Tall Stature without Growth Hormone: Four Male Patients with Aromatase Deficiency

Abstract

Context: From preliminary observations, GH-IGF-I seems to be compromised in men with aromatase deficiency. The GH deficiency (GHD) coexists paradoxically with tall stature, raising the question whether or not a true GHD is part of this rare syndrome.

Objective: To evaluate the GH secretion in aromatase-deficient men, their GH response to the GHRH plus arginine (GHRH-ARG) test was compared with that of normal subjects. The effect of estrogen replacement treatment on the GH-IGF-I axis in aromatase-deficient men was evaluated before and during therapy.

Design and Setting: A case-control study was conducted.

Patients: Four adult men with aromatase deficiency were compared with 12 normal subjects.

Main Outcome Measures: We measured the GH response to GHRH-ARG in aromatase-deficient men (at baseline and during estrogen treatment) and in normal subjects. Basal serum IGF-I was measured in both patients and controls.

Results: The response of GH to GHRH-ARG was severely impaired in men with aromatase deficiency and resulted in significantly lower (P < 0.001) levels than in normal subjects. Although normal, serum IGF-I levels were also significantly lower (P < 0.001) than in normal subjects. Both GH peak and IGF-I concentrations were not modified by estrogen therapy in men with aromatase deficiency.

Conclusions: In aromatase-deficient men, GH response to potent provocative stimuli is impaired and is not restored by exogenous estrogens. Furthermore, a tall stature may be reached, notwithstanding the coexistence of GHD, if a prolonged time for growth is available due to a delay in bone maturation, and other growth factors different from GH (mainly insulin) promote growth.

The knowledge of the role of estrogen in male physiology (1) has received growing attention with the discovery of natural models of human male estrogen deprivation, such as estrogen resistance (2) and aromatase deficiency (3), together with the development of both estrogen receptor and aromatase gene knockout mice (4).

The characterization of the few cases of male aromatase deficiency diagnosed until now (a total of eight male subjects) highlighted both the relevant clinical phenotype and the underlying genetic defects (5). In particular, men with aromatase deficiency present tall stature with a height exceeding the genetic target, eunuchoid proportions of the skeleton (increased arm span, genu valgum, and increased upper-to-lower segment ratio), and overweight (5). In men, estrogens are required for the achievement of bone maturation, epiphyseal closure, and normal skeletal proportions (6–13). In addition, estrogen action is important for the acquisition of adequate peak and subsequent maintenance of bone mass (12, 14).

Estrogens, mainly the circulating fraction, also modulate the gonadotropin feedback in men (15, 16), both at the pituitary (17) and the hypothalamic level (18–20), but the role of estrogen on the secretion of other anterior pituitary hormones is not completely explained. Preliminary observations suggest that the GH-IGF-I axis may be compromised in men with aromatase deficiency (8, 11, 16, 21) and that estrogen replacement treatment does not increase GH secretion (16), differently from normal men (22, 23).

With this in view, in the setting of aromatase deficiency the supposed GH deficiency (GHD) coexists with a condition of tall stature in a paradoxical way, thus raising the question whether or not a true GHD is part of this rare syndrome.

The aim of this study was to further evaluate the GH secretion in men with aromatase deficiency by comparing the GH response to the combined GHRH plus arginine (GHRH-ARG) test of patients with that of matched normal healthy men. In addition, the role of exogenous estrogens in the GH-IGF-I axis in men with aromatase deficiency was evaluated before and during estrogen replacement treatment by using amounts of estrogens that ensured serum physiological estradiol levels for an adult male.

