| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Institute of Reproductive Medicine of the University, D-48129 Münster, Germany
Address all correspondence and requests for reprints to: Prof. E. Nieschlag, FRCP, Institute of Reproductive Medicine of the University, Domagkstr. 11, D-48129 Münster, Germany. E-mail: nieschl{at}uni-muenster.de.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
T effects are mediated via the androgen receptor (AR) and the significant amount of variation in T responsiveness that is observed in hypogonadal men receiving substitution therapy could hypothetically be facilitated by a pharmacogenetic mechanism involving the AR. A respective candidate is the CAG repeat [(CAG)n] polymorphism of the AR gene on the X chromosome, which normally ranges between 9 and 37 triplets (8, 9, 10) and encodes for a variable number of glutamine residues in the aminoterminal domain of the receptor. The number of (CAG)n is inversely associated with the transcriptional activity of testosterone target genes (11, 12). A length beyond 37 repeats leads to Kennedy disease (spinobulbar muscular atrophy), which is caused by irregular processing of the AR protein (13). The disease is also characterized by marked hypoandrogenic traits (e.g.14).
Frequently, but not uniformly, (CAG)n length within the normal range has clinical implications for androgen effects in eugonadal men, which have been demonstrated on a cross-sectional basis. This refers to bone density (15), endothelial function and lipid levels (16), body composition in relation to leptin and insulin levels (17), and spermatogenesis (18, 19) as well as depression (20), alopecia (21), and behavioral aspects (22). Effects on prostate tissue as the most prominent androgen-sensitive organ have also been reported: studies stratifying for lifestyle factors and multidimensional matching of controls in a sufficient number of subjects show an increased risk for prostate cancer (9, 23, 24, 25, 26) as well as benign prostate hyperplasia in patients with lower numbers of (CAG)n, hence increased androgenicity (27), e.g. the odds ratio (OR) for men with (CAG)n = 19 vs. those with (CAG)n = 26 to be diagnosed with metastatic prostate cancer was 2.44 (95% confidence interval, 1.32–4.55) (23) and obstructive benign prostate hyperplasia was 3.62 (95% confidence interval, 1.51–8.76) (27).
Extending these findings to pharmacogenetic considerations, an influence of the (CAG)n polymorphism on efficacy of hormonal male contraception was recently reported in 85 healthy men, showing better inhibition of spermatogenesis in volunteers with longer triplet residues in the subgroup of those subjects with not fully suppressed gonadotropin secretion (28). An earlier study of 75 men in contraceptive trials without such a subgroup classification could not report such results (29).
The above-mentioned modulations of androgen activity by the (CAG)n polymorphism were observed in eugonadal men. It can be speculated that the sudden exposure to androgens on initiation of substitution therapy in hypogonadal men, which is often accompanied by impressive changes in androgen-sensitive tissues, is modulated to an even larger degree by the AR (CAG)n polymorphism. We addressed this question in the first pharmacogenetic study on androgen substitution in hypogonadal men by using a longitudinal design.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
One hundred thirty-one untreated hypogonadal men aged 19–69 yr (Table 1
) presenting for examination and possible treatment of hypogonadism-related symptoms lasting for at least 2 yr were included in this retrospective analysis. Such symptoms were fatigue, loss of libido, depression, change in body composition/weight, decreased physical performance, decreased aggressive behavior, disability to cope, decreased performance at work, and lack of androgenization. At least one of these symptoms had to be accompanied by low total T levels (<12 nmol/liter). Previous androgen treatment as well as diabetes mellitus, arterial hypertension, dyslipoproteinemia, medication of any kind, or drug abuse led to exclusion of patients. Hypogonadism was defined by symptoms of androgen deficiency and morning total serum T levels of less than 12 nmol/liter (see also Hormone measurements). The threshold of 12 nmol/liter is in agreement with the World Health Organization consensus guidelines (30) and data obtained in a large trial involving older men (31). Hence, we did not perform an age adjustment of the normal range for total T of 12–35 nmol/liter. Patients gave written informed consent to the investigations.
