|Home | About | Journals | Submit | Contact Us | Français|
To identify adrenocortical hormone abnormalities as indicators of endocrine dysfunction in CP/CPPS.
We simultaneously measured the serum concentrations of 12 steroids in CP/CPPS and control patients, using isotope dilution liquid chromatography followed by atmospheric pressure photospray ionization and tandem mass spectrometry.
Twenty-seven CP/CPPS patients and 29 age-matched asymptomatic healthy controls were evaluated. In the mineralocorticoid pathway, progesterone was significantly higher, whereas corticosterone and aldosterone concentrations were significantly lower, in CP/CPPS than in controls. In the glucocorticoid pathway, 11-deoxycortisol was significantly lower, and cortisol concentrations were not different between patients and controls. In the sex steroid pathway, androstenedione and testosterone concentrations were significantly higher in CP/CPPS than in controls. Estradiol, dehydroepiandrosterone (DHEA) and dehydroepiandrosterone sulfate (DHEAS) concentrations were not different between patients and controls. NIH-CPSI total and pain domain scores correlated positively with 17-hydroxyprogesterone and aldosterone (P<0.001) and negatively with cortisol concentrations (P<0.001).
Results suggest reduced activity of CYP21A2 (P450c21), the enzyme that converts progesterone to corticosterone, and 17-hydroxyprogesterone to 11-deoxycortisol. Furthermore, these results provide insights into the biological basis of CP/CPPS. Follow-up studies should explore the possibility that CP/CPPS patients meet the diagnostic criteria for nonclassical CAH and if hormonal findings improve or worsen in parallel with symptom severity.
The search for an etiology or biomarker for the prevalent and clinically important disease 1,2, chronic prostatitis/chronic pelvic pain syndrome (CP/CPPS), has thus far proven elusive.1 Evidence suggest the possibility that CP/CPPS symptoms result from a general, systemic condition, not localized prostate infection, inflammation or dysfunction.2 A significant number of CP/CPPS patients report associated cardiovascular, neurological, psychiatric, and immunologic disease; many also meet diagnostic criteria for fibromyalgia, chronic fatigue syndrome, and irritable bowel syndrome.3
A variety of vulnerability factors have been implicated for the development of these various systemic conditions, including complex but subtle hormonal abnormalities related to the hypothalamic-pituitary-adrenal (HPA) axis.4 The goal of this study was to investigate and begin to characterize such adrenocortical hormone abnormalities as indicators of endocrine dysfunction in CP/CPPS patients and thereby improve our understanding of the condition and identify potential biomarkers for the disease.
This was a double-blind biomarker discovery study. Serum samples were obtained from CP/CPPS patients and age-matched healthy asymptomatic controls at two tertiary referral medical centers specializing in the diagnosis of CP/CPPS during 2006. Following subject recruitment and sample collection, all specimens were analyzed in a single batch at a single blinded research institution using isotope dilution liquid chromatography followed by atmospheric pressure photospray ionization and tandem mass spectrometry (LC-APPI-MS/MS). The institutional review boards at the respective institutions approved the study.
The primary diagnostic criterion for CPPS was pain or discomfort in the pelvic region for at least 3 months of the previous 6 months. Eligible men were required to have at least "moderate" symptoms, defined as a total NIH-CPSI5 score > 15. Eligibility criteria were identical to those of the NIH-CPCRN and are available elsewhere.6
Healthy asymptomatic men from the general population were matched to the cases by age and served as voluntary controls. In addition to an NIH-CPSI score of zero, the controls were required to have a prostate-specific antigen level < 2.5 ng/mL and normal digital rectal exam.
The study included one clinic visit during which subjects provided written informed consent, completed the NIH-CPSI questionnaire, and provided blood and urine samples. Because of the effect of the circadian rhythm on adrenocortical function,7,8 all samples were obtained at the time of peak hormone concentration between 7:00 and 8:00 AM after an overnight fast. To minimize the effect of inter-assay variability, CP/CPPS and control samples were assayed in the same run. Blood samples were centrifuged to separate serum and all samples were frozen at −70°C and stored until shipped and processed.
An API-5000 triple-quadrupole mass spectrometer (Applied Biosystems/MDS SCIEX, Foster City, CA/Concord, Ontario, Canada) equipped with an atmospheric pressure photoionization source, 3 Shimadzu LC-10ADvp pumps, a Shimadzu SIL-HT autosampler, and a Shimadzu DUG-14A degasser (Shimadzu Scientific Instruments, Columbia, MD) employing isotope dilution with deuterium-labeled internal standard for each analyte was used to measure hormone concentrations.9 200 μL of serum was deproteinized by adding 300 μL of acetonitrile containing internal standards. After centrifugation, 450 μL of supernatant was diluted with 900 μL of water, and then injected a 1,000 μL aliquot into the LC-APPI-MS/MS system. After 3 minutes of washing, the steroids were eluted from the column with a water/methanol gradient at a flow rate of 0.6 ml/min, and introduced the sample into the mass spectrometer. Quantitation by multiple reaction monitoring (MRM) analysis in positive ion mode was performed for 11 analytes, and in negative ion mode for aldosterone.
