Purpose: Polymorphisms in the genes SRD5A1 and SRD5A2 encoding androgen biosynthetic 5α-reductase enzymes have been associated with an altered risk of biochemical recurrence after radical prostatectomy in localized prostate cancer.

Experimental Design: To gain potential insights into SRD5A biologic effects, we examined the relationship between SRD5A prognostic markers and endogenous sex-steroid levels measured by mass spectrometry in plasma samples and corresponding prostatic tissues of patients with prostate cancer.

Results: We report that five of the seven SRD5A markers differentially affect sex-steroid profiles of dihydrotestosterone and its metabolites in both the circulation and prostatic tissues of patients with prostate cancer. Remarkably, a 32% increase in intraprostatic testosterone levels was observed in the presence of the high-risk SRD5A rs2208532 polymorphism. Moreover, SRD5A2 markers were associated predominantly with circulating levels of inactive glucuronides. Indeed, the rs12470143 SRD5A2 protective allele was associated with high circulating androstane-3α, 17β-diol-17-glucuronide (3α-diol-17G) levels as opposed to lower levels of both 3α-diol-17G and androsterone-glucuronide observed with the rs2208532 SRD5A2 risk allele. Moreover, SRD5A2 rs676033 and rs523349 (V89L) risk variants, in strong linkage disequilibrium, were associated with higher circulating levels of 3α-diol-3G. The SRD5A2 rs676033 variant further correlated with enhanced intraprostatic exposure to 5α-reduced steroids (dihydrotestosterone and its metabolite 3β-diol). Similarly, the SRD5A1 rs166050C risk variant was associated with greater prostatic exposure to androsterone, whereas no association was noted with circulating steroids.

Conclusions: Our data support the association of 5α-reductase germline polymorphisms with the hormonal milieu in patients with prostate cancer. Further studies are needed to evaluate if these variants influence 5α-reductase inhibitor efficacy. Clin Cancer Res; 20(3); 576–84. ©2013 AACR.

Translational Relevance

Prostate cancer is a heterogeneous disease at the molecular and clinical levels requiring an individualized approach to patient care. The importance of androgen in cancer initiation and progression is clearly established at all disease stages. Moreover, 5α-reductase inhibitors (5-ARI) have shown potential clinical efficacy as chemopreventive medications. Likewise, SRD5A polymorphisms were recently associated with altered risk of biochemical recurrence in localized prostate cancer disease. To gain insights into SRD5A biologic effects, we examined the relationship between SRD5A prognostic markers and endogenous sex-steroid levels measured by gold-standard mass spectrometry methods in plasma and corresponding prostatic tissues of patients with prostate cancer. Our findings sustain the significance of common inherited SRD5A polymorphisms on the modulation of the androgenic milieu of patients with cancer. The utility of understanding the impact of SRD5A germline determinants on androgen exposure is that it could potentially be used to identify subgroups of patients, with a particular androgen microenvironment profile, that would most benefit from 5-ARI. Therefore, protective and high-risk germline SRD5A alleles associated with changes in hormone levels may lead to more personalized approaches to refining 5-ARI therapy, especially early in disease course.

Prostate cancer is the most frequent malignancy in men and the second leading cause of cancer death in Western countries (1). It is well established that androgens play a central role in prostate cancer progression even until it reaches advanced stages. Indeed, the conversion of testosterone (T) by 5α-reductase (5-AR) enzymes (types 1 and 2) leads to dihydrotestosterone (DHT), the most potent androgen receptor agonist in prostate cells. Under normal physiologic conditions, SRD5A2, which encodes a 5-AR, is preferentially expressed over SRD5A1 in the prostate (2, 3). In prostate cancer cells, however, the balance in expression of these genes shifts toward predominant expression of SRD5A1 (3, 4), supporting a role for both enzymes in DHT bioavailability and carcinogenesis. In fact, 5-AR enzymes represent attractive targets for preventing prostate cancer development. In that context, the importance of androgens in early cancer initiation is emphasized by the fact that finasteride, a 5-AR type 2 inhibitor, and dutasteride, a dual 5-AR inhibitor targeting both 5-AR type 1 and type 2 enzymes, have been shown to reduce by almost 23% the risk of prostate cancer incidence (5, 6). More recently, the REDEEM trial conducted with patients with low-risk prostate cancer showed that dutasteride also reduced by 10% the likelihood of cancer progression (7).

However, prostate cancer is clearly a heterogeneous disease, and prediction of response to treatment and progression remains a major challenge for the field. In addition to established tumor markers, certain patient genetic factors seem clearly associated with cancer progression. For instance, germline genetic polymorphisms in SRD5A, UGT2B, and HSD17B were shown to be associated with progression in Caucasian and Asian patients with prostate cancer after radical prostatectomy (8–10). In a previous study, 4 SRD5A2 single nucleotide polymorphisms (SNP; rs2208532, rs12470143, rs523349 (V89L), and rs4952197) were associated with biochemical recurrence (BCR) after radical prostatectomy in Caucasians and Asians (9). The strongest effect was conferred by the SRD5A2 V89L nonsynonymous SNP (rs523349C allele) with an HR of 2.87 [95% confidence interval (CI), 2.07–4.00; P = 4 × 10−10; 48% BCR). In addition, 2 SNPs, rs518673T in SRD5A1 and rs12470143A in SRD5A2, were associated with a reduced BCR rate (HR = 0.37; 95% CI, 0.19–0.71; P = 0.003 when combined; 16% BCR) compared with noncarriers (38% BCR). Such results support a significant effect of inherited genetic variations in the androgenic pathways on hormonal homeostasis and prostate cancer recurrence. Despite the identification of such potential biomarkers, however, there is a clear lack of biologic data explaining these findings.