|
We also report results of a patient with (CAG)n = 38. Because this length is out of the normal range (9–37, see above) and was distinct from the distribution of the other patients (see Table 1
), his data were not included in the analysis but are reported separately. The 22-yr-old patient was diagnosed with idiopathic hypogonadotropic hypogonadism (LH and FSH levels below normal with 0.8 and 1.0 IU/liter, respectively, normal response to GnRH challenge, total T levels 1.2 nmol/liter). Because of suspicion of Kennedy disease, the patient was referred to the neurological department. Although the diagnosis of spinobulbar muscular atrophy could not be confirmed by electromyography, the patient showed distinct features of the disease such as muscle cramps after physical exertion as well as gynecomastia and tongue fasciculations after prolonged speech. According to current knowledge, onset of the disease at a later time point in life is probable for this patient (33). The patient was treated with two weekly injections of T enanthate and demonstrated a weak improvement of androgen-dependent parameters.
Treatment
Reliable continuity of treatment was documented by regular visits including assessment of T levels indicating effective therapy. Treatment modalities included T enanthate 250 mg 2–3 wk im, scrotal T patches, oral T undecanoate (at least two to four capsules per day with meals); patients with secondary hypogonadism desiring paternity received combined treatment of sc injections of human menopausal gonadotropin (three times per week) and human chorionic gonadotropin (hCG) (two times per week). Table 2 details distribution of treatment modalities among patients.
|
All blood sampling was performed between 0800 and 1200 h. To test for effective T substitution, samples from patients being treated with im-injected T enanthate were obtained at time points indicating individual average levels, preferentially in the second week after injection (34). Patients treated with the transdermal scrotal T system or oral T undecanoate were sampled 2–5 h after administration of the patch or capsules. Serum from those men treated with hCG was obtained 2–3 d after the last sc injection. Serum T levels were measured with a commercial ELISA kit (DRG Instruments GmbH, Marburg, Germany), prostate-specific antigen (PSA) with highly specific time-resolved fluoroimmunoassay (Autodelfia, Freiburg, Germany), with a normal range less than 4 µg/liter. Mean intra- and interassay coefficients of variation for T were less than 5% and 10% and PSA less than 2% and 5%, respectively. Sampling was performed before prostate palpation and transrectal ultrasonography.
Determination of (CAG)n
Analysis of the AR gene microsatellite residues was performed as previously published (16). In the Klinefelter patients, two X chromosomes were present (patients with more than two X chromosomes were not included); hence, two alleles of the AR gene were found. In case of different (CAG)n alleles (22 of 40 patients), the mean was calculated and used for further analysis because, without biopsy, it could not be determined whether inactivation of X chromosomes was skewed in the prostate tissue.
Prostate measurement
A high-resolution transrectal ultrasonography probe (7.5 MHz for sagittal and transversal scans) with an ultrasound scanner type 2002 ADI (B-K Medical, Gentofte, Denmark) was used. Contact gel was used inside and outside a sterile latex sheath, and patients were examined while lying on the left side with knees drawn upward. Anterior-posterior, cephalocaudal, and transverse rectangular diameters (d1–3) of the prostate were determined, and volume was calculated by the procedure provided by the machine (volume = 
/6 x d1 x d2 x d3). Accuracy of the method was determined by two investigators in 19 healthy men with reassessment of prostate size after 8 wk. Prostate volume in these men ranged from 13.8 to 31 ml with a mean of 19.8 ± 4.6 ml; interobserver coefficient of variation was 4.2%.
Statistics
All variables were checked for normal distribution by the Kolmogorov-Smirnov one-sample test for goodness of fit. In case of PSA levels, analysis was performed on logarithmically transformed data. All data are presented as mean ± SD.