Subjects and Methods

Subjects

We compared the GH-IGF-I axis of four adult men with aromatase deficiency, whose clinical and genetic aspects have been already published (6, 9–11, 16), with that of 12 normal male subjects. All patients with aromatase deficiency sought medical consultation for a persistent linear growth in adulthood that resulted in a condition of tall stature (higher than that expected on the basis of the genetic target), associated with a progressive derangement of skeletal proportions (6, 9–11). Among patients with aromatase deficiency, subject 1 is the case first described by Carani et al. (6, 16); subject 2 and subject 3 have been described by Maffei et al. (9, 10); and subject 4 has been described recently by Lanfranco et al. (11). The data coming from subject 1 have been published already (16). Mutations of the aromatase gene were different among the four patients, who came from different geographical areas, and the four patients were not related.

The results of patients’ genetic analysis have been previously described in detail (6, 9–11) and summarized, at least in part, in a review (5). Comparisons with the published sequence of the human CYP19A1 gene revealed: 1) a homozygous point G→A mutation at bp 1094 in exon IX in subject 1 (5, 6); 2) a homozygous point G→A mutation at the last nucleotide of exon V in subject 2 (5, 9); 3) compound heterozygosity for two point mutations at bp 380 (T→G) in exon IV and bp 1124 (G→A) in exon IX in subject 3 (5, 10); and 4) compound heterozygosity due to 23-bp deletion in exon IV and a point G→T mutation in the first nucleotide of intron IX in subject 4 (11). The results of the genetic defects were: 1) the abolishment of a site of cleavage in subject 1 (5, 6); 2) an aberrant splicing of the mRNA, resulting in a truncated inactive aromatase protein lacking the heme-binding region in subject 2 (5, 9); 3) a change in protein polarization with a distortion of the tertiary or quaternary structure in subject 3 (5, 10); and 4) a frame shift with a premature stop codon at nucleotide 361 in exon IV resulting in a truncated, shorter, and inactive aromatase protein in subject 4 (11). In all four subjects, transfection studies showed a loss of aromatase function (5, 6, 9–11).

Normal subjects were recruited at the Endocrine Unit of Turin and matched to patients with aromatase deficiency for age (as mean ± sd, aromatase-deficient men, 27.75 ± 6.4 yr; normal subjects, 29.9 ± 5.6 yr), sex, and body mass index (BMI) (as mean ± sd, aromatase-deficient men, 29.7 ± 4.03 kg/m²; normal subjects, 29.7 ± 2.9 kg/m²). Inclusion criteria for normal subjects were normal serum estradiol and total testosterone levels and absence of endocrine dysfunction or other major chronic diseases.

All patients with aromatase deficiency underwent estrogen replacement therapy with estradiol patch [Estraderm TTS patch system (Novartis Pharmaceuticals, Basel, Switzerland), at the dosage of 25 μg of estradiol twice weekly for subjects 1, 2, and 3] and estradiol gel [Gelestra (Abiogen Pharma SpA, Pisa, Italy), at the dosage of 0.75 mg of estradiol per day for subject 4], leading to estradiol circulating concentrations similar to those recorded in the normal male population. Aromatase-deficient men were studied before and during estrogen replacement treatment. In particular, the study under estrogen therapy was performed after the normalization of serum estradiol levels and at least 6 months after the onset of the treatment. In particular, the GH-IGF-I axis was studied after 15 months of replacement treatment in subject 1, after 6 months in subject 2, after 12 months in subject 3, and after 60 months in subject 4. At that time, all patients with aromatase deficiency had detectable serum estradiol within the normal range for male population (subject 1, 32 pg/ml; subject 2, 23.5 pg/ml; subject 3, 13 pg/ml; and subject 4, 15.2 pg/ml) as already published (6, 9–11, 16).

