Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Gynecomastia wikipedia , lookup
Testosterone wikipedia , lookup
Hormone replacement therapy (menopause) wikipedia , lookup
Hypothalamus wikipedia , lookup
Sexually dimorphic nucleus wikipedia , lookup
Hormonal breast enhancement wikipedia , lookup
Growth hormone therapy wikipedia , lookup
Hormone replacement therapy (female-to-male) wikipedia , lookup
Hormone replacement therapy (male-to-female) wikipedia , lookup
Hypopituitarism wikipedia , lookup
Human Reproduction, Vol.30, No.11 pp. 2587– 2596, 2015 Advanced Access publication on September 6, 2015 doi:10.1093/humrep/dev213 ORIGINAL ARTICLE Puberty, aging and HRT Sexual dichotomy of gonadal function in Prader – Willi syndrome from early infancy through the fourth decade H.J. Hirsch 1,2,*,†, T. Eldar-Geva 1,3,†, F. Bennaroch 4, Y. Pollak 1,5, and V. Gross-Tsur 1,2,6,† 1 Neuropediatric Unit, Shaare Zedek Medical Center, Jerusalem 91031, Israel 2Department of Pediatrics, Shaare Zedek Medical Center, Jerusalem 91031, Israel 3Reproductive Endocrinology and Genetics Unit, Department of Obstetrics and Gynecology, Shaare Zedek Medical Center and The Hebrew University School of Medicine, Jerusalem 91031, Israel 4The Herman Dana Child and Adolescent Psychiatry Unit, Hadassah-Hebrew University Medical Center, Jerusalem 91031, Israel 5The School of Education, The Hebrew University, Jerusalem 91031, Israel 6The Hebrew University School of Medicine, Jerusalem 91031, Israel *Correspondence address. E-mail: [email protected] Submitted on April 27, 2015; resubmitted on July 19, 2015; accepted on August 10, 2015 study question: At what age does the type of hypogonadism, namely hypothalamic or primary gonadal defect, become established in men and women with Prader–Willi syndrome (PWS)? summary answer: The type of hypogonadism becomes established only in late adolescence and early adulthood. what is known already: The etiology of hypogonadism in PWS is heterogeneous and the clinical expression is variable. Primary testicular failure is common in PWS men, while combinations of ovarian dysfunction and gonadotrophin deficiency are seen in women. study design, size, duration: This is a prospective study of a cohort of 106 PWS patients followed for a mean duration of 4.5 years. Serial blood samples were obtained and assayed for gonadotrophins, inhibin B, anti-Mullerian hormone (AMH), dehydroepiandrosterone sulfate (DHEAS), testosterone (males), and estradiol (females). Results were compared with normal reference values obtained from the literature. For the purpose of this study, we defined the following age groups: infants ,1 year; children 1–10 years; adolescents 11– 20 years and adults .20 years. participants/materials, setting, methods: Study participants were 49 males (aged 2 months to 36 years) and 57 females (aged 1 month to 37 years) with genetically confirmed diagnoses of PWS (deletions 60, uniparental disomy 54, imprinting center defect 2) followed in the Israel national multidisciplinary PWS clinic. main results and the role of chance: Serum LH levels were in the normal range (1.0 – 6.0 mIU/ml) for 7/10 adult men, and high in 3, while FSH (normal range 1.0 – 6.1 mIU/ml) was elevated (34.4 + 11.5 mIU/ml) in 6 and normal (3.5 + 1.6 mIU/ml) in 4 men. Testosterone was low (5.7 + 3.4 nmol/l) compared with the normal range of 12.0 – 34.5 nmol/l in the reference population in all men .20 years. AMH showed a normal decrease with age, despite low testosterone levels. Inhibin B was normal (241 + 105 pg/ml) in infant boys, but low or undetectable in most adult men. Hormonal profiles were more heterogeneous in women than in men. Estradiol was consistently detectable in only 7/13 adult women. Inhibin B was low or undetectable in all PWS females although occasional samples showed levels within the normal range of 15 – 95 pg/ml. Vaginal bleeding was reported to occur for the first time in eight women at a median age of 20 years (13 – 34 years), but only one had regular monthly menses. The type of hypogonadism (primary or secondary) in PWS can be determined only after age 20 years. limitations, reasons for caution: The study cohort was heterogeneous, showing variability in BMI, cognitive disability and medical treatment. wider implications of the findings: Demonstration of the natural history of reproductive hormone development in PWS suggests that androgen replacement may be indicated for most PWS boys in mid-adolescence. Recommendations for hormone replacement in PWS women need to be individually tailored, serial measurements of inhibin B should be performed, and contraception should be considered in those women who may have the potential for fertility. † H.J.H., T.E.-G. and V.G.-T., contributed equally to this study. & The Author 2015. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: [email protected] 2588 Hirsch et al. study funding/competing interest(s): This study was supported by grants from Pfizer Pharmaceuticals and from the Foundation for Prader – Willi Research. T. Eldar-Geva and V. Gross-Tsur received grant/research support funding from the Foundation for Prader – Willi Research and from Pfizer Pharmaceuticals. trial registration number: Not applicable. Key words: Prader– Willi syndrome / delayed puberty / gonadotrophins / primary testicular failure / primary ovarian insufficiency / inhibin B / anti-Mullerian hormone / hypogonadotrophic hypogonadism / hypergonadotrophic hypogonadism Introduction Hypogonadism is a major and prominent feature of Prader–Willi syndrome (PWS), a complex neurogenetic disorder characterized by hypotonia, onset of extreme weight gain and obesity in early childhood, neuro-cognitive defects and behavioral disorders (Burman et al., 2001; Crino et al., 2003; Goldstone et al., 2008; Cassidy and Driscoll, 2009; Diene et al., 2010). Hypogonadism significantly contributes to poor body image, muscle weakness, abnormal body composition, decreased bone density and infertility in this syndrome. The etiology and expression of hypogonadism varies among affected individuals. Hypogonadism in PWS results from variable combinations of primary gonadal failure along with hypothalamic dysfunction (Eiholzer et al., 2006; Hirsch et al., 2009, 2013; Eldar-Geva et al., 2010a; Gross-Tsur et al., 2012; Siemensma et al., 2012a,b). Longitudinal studies of reproductive hormones in PWS have, to date, focused on children and adolescents (Siemensma et al., 2012a,b), with little or no data for individuals older than age 20 years. Since puberty and age at menarche may be markedly delayed in PWS, information regarding evolution of hormone patterns from adolescence into adulthood is needed to understand the timing and tempo of hypothalamic –pituitary –gonadal (HPG) axis development in this syndrome. Longitudinal tracking of reproductive hormones in individual patients may provide important data for understanding the natural history of the HPG axis in PWS and lead to further refinement of treatment guidelines, particularly with regard to determining when hormone replacement should be initiated. We performed a prospective study of reproductive hormones in males and females with PWS from infancy through the fourth decade. The aims of our study were to investigate when one can determine the type of hypogonadism (hypothalamic or primary gonadal defect) in PWS individuals, to observe if the type of hypogonadism remains consistent for each individual, and assess the relative contributions of hypothalamic versus gonadal dysfunction in a cohort of PWS patients across a wide age range. Methods Patients Study participants included 106 (49 males, 57 females) individuals (aged 1 month to 37 years) with PWS who were followed in our national multidisciplinary PWS clinic. Initial blood samples were obtained from 7 male and 9 female infants, 19 male and 30 female children, 13 male and 9 female adolescents, and 10 male and 9 female adults. Two or more sequential blood samples were obtained from 10 males and 13 females after age 20 years. This cohort includes nearly all known PWS individuals in Israel. Data from the first sample drawn from these patients were described in previous reports from our group, but sequential data have not yet been published (Hirsch et al., 2009, 2013, 2014; Eldar-Geva et al., 2010a; Gross-Tsur et al., 2012). Genetic diagnoses were deletion (DEL) in 60, uniparental disomy (UPD) in 44, and an imprinting center defect (IC) in 2. Forty-eight children aged 6 months to 17 years were treated with growth hormone (GH). None of the adult patients received GH prior to or during the study. During the study period, 28 patients received risperidone and/or selective serotonin-reuptake inhibitors (SSRI). Nine were taking methylphenidate and three received anticonvulsant medication (topiramate or valporate). Of four individuals with diabetes mellitus, three were on insulin treatment and one was taking metformin. Three others with hypothyroidism were wellcontrolled on l-thyroxine. None of the study participants were treated with androgens or sex steroid replacement prior to or during the study period. The study was approved by the institutional review board of Shaare Zedek Medical Center. Signed informed consent was obtained from parents, guardians or adult participants. Between 2007 and 2013, blood samples for hormone determinations were obtained on 2 – 6 occasions from each patient. Physical examination at each visit was performed by a pediatric endocrinologist (H.J.H.) or gynecologist (T.E.-G.) and included height measured with a wall-mounted stadiometer, weight, and assessment of pubertal stage according to the Tanner classification (Tanner, 1962). Data from the CDC-2000 North American growth standards were used for the normal age-related ranges of BMI (Kuczmarski et al., 2000). Hormone measurements Blood samples were assayed for serum levels of LH, FSH, total testosterone, estradiol (E2), prolactin and thyroid-stimulation hormone (TSH) using the DxI 800 immunoassay system (Beckman Coulter Instruments, Inc., USA). Assay sensitivities were 0.1 IU/l, 0.1 IU/l, 0.35 nmol/l, 37 pmol/l, 0.5 ng/ ml, and 0.03 U/ml for LH, FSH, testosterone, E2, prolactin and TSH, respectively. For testosterone, the intra-assay coefficients of variation were ,20% for a level of 1.74 nmol/l and ,10% for levels of 6.94– 34.7 nmol/l. For E2, the intra-assay coefficients of variation (CV) were 6.3 – 15% for levels ≥147 pmol/l and 20% for levels ,147 pmol/l. Serum inhibin B and antiMüllerian hormone (AMH) concentrations were measured using highly sensitive two-site enzyme-linked immunosorbent assays (ELISAs) (DSL, Webster, TX, USA) and the assay sensitivities were 10 and 0.017 ng/ml, respectively. Inter- and intra-assay CVs were 15 and 7% for inhibin B and 8.7 and 5.3% for AMH, respectively. Beginning in 2011, inhibin B was assayed using the GEN II ELISA kit (Beckman-Coulter). Results of inhibin B obtained using the older assay were adjusted for comparison with values obtained in the GEN II assay by using the formula ‘inhibin B value in old assay × 1.57 + 11.29 ¼ value for new