, and >50% subgroups, respectively. For cytoplasmic staining, the number of (+/+) UGT2B28 patients was 24, 68, and 91; in (+/−) UGT2B28 individuals, the distribution was 4, 24, and 28 in the ≤10, >10–50, and >50% subgroups, respectively. However, compared with patients with PCa tumors with absent or weak UGT2B28 staining, those with tumors presenting strong UGT2B28 nuclear staining (>50% of cells with +2/+3 staining scores) had significantly lower prostate-specific antigen (PSA) levels at diagnosis (Fig. 3B; Supplementary Fig. 2). The UGT2B28 tumor expression in both cellular compartments was also associated with tumor volume (Fig. 3C; Supplementary Fig. 3).
Tumors with higher UGT2B28 expression compared with individuals with low/absent expression were significantly smaller in volume by 30% (p ≤ 0.002). Also, a positive margin was associated with UGT2B28 nuclear staining, whereas positive nodal status and higher Gleason score were associated with higher UGT2B28 cytoplasmic staining (Fig. 3E and 3F). The association between higher expression of UGT2B28 and Gleason score values was further confirmed using The Cancer Genome Atlas data set (http://cancergenome.nih.gov/) and data from Taylor and colleagues (Fig. 3G and 3H) . In these two publicly available data sets and as compared with normal tissues, increased expression of UGT2B28 messenger RNA was apparent in primary tumors and further enhanced in metastasis (Supplementary Fig. 4). Although a trend was observed, the relationship between PSA levels and the Gleason scores was not statistically significant (Supplementary Fig. 5).
After adjustments for known prognostic markers including PSA, Gleason score, positive margins, nodal status, tumor size, smoking status, and germline UGT2B28 deletion status , overexpression of UGT2B28 was an adverse and independent prognostic factor associated with disease recurrence and/or death with a hazard ratio of 2.83 (95% confidence interval, 1.09–7.36; p = 0.033) (Supplementary Table 3). Similar effects of UGT2B28 on clinical outcomes were also observed in independent data sets  and .
3.3. UGT2B28 and circulating androgen levels
Of the 13 steroids measured in the circulation of the same cohort of 239 PCa patients studied by TMA, we observed higher circulating levels of T (p = 0.0043) and DHT (p = 0.002) by 30% associated with UGT2B28 nuclear staining (>50%) compared with individuals with absent expression (≤10%) (Table 2). Upon adjustment for UGT2B28 expression, there was an association between T level, tumor volume, and nodal status (Supplementary Table 4). Upon adjustment for T levels, we showed that the association of UGT2B28 on all prognostic factors remained statistically significant (Supplementary Table 4).
|UGT2B28 nuclear staining||UGT2B28 cytosolic staining|
|Steroid||Staining score, %||n||Mean*||95% CI||p value†||n||Mean*||95% CI||p value†|
* Total steroid levels (geometric mean).
† Only steroids with positive findings are shown. General linear model F test p values ≤0.05 are bold.
5-diol = androstenediol; DHT = dihydrotestosterone; T = testosterone.
Data are presented for patients with available data for tissue microarray and steroid levels.
3.4. A complete UGT2B28 deficiency is associated with an altered profile of circulating steroids
Characteristics of UGT2B28-deficient patients are depicted in Supplementary Table 1. Relative to UGT2B28 (+/+) carriers, patients carrying no copies of UGT2B28 (−/−) in their genomes (n = 51) presented significantly lower levels of circulating T, DHT, and their glucuronide metabolites, combined with a simultaneous 36% increase in the adrenal precursor androstenedione (p < 0.0001) (Table 3). Of these patients, 22 had T values in the hypogonadism range (<3 ng/ml), and 11 displayed extremely impaired androgen conjugation capacity with 3α-diol-17-glucuronide levels <1 ng/ml.
|Steroids||Unit||UGT2B28 germline status||Patients, n||Geometric mean||95% CI||% change (p value)|
4-dione = androstenedione; 3α-diol-3G = androstane-3α, 17β-diol-3-glucuronide; 3α-diol-17G = androstane-3α, 17β-diol-17-glucuronide; ADT-G = androsterone glucuronide; CI = confidence interval; DHT = dihydrotestosterone.
Total steroid levels (geometric mean). The symbol +/+ indicates two copies of UGT2B28; −/− indicates no copy of UGT2B28.
