Isolated Growth Hormone Deficiency and Genetic Forms of Short Stature due to GH Deficiency

Monogenic isolated growth hormone deficiency (GH1, GHRHR, GHSR and X-linked forms), with genetic panhypopituitarism and growth hormone insensitivity as the main differential diagnoses.

Literature and regulatory information updated: 10 October 2026.

0. Initial Reference Table

1. Introduction

1.1. Historical overview

The molecular history of isolated growth hormone (GH, growth hormone) deficiency begins with a clinical observation. In 1970, Ruth Illig described three Swiss children with unusually severe growth failure who responded well to exogenous pituitary-derived GH and then stopped growing when they developed antibodies against the administered hormone [4]. In 1981, Phillips, Hjelle, Seeburg and Zachmann studied DNA from four individuals with the form then called IGHD “type A” using restriction enzymes and a human GH complementary DNA (cDNA) probe, and demonstrated homozygosity for a deletion of at least 7.5 kb removing the normal GH gene (GH-N, now GH1) while sparing the placental variant gene (GH-V, GH2), with autosomal recessive segregation consistent with the phenotype [28]. Complete absence of hormone exposure during development explained the immune response against therapeutic GH [4, 5].

The X-linked form was described in 1980 by Fleisher and colleagues in a family with two brothers and two maternal uncles who had hypogammaglobulinaemia, recurrent sinopulmonary infections, short stature, delayed bone age and an impaired GH response to insulin and arginine or levodopa; three of the four had no circulating B lymphocytes, and the pattern was X-linked recessive [21]. In 1994, Duriez and colleagues identified, in a sporadic patient with X-linked agammaglobulinaemia and IGHD, an intronic BTK variant (1882+5G>A, the authors' nomenclature) causing exon skipping in the tyrosine kinase domain and suggested that the same gene could explain the combined phenotype, although the report was based on a single case [29].

The growth hormone-releasing hormone (GHRH, growth hormone-releasing hormone) receptor was cloned in 1992, and the little mouse, with a missense variant abolishing GHRH binding, provided an animal model of GH deficiency of hypothalamic–pituitary origin [4]. In 1996, Wajnrajch and colleagues described the first human nonsense GHRHR variant in an Indian Muslim family and called it analogous to the little mouse [4, 30]. In 1998, following a newspaper report, Maheshwari, Baumann and colleagues organised an expedition to two villages in Sindh province, Pakistan, and characterised 18 affected members of a consanguineous kindred with what they called “Sindh dwarfism”: the variant was identical to that in Wajnrajch's families [2]. In Brazil, the Itabaianinha kindred in Sergipe state, studied since 1994, revealed more than one hundred individuals over eight generations with a homozygous splice variant in the same gene, most never treated [3, 31].

Other milestones complete the picture. In 1978, Kowarski described children with short stature, normal or slightly elevated immunoreactive GH and low IGF-I who grew with exogenous GH, termed “bioinactive GH” [4]. In 2006, Pantel and colleagues identified a missense variant in the growth hormone secretagogue receptor (GHSR, the ghrelin receptor) segregating with short stature in two families and selectively abolishing receptor constitutive activity while preserving the response to ghrelin [32]. Among the differential diagnoses, Laron syndrome was described in 1966 in two siblings from consanguineous Yemenite Jewish families, and its cause (partial GHR deletion) identified in 1989 [33]; septo-optic dysplasia was delineated by de Morsier, who associated absence of the septum pellucidum with optic nerve hypoplasia, and Hoyt and colleagues expanded the spectrum to hypopituitarism in 1970 [34].

1.2. Nomenclature and classification

The classical classification of Mendelian IGHD, consolidated in reviews by Mullis and Alatzoglou, recognises at least four disorders: types IA and IB, autosomal recessive; type II, autosomal dominant; and type III, X-linked [4, 35]. Type IA corresponds to complete absence of GH, almost always due to homozygous GH1 deletion or variants preventing production of any mature hormone. Type IB denotes the recessive form with low but detectable GH and immune tolerance to exogenous GH; it was historically used both for GH1 splice variants and biallelic GHRHR variants [4]. Type II results predominantly from variants causing GH1 exon 3 skipping and producing a 17.5-kDa isoform with a dominant-negative effect [4]. X-linked type III groups the association with agammaglobulinaemia (OMIM 307200) and SOX3-related forms (OMIM 300123) [5, 13].

Current nomenclature differs between databases, and clinicians encounter both conventions in reports. OMIM created entry 618157 (“type IV”) for GHRHR-related IGHD, a designation adopted by ClinVar (“isolated growth hormone deficiency, type 4”) and recent publications, whereas classical literature and some laboratories retain “type IB” for these families [5, 10, 36]. Orphanet's current interface separates type IV (ORPHA:684247, GHRHR) from type IB (ORPHA:231671, GH1 only) [7, 11]. Orphanet also associates RNPC3 with type IA (OMIM 618160) and includes OMIM entry 613038 under type II, referring to POU1F1, with a “gain of function” annotation, reflecting the existence of dominant POU1F1 variants with an isolated IGHD phenotype rather than a reclassification of type II [6, 9, 12]. For the form associated with agammaglobulinaemia, Orphanet lists BTK and ELF4 [14]. This article uses “type IV (GHRHR)” with the historical equivalence “type IB” made explicit.

Kowarski syndrome (OMIM 262650) is a qualitative form: GH circulates in normal amounts but has low biological activity, distinguishing it from quantitative deficiency and explaining why measurement of GH alone does not detect it [1, 5]. Short stature due to GHSR deficiency (OMIM 615925; ORPHA:314811) and CAGSSS syndrome (cataracts, GH deficiency, sensory neuropathy, sensorineural deafness and skeletal dysplasia; IARS2, ORPHA:436174), a mitochondrial disease, complete the rare entities grouped by Orphanet under GH deficiency [8, 15].

1.3. Definitions and scope of the article

IGHD is defined as impaired GH secretion without deficiency of other anterior pituitary hormones at diagnosis, with growth failure, delayed bone age and short stature of variable severity and age at onset, and a subtype-dependent response to recombinant GH [1]. Combined pituitary hormone deficiency (CPHD, combined pituitary hormone deficiency) is defined as GH deficiency associated with at least one other deficiency (TSH, thyroid-stimulating hormone; LH/FSH, gonadotropins; PRL, prolactin; ACTH, corticotropin) [19]. The boundary is dynamic: IGHD type II and forms associated with HESX1, GLI2, SOX3 or POU1F1 may begin as isolated deficiencies and progress to CPHD, and isolated hypogonadotropic hypogonadism may overlap with CPHD [9, 19]. Growth hormone insensitivity (GHI, growth hormone insensitivity) is the biochemical mirror image of deficiency: low IGF-I with normal or elevated GH and no IGF-I response to exogenous GH [33].

This article addresses monogenic IGHD and, in dedicated differential diagnosis subsections, genetic panhypopituitarism and primary GH insensitivity. SHOX haploinsufficiency and Léri–Weill dyschondrosteosis, Silver–Russell syndrome and X-linked agammaglobulinaemia, regarding the immunodeficiency, appear here only as differential diagnoses. Acquired GH deficiency (tumours, surgery, irradiation, head injury) is mentioned only when it affects the interpretation of tests or guidelines.

2. Molecular Genetics and Pathophysiology

2.1. The GHRH–GH–IGF-I axis

GH secretion by anterior pituitary somatotrophs is stimulated by hypothalamic GHRH, acting at GHRHR, a G protein-coupled receptor expressed almost exclusively in somatotrophs; adenylate cyclase activation promotes somatotroph proliferation and GH transcription and secretion [5]. Ghrelin potentiates secretion through GHSR, and this receptor has constitutive (ligand-independent) activity whose loss has been associated with familial short stature [4, 32]. Circulating GH stimulates hepatic and peripheral IGF-I production, promotes myoblast proliferation and differentiation, amino acid uptake and protein synthesis [5]. IGF-I circulates predominantly in a ternary complex with IGFBP-3 and the acid-labile subunit (ALS, encoded by IGFALS) and acts at the IGF1R receptor; defects downstream of GH (GHR receptor, STAT5B signalling, IGF-I synthesis, ALS-mediated transport, IGF1R receptor) constitute the GH insensitivity addressed in section 6 [33].

Pituitary ontogeny depends on a cascade of transcription factors. HESX1, a paired-like repressor, acts in the early forebrain and subsequently Rathke's pouch and antagonises PROP1; PROP1 precedes and is required for POU1F1 (PIT-1) expression, which specifies somatotrophs, lactotrophs and thyrotrophs and activates the GH and PRL genes [5, 19]. Developmental genes explain why some apparently isolated forms progress to multiple deficiencies and why they are associated with midline, ocular or posterior pituitary defects [37].

2.2. GH1 and the growth hormone cluster

GH1 is located at 17q23.3 (OMIM 139250; MANE transcript NM_000515.5) [5, 36]. It forms part of a cluster of around 65 kb with five highly homologous genes (92–98%), arranged from 5′ to 3′: GH1, CSHL1 (pseudogene), CSH1, GH2 and CSH2 [4]. Only GH1 is expressed in the pituitary; the others are placental, and GH2 encodes variant GH (GH-V), replacing pituitary GH in the maternal circulation during the second half of pregnancy [4]. GH1 has five exons and four introns; the 217-amino-acid prehormone gives rise to mature GH of 191 amino acids and 22 kDa (around 75% of circulating pituitary GH), and a cryptic acceptor site in exon 3 generates the 20-kDa isoform (5–10%), also bioactive [4]. Cluster homology favours unequal recombination and explains the frequency of deletions; it also has technical consequences for diagnosis (section 5) [4].

Type IA results from complete absence of GH. Homozygous deletions are heterogeneous: the 6.7-kb deletion accounts for 70–80% of cases, and deletions of 7.6, 7.0 and 45 kb and double deletions have been described, all generated by unequal recombination during meiosis [4]. Deletions of 1 bp and nonsense variants in the signal peptide also abolish mature GH and may induce antibodies [4]. As the immune system has never been exposed to endogenous GH, the exogenous hormone behaves as a foreign protein and many patients develop neutralising antibodies that abolish the response; compound heterozygotes whose second allele produces a non-truncated protein are usually tolerant [5, 38]. Antibodies are nevertheless an inconsistent finding, even among patients with the same homozygous deletion [4]. In the French GENHYPOPIT cohort, the only patient with recessive GH1-related IGHD had a homozygous 16-bp deletion in exon 2 (NM_000515.5:c.16_31del p.(Arg6Trpfs*89)), anterior pituitary hypoplasia and transient partial thyrotrophic and corticotrophic deficiency, with an equally affected homozygous sister and a healthy heterozygous mother; no patient in this cohort had the classical complete homozygous deletion [9].

GH1-related type IB denotes biallelic variants, including splice variants, that reduce rather than abolish GH: low but detectable GH (below 7 mU/L, or 2.5 ng/mL, in the criteria quoted by Mullis), height below −2 standard deviations (SD), growth velocity below the 25th percentile, markedly delayed bone age, normal-stature parents, a good GH response and immune tolerance [4]. The phenotype is more variable than type IA: in some families it is similar, while in others growth is normal in early childhood and fails only in mid-childhood [4].

Autosomal dominant type II results predominantly from variants in the first six nucleotides of intron 3 causing exon 3 skipping, and also from variants in exonic (ESE1 and ESE2) and intronic splicing enhancers and intronic deletions in intron 3 [4]. The product is a 17.5-kDa GH lacking amino acids 32–71, including the loop between helices 1 and 2 [4]. This isoform is retained in the endoplasmic reticulum, disrupts the Golgi apparatus, impairs trafficking of normal GH from the intact allele and other hormones, and reduces stability of the 22-kDa isoform; transgenic mice develop pituitary hypoplasia due to somatotroph loss [4]. Small amounts of the 17.5-kDa isoform normally exist, and the phenotype depends on the isoform ratio: variant K41R, causing around 20% exon 3 skipping, is associated with normal or short stature [4]. This threshold effect explains variability in pituitary size, age at onset and severity, and why the most severe and rapidly presenting forms may progress to ACTH, TSH and gonadotropin deficiencies [4, 39]. Missense variants, designated in mature protein numbering as P89L, V110F and R183H, have also been associated with type II; the latter corresponds in current nomenclature to NM_000515.5:c.626G>A p.(Arg209His) [4, 36]. In GENHYPOPIT, 90% of GH1 variants were dominant, and exon 3 donor region variants (c.291+1G>A, c.291+1G>C and c.291+6T>C) accounted for more than half of GH1 variants [9].

Kowarski syndrome corresponds to structurally abnormal GH. The heterozygous variants R77C and D112G (historical mature-protein numbering, = NM_000515.5:c.307C>T p.(Arg103Cys) and c.413A>G p.(Asp138Gly)) were described in Japan, but R77C was also present in the normal-stature father, with no functional difference in vitro; the most convincing case is a Serbian boy (−3.6 SD) homozygous for C53S (historical numbering, = NM_000515.5:c.236G>C p.(Cys79Ser)), which disrupts the Cys53–Cys165 disulphide bridge and reduces GHR binding and JAK2/STAT5 signalling [4]. The heterozygous variant R178H (published designation, in mature-protein numbering) produces a partial agonist GH with IGHD type II and −6.0 SD at age 5 [4]. These examples show the need for functional studies before attributing causality to a missense variant [4].