Methods

Aromatase-deficient men and normal subjects underwent hormonal analyses to investigate GH secretion and GH-IGF-I status. Only aromatase-deficient men were investigated twice (at baseline and during estrogen treatment). Blood samples were taken at 0800 h after an overnight fast. All subjects underwent a standard GHRH-ARG test [GHRH 1–29 (Geref, Serono, Italy); 1 μg/kg iv at 0 min; arginine hydrochloride, 0.5 g/kg iv over 30 min from 0 to +30 min, up to a maximum of 30 g]. Blood samples for GH evaluation were taken every 15 min from 0 to +90 min, as previously standardized (24). Cutoffs for a normal GH response to GHRH-ARG test were chosen according to reference data available in literature for normal healthy men (24, 25) and for overweight/obese men (25, 26). Basal serum IGF-I, insulin, glucose, and prolactin were also measured in aromatase-deficient men as well as IGF-I in normal subjects.

Hormone assays

All serum samples were stored at −80 C until assayed. Serum hormones were measured by commercially available kits.

Serum estradiol levels were measured by a double antibody RIA (Third-Generation DSL-39100; Diagnostic System Laboratories, Inc., Webster, TX). The sensitivity of the assay was 0.6 pg/ml, with the lowest standard at 1.5 pg/ml. The ranges of inter- and intraassay coefficients of variation were 4.1–9.9 and 3.4–3.9%, respectively.

Serum GH was assayed by immunoradiometric method assay method (HGH-CTK IRMA; Diasorin, Saluggia, Italy). All samples from an individual subject were analyzed together. The sensitivity of the method was 0.15 μg/liter. The inter- and intraassay coefficients of variation were 3.5–4.4 and 5.1–7.5%, respectively, at GH levels of 1.98–41.92 and 2.99–42.45 μg/liter, respectively.

Serum IGF-I was assayed by radioimmunometric method (RIA) (SM-C-RIACT; Pantec, Turin, Italy) after acid-ethanol extraction to avoid interference by binding proteins. The sensitivity of the method was 0.1 ng/ml, with a normal range of 101 to 103 ng/ml. The inter- and intraassay coefficients of variation were 5.0–9.5 and 8.8–10.8%, respectively, at IGF-I levels of 79.41–712.3 and 79.6–766.4 ng/ml, respectively.

Serum prolactin was measured by a fluoroimmunoassay (Autodelfia hPRL kit; Wallac, Oy, Turku, Finland) with a sensitivity of 0.04 ng/ml and a normal range of 2.4 to 11.5 μg/liter. The inter- and intraassay coefficients of variation for prolactin were 3.5 and 2.9%, respectively.

Serum insulin was detected by a immunochemiluminometric assays (ADVIA Centaur Insulin; Bayer Diagnostics, Leverkusen, Germany). The normal range varied from 5 to 20 μU/ml. The inter- and intraassay coefficients of variation were 5 and 4%, respectively.

Serum glucose was assayed by means of commercially available kits.

Ethics and statistical analysis

The local Ethics Committees at the University of Modena and Reggio Emilia and at the University of Torino approved the study. All subjects provided informed consent to participate to the study. All samples from an individual subject were measured in the same assay.

After testing the normal distribution of the variable by means of the Kolmogorov-Smirnov test, the statistical analysis was carried out using a paired t test for comparing patients before and during estrogen treatment and using an unpaired t test when comparing patients with normal subjects. Levels of statistical significance were set at a P value less than 0.05. After Bonferroni’s correction for multiple comparisons, a P < 0.02 was considered statistically significant. Computations were performed using the statistical software package SPSS (Windows version 14; SPSS, Chicago, IL).

Results

Before estrogen therapy, aromatase-deficient men showed an impaired GH response to the provocative GHRH-ARG test, consistent with a condition of severe GHD, because the GH peak was less than the standardized BMI-related (for overweight) cutoff of 9 μg/liter (25, 26) (Fig. 1).

GH peak after GHRH-ARG did not change before (2.0, 1.5, 1.0, 2.8 μg/liter for subjects 1, 2, 3, and 4, respectively; mean ± sd, 1.825 ± 0.96 μg/liter) and during estrogen treatment (paired t test) (1.3, 1.8, 1.5, 2.6 μg/liter for subjects 1, 2, 3, and 4, respectively; mean ± sd, 1.8 ± 0.96 μg/liter) (Fig. 1).