It is hypothesized that polymorphisms in SRD5A genes affect the corresponding exposure to sex steroids of androgen-responsive cells, which may impact cell proliferation and cancer progression. At present, the influence of SRD5A polymorphisms linked to recurrence and progression in patients with prostate cancer on the systemic and prostatic sex-steroid hormonal environment remains undetermined. To gain potential insight into their biologic effects, we sought to evaluate the association of previously reported prognostic markers in SRD5A1 and SRD5A2 with levels of circulating and prostatic sex steroids in the same cohort of patients with prostate cancer.

This study was based on a cohort of 526 Caucasians diagnosed with prostate cancer recruited at the CHU de Québec-Hotel-Dieu de Québec Hospital (Québec, Canada) between 1999 and 2002, as described (8, 9). For the specific purpose of this study, available plasma and tissue samples were studied (numbers are indicated in Tables), and we excluded patients who received hormonal treatment. Patients did not receive 5-AR inhibitors (5-ARI) in this study. Two high-volume surgeons performed all surgeries for these patients (Drs. L. Lacombe and Y. Fradet). A fragment of fresh prostatic tissue was selected by the pathologist from the radical prostatectomy specimen in the area containing tumor tissue. Microdissection of tumors was not performed. Each specimen was then immediately snap-frozen and kept at −80°C. The remaining prostatic tissue was fixed and submitted in its entirety for histologic examination. Gleason score and pTNM stage were evaluated with paraffin sections. Detailed clinical information was available from a preoperative evaluation and medical records. All participants provided written consent before surgery for the analysis of their genome and corresponding hormone levels. The local research ethics committee approved the research protocol. Analyses described below were conducted blinded.

Genotyping

Genomic DNA was prepared from peripheral blood mononuclear cells collected from patients on the morning of a preoperative ambulatory clinical visit. All samples were kept frozen at −80°C until the time of study. Genomic DNA was extracted using the QIAamp DNA Blood Mini Kit (QIAGEN Inc.) and stored at −80°C. For genotyping, PCR was performed using Sequenom iPLEX matrix-assisted laser desorption/ionization-time-of-flight mass spectrometry, as described (9).

Analyses of steroid levels in plasma and prostatic tissue samples

Dehydroepiandrosterone (DHEA), 5-androsten-3β, 17β-diol (5-diol), testosterone (testo), DHT, androsterone (ADT), androstane-3β-17β-diol (3β-diol), estrone (E1), estradiol (E2), 4-androstenedione (4-dione), ADT-glucuronide (ADT-G), androstane-3α, 17β-diol 3-glucuronide (3α-diol-3G), androstane-3α, 17β-diol-17 glucuronide (3α-diol-17G), DHEA-sulfate (DHEA-S), and E1-Sulfate (E1-S) were purchased from Steraloids. Deuterated isotopes of hormones, namely DHEA-d3, testo-d3, DHT-d3, E1-d4, E2-d4, ADT-d3, 3β-diol-d3, 4-dione-d7, androstane-3α, 17β-diol 17-glucuronide-d3 (3α-diol-17G-d3), E1-S d4, and ADT-S d2, were purchased from C/D/N Isotopes. Plasma steroid measures were based on a procedure previously described (11, 12). Steroids measured in plasma were DHEA-S, DHEA, 5-diol, testo, DHT, 3β-diol, 4-dione, ADT, E1-S, E1, E2, ADT-G, 3α-DIOL-3G, and 3α-DIOL-17G. For tissue, a 50-mg frozen tissue section of prostate was transferred to a 12 × 75 mm glass tube containing 250 μL of 12.5 mmol/L ammonium bicarbonate and homogenized. To assess total steroids, namely total (unconjugated and conjugated) T, DHEA, 5-diol, DHT, 3β-diol, and ADT, 50 μL of the deuterated internal standards were added to the homogenized samples, following by the addition of 0.5 mL of freshly prepared hydrolysis buffer containing β-glucuronidase/sulfatase enzyme (Helix pomatia, type HP-2, ≥500 U β-glucuronidase, and ≥37.5 U of sulfatase activity) as described (13). Samples were incubated overnight at 37°C for hydrolysis and then extracted with 4 mL of chlorobutane:ethyl acetate (3:1) mixture. The organic layer was transferred to a clean glass tube, evaporated to dryness at 35°C under nitrogen gas, and dissolved with 100 μL of sodium bicarbonate (pH 9.0) and 100 μL of dansyl chloride (1 mg/mL in acetone). Then, 3 mL methanol:water (1:4) mixture was added to the sample, and solid-phase extraction was performed with a Strata-X Reversed SPE Phase Sorbents 60-mg columns (Phenomenex) previously conditioned with 1 mL methanol followed by 2 mL of water. The loaded cartridges were then sequentially washed with 3 mL of water and 3 mL of 55% methanol. The cartridges were dried under full vacuum. The androgenic compounds were eluted with 3 mL of 90% methanol and evaporated to dryness at 45°C under nitrogen gas. Next, androgens were derivatized and analyzed by gas chromatography–mass spectrometry as described (8). The sulfatase and β-glucuronidase mixture from H. pomatia was tested for steroid-converting enzymatic activities, and it was noted that this specific preparation presents some HSD3B activity (<3%, data not shown). Lower limit of quantification for T, DHEA, 5-diol, DHT, 3β-diol, and ADT were 300, 1,000, 500, 50, 100, and 250 pg/g, respectively.