2.3. GHRHR

GHRHR is located at 7p14.3 (OMIM 139191; MANE transcript NM_000823.4) [5, 36]. Biallelic loss-of-function variants cause a signalling defect rather than a defect in the GH molecule: GH1, the hormone and the peripheral receptor are intact, so exogenous GH is effective [5]. By 2010, six splice donor variants, two microdeletions, two nonsense variants, seven missense variants and one promoter variant had been described, affecting ligand binding or transduction [4]. The best-known founder variants are the nonsense variant NM_000823.4:c.214G>T p.(Glu72*), called “Glu50Stop” in the original literature (mature protein numbering), present in Wajnrajch's Indian families, Sindh, Delft Island (Sri Lanka) and Pakistani families, and the intron 1 donor variant c.57+1G>A (IVS1+1G>A) in the Itabaianinha kindred [2, 3, 5, 36]. An estimated common ancestor around 1350–2700 years ago for p.(Glu72*) explains why it is the main cause of familial IGHD in western India [5]. The same variant was identified in homozygosity in a consanguineous Pakistani family with four affected members, and the authors screened for it in the Qatar Genome Programme [40].

The GENHYPOPIT cohort expanded the spectrum: two of three index cases were compound heterozygotes (NM_000823.4:c.214G>T p.(Glu72*) with NM_000823.4:c.1141del p.(Gln381Lysfs*3); NM_000823.4:c.856A>T p.(Lys286*) with NM_000823.4:c.882+1G>T p.?), in contrast with the homozygosity in consanguineous settings that dominated the literature, and a third was homozygous for the synonymous variant NM_000823.4:c.1146G>A p.(Glu382=), which alters the last base of exon 12 and causes skipping of this exon [9]. Heterozygosity for null GHRHR alleles does not reduce adult height: in Itabaianinha, heterozygotes had a height of −1.84 SD, no different from wild-type homozygotes (−1.85 SD), with a tendency towards lower IGF-I, and in Sindh heterozygotes had moderately low IGF-I and slightly reduced GH responses without a relevant height deficit [2, 31]. Heterozygote body composition may nevertheless be altered [9].

2.4. GHSR and other secretion genes

The GHSR variant NM_198407.2:c.611C>A p.(Ala204Glu) reduces receptor membrane expression and abolishes constitutive activity [32, 36]. In GENHYPOPIT, GHSR was the second most frequent gene (4 of 23 diagnoses), always with this variant, in two homozygotes and two heterozygotes; homozygotes were not more severely affected, some heterozygous relatives had no deficiency, and inheritance (recessive or dominant with incomplete penetrance) remains debated [9]. ClinVar records conflicting classifications for this variant [36], and the physiological role of GHSR in human short stature remains controversial [4]. Muscarinic receptors M1–M5 were studied in 39 families with IGHD IB without causal variants (frequency below 2.6%), and by 2010 no pathogenic GHRH variants had been described [4].

2.5. X-linked forms

Type III combines IGHD with X-linked agammaglobulinaemia or hypogammaglobulinaemia. In the original family, the pattern was X-linked recessive and absence of B lymphocytes coexisted with GH deficiency [21]; the BTK variant causing exon skipping in a sporadic patient suggests that BTK itself explains the combined phenotype, but deletion of an Xq region involving two loci was also hypothesised [4, 29]. The X-linked agammaglobulinaemia GeneReviews chapter does not mention GH deficiency, indicating that the association is rare and is not part of the usual phenotype in patients with BTK variants [20].

SOX3 (Xq27) is dosage-sensitive: both loss and gain of function cause hypopituitarism [4, 5]. Expansions of the first polyalanine tract (+11 alanines) are associated with diabetes insipidus and GH deficiency, expansions of +7 alanines with panhypopituitarism without diabetes insipidus, and Xq26–q27 duplications including SOX3 (685.6 kb or smaller, down to 321 kb) cause a similar pituitary phenotype [5, 37]. Affected males have IGHD or CPHD, anterior pituitary hypoplasia, an ectopic posterior pituitary, an absent or hypoplastic infundibulum and, sometimes, corpus callosum abnormalities and intellectual disability; female carriers are usually unaffected or mildly affected, and several boys inherited duplications from mothers without a phenotype [5]. Regulatory SOX3 rearrangements also cause SRY-negative 46,XX testicular disorder of sex development, and position effects are associated with congenital hypertrichosis, neural tube defects and X-linked hypoparathyroidism [5]. ClinGen classifies SOX3 validity for X-linked hypopituitarism with or without diabetes insipidus as moderate [5].

2.6. Pituitary developmental genes with an isolated presentation

Presentation as IGHD does not exclude a developmental gene. In GENHYPOPIT, 30% of diagnoses (7 of 23) involved pituitary developmental genes: GLI2 (three patients, all with an ectopic posterior pituitary and pituitary stalk interruption syndrome, without other typical GLI2 features), POU1F1 (two), HESX1 (one, with complete IGHD diagnosed at age 33) and a SOX3 duplication [9]. The heterozygous POU1F1 variant NM_000306.4:c.227C>T p.(Pro76Leu) causes pure IGHD through a dominant-negative effect specific to the GH promoter, whereas the homozygous variant NM_000306.4:c.10C>T p.(Gln4*) produced, at age 16, −7 SD (final height 125 cm), peak GH below 0.1 ng/mL and prolactin deficiency without thyrotrophic deficiency at age 17, an exceptional course for a gene that usually affects GH, PRL and TSH [9]. In a previous study, 25% of GLI2 variant carriers had IGHD, always with an ectopic posterior pituitary [9]. HESX1 heterozygosity may cause mild IGHD, and a heterozygous 211-bp deletion in RARA was described as a cause of IGHD IB [4].

2.7. Normalised variants

GENHYPOPIT designations used earlier transcript versions (NM_000515.4 and NM_000823.3), with the same c. coordinates for the above variants [9]. The correspondence between ‘Glu50Stop’ and p.(Glu72*) results from the 22-amino-acid signal peptide difference between mature protein numbering and precursor numbering, the current convention [2, 36]. Huang and colleagues annotated c.291+1G>A as ‘p.Glu58Asnfs*120’, a frameshift prediction; the demonstrated mechanism for intron 3 donor variants is in-frame exon 3 skipping, with the 17.5 kDa isoform [4, 9, 10].

2.8. Genotype–phenotype correlations, penetrance and modifiers

The most robust correlations concern type II. In the review of 180 individuals with GH1-related IGHD II, splice variants (n = 62) were associated with significantly shorter stature than missense variants (n = 73; P = 0.0002) and large deletions (n = 3; P = 0.0104), and earlier diagnosis (median 2.2 years, interquartile range 1.0–4.3) [10]. Stature in type II ranges from −4.5 to −1.0 SD, with intrafamilial heterogeneity [10]. In Alatzoglou's cohort of 224 patients with congenital IGHD and a eutopic posterior pituitary screened for GH1 and GHRHR, variant-positive patients had more severe growth failure (mean height −4.7 versus −3.4 SD) but did not differ endocrinologically or on MRI [5, 41].

Interpretation of heterozygous GH1 deletions is controversial. Huang and colleagues attributed IGHD type II to heterozygous whole-gene deletions (approximately 1.5 and 1.7 kb) inherited from parents with heights between −1.0 and −2.89 SD, whereas GENHYPOPIT considered complete heterozygous GH1 deletion insufficient to explain the phenotype, because the expected inheritance is recessive [9, 10]. The same cohort classified as negative the cases with a single pathogenic allele in recessive genes (including GH1 NM_000515.5:c.114_118del p.(Met40Profs*11) and NM_000515.5:c.215del p.(Gln72Argfs*28), GHSR NM_198407.2:c.709A>T p.(Arg237Trp) and PROKR2 NM_144773.4:c.691G>A p.(Glu231Lys)) [9]. Evidence for heterozygous deletions as a monogenic cause is therefore weak. The GENHYPOPIT authors also acknowledge a possible polygenic burden [9].

Mosaicism is relevant in type II. In GENHYPOPIT, of four cases with an exon 3 donor variant in whom parents were studied, three were de novo and, in the fourth, the mother, apparently heterozygous by Sanger sequencing, had mosaicism with 30% variant allele in blood and saliva by next-generation sequencing (NGS), final stature +1 SD and normal IGF-I; the variant had arisen de novo in her [9]. This was the first described GH1 mosaicism, and the proportion in accessible tissues may not reflect the pituitary [9].

3. Clinical Spectrum and Manifestations by System

3.1. Neonatal period and infancy

GH is not necessary for intrauterine growth, and newborns with GHRHR-related IGHD have normal weight and length [3, 5]. In type IA, birth length is occasionally reduced and neonatal hypoglycaemia may occur, but severe growth retardation is uniformly established around 6 months [4]. Neonatal hypoglycaemia with low GH is the early clue in severe recessive forms [5]. In the review of 180 IGHD II patients, neonatal hypoglycaemia was recorded in only 1.7% and micropenis in 2.2%, showing that neonatal manifestations mainly suggest CPHD or severe recessive forms [10]. When GH deficiency coexists with ACTH deficiency, the newborn may have hypoglycaemia, prolonged jaundice, apnoea, hypotonia or seizures, and micropenis and midline lesions (cleft palate) point to pituitary dysfunction [34, 42].

3.2. Growth and auxology by subtype

IGHD short stature is proportionate. In type IA, growth failure is the most severe among Mendelian forms, and initial response to exogenous GH may be followed by dramatic slowing when neutralising antibodies emerge [4, 5]. In type IB, GH is low but detectable and bone age greatly delayed [5]. In GHRHR-related disease, growth failure is early and severe, with height down to −7.4 SD: in Sindh, the 18 affected individuals (15 male, 3 female, from newborn to 28 years) had mean height −7.2 SD, mean adult height 130 cm in men and 113.5 cm in women, head circumference −4.1 SD, blood pressure approximately −3 SD, normal proportions and no dysmorphism, micropenis or hypoglycaemia [2, 5]. In Itabaianinha, 21 homozygotes aged 8 to 63 years had height −7.35 ± 1.37 SD [31]. In type II, severity varies (−4.5 to −1.0 SD) and correlates with the quantity of 17.5 kDa isoform; splice variants cause more severe forms with earlier diagnosis than missense variants [5, 10]. In Huang and colleagues' series, six Chinese children with GH1 variants had mean age 4.64 ± 1.15 years, mean height −3.95 ± 1.41 SD and mean peak GH 2.83 ± 2.46 ng/mL at diagnosis [10]. In GENHYPOPIT, mean age at IGHD diagnosis was 3.9 years (median 3; 0.8–16), and genetic testing was performed at a mean 12.7 years [9].

3.3. Untreated phenotype: Sindh and Itabaianinha

GHRHR cohorts offer the most complete description of untreated natural history because most affected Itabaianinha individuals never received GH [3]. The phenotype includes proportionate short stature, a ‘doll-like’ face, high-pitched prepubertal voice, reduced muscle mass, central obesity, wrinkled skin and youthful-looking hair with late depigmentation and almost no grey hair [3, 5]. Quality of life and longevity were described as normal, and authors propose that lifelong GH deficiency in this kindred has more beneficial than harmful consequences; this interpretation concerns a single-founder-variant cohort and cannot be transferred to IGHD with hypoglycaemia or associated deficiencies [3]. Intelligence is normal, and fertility was documented in both sexes in Sindh, including an affected couple with a child [2]. Biochemically, GH fails to respond to GHRH, levodopa or clonidine, mean IGF-I was 5.2 ± 2.0 ng/mL and IGFBP-3 0.42 ± 0.13 µg/mL, both increasing normally with exogenous GH [2]. Anterior pituitary hypoplasia may result from loss of trophic signalling, but pituitary size varies even between siblings, and NREM (non-rapid-eye-movement) sleep may be altered [4, 5].

3.4. Progression to multiple pituitary deficiencies

In type II, ACTH, TSH and gonadotrophin deficiencies may develop, and endocrine follow-up is lifelong, without discharge on reaching final height [5, 39]. GENHYPOPIT documented this progression in SOX3 duplication carriers: the index case had intrauterine growth restriction, global developmental delay, −3.5 SD, complete GH deficiency, ectopic posterior pituitary and Duane syndrome, the carrier brother also developed corticotrophin deficiency and the uncle thyrotrophin deficiency [9]. In Alatzoglou's cohort, some initially isolated patients developed additional deficiencies, justifying long-term follow-up [5, 41]. GENHYPOPIT had no systematic longitudinal follow-up and did not assess progression to CPHD [9].

3.5. Extrapituitary manifestations

GH1 and GHRHR forms generally lack manifestations outside the somatotrophic axis. In the review of 180 IGHD II individuals, prominent forehead occurred in 11.7% (21/180), depressed nasal bridge in 6.7%, obesity in 3.9%, midface hypoplasia in 2.8%, micropenis in 2.2%, cryptorchidism in 1.7% and high-pitched voice in 1.7% [10]. Huang and colleagues described pulmonary hypertension in a child with IGHD II and suggested a possible association, with insufficient evidence [10]. GENHYPOPIT had isolated cryptorchidism (two patients), psychomotor delay, central precocious puberty and a bone malformation (fifth toe agenesis and fibular hypoplasia in a GHRHR patient, of uncertain causal relationship) [9]. Type III with BTK adds humoral immunodeficiency: recurrent sinopulmonary infections, absent specific antibodies and circulating B lymphocytes, delayed puberty in adults [21]. Childhood IGHD may affect health and neurodevelopment beyond height, and reversibility with treatment is unestablished [5, 35].