assay’ as recommended by Kalra et al. (2010). Sex hormone-binding globulin (SHBG) concentrations were measured using immunochemiluminescence and IMMULITE (Siemens, Diagnostic Product Corporation, Los Angeles, CA, USA). Assay sensitivity for SHBG was 0.02 nmol/l and inter- and intra-assay CVs were ,10%. For each hormone, normal values for age were obtained from the literature (Esoterix Laboratory 2589 Reproductive hormones in Prader – Willi syndrome We compared our findings with normal reference data from published reports as described in the methods section. A two-tailed T-test was used to compare hormonal levels in women with PWS who experienced spontaneous vaginal bleeding with those with no history of vaginal bleeding (IBM Statistical Package for the Social Sciences for Windows, Version 19.0, IBM Corp Armonk, NY, USA). A value of P , 0.05 was considered significant. Tanner stage 3 testes size was achieved in seven men, only three reached Tanner stage 4, however most remained at Tanner stage 1. Penile length (Fig. 2A) was within the normal reference range in infancy and early childhood (Sun et al., 2002). Although some increase in penis size did occur during adolescence, penile lengths were below the normal limit in all adult men except for two whose phallic length was minimally above the lower limit of normal. Two PWS boys showed onset and progression of pubic hair (Fig. 2B) prior to the normal mean age of 11.8 years (Sun et al., 2002). Progression to Tanner stages 4–5 was often achieved only in late adolescence or early adulthood. Results Pituitary hormones services; Rey et al., 1993; Lee et al., 1996; Andersson et al., 1997; Sehested et al., 2000; Jensen et al., 2004; Sørensen et al., 2007; Aksglaede et al., 2010). Results of hormone data are presented as mean + SD. Statistics PWS males Clinical features: BMI and sexual development During childhood and adolescence, BMI was markedly elevated in most PWS boys (Fig. 1). BMI increased in younger boys, but decreased in older adolescents. Adult men had BMI values within the normal range, most likely due to strict diet control and exercise routines in the residential hostel. Undescended testes (bilateral in 34 and unilateral in 12) were noted at birth in 46 of 49 PWS males (Table I). The mean age when orchiopexy was performed in 37 boys was 2.7 years. Three boys had spontaneous testicular descent. Testes remained small in almost all PWS men. Figure 3A –C shows gonadotrophin and prolactin levels versus age in PWS males. LH and FSH decreased from early infancy to the expected low levels seen in prepubertal boys. In most PWS boys, FSH levels began to increase by 8–10 years of age and LH levels began to rise at age 12 –13 years. During adolescence, LH levels were in the normal range, but FSH was increased in some adolescent boys. In the 10 adult Table I Clinical description of study participants with Prader–Willi syndrome (PWS). PWS males PWS females ........................................................................................ Figure 1 BMI in Prader – Willi syndrome. BMI in 49 males (A) and 57 females (B) with Prader – Willi syndrome (PWS) according to age. The solid gray lines indicate the upper (+2 SD) and lower (22 SD) limits of normal, according to growth data from the Center for Disease Control North American growth database (CDC-2000) (Kuczmarski et al., 2000). Number of participants 49 57 Genetic diagnosis (DEL/UPD/IC) 27/21/1 33/23/1 Age range 2.8 months to 36 years 1 month to 37 years Age (years) at first sample (median, range) 7.8 (0.2– 32.4) 6.6 (0.1–31.8) Duration of follow-up in years (median and range) 4.8 (1.0– 6.2) 4.4 (0.8–5.9) Number of blood samples taken 149 190 Samples per patient (mean, median, range) 3.2, 3 (2– 5) 3.0, 3.4 (2–6) Cryptorchidism at birth (%): 46/49 (93.8%) Bilateral, n (%) 34/46 (73.9%) Unilateral, n (%) 12/46 (26.1%) Scrotal testes, n (%) 3/49 (6.1%) *Number of boys who underwent orchiopexy (%) 37/46 (80.4%) Number of boys with spontaneous descent of testes 3/46 (6.5%) Age (years) at orchiopexy (mean, median, range) 2.7, 2.0 (0.3– 7.0) No. of women with spontaneous vaginal bleeding 8 Age (years) at first reported vaginal bleeding (median, range) 20 (13.5– 34) DEL, deletion; UPD, uniparental disomy; IC, imprinting center defect. *Six boys with cryptorchidism had not yet undergone orchiopexy at the time of this study, primarily because of their young age or because of other medical problems, e.g. sleep apnea, which led to a delay in the surgery. 2590 Hirsch et al. Figure 2 Secondary sexual characteristics of the patients with PWS. Development of secondary sexual characteristics in 49 PWS males and 57 PWS females, including penile length (A), pubic hair according to Tanner stage in males (B), Tanner stage for breast development in PWS females (C), and pubic hair according to Tanner stage for PWS females (D). For penile lengths, the solid gray lines indicate the upper (+2 SD) and lower (22 SD) limits of normal (Schonfeld and Beebe, 1942). The average ages for achieving Tanner stage 2 for pubic hair (B and D) and Tanner stage 2 for breast development (C) are shown by the blank arrows. The average ages for achieving Tanner stage 5 for pubic hair (B and D) and Tanner stage 5 for breast development (C) are shown by the solid arrows. Normal ages for achieving breast and pubic hair development are described in Sun et al. (2002). men aged 20 –35 years, values of LH were normal in 7 and elevated in 3 samples. None of the adult men had LH levels below the lower normal limit of 0.8 