Accuracy of the method was evaluated by: (a) comparing our methods with previously validated MS/MS methods conducted at the Mayo Clinic and (b) performing recovery studies. The results of the comparison study yielded correlations ranging from r= 0.908 to 0.999. Within-day coefficients of variation (CVs) were less than 11.5% for all analytes tested and between-day CVs ranged from 3.5% to 12.2%.
The amount of analyte we recovered was close to the target value spiked into a serum pool. Mean recoveries of the analytes under study ranged from 90% to 110%. The lower limit of detection (LOD = a reading of 3 standard deviations (SD) above baseline noise) for each steroid ranged from 1.5 to 10 pg/ml. We have previously published details of the protocol.9
CP/CPPS patients and controls were compared using t-tests, chi-square tests, and Kruskal-Wallis analysis of variance with Scheffé contrasts. Jonckheere’s test was employed to analyze trends among the groups. Values were expressed as means ±SD, unless specified otherwise. For all steroids wherein subject values fell below the limit of detection of the instrument, these values were set at the limit of detection and the groups were compared using the Mann-Whitney nonparametric test. Correlational analyses were conducted using the Spearman Rank Method; p values<0.05 (two-tailed) were considered statistically significant. All tests were conducted using Version 9.1.3 of the SAS software package (SAS Institute Inc., 100 SAS Campus Drive, Cary, NC 27513).
Twenty seven CP/CPPS patients and 29 age-matched healthy controls were included in the analysis. Three CP/CPPS patients and 4 control subjects were excluded from the final analysis due to multiple values more than 2 SD greater than the mean value for the respective group. No differences in demographic parameters between CP/CPPS cases and controls were identified (Table 1). Patients had an average (±SD) of 2.4 ±1.0 co-morbid disorders in addition to CP/CPPS: nervous in 16 (anxiety, numbness, tingling), allergic in 9 (sinusitis, allergies, asthma), musculoskeletal in 7 (chronic fatigue syndrome, fibromyalgia, arthritis), gastrointestinal in 6 (irritable bowel syndrome), cardiovascular in 7 (hypertension, palpitations, sweating), and other urogenital in 11 (erectile dysfunction, balanitis, stones).
In the mineralocorticoid synthetic pathway (see Figure 1), concentrations of progesterone were found to be significantly higher in CP/CPPS than in controls, whereas corticosterone and aldosterone were significantly lower in patients than in controls (Table 2).
In the glucocorticoid synthesis pathway (see Figure 1), serum concentrations of 11-deoxycortisol (11-DOC) were significantly lower, and those of cortisol were not different in patients than in controls (Table 2).
In the sex steroid synthetic pathway (see Figure 1), serum concentrations of androstenedione (ASD) and testosterone were significantly higher in CP/CPPS than in controls, whereas those of estradiol were not different. ASD also can be synthesized from dehydroepiandrosterone (DHEA), which also can interconvert with dehydroepiandrosterone sulfate (DHEAS). Serum concentrations of DHEA and DHEAS were not different between CP/CPPS patients and controls (Table 2).
These results suggested reduced activity of CYP21A2 (P450c21), the enzyme that converts progesterone (via 11-deoxycorticosterone, which was not measured in the current study) to corticosterone, and 17-OHP to 11-DOC (Figure 1). To further investigate this possibility, we calculated precursor/product ratios and found them significantly elevated in CP/CPPS patients relative to controls (Table 2).
NIH-CPSI total and pain domain scores correlated positively (P<0.001), with 17-OHP and aldosterone and negatively (P<0.001) with cortisol concentrations (Table 3).
CP/CPPS patients had abnormally high levels of progesterone and androstenedione, and abnormally low levels of corticosterone, aldosterone, and 11-DOC compared to healthy volunteers. Furthermore, lower levels of cortisol and higher levels of aldosterone were associated with higher pain and CPSI scores. These clinical correlates provide tantalizing insights into the systemic nature of CP/CPPS.
In contrast to traditional assumptions about CP/CPPS—that it is a disease that originates in the prostate,10,11 that it is an infectious disease treatable with antibiotics,12–14 that it is a psychosomatic disorder with no biological basis2 —recent studies have identified a variety of co-morbid disorders in patients with CP/CPPS, suggesting it may be a systemic disorder.3 Our interest in characterizing adrenocortical hormone levels in CP/CPPS was prompted by findings of HPA disturbances in some of these co-morbid conditions.3 The present results confirm the presence of HPA abnormalities in CP/CPPS patients. In particular, the hormonal abnormalities are characterized by an increased concentration of proximal adrenocortical hormones and a corresponding decrease of distal hormones within their respective steroidal synthetic pathways (Figure 1). These findings suggest a defect in CYP21A2 (also known as 21-hydroxylase), the enzyme responsible for converting progesterone to deoxycorticosterone (and subsequently corticosterone and aldosterone), and 17-OHP to 11-DOC.