3.6. Treated disease and adulthood

With treatment, children with genetic IGHD respond well to GH, except type IA with antibodies, and more severe deficiency (low GH peak, low IGF-I, additional deficiencies, abnormal MRI or genetic defect) responds to lower doses than partial deficiency [22]. In Huang's series, four children treated from 4.5 ± 0.7 years gained 1.21 ± 0.3 SD at 6 months and 1.79 ± 0.15 SD after the first year [10]. Adulthood raises another issue: patients with a proven genetic cause retain GH deficiency, and guidelines recommend continuing treatment through transition to complete skeletal and muscle maturation (sections 7.10 and 11) [43, 44]. Adults with GH deficiency have increased osteopenia, osteoporosis and cardiovascular morbidity risk, without definitive data that treatment reduces the latter [44].

4. Clinical Recognition and Warning Signs

4.1. When to suspect, by age

In newborns and infants, genetic IGHD should be suspected with hypoglycaemia (especially with micropenis, prolonged jaundice or midline defects), length progressively diverging from the curves within the first 6 to 12 months, and consanguinity or an affected sibling [4, 22]. At this age, diagnosis requires no height cut-off, particularly in infants with hypoglycaemia or midline defects [22]. Every infant with unilateral or bilateral optic nerve hypoplasia should be assessed by paediatric endocrinology because of its association with septo-optic dysplasia and pituitary deficiencies [45]. In hypoglycaemia with ketosis, isolated GH deficiency and hypopituitarism are among the causes to be excluded before a diagnosis of idiopathic ketotic hypoglycaemia, alongside drugs (beta-blockers, alcohol), liver disease, adrenal insufficiency, glycogen storage diseases and gluconeogenesis defects [46]; in the newborn, the investigation of hypoglycaemia includes the exclusion of GH deficiency and panhypopituitarism [42].

In children, assessment is indicated with height below −2 SD, height discordant with the family context or deceleration of at least 0.3 SD per year not explained by channelling in early childhood, prepubertal deceleration or delayed puberty [22]. Short stature with obesity is a warning sign, because overweight children tend to be tall [47]. Boys and girls should be assessed by the same criteria, because short girls have historically been underinvestigated [22]. Severe short stature (below −3 SD or more than 3 SD below target height), a family history with autosomal dominant or recessive segregation and consanguinity increase the probability of a monogenic cause [22]. In a child already receiving GH, loss of response with a GH1 deletion should suggest neutralising antibodies [5, 22].

In adolescents and adults, genetic IGHD re-emerges during transition: a patient treated in childhood who stops GH may develop changes in body composition, bone mineral density and metabolic profile, and IGHD type II carriers may develop ACTH, TSH or gonadotrophin deficiencies at any age [5, 44].

4.2. Emergency signs

The main emergency is unrecognised adrenal insufficiency, the deficiency that kills in congenital hypopituitarism: in acute presentation with hypoglycaemia, cortisol and TSH deficiencies should be excluded and treated promptly [5]. In a series of 88 individuals with septo-optic dysplasia (median 12.4 years), five deaths occurred, three from adrenal crisis [34]. GH initiation increases cortisol metabolism and may unmask latent adrenal insufficiency, and levothyroxine should not be started before assessing and treating the adrenal axis [19]. Seizures in infants with pituitary deficiencies are often secondary to hypoglycaemia rather than a primary brain anomaly [48]. In type III patients, recurrent bacterial infections are the immunological warning sign [21].

4.3. Newborn screening

There is no newborn screening for GH deficiency. TSH-based programmes do not detect central hypothyroidism, which in POU1F1 forms may precede clinically evident GH deficiency (median 6 months for TSH versus 3 years for GH in a western Indian cohort), so free T4 should be measured when hypopituitarism is suspected [5].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. General principle

The 2019 Growth Hormone Research Society (GRS) consensus, bringing together 46 experts from 14 countries and non-voting European Medicines Agency (EMA), Food and Drug Administration (FDA) and industry representatives, states that GH deficiency diagnosis remains clinical: it synthesises auxology, anatomy and laboratory findings and should not rely on tests alone [22]. Childhood IGHD diagnosis is multistep (history, examination with detailed auxology, biochemistry, imaging and increasingly genetics) [35]. There is no formally validated criteria set for genetic IGHD; molecular diagnosis rests on biochemical demonstration of isolated GH deficiency and identification of pathogenic or likely pathogenic variants with the expected inheritance pattern [5, 9].

5.2. Auxological assessment and first-line investigations

History should include three generations with parental heights and consanguinity, parental pubertal age, pregnancy, birth weight, length and head circumference corrected for gestational age, growth curve, chronic diseases and development [47]. Examination assesses disproportion (arm span, sitting-height-to-height ratio, preferable to segment ratio), deformities, asymmetry, pubertal stage, hypothyroidism signs and dysmorphism, and includes head circumference [22, 47]. Target height is calculated for girls as [(father's height − 13) + mother's height]/2 with ±9 cm range and for boys as [(mother's height + 13) + father's height]/2 with ±10 cm range; the formula misleads if a parent has a genetic short stature cause, frequent in IGHD type II [47].

First-line investigations include biochemistry (electrolytes, urea, thyroid function), full blood count and erythrocyte sedimentation rate, urine, coeliac antibodies and bone age (left wrist, difficult to interpret below 4 years), plus karyotype or microarray in all girls to exclude Turner syndrome [47]. A tertiary-centre study showed low laboratory screening yield in children with growth velocity above 5 cm/year, except for coeliac disease [22]. Bone age is typically delayed in GH deficiency, and investigating this deficiency is unnecessary in a child with longstanding short stature without delayed bone age, except recent acquired deficiency; it is less useful in obesity and below 2 years, and Greulich-Pyle standards derive from white North American children of Northern European ancestry [22].

5.3. IGF-I and IGFBP-3

IGF-I should be measured with an assay having age-, sex- and pubertal-stage reference values; it varies with method, malnutrition, chronic disease and liver disease [22]. Below 3 years, the normal range includes the detection limit and IGFBP-3 is more reliable; after 3 years, IGFBP-3 is less sensitive [22]. IGF-I above 0 SD at any age makes GH deficiency unlikely [22]. Low IGF-I and IGFBP-3 suggest GH deficiency but also prolonged malnutrition and GH insensitivity; low IGF-I with normal or high IGFBP-3 may suggest an IGF1 defect; high IGF-I suggests IGF1R, some IGF1 forms or PAPPA2 variants [22]. In Kowarski syndrome, dissociation between normal or high GH and low IGF-I, with recovery under bioactive hormone, is diagnostic, requiring IGF-I measurement rather than inferring activity from GH [5]. According to the Endocrine Society, in a patient with three or more pituitary deficiencies, low IGF-I is as specific as any stimulation test, and the GRS Delphi panel considered subnormal IGF-I diagnostic in panhypopituitarism in children and adults [43, 49].

5.4. GH stimulation tests

Testing is performed fasting and after replacing other deficiencies (hypothyroidism, hypogonadism) [22, 44]. Stimuli in paediatric use are insulin-induced hypoglycaemia (ITT, insulin tolerance test), glucagon, arginine, clonidine, levodopa and GHRP-2 (growth hormone-releasing peptide 2, used intravenously in Japan); GHRH alone is unsuitable in children because it misses hypothalamic or stalk deficiency, and there are no data showing superiority of one stimulus [22]. Diagnosis requires insufficient response to two stimuli, reducing but not eliminating false positives; with high suspicion, a single test may suffice, and a sufficient response excludes deficiency in most cases [22]. The GRS Delphi panel reached consensus recommending arginine and glucagon tests in children and considered a peak below 5 µg/L consistent with severe deficiency [49]. Paediatric consensus was weaker than adult consensus (17 of 29 paediatric statements versus 28 of 32 adult statements), showing that an isolated borderline test is not definitive in paediatrics [49].

The traditional paediatric 10 µg/L cutoff was defined using polyclonal radioimmunoassays. Monoclonal antibody assays and recombinant standards give values approximately 40% lower, so cutoffs should fall; most GRS experts suggested approximately 7 ng/mL (equivalent to 7 µg/L), assay-dependent, already adopted in Australia, New Zealand, Canada, most of Europe and Japan, without requiring retesting of children diagnosed using other thresholds [22, 50]. Outside the neonatal period, cutoffs between 4.7 and 6.5 µg/L probably establish deficiency, varying with assay and region [50]. Severe deficiency is frequently defined as peak below 3 ng/mL with high prior probability [22]. The PROP1 GeneReviews expresses the threshold as ‘GH < 7 ng/dL’; the unit is a source typographical error and the threshold corresponds to 7 ng/mL [19, 22]. Interlaboratory variability is high: in the United Kingdom, 96 laboratories had a geometric coefficient of variation of 25% (mean 7 µg/L corresponding to 5–10 µg/L), and GH should be expressed in mass units with international standard 98/574 (1 mg = 3 IU) [44, 50].

Peak GH correlates inversely with body mass index (BMI): a meta-analysis estimated a 12% peak reduction per BMI SD, but data are insufficient to adjust paediatric cutoffs [22, 50]. Girls and women secrete more GH through oestrogen effects, and without sex-specific cutoffs boys risk overdiagnosis [50]. Sex steroids increase response before Tanner stage III; priming is unstandardised, and GRS consensus limits it to adolescents with suspected constitutional delay, whereas the 2016 Pediatric Endocrine Society (PES) supported it in prepubertal children older than 11 years (boys) and 10 years (girls) [22, 50]. Most peripubertal GH deficiency diagnoses prove to be constitutional delay, and with normal MRI the diagnosis should be reconsidered at growth completion [22]. Deficiency may evolve, and retesting is indicated in certain genetic conditions [22]. Rare patients with true deficiency have peaks above cutoffs (marked short stature, low velocity, delayed bone age, very low IGF-I, abnormal MRI) and may need GH with annual reassessment [22]. ITT requires caution because of severe hypoglycaemia [22].

5.5. Situations in which stimulation testing is unnecessary

Testing is unnecessary in newborns and infants with high pretest probability, that is, hypoglycaemia, hyperbilirubinaemia, poor growth, midline defects, micropenis, low IGF-I and IGFBP-3, TSH or ACTH deficiencies and abnormal MRI [22]. It is also unnecessary when MRI shows a large sellar mass or pituitary defect (hypoplasia, absent or hypoplastic stalk, ectopic posterior pituitary) with other clinical findings, and with an obvious alternative diagnosis (Turner, Noonan, Prader-Willi, ACAN, SHOX, chronic kidney failure, Silver-Russell, small for gestational age [SGA] without catch-up), unless GH deficiency is additionally suspected [22]. Dynamic tests cannot be performed neonatally; neonatal deficiency is inferred from auxology, MRI and low growth factors, and posterior pituitary function is assessed through plasma and urine osmolalities [5].

5.6. Adults and transition: cutoffs and tests

In adolescents and young adults, cutoffs may be higher (6.1 µg/L in ITT and 19.0 µg/L with GHRH-arginine, according to studies cited by the Endocrine Society), and a study in young people aged 17 to 19 years identified the at-risk group with ITT at 6 µg/L, whereas earlier consensus documents prefer the adult criterion (ITT < 3 µg/L) [43, 50]. AACE recommends ITT with a 5.0 µg/L cutoff during transition or, alternatively, glucagon and macimorelin with the same adult cutoffs (AACE recommendation 24 expresses the macimorelin cutoff as ‘2.8 mg/L’, a source unit error, thus read as 2.8 µg/L) [44]. In adults, diagnosis requires no stimulation test with organic hypothalamic–pituitary disease and three or more pituitary deficiencies and IGF-I below −2 SD, a genetic hypothalamic–pituitary axis defect or a structural hypothalamic–pituitary brain defect [44]. Isolated low IGF-I is insufficient with two or fewer deficiencies, and testing is then required [44].

Macimorelin, an oral ghrelin receptor agonist, is administered after fasting at least 8 h and withholding GH for at least a week, with sampling at 30, 45, 60 and 90 min [51]. In the crossover study with ITT (140 patients with both tests), negative agreement was 94% (lower 95% confidence interval [95% CI] limit 85%), positive agreement 74% (lower limit 63%) and overall agreement 84%; in the low-probability group, including idiopathic childhood-onset IGHD, positive agreement was only 33% (2/6), and 17.2% of ITTs failed to achieve hypoglycaemia [51]. On 23 July 2026, FDA approved a label amendment stating diagnostic accuracy was not demonstrated in a multicentre, open-label, randomised crossover study in 102 children aged 3 to 17 years with suspected GH deficiency; the indication remains adult-only [52]. Macimorelin should therefore not be used to diagnose childhood IGHD.

5.7. Hypothalamic–pituitary magnetic resonance imaging

MRI is indicated in all patients with diagnosed GH deficiency because it identifies anatomical defects and tumours and predicts other deficiencies, genetic testing utility and persistence of deficiency [22]. It is not a diagnostic test for deficiency, and an isolated small pituitary is insufficient; the most suggestive findings are absent anterior pituitary, ectopic posterior pituitary and stalk or pituitary hypoplasia, interpreted by pubertal stage [22]. In IGHD, MRI detects hypothalamic–pituitary anomalies in only 12–20% of cases [4]. Normal MRI does not exclude GHRHR: in Alatzoglou's 224-patient cohort, MRI did not differ between variant-positive and variant-negative patients [5, 41]. In GENHYPOPIT, of 19 available MRIs, 47% showed an aplastic or hypoplastic pituitary (three GH1, three GLI2, two GHRHR, one GHSR) and 21% an ectopic or absent posterior pituitary (all three GLI2 and the SOX3 duplication) [9]. Ectopic posterior pituitary points towards developmental genes (GLI2, SOX3, HESX1, LHX4) rather than GH1 or GHRHR [9, 19]. Published mean pituitary volumes were 231 ± 146 mm³ in idiopathic deficiency, 287 ± 108 mm³ in idiopathic short stature and 344 ± 146 mm³ in controls [50].