Before estrogen therapy, mean ± sd serum IGF-I levels in aromatase-deficient men were 114.5 ± 41.32 ng/ml, remaining in the lowest quartile of the normal range for age-matched male subjects (27), and were not significantly modified (mean ± sd, 119.25 ± 34.88 ng/ml) (paired t test) by estrogen replacement therapy (Fig. 2). Thus, IGF-I levels remained in the lower limit of the normal range throughout the study (Fig. 2).

As expected, GH peak response to GHRH-ARG in normal subjects was normal on the basis of the standardized BMI-related cutoffs (mean ± sd, 37.6 ± 19.6 μg/liter) (25, 26) (Fig. 1), associated with normal IGF-I concentrations (mean ± sd, 281.2 ± 67.7 ng/ml) (Fig. 2). Thus, the GH response to GHRH-ARG in patients with aromatase deficiency was significantly lower (P < 0.001) than in normal subjects (unpaired t test) (Fig. 1). In the same way, the IGF-I levels in aromatase-deficient men were significantly (P < 0.001) lower than normal subjects (unpaired t test) (Fig. 2).

Except for subject 2, who had a mild concomitant hypotestosteronemia (9), in all patients with aromatase deficiency, before estrogen replacement therapy, basal testosterone concentrations were within the normal range, whereas serum estradiol was undetectable (6, 10, 11). Transdermal estrogen therapy normalized estradiol levels and reduced LH, FSH, and total testosterone concentrations in all aromatase-deficient men (6, 9–11). Serum prolactin was at the lower end of the normal range (mean ± sd, 12.6 ± 6.8 μg/liter) and did not change during estradiol treatment (mean ± sd, 3.8 ± 1.6 μg/liter). Serum basal insulin was higher than the highest end of the normal range before estradiol treatment (mean ± sd, 55.2 ± 30.7 μU/ml), decreased during estradiol treatment (mean ± sd, 34.6 ± 21.2 μU/ml), but remained above the normal range (Table 1). Conversely, no difference in fasting serum glucose was found before (mean ± sd, 111 ± 47.5 mg/dl) and during (mean ± sd, 103.7 ± 28.5 mg/dl) estradiol treatment.

Transdermal estrogen therapy led to a closure of epiphyseal cartilage and to a slight increase in bone age (from a mean bone age of 15 yr before estradiol treatment to a mean bone age higher than 17 yr after at least 6 months of estradiol treatment) and height (with a mean gain of 1 cm in height) in all aromatase-deficient men (6, 9–11) (Table 2).

Discussion

The results of this study suggest that in men with aromatase deficiency: 1) GH response to potent provocative stimuli is impaired; 2) exogenous estrogens are not able to restore a normal GH secretion; and 3) a tall stature (higher than expected) may be reached notwithstanding the coexistence of GHD. In the present setting, the deranged GH secretion in adult men with aromatase deficiency is consistent with a condition of severely impaired GH peak after GHRH-ARG (25, 28, 29) that resembles that of patients with acquired or congenital hypopituitarism, due to hypothalamic-pituitary diseases and with manifest GHD. The GHD is confirmed in our aromatase-deficient men also when the BMI-weighted cutoffs of the GH peak after GHRH-ARG are considered (Table 1) (25, 26), when their GH peak is compared with that of normal subjects (Fig. 1), and finally when serum IGF-I is considered. Accordingly, patients with aromatase deficiency significantly differ from normal subjects, because aromatase-deficient men have lower IGF-I concentrations than controls, although still at the lower limit of the normal range (Fig. 2).

