Growth Hormone Receptor Deficiency in Ecuador1

Arlan L. Rosenbloom, Jaime Guevara-Aguirre, Ron G. Rosenfeld and Uta Francke

Department of Pediatrics (A.R.), University of Florida College of Medicine, Children’s Medical Services Center, Gainesville Florida 32608; Instituto Endocrinologia, Metabolismo y Reproduccion (J.G-A.), Quito, Ecuador; Department of Pediatrics (R.G.R.), Oregon Health Sciences University, Portland Oregon 97201-3098; and Department of Genetics and Howard Hughes Medical Institute (U.F.), Stanford University Medical Center, Stanford, California 94305-5323

Address correspondence to: Arlan Rosenbloom, Department of Pediatrics, Children’s Medical Services Center, 1701 SW 16th Avenue, Room 2163, Gainesville, Florida 32608.


    Introduction
 Top
 Introduction
 Background and History
 Epidemiology
 An Unusual Mutation in...
 Clinical Phenotype and Spectrum...
 Biochemical Features of IGF-I...
 Administration of IGF-I to...
 References
 
THE JOY OF understanding is both the closest rival of the joy of service and its most effective partner— Dana W. Atchley.


    Background and History
 Top
 Introduction
 Background and History
 Epidemiology
 An Unusual Mutation in...
 Clinical Phenotype and Spectrum...
 Biochemical Features of IGF-I...
 Administration of IGF-I to...
 References
 
When recombinant human GH became available late in 1985, Genentech, Inc. generously agreed to supply GH as part of their free goods program to the two GH-deficient patients from Ecuador who had been followed for several years at the University of Florida in Gainesville. By mid 1988, the generosity of Genentech, Inc.’s free goods program had tripled, with six Ecuadorian patients needing to come to Gainesville every 6 months, a requirement of the program. The limited means of some of the families and deterioration of the Ecuadorian economy made it difficult or impossible for them to pay for the journey, and they decided to pool resources and obtain donations so that the physician (AR) could come to them. The first trip was in December 1988; in addition to the six patients for whom GH was brought, sisters aged 6 and 8 yr were seen who had height ages of 11 and 15 months (Fig. 1Go). They had been found to have very high random GH levels (>50 ug/L) despite the appearance of severe GH deficiency, their weights were at the 25th percentile for height despite the appearance of obesity, and their bone ages were advanced for height age (2.5 and 3.5 yr).



View larger version (104K):
[in this window]
[in a new window]
 
Figure 1. Sisters 8.2 (left) and 6.1 (right) years of age with the appearance of GH deficiency, but with markedly elevated random GH levels. SDS for height were -8.3 and -9.4 with height ages of 15 and 11 months. A 10.5-yr-old sister (homozygous normal for the E180 splice mutation) had a height SDS of -0.3 as did the (proven) heterozygous father. Their mother’s height SDS was -1.9.

 
Six months later, JG-A and AR saw six adult women and a toddler who came from southern Ecuador to the Institute in Quito for evaluation. Over the next several years, a cohort of over 70 individuals would be identified spanning the entire age range, who were inbred whites without apparent Indian admixture, considered to be descendants of conversos (Spanish Jews who became Catholic during the Inquisition) and all originating from an area 120 kilometers in diameter in the Andes of southern Ecuador (1).

The recognition of this population could not have come at a more propitious time. In 1987, GH binding protein (GHBP) was identified as being identical to the extracellular domain of the GH receptor (GHR) and to be absent in several patients with Laron syndrome. During the year that we were identifying the first 45 patients, 1989, the human GHR gene was characterized and a partial gene deletion found in two Israeli patients with Laron syndrome, and a single nucleotide substitution resulting in a missense mutation was described in four patients in a Mediterranean family (reviewed in Ref. 2). The unique concentration in time and place of the Ecuadorian clinical experience and the finding that a single mutation of the GHR accounted for GHRD in all but one of the patients permitted description of new clinical and biochemical features, the quantification of other features, appreciation of phenotypic variability with the same GHR mutation, and provided opportunity for treatment study. This review is of the findings in the Ecuadorian population. Other reviews are available covering the basic physiology, genetics, and global experience with this condition (1, 2, 3, 4, 5).


    Epidemiology
 Top
 Introduction
 Background and History
 Epidemiology
 An Unusual Mutation in...
 Clinical Phenotype and Spectrum...
 Biochemical Features of IGF-I...
 Administration of IGF-I to...
 References
 
Twenty of the 21 patients initially identified from the province of Loja were female (6). Eventually, a total of 8 males and 27 females with GHRD (living and dead) were identified in sibships from Loja comprising 119 children, with 17% of males and 35% of females affected (P < 0.05). The sex ratio in the Loja sibships was three females to two males (P < 0.05), the discrepancy accountable to the missing male cases. In neighboring El Oro province, there was a normal sex distribution in affected sibships and cases (22 males, 23 females) (2, 5, 7). This difference between two neighboring large groups of patients made genetic analysis particularly important.

Mortality with GHRD has only been studied in the Ecuadorian cohort. Childhood mortality before the age of 7 yr was 19% for affected children (15 of 79) compared with 9.7% (21 of 216) for unaffected siblings (P < 0.05). Causes of death (meningitis, pneumonia, diarrhea) did not differ between affected and unaffected children, suggesting greater vulnerability of the children with GHRD.