5.8. Genetic testing

Genetic testing is not indicated in all short children. GRS indicates it in severe familial IGHD, syndromic multiple-deficiency forms, severe short stature (below −3 SD or more than 3 SD below target height), disproportion or skeletal dysplasia and SGA without catch-up; probability increases with autosomal dominant or recessive segregation and consanguinity [22]. With severe growth failure and family history, GH1 and GHRHR should be studied: in Alatzoglou's cohort, yield was 11.1% overall and 38.6% in familial cases, and among variant-positive families the recessive group (types IB and IV) represented 42.8% [5, 41]. Without a targeted hypothesis, GRS proposes single-nucleotide polymorphism microarray (SNP-array, detecting copy number variants [CNV] and uniparental disomies) followed by an exome-based growth panel, with exome sequencing particularly below −3 SD and in syndromic presentations [22].

GENHYPOPIT is the current European reference for panel yield. Between 2017 and 2024, 205 index cases with IGHD without other deficiencies were studied using a 13-gene panel (FGF8, FGFR1, GH1, GHRHR, GHSR, GLI2, HESX1, LHX4, POU1F1, PROKR2, PROP1, RNPC3, SOX3), with 100% coverage at 30×, the 2015 American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP) classification, NGS and multiplex ligation-dependent probe amplification (MLPA) copy number analysis and quantitative polymerase chain reaction (PCR) for SOX3 [9]. A cause was found in 23 (11.2%): 22 single-nucleotide variants and one SOX3 duplication; 70% in GH secretion genes (GH1 39%, GHSR 17%, GHRHR 13%) and 30% in developmental genes (GLI2 13%, POU1F1 9%, HESX1 4%, SOX3 4%), without diagnoses in RNPC3, FGF8, FGFR1, LHX4, PROKR2 or PROP1; ten variants were novel [9]. This yield compares with 6.5% in an earlier study of 475 patients without GHSR and 6–7% in the era of iterative Sanger sequencing of 8–12 genes [9]. In a Chinese cohort of 792 short children studied by exome sequencing, prevalence of pathogenic or likely pathogenic GH1 variants was 0.76% (6/792) [10].

5.9. Genetic test limitations

The technical limitations are specific to this gene group. The high homology of GH1 with GH2, CSH1, CSH2 and CSHL1 underlies the unequal-recombination deletions that constitute the main mechanism of type IA (6.7, 7.0, 7.6 and 45 kb), and these deletions are not detected by sequencing alone: they require copy number analysis, which in GENHYPOPIT combined NGS data with MLPA [4, 9]. SOX3 duplications require copy number analysis and in GENHYPOPIT were confirmed by quantitative PCR [9]. Splice variants outside the canonical dinucleotides (such as GH1 NM_000515.5:c.291+6T>C or NM_000515.5:c.172-8_172-3del) and synonymous variants that alter the last base of an exon (GHRHR NM_000823.4:c.1146G>A) may be underclassified without RNA studies; GENHYPOPIT studied GH1 splicing in blood mRNA in three patients [9]. Low-level parental mosaicism may escape Sanger sequencing, which justifies studying the parents by NGS [9]. In PROP1 deficiency, sequencing detects more than 98% of the reported pathogenic variants and deletion/duplication analysis less than 2%, but these values measure the proportion of detectable variants, not clinical sensitivity [19]. A negative result does not exclude a genetic cause: approximately 89% of GENHYPOPIT IGHD cases remained undiagnosed, and 84–90% of CPHD cases have no identified aetiology [9, 37].

Interpretation requires caution regarding inheritance. A single pathogenic allele in a recessive gene does not establish diagnosis, and GENHYPOPIT excluded variants of uncertain significance (VUS) without functional studies [9]. In PROP1 forms, biallelic VUS or a pathogenic variant with a VUS neither establish nor exclude diagnosis [19]. VUS should be interpreted with family segregation and functional studies, and reports should explicitly state limitations [22]. The report should indicate gene, versioned transcript (for example NM_000515.5 for GH1 and NM_000823.4 for GHRHR), HGVS c. and p. nomenclature, zygosity, ACMG/AMP class and, for splice variants, whether the effect was demonstrated in RNA [9, 36].

Identifying the cause has practical consequences enumerated by GRS: guiding GH treatment in some patients, providing prognostic information, facilitating surveillance for associated conditions, rarely identifying diseases contraindicating growth therapy (such as Bloom syndrome), ending the diagnostic odyssey, enabling genetic counselling and identifying relatives; pharmacogenomics is an unproven future possibility [22]. In genetic IGHD, identifying a variant also removes the need for transition retesting (section 7.10) [43, 44].

5.10. Practical algorithm

5.11. Prenatal and preimplantation diagnosis

Prenatal diagnosis and preimplantation genetic testing (PGT) are possible when the familial variant is known, and serial ultrasound monitors fetal growth [19, 47]. Because GH is unnecessary for intrauterine growth, ultrasound is not a diagnostic tool for isolated IGHD [3].

6. Differential Diagnosis

6.1. Comparative overview

6.2. Genetic panhypopituitarism (combined pituitary hormone deficiency)

Congenital CPHD has more than 30 associated genes, syndromic and non-syndromic, but 84–90% of cases remain without an aetiology; the spectrum ranges from holoprosencephaly to IGHD, with incomplete penetrance and variable expressivity [19, 37]. In a meta-analysis of 21 studies, HESX1, LHX3, LHX4, PROP1 and POU1F1 explained 12.4% of CPHD (11.2% of simplex and 63% of familial cases) [19]. Molecular diagnosis helps predict progression, avoid unnecessary surgery for PROP1 pituitary masses and plan families [37].

6.2.1. PROP1

PROP1 (5q35.3; NM_006261.5) encodes a homeodomain factor necessary for somatotroph, lactotroph, thyrotroph and gonadotroph ontogeny; missense variants in the two conserved basic homeodomain regions responsible for nuclear localisation and DNA binding cause loss-of-function CPHD [19]. It is the world's most common genetic CPHD cause (approximately 11% of cases) [19, 37]. Unlike other congenital forms, newborns have normal weight and length, and hypoglycaemia and prolonged jaundice are infrequent [19]. Deficiencies develop progressively, usually in the order GH, LH/FSH, TSH and ACTH, with variable degree and age even within families [19].

Source: PROP1 GeneReviews, Table 2 [19].

Growth decelerates moderately to severely in the first year (−1.5 ± 0.9 SD at 18 months) and more markedly between 1.5 and 3 years (−3.6 ± 1.3 SD at 3 years), with greatly delayed bone age (−4.0 SD at 2.5 years) and small hands and feet [19]. Hypothyroidism is mild and late, without intellectual disability; gonadotrophin deficiency presents with micropenis (stretched length below 2.5 cm at term), cryptorchidism, absent, delayed or incomplete puberty and infertility, and may be the initial presentation, with GH and TSH failing only in adulthood [19]. PRL deficiency compromises lactation, and ACTH deficiency presents with persistent weakness, abdominal pain, anorexia and weight loss, without a described neonatal presentation, and adrenal function continues declining even after more than four decades [19]. Intelligence is usually normal; immature facies, depressed nasal bridge and reduced vertical facial dimensions are described, and obesity is rare in childhood and more common in adults [19]. On MRI, stalk and posterior pituitary are normal and anterior pituitary may be hypoplastic, normal or diffusely enlarged, mimicking pseudotumour, non-functioning adenoma, craniopharyngioma or Rathke's cleft cyst, with initial childhood enlargement, fluctuation and complete involution in adolescence or adulthood [19, 37].

The frequency of PROP1 among CPHD cases varies by population: 70.1% (47/67) in Lithuania, approximately 53% (16/30) in Brazil and 50.8% (32/63) in Poland, versus 0% in the Netherlands (0/50), the United Kingdom (0/27), Japan (0/71), Korea (0/12) and Australia (0/33) [19]. PROP1 is present in 6.7% of simplex and 48.5% of familial cases [19]. The GA repeat at positions 296–302 is a mutational hotspot [4]. Genotype–phenotype correlations are limited, and a patient homozygous for NM_006261.5:c.582G>A p.(Trp194*) had initial hypogonadotropic hypogonadism with progressive GH and TSH deficiencies in adulthood, whereas another attained normal height and developed CPHD only in adulthood [19, 37].

6.2.2. POU1F1

POU1F1 (PIT-1; 3p11.2; NM_000306.4) specifies somatotrophs, lactotrophs and thyrotrophs and activates the GH and PRL genes and, with GATA2, the TSH beta-subunit gene [5]. The hormonal signature is GH, PRL and TSH deficiency with preserved ACTH, LH and FSH, which distinguishes it from PROP1 [5]. Growth failure is severe and often early, with a prominent forehead, marked midface hypoplasia, depressed nasal bridge, deep-set eyes and a short nose [5]. In a systematic review of 114 patients from 58 studies, all had GH deficiency, TSH was preserved in only 12.5% and PRL in 4.4%; the gonads are intact, and central precocious puberty has been described in two sisters [5]. MRI is normal or shows hypoplasia (63.6% of heterozygotes and 86.3% of biallelic cases), without stalk interruption or ectopic posterior pituitary, and low or undetectable PRL is the biochemical clue [5]. Variants are mainly recessive, but dominant-negative variants occur: NM_000306.4:c.811C>T p.(Arg271Trp) is a heterozygous hotspot with incomplete penetrance in some relatives, and splice variants shift production towards the repressive beta isoform [4, 5, 37]. Approximately 3% of congenital pituitary deficiency, which affects approximately 1 in 4000 live births, results from variants of the POU1F1 alpha isoform [5]. With partial-transactivation variants, TSH deficiency may develop after GH is started, so thyroid function should be reassessed after initiation of treatment [5].

6.2.3. HESX1 and septo-optic dysplasia

Septo-optic dysplasia (SOD) is diagnosed by the presence of at least two of three features (optic nerve hypoplasia, absent septum pellucidum, pituitary dysfunction) [34, 47]. It is aetiologically heterogeneous and mostly sporadic; teratogens (intrauterine viral infection, valproic acid, alcohol), vascular disruption, young maternal age, primiparity and first-trimester bleeding, and variants of HESX1, SOX2, SOX3 and OTX2 have been implicated [34]. In one series, 80% had GH deficiency, 60% ACTH deficiency, 40% LH/FSH deficiency and 35% diabetes insipidus; half had a septum pellucidum present, and approximately half had moderate or severe developmental delay [48]. In a series of 80 patients with SOD, 55% had hypothalamic–pituitary dysfunction, diagnosed by 2 years of age in 86% of these, with the greatest risk when optic nerve hypoplasia was combined with septal or corpus callosum dysgenesis [5]. Intellectual disability occurs in approximately 30%, especially with other central nervous system defects [34]. HESX1 (3p14.3) explains less than 1% to approximately 1.1% of screened cases (8 of 724 in a cohort of 1765 with congenital hypopituitarism), with phenotypes ranging from severe SOD to mild CPHD and IGHD [5, 48]. The homozygous variant NM_003865.3:c.478C>T p.(Arg160Cys) (Arg53Cys in homeodomain numbering), which abolishes DNA binding, causes familial SOD, and many heterozygous variants have incomplete penetrance, so an unaffected parent may be a carrier [5, 37]. SOD is considered mainly sporadic, with a low recurrence risk, and the evidence points to a multigenic disorder with environmental interaction [48]. Optic nerve hypoplasia, a non-progressive congenital anomaly, unilateral or bilateral, with a small pale disc, nystagmus when bilateral and strabismus when unilateral, has a differential diagnosis that includes the tubulinopathies (with lissencephaly), papillorenal syndrome (PAX2, autosomal dominant), PAX6 variants (colobomas, morning glory anomaly) and environmental causes (alcohol, anticonvulsants, maternal diabetes); assessment includes electroretinography (normal) and visual evoked potentials (abnormal), MRI of the optic nerves and midline, pituitary investigation and renal ultrasound [45]. Recurrence of isolated optic nerve hypoplasia is low, but it may manifest as SOD [45].

6.2.4. LHX3, LHX4, SOX2, OTX2 and GLI2

Recessive LHX3 (9q34.3) causes GH, TSH, LH, FSH and PRL deficiency with ACTH usually preserved, cervical spine stiffness with limited head rotation, vertebral anomalies and sensorineural deafness; the pituitary may be hypoplastic or secondarily enlarged [5, 19, 37]. LHX4, mainly dominant with incomplete penetrance, is associated with a small sella turcica, cerebellar malformations (Chiari), small anterior pituitary and ectopic posterior pituitary; in a cohort of 417 probands with IGHD or CPHD and ectopic posterior pituitary or sellar anomaly, it accounted for approximately 1.4%, and biallelic variants are neonatally lethal [5, 37]. SOX2 and OTX2 are associated with anophthalmia and microphthalmia: truncating SOX2 variants were found in 11% (4/35) of anophthalmia, with hypogonadotrophic hypogonadism or CPHD, maternal gonadal mosaicism and non-progressive pituitary masses; OTX2 explains 2–3% of anophthalmia and microphthalmia, approximately one third with pituitary deficiencies (19/55), independent of ocular severity, with 42% de novo variants and documented parental mosaicism [5, 37, 54]. Heterozygous loss-of-function GLI2 variants are associated with ectopic posterior pituitary, polydactyly and midline defects from hypotelorism to holoprosencephaly; more than 20% of carrier parents are unaffected, one third have CPHD and 13% IGHD [19, 37]. Pallister-Hall syndrome (GLI3) combines hypothalamic hamartoblastoma and panhypopituitarism, with hypoadrenalism the main neonatal cause of death [55].