Data were analyzed using a cross-sectional and longitudinal approach. The cross-sectional approach demonstrates differences among subjects, and parameters modulating treatment effects in individuals (inner-patient contrasts) can be detected only by longitudinal methods. Thus, predictors of baseline and final prostate volume were subjected to analysis of covariance (ANCOVA) using parameters mentioned in Table 3 for the cross-sectional approach, and, similarly, ANCOVA for repeated measurements was calculated for the longitudinal analysis (Table 4). Prostate growth per year was slightly, but nonsignificantly, dependent on the duration of treatment with higher values in shorter treatment periods. This suggests that prostate growth could be more pronounced shortly after initiation of therapy in comparison with longer-term treatment. Accordingly, growth per year was corrected for this slight, nonsignificant linear association with duration of treatment. Growth per year was calculated on the basis of treatment days.
|
|
or >40 yr) as covariate. Significant differences in distribution were calculated by Fisher’s exact test (Table 5). The overall significance of ORs was calculated using asymptotic significance according to Mantel-Haenszel; homogeneity of subgroups was tested according to Breslow-Day.
|
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
). In addition, no such differences were observable among the groups receiving different treatment modalities. Time between start of therapy and final investigation was homogenous among the respective subgroups as well as the significant increase of T levels from the clearly hypogonadal into the midnormal range (Table 2).
Cross-sectional analysis of initial prostate volume revealed no significant relation to the length of AR (CAG)n (Pearson’s r = -0.12, P = 0.18, see Fig. 1). Instead, initial prostate size was positively dependent on age and baseline T levels (ANCOVA; see Table 3). The volume of the prostate gland increased significantly during treatment, an observation that was homogenous among subgroups concerning diagnosis or treatment modality (Table 2).
|
). Further analysis by ANCOVA included influences of age and initial prostate volume as well as duration of treatment and T levels achieved by treatment because these parameters also had an effect on outcome (Table 3). Differences among patients regarding speed of prostate growth (per year) on initiation of T substitution were also highly dependent on the AR polymorphism: the shorter the AR microsatellite residues, the more pronounced prostate growth was (Pearson’s r = -0.32, P < 0.001, Fig. 2). Advanced age contributed to prostate growth as well as higher T levels (Table 3). A larger initial prostate size resulted in slower growth, although the total size under therapy was larger in these patients (Table 3). A trend to lower speed of organ growth was observed in patients being treated with hCG or oral testosterone undecanoate vs. those treated with a transdermal patch or im testosterone enanthate; nevertheless, final prostate size did not differ significantly between these groups (Table 3).
|
). This is in good agreement with the cross-sectional results (Table 3).
For estimation of clinically useful measures, the distribution of (CAG)n was categorized into tertiles: one third of the patients had residues shorter than 20 (group 1, n = 40), one third had a repeat length of 20 or more and less than 23 (group 2, n = 46), and the last group had (CAG)n of 23 or more (group 3, n = 45). The groups were then analyzed for the presence of a prostate size of 30 ml or more under treatment. No difference was seen between groups 2 and 3; hence, these were combined to increase power for the calculation of ORs between group 1 and the others. It could be shown that the chance to develop a large prostate under T substitution was, on average, 8 times higher for the group with (CAG)n less than 20. This effect was age dependent and more pronounced in men older than 40 yr (Table 5).
PSA levels increased significantly during the study (Table 2) and were significantly dependent only on prostate size (for initial and final PSA both P = 0.003) but, under treatment, showed a trend to be elevated in the presence of higher T levels (for final PSA: P = 0.06) (all results according to ANCOVA). PSA levels remained below the limit of normal (<4 µg/liter) in all cases.
Figures 1 and 2 also display the results of the patient with (CAG)n = 38 (see Patients) who was not included in the statistical analysis: He was treated with testosterone enanthate and received im injections every 2 wk. Small prostate size and poor growth are in agreement with the results mentioned above.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
It has been demonstrated previously that T administration to hypogonadal men increases prostate volume, but marked differences among patients were noted (1, 2, 3, 4, 5, 6, 7). The average growth of the prostate under T substitution (see Table 2) reported here is in good agreement with previous results. Data on the influence of age and T levels on prostate size in hypogonadal men (1, 7) are confirmed by the cross-sectional analysis of baseline values.