Among adults over 50 yr of age followed for 7 yr, two died of heart disease (an uncommon problem in the Andes): a man at 55 yr and a woman at 67 years, (5). Compared with sibling and community controls, adults with GHRD had significantly elevated total cholesterol and low-density lipoprotein (LDL) cholesterol, which was unrelated to adiposity [determined by dual-energy x-ray absorptiometry (DEXA)] or insulin resistance [inferred by fasting insulin-like growth factor binding protein (IGFBP)-1 concentration]. These findings were thought to reflect decreased activity of the hepatic LDL receptor, which is influenced by GH action (8).


    An Unusual Mutation in the GHR Gene Causes GHRD in Ecuador
 Top
 Introduction
 Background and History
 Epidemiology
 An Unusual Mutation in...
 Clinical Phenotype and Spectrum...
 Biochemical Features of IGF-I...
 Administration of IGF-I to...
 References
 
The gene for the GHR was mapped to the proximal short arm of human chromosome 5 (5p13.1-p12) and to a previously unknown conserved syntenic region on mouse chromosome 5 in the laboratory of UF (9). This assignment was consistent with the reported autosomal recessive inheritance of Laron syndrome and structural defects in the GHR gene in affected individuals from Israel (3). When the first cohort of patients with Laron syndrome in Ecuador was reported to include almost exclusively females (6), alternative genetic mechanisms for this skewed gender distribution needed to be considered. A collaboration was established between JG-A, AR, RR, and UF with the initial goal of molecular study to determine whether the Ecuadorean Laron syndrome was due to mutations in the GHR gene on chromosome 5 or whether another gene, possibly located on the X chromosome, could be the culprit. Blood samples from related and unrelated affected individuals and their parents were obtained and lymphoblastoid cell lines established from 30 individuals by transformation with Epstein-Barr virus.

The GHR gene spans 86 kilobase pairs and includes nine exons, numbered 2–10, that encode the receptor and four additional exons in the 5' untranslated region. Exon 2 encodes a secretion signal sequence, exons 3 through 7 the large extracellular GH binding domain, exon 8 the transmembrane domain, and exons 9 and 10 the cytoplasmic domain and 3' untranslated region (3). To look for mutations in the GHR gene, Mary Anne Berg, in the laboratory of UF, first excluded major structural rearrangements by Southern blot analysis with a human GHR complementary DNA probe. Because automated sequencing was not readily available to look for more subtle mutations, Berg applied a new mutation screening technique, denaturing gradient gel electrophoresis; this most sensitive of mutation detection methods was perfected in the laboratory of Richard Myers at University of California–San Francisco (San Francisco, CA), who provided generous advice. Based on the published sequences of the GHR gene exons and flanking introns, primers were designed to amplify each coding exon. PCR products from genomic DNA of obligate heterozygotes contain a mixture of normal and mutant amplimers. Heteroduplex DNA molecules show altered melting behavior under denaturing gradient gel electrophoresis conditions. By this analysis, only GHR exon 6 amplimers produced evidence of a sequence variant (10). On sequencing of cloned PCR products containing exon 6, a single nucleotide substitution was identified that changed the glutamic acid codon GAA in position 180 of the amino acid sequence to a GAG codon, also encoding glutamic acid (10, 11).

Single nucleotide polymorphisms leading to such synonymous changes are rather common in the human genome and are considered normal variants. Therefore, the mechanism by which this mutation could be pathogenic was not immediately apparent. Inspection of the surrounding nucleotide sequence, however, suggested that this A to G substitution could have activated a cryptic splice site. The normal exon 6/intron 6 splice junction is ATG/GTAAGT. The mutation creates a sequence GAG/GTAAAT within the exon 6 coding sequence. Therefore, if splicing occurred after the substituted G nucleotide, the resultant shortened messenger RNA would lack the 24 nucleotides encoding the most C-terminal eight amino acids of exon 6. Expression of the GHR gene in immortalized lymphoblastoid cell lines allowed us to test this hypothesis directly. RT-PCR analysis of cell lines from affected individuals and heterozygotes led to the surprising result that the new splice site created by the mutation was exclusively used, although the original site remained intact (10, 11). It is now known that similar mutations in other genes often lead to transcript heterogeneity that may modulate the clinical phenotype (12). No evidence for transcript heterogeneity was found at the molecular level for the GHR E180 splice mutation, a finding that is consistent with the rather uniform clinical picture.

The eight deleted amino acids are part of the extracellular domain and are located near the receptor dimerization site. They also include a putative glycosylation site. We consider it most likely that the mutant protein is misfolded and degraded intracellularly (13). This hypothesis is consistent with the lack of serum GHBP in most of the Ecuadorean GHRD patients (2). In contrast, a putative stable mutant protein that may lack the ability to transmit signal would be expected to exert a dominant negative effect. No effect on growth, however, was observed in individuals heterozygous for this mutation (14).