Only steroids with positive findings are shown. General linear model F test p values ≤ 0.05 are bold.
Steroid biotransformation in PCa cells is a dynamic process involving synthesis and catabolism. Based on the important clinical role of steroid biosynthesis and action in target cells , , , , and , we postulated that steroid inactivation by UGT2B28 is likely to have a role in the disease through the control of bioavailability of systemic and local potent androgens, especially in tumor cells. Our findings identify a new aberrant role for the inactivation pathway of sex steroids, which can also potentially serve as a useful marker to identify disease with aggressive potential, and clearly expose UGT2B28 as a regulator of steroidogenesis. To our knowledge, this is the first demonstration linking a disruption of steroid inactivation by UGT-mediated glucuronidation, modification of potent circulating steroids, and altered PCa disease characteristics and outcomes.
UGT2B28 expression was higher in primary prostate tumors and further amplified in metastases. We also exposed a differential compartmentalization in PIN lesions and cancer cells compared with normal prostate, further indicating an alteration in enzyme expression at a very early stage of tumorigenesis. This change in subcellular localization of UGT2B28 might be secondary to alternative splicing processes, as suggested by the existence of several alternative UGT2B28 variant transcripts in human tissues . This regulatory mechanism, combined with potential posttranslational modifications and/or interactions with unknown binding partners, may influence the association of UGT2B28 with the endoplasmic reticulum membrane and other cellular compartments , , and . Additional studies are required to address the underlying mechanisms of this differential compartmentalization that may participate in the regulation of UGT2B28 steroidogenic functions in PCa.
One key finding was that high UGT2B28 levels might serve as an independent predictor of risk of progression. In patients carrying one or two germline copies of UGT2B28, its overexpression in tumors had an impact on circulating hormone levels, disease characteristics, and risk of clinical progression. UGT2B28-mediated inactivation of steroid hormones likely modifies local steroid bioavailability required for aggressive growth of PCa cells while altering the transcription output of the androgen-dependent genes leading to enhanced metastatic potential. This was supported by the association of high UGT2B28 expression with more aggressive phenotypes defined by smaller tumor volume, low PSA levels, positive nodes, and higher Gleason score. This tumor UGT2B28 overexpression occurred in a high-androgen environment as shown by its association with elevated circulating levels of T and DHT.
Patients with complete UGT2B28 deficiency (ie, no copies of UGT2B28) had a substantial reduction in circulating testicular androgens and glucuronide metabolites. Increased exposure of circulating T and DHT would have been expected based on lower inactivation capacity created by the absence of the UGT2B28 inactivation pathway. However, in UGT2B28-deficient patients, 22 of 51 men had T values in the hypogonadism range (<3 ng/ml) with concomitant reduction in androgen conjugation capacity. It is thus tempting to hypothesize that UGT2B28 deficiency leads to a reprogramming of the endocrine system, adapting hormone bioavailability to conjugation capacity.
We previously showed that in patients with an active UGT2B28 pathway, a variation in the number of inherited copies of the gene is associated with an increased risk of BCR after prostatectomy in whites and Asians . These observations and those of the current study suggest the hypothesis that tumor UGT2B28 enzyme overexpression and the inherited deficiency of this inactivation pathway, as well as their accompanying consequences on hormone levels, are both associated with disease progression (Fig. 4). In agreement, a prior study showed that both the lowest and the highest preoperative serum T levels were associated with aggressive PCa . Moreover, low intraprostatic DHT level was previously associated with disease aggressiveness  and . However, knowing that a lack of correlation between circulating and intraprostatic hormone levels was previously reported , intraprostatic steroid hormone profiles in patients with UGT2B28 deficiency compared with patients with UGT2B28-overexpressing tumors will be required to capture all the consequences of this enzyme on steroidogenesis in cancer cells. However, thus far, the rarity of fresh-frozen prostate specimens to assess hormone levels complicates such analyses. Additional research is needed to establish the potential of a reciprocal feedback regulatory loop between androgens and the androgen-inactivating UGT2B28 enzyme. Because the other androgen-inactivating UGT2B15 and UGT2B17 enzymes are repressed by androgens , our data suggest that UGT2B28 is the active androgen-inactivating UGT isoform under high-androgen exposure.
