Published OnlineFirst August 22, 2017; DOI: 10.1158/1055-9965.EPI-17-0215 Research Article Cancer Epidemiology, Biomarkers Relationships between Circulating and & Prevention Intraprostatic Sex Steroid Hormone Concentrations Michael B. Cook, Frank Z. Stanczyk, Shannon N. Wood, Ruth M. Pfeiffer, Muhannad Hafi, Carmela C. Veneroso, Barlow Lynch, Roni T. Falk, Cindy Ke Zhou, Shelley Niwa, Eric Emanuel, Yu-Tang Gao, George P. Hemstreet, Ladan Zolfghari, Peter R. Carroll, Michael J. Manyak, Isabell A. Sesterhann, Paul H. Levine, and Ann W. Hsing Abstract Background: Sex hormones have been implicated in prostate Results: Circulating sex steroid hormone concentrations carcinogenesis, yet epidemiologic studies have not provided had low-to-moderate correlations with, and explained small substantiating evidence. We tested the hypothesis that circulat- proportions of variations in, intraprostatic sex steroid hor- ing concentrations of sex steroid hormones reflect intraprostatic mone concentrations. Androstane-3a,17b-diol glucuronide concentrations using serum and adjacent microscopically ver- (3a-diol G) explained the highest variance of tissue concen- ified benign prostate tissue from prostate cancer cases. trations of 3a-diol G (linear regression r2 ¼ 0.21), followed Methods: Incident localized prostate cancer cases scheduled by serum testosterone and tissue dihydrotestosterone (r2 ¼ for surgery were invited to participate. Consented participants 0.10), and then serum estrone and tissue estrone (r2 ¼ 0.09). completed surveys, and provided resected tissues and blood. There was no effect modification by Gleason score, race, Histologic assessment of the ends of fresh frozen tissue confirmed or age. adjacent microscopically verified benign pathology. Sex steroid Conclusions: Circulating concentrations of sex steroid hormones in sera and tissues were extracted, chromatographically hormones are poor surrogate measures of the intraprostatic separated, and then quantitated by radioimmunoassays. Linear hormonal milieu. regression was used to account for variations in intraprostatic Impact: The high exposure misclassification provided by cir- hormone concentrations by age, body mass index, race, and study culating sex steroid hormone concentrations for intraprostatic site, and subsequently to assess relationships with serum hormone levels may partly explain the lack of any consistent association of concentrations. Gleason score (from adjacent tumor tissue), race, circulating hormones with prostate cancer risk. Cancer Epidemiol and age were assessed as potential effect modifiers. Biomarkers Prev; 26(11); 1660–6. Ó2017 AACR. Introduction developed male phenotype, including a small and immature prostate gland (1, 2). The Nobel Prize–winning studies by Prostate cancer has long been hypothesized to have a hor- Huggins and Hodges in 1941 reported that castration and monal pathogenesis. Endogenous sex steroid hormones, par- injection of estrogen cause temporary regression of metastatic ticularly androgens, are undoubtedly essential for normal prostate cancer, implicating androgenic action in prostate can- physiological development, maintenance, and function of the cer progression (3). This led to development of androgen prostate gland. Prepubertally castrated men and male pseudo- deprivation therapy, which remains the mainstay therapy for hermaphrodites with deficient 5a-reductase type II have a mal- men with advanced prostate cancer. Androgen signaling also functions in cell proliferation, differentiation, and apoptosis, and evidence from basic science indicates that androgens, and Division of Cancer Epidemiology and Genetics, National Cancer Institute, NIH, possibly estrogens, are critically important for prostate carcino- Bethesda, Maryland. genesis (4–6). Note: Supplementary data for this article are available at Cancer Epidemiology, Despite this evidence that implicates sex steroid hormones Biomarkers & Prevention Online (http://cebp.aacrjournals.org/). in prostate cancer pathogenesis, epidemiologic studies that Corresponding Authors: Michael B. Cook, National Cancer Institute, NIH, have assessed prediagnostic circulating hormone concentra- 9609 Medical Center Drive, Rm 7-E106, MSC 9774, Bethesda, MD 20892- tions have not found any consistent association with subse- 9774. Phone: 240-276-7298; E-mail: [email protected]; and Ann W. quent prostate cancer risk (7–9). There are various explana- Hsing, Stanford Cancer Institute, Stanford School of Medicine, Stanford tions for why a true association may have been missed, includ- University, 780 Welch Road, CJ250D, Stanford, CA 94305. E-mail: ing interassay variability, lack of assay standardization, use [email protected]. of a single peripheral blood measurement typically at middle doi: 