Statistical analyses

To adjust for differences in the absolute levels of sex-steroid hormones, we calculated residuals of the natural logarithm of the hormone level regressed on age at blood donation and smoking status. We assessed the association of SNPs with variation in hormone levels by performing the regression of hormone residuals on each SNP independently for 2 models: recessive and dominant with one degree of freedom. We considered the association of a SNP with variation in hormone levels to be significant if the P value was ≤0.05. To facilitate comparisons between groups, we displayed the hormone level as untransformed data by using the geometric mean and SEM. Statistical analyses were performed using SAS Statistical Software version 9.2 (SAS Institute) and using PASW statistics version 17 (SPSS Inc.). Owing to the exploratory nature of this study on androgen metabolism in patients with prostate cancer, no correction was made for multiple testing.

Characteristics and descriptive statistics of the study cohort are listed in Table 1. The cohort was mostly composed of white men who underwent radical prostatectomy at L'Hôtel-Dieu de Québec Hospital between February 1999 and December 2002 (8). Tables 2 to 4 present data for prostatic and circulating steroid levels in samples from patients heterozygous or homozygous for the minor allele in comparison with levels measured in samples from homozygous carriers of the major allele.

Table 1.

Clinical and pathologic characteristics of the study population

CharacteristicsLocalized prostate cancer (n = 526)
Age at diagnosis (y)  
Mean 63.3 
SD 6.8 
Range 43.5–80.7 
Median follow up (mo) 88.8 
 Number (%) 
Biochemical recurrence 130 (24.7) 
PSA at diagnosis (ng/mL) 
≤10 362 (69) 
>10–20 103 (20) 
>20 56 (11) 
Pathologic Gleason score 
≤6 158 (31) 
244 (48) 
≥8 107 (21) 
Pathologic T stage 
pT = T2 313 (60) 
pT = T3a 131 (25) 
pT ≥ T3b 77 (15) 
Nodal invasion  
N0 481 (92) 
N+ 44 (8) 
Neoadjuvant hormonotherapy 
Yes 31 (6) 
No 495 (94) 
Adjuvant hormone therapy 
Yes 30 (6) 
No 496 (94) 
Margin status 
Negative 368 (70) 
Positive 154 (30) 
D'Amico risk classification 
Low 187 (36) 
Intermediate 208 (40) 
High 122 (24) 
CharacteristicsLocalized prostate cancer (n = 526)
Age at diagnosis (y)  
Mean 63.3 
SD 6.8 
Range 43.5–80.7 
Median follow up (mo) 88.8 
 Number (%) 
Biochemical recurrence 130 (24.7) 
PSA at diagnosis (ng/mL) 
≤10 362 (69) 
>10–20 103 (20) 
>20 56 (11) 
Pathologic Gleason score 
≤6 158 (31) 
244 (48) 
≥8 107 (21) 
Pathologic T stage 
pT = T2 313 (60) 
pT = T3a 131 (25) 
pT ≥ T3b 77 (15) 
Nodal invasion  
N0 481 (92) 
N+ 44 (8) 
Neoadjuvant hormonotherapy 
Yes 31 (6) 
No 495 (94) 
Adjuvant hormone therapy 
Yes 30 (6) 
No 496 (94) 
Margin status 
Negative 368 (70) 
Positive 154 (30) 
D'Amico risk classification 
Low 187 (36) 
Intermediate 208 (40) 
High 122 (24) 

Abbreviations: N, node; PSA, prostate-specific antigen; T, tumor.

Table 2.

Prostatic steroid levels in relation to SRD5A1 markers (n = 247)

Prostatic steroid levels in relation to SRD5A1 markers (n = 247)
Prostatic steroid levels in relation to SRD5A1 markers (n = 247)

SRD5A1 rs166050, which is associated with higher risk of BCR, was associated with higher levels of numerous prostatic steroids, particularly DHEA (60%, 25.94 ng/g vs. 15.76 ng/g; P = 0.045) and ADT (100%, 13.51 ng/g vs. 6.77 ng/g; P = 0.015), in homozygous carriers for the minor allele rs166050C (Table 2). No significant change in the concentrations of steroid hormones was observed in the circulation in relation to this SNP (Supplementary Table S1), both in circulating blood and prostatic tissues (Tables 2–4).

Table 3.

Prostatic steroid levels in relation to SRD5A2 markers (n = 247)

Prostatic steroid levels in relation to SRD5A2 markers (n = 247)
Prostatic steroid levels in relation to SRD5A2 markers (n = 247)
Table 4.