The possible relation between benign hyperplasia of the prostate (BPH) and the (CAG)n polymorphism of the AR gene has previously been addressed by cross-sectional studies in eugonadal men. The overgrowth of the tissue of the transition zone and periurethral area of the prostate is the pathophysiological correlate of BPH and can be histologically described as epithelial and fibromuscular hyperplasia (35). Androgens modulate this process by binding to the AR under the conditions the prostate cell provides in terms of specific coactivators (36). Binding of these coactivators is modulated by the length of (CAG)n, which is the underlying mechanism suspected to influence the repeat polymorphism concerning transcriptional activity of target genes (37). The putative influence of the (CAG)n polymorphism on prostate size in eugonadal men has been investigated by several groups in cross-sectional approaches. The two largest studies compared matched healthy controls (n = 1041 and n = 499) and BPH patients (n = 310 and n = 449), finding the OR for an enlarged prostate gland or BPH surgery to be 1.92 (P = 0.0002) in men with (CAG)n length of 19 or less, compared with those with 25 repeats or more. For a six-repeat decrease in (CAG)n length, the OR for moderate or severe urinary obstructive symptoms induced by an enlarged prostate gland was calculated as 3.62 (P = 0.004) (27). In agreement, adenoma size was found to be inversely associated with the number of CAG residues in 176 patients vs. 41 controls (38). This was confirmed in 68 BPH patients and 197 controls (39) but not in a European study involving 98 patients and 61 control patients; however, controls consisted of patients with bladder cancer, not healthy men (40). In summary, the development of BPH in eugonadal men is likely to be influenced by the AR polymorphism.
It is, therefore, in good agreement with previous cross-sectional models in eugonadal men that the effects of an elevation of T levels in hypogonadal men are modulated by the (CAG)n polymorphism of the AR. It has to be noted that androgen levels within the hypogonadal range were associated with prostate size, but no effect of the genetically determined AR activity on prostate size was seen; it can be speculated that within the hypogonadal range, the association of androgen target tissues to changes in T levels is stronger than within the eugonadal range, overriding the effects of the polymorphism. We have previously shown that the association of T with bone tissue and vascular functions follows a nonlinear relationship with about 4-fold stronger effects for increments of T levels below the lower limit of normal in comparison with fluctuations of T levels within the normal range (41, 42). This suggests a plateau effect of androgens on target tissues and provides reason to assume that the AR polymorphism will show only significant effects in the environment of sufficient androgen concentrations, hence full AR activation. To date, there are no other data available concerning hypogonadal men and the AR polymorphism.
Accordingly, the modulatory activity of the (CAG)n polymorphism became visible on T treatment that provided an amount of androgens sufficient for regular receptor activation. Correspondingly, the effects of various T levels were still visible but weaker than in the untreated men. The relationship between (CAG)n length and prostate size and growth under T substitution is obviously linear over the normal range (see Figs. 1 and 2). Nevertheless, when 30-ml prostate size was chosen as a specific threshold, especially patients with (CAG)n less than 20, representing one third of the study population, were likely to be diagnosed with a large prostate. The OR for these men in comparison with the patients with longer triplet chains to develop such an organ size were, on average, 8 times higher (Table 5). This observation was largely age dependent, with a higher risk in men older than 40 yr (Table 5). This may suggest that the prostate especially in older men with residues shorter than 20 should be subject to closer surveillance. Considering the effects of T levels, a lower substitution dose could also be considered. These aspects should be addressed by larger prospective studies.
As mentioned in the introduction, a relation of the AR (CAG)n polymorphism with the development of prostate cancer is likely because these cells are, at least initially, susceptible to androgenic influence. This study did not address this specific topic, and the potential influence of the AR polymorphism on the risk of hypogonadal men under androgen substitution to develop prostate cancer cannot be estimated by our data. In patients with prostate cancer, PSA levels are most often elevated. These elevated PSA levels usually do not parallel the changes in prostate volume but exceed them pronouncedly. The overall increase in prostate volume in this study was accompanied by a moderate increase of PSA levels, which remained within the normal range.