IU/l. Relatively little variation in LH levels was seen in individual adult PWS men except for one 23 year-old whose LH levels fluctuated between 12.7 and 0.8 IU/l despite FSH values which were persistently elevated. After age 20 years, FSH was markedly high (34.4 + 11.5 mIU/ml) in 6/10 men, but was normal (3.5 + 1.6 mIU/ml) in the other four. Three young boys had high prolactin levels, possibly due to stressful venipunctures, however prolactin levels were normal in adolescent males. Hyperprolactinemia was noted, however, in 10 adult PWS men all of whom were receiving risperidone or SSRI medications. TSH levels were normal except for one 4-year-old boy whose TSH was 6.35 mIU/ml. For PWS males, the TSH level (mean + SD) was 2.07 + 1.03 (0.77 –5.14) mIU/ml. qualitatively similar to the pattern seen in normal boys. Levels were highest in early infancy and gradually declined to low levels by adolescence, remaining low throughout adulthood. Inhibin B was in the normal range (241 + 105 pg/ml) in infant males (,1 year) and fell to lower levels by 2 years (Fig. 4D). Some boys showed a transient rise in inhibin B between ages 9 and 20 years, but levels were low to undetectable in adults (65 + 58 pg/ml). SHBG levels were in the normal range for young infants up until age 2 years (Fig. 4B). From 2 years until early adolescence, SBHG levels were generally below the lower limit of normal. After age 15 years, levels were again in the normal range and were slightly elevated in the two oldest adult men, aged 31 and 36 years. DHEAS levels were within the normal range in most PWS males at all ages, but eight children and adolescent boys had high values (Fig. 4E). Testicular hormones, SHBG and DHEAS PWS females Testosterone fell from the normal mini-puberty values in infant boys to the low or undetectable range expected during childhood. Nine boys aged 10–15 years had normal testosterone levels (2.2 + 2.7 nmol/l) for their age, however testosterone was low (5.7 + 3.4 nmol/l) in all 10 men above age 20 years (Fig. 4A). Values for testosterone were relatively stable in all adults older than age 20 years. AMH levels in men showed the expected decrease with age, despite low testosterone (Fig. 4C). Although AMH levels in PWS were in the lower half of the reference range, changes in levels with age were Clinical features: BMI and sexual development BMI values for young PWS girls were near or within the normal range, increased as the girls entered adolescence, and were elevated in adult women. Three PWS girls began breast development before age 10 years (Fig. 2C). One girl had Tanner 3 breasts and pubic hair at age 10 years and a bone age of 11 –1/2 years. At age 12 years, her bone age was 14, however there was no progression of breast or pubic hair development and she had no menses. Subsequent to completion of this study, 2591 Reproductive hormones in Prader – Willi syndrome Figure 3 Serum levels of pituitary hormones. Changes in pituitary hormone levels with age in 49 individual PWS males and 57 PWS females. For PWS men, levels of LH, FSH and prolactin are shown in (A), (B) and (C) respectively. Levels of LH, FSH and prolactin for PWS women are shown in (D), (E) and (F), respectively. The upper and lower limits of each hormone for a normal, reference population are indicated by the solid gray lines (Esoterix Laboratory Services, Inc. http://www.esoterix.com/files/expected values.pdf; Hirsch et al., 2013). another 7-year-old girl showed spontaneous onset of puberty. Most women showed spontaneous progression of breast development, but achievement of breast stage 5 generally occurred well past the normal age of 15.5 years (Sun et al., 2002). Young PWS girls often had onset of pubic hair prior to the normal age of 10.6 years (Fig. 2D). Most adolescent girls and adult women reached pubic hair Tanner stage 4–5. Eight PWS women (5 DEL, 3 UPD) reported having spontaneous vaginal bleeding (Table I). The median age for the first episode of bleeding was 20 years with a range of 13.5 – 34 years. One woman had regular monthly menses, but all others had oligomenorrhea or secondary amenorrhea. The girl with vaginal bleeding at age 13.5 years had a BMI of 62.6 kg/m2, the highest BMI in our study. Mean BMI for the other seven women with spontaneous bleeding was 37.9 kg/m2, similar to the BMI in amenorrheic PWS women. No differences in LH, FSH, inhibin B, or AMH levels were seen between women who experienced vaginal bleeding compared with those PWS women above age 12 years who did not. E2, however, was significantly higher in women who had at least one episode of bleeding compared with those who had none (185 + 146 versus 95 + 106 pmol/l, P , 0.01). Pituitary hormones In females (Fig. 3D), LH and FSH were relatively high in infancy but decreased to low levels in early childhood. LH remained low in most of the young girls, with two exceptions. One 2-year-old girl had an LH level of 15.6 mIU/ml which then fell to 1.3 mIU/ml. On further followup, LH levels in this girl remained above the age-related normal range. Another girl had high LH levels ranging from 5.9 to 11.8 mIU/ml at ages 6–10 years. LH was normal in adolescent girls. In adult women over age 20 years, LH was normal in 6/13 (3.77 + 1.67 mIU/ml) but low in 7/13 (0.21 + 0.18 mIU/ml). Sequential samples showed relatively little variation for individual PWS women .20 years, i.e. no crossover between normal and low levels of LH was seen except for one 26-year-old with a low LH level whose levels subsequently increased to within the normal range. Low LH levels