Once the E180 splice mutation was identified, rapid assays for its detection were developed by use of allele-specific oligonucleotide hybridization (10, 11) and by an allele-specific restriction enzyme analysis assay with MnlI, for which the mutation generates a new cleavage site (15). To test all available GHRD patients in Ecuador and their relatives for the presence of this mutation, we used dried blood spots collected onto filter papers. The DNA that was PCR-amplified from these blood spots appeared to be stable at room temperature for several months. The results, to date, indicate that all but one of the individuals with GHRD from Ecuador were homozygous for the E180 splice mutation. The single exception, an offspring of a consanguineous marriage, was homozygous for a recurrent R43X nonsense mutation in exon 4. This mutation involves a CpG hotspot and has occurred independently in different populations as indicated by different haplotypes (3, 16). Genotyping potential heterozygotes by PCR from blood spots enabled the systematic comparison of carrier and noncarrier siblings for a variety of clinical parameters (8, 14, 17). The carrier detection results were also used for genetic counseling (Fig. 2Go) (1).



View larger version (12K):
[in this window]
[in a new window]
 
Figure 2. Four-generation pedigree reflecting the consanguinity in the Ecuadorian population with GHRD and the results of testing for the E180 splice mutation. Square, males; circle, females; dot, tested homozygous normal individuals; half black symbol, heterozygotes; solid symbol, homozygotes with GHRD. Deceased children are indicated by the diagonal line through the symbol. The shaded symbol is for a child with the clinical features of GHRD who died before the start of the study. The double horizontal lines indicate consanguineous unions.

 
To determine whether the E180 splice mutation may have been introduced into Ecuador by early Spanish settlers or may have arisen on this continent before or after the conquest, we searched for it in nearly 20 GHRD patients from many geographic regions. Nearly 40 other mutations were identified by us and others (3, 4, 13, 16). Eventually, however, a single Jewish patient from Israel, with bilateral Moroccan parentage, was found to carry the Ecuadorian E180 splice mutation (15).

The gender bias in the initially identified population from Loja was not due to the GHRD, as the sex ratio was normal in patients from El Oro province, and they carry the same mutation. An independent X-linked lethal allele segregating in the inbred Loja population could reduce the number of viable male births. This possibility has not been systematically pursued. The molecular genetic GHR studies of this population revealed an instructive example of a founder effect, where a rare recessive mutation is amplified in a geographically isolated population and leads to an unusually high disease frequency.


    Clinical Phenotype and Spectrum with Homogeneity for the GHR Mutation
 Top
 Introduction
 Background and History
 Epidemiology
 An Unusual Mutation in...
 Clinical Phenotype and Spectrum...
 Biochemical Features of IGF-I...
 Administration of IGF-I to...
 References
 
The clinical features that have been described in the Ecuadorian patient group with identical mutations in the GHR and in the two other large cohorts that are genetically heterogeneous, the European Pharmacia and Upjohn treatment study group (18) and the Israeli patients (19), are presented in Table 1Go. Features that have been more extensively described or quantitated and those originally described in the Ecuadorian population are noted. The clinical phenotype of GHRD appears identical to that of severe GH deficiency, including those features originally described from Ecuador (blue scleras, limited elbow extensibility) (4).


View this table:
[in this window]
[in a new window]
 
Table 1. Clinical features of GHRD

 
Growth

Despite a characteristic clinical appearance at birth indicating an effect of GHRD in utero (patient 1 in Fig. 3Go), affected newborns are of normal weight and most are normal in length, similar to what is seen with severe GH deficiency due to GH gene deletion (20). There is rapid postnatal decline in SD score), and growth velocity is approximately half normal. In both sexes there is absence of the pubertal growth spurt, with persistent growth beyond the normal time of adolescence and substantial delay in the onset of puberty in 50% (20). Adult stature varies from -5.3 to -12 SDS, using United States standards, a range of 95–124 cm for women and 106–141 cm for men. This wide variation, despite homogeneity for the GHR mutation, was seen within families, as well as within the entire population, and is comparable to the range of SDS for height in a genetically diverse group of patients (18).



View larger version (110K):
[in this window]
[in a new window]
 
Figure 3. Front and side views of 14 Ecuadorian patients with GHRD, ranging in age from 1 month to 66 yr, demonstrating craniofacial abnormalities and phenotypic heterogeneity despite genetic homogeneity for the mutation in the GHR. From left to right: 1) 1-month-old girl of normal length and weight for family; 2) her 26-month-old brother (height, -8.2 SD); 3) 4.8-yr-old patient (height, -7.4 SD); 4) 6.5-yr-old patient (height, -9.4 SD); 5) 8.2-yr-old patient (height, -8.3 SD); 6) her 10-yr-old sister (height, -7.5 SD); 7) 14.8-yr-old patient (height, -8.3 SD; bone age 7 yr); 8) 17-yr-old woman who had menarche at 15 yr of age and stopped growing at 16 yr of age (height, -7.8 SD); 9) 19-yr-old woman [tallest female (height, 123.9 cm, -6.5 SD)] with menarche at 15 yr of age; 10) 21-yr-old man with normal timing of puberty and cessation of growth at 16 yr of age (height, 126.6 cm; -7.6 SD); 11) 28.5-yr-old man (height, -9.2 SD; 12) 40-yr-old woman (height, -8.6 SD) with advanced aging of skin; 13) 52-yr-old woman (height, 94.8 cm; -11.5 SD); 14) 66-yr-old woman (height, -9.5 SD). (Photography and production by A. L. Rosenbloom. Reproduced with permission from Rosenfeld RG, Rosenbloom AL, Guevara-Aguirre J. Growth hormone (GH) insensitivity due to primary GH receptor deficiency. Endocr Rev. 15(3 ):369–390, 1994 211 The Endocrine Society.)