Our findings reinforce our previously published observation that GH response to the two most potent stimuli (insulin-induced hypoglycemia and GHRH-ARG infusion) was impaired in a subject with aromatase deficiency (subject 1 in this study) (16). Accordingly, repeated measurements of serum IGF-I were below the normal range in another aromatase-deficient man studied by Herrmann et al. (21), even during estrogen replacement treatment. In addition to humans (16, 21), congenital estrogen deficiency was also associated with impaired pituitary GH secretion in a mice model of aromatase deficiency (4, 30).

In addition, estradiol replacement treatment was not able to restore GH response to GHRH-ARG in all four aromatase-deficient men, differently from what happens in normal men (23), suggesting that the pituitary may be partially insensitive to sex steroids, particularly estrogens, in men with aromatase deficiency. This condition seems not to be dependent from the time course of estrogen replacement treatment because no change was recorded even for a long period after the initiation of the adequate estrogen substitutive therapy (15 months for subject 1, and 60 months for subject 4). It is well known that estrogen plays an important role in the normal GH-IGF-I axis in normal men by increasing GH-IGF-I secretion (23, 31), an effect that is manipulated at the time of puberty as a strategy to maximize growth in boys with short stature (32). Generally, estrogens increase GH secretion in normal men (22, 23, 31, 32) and in mice (4, 30), whereas the four aromatase-deficient men were unresponsive to a dosage of estrogen able to modify the gonadotropin secretion (15, 19). This partial pituitary insensitivity to estrogens is difficult to explain in this setting. The congenital lack of estrogen may have interfered with the normal process of development of the somatotrophic axis in these patients because preliminary results suggest a possible role of estrogens as one of the factors involved in the modulation of both pituitary cell function (33) and pituitary cell proliferation (34), early during the maturation of the hypothalamic-pituitary unit (30). Accordingly, some clinical evidences speak in favor of a strict linkage between estrogen and GH-IGF-I axis maturation, at least during early development (4, 30, 33, 34) and at puberty (16, 32), although details of the underlying mechanisms still remain unknown. Thus, it is possible that a mechanism of priming like that operating for other pituitary axis (15, 19) may also be needed for a complete maturation of the GH-IGF-I axis and that this process may fail in men with aromatase deficiency due to the congenital estrogen deprivation.

The results of this study confirm that longitudinal bone growth and a slow, progressive, and continuous increase in height during adulthood (Table 2) are possible even when the GH-IGF-I axis is compromised by the presence of low levels of GH and low to normal levels of IGF-I; indeed, this clinical condition may still paradoxically result in tall stature. Several similar clinical settings have been described in literature as characterized by concomitant GHD and normal or accelerated growth progression. Of these, panhypopituitarism due to absence of the pituitary stalk (35) and GHD due to craniopharyngioma removal from the hypothalamic-pituitary area (36–38) need to be remembered. How it is possible that growth continues to progress, although slowly, in men with impaired GH-IGF-I axis remains challenging, and the setting of this study does not allow providing evidence, but some speculations may be issued to launch possible explanations that need to be confirmed by further evidence in the future. In particular, the following hypotheses may explain how growth in height had been possible in these patients even in the absence of GH.

1. 1.

   Definitively, the patients with aromatase deficiency had a prolonged time available for linear growth due to the delayed diagnosis of aromatase deficiency and the consequent unfused epiphyses during adulthood (5, 39), as happens in rare cases of gigantism associated with severe hypopituitarism in which the concomitant hypogonadism leads to a delay in bone maturation (35). Indeed, despite the presence of severe growth retardation in childhood, the open epiphyses surely allow the continuing linear growth during adulthood, counterbalancing the short stature and resulting in a final height taller than the genetic target (35). The increase in height may be possible thanks to other factors able to promote a slow growth and/or to a minimal residual aromatase activity resulting in very small amounts of estrogen produced in situ that may directly or indirectly (through GH, IGF-I, or other growth factors) promote linear bone growth (40). A similar pattern of growth is obtained in pubertal boys by administering an aromatase inhibitor that is able to prolong the time window useful to growth even if a concomitant reduction of circulating IGF-I occurs (41). Even in this model of pharmacologically induced partial aromatase insufficiency, growth may proceed even with a concomitant decrease of circulating IGF-I and estradiol (41).