Levels of circulating steroid glucuronides in relation to SRD5A1 and SRD5A2 markers (n = 495)

Levels of circulating steroid glucuronides in relation to SRD5A1 and SRD5A2 markers (n = 495)
Levels of circulating steroid glucuronides in relation to SRD5A1 and SRD5A2 markers (n = 495)

Five SNPs in SRD5A2 were evaluated, of which 4 (rs2208532, rs4952197, rs523349 (V89L) and rs676033) were significantly associated with a higher risk of BCR, and 1 marker (rs12470143) with lower risk of BCR in Caucasians and Asians (9). Levels of prostate DHT and 3β-diol were particularly affected by one SRD5A2 polymorphism (rs676033). Indeed, patients homozygous for the risk allele rs676033A, in close genetic linkage with rs523349 (V89L), displayed 30% higher prostatic levels of DHT (P = 0.04) and 40% higher 3β-diol (P = 0.03) compared with GG/AG. In addition, plasma ADT levels were also significantly higher (17%; P = 0.04; Supplementary Table 2) in the presence of this polymorphism. Interestingly, tissue levels of 5-diol were higher for most unfavorable SNPs, reaching significance for rs4952197 (37% higher; P = 0.012) and rs523349 (30%; P = 0.030) but not attaining significance for rs2208532 (38%; P = 0.052) and rs676003 (26%; P = 0.058). Remarkably, we found a significant association between the SRD5A rs2208532 SNP and intraprostatic testosterone levels. Indeed, carriers of this high-risk allele had a 32% increase in prostatic testosterone levels (P = 0.038). An apparent lower level (20%) of prostatic testosterone was observed with the protective rs12470143 allele, but this difference was not statistically significant (P = 0.13). Table 3 lists details of the comprehensive effects of SRD5A2 SNPs on prostatic steroid hormone levels.

Of 14 steroids measured in plasma, significant associations were observed for glucuronide (-G) metabolites of DHT in relation to SRD5A2 markers (Fig. 1, Table 4, and Supplementary Table S2). The protective marker rs12470143A was associated with significantly higher 3α-diol-17G levels (15% for homozygotes; P = 0.048), whereas the opposite was observed for carriers of an unfavorable allele of that gene. In particular, rs2208532G was associated with significantly lower concentrations (10–20%) of ADT-G (P = 0.028) and 3α-diol-17G (P < 0.001). Interestingly, the risk alleles rs676033 and rs523349, in strong linkage disequilibrium in Caucasians (r2 = 0.90), were both associated with elevated levels of 3α-diol-3G (24% for rs523349CC, P = 0.036; 20% for rs676033AA, P = 0.039). Two SRD5A2 markers (rs12470143 and rs2208532) were also associated with an increase and decrease, respectively, in prostate volume at time of surgery (not shown).

Figure 1.

Positive associations observed for SRD5A2 markers and circulating androgen metabolite levels. Graphic representation of variation associated with 4 SNPs in SRD5A2 [rs2208532, rs4952197, rs523349 (V89L), and rs676033; gray tone histograms], which were significantly associated with a higher risk of biochemical recurrence (BCR), and 1 SNP (rs12470143; white histograms) with lower risk of BCR in this cohort of patients. Positive associations were observed with circulating levels of 3α-diol-3G (A), 3α-diol-17G (B), and ADT-G (C). Variation (%) compared with homozygotes of the major allele is shown on the y-axis.

Figure 1.

Positive associations observed for SRD5A2 markers and circulating androgen metabolite levels. Graphic representation of variation associated with 4 SNPs in SRD5A2 [rs2208532, rs4952197, rs523349 (V89L), and rs676033; gray tone histograms], which were significantly associated with a higher risk of biochemical recurrence (BCR), and 1 SNP (rs12470143; white histograms) with lower risk of BCR in this cohort of patients. Positive associations were observed with circulating levels of 3α-diol-3G (A), 3α-diol-17G (B), and ADT-G (C). Variation (%) compared with homozygotes of the major allele is shown on the y-axis.

Close modal

SRD5A genes encode rate-limiting enzymes involved in DHT formation in the prostate and several other tissues, particularly the liver and skin. Inhibition of these enzymes is the theoretical basis for chemoprevention strategies (5, 6) and was also recently shown to reduce prostate cancer progression in low-risk disease (14). In agreement with these previous findings, inherited germline variations in SRD5A1 and SRD5A2 were positively associated with BCR after radical prostatectomy in localized disease in both Caucasians and Asians (9). These results suggest that SRD5A genes may represent potential biomarkers of disease progression and response to treatment.

Our data suggest that SRD5A polymorphisms affect androgen metabolism in patients with prostate cancer as reflected by significant changes in circulating steroid glucuronide metabolites of the potent hormone DHT. Indeed, 4 of the 5 previously identified SRD5A2 markers (rs12470143, rs2208532, rs523349, rs676033) affect circulating glucuronide levels. The SRD5A2 rs12470143 protective marker is associated with higher levels of 3α-diol-17G in patients with prostate cancer, whereas the SRD5A2 rs2208532 risk allele is associated with lower levels of both 3α-diol-17G and ADT-G. Also, the genetically linked rs676033 and rs523349 risk variants are associated with increased levels of androgen glucuronides, namely 3α-diol-3G. Thus, these SRD5A markers significantly affect sex-steroid profiles of DHT metabolites in circulation of patients with prostate cancer. Indeed, T and DHT are rapidly transformed in the human prostate by several metabolic enzymes (15, 16). Moreover, DHT and its metabolites are further metabolized by UDP-glucuronosyltransferases (UGT), namely UGT2B15, UGT2B17, and UGT2B28, to yield their downstream inactive metabolites such as 3α-diol-3G, 3α-diol-17G, and ADT-G that are subsequently released into the circulation (Fig. 2; refs. 17–19). These major inactive steroids are considered biomarkers for intraprostatic DHT synthesis and exposure (20). Therefore, the presence of SRD5A2 prognostic markers impacts androgen formation in these men, whereas additional steroidogenic enzymes would be involved in DHT biotransformation before their inactivation by UGTs. In support of our findings, results obtained with the SRD5A2 rs12470143 and rs2208532 SNPs are in complete agreement with previous findings indicating an impact of these key genetic variations on circulating androgen metabolites rather than on circulating SRD5A substrates (21, 22).