 
In an effort to determine factors that might provide insight into the variability in the statural affects of this mutation of the GHR, we determined that height SDS of affected adults correlated significantly with their serum concentrations of IGF-I (r = 0.60; P < 0.001) and IGFBP-3 (r = 0.61, P < 0.001), even though the levels of IGF-I and IGFBP-3 were markedly below normal. Children’s levels of IGF-I were too low for analysis, but they also had significant correlation between IGFBP-3 serum concentrations and statural deviation (r = 0.58; P < 0.05) (7, 20).

As noted above, heterozygosity for the E180 splice mutation did not affect stature among 53 relatives of probands compared with 37 homozygous normal relatives. Furthermore, if heterozygosity were to influence stature, the heights of heterozygous parents of probands would be expected to correlate with those of probands and of carriers who are their offspring and not with heights of their homozygous normal children. In fact, parental height SDS did not correlate with height SDS of affected offspring, but correlated strongly with those of both heterozygous and homozygous normal offspring (14).

Craniofacial characteristics

As with statural effects, there is wide variability of craniofacial features among affected individuals, despite identity of the GHR mutation (Fig. 3Go). Affected children are often recognized by knowledgeable family members at birth because of frontal prominence, depressed nasal bridge, sparse hair, small hands or feet, and hypoplastic fingernails (patient 1 in Fig. 3Go). Because of the prominent forehead, small face, prominence of the scalp veins with thin skin and hypotrichosis, and setting sun sign, patients may be subjected to evaluation for hydrocephalus (patient 2 in Fig. 3Go). Decreased vertical dimension of the face is demonstrable by computer analysis of the relationships between facial landmarks and is present in all patients, compared with their relatives, including those with reasonably normal facial appearance (patient 10 in Fig. 3Go) (21). The occasional occurrence of unilateral ptosis with facial asymmetry may result from positional deformity due to decrease in muscular activity in utero (5).

Musculoskeletal and body composition

Delayed walking, despite normal time of speech onset, is considered to be the result of hypomuscularity, which is apparent on roentgenograms. Osteopenia is also demonstrated on radiographs and by DEXA (5, 6). In a detailed study of 11 adult probands matched to 11 unaffected siblings, areal bone mineral density was reduced in the patients with GHRD, but estimated volumetric bone density (bone mineral apparent density) was normal. Iliac crest dynamic histomorphometric studies from bone biopsies confirm the interpretation of preserved bone mineral apparent density, with the only difference being a reduced trabecular connectivity in GHRD (22). It remains uncertain whether or not these patients have true osteoporosis; fractures do not seem to be more frequent, including in the elderly.

Limited elbow extensibility seen in 85% of Ecuadorian patients with GHRD over 5 yr of age and with increasing severity with age has been described in at least one other individual with GHRD and with familial anterior hypopituitarism due to PROP1 deficiency (4). The reason for this acquired problem with severe IGF-I deficiency is unknown.

Children are underweight to normal weight for height despite the appearance of obesity (Fig. 1Go), whereas all females and most males after puberty are substantially overweight with marked decrease in lean to fat ratios by DEXA (8, 20).

Intellectual function

We noted exceptional school performance with seriously constrained social and economic independence of most women, but only a few of the men in the Ecuadorian patient population (23). The only controlled evaluation of intellectual ability of patients with GHRD was carried out in Ecuador with intelligence tests that had been validated in cross-cultural research, designed to minimize the effects of physical size, motor coordination, and cultural background. The intellectual ability of the patients with GHRD did not differ from that of their relatives or of community controls, and there was no effect of heterozygosity for the E180 splice mutation (17). The studies indicate that neither fetal nor postnatal brain growth or intellectual development is dependent on GH stimulated IGF-I production.


    Biochemical Features of IGF-I Deficiency Resulting from GHR Failure (Table 2Go)
 Top
 Introduction
 Background and History
 Epidemiology
 An Unusual Mutation in...
 Clinical Phenotype and Spectrum...
 Biochemical Features of IGF-I...
 Administration of IGF-I to...
 References
 
GH

Ecuadorian children with GHRD have elevated random GH levels that may be as high as 200 ug/L, and their response to stimulation is increased, with paradoxical elevations after oral or iv glucose, as in acromegaly (2, 6). Diurnal variation is normal with normal suppression by exogenous recombinant human IGF-I (24). Thus, the normal sensitivity of the GH secretion is preserved, despite elevated levels and lack of feedback suppression from IGF-I.


View this table:
[in this window]
[in a new window]
 
Table 2. The GH/IGF/IGFBP axis with GHRD in Ecuadorean patients

 
Postpubertal patients may have normal or elevated basal levels of GH but invariably demonstrate hyper-responsiveness to stimulation, which is all the more impressive with their obesity, which suppresses GH responses in normal individuals. In the Ecuadorian population, mean basal GH level in adults was significantly lower than that in children (11 ± 11 ng/mL vs. 32 ± 22 ng/mL, P < 0.0001). This is thought to be related to the greater, though still markedly abnormal, IGF-I levels in the adults, resulting in some feedback inhibition of GH secretion (2).