were more common in PWS women than in men. FSH (Fig. 3E) was relatively high in PWS infant girls, similar to levels seen in normal infants, but fell to the normal, lower values during childhood. One 15-year-old girl had a high FSH level which increased further in repeated samples and subsequently showed a modest decrease. FSH levels in older adolescents and adults were generally in the 2592 Hirsch et al. Figure 4 Serum levels of testicular hormones, SHBG and DHEAS in 49 PWS males. Values obtained for serum testosterone (A), sex hormone-binding globulin (SHBG) (B), anti-Müllerian hormone (AMH) (C), inhibin B (D) and dehydroepiandrosterone sulfate (DHEAS) (E) for individual PWS males are shown in the respective panels. The solid gray lines show the upper and lower limits for a normal, reference population (Rey et al., 1993; Lee et al., 1996; Andersson et al., 1997; Aksglaede et al., 2010; Hirsch et al., 2013; Josso et al., 2013). normal range, although three PWS women had consistently low values. In women aged 20 years or older, FSH was normal in 10/13 (6.42 + 3.78 mIU/ml) and low in three (0.71 + 0.60 mIU/ml). Except for a 1-month old PWS infant girl whose TSH level was 7.2 mIU/ml, all other girls and women had normal TSH levels (1.87 + 1.06 mIU/ml). Ovarian hormones, SHBG and DHEAS E2 levels were in the normal prepubertal range of ,73 pmol/l in PWS young girls and adolescents (Fig. 5A). E2 values were relatively low in adult PWS women, although measurements in two women were above the upper limit of normal for follicular phase levels. Only seven adult women had E2 levels (157 + 133 pmol/l) consistently above the lower limit of detection. SHBG (Fig. 5B) levels in the PWS females were mostly within the normal reference range, showing the expected decline from higher values in infancy to lower values in childhood and adolescence. Levels were slightly below the lower limit of normal in some PWS females during childhood. AMH (Fig. 5C) was below the normal median in most PWS females, although high levels (3.67 + 2.66 ng/ml) were measured in seven girls below age 12 years. Inhibin B (Fig. 5D) was low (19.6 + 16.8 pg/ml) or undetectable in all PWS females. DHEAS levels (Fig. 5E) were in the normal range for most PWS women, although an early and exaggerated rise in DHEAS was seen in Reproductive hormones in Prader – Willi syndrome 2593 Figure 5 Serum levels of ovarian hormones, SHBG and DHEAS in PWS females. Values obtained for serum estradiol (A), SHBG (B), AMH (C), inhibin B (D) and DHEAS (E) for 57 individual PWS females are shown in the respective panels. Levels of DHEAS in PWS females are shown in (B). The solid gray lines show the upper and lower limits for a normal, reference population (Esoterix Laboratory Services, Inc. http://www.esoterix.com/files/expected values. pdf; Sehested et al., 2000; Sørensen et al., 2007). seven PWS girls aged 5–9 years. Two women had DHEAS levels consistently above the upper limit of normal, but all other adult women had levels in the normal range. Discussion This is the first study to report sequential changes in pubertal development and reproductive hormones including inhibin B, AMH, DHEAS and SHBG, in a cohort of PWS individuals over a wide age-span from early infancy into the fourth decade of adult life. The specific profile of hypogonadism (hypothalamic versus primary gonadal dysfunction) in each individual becomes apparent in late adolescence and early adulthood in both sexes. Hormone patterns for each individual are consistent and stable after age 20 years. Our findings for LH, FSH and testosterone confirm results reported in our previous cross-sectional study (Hirsch et al., 2009), and are consistent with the longitudinal study on 68 PWS boys by Siemensma et al. (2012a). Their study, however, included only two men older than 20 years and did not report results for prolactin, AMH, DHEAS or SHBG. In their paper, one patient was diagnosed as having hypogonadotrophic hypogonadism at age 17 –8/12 years, but subsequently developed high gonadotrophin levels at age 24 years. We found that the hormonal profiles typical of primary or secondary hypogonadism become apparent only after age 20 years. For this reason, longitudinal 2594 follow-up of hormone levels into adulthood are needed in order to determine the type of hypogonadism for individual patients. Our study is a representative cross-section of individuals with PWS, including nearly all cases of this syndrome in Israel. Approximately 70% of the adults live in one of several residential hostels. PWS individuals who live in a hostel setting are more likely to maintain strict compliance with a diet and exercise regimen and thereby avoid severe obesity, which is more common in PWS individuals living at home with their families. We obtained samples from PWS infants, children and adults. The wide age range represented in our study clearly shows the decline in gonadotrophin, testosterone, and inhibin B levels from the ‘minipuberty’ values in infancy, to low levels in childhood, followed by an increase in gonadotrophin levels in adolescents and adults. The low levels of testosterone and inhibin B and very high levels of FSH in adult males demonstrate that a primary testicular defect is the major contributor to hypogonadism in PWS males. Although hyperprolactinemia, present in all PWS adult men (most probably related to the use of SSRI medications), has an inhibitory effect on gonadotrophin secretion, nevertheless FSH was elevated. High FSH despite elevated prolactin provides additional support