6.2.5. SOX3 and other X-linked genes

SOX3 was described in section 2.5. Central hypothyroidism forms part of isolated TSH deficiency (TSHB, TSHR) or CPHD (POU1F1, PROP1, HESX1, LHX3, LHX4, SOX3, LEPR) [56]. X-linked central hypothyroidism due to IGSF1, with macro-orchidism and variable PRL and GH deficiencies, is another differential in boys with central hypothyroidism and GH deficiency [56]. Isolated hypogonadotrophic hypogonadism (more than 30 genes, including ANOS1, CHD7, FGFR1, PROKR2 and SOX10) may overlap with CPHD or evolve into it [19].

6.3. Growth hormone insensitivity

Growth hormone insensitivity (Laron syndrome and primary IGF-1 deficiency) is part of the differential diagnosis. GH insensitivity is defined by low IGF-I with normal or high GH and absent IGF-I response to exogenous GH; IGF-I deficiency is the pathognomonic feature [33]. Primary forms result from defects of GHR, of signalling (STAT5B, STAT3), of IGF-I synthesis, of the IGF1R receptor or of IGFALS; secondary forms include anti-GH or anti-GHR antibodies, malnutrition, inflammatory bowel disease, severe illness, liver disease and poorly controlled diabetes [33]. The EMA defines severe primary IGF-I deficiency as height at or below −3.0 SD, basal IGF-I below the 2.5th centile for age and sex, GH sufficiency and exclusion of secondary causes, with optional IGF-I generation testing; the FDA uses height and basal IGF-I at or below −3.0 SD with normal or high GH [25, 26].

Source: adapted from Mastromauro and colleagues, Table 1, with the GRS consensus note on high IGF-I in IGF1R variants [22, 33].

Laron syndrome (OMIM 262500) is the most frequent primary insensitivity form: autosomal recessive with complete penetrance, final height −5 to −10 SD, facial hypoplasia, obesity, small genitalia and delayed puberty, phenotypically indistinguishable from severe congenital GH deficiency [33]. More than 90 GHR variants have been described (68 extracellular, 13 intronic, 10 intracellular and two transmembrane), including deep intronic pseudoexons; GH-binding protein (GHBP) is low or undetectable in approximately 80% of extracellular variants and normal or high in transmembrane and intracellular variants [33]. Some heterozygous dominant-negative variants cause milder partial insensitivity (OMIM 604271) in children previously labelled idiopathic short stature [5, 33]. Hypoglycaemia is frequent: in the long-term recombinant IGF-I trial, it affected 49% of children before and during treatment [5]. Oxford Desk Reference attributes Laron syndrome to a ‘growth hormone-releasing hormone receptor mutation’, an error: Laron syndrome results from GHR variants, the GH receptor, whereas GHRHR variants cause GH deficiency with IGF-I responsive to exogenous GH [2, 33, 42]. Oxford Desk Reference's clinical observation that GH insensitivity may present in infancy with hypoglycaemia, short stature, micropenis and obesity remains valid [42].

Biallelic STAT5B deficiency (OMIM 245590) combines severe progressive postnatal growth failure (the first case in 2003, a 16-year-old Argentine girl, had height below −10 SD) with immunodeficiency, eczema, elevated IgE, autoimmunity, herpes infections and potentially fatal chronic lymphocytic interstitial pneumonia [5, 33]. Heterozygous dominant-negative variants (OMIM 618985) cause a milder picture, and somatic STAT5B variants in haematological malignancies have no germline significance [5]. Mecasermin response is lower than in GHR defects, and the distinguishing clinical clue is immune and lung involvement [5]. Heterozygous STAT3 gain of function causes early multiorgan autoimmunity, immunodeficiency and short stature [33].

Biallelic IGF1 deficiency (OMIM 608747) combines intrauterine growth restriction and postnatal failure (below −3 SD), microcephaly, intellectual disability and sensorineural deafness, with high GH, very low or undetectable IGF-I and normal IGFBP-3; heterozygosity modestly reduces height and head circumference [5, 33]. IGFALS-related ALS deficiency (OMIM 615961) has very low IGF-I with even lower IGFBP-3, only mildly to moderately reduced stature, delayed puberty particularly in boys, insulin resistance and reduced bone mineral density, with preserved male fertility [5, 33]. Isolated IGF1R haploinsufficiency or 15q26 deletion causes intrauterine growth restriction and postnatal failure with height −2.5 to −3 SD, delayed bone age, sometimes microcephaly and mild learning difficulties [47].

6.4. Other causes of short stature

The most common causes of short stature (75% of cases in a British source) are familial short stature, with bone age equal to chronological age and final height within the target range, and constitutional delay of growth and puberty, with bone age equal to height age and a good prognosis for normal adult height [47]. They are followed by chronic diseases (10%), syndromes (6%), chromosomal anomalies (5%), skeletal dysplasias (1%) and GH deficiency and GH insensitivity (1–2%) [47]. SHOX haploinsufficiency explains 2.4–7% of idiopathic short stature, with a prevalence similar to that of GH deficiency and Turner syndrome [47]. Silver–Russell syndrome results from ICR1 hypomethylation at 11p15 in up to 60% of cases and from maternal uniparental disomy of chromosome 7 in 5–10% [47]. Other differentials include 3M syndrome (CUL7, OBSL1, CCDC8; final height −5 to −6 SD), hypochondroplasia (FGFR3), Noonan syndrome, Albright hereditary osteodystrophy (GNAS), Floating-Harbor syndrome (SRCAP), Aarskog, Rubinstein–Taybi and Cornelia de Lange syndromes, the osteodysplastic dwarfisms and DNA repair defects [47]. Among low-birth-weight newborns, 88% catch up in the first year [47].

7. Treatments and Risk Management Strategies

7.1. Principles

Recombinant GH (somatropin) is established replacement therapy in IGHD due to GH1 (except with neutralising antibodies), GHRHR, GHSR and developmental genes, because peripheral receptor and downstream pathways are intact [5]. Approved indications are defined by mechanism (‘growth failure due to inadequate endogenous GH secretion’), not gene, and there is no specific trial in genetic cohorts such as GHRHR [5, 23]. Daily somatropin and long-acting GH (LAGH, long-acting growth hormone) have comparable short- and medium-term efficacy; choice depends on age, regulatory approval, adherence and hypoglycaemia risk [24]. Mecasermin is reserved for primary GH insensitivity and, in the USA, GH1 deletion with neutralising antibodies [25]. In multiple-deficiency forms, replacement sequence is critical: cortisol is assessed and replaced before levothyroxine, and GH may unmask adrenal insufficiency and hypothyroidism [19, 57].

7.2. Daily somatropin

Somatropin is administered subcutaneously in 6 to 7 injections weekly [23, 58]. Dose is individualised, seeking the lowest effective dose [22]. Starting doses differ by region: 25 µg/kg/day (0.18 mg/kg/week) in most of Europe, Canada and Japan, 4.5 mg/m²/week in Australia and up to 43 µg/kg/day (0.3 mg/kg/week) in the USA [22]. More severe deficiency, including genetic deficiency, responds to lower doses: 17–35 µg/kg/day (0.16–0.24 mg/kg/week), approximately 0.7–1.0 mg/m²/day, may suffice [22]. Main adjustment is based on growth (velocity and height SD change) every 6 to 12 months, with annual IGF-I or after adjustment; target IGF-I is near 0 SD, although severe deficiency may grow well with IGF-I below 0 SD, and a 20% dose adjustment changes IGF-I by approximately 1 SD [22]. Consecutive values above +2 SD justify dose reduction unless insensitive [22].

An inadequate response in the first year (6 to 12 months) is defined by an increase in velocity below 2 cm/year, velocity SD below 0 or a gain below 0.3 height SD per year [22]. This response warrants a review of adherence and technique, investigation of other causes (scoliosis, coeliac disease, inflammatory bowel disease, hypothyroidism, nutrition, drugs, psychosocial factors, dysplasias) and, in complete GH1 deletions, testing for neutralising anti-GH antibodies; with IGF-I below target, the dose is increased, and GH is discontinued if the suboptimal response persists [22]. The labels add hypothyroidism, undernutrition and advanced bone age to the causes to exclude [23, 57]. Oral oestrogen reduces the biological effect of GH and requires higher doses, unlike the transdermal route [43, 59]. Aromatase inhibitors in pubertal boys are a controversial, off-label option; gonadotrophin-releasing hormone (GnRH) analogues alone are not indicated, and their combination with GH may be considered, off-label and in reference centres, in GH deficiency with short stature at the onset of puberty [22]. Genetic GH insensitivity may respond to recombinant IGF-I (section 7.7) [22].

7.3. Long-acting GH

The first approved LAGH, Nutropin Depot® (FDA, 1999), was withdrawn in 2004 for manufacturing and commercial reasons; South Korea and China approved LAGH in 2014 [24]. Current molecules use different technologies (reversible albumin binding, a prodrug with transient PEGylation, protein fusion, PEGylation), with different pharmacokinetics, doses and IGF-I interpretation; each molecule is a new biological entity, and doses in mg are not interchangeable with one another or with daily GH [22, 24].

Sources: FDA and EMA labels, GRS 2026 consensus and international consensus by Maniatis and colleagues [24, 57, 59–64].

In the EU, paediatric somapacitan adjustment uses mean IGF-I sampled 4 days after the dose, with a target between −2 and +2 SD (close to 0): if above +2 SD, it is reassessed after the next dose and, if persistent, the dose is reduced by 0.04 mg/kg/week; doses above 0.16 mg/kg/week have not been studied in paediatric deficiency and are not recommended [60]. For somatrogon, IGF-I above the mean + 2 SD leads to a 15% dose reduction [64]. For lonapegsomatropin, the European summary of product characteristics (SmPC) provides for a reduction of three strengths (cartridges) with mean IGF-I above +4 SD, two between +3 and +4, one between +2 and +3, none between −2 and +2, and an increase of one with a value below −2 SD [62]. When switching, the first LAGH dose is given the day after the last daily dose (at least 8 h later), and between LAGHs the new dose is given 7 days after the last dose of the previous product [59, 61]. Because IGF-I varies throughout the week, correction factors have been published to estimate mean weekly IGF-I from samples taken outside day 4: for example, sampling on day 2 underestimates the mean by 0.67 SD with lonapegsomatropin, 1.0 SD with somapacitan and 0.7 SD with somatrogon [59].

The 2026 GRS consensus concludes that LAGHs have growth efficacy comparable to daily GH at the recommended doses in the first years (grade A), without major new safety signals, with injection-site reactions as the most frequent adverse effect (5–42% with LAGH and 6–25% with daily GH in pivotal paediatric trials) and anti-drug antibodies without a demonstrated impact on efficacy [24]. It recommends starting with the labelled dose and titrating (grade A), using ideal weight in obese patients and recognising that labelled doses may be excessive in severe deficiency or panhypopituitarism, which require low daily doses [24]. The most relevant point for severe genetic IGHD concerns infants and very young children, particularly with a history of hypoglycaemia or multiple deficiencies: fasting hypoglycaemia at the end of the dosing interval should be monitored, with explicit instructions to the family, because the trough occurs 1 to 2 days before each injection and few patients under 3 years entered the trials [24, 59]. Adult trials underrepresented IGHD, childhood-onset deficiency and genetic causes [24]. LAGH may be preferred when adherence, injection burden or anxiety are relevant (alternating carers, several homes, neurodivergence, polypharmacy, travel), but does not alleviate needle phobia (grade C) [24]. In persistent non-adherence, administration by third parties (school or community nursing) is an alternative, and recording the time and site of each injection helps rotate sites [59]. Reducing the number of injections does not guarantee adherence: in a Chinese cohort of 8621 children, LAGH was associated with modestly better adherence in the first 1 to 2 years [24]. In a nationwide New Zealand study, two-thirds of patients on daily GH missed more than one dose per week, and non-adherence was associated with a progressive fall in growth velocity [59].

7.4. Comparative regulatory status

National availability of these medicines in Portugal and Brazil was not studied in the sources consulted, except for somatrogon approval in Brazil reported by the GRS [24].

7.5. Main trials

Pivotal LAGH trials compared each molecule with daily somatropin in prepubertal children with GH deficiency, mostly idiopathic, and did not stratify by genetic cause; their results apply to genetic IGHD by extrapolation from the mechanism [24]. There are no direct comparisons between LAGHs, and differences between trials (comparator doses of 0.034 versus 0.24 mg/kg/week, age, proportion of boys) preclude indirect comparisons; the superiority of lonapegsomatropin in heiGHt was 0.9 cm/year for on-treatment velocities of 10–11 cm/year, and smaller differences are attributed to baseline characteristics, dose and IGF-I [24, 61]. The adult somatrogon trial was negative, as was the paediatric macimorelin study, which should be taken into account when interpreting indication extensions [52, 66]. Mecasermin data come from mostly open-label studies without a control group, with samples decreasing during follow-up (13 patients at year 8) [25].