The strengths of our study include (1) complete pathologic and clinical data combined with available circulating steroid hormone levels obtained through gold standard mass spectrometry assays, (2) the role of UGT2B28 in steroid metabolism, (3) a long follow-up time to assess clinical progression adequately, and (4) data for the first time from UGT2B28-deficient patients. Limitations are related to the availability of only one single blood steroid measurement and the fact that smaller associations may have been missed because of sample size limitations.
A significant heterogeneity in UGT2B28 expression exists in PCa tumors that affects plasma levels of potent androgens, disease aggressiveness, and risk of progression. Our data also suggest an essential role for UGT2B28 within the endocrine system; UGT2B28 has major repercussions on circulating hormone levels despite being placed at the end of the androgen-signaling pathway. Complete deficiency of UGT2B28 modifies the steroid hormone profile and is associated with low circulating levels of potent androgens and their metabolites, clearly indicating the impact of the enzyme on steroidogenesis. These findings may also have important clinical implications in other hormone-dependent cancers. The integration of all data regarding the steroid biotransformation/inactivation process in PCa will enable us to enhance our understanding of the disease and help capture at least a fraction of the fine regulation of steroid metabolism that occurs in these patients to improve therapeutic approaches. Additional studies performed in men lacking UGT2B28 and in high UGT2B28-expressing tumors will help us to better understand the critical role of the steroid-inactivation process that occurs throughout cancer progression.
Author contributions: Eric Lévesque had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Guillemette, Lacombe, Lévesque.
Acquisition of data: Belledant, Hovington, Garcia, Caron, Brisson, Villeneuve, Simonyan, Têtu, Fradet, Lacombe, Guillemette, Lévesque.
Analysis and interpretation of data: Belledant, Hovington, Garcia, Caron, Brisson, Villeneuve, Simonyan, Têtu, Fradet, Lacombe, Guillemette, Lévesque.
Drafting of the manuscript: Lévesque, Guillemette.
Critical revision of the manuscript for important intellectual content: Belledant, Hovington, Garcia, Caron, Brisson, Villeneuve, Simonyan, Têtu, Fradet, Lacombe, Guillemette, Lévesque.
Statistical analysis: Belledant, Garcia, Villeneuve, Simonyan.
Obtaining funding: Lévesque, Guillemette, Lacombe, Fradet.
Administrative, technical, or material support: Lévesque, Guillemette, Lacombe, Fradet.
Supervision: Guillemette, Lacombe, Lévesque.
Other (specify): None.
Financial disclosures: Eric Lévesque certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.
Funding/Support and role of the sponsor: This work was supported by Canadian research grants from Prostate Cancer Canada (to Eric Lévesque), Cancer Research Society (to Chantal Guillemette), the Canada Research Chair Program (to Chantal Guillemette), and the Fonds de Recherche du Québec-Santé (FRQ-S) Innovation fund to the CHU Research Centre (to Chantal Guillemette, Eric Lévesque, Louis Lacombe, and Yves Fradet). Eric Lévesque is the recipient of a Prostate Cancer Canada rising star award (RS2013-55). He is also the holder of a Canadian Institutes of Health Research clinician-scientist award. Chantal Guillemette holds the Canada Research Chair in Pharmacogenomics (Tier I).
Appendix A. Supplementary data
Supplementary Fig. 1 – Assessment of the specificity of the polyclonal antibody (Ab2321) against UGT2B28. (a) Peptide selected for the generation of the novel UGT2B28 polyclonal antibody aligned with homologous regions from six other UGT subfamily members. Amino acid sequence similarity is shown in black between members of this subfamily. (b) Western blot analysis using the polyclonal UGT2B28 antibody. Microsomal proteins (20 μg) from HEK293 cells stably expressing each of the UGT2B family members shown in (a) were separated by 10% SDS-PAGE, transferred to a nitrocellulose membrane, and probed with the polyclonal UGT2B28 antibody. (c) Microsomal proteins as in (a) were probed with the pan-UGT2B polyclonal EL-93 antibody as a positive control. Arrow indicates detected UGT proteins.
Supplementary Fig. 2 – Relationships between (a) prostate-specific antigen (PSA) and UGT2B28 protein expression (n = 239) in our study and (b) PSA and UGT2B28 mRNA expression generated from the Taylor data set .