to our findings of a primary testicular defect in this population. Although cryptorchidism is a common feature in PWS (Crino et al., 2003; Bakker et al., 2015a), we found that 3/49 boys had normally descended testes at birth and three others had spontaneous descent of undescended testes. Despite previous reports stating that micropenis is common in PWS infants (Crino et al., 2003), we found that penile length measurements were within the normal range during infancy and early childhood (Fig. 2A). The normal ‘mini-puberty’ gonadotrophin and testosterone levels found in PWS infant boys together with normal penile lengths despite the high prevalence of undescended testes, suggest that factors, other than hormones, are responsible for the lack of testicular descent in these infants (Fillion et al., 2006; Hirsch et al., 2013, 2014). The age at puberty onset in our cohort is similar to that reported by Radiocini who found that most PWS adolescent males experience the onset of puberty but do not progress beyond Tanner stage 3 (Radicioni et al., 2012). Testes were small in all of our PWS men, with only three adults reaching Tanner stage 4 and none achieved Tanner stage 5 for testes volume. Most adult men had penile lengths below the lower limit of normal. AMH levels showed a qualitatively normal decrease with age during childhood, and were low in all PWS adult men. By contrast, AMH levels are elevated in men with hypogonadotrophic hypogonadism (Eldar-Geva et al., 2010b; Josso et al., 2013). The low intratesticular testosterone levels in our patients may be sufficient to inhibit AMH secretion by Sertoli cells, but not high enough to result in full pubertal development or spermatogenesis. Alternatively, the low AMH levels in PWS may indicate progressive deterioration in Sertoli cell function. Our data showing that inhibin B levels were in the normal range through adolescence, but were undetectable in nearly all adults older than age 20 years (Hirsch et al., 2009) support the suggestion that Sertoli cell function declines during or after puberty in PWS males. Testicular histology from PWS males showed decreased or absent spermatogonia (Vogels et al., 2008). By age 15 years, testosterone levels were low suggesting that testosterone replacement might be appropriate as early as age 15 or 16 years, although careful monitoring of growth and skeletal maturation would be needed to avoid deleterious effects on final height. Hirsch et al. Despite low levels of testosterone, AMH and inhibin B in PWS adult males, we reported normal levels of insulin-like peptide-3, indicating that testicular failure in PWS is not due to atrophic changes of entire testes, but rather represents a specific and unique pattern of hypogonadism in PWS (Hirsch et al., 2013). Most infants and children in our study were treated with GH, however, none of the adults were treated with GH during or prior to the study. Although it appears that GH treatment does not affect reproductive hormones, particularly AMH, in children, some long-term effects are possible. A recent study (Tinggaard et al., 2014) found no changes in gonadotrophins and ovarian steroids in 18 short pre-pubertal, small for gestational age (SGA) girls following up to 3 years of GH treatment. AMH levels and ovarian volumes increased, but the differences were not significant compared with findings in SGA girls who had not received GH. Other studies found no direct effect of GH on AMH levels in Turner syndrome (Visser et al., 2013), SGA boys (Boonstra et al., 2008), or in SGA girls (Lem et al., 2011). Hypogonadism in PWS females is more heterogeneous than in males and is characterized by low E2, normal AMH, but very low or undetectable levels of inhibin B. Gonadotrophin levels are generally in the normal range. As for PWS infant males, LH and FSH were in the minipuberty range for PWS infant females and levels decreased in early childhood. Many of the PWS women in our study achieved full, Tanner stage 5 breast development. Most were amenorrheic, but eight women did have at least one episode of vaginal bleeding, although the median age was 20 years (range 13– 34 years). In the study by Siemensma et al., 5 of 61 PWS girls had menarche at ages ranging from 12 years to 16.5 years (Siemensma et al., 2012b), however, only one patient was followed beyond 19 years of age. One of these women continued to have regular menstrual periods, while two had oligomenorrhea and the other two had secondary amenorrhea. Nevertheless, vaginal bleeding in PWS obese adolescent or adult women may not indicate normal ovarian function or true menarche, but more likely is due to unopposed estrogens of adipose tissue origin. Measurement of serum inhibin B levels may be important for management of adult PWS women. Inhibin B levels in individual women showed considerable variability with levels ranging from undetectable to values within the normal range. Women whose inhibin B levels are within the normal range may have potential for fertility, and contraception may be indicated for this subgroup of PWS women (Eldar-Geva et al., 2013). Since inhibin B levels in individual women are variable, recommendations for treatment should be made only after the results of multiple samples are available. Although precocious puberty is uncommon in PWS, several case reports have been published (Crino et al., 2008; Pusz and Rotenstein, 2008; Lee and Hwang, 2013). In our cohort, one 10-year-old girl had rapidly advancing puberty, but by age 17 years she was still premenarchal. Mutations in MKRN3, an