Figure 2.

Schematic representation of sex-steroid biosynthesis pathways in patients with prostate cancer. SRD5A2 markers significantly affect the levels of prostatic androgens and circulating androgen metabolites (ADT-G, 3α-DIOL-3G, 3α-DIOL-17G), whereas the SRD5A1 rs166050C risk variant is correlated with greater prostatic exposure to DHT. UGT, UDP-glucuronosyltransferase; 4-dione, androstenedione; ADT, androsterone; ADT-G, androsterone-glucuronide; 3α-diol-3G, androstane-3α, 17β-diol 3-glucuronide; 3α-diol-17G, androstane-3α, 17β-diol 17-glucuronide; 5-diol, androst-5-ene-3β,17β-diol; 3α-diol, androstane-3α, 17β-diol; Testo, testosterone.

Figure 2.

Schematic representation of sex-steroid biosynthesis pathways in patients with prostate cancer. SRD5A2 markers significantly affect the levels of prostatic androgens and circulating androgen metabolites (ADT-G, 3α-DIOL-3G, 3α-DIOL-17G), whereas the SRD5A1 rs166050C risk variant is correlated with greater prostatic exposure to DHT. UGT, UDP-glucuronosyltransferase; 4-dione, androstenedione; ADT, androsterone; ADT-G, androsterone-glucuronide; 3α-diol-3G, androstane-3α, 17β-diol 3-glucuronide; 3α-diol-17G, androstane-3α, 17β-diol 17-glucuronide; 5-diol, androst-5-ene-3β,17β-diol; 3α-diol, androstane-3α, 17β-diol; Testo, testosterone.

Close modal

Moreover, SRD5A variants are associated with sex-steroid exposure in the prostate of patients with cancer. Remarkably, the rs2208532 SNP was linked with a 32% increase of intraprostatic testosterone levels, combined with reduced circulating glucuronide levels, suggesting that this high-risk allele is associated with reduced 5α-reductase enzyme efficiency. In the same line of thought, in the presence of the low-risk allele rs12470143, a non-significant reduction (16%) in prostatic testosterone levels combined with higher glucuronide levels suggest that this SNP is associated with high enzyme activity. Interestingly, the nonsynonymous V89L variant of SRD5A2 (rs523349), previously associated with aggressive forms of the disease (23), has been described as a low-activity allele in vitro (24). This coding SRD5A2 variant has been reported to impact sex-steroid concentrations in different ethnic groups, whereas an ethnic-specific distribution and linkage with other SRD5A2 alleles also has been observed (21, 24–27). Measurement of steroid levels in the prostate revealed that individuals carrying a tightly linked variation to rs523349 (e.g., rs676033) display significantly higher intraprostatic DHT and 3β-diol levels, higher circulating levels of ADT (a 5α-reduced metabolite of DHT; Fig. 2), and no significant alteration in circulating T or DHT levels. According to these observations, we hypothesize that the V89L rs523349 SNP or another variation in linkage disequilibrium such as rs676033 located in the promoter region of the gene would modulate 5α-reductase activity/expression within prostate cells. Further studies are definitely required to evaluate the molecular impact of the rs523349/rs676033 in target cells of patients with cancer.

Regarding SRD5A1 markers, no significant changes were observed in circulating sex-steroid levels. Among the statistically significant results, our data reveal an accumulation of ADT without changes in steroid glucuronide levels, as observed for SRD5A2 markers. These higher levels of ADT indicate that this variation might enhance enzyme activity, also hypothetically favoring BCR in these patients. As previously suggested (28), this accumulation of ADT combined with the absence of significant changes in T levels in SRD5A carriers may also indicate that the alternate route of androgen biosynthesis not involving T may perhaps predominate in these individuals. In addition, high levels of ADT in these patients could theoretically fuel DHT synthesis by the concerted action of other pathways such as the HSD17B3 and HSD17B6 enzymes, supporting androgen formation and conceivably recurrence of elevated prostate-specific antigen.