7.6. Particular features by genetic subtype

In type IA, treatment starts with somatropin, but the response should be monitored closely because many patients develop neutralising antibodies that abolish growth [5, 38]. Neutralising anti-GH antibodies often disappear after discontinuation or a change of preparation, but in GH1 deletion with neutralising antibodies, mecasermin is the alternative, labelled in the USA and off-label in the EU [25, 26, 33]. In the mecasermin trial, 11% of patients had GH gene deletion [25].

In type IV (GHRHR), patients retain immunological tolerance and do not usually develop neutralising antibodies; GH is effective, and in Itabaianinha two children treated for one year had a normal response [5, 31]. If a child with a GHRHR genotype stops growing, the explanation should be sought in adherence, dose or diagnosis [5]. Monitoring includes growth velocity, IGF-I, thyroid function and glucose tolerance, with reassessment at transition [5].

In type II, the response to GH is good, but thyroid, adrenal and gonadal function should be reassessed periodically because additional deficiencies may develop [5, 19]. Correction of GH1 splicing (exon 3 retention) with antisense oligonucleotides and small molecules has been studied in cellular and animal models; this is preclinical research, and no gene or splicing therapy has marketing authorisation [5].

In type III with BTK, GH deficiency is treated as in the other forms, and the immunodeficiency follows the management of X-linked agammaglobulinaemia [21]. In forms due to SOX3, HESX1, GLI2 and POU1F1, GH is combined with surveillance of the other axes, and in POU1F1 levothyroxine is titrated by free T4 in the upper half of the reference range, not by TSH [5, 68]. In GHSR deficiency, the response to GH has not been studied in specific trials in the sources consulted.

7.7. Mecasermin

Mecasermin (Increlex®) is recombinant human IGF-I [25]. In the USA, it is indicated in children aged 2 years or older with severe primary IGF-I deficiency (height and basal IGF-I at or below −3.0 SD with normal or high GH) or with GH gene deletion and neutralising antibodies; in the EU, from 2 to 18 years with confirmed severe primary IGF-I deficiency, which includes variants in GHR, the post-receptor pathway and IGF1 [25, 26]. It does not replace GH in its indications and is not indicated in secondary IGF-I deficiency (GH deficiency, malnutrition, hypothyroidism, chronic corticosteroid therapy) [25]. Somatropin is ineffective in complete insensitivity; high-dose GH produces only a slight and transient increase in IGF-I, and combining GH with IGF-I may be useful in partial insensitivity [5, 33].

Sources: FDA label and European SmPC [25, 26].

Hypoglycaemia is the main adverse effect: in the 71 patients in the clinical programme, it occurred in 42% (30), mostly mild or moderate, with five severe episodes and four with seizures or loss of consciousness, more frequently in the first month and in younger children; 14 of the 30 had previous hypoglycaemia [25]. The March 2024 label update strengthened the warning about severe hypoglycaemia with seizures [67]. A narrative review reports hypoglycaemia in 8%, a figure differing from the label because of different definitions and populations; the label figure should guide counselling [25, 33]. Other effects include tonsillar hypertrophy (15%, with seven tonsillectomies), intracranial hypertension (ICH, three cases in the programme; approximately 5% in a review), increased transaminases and lactate dehydrogenase, rapid renal and splenic growth in the first years, slipped capital femoral epiphysis, scoliosis, lipohypertrophy and post-marketing malignancies; bone age advanced by 5.3 ± 3.4 years over 4.9 ± 3.4 chronological years [25, 33]. Contraindications in the USA include hypersensitivity, closed epiphyses and active or previous malignancy, and benzyl alcohol in the excipient requires caution in infants [25]. Fundoscopy is performed at baseline and periodically [25]. Final height with mecasermin generally remains below the 3rd percentile, nearly normal when started early [33].

7.8. Replacement of the other axes in combined deficiency

Modified-release and dual-release hydrocortisone preparations have been studied in hypopituitarism, and the 2016 Endocrine Society guidelines on hormonal replacement in adult hypopituitarism and the 2018 European Thyroid Association (ETA) guidelines on central hypothyroidism are the references for these axes [19, 68, 71].

7.9. GH safety, contraindications and monitoring

GH contraindications in the US label include acute critical illness after cardiac or abdominal surgery, multiple trauma or acute respiratory failure, hypersensitivity, closed epiphyses when the goal is growth, active malignancy, active proliferative or severe non-proliferative diabetic retinopathy and Prader–Willi syndrome with severe obesity or respiratory impairment [57]. The European somapacitan SmPC requires intracranial tumours to be inactive and antitumour therapy to be complete [60]. Class warnings are ICH (symptoms usually in the first 8 weeks, reversible with discontinuation or reduction; fundoscopy before treatment and periodically), glucose intolerance, fluid retention, unmasking of hypoadrenalism and hypothyroidism, slipped capital femoral epiphysis (limping, hip or knee pain), scoliosis progression, pancreatitis, lipohypertrophy or lipoatrophy, neoplasm progression and second neoplasms in irradiated survivors [57]. Adverse effects of GH in children are infrequent, and no upper IGF-I limit has been associated with adverse effects [22]. There were no confirmed cases of ICH or slipped capital femoral epiphysis in paediatric LAGH trials (grade A), but the small samples do not exclude rare events [24]. Lipoatrophy has been described with several LAGHs and daily GH, requiring site rotation [24]. In adults, adverse effects relate mainly to fluid retention, occur at initiation and on dose increases, and respond to reduction; obese and older patients are more susceptible [44].

In cancer survivors, who were not included in paediatric LAGH trials, daily GH rules apply: in most childhood cancer survivors, start at least one year after treatment ends, with imaging confirming tumour absence or stability; in adults, the guidelines specify five disease-free years, with an individual decision involving the oncologist [24, 44]. AACE contraindicates GH with active neoplasia (except basal and squamous cell skin carcinomas) and advises caution with a strong family history of cancer [44].

7.10. End of growth and transition to adulthood

Paediatric treatment is discontinued on reaching final or near-final height, defined in the European SmPC by a velocity below 2 cm/year and bone age above 14 years in girls and 16 years in boys; the lonapegsomatropin SmPC adds discontinuation when velocity SD is below +1 after the first year [60, 62, 64]. Once the epiphyses have fused, the need is reassessed, and if deficiency persists GH continues at adult doses to complete somatic development (lean mass and bone mineralisation) [60]. After skeletal maturation (bone age approximately 15 years in girls and 17 in boys), there is no height benefit, and continuation targets bone mineral density, body composition, strength and quality of life [50].

The distinguishing feature of genetic IGHD is exemption from retesting. The Endocrine Society recommends retesting childhood-onset deficiency before adult GH, except when there are known mutations, embryopathic lesions with multiple deficiencies or irreversible structural lesions (strong recommendation, high-quality evidence), and considers low IGF-I at least one month after discontinuing GH sufficient in patients with a proven genetic cause [43]. AACE exempts patients with genetic defects of the hypothalamic–pituitary axes from retesting, allowing GH to continue without interruption, and recommends retesting most other patients, particularly idiopathic IGHD with IGF-I below 0 SD, at least one month after discontinuing GH; it dispenses with retesting and therapy in idiopathic IGHD with IGF-I at or above 0 SD [44]. The review by Yuen and colleagues lists IGHD with an identified mutation and transcription factor mutations among patients not requiring retesting, and recalls that 25–100% of patients with childhood-onset deficiency normalise on retesting, particularly IGHD with a normal or small pituitary [50]. This contrast justifies genetic testing before transition in patients with severe, early IGHD.

The 2005 European Society for Paediatric Endocrinology (ESPE) consensus on the GH-treated adolescent at transition, prepared at a workshop in Manchester in December 2003 with paediatricians, adult endocrinologists and industry representatives, was accessible only as an abstract, and its recommendations and thresholds are not reproduced here [72]. AACE emphasises that transition is a vulnerable period in which adolescents drop out of follow-up, and recommends that the paediatrician start counselling early and collaborate with the adult endocrinologist [44].

7.11. Pregnancy, lactation and fertility

Studies support GH use when seeking fertility, and continuation during pregnancy does not appear to affect the mother or fetus, but AACE does not recommend routine use for conception or continuation during pregnancy in women with GH deficiency because of the lack of safety data [44]. In the second half of pregnancy, placental variant GH replaces pituitary GH in the maternal circulation [4]. In Sindh, men and women with GHRHR deficiency were fertile without treatment [2]. In PROP1-related CPHD, fertility is possible with gonadotrophins; two women achieved pregnancy after ovulation induction without GH and were unable to breastfeed, and PRL deficiency in POU1F1 likewise predicts inability to breastfeed [5, 19]. Oral oestrogen increases the required GH dose [43].

7.12. Perioperative care and agents to avoid

In CPHD with ACTH deficiency, surgery or significant illness requires parenteral hydrocortisone (50–100 mg) with fluids, and minor illness requires double or triple the dose until resolution [19]. GH is contraindicated in acute critical illness after cardiac or abdominal surgery [57]. Levothyroxine should not be started before treating adrenal insufficiency [19]. Clomifene does not induce ovulation in hypogonadotrophic hypogonadism because it requires an intact pituitary [19]. In the macimorelin test, strong CYP3A4 inducers should be avoided [51]. LAGH interactions with glucocorticoid and thyroxine doses have not been studied [59].

7.13. Developments in the last 12 months and research

LUM-201 had already been studied in phase 2 (OraGrowtH210, 104 children); the disclosed results appear only in press releases and are not used here [76]. In 2019, the GRS considered oral ghrelin agonists unlikely to be effective in severe pituitary deficiency, with potential in hypothalamic or mild deficiency, limiting the applicability of these drugs to severe genetic IGHD [22]. No European decision extending lonapegsomatropin to adults or any gene therapy or new specific therapy for types IA or II was identified in the sources consulted; these negative searches do not demonstrate the absence of developments.

7.14. Guideline recommendations and divergences

The most relevant divergences are four. The first is the paediatric cut-off: the 2019 GRS suggests approximately 7 ng/mL with current assays, the Delphi panel accepts below 5 µg/L as severe deficiency, and the tradition of 10 µg/L persists in some centres, while the PROP1 GeneReviews contains a unit error [19, 22, 49, 50]. The second is priming: limited by the GRS to adolescents with suspected constitutional delay and supported more broadly by the PES [22, 50]. The third is the paediatric dose, higher in the USA than in Europe [22]. The fourth is regulatory: mecasermin in GH1 deletion with antibodies, LAGHs below 3 years and in adults, and somapacitan indications without GH deficiency exist in the USA and not in the EU [25, 26, 57, 60–62].

8. Reproductive Options and Family Counselling

8.1. Recurrence risks by form

8.2. Family investigation and cascade testing

Parental testing confirms the phase of variants in recessive genes and de novo origin in dominant genes. When only one parent of a patient with recessive disease is a carrier, the possibilities are a de novo variant in the proband or postzygotic parental mosaicism [19]. In type II, parents of an apparently de novo case may be mosaics undetected by Sanger sequencing, and family investigation should favour NGS [9]. In younger siblings of a patient with PROP1-related CPHD, testing enables early treatment; without testing, growth is monitored [19]. The same reasoning applies to siblings of patients with recessive IGHD and children of patients with type II IGHD, in whom early detection prevents height loss and, in forms at risk of additional deficiencies, unrecognised adrenal insufficiency [5, 19]. Variable expressivity within families (in Huang's series, carrier relatives with heights between −1.0 and −7.89 SD, including carriers of deletions of controversial interpretation) requires carrier relatives to be assessed rather than merely counting affected individuals [10]. Carrier testing and preconception counselling are indicated in families with known recessive variants, particularly with consanguinity and in populations with founder variants [19, 40].

8.3. Prenatal and preimplantation diagnosis

With the familial variant identified, prenatal diagnosis and PGT are technically possible [19, 47]. The decision involves the severity of the form concerned and treatment efficacy: IGHD responds to GH in most subtypes and, in the Itabaianinha cohort, untreated affected individuals had normal quality of life and longevity, whereas CPHD with a risk of neonatal adrenal crisis and syndromic forms (OTX2, SOX2, biallelic LHX4) have different implications [3, 5]. Counselling should present these data without converting them into a recommendation.

8.4. Counselling of minors and counselling content

Testing at-risk children has direct benefit because early GH treatment improves final height and because identifying forms at risk of additional deficiencies guides surveillance [5, 19]. Counselling should explain the inheritance pattern, the possibility of mosaicism and incomplete penetrance, possible progression to multiple deficiencies (type II, SOX3, HESX1, GLI2), exemption from retesting at transition when the genetic cause is established and the need for lifelong endocrine follow-up in progressive forms [5, 9, 43]. For heterozygous GHRHR carriers, the available information indicates normal stature [31].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Prevalence and what each number measures

There is no published population frequency for GHRHR-related IGHD or a reliable gene-specific figure for GH1 [5]. The absence of data in the sources consulted does not demonstrate that none exist.

9.2. Founder populations and geography

Recessive IGHD is concentrated in populations with consanguinity and founder effects. The GHRHR p.(Glu72*) variant was found in families from Bombay, Sindh, Delft Island and Pakistan, with a common ancestor estimated approximately 1350–2700 years ago, and is the main cause of familial IGHD in western India (17 patients from one Muslim family and four Hindu families with identical homozygous haplotypes) [2, 5, 40]. Families with GHRHR variants have also been described in Sri Lanka, Brazil, the USA, Spain and Japan [4, 5]. In France, GENHYPOPIT showed that GHRHR variants may occur in compound heterozygosity without consanguinity [9]. For CPHD, PROP1 is more frequent in Eastern Europe and Russia, with founder c.301_302del in Lithuania and the Hutterites, possibly founder c.112_124del in the Indian subcontinent, and targeted testing recommended in the Hutterites, India, Eastern Europe and the Iberian Peninsula [19, 37]. POU1F1 is relatively prevalent in Indian cohorts [5].