GHBP

The ligand-mediated immunofunction assay used to measure GHBP serum levels since 1990, uses an anti-GH monoclonal antibody to determine the amount of GH bound to GHBP. This largely functional assay should not detect structurally abnormal, but expressed, GHBP (2).

Despite in vitro evidence for failure of production of normally spliced receptor, four children and four adults of 49 Ecuadorian GHRD patients initially studied had serum GHBP levels higher than 40% of the sex-specific lower limit for controls, and one adult male had a level in the lower portion of the normal adult male range. The presence or amount of GHBP measured did not relate to stature. There were no age-dependent changes, indicating that the difference in IGF values between children and adults was not related to the GHBP levels (7).

IGF

Serum IGF-I concentrations are profoundly reduced, with prepubertal patients having levels consistently less than 10 ng/mL and adults less than 100 ng/mL (6, 7). Thus, prepubertal IGF-I levels in GHRD are totally dependent on GH secretion and action, with the elevations in sexually mature individuals indicating that sex steroids may be able to stimulate IGF-I production independently of GH. Patients with GHRH receptor deficiency, however, have identical extremely low IGF-I levels as children and as sexually mature adults, which argues against this hypothesis (25).

Serum IGF-II concentrations are also significantly reduced, but not as dramatically as those of IGF-I. The reduction in IGF-II levels likely reflects the marked decrease in serum levels of IGFBP-3 and ALS (acid labile subunit).

IGFBP

The principal binding protein for circulating IGF, IGFBP-3, is considered to serve as a reservoir and delivery mechanism for IGF to tissues. In both prepubertal and adult patients with GHRD, serum IGFBP-3 concentrations were found to be less than 1 ug/mL, approximately 20% of normal adult concentrations (7). ALS, which combines with IGFBP-3 and IGF to form the circulating ternary complex and is also GH dependent, is profoundly reduced, averaging less than 1 ug/mL, compared to normal levels of 16–20 ug/mL (26). Modest reductions were observed in serum concentrations of IGFBP-4 and IGFBP-5; because IGFBP-5 is able to complex with ALS, this reduction may reflect ALS deficiency rather than specific GH dependency for this IGFBP (26). IGFBP-I is elevated in GHD and GHRD; in GHRD it is the most abundant IGFBP and is strongly inversely related to insulinemia. IGFBP-2 is present at a mean 300 percent of control concentrations in children with GHRD and 175 percent of control in affected adults (7).


    Administration of IGF-I to Ecuadorian Subjects with GHRD
 Top
 Introduction
 Background and History
 Epidemiology
 An Unusual Mutation in...
 Clinical Phenotype and Spectrum...
 Biochemical Features of IGF-I...
 Administration of IGF-I to...
 References
 
Short-term administration

Before undertaking treatment of the large number of children in the Ecuadorian population with GHRD, Vaccarello et al. (24) studied six Ecuadorian adults with GHRD on the Clinical Research Center at the University of Florida. They administered IGF-I in a sc dose of 40 ug/kg body weight every 12 h over 7 days. Mean integrated 24-h GH levels were suppressed, as were the number of peaks, the area under the curve, and clonidine-stimulated GH release. The mean peak serum IGF-I levels were 253 ± 11 ng/mL, reached between 2–6 h after injection, and mean trough levels were 137 ± 8 ng/mL before the next injection, values not significantly different from those of normal control Ecuadorian adults. There were no significant changes in the half-life or metabolic clearance of IGF-I between days 1 and 7, but the distribution volume did increase over this time. Although IGFBP-3 levels did not increase, elevated baseline IGFBP-2 levels (153% of control) increased 45% (P < 0.01). In these short-term studies, there was no increased risk of hypoglycemia despite low levels of IGFBP-3. There remained, however, concern whether the low IGFBP-3 levels would result in more rapid clearance of IGF-I and blunting of the growth response.

Additional studies were carried out on composition and distribution of IGF and IGFBP in patients from this study. It was found that the two forms of IGFBP-3 associated with IGF and ALS, which are able to form the ternary 150-kDa complex are abnormally distributed in GHRD patients. This distribution was unchanged by IGF-I treatment, which again demonstrated failure to increase IGFBP-3, as well as ALS, levels (26, 27).

Long-term treatment

In addition to concern about replacement therapy with IGF-I related to the failure of short-term therapy to increase IGFBP-3 levels, there was a question whether catch-up growth in children with GHRD would be as substantial as occurs with GH replacement therapy in patients with GHD, in the absence of a direct effect of GH on bone. This direct effect is considered to be the differentiation of prechondrocytes into early chondrocytes, which then secrete IGF-I, that in turn stimulates clonal expansion and maturation of the chondrocytes, or growth, thought to account for 20% of GH-induced growth (4).