Supplementary Fig. 3 – Relationship between UGT2B28 staining and pathologic staging in 239 patients with prostate cancer.
Supplementary Fig. 4 – UGT2B28 mRNA expression increases with disease progression. With the publicly available (a) Taylor et al  and (b) TCGA data sets, relative mRNA expression of UGT2B28 mRNA was evaluated in normal prostate gland, prostate cancer primary tumors, and in metastasis. *p ≤ 0.05; **p ≤ 0.01.
Supplementary Fig. 5 – Relationship between PSA and Gleason score in 239 patients with prostate cancer.
-  C.E. DeSantis, C.C. Lin, A.B. Mariotto, et al. Cancer treatment and survivorship statistics, 2014. CA Cancer J Clin. 2014;64:252-271 Crossref
-  C.J. Ryan, M.R. Smith, J.S. de Bono, et al. Abiraterone in metastatic prostate cancer without previous chemotherapy. N Engl J Med. 2013;368:138-148 Crossref
-  T.M. Beer, A.J. Armstrong, D.E. Rathkopf, et al. Enzalutamide in metastatic prostate cancer before chemotherapy. N Engl J Med. 2014;371:424-433 Crossref
-  H.I. Scher, K. Fizazi, F. Saad, et al. Increased survival with enzalutamide in prostate cancer after chemotherapy. N Engl J Med. 2012;367:1187-1197
-  J.S. de Bono, C.J. Logothetis, A. Molina, et al. Abiraterone and increased survival in metastatic prostate cancer. N Engl J Med. 2011;364:1995-2005 Crossref
-  R.B. Montgomery, E.A. Mostaghel, R. Vessella, et al. Maintenance of intratumoral androgens in metastatic prostate cancer: a mechanism for castration-resistant tumor growth. Cancer Res. 2008;68:4447-4454 Crossref
-  S.M. Green, E.A. Mostaghel, P.S. Nelson. Androgen action and metabolism in prostate cancer. Mol Cell Endocrinol. 2012;360:3-13 Crossref
-  A. Bélanger, G. Pelletier, F. Labrie, O. Barbier, S. Chouinard. Inactivation of androgens by UDP-glucuronosyltransferase enzymes in humans. Trends Endocrinol Metab. 2003;14:473-479
-  C. Guillemette, E. Levesque, M. Rouleau. Pharmacogenomics of human uridine diphospho-glucuronosyltransferases and clinical implications. Clin Pharmacol Ther. 2014;96:324-339 Crossref
-  E. Levesque, D. Turgeon, J.S. Carrier, V. Montminy, M. Beaulieu, A. Belanger. Isolation and characterization of the UGT2B28 cDNA encoding a novel human steroid conjugating UDP-glucuronosyltransferase. Biochemistry. 2001;40:3869-3881 Crossref
-  M. Beaulieu, E. Levesque, D.W. Hum, A. Belanger. Isolation and characterization of a novel cDNA encoding a human UDP-glucuronosyltransferase active on C19 steroids. J Biol Chem. 1996;271:22855-22862
-  E. Levesque, M. Beaulieu, M.D. Green, T.R. Tephly, A. Belanger, D.W. Hum. Isolation and characterization of UGT2B15(Y85): a UDP-glucuronosyltransferase encoded by a polymorphic gene. Pharmacogenetics. 1997;7:317-325 Crossref
-  S.A. McCarroll, T.N. Hadnott, G.H. Perry, et al. Common deletion polymorphisms in the human genome. Nat Genet. 2006;38:86-92 Crossref
-  I. Bronstad, A.S. Wolff, K. Lovas, P.M. Knappskog, E.S. Husebye. Genome-wide copy number variation (CNV) in patients with autoimmune Addison's disease. BMC Med Genet. 2011;12:111 Crossref
-  G. Nadeau, J. Bellemare, E. Audet-Walsh, et al. Deletions of the androgen-metabolizing UGT2B genes have an effect on circulating steroid levels and biochemical recurrence after radical prostatectomy in localized prostate cancer. J Clin Endocrinol Metab. 2011;96:E1550-E1557 Crossref
-  Y. Zhang, Y. Dai, Z. Tu, Q. Li, L. Wang, L. Zhang. Array-CGH detection of UGT2B28 gene deletion in a girl with primary amenorrhea and hyperandrogenism. Int J Gynaecol Obstet. 2010;109:164-166 Crossref