Unfavorable SNPs in SRD5A2 are also linked to an elevation at the target cell level of other potent steroid hormones such as 5-diol secreted by the adrenals. The mechanism leading to enhanced exposure of 5-diol in these individuals may be a result of additional and complex interplay between 5α-reductases and other key steroidogenic enzymes expressed in the prostate, for example, CYP17A, HSD3B, and HSD17B. If further confirmed in additional independent studies, this enhanced exposure to 5-diol may be of biologic significance in the context of prostate cancer progression because this adrenal precursor was elegantly shown to be a natural hormone with androgenic properties in human prostate cancer cells (29, 30). Indeed, 5-diol can activate the androgen receptor without being metabolized to T or DHT, especially in the presence of the coactivator ARA70 that notably enhances its androgenic properties (29). Moreover, 5-diol also has estrogenic activity at physiologic concentrations and can bind the estrogen receptor α (ERα), although at low levels in the normal prostate, and trigger an estrogenic response (31). Thus, additional studies of this sex-steroid hormone are warranted in the context of cancer progression at different stages of the disease. Moreover, the level of 3β-diol, an important metabolite of DHT, is also altered in the presence of SRD5A prognostic marker (rs676033) and may potentially impact cancer progression as well because it is the endogenous ligand of ERβ in human prostate (32). ERβ is the most abundant estrogenic receptor in prostatic basal cells and is involved in major cellular pathways, including inflammation processes, differentiation, and apoptosis, thus supporting a role for this nonaromatized estrogenic steroid in the prostate (32, 33).

Overall, inherited SRD5A variations seem to have noteworthy physiopathologic consequences. Indeed, any subtle and persistent variation in sex-steroid hormone levels and/or their relative ratios may modify the course of a slow-progressing disease such as prostate cancer over several years. Data further reinforce the significance of SRD5A genes in hormone metabolism and the biologic relevance of previous associations observed between this pathway and clinical outcomes. The assessment of SRD5A markers based on individual patients' germline genetic variations may also lead to a better patient stratification in future 5-ARI clinical trials, targeting therapeutic interventions to optimize hormonal manipulation in patients with a high risk of recurrence and helping to identify patients who would more likely benefit from treatment. However, it is currently unknown if any SRD5A genotypes, such as the low-risk SRD5A2 allele rs12470143 or any of the high-risk variants (rs2208532, rs676033, rs523349), may modify 5-ARI response. The impact of these common germline polymorphisms should certainly be addressed in future studies.

To our knowledge, this is the first report of sex-steroid measurements in both the circulation and prostatic tissues of patients with prostate cancer in relation to inherited genetic markers associated with recurrence. The strengths of the study include the use of gold standard mass spectrometry–based sex-steroid assays, fasting blood sampling on the morning of the surgery for all patients with paired prostatic tissues collected at prostatectomy, the selection of SRD5A markers (n = 7) associated with biochemical recurrence (9) and the biologic plausibility of the association. Limitations are related to (i) the availability of only single blood and tissue samples for measurement of steroids and (ii) quantification of steroids in available prostate cancer samples containing surrounding peritumoral tissues, because no tumor microdissection was performed. The latter may have resulted in potential underestimation of the impact of these germline variants on overall androgen metabolism in cancer samples. Hence, although this is the first report of sex steroids measured in human prostatic tissues in relation to germline SNPs in key androgenic genes, one must bear in mind that the number of cases assessed was limited owing to the infrequency of acquiring samples from which a relatively large quantity of fresh-frozen prostate cancer tissue could be obtained for steroid measurements. Therefore, these analyses are exploratory, with no attempt to correct for multiplicity. Additional studies analyzing both circulating and prostatic tissue levels of a wide range of sex-steroid hormones in patients with prostate cancer are clearly necessary to gain important information on steroid biotransformation and prostate cancer progression.

We conclude that the assessment of host genetic variants in key steroidogenic pathways, such as those governed by SRD5A genes, represents additional indications that the inherited genetic background influences the hormonal microenvironment to which cancer cells are exposed. Our results support that SRD5A genetic variations modify sex-steroid exposure to potentially promote cancer growth and proliferation. Further studies are required to fully characterize at the molecular level the impact of functional variations in SRD5A genes in both normal and prostate cancer cells. Moreover, markers in SRD5A genes, especially the SRD5A1 rs166050, SRD5A2 rs12470143, rs2208532, and rs676033/rs523349 SNPs, may ultimately represent clinically relevant prostate cancer indicators and lead to more personalized management of the most prevalent cancer in men, especially early in the course of the disease.

E. Lévesque, Y. Fradet, L. Lacombe, C. Guillemette, have ownership interest (including patents). No potential conflicts of interest were disclosed by the other authors.

Conception and design: E. Lévesque, C. Guillemette

Development of methodology: E. Lévesque, C. Guillemette

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): E. Lévesque, I. Laverdière, L. Lacombe, P. Caron, V. Turcotte, B. Têtu, C. Guillemette

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): E. Lévesque, I. Laverdière, M. Rouleau, V. Turcotte, C. Guillemette

Writing, review, and/or revision of the manuscript: E. Lévesque, I. Laverdière, L. Lacombe, M. Rouleau, B. Têtu, Y. Fradet, C. Guillemette

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): E. Lévesque, L. Lacombe, C. Guillemette

Study supervision: E. Lévesque, C. Guillemette

The authors thank Dr. A. Bélanger for helpful discussion, the biostatistics services of the clinical research platform (CHUQ Research Center) for their support, as well as M. Orain, J. Ouellette, É. Audet-Walsh, and C. Flageole for their technical support in the preparation of tissue samples before extraction and measurement of hormones by mass spectrometry.