The first IGF-I treatment report from the large Ecuadorian cohort was of growth and body composition changes in two adolescent patients treated with a combination of IGF-I (120 mcg/kg bid) and LRH analog to forestall puberty. A girl age 18 yr and boy age 17.2 yr, with bone ages of 13.5 and 13 yr, experienced an approximate tripling of growth velocity, increased bone mineral density, and maturation of facial features on IGF-I for 1 year. There was initial hair loss, followed by recovery of denser and curly hair with filling of the fronto-temporal baldness, the appearance of axillary sweating, loss of deciduous teeth, and appearance of permanent dentition. The boy had coarsening of his facial features indicative of supraphysiologic tissue levels of IGF-I (Fig. 5Go). The fading of premature facial wrinkles was noted in one patient. The other patient had submaxillary gland swelling. Serum IGF-I levels were seen to increase into the normal range during the first 2–8 h after IGF-I injection (28). Studies were done at doses of 40, 80, and 120 ug/kg body weight with pharmacokinetic profiles suggesting a plateau effect between 80 and 120 ug/kg per dose. It was considered that the carrying capacity of the IGFBP was saturated at this level. Mean serum IGF-II levels decreased concurrently with the increase in IGF-I, and serum IGFBP-3 levels did not respond to IGF-I treatment. There was no apparent change in the half-life of IGF-I during the treatment period, indicating no alteration of IGF-I pharmacokinetics induced by prolonged treatment (29).



View larger version (86K):
[in this window]
[in a new window]
 
Figure 5. Face and hair changes in 17.7-yr-old patient (bone age 13 yr) with GHRD during 6 months of treatment with IGF-I, 120 ug/kg body weight twice daily and depot GnRH agonist begun at age 16.5 yr. During this period, he grew 4 cm and increased his height SDS from -9.1 to -8.5. (Reproduced by permission from Rosenbloom AL. IGF-I treatment of growth hormone insensitivity. In: Rosenfeld RG, Roberts CT, eds. The IGF system: molecular biology, physiology, and clinical applications. Totowa, NJ: Humana Press, Inc.; 739–769, 1999.)

 
Tachycardia was noted in these adolescent patients and further studied in 16 prepubertal Ecuadorian patients during induction of treatment at progressive dosages of 40, 80, and 120 mcg/kg twice daily. There was a direct dose-related increase to approximately 125% of baseline rate at the highest dosage. This was unrelated to glucose or electrolyte changes that were not significant (30).

Seventeen prepubertal Ecuadorian patients were entered into a randomized double-blind, placebo-controlled trial of IGF-I at 120 ug/kg sc twice daily for 6 months, following which all subjects received IGF-I. Such a study was considered necessary because of the observation of spontaneous periods of normal growth in these youngsters, the suggestion that nutritional changes that might accompany intervention would be an independent variable, and the need to control for side effects, particularly hypoglycemia, which occur in the untreated state. The nine placebo-treated patients had a modest, but not significant, increase in height velocity from 2.8 ± 0.3 to 4.4 ± 0.7 cm/yr, entirely attributable to three individuals with 6 month velocities of 6.6 to 8 cm/yr. Although this response was attributed to improved nutritional status, there was no accompanying increase in IGFBP-3, as was noted with nutrition-induced catch-up growth by Crosnier et al. (31) in their GHRD patient with anorexia. For those receiving IGF-I, the height velocity increased from 2.9 ± 0.6 to 8.8 ± 0.6 cm/yr and all 16 patients had accelerated velocities during the second 6 month period when all were receiving IGF-I. Again, no changes or differences in IGFBP-3 were noted. There was no difference in the rate of hypoglycemia events, nausea or vomiting, headaches, or pain at the injection site between the placebo and IGF-I-treated groups. Six of the seven IGF-I-treated patients experienced hair loss (32).

Two-year treatment results in the Ecuadorian group have been reported, comparing the 120 ug/kg bid dosage to 80 ug/kg bid and also comparing responses to those of GH-treated GHD in the Ecuadorian population. There were no baseline differences between the low- and high-dose groups for growth velocity, bone age, SDS for height, or mean percent body weight for height. No differences were seen between the two dosage groups in growth velocity or changes in height SDS, height age, or bone age. A group of six subjects receiving the higher dose followed for a 3rd yr continued to maintain 2nd yr growth velocities. The annual changes in height age in both the 1st and the 2nd yr of treatment correlated with IGF-I trough levels, which tended to be in the low normal range despite a failure of serum IGFBP-3 levels to increase. The comparable growth responses to the two dosage levels and the similar IGF-I trough levels were thought to indicate a plateau effect at or below 80 mcg/kg body weight twice daily (33).

When all 22 of these patients were compared to 11 GH-treated GHD subjects, the GHD group was found to have a greater change in SDS for height and height age but did not differ in the ratio of height age to bone age changes over the 2-yr period. There was, however, a greater change in mean percent body weight for height in the GHRD group treated with IGF-I, thought to indicate comparable effects on lean body mass of the two treatment programs, without the lipolytic effects of GH in the IGF-I-treated GHRD subjects. The difference in growth response between IGF-I-treated GHRD and GH-treated GHD was consistent with the hypothesis that 20% or more of GH-influenced growth is due to the direct effects of GH on bone (20). Nonetheless, the comparable ratios of height age change to bone age change suggested similar long-term effects for replacement therapy in these two conditions (33).