10.1158/1055-9965.EPI-17-0215 age or later, and case heterogeneity with inclusion of a vari- Ó2017 American Association for Cancer Research. able proportion of indolent disease. Regardless of the true 1660 Cancer Epidemiol Biomarkers Prev; 26(11) November 2017 Downloaded from cebp.aacrjournals.org on September 26, 2021. © 2017 American Association for Cancer Research. Published OnlineFirst August 22, 2017; DOI: 10.1158/1055-9965.EPI-17-0215 Circulating and Intraprostatic Hormones explanation, an inherent assumption of the prior observation- mont Report, U.S. Common Rule) and was preapproved by the al studies is that circulating concentrations are proxies of the required institutional review boards. intraprostatic environment. Testosterone (T) and the more potent metabolite, dihydrotestosterone (DHT), bind the Serum hormone quantitation androgen receptor within the prostate eliciting gene expression Serum levels of androstenedione (A), testosterone (T), DHT, profiles and biological effects that maintain prostate function. 5a-androstane-3a,17b-diol glucuronide (3a-diol G), estrone T is predominantly produced by the testes and released into the (E1), and estradiol (E2) were measured at the University of circulation. DHT, however, is primarily produced within the Southern California (under the direction of FZS) by immunoassay prostate gland, thus circulating DHT precursors [T, androstene- methods during 2007. A, T, DHT, E1, and E2 were measured by RIA dione (A)] and metabolites [5a-androstane-3a,17b-diol glu- with preceding purification steps, including extraction of steroids curonide (3a-diol G)] have traditionally been assessed as with ethyl acetate:hexane (3:2) followed by Celite column par- proxies. The validity of these proxies has not been tested. tition chromatography of individual hormones using ethylene Therefore, we set out to test the hypothesis that circulating glycol as the stationary phase (10–13). For optimum chro- sex steroid hormone concentrations are valid proxies of intra- matographic steroid separation, it was necessary to process A, prostatic concentrations using a large set of blood samples T, E1 and E2 in one assay (hormone set 1 tubes) and DHT in paired with microscopically verified benign tissue samples another assay (hormone set 2 tubes); 0.8 mL serum aliquots were adjacent to prostate cancers. used in each assay. In the first assay, chromatographic elution of A, T, E1, and E2 was achieved with isooctane, 40% toluene in Materials and Methods isooctane, 25% toluene in ethyl acetate, and 40% toluene in isooctane, respectively. In the second assay, DHT was eluted Study population with 10% toluene in isooctane. In each RIA, a highly specific Patients were enrolled in the study between January 2000 and antiserum was used in conjunction with an iodinated radio- April 2004 at five locations: George Washington University Med- ligand, and an appropriate 8-point standard curve was con- ical Center (Washington, DC), University of California at San structed. After an appropriate incubation period, antibody-bound Francisco (San Francisco, CA), Doctor's Community Hospital and unbound hormones were separated by use of a second (Lanham-Seabrook, MD), Washington Hospital Center (Washing- antibody. The resulting raw values were corrected for dilution ton, DC), and INOVA Fairfax Hospital (Falls Church, VA), the factors and procedural losses. latter three of which were primarily coordinated by the staff at 3a-diol G was measured using a commercial 3a-diol G RIA George Washington University Medical Center. Study subject Kit (Diagnostics Systems Laboratories, Webster TX, presently eligibility included: 18 years of age or older; scheduled for radical Beckman-Coulter). The assay measured both isomers of 3a-diol prostatectomy; and newly diagnosed with localized prostate can- G (14). It required no preceding purification steps and was cer. Patients provided written informed consent to be part of the validated extensively in the laboratory. study. Prior to surgery, study patients had standard anthropomet- The assay sensitivities were as follows: 3, 0.5, 1.5, 50, 0.4, and ric measures taken and were administered a questionnaire to 0.2 ng/dL for A, DHT, T, 3a-diol G, E1, and E2, respectively. confirm that they were fasting and had not taken any hormones Coefficients of variation (CV), calculated on the logarithmic scale (e.g., DHEA) or substances that could potentially affect hormone using a mixed model which included an average of four blinded concentrations (e.g., finasteride) in the preceding 24 hours. Study technical replicates from each of seven different individuals subjects also provided 30 mL of blood, which were processed assayed at the same time as the sera for the main analysis across within 4 hours into aliquots of serum,