This work was supported by Canadian research grants from the Cancer Research Society (C. Guillemette), Fonds de recherche du Québec–Santé (FRQS), and the Canada Research Chair Program (C. Guillemette). E. Lévesque is the recipient of a Prostate Cancer Canada rising star award (RS2013-55). C. Guillemette holds the Canada Research Chair in Pharmacogenomics (Tier II). I. Laverdière and M. Rouleau are both recipients of a Frederick Banting and Charles Best Canada Graduate Scholarship award from the CIHR. I. Laverdière is a recipient of a clinician-scientist graduate scholarship from FRQS.

1.
Jemal
A
,
Bray
F
,
Center
MM
,
Ferlay
J
,
Ward
E
,
Forman
D
. 
Global cancer statistics
.
CA Cancer J Clin
2011
;
61
:
69
90
.
2.
Thomas
LN
,
Lazier
CB
,
Gupta
R
,
Norman
RW
,
Troyer
DA
,
O'Brien
SP
, et al
Differential alterations in 5α-reductase type 1 and type 2 levels during development and progression of prostate cancer
.
Prostate
2005
;
63
:
231
9
.
3.
Titus
MA
,
Gregory
CW
,
Ford
OH
 3rd
,
Schell
MJ
,
Maygarden
SJ
,
Mohler
JL
. 
Steroid 5α-reductase isozymes I and II in recurrent prostate cancer
.
Clin Cancer Res
2005
;
11
:
4365
71
.
4.
Stanbrough
M
,
Bubley
GJ
,
Ross
K
,
Golub
TR
,
Rubin
MA
,
Penning
TM
, et al
Increased expression of genes converting adrenal androgens to testosterone in androgen-independent prostate cancer
.
Cancer Res
2006
;
66
:
2815
25
.
5.
Andriole
GL
,
Bostwick
DG
,
Brawley
OW
,
Gomella
LG
,
Marberger
M
,
Montorsi
F
, et al
Effect of dutasteride on the risk of prostate cancer
.
N Engl J Med
2010
;
362
:
1192
202
.
6.
Thompson
IM
,
Goodman
PJ
,
Tangen
CM
,
Lucia
MS
,
Miller
GJ
,
Ford
LG
, et al
The influence of finasteride on the development of prostate cancer
.
N Engl J Med
2003
;
349
:
215
24
.
7.
Fleshner
NE
,
Lucia
MS
,
Egerdie
B
,
Aaron
L
,
Eure
G
,
Nandy
I
, et al
Dutasteride in localised prostate cancer management: the REDEEM randomised, double-blind, placebo-controlled trial
.
Lancet
2012
;
379
:
1103
11
.
8.
Nadeau
G
,
Bellemare
J
,
Audet-Walsh
E
,
Flageole
C
,
Huang
SP
,
Bao
BY
, 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
7
.
9.
Audet-Walsh
E
,
Bellemare
J
,
Nadeau
G
,
Lacombe
L
,
Fradet
Y
,
Fradet
V
, et al
SRD5A Polymorphisms and biochemical failure after radical prostatectomy
.
Eur Urol
2011
;
60
:
1226
34
.
10.
Audet-Walsh
E
,
Bellemare
J
,
Lacombe
L
,
Fradet
Y
,
Fradet
V
,
Douville
P
, et al
The impact of germline genetic variations in hydroxysteroid (17-β) dehydrogenases on prostate cancer outcomes after prostatectomy
.
Eur Urol
2012
;
62
:
88
96
.
11.
Audet-Walsh
E
,
Lepine
J
,
Gregoire
J
,
Plante
M
,
Caron
P
,
Têtu
B
, et al
Profiling of endogenous estrogens, their precursors, and metabolites in endometrial cancer patients: association with risk and relationship to clinical characteristics
.
J Clin Endocrinol Metab
2011
;
96
:
E330
9
.
12.
Lepine
J
,
Audet-Walsh
E
,
Gregoire
J
,
Têtu
B
,
Plante
M
,
Menard
V
, et al
Circulating estrogens in endometrial cancer cases and their relationship with tissular expression of key estrogen biosynthesis and metabolic pathways
.
J Clin Endocrinol Metab
2010
;
95
:
2689
98
.
13.
Blonder
J
,
Johann
DJ
,
Veenstra
TD
,
Xiao
Z
,
Emmert-Buck
MR
,
Ziegler
RG
, et al
Quantitation of steroid hormones in thin fresh frozen tissue sections
.
Anal Chem
2008
;
80
:
8845
52
.
14.
Fleshner
NE
. 
Dutasteride and active surveillance of low-risk prostate cancer
.
Lancet
2012
;
379
:
1590
.
15.
Montgomery
RB
,
Mostaghel
EA
,
Vessella
R
,
Hess
DL
,
Kalhorn
TF
,
Higano
CS
, et al
Maintenance of intratumoral androgens in metastatic prostate cancer: a mechanism for castration-resistant tumor growth
.
Cancer Res
2008
;
68
:
4447
54
.
16.
Mitsiades
N
,
Sung
CC
,
Schultz
N
,
Danila
DC
,
He
B
,
Eedunuri
VK
, et al
Distinct patterns of dysregulated expression of enzymes involved in androgen synthesis and metabolism in metastatic prostate cancer tumors
.
Cancer Res
2012
;
72
:
6142
52
.
17.