-  E. Audet-Walsh, J. Bellemare, L. Lacombe, et al. The impact of germline genetic variations in hydroxysteroid (17-beta) dehydrogenases on prostate cancer outcomes after prostatectomy. Eur Urol. 2012;62:88-96 Crossref
-  F. Brimo, A. Aprikian, M. Latour, et al. Strategies for biochemical and pathologic quality assurance in a large multi-institutional biorepository: the experience of the PROCURE Quebec Prostate Cancer Biobank. Biopreserv Biobank. 2013;11:285-290 Crossref
-  V. Menard, O. Eap, M. Harvey, C. Guillemette, E. Levesque. Copy-number variations (CNVs) of the human sex steroid metabolizing genes UGT2B17 and UGT2B28 and their associations with a UGT2B15 functional polymorphism. Hum Mutat. 2009;30:1310-1319
-  E. Levesque, S.P. Huang, E. Audet-Walsh, et al. Molecular markers in key steroidogenic pathways, circulating steroid levels, and prostate cancer progression. Clin Cancer Res. 2013;19:699-709 Crossref
-  E. Levesque, I. Laverdiere, E. Audet-Walsh, et al. Steroidogenic germline polymorphism predictors of prostate cancer progression in the estradiol pathway. Clin Cancer Res. 2014;20:2971-2983 Crossref
-  B.S. Taylor, N. Schultz, H. Hieronymus, et al. Integrative genomic profiling of human prostate cancer. Cancer Cell. 2010;18:11-22 Crossref
-  A.K. Kaushik, S.K. Vareed, S. Basu, et al. Metabolomic profiling identifies biochemical pathways associated with castration-resistant prostate cancer. J Proteome Res. 2014;13:1088-1100 Crossref
-  Tourancheau A, Margaillan G, Rouleau M, et al. Unravelling the transcriptomic landscape of the major phase II UDP-glucuronosyltransferase drug metabolizing pathway using targeted RNA sequencing. Pharmacogenomics J. In press. http://dx.doi.org/10.1038/tpj.2015.20
-  M. Rouleau, J. Roberge, J. Bellemare, C. Guillemette. Dual roles for splice variants of the glucuronidation pathway as regulators of cellular metabolism. Mol Pharmacol. 2014;85:29-36
-  A. Salonia, F. Abdollah, U. Capitanio, et al. Serum sex steroids depict a nonlinear u-shaped association with high-risk prostate cancer at radical prostatectomy. Clin Cancer Res. 2012;18:3648-3657 Crossref
-  T. Nishiyama, T. Ikarashi, Y. Hashimoto, K. Wako, K. Takahashi. The change in the dihydrotestosterone level in the prostate before and after androgen deprivation therapy in connection with prostate cancer aggressiveness using the Gleason score. J Urol. 2007;178:1282-1288 discussion 1288-9
-  T. Nishiyama, F. Ishizaki, T. Anraku, H. Shimura, K. Takahashi. The influence of androgen deprivation therapy on metabolism in patients with prostate cancer. J Clin Endocrinol Metab. 2005;90:657-660 Crossref
-  J. Heracek, R. Hampl, M. Hill, et al. Tissue and serum levels of principal androgens in benign prostatic hyperplasia and prostate cancer. Steroids. 2007;72:375-380 Crossref
-  C. Guillemette, E. Levesque, M. Beaulieu, D. Turgeon, D.W. Hum, A. Belanger. Differential regulation of two uridine diphospho-glucuronosyltransferases, UGT2B15 and UGT2B17, in human prostate LNCaP cells. Endocrinology. 1997;138:2998-3005
a Pharmacogenomics Laboratory, Centre Hospitalier Universitaire de Québec Research Center and Faculty of Pharmacy, Laval University, Québec, Canada
b Centre Hospitalier Universitaire de Québec Research Center, Faculty of Medicine, Laval University, Québec, Canada
c Clinical and Evaluative Research Platform, Centre Hospitalier Universitaire de Québec Research Center, Québec, Canada
Corresponding author. CHU de Québec Research Center, R4720, 2705 Boul. Laurier, Québec, G1V 4G2, Canada. Tel. +1 418 654 2296; Fax: +1 418 654 2298.
These authors contributed equally.
© 2015 European Association of Urology, Published by Elsevier B.V.