2. 2.

   We do not know in detail the natural history of aromatase deficiency, the precise growth rate of these patients during childhood, or their GH and IGF-I status at that time (42); therefore, it is also possible that GH secretion had been normal for a long time during infancy and/or puberty and that a secretory dysfunction had appeared only in adulthood, i.e. an adult-onset phenomenon. In this case, a normal GH-IGF-I status ensured a normal growth during childhood, puberty, and finally during late adolescence thanks to the absence of epiphyseal closure.

3. 3.

   The GH response to GHRH-ARG indicates that the GH secretory reserve of the pituitary is impaired in aromatase-deficient men, but it does not provide evidence on the impaired daily tonic GH secretion that may be conserved in these patients. With this in view, the severe GHD in these patients may also have contributed to the lack of the growth spurt at puberty because the combined actions of sex steroids, especially estrogens, and GH are assumed to be necessary for the occurrence of the normal pubertal growth spurt in boys (32), but continuous daily tonic GH secretion may have stimulated growth in these males whose epiphyseal growth plates remained open due to the lack of normal tissue estrogen levels.

4. 4.

   In different settings (35, 36–38), several authors suggested that hyperprolactinemia might also maintain a normal growth without GH by interacting, at least in part, with the GH receptor. Serum prolactin, however, in aromatase-deficient men was at the lower end of the normal range (Table 1); thus, this mechanism seems not to be involved in these patients.

5. 5.

   Among factors other than GH that may have supplied GH and IGF-I deficiency, insulin should be considered as the most important because it is higher than normal in aromatase-deficient men (3, 5, 6–12, 42) and the excess of insulin is considered one of the possible mechanisms involved in overgrowth in some clinical conditions, such as obesity, infants of diabetic mothers, congenital lipodystrophy, and syndromes characterized by neonatal hyperinsulinemia (43, 44). In addition, insulin is able to stimulate IGF-I synthesis and to directly promote the growth of chondrocytes within the growth plate (43), thus contributing directly and indirectly to body growth. The same mechanism has also been suggested in similar clinical conditions characterized by the paradox of excessive growth without GH (35, 36–38). Thus, it is possible to speculate that hyperinsulinism may have ensured low to normal serum IGF-I concentrations in aromatase-deficient men and that both insulin and IGF-I may have acted within the growth plate by maintaining a continuous but slow pattern of growth. Accordingly, there is evidence that IGF-I alone is able to promote growth even in conditions of severe GH resistance as in Laron syndrome (43, 45), and that insulin may have acted as a growth factor instead of GH and IGF-I in these patients.

Conclusions

The congenital severe estrogen deficiency may have interfered with the normal development of the somatotropic axis in patients with aromatase deficiency, resulting in GHD and low sensitivity of the GH-IGF-I axis to circulating estrogens in adulthood. A real estimate of true GHD and its severity remains a matter of debate in patients with aromatase deficiency, and the results of provocative tests for GH secretion in male subjects with congenital estrogen deficiency need to be interpreted carefully.

Finally, growth and an increase in height are possible even when the GH-IGF-I axis is impaired if a prolonged time is available for growth due to a delay in bone maturation, and other growth factors different from GH may act as surrogates in promoting growth.

Acknowledgments

We are grateful to Giuseppina Rossi (Department of Medicine, Endocrinology and Metabolism, and Geriatrics, University of Modena and Reggio Emilia) for editing the manuscript.

This work was supported by a grant from the Ministero dell’Università e della Ricerca (ex-40%-2005).

Disclosure Summary: The authors have nothing to disclose.

Abbreviations

 
• BMI

  Body mass index

 
• GHD

  GH deficiency

 
• GHRH-ARG

  GHRH plus arginine