Beaulieu
M
,
Lévesque
E
,
Hum
DW
,
Belanger
A
. 
Isolation and characterization of a novel cDNA encoding a human UDP-glucuronosyltransferase active on C19 steroids
.
J Biol Chem
1996
;
271
:
22855
62
.
18.
Belanger
A
,
Pelletier
G
,
Labrie
F
,
Barbier
O
,
Chouinard
S
. 
Inactivation of androgens by UDP-glucuronosyltransferase enzymes in humans
.
Trends Endocrinol Metab
2003
;
14
:
473
9
.
19.
Lévesque
E
,
Turgeon
D
,
Carrier
JS
,
Montminy
V
,
Beaulieu
M
,
Belanger
A
. 
Isolation and characterization of the UGT2B28 cDNA encoding a novel human steroid conjugating UDP-glucuronosyltransferase
.
Biochemistry
2001
;
40
:
3869
81
.
20.
Labrie
F
,
Belanger
A
,
Belanger
P
,
Berube
R
,
Martel
C
,
Cusan
L
, et al
Androgen glucuronides, instead of testosterone, as the new markers of androgenic activity in women
.
J Steroid Biochem Mol Biol
2006
;
99
:
182
8
.
21.
Jiang
J
,
Tang
NL
,
Ohlsson
C
,
Eriksson
AL
,
Vandenput
L
,
Liao
C
, et al
Association of SRD5A2 variants and serum androstane-3α,17β-diol glucuronide concentration in Chinese elderly men
.
Clin Chem
2010
;
56
:
1742
9
.
22.
Ahn
J
,
Schumacher
FR
,
Berndt
SI
,
Pfeiffer
R
,
Albanes
D
,
Andriole
GL
, et al
Quantitative trait loci predicting circulating sex steroid hormones in men from the NCI-Breast and Prostate Cancer Cohort Consortium (BPC3)
.
Hum Mol Genet
2009
;
18
:
3749
57
.
23.
Cussenot
O
,
Azzouzi
AR
,
Nicolaiew
N
,
Mangin
P
,
Cormier
L
,
Fournier
G
, et al
Low-activity V89L variant in SRD5A2 is associated with aggressive prostate cancer risk: an explanation for the adverse effects observed in chemoprevention trials using 5-α-reductase inhibitors
.
Eur Urol
2007
;
52
:
1082
7
.
24.
Makridakis
N
,
Ross
RK
,
Pike
MC
,
Chang
L
,
Stanczyk
FZ
,
Kolonel
LN
, et al
A prevalent missense substitution that modulates activity of prostatic steroid 5α-reductase
.
Cancer Res
1997
;
57
:
1020
2
.
25.
Hsing
AW
,
Chen
C
,
Chokkalingam
AP
,
Gao
YT
,
Dightman
DA
,
Nguyen
HT
, et al
Polymorphic markers in the SRD5A2 gene and prostate cancer risk: a population-based case-control study
.
Cancer Epidemiol Biomarkers Prev
2001
;
10
:
1077
82
.
26.
Febbo
PG
,
Kantoff
PW
,
Platz
EA
,
Casey
D
,
Batter
S
,
Giovannucci
E
, et al
The V89L polymorphism in the 5α-reductase type 2 gene and risk of prostate cancer
.
Cancer Res
1999
;
59
:
5878
81
.
27.
Allen
NE
,
Reichardt
JK
,
Nguyen
H
,
Key
TJ
. 
Association between two polymorphisms in the SRD5A2 gene and serum androgen concentrations in British men
.
Cancer Epidemiol Biomarkers Prev
2003
;
12
:
578
81
.
28.
Chang
KH
,
Li
R
,
Papari-Zareei
M
,
Watumull
L
,
Zhao
YD
,
Auchus
RJ
, et al
Dihydrotestosterone synthesis bypasses testosterone to drive castration-resistant prostate cancer
.
Proc Natl Acad Sci U S A
2011
;
108
:
13728
33
.
29.
Miyamoto
H
,
Yeh
S
,
Lardy
H
,
Messing
E
,
Chang
C
. 
Δ5-androstenediol is a natural hormone with androgenic activity in human prostate cancer cells
.
Proc Natl Acad Sci U S A
1998
;
95
:
11083
8
.
30.
Chang
HC
,
Miyamoto
H
,
Marwah
P
,
Lardy
H
,
Yeh
S
,
Huang
KE
, et al
Suppression of Δ(5)-androstenediol-induced androgen receptor transactivation by selective steroids in human prostate cancer cells
.
Proc Natl Acad Sci U S A
1999
;
96
:
11173
7
.
31.
Hackenberg
R
,
Turgetto
I
,
Filmer
A
,
Schulz
KD
. 
Estrogen and androgen receptor mediated stimulation and inhibition of proliferation by androst-5-ene-3β,17β-diol in human mammary cancer cells
.
J Steroid Biochem Mol Biol
1993
;
46
:
597
603
.
32.
Oliveira
AG
,
Coelho
PH
,
Guedes
FD
,
Mahecha
GA
,
Hess
RA
,
Oliveira
CA
. 
5alpha-Androstane-3β,17β-diol (3β-diol), an estrogenic metabolite of 5α-dihydrotestosterone, is a potent modulator of estrogen receptor ERβ expression in the ventral prostrate of adult rats
.
Steroids
2007
;
72
:
914
22
.
33.
Hartman
J
,
Strom
A
,
Gustafsson
JA
. 
Current concepts and significance of estrogen receptor β in prostate cancer
.
Steroids
2012
;
77
:
1262
6
.