In conclusion, the influence of T substitution in hypogonadal men on the prostate gland is markedly influenced by the (CAG)n polymorphism of the AR gene, a process that is pronounced in older men and seems to put patients with a repeat chain of 20 or less triplets at an increased risk to develop an enlarged organ. These pharmacogenetic findings may provide the basis for individualized T substitution therapy by adjusting the dose to the AR polymorphism.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: ANCOVA, Analysis of covariance; AR, androgen receptor; BPH, benign prostate hyperplasia; (CAG)n, CAG repeat; hCG, human chorionic gonadotropin; OR, odds ratio; PSA, prostate-specific antigen; T, testosterone.
Received December 11, 2002.
Accepted February 10, 2003.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Zitzmann and E. Nieschlag Androgen Receptor Gene CAG Repeat Length and Body Mass Index Modulate the Safety of Long-Term Intramuscular Testosterone Undecanoate Therapy in Hypogonadal Men J. Clin. Endocrinol. Metab., October 1, 2007; 92(10): 3844 - 3853. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Crabbe, V. Bogaert, D. De Bacquer, S. Goemaere, H. Zmierczak, and J. M. Kaufman Part of the Interindividual Variation in Serum Testosterone Levels in Healthy Men Reflects Differences in Androgen Sensitivity and Feedback Set Point: Contribution of the Androgen Receptor Polyglutamine Tract Polymorphism J. Clin. Endocrinol. Metab., September 1, 2007; 92(9): 3604 - 3610. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. E. Borst, C. F. Conover, C. S. Carter, C. M. Gregory, E. Marzetti, C. Leeuwenburgh, K. Vandenborne, and T. J. Wronski Anabolic effects of testosterone are preserved during inhibition of 5{alpha}-reductase Am J Physiol Endocrinol Metab, August 1, 2007; 293(2): E507 - E514. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Zitzmann, S. Faber, and E. Nieschlag Association of Specific Symptoms and Metabolic Risks with Serum Testosterone in Older Men J. Clin. Endocrinol. Metab., November 1, 2006; 91(11): 4335 - 4343. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. K Chandolia, C. M. Luetjens, J. Wistuba, C.-H. Yeung, E. Nieschlag, and M. Simoni Changes in endocrine profile and reproductive organs during puberty in the male marmoset monkey (Callithrix jacchus). Reproduction, August 1, 2006; 132(2): 355 - 363. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Zinn, P. Ramos, F. F. Elder, K. Kowal, C. Samango-Sprouse, and J. L. Ross Androgen Receptor CAGn Repeat Length Influences Phenotype of 47,XXY (Klinefelter) Syndrome J. Clin. Endocrinol. Metab., September 1, 2005; 90(9): 5041 - 5046. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Zitzmann, M. Depenbusch, J. Gromoll, and E. Nieschlag X-Chromosome Inactivation Patterns and Androgen Receptor Functionality Influence Phenotype and Social Characteristics as Well as Pharmacogenetics of Testosterone Therapy in Klinefelter Patients J. Clin. Endocrinol. Metab., December 1, 2004; 89(12): 6208 - 6217. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Nieschlag, H.M. Behre, P. Bouchard, J.J. Corrales, T.H. Jones, G.K. Stalla, S.M. Webb, and F.C.W. Wu Testosterone replacement therapy: current trends and future directions Hum. Reprod. Update, September 1, 2004; 10(5): 409 - 419. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Heinlein and C. Chang Androgen Receptor in Prostate Cancer Endocr. Rev., April 1, 2004; 25(2): 276 - 308. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. H. Field Fractures, Osteoporosis, and the Endocrinologist Arch Intern Med, December 8, 2003; 163(22): 2796 - 2796. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) and under androgen substitution (•) with corresponding regression lines (line a: before treatment, slope not significantly different from zero; correlation coefficient, r = -0.12, P = 0.18; line b: under treatment, slope significantly different from zero, correlation coefficient: r = -0.45, P < 0.001; slopes of lines a and b are significantly different, P = 0.002 by F test). Data from the patient with 38 triplets, which were not included in the statistical evaluation, are shown in parentheses.