Unlike the results in the short-term adult studies, long-term administration of IGF-I in children was associated with elevated but not normal trough levels (12 h after previous injection) (26). This reduction in half-life of IGF-I compared to normals is almost certainly the result of deficient IGFBP-3 and ALS. In addition to a lack of change in serum concentrations of IGFBP-3 and ALS, there was no significant change in IGFBP-4 or IGFBP-5 concentrations (26).



View larger version (23K):
[in this window]
[in a new window]
 
Figure 4. Length SDS of nine female Ecuadorian patients (open symbols, solid lines) and two brothers from Russia (solid symbols, dashed lines) with known birth lengths, followed during the first 2–3 yr of life. (Reprinted from Rosenbloom AL, Guevara-Aguirre J, Rosenfeld RG, Pollock BH. Growth in growth hormone insensitivity. Trends Endocrinol Metab. 5:296–303; copyright 1994; with kind permisssion from Elsevier Science Ltd. The Boulevard, Langford Lane, Kidlington OX5 1GB, UK.)

 

    Footnotes
 
1 Supported by grants from the NIH (DK-45830, HD-28703, DK-36054, DK-08516, DK-28229, HG-00298, and DK 45266), the General Clinical Research Centers (University of Florida, MO1 RR-00082), the Howard Hughes Medical Institute, The March of Dimes-Birth Defects Foundation, an orphan drug grant from the United States Food and Drug Administration, and Pharmacia & Upjohn, Inc. Back

Received September 13, 1999.

Accepted October 14, 1999.


    References
 Top
 Introduction
 Background and History
 Epidemiology
 An Unusual Mutation in...
 Clinical Phenotype and Spectrum...
 Biochemical Features of IGF-I...
 Administration of IGF-I to...
 References
 