9.3. Brazil and Portugal

Brazil has two central contributions. The Itabaianinha kindred in Sergipe is the largest described cohort of single-gene IGHD: 105 individuals across eight generations homozygous for GHRHR c.57+1G>A, most never treated, studied since 1994 [3, 31]. In CPHD, PROP1 explains approximately 53% (16/30) of cases in a Brazilian series, and the PROP1 GeneReviews is authored by researchers from the University of São Paulo [19]. Somatrogon is approved in Brazil [24]. The sources consulted contain no Portuguese data on genetic IGHD, CPHD or access to medicines.

9.4. Special populations

Greulich–Pyle bone age standards derive from white North American children of Northern European ancestry, limiting their application to other populations [22]. Girls with short stature have historically been underinvestigated [22]. Obesity reduces peak GH in tests and may advance bone age, requiring BMI-based cut-offs in adults and ideal-weight dosing with LAGH [24, 44, 50]. Infants with severe deficiency are the group at greatest risk of hypoglycaemia with LAGH [24]. Women taking oral oestrogen need higher GH doses [43, 60]. Very young children (under 3 years) were poorly represented in LAGH trials, and in the EU there is no recommended posology below 3 years [59, 60, 62].

9.5. European networks, registries and patient associations

Endo-ERN (https://endo-ern.eu/) is the European Reference Network for rare endocrine diseases, with thematic groups for growth and genetic obesity (MTG5) and hypothalamic and pituitary conditions (MTG6) [27]. GloBE-Reg (https://globe-reg.net) is an international product-independent registry for GH and LAGH, and the GRS recommends enrolling all patients on LAGH in a registry [24, 80]. Patient associations include the MAGIC Foundation (USA; https://www.magicfoundation.org/), the Human Growth Foundation (USA; https://www.hgfound.org/) and the Child Growth Foundation (United Kingdom; https://childgrowthfoundation.org/), as well as the Pituitary Foundation (United Kingdom) and, for septo-optic dysplasia, FOCUS Families [19, 45, 81–83].

10. Illustrative Clinical Case

In 2024, Huang and colleagues described six Chinese children from Fujian province with GH1 variants and type II IGHD, identified by exome sequencing in a cohort of 792 children with short stature [10]. Patient 3 illustrates the de novo dominant form. He was a boy born at term by caesarean section, weighing 3500 g and measuring 53 cm, assessed at 4.1 years with a height of 87.1 cm (−4.51 SD), weight of 13.5 kg (+1.17 SD) and BMI of 17.8 [10]. Bone age was 3.08 years, peak GH on stimulation tests was 0.29 ng/mL, IGF-I was below 25 ng/mL (below −2.39 SD) and IGFBP-3 was 0.77 mg/L; pituitary MRI was normal [10]. The mother measured 156 cm (−0.85 SD) and the father 177 cm (+1.4 SD), making the short stature disproportionate to the familial potential [10].

Exome sequencing identified the heterozygous variant NM_000515.5(GH1):c.291+1G>A, at the intron 3 splice donor, absent in both parents by Sanger sequencing, classified as pathogenic (ACMG criteria PVS1, PM1, PP3 and PP4) [10, 36]. Figure 2C of the article (https://pmc.ncbi.nlm.nih.gov/articles/PMC11491352/figure/f2/) shows the pedigree, with the proband at −4.51 SD, the father at +1.4 SD and the mother at −0.85 SD, and chromatograms with the heterozygous peak only in the proband [10]. The authors annotated the protein consequence as “p.Glu58Asnfs*120”, but the mechanism described for variants at this position is in-frame exon 3 skipping, producing the 17.5 kDa isoform with a dominant-negative effect [4, 9, 10]. The same variant was the most frequent among patients with GH1 in the French GENHYPOPIT cohort, where three of four cases studied were de novo and the fourth resulted from maternal mosaicism [9].

The child started somatropin at 4.1 years at 0.10 IU/kg/day, a dose maintained during the first year: at 6 months he measured 96.4 cm (−2.98 SD), with IGF-I of 267 ng/mL, and after one year 102.8 cm (−2.12 SD), with IGF-I of 211 ng/mL and IGFBP-3 of 3.86 mg/L, a gain of 2.37 SD according to the authors; Figure 1C (https://pmc.ncbi.nlm.nih.gov/articles/PMC11491352/figure/f1/) shows this curve [10]. The authors reported no changes in safety laboratory tests [10]. By the international standard (1 mg = 3 IU), the dose corresponds to approximately 33 µg/kg/day, above the European initial dose of 25 µg/kg/day and below the US upper limit of 43 µg/kg/day [22, 50]. The report does not describe surveillance of the other pituitary axes; with a splicing variant causing severe, early deficiency, the sources recommend lifelong endocrine follow-up because of the risk of progression to CPHD, and parental NGS testing to exclude mosaicism before considering the recurrence risk low [5, 9].

Patient 1 in the same series completes the picture of treated history: with a de novo heterozygous deletion of GH1 exons 2–5, he started GH at 3.8 years, reached −0.19 SD after 2.3 years, stopped for 3.25 years, restarted at 9.4 years and, at 16 years, measured 175.0 cm (+0.54 SD), above his target height of 169.5 cm [10]. However, the interpretation of heterozygous GH1 deletions as a cause of type II IGHD is disputed (section 2.8) [9].

11. Prognosis and Long-Term Follow-Up

11.1. Prognosis

Without treatment, severe genetic IGHD leads to marked adult short stature: in Sindh, 130 cm in men and 113.5 cm in women, and in Itabaianinha −7.35 SD [2, 31]. The Itabaianinha cohort shows that isolated lifelong GH deficiency due to GHRHR is compatible with normal quality of life and longevity, but with central obesity, reduced muscle mass and skin changes [3]. This prognosis does not apply to forms with ACTH deficiency: in septo-optic dysplasia, 5 of 88 patients died (median 12.4 years), three from adrenal crisis, and in Pallister–Hall syndrome hypoadrenalism is the main cause of neonatal death [34, 55].

With treatment, the response is good in types IB, II and IV and developmental forms, particularly when GH is started early; in PROP1-related CPHD, early initiation allows the expected familial height to be reached [10, 19]. In type IA, prognosis depends on antibody development; with mecasermin, velocity rises from 2.8 to 8.0 cm/year in the first year and decreases thereafter, and final height generally remains below the 3rd percentile [25, 33]. In type II, long-term prognosis is influenced by possible progression to ACTH, TSH and gonadotrophin deficiencies [5, 39]. In adulthood, persistent GH deficiency is associated with poorer bone mineral density, body composition and cardiovascular risk markers, and continuing GH at transition improves body composition, bone health, quality of life and lipid metabolism [43, 44].

11.2. Surveillance

The frequency of reassessment for additional deficiencies in type II IGHD and developmental forms is not defined by a guideline; the sources recommend lifelong endocrine follow-up, without discharge on reaching final height [5, 39].

12. Summary and Final Considerations

Genetic isolated GH deficiency accounts for a small fraction of childhood GH deficiency, but has clinical consequences that justify its identification. The four classical types correspond to distinct mechanisms: complete absence of GH due to biallelic GH1 deletion in type IA, with a risk of neutralising antibodies; reduced production in type IB; interrupted hypothalamic signalling due to biallelic GHRHR variants, now designated type IV in OMIM and ClinVar; the dominant-negative effect of the 17.5 kDa isoform in type II; and X-linked forms, with BTK and agammaglobulinaemia or with SOX3 and neurohypophyseal abnormalities. These are supplemented by GHSR deficiency, with inheritance still uncertain, and Kowarski bioinactive GH.

Diagnosis remains clinical. It rests on auxology, IGF-I and IGFBP-3, two stimulation tests (one suffices with high probability, none in a neonate with a typical presentation), with assay-dependent cut-offs, and MRI, which does not exclude GH1 and GHRHR forms when normal. Genetic testing, preferably a panel with copy-number analysis, finds a cause in approximately one-tenth of IGHD and more than one-third of severe familial forms. Its limitations (GH1 deletions invisible to simple sequencing, splicing variants requiring RNA analysis, parental mosaicism, heterozygosity in recessive genes) make a negative result poorly informative. Identifying a variant guides surveillance for evolving deficiencies in type II and developmental forms, explains loss of response in type IA and dispenses with retesting at transition.

Treatment uses daily somatropin or long-acting GH, with comparable efficacy in paediatric trials, caution regarding trough hypoglycaemia with LAGHs in infants with severe deficiency and dose and IGF-I sampling rules specific to each molecule. Regulatory differences between the USA and EU have practical implications: minimum ages, adult indications, indications without GH deficiency and mecasermin in GH1 deletion with antibodies, labelled only in the USA. Macimorelin is unsuitable for paediatric diagnosis. In forms with multiple deficiencies, cortisol replacement before thyroxine and education on stress dosing prevent the main cause of death. The Itabaianinha kindred and Sindh families show untreated natural history and the role of founder populations, and the French GENHYPOPIT cohort shows genetic heterogeneity outside them. Gene-specific population data, Portuguese data in the sources consulted and trials stratified by genetic cause are lacking.