imprinted gene located in the critical PWS region of chromosome 15, have recently been shown to be associated with familial precocious puberty (Abreu et al., 2013). Most PWS individuals, however, have delayed or incomplete pubertal development. In conclusion, we described the development of reproductive hormone patterns in PWS males and females. The type of hypogonadism (hypothalamic versus primary gonadal dysfunction) becomes apparent only in late adolescence and early adulthood in PWS men and women. Hormonal profiles are stable for each PWS individual after age 20 years. Primary gonadal dysfunction is a major component of hypogonadism in Reproductive hormones in Prader – Willi syndrome PWS, particularly in men, who show a steadily progressive deterioration in testicular function. Our findings support recommendations for earlier initiation of testosterone replacement in PWS adolescent boys (Bakker et al., 2015b). Recognition of age-related changes in reproductive hormones is important for tailoring individualized recommendations for contraception or hormone replacement and counseling. Acknowledgements We are grateful to Nava Badichi, BA, for administrative assistance in coordinating blood sampling and data analysis and to Nurit Algor, MSc, for performing the laboratory tests. Authors’ roles H.J.H., T.E.-G. and V.G.-T. each participated in planning the study, examining the patients, analyzing the data and writing the manuscript. F.B. contributed by evaluating the patients and reviewing the manuscript. Y.P. performed statistical analyses of the data and assisted in preparing the manuscript. Funding This study was funded by grants from the Foundation for Prader-Willi Research and Pfizer Pharmaceuticals. Conflict of interest T.E.-G. and V.G.-T. received grant/research support funding from the Foundation for Prader-Willi Research and Pfizer Pharmaceuticals. References Abreu AP, Dauber A, Macedo DB, Noel SD, Brito VN, Gill JC, Cukier P, Thompson IR, Navarro VM, Gagliardi PC et al. Central precocious puberty caused by mutations in the imprinted gene MKRN3. N Engl J Med 2013;386:2467 – 2475. Aksglaede L, Sørensen K, Boas M, Mouritsen A, Hagen CP, Jensen RB, Petersen JH, Linneberg A, Andersson AM, Main KM et al. Changes in anti-Müllerian hormone (AMH) throughout the life span: a population-based study of 1027 healthy males from birth (cord blood) to the age of 69 years. J Clin Endocrinol Metab 2010;95:5357 – 5364. Andersson AM, Juul A, Petersen JH, Müller J, Groome NP, Skakkebaek NE. Serum Inhibin B in healthy pubertal and adolescent boys: relation to age, stage of puberty, and follicle-stimulating hormone, luteinizing hormone, testosterone, and estradiol levels. J Clin Endocrinol Metab 1997; 82:3976– 3981. Bakker NE, Wolffenbuttel KP, Looijenga LH, Hokken-Koelega AC, Burman P, Ritzen EM, Lindgren AC. Testes in infants with Prader-Willi syndrome: human chorionic gonadotropin treatment, surgery and histology. J Urol 2015a;193:291 – 298. Bakker NE, Kuppens RJ, Siemensma EP, Tummers-de Lind van Wijngaarden RF, Festen DA, Bindels-de Heus GC, Bocca G, Haring DA, Hoorweg-Nijman JJ, Houdijk EC et al. Bone mineral density in children and adolescents with Prader-Willi syndrome: a longitudinal study during puberty and 9 years of growth hormone treatment. J Clin Endocrinol Metab 2015b;100:1609 – 1618. 2595 Boonstra VH, Weber RF, de Jong FH, Hokken-Koelega AC. Testis function in prepubertal boys and young men born small for gestational age. Horm Res 2008;70:357 – 363. Burman P, Ritzen EM, Lindgren AC. Endocrine dysfunction in Prader-Willi syndrome: a review with special reference to GH. Endocr Rev 2001; 22:787– 799. Cassidy SB, Driscoll DJ. Prader-Willi syndrome. Euro J Hum Genet 2009; 17:3– 13. Crino A, Schiaffini R, Ciampalini P, Spera S, Beccaria L, Benzi F, Bosio L, Corrias A, Gargantini L, Salvatoni A et al. Hypogonadism and pubertal development in Prader-Willi syndrome. Eur J Pediatr 2003;162:327– 333. Crino A, Di Giorgio G, Schiaffini R, Fierabracci A, Spera S, Maggioni A, Gattinara GC. Central precocious puberty and growth hormone deficiency in a boy with Prader-Willi syndrome. Eur J Pediatr 2008; 167:1455– 1458. Diene G, Mimoun E, Feigerlova E, Caula S, Molinas C, Grandjean H, Tauber M; French Reference Centre for PWS. Endocrine disorders in children with Prader-Willi syndrome—data from 142 children of the French database. Horm Res Paediatr 2010;74:121 – 128. Eiholzer U, l’Allemand D, Rousson V, Schlumpf M, Gasser T, Girard J, Gruters A, Simoni M. Hypothalamic and gonadal components of hypogonadism in boys with Prader-Labhardt-Willi syndrome. J Clin Endocrinol Metab 2006;91:892 – 898. Eldar-Geva T, Hirsch HJ, Benarroch F, Rubinstein O, Gross-Tsur V. Hypogonadism in females with Prader-Willi syndrome from infancy to adulthood: variable combinations of a primary gonadal defect and hypothalamic dysfunction. Eur J Endocrinol 2010a;162:377 – 338. Eldar-Geva T, Liberty G, Chertin B, Fridmans A, Farkas A, Margalioth AJ, Spitz IM. Relationships between FSH, Inhibin B, anti-Mullerian hormone, and testosterone during long-term treatment with the GnRH-agonist histrelin in patients with prostate cancer. Eur J Endocrinol 2010b; 162:177– 181. Eldar-Geva T, Hirsch HJ, Pollak Y, Benarroch F, Gross-Tsur V. Management of hypogonadism in adolescent girls and adult women With Prader– Willi syndrome. Am J Med Genet A 2013;161A:3030– 3034. Endocrinology expected values and SI unit conversion pocket book. Austin, TX: Esoterix Laboratory Services, Inc. http://www.esoterix.com/files/ expectedvalues.pdf. Fillion