CYP21A2 defects traditionally have been described in patients with congenital adrenal hyperplasia (CAH). The hormonal defects in our CP/CPPS population suggest that some may have an inherited or acquired form of non-classical CAH due to CYP21A2 deficiency. Classical CAH presents with salt wasting or genital ambiguity in infants. In contrast, non-classical (also known as mild or late-onset) CAH is characterized by partial CYP21A2 deficiency and varying signs of hyperandrogenism—abnormalities that are generally thought to be asymptomatic in men.15,16 In our study, the CP/CPPS patients also had hormonal evidence of hyperandrogenism, elevated androstenedione and elevated testosterone levels compared to controls, a finding that further supports the presence of reduced CYP21A2 activity. These results suggest that follow-up endocrinological and genetic testing could determine the role of CYP21A2 abnormalities in men with CP/CPPS.17
Reduced morning cortisol concentrations have been described in several other pain syndromes, including fibromyalgia,18 some functional gastrointestinal disorders,19 and low back pain.20 However in our study, unstimulated morning cortisol concentrations were not different between CP/CPPS cases and controls. In women with interstitial cystitis, a condition similar to CP/CPPS, patients with lower morning cortisol had significantly more pain and urgency, and those with lower urinary free cortisol reported more overall symptomatology (P<0.05),21 although mean urinary or salivary cortisol did not differ between patients and controls.21 This negative correlation is similar to that found for serum cortisol in our patients (Table 3). The underlying biological basis for hypocortisolemia in such chronic pain conditions is incompletely understood, and may be a protective adaptation to chronic stress.22
Although direct measurements of cortisol concentrations were not different between patients and controls in our study, the excess concentrations of progesterone and androstenedione still suggest a relative hypocortisolemia (because reduced negative feedback by cortisol on the corticotrophs of the anterior pituitary gland causes increased adrenocorticotropin [ACTH] secretion that stimulates adrenal hormone steroidogenesis, leading to excess hormone concentrations proximal to the CYP21A2 defect). We could not obtain ACTH measurements in our patients or results from cosyntropin testing in our patients to further investigate this possibility, but if these patients do in fact have hypocortisolemia, it would further implicate a defect in CYP21A2 and an association with non-classical CAH. To test this hypothesis, future prospective cohort studies should measure both morning ACTH and diurnal cortisol concentrations and perform cosyntropin stimulation testing in patients with CP/CPPS.
Limitations of the current study include small sample size, cortisol concentration measurements taken only from morning samples, a cross-sectional design that prevents conclusions about causality, and the lack of measurement of some adrenal hormones due to the unavailability of quantitative MS assays for those substances. Compared to results obtained from immunoassays used routinely in clinical practice, our quantitative results appear to be lower than the traditional reference ranges. This can be explained by the lack of specificity in the traditional assays, which cross-react with molecules of similar structure and result in readings that are higher than those obtained by tandem mass spectrometry. Unlike immunoassays, isotope dilution tandem mass spectrometry is quite specific for the steroids tested and does not quantify other steroids of similar structure (Tables 5–9 in reference 21).23
The hormonal differences we identified in CP/CPPS patients may help us understand the etiology of CP/CPPS and/or serve as biomarkers for the disease. Our study suggested that some men with CP/CPPS may have a defect in CYP21A2.. Future studies should explore the possibility that CP/CPPS patients meet the diagnostic criteria for nonclassical CAH due to 21-hydroxylase deficiency by performing a CRF and cosyntropin stimulation test with 17-OHP measurements, and, if confirmatory, DNA testing for CYP21A2 mutations. Follow-up studies should further assess whether the hormonal findings described in this paper improve or worsen in parallel with symptom severity. Finally, if our work is confirmed, consideration may be given to performing a randomized, controlled trial of low-dose dexamethasone (the treatment of choice for patients with CAH) in patients with CP/CPPS to assess whether normalization of these hormonal abnormalities improves symptoms.
The authors gratefully acknowledge Leroy M. Nyberg for helpful discussions and Michael D. Smith for editorial assistance in the preparation of the manuscript.
Sources of support: Grants R01 DK065990 and R21 DK070672 to Jordan Dimitrakov
Name and address of the author to whom requests for reprints should be addressed:
Jordan Dimitrakov, M.D., Ph.D.,Harvard Medical School, Harvard Urological Diseases Research Center, Enders Research Building, Room 1061, 300 Longwood Ave, Boston, MA 02115, Phone: (617) 919–2521, Fax: (617) 249–2035, E-mail: Jordan.Dimitrakov/at/childrens.harvard.edu
Conflict of Interest Disclosures:
The following individuals have NO relevant financial interests in the manuscript: Jordan D. Dimitrakov, Hylton V. Joffe, Steven J. Soldin, Roger Bolus, CA Tony Buffington
The following individual has disclosed financial interests relevant to the manuscript as follows:
J Curtis Nickel: Consultant: Bayer, Boehringer Ingelheim, Farr Laboratories, Glaxo-Smith-Kline, Merck Frosst Canada, Ortho-McNeil, Boston Scientific
Investigator: Bayer, Glaxo-Smith-Kline, Merck Frosst Canada, Ortho-McNeil, Sanofi-Aventis, Farr Laboratories, Stellar Laboratories
Honoraria: Bayer, Boehringer Ingelheim, Ortho-McNeil
Grants received: Farr Laboratories, Stellar Laboratories