13. References

1. Orphanet. Non-acquired isolated growth hormone deficiency (ORPHA:631). https://www.orpha.net/en/disease/detail/631
2. Maheshwari HG, Silverman BL, Dupuis J, Baumann G. Phenotype 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. 1998;83(11):4065-74. doi:10.1210/jcem.83.11.5226.
3. Aguiar-Oliveira MH, Salvatori R. Disruption of the GHRH receptor and its impact on children and adults: the Itabaianinha syndrome. Rev Endocr Metab Disord. 2021;22(1):81-89. doi:10.1007/s11154-020-09591-4.
4. Mullis PE. Genetics of isolated growth hormone deficiency. J Clin Res Pediatr Endocrinol. 2010;2(2):52-62. doi:10.4274/jcrpe.v2i2.52.
5. GeneGuidance. Gene records for GH1, GHRHR, PROP1, POU1F1, HESX1, LHX3, LHX4, SOX3, OTX2, GHR, STAT5B, IGF1, IGFALS and BTK: identifiers, clinical curation, frequency, treatment and gene–disease validity. Accessed 10 Oct 2026.
6. Orphanet. Isolated growth hormone deficiency type IA (ORPHA:231662). https://www.orpha.net/en/disease/detail/231662
7. Orphanet. Isolated growth hormone deficiency type IV (ORPHA:684247). https://www.orpha.net/en/disease/detail/684247
8. Orphanet. Short stature due to GHSR deficiency (ORPHA:314811). https://www.orpha.net/en/disease/detail/314811
9. Aouchiche K, Romanet P, Charnay T, Barlier A, Roche C, Grunenwald S, et al. Unraveling the genetic heterogeneity of isolated growth hormone deficiency: insights from the GENHYPOPIT cohort. Horm Res Paediatr. 2026:1-12. doi:10.1159/000551700.
10. Huang X, Chen H, Shangguan H, Wu W, Ai Z, Chen Z, et al. The clinical and genetic aspects of six individuals with GH1 variants and isolated growth hormone deficiency type II. Front Endocrinol (Lausanne). 2024;15:1363050. doi:10.3389/fendo.2024.1363050.
11. Orphanet. Isolated growth hormone deficiency type IB (ORPHA:231671). https://www.orpha.net/en/disease/detail/231671
12. Orphanet. Isolated growth hormone deficiency type II (ORPHA:231679). https://www.orpha.net/en/disease/detail/231679
13. Orphanet. Isolated growth hormone deficiency type III (ORPHA:231692). https://www.orpha.net/en/disease/detail/231692
14. Orphanet. Short stature due to isolated growth hormone deficiency with X-linked hypogammaglobulinemia (ORPHA:632). https://www.orpha.net/en/disease/detail/632
15. Orphanet. Cataract-growth hormone deficiency-sensory neuropathy-sensorineural hearing loss-skeletal dysplasia syndrome (ORPHA:436174). https://www.orpha.net/en/disease/detail/436174
16. Orphanet. Non-acquired combined pituitary hormone deficiency (ORPHA:467). https://www.orpha.net/en/disease/detail/467
17. Orphanet. Septo-optic dysplasia spectrum (ORPHA:3157). https://www.orpha.net/en/disease/detail/3157
18. Orphanet. Laron syndrome (ORPHA:633). https://www.orpha.net/en/disease/detail/633
19. Carvalho LR, Nishi MY, Correa FA, Marques JM, Arnhold IJP, Mendonca BB. PROP1-Related Combined Pituitary Hormone Deficiency. 2000 Dec 7 [updated 2022 Feb 24]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1347/
20. Smith CIE, Berglöf A. X-Linked Agammaglobulinemia. 2001 Apr 5 [updated 2024 Jun 27]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1453/
21. Fleisher TA, White RM, Broder S, Nissley SP, Blaese RM, Mulvihill JJ, et al. X-linked hypogammaglobulinemia and isolated growth hormone deficiency. N Engl J Med. 1980;302(26):1429-34. doi:10.1056/NEJM198006263022601.
22. Collett-Solberg PF, Ambler G, Backeljauw PF, Bidlingmaier M, Biller BMK, Boguszewski MCS, et al. Diagnosis, genetics, and therapy of short stature in children: a Growth Hormone Research Society international perspective. Horm Res Paediatr. 2019;92(1):1-14. doi:10.1159/000502231.
23. Pfizer. GENOTROPIN (somatropin) for injection. Prescribing information. DailyMed set id ffebf88b-d257-4542-9808-74d9b7167765. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=ffebf88b-d257-4542-9808-74d9b7167765
24. Schilbach K, Clayton P, Agrawal N, Arlien-Søborg MC, Backeljauw PF, Biller BMK, et al. Long-acting growth hormone: an updated Growth Hormone Research Society consensus statement. Eur J Endocrinol. 2026;195(3):R73-R96. doi:10.1093/ejendo/lvag167.
25. Eton Pharmaceuticals, Inc. INCRELEX (mecasermin) injection, for subcutaneous use. Prescribing information. DailyMed set id d5b2b0e1-c523-468d-9a0b-4ae4e4a8dbce. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=d5b2b0e1-c523-468d-9a0b-4ae4e4a8dbce
26. European Medicines Agency. Increlex (mecasermin): EPAR — product information. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/increlex
27. European Reference Network on Rare Endocrine Conditions (Endo-ERN). Available from: https://endo-ern.eu/
28. Phillips JA 3rd, Hjelle BL, Seeburg PH, Zachmann M. Molecular basis for familial isolated growth hormone deficiency. Proc Natl Acad Sci U S A. 1981;78(10):6372-5. doi:10.1073/pnas.78.10.6372.
29. Duriez B, Duquesnoy P, Dastot F, Bougnères P, Amselem S, Goossens M. An exon-skipping mutation in the btk gene of a patient with X-linked agammaglobulinemia and isolated growth hormone deficiency. FEBS Lett. 1994;346(2-3):165-70. doi:10.1016/0014-5793(94)00457-9.
30. Wajnrajch MP, Gertner JM, Harbison MD, Chua SC Jr, Leibel RL. Nonsense mutation in the human growth hormone-releasing hormone receptor causes growth failure analogous to the little (lit) mouse. Nat Genet. 1996;12(1):88-90. doi:10.1038/ng0196-88.
31. Hayashida CY, Gondo RG, Ferrari C, Toledo SP, Salvatori R, Levine MA, et al. Familial growth hormone deficiency with mutated GHRH receptor gene: clinical and hormonal findings in homozygous and heterozygous individuals from Itabaianinha. Eur J Endocrinol. 2000;142(6):557-63. doi:10.1530/eje.0.1420557.
32. Pantel J, Legendre M, Cabrol S, Hilal L, Hajaji Y, Morisset S, et al. Loss of constitutive activity of the growth hormone secretagogue receptor in familial short stature. J Clin Invest. 2006;116(3):760-8. doi:10.1172/JCI25303.
33. Mastromauro C, Giannini C, Chiarelli F. Short stature related to growth hormone insensitivity (GHI) in childhood. Front Endocrinol (Lausanne). 2023;14:1141039. doi:10.3389/fendo.2023.1141039.
34. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 880–881 (Septo-optic dysplasia sequence).
35. Alatzoglou KS, Webb EA, Le Tissier P, Dattani MT. Isolated growth hormone deficiency (GHD) in childhood and adolescence: recent advances. Endocr Rev. 2014;35(3):376-432. doi:10.1210/er.2013-1067.
36. National Center for Biotechnology Information. ClinVar. Variation IDs 15970, 15971, 15968, 15983 (GH1); 15989, 15990, 3720730 (GHRHR); 7632 (GHSR). Available from: https://www.ncbi.nlm.nih.gov/clinvar/
37. Fang Q, George AS, Brinkmeier ML, Mortensen AH, Gergics P, Cheung LY, et al. Genetics of combined pituitary hormone deficiency: roadmap into the genome era. Endocr Rev. 2016;37(6):636-675. doi:10.1210/er.2016-1101.
38. Cogan JD, Phillips JA 3rd. GH1 gene deletions and IGHD type 1A. Pediatr Endocrinol Rev. 2006;3 Suppl 3:480-8. PMID: 17551470.
39. Alatzoglou KS, Kular D, Dattani MT. Autosomal dominant growth hormone deficiency (type II). Pediatr Endocrinol Rev. 2015;12(4):347-55. PMID: 26182479.
40. Ahmad S, Ali MZ, Abbasi SW, Abbas S, Ahmed I, Abbas S, et al. A GHRHR founder mutation causes isolated growth hormone deficiency type IV in a consanguineous Pakistani family. Front Endocrinol (Lausanne). 2023;14:1066182. doi:10.3389/fendo.2023.1066182.
41. Alatzoglou KS, Turton JP, Kelberman D, Clayton PE, Mehta A, Buchanan C, et al. Expanding the spectrum of mutations in GH1 and GHRHR: genetic screening in a large cohort of patients with congenital isolated growth hormone deficiency. J Clin Endocrinol Metab. 2009;94(9):3191-9. doi:10.1210/jc.2008-2783.
42. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 182–185 (Hypoglycaemia in neonate and infant).
43. Molitch ME, Clemmons DR, Malozowski S, Merriam GR, Vance ML; Endocrine Society. Evaluation and treatment of adult growth hormone deficiency: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2011;96(6):1587-609. doi:10.1210/jc.2011-0179.
44. Yuen KCJ, Biller BMK, Radovick S, Carmichael JD, Jasim S, Pantalone KM, et al. American Association of Clinical Endocrinologists and American College of Endocrinology guidelines for management of growth hormone deficiency in adults and patients transitioning from pediatric to adult care. Endocr Pract. 2019;25(11):1191-1232. doi:10.4158/GL-2019-0405.
45. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 270–271 (Optic nerve hypoplasia).
46. Morris AAM. Disorders of Ketogenesis and Ketolysis. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 13 (13.4.3). doi:10.1007/978-3-662-63123-2_13
47. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 312–314 (Short stature).
48. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 320–323 (Structural intracranial anomalies: septo-optic dysplasia).
49. Arlien-Søborg MC, Radovick S, Boguszewski MCS, Bidlingmaier M, Johannsson G, Grimberg A, et al. Consensus and controversies about diagnosing GH deficiency: a Delphi survey by the GH research society. Pituitary. 2025;28(3):57. doi:10.1007/s11102-025-01526-z. Correction: doi:10.1007/s11102-026-01681-x (27 Jul 2026).
50. Yuen KCJ, Johannsson G, Ho KKY, Miller BS, Bergada I, Rogol AD. Diagnosis and testing for growth hormone deficiency across the ages: a global view of the accuracy, caveats, and cut-offs for diagnosis. Endocr Connect. 2023;12(7):e220504. doi:10.1530/EC-22-0504.
51. Aeterna Zentaris GmbH. MACRILEN (macimorelin) for oral solution. Prescribing information. DailyMed set id aeb30f1a-4815-47ae-a2cb-b6b4e26372e1. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=aeb30f1a-4815-47ae-a2cb-b6b4e26372e1
52. U.S. Food and Drug Administration. Supplement approval letter NDA 205598/S-005 (Macrilen) and label. 23 Jul 2026. Available from: https://www.accessdata.fda.gov/drugsatfda_docs/appletter/2026/205598Orig1s005ltr.pdf ; label: https://www.accessdata.fda.gov/drugsatfda_docs/label/2026/205598s005lbl.pdf
53. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 112–113 (Silver-Russell syndrome).
54. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 236–239 (Microphthalmia and anophthalmia).
55. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 248–249 (Pallister-Hall syndrome).
56. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 116–119 (Congenital hypothyroidism).
57. Novo Nordisk Inc. SOGROYA (somapacitan-beco) injection, for subcutaneous use. Prescribing information. Revised 02/2026. DailyMed set id eadd03b4-adc3-4b88-8a24-9d205c5b85aa. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=eadd03b4-adc3-4b88-8a24-9d205c5b85aa
58. Novo Nordisk. NORDITROPIN (somatropin) injection. Prescribing information. DailyMed set id 1058e17c-9261-459c-a3e6-fae38d196c14. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=1058e17c-9261-459c-a3e6-fae38d196c14
59. Maniatis A, Cutfield W, Dattani M, Deal C, Collett-Solberg PF, Horikawa R, et al. Long-acting growth hormone therapy in pediatric growth hormone deficiency: a consensus statement. J Clin Endocrinol Metab. 2025;110(4):e1232-e1240. doi:10.1210/clinem/dgae834. Correction: J Clin Endocrinol Metab. 2025;110(5):e1729.
60. European Medicines Agency. Sogroya (somapacitan): EPAR — product information. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/sogroya
61. Ascendis Pharma Endocrinology Division A/S. SKYTROFA (lonapegsomatropin-tcgd) for injection. Prescribing information. Revised 07/2025. DailyMed set id 9229f871-bc61-4ddc-8122-d126d40e36ac. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=9229f871-bc61-4ddc-8122-d126d40e36ac
62. European Medicines Agency. Skytrofa (lonapegsomatropin): EPAR — product information. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/skytrofa
63. Pfizer Inc. NGENLA (somatrogon-ghla) injection, for subcutaneous use. Prescribing information. DailyMed set id 06dce529-b741-4a1b-8842-3ad77be6e308. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=06dce529-b741-4a1b-8842-3ad77be6e308
64. European Medicines Agency. Ngenla (somatrogon): EPAR — product information. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/ngenla
65. U.S. Food and Drug Administration. Supplement approval letter BLA 761156/S-012, S-014, S-015 (Sogroya). 27 Feb 2026. Available from: https://www.accessdata.fda.gov/drugsatfda_docs/appletter/2026/761156Orig1s012,%20761156Orig1s014,%20761156Orig1s015ltr.pdf
66. European Medicines Agency. Questions and answers on the withdrawal of application to change the marketing authorisation for Ngenla (somatrogon) (II-16). EMA/134483/2025. 25 Apr 2025. Available from: https://www.ema.europa.eu/en/medicines/human/variation/ngenla
67. U.S. Food and Drug Administration. Supplement approval letter BLA 021839/S-031 (Increlex). 15 Mar 2024. Available from: https://www.accessdata.fda.gov/drugsatfda_docs/appletter/2024/021839Orig1s031ltr.pdf
68. Persani L, Brabant G, Dattani M, Bonomi M, Feldt-Rasmussen U, Fliers E, et al. 2018 European Thyroid Association (ETA) guidelines on the diagnosis and management of central hypothyroidism. Eur Thyroid J. 2018;7(5):225-237. doi:10.1159/000491388.
69. van Trotsenburg P, Stoupa A, Léger J, Rohrer T, Peters C, Fugazzola L, et al. Congenital Hypothyroidism: A 2020-2021 Consensus Guidelines Update-An ENDO-European Reference Network Initiative Endorsed by the European Society for Pediatric Endocrinology and the European Society for Endocrinology. Thyroid. 2021;31(3):387-419. doi:10.1089/thy.2020.0333. PMID: 33272083. PMCID: PMC8001676.
70. AbbVie Inc. Synthroid (levothyroxine sodium) — prescribing information [Internet]. Revised 2024-02-20 (NDA021402). Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=1e11ad30-1041-4520-10b0-8f9d30d30fcc
71. Fleseriu M, Hashim IA, Karavitaki N, Melmed S, Murad MH, Salvatori R, et al. Hormonal replacement in hypopituitarism in adults: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2016;101(11):3888-3921. doi:10.1210/jc.2016-2118.
72. Clayton PE, Cuneo RC, Juul A, Monson JP, Shalet SM, Tauber M; European Society of Paediatric Endocrinology. Consensus statement on the management of the GH-treated adolescent in the transition to adult care. Eur J Endocrinol. 2005;152(2):165-70. doi:10.1530/eje.1.01829.
73. ClinicalTrials.gov. NCT06948214: Phase 3 efficacy and safety study of daily oral LUM-201 in naïve-to-treatment, prepubertal children with GHD (OraGrowtH). Available from: https://clinicaltrials.gov/study/NCT06948214
74. ClinicalTrials.gov. NCT07264595: Phase 1b/2 study of oral GS3-007a in children with GHD. Available from: https://clinicaltrials.gov/study/NCT07264595
75. ClinicalTrials.gov. NCT07221851: Phase 3 basket trial of once-weekly lonapegsomatropin vs daily somatropin (TS, SHOX-D, SGA, ISS). Available from: https://clinicaltrials.gov/study/NCT07221851
76. ClinicalTrials.gov. NCT04614337: Phase 2 study of LUM-201 in children with GHD (OraGrowtH210). Available from: https://clinicaltrials.gov/study/NCT04614337
77. Grimberg A, DiVall SA, Polychronakos C, Allen DB, Cohen LE, Quintos JB, et al. Guidelines for growth hormone and insulin-like growth factor-I treatment in children and adolescents: growth hormone deficiency, idiopathic short stature, and primary insulin-like growth factor-I deficiency. Horm Res Paediatr. 2016;86(6):361-397. doi:10.1159/000452150.
78. Martari M, Salvatori R. Diseases associated with growth hormone-releasing hormone receptor (GHRHR) mutations. Prog Mol Biol Transl Sci. 2009;88:57-84. doi:10.1016/S1877-1173(09)88003-4.
79. Orphanet. Non-acquired panhypopituitarism (ORPHA:90695). https://www.orpha.net/en/disease/detail/90695
80. Global Registry for Novel Therapies in Rare Bone & Endocrine Conditions (GloBE-Reg). Available from: https://globe-reg.net
81. MAGIC Foundation. Available from: https://www.magicfoundation.org/
82. Human Growth Foundation. Available from: https://www.hgfound.org/
83. Child Growth Foundation. Available from: https://childgrowthfoundation.org/