M, Deal CL, Van Vliet G. Normal minipuberty of infancy in boys with Prader-Willi syndrome. J Pediatr 2006;149:874– 876. Goldstone AP, Holland AJ, Hauffa BP, Hokken-Koelega AC, Tauber M; speakers contributors at the Second Expert Meeting of the Comprehensive Care of Patients with PWS. Recommendations for the diagnosis and management of Prader-Willi syndrome. J Clin Endocrinol Metab 2008;93:4183– 4197. Gross-Tsur V, Hirsch HJ, Benarroch F, Eldar-Geva T. The FSH-Inhibin axis in Prader-Willi syndrome: heterogeneity of gonadal dysfunction. Reprod Biol Endocrinol 2012;10:39. doi:10.1186/1477-7827-10-39. Hirsch HJ, Eldar-Geva T, Benarroch F, Rubinstein O, Gross-Tsur V. Primary testicular dysfunction is a major contributor to abnormal pubertal development in males with Prader-Willi syndrome. J Clin Endocrinol Metab 2009;94:2262– 2268. Hirsch HJ, Eldar-Geva T, Gross-Tsur V, Benarroch F, Roger M, Lahlou N. Normal insulin-like peptide-3 levels despite low testosterone in adult males with Prader-Willi syndrome: variations in leydig cell function from infancy through adulthood. J Clin Endocrinol Metab 2013; 98:E135 – E114. Hirsch HJ, Eldar-Geva T, Erlichman M, Pollak Y, Gross-Tsur V. Characterization of minipuberty in infants with Prader-Willi syndrome. Horm Res Paediatr 2014;82:230 – 237. Jensen TK, Andersson AM, Jørgensen N, Andersen AG, Carlsen E, Petersen JH, Skakkebaek NE. Body mass index in relation to semen 2596 quality and reproductive hormones among 1,558 Danish men. Fertil Steril 2004;82:863 – 870. Josso N, Rey RA, Picard JY. Anti-mullerian hormone: a valuable addition to the toolbox of the pediatric endocrinologist. Int J Endocrinol 2013; 2013:674105. doi: 10.1155/2013/674105. Epub 2013 Dec 8. Kalra B, Kumar A, Patel K, Patel A, Khosravi MJ. Development of a second generation inhibin B ELISA. J Immunol Methods 2010;362: 22 – 31. Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, Flegal KM, Guo SS, Wei R, Mei Z, Curtin LR, Roche AF, Johnson CL. CDC growth charts: United States. Advance data from vital and health statistics; no. 314. National Center for Health Statistics, 2000. Lee HS, Hwang JS. Central precocious puberty in a girl with Prader-Willi syndrome. J Pediatr Endocrinol Metab 2013;26:1201– 1204. Lee MM, Donahoe PK, Hasegawa T, Silverman B, Crist GB, Best S, Hasegawa Y, Noto RA, Schoenfeld D, MacLaughlin DT. Mullerian Inhibiting Substance in humans: normal levels from infancy to adulthood. J Clin Endocrinol Metab 1996;81:571 – 576. Lem AJ, Boonstra VH, Renes JS, Breukhoven PE, de Jong FH, Laven JS, Hokken-Koelega AC. Anti-Mullerian hormone in short girls born small for gestational age and the effect of growth hormone treatment. Hum Reprod 2011;26:898 – 903. Pusz ER, Rotenstein D. Treatment of precocious puberty in a female with Prader-Willi syndrome. J Pediatr Endocrinol Metab 2008; 21:495 – 500. Radicioni AF, Di Giorgio G, Grugni G, Cuttini M, Losacco V, Anzuini A, Spera S, Marzano C, Lenzi A, Cappa M et al. Multiple forms of hypogonadism of central, peripheral or combined origin in males with Prader-Willi syndrome. Clin Endocrinol (Oxf) 2012;76:72 – 77. Rey R, Lordereau-Richard I, Carel JC, Barbet P, Cate RL, Roger M, Chaussain JL, Josso N. Anti-Mullerian hormone and testosterone serum levels are inversely related during normal and precocious pubertal development. J Clin Endocrinol Metab 1993;77:1220 – 1226. Hirsch et al. Schonfeld WA, Beebe GW. Normal growth and variation in the male genitalia from birth to maturity. J Urol 1942;48:759 – 777. Sehested A, Juul AA, Andersson AM, Petersen JH, Jensen TK, Muller J, Skakkebaek NE. Serum inhibin A and inhibin B in healthy prepubertal, pubertal, and adolescent girls and adult women: relation to age, stage of puberty, menstrual cycle, follicle-stimulating hormone, luteinizing hormone, and estradiol levels. J Clin Endocrinol Metab 2000;85:1634–1640. Siemensma EP, de Lind van Wijngaarden RF, Otten BJ, de Jong FH, Hokken-Koelega AC. Testicular failure in boys with Prader-Willi syndrome: longitudinal studies of reproductive hormones. J Clin Endocrinol Metab 2012a;97:E452–E459. Siemensma EP, van Alfen-van der Velden AA, Otten BJ, Laven JS, Hokken-Koelega AC. Ovarian function and reproductive hormone levels in girls with Prader-Willi Syndrome: a longitudinal study. J Clin Endocrinol Metab 2012b;97:E1766–E1773. Sørensen K, Andersson AM, Skakkebæk NE, Juul A. Serum sex hormone-binding globulin levels in healthy children and girls with precocious puberty before and during gonadotropin-releasing hormone agonist treatment. J Clin Endocrinol Metab 2007;92:3189 – 3196. Sun SS, Schubert CM, Chumlea WC, Roche AF, Kulin HE, Lee PA, Himes JH, Ryan AS. National estimates of the timing of sexual maturation and racial differences among US children. Pediatrics 2002;110:911– 919. Tanner JM. Growth at Adolescence. Oxford: Blackwell Scientific, 1962, 88–102. Tinggaard J, Jensen RB, Sundberg K, Birkebæk N, Christiansen P, Ellermann A, Holm K, Jeppesen EM, Kremke B, Marcinski P et al. Ovarian morphology and function during growth hormone therapy of short girls born small for gestational age. Fertil Steril 2014;102:1733 – 1741. Visser JA, Hokken-Koelega AC, Zandwijken GR, Limacher A, Ranke MB, Flück CE. Anti-Müllerian hormone levels in girls and adolescents with Turner syndrome are related to karyotype, pubertal development and growth hormone treatment. Hum Reprod 2013;28:1899– 1907. Vogels A, Moerman P, Frijns JP, Bogaert GA. Testicular histology in boys with Prader-Willi syndrome: fertile or infertile? J Urol 2008;180:1800– 1804.