  1. Rosenbloom AL, Guevara-Aguirre J. 1994 Bienvenidos a mi tierra de soledad: from poetry to molecular biology in Southern Ecuador. J Clin Endocrinol Metab. 79:695–702.[Medline]
  2. Rosenfeld RG, Rosenbloom AL, Guevara-Aguirre J. 1994 Growth hormone (GH) insensitivity due to primary GH receptor deficiency. Endocr Rev. 15:369–390.[Abstract]
  3. Rosenbloom AL, Guevara-Aguirre J. 1998 Lessons from the genetics of Laron syndrome. Trends Endocrinol Metab. 9:276–283.[CrossRef]
  4. Rosenbloom AL. 1999 Growth hormone insensitivity: physiologic, and genetic basis, phenotype, and treatment. J Pediatr. 135:280–289.[Medline]
  5. Rosenbloom AL, Rosenfeld RG, Guevara-Aguirre J. 1997 Growth hormone insensitivity. Pediatr Clin North Am. 44:423–442.[Medline]
  6. Rosenbloom AL, Guevara-Aguirre J, Rosenfeld RG, Fielder PJ. 1990 The little women of Loja: growth hormone receptor-deficiency in an inbred population of southern Ecuador. N Engl J Med. 323:1367–1374.[Abstract]
  7. Guevara-Aguirre J, Rosenbloom AL, Fielder PJ, Diamond FB Jr, Rosenfeld RG. 1993 Growth hormone receptor deficiency in Ecuador: clinical and biochemical phenotype in two populations. J Clin Endocrinol Metab. 76:417–423.[Abstract]
  8. Rosenbloom AL, Martinez V, Kranzler J, Bachrach LK, Rosenfeld RG, Guevara-Aguirre J. 1999 Natural history of growth hormone receptor deficiency. Acta Paediatr Suppl. 428:153–156.
  9. Barton DE, Foellmer BE, Wood WI, Francke U. 1989 Chromosome mapping of the growth hormone receptor gene in man and mouse. Cytogenet Cell Genet. 50:137–141.[Medline]
  10. Berg MA, Guevara-Aguirre JG, Rosenbloom AL, Rosenfeld RG, Francke U. 1992 Mutation creating a new donor splice site in the growth hormone receptor genes of 37 Ecuadorean patients with Laron syndrome. Hum Mutat. 1:24–34.[Medline]
  11. Berg MA, Guevara-Aguirre J, Rosenbloom AL, Rosenfeld RG, Francke U. 1993 Laron syndrome in Ecuador: point mutation in the growth hormone receptor gene causes deletion of 8 amino acids from the extracellular domain. Pediatr Adolesc Endocrinol. 24:140–152.
  12. Lemahieu V, Gastier JM, Francke U. 1999 Novel mutations in the Wiskott-Aldrich syndrome protein gene and their effects on transcriptional, translational and clinical phenotypes. Hum Mutat. 14:54–66.[CrossRef][Medline]
  13. Francke U, Berg MA. 1993 Genetic heterogeneity in Laron syndrome. Acta Paediatr Suppl. 391:3–7.
  14. Rosenbloom AL, Guevara-Aguirre J, Berg, MA, Francke U. 1998 Stature in Ecuadorians heterozygous for the growth hormone receptor gene E180 splice mutation does not differ from that of homozygous normal relatives. J Clin Endocrinol Metab. 83:2373–2375.[Abstract/Free Full Text]
  15. Berg MA, Peoples R, Pérez-Jurado L, et al. 1994 Receptor mutations and haplotypes in growth hormone receptor deficiency: a global survey and identification of the Ecuadorean E180 splice mutation in an Oriental Jewish patient. Acta Paediatr Suppl. 399:112–114.[Medline]
  16. Berg MA, Argente J, Chernausek S, et al. 1993 Diverse growth hormone receptor gene mutations in Laron syndrome. Am J Hum Genet. 52:998–1005.[Medline]
  17. Kranzler JH, Rosenbloom AL, Martinez V, Guevara-Aguirre J. 1998 Normal intelligence with severe insulin-like growth factor I deficiency due to growth hormone receptor deficiency: a controlled study in a genetically homogeneous population. J Clin Endocrinol Metab. 83:1953–1958.[Abstract/Free Full Text]
  18. Woods KA, Dastot F, Preece MA, et al. 1977 Extensive personal experience. Phenotype:genotype relationships in growth hormone insensitivity syndrome. J Clin Endocrinol Metab. 82:3529–3535.[Abstract/Free Full Text]
  19. Laron Z, Pertzelan A, Karp M, Keret R, Eshet R, Silbergeld A. 1993 Laron syndrome—a unique model of IGF-I deficiency. Pediatr Adolesc Endocrinol. 24:140–152.
  20. Rosenbloom AL, Guevara-Aguirre J, Rosenfeld RG, Pollock BH. 1994 Growth in growth hormone insensitivity. Trends Endocrinol Metab. 5:296–303.
  21. Schaefer GB, Rosenbloom AL, Guevara-Aguirre J, et al. 1994 Facial morphometry of Ecuadorian patients with growth hormone receptor deficiency/Laron syndrome. J Med Genet. 31:635–639.[Abstract]
  22. Bachrach LB, Marcus R, Ott SM, et al. 1998 Bone mineral, histomorphometry, and body composition in adults with growth hormone receptor deficiency. J Bone Miner Res. 13:415–421.[Medline]
  23. Guevara-Aguirre J, Rosenbloom AL. 1993 Psychosocial adaptation of Ecuadorian patients with growth hormone receptor deficiency/Laron syndrome. Pediatr Adolesc Endocrinol. 24:61–64.
  24. Vaccarello MA, Diamond FB Jr, Guevara-Aguirre J, et al. 1993 Hormonal and metabolic effects and pharmacokinetics of recombinant human insulin-like growth factor-I in growth hormone receptor deficiency/Laron syndrome. J Clin Endocrinol Metab. 77:273–280.[Abstract]
  25. Maheshwari HG, Silverman BL, Dupuis J, Baumann G. 1998 Phenotypic and genetic analysis of a syndrome caused by an inactivating mutation in the growth hormone-releasing hormone receptor: dwarfism of Sindh. J Clin Endocrinol Metab. 83:4065–4074.[Abstract/Free Full Text]
  26. Burren CP, Wanek D, Mohan S, Guevara-Aguirre J, Rosenfeld RG. 1999 Serum levels of insulin like growth factor binding proteins in Ecuadorean children with growth hormone insensitivity. Acta Paediatr Suppl. 428:185–191.
  27. Gargosky SE, Wilson KF, Fielder PJ, et al. 1993 The composition and distribution of insulin-like growth factors (IGF) and IGF binding proteins (IGFBPs) in the serum of growth hormone receptor-deficient patients: effects of IGF-I therapy on IGFBP-3. J Clin Endocrinol Metab. 77:1683–1689.[Abstract]
  28. Martinez V, Vasconez O, Martinez A, et al. 1994 Body changes in adolescent patients with growth hormone receptor deficiency receiving recombinant human insulin-like growth factor I and luteinizing hormone-releasing hormone analog: preliminary results. Acta Paediatr Suppl. 399:133–136.[Medline]
  29. Wilson KF, Fielder PJ, Guevara-Aguirre J, et al. 1995 Long-term effects of insulin-like growth factor (IGF)-I treatment of serum IGFs and IGF binding proteins in adolescent patients with growth hormone receptor deficiency. Clin Endocrinol. 42:399–407.[Medline]
  30. Vasconez O, Martinez V, Martinez AL, et al. 1994 Heart rate increase in patients with growth hormone receptor deficiency treated with insulin-like growth factor I. Acta Paediatr Suppl. 399:137–139.[Medline]
  31. Crosnier H, Gourmelen M, Prëvot C, Rappaport R. 1993 Effects of nutrient intake on growth, insulin-like growth factors, and their binding proteins in a Laron-type dwarf. J Clin Endocrinol Metab. 76:248–250.[Abstract]
  32. Guevara-Aguirre J, Vasconez O, Martinez V, et al. 1995 A randomized double-blind, placebo-controlled trial of safety and efficacy of recombinant insulin-like growth factor-I in children with growth hormone receptor deficiency. J Clin Endocrinol Metab. 80:1393–1398.[Abstract]
  33. Guevara-Aguirre J, Rosenbloom AL, Vasconez O, et al. 1997 Two-year treatment of growth hormone receptor deficiency (GHRD) with recombinant insulin-like growth factor-I in 22 children: comparison of two dosage levels and to GH treated GH deficiency. J Clin Endocrinol Metab. 82:629–633.[Abstract/Free Full Text]