Martin Savage

Idiopathic short stature is not a diagnosis: A strategy is needed to identify the true pathogenesis in cases of unexplained short stature

Introduction

The designation “Idiopathic short stature” (ISS) was first used in the 1980s as a description for short children who had normal growth hormone (GH) secretion and otherwise unexplained short stature. ISS acquired clinical significance following the introduction of recombinant human (h)GH in 1985 and the approval of GH deficiency (GHD) by the FDA and EMA for hGH replacement therapy. Trials of hGH in ISS subjects were started, guided by collaborations between pharma companies and academic institutions, because patients with the ISS label were shown to have comparable degrees of short stature to GHD patients.

These short children therefore deserved an opportunity to possibly benefit from hGH prescribed in a higher pharmacological dose. Positive data demonstrated efficacy of hGH in the ISS subjects in randomised placebo-controlled trials. Although the growth responses were highly variable, highlighting the heterogeneous nature of these patients, the overall positive responses and absence of safety signals led to successful applications to the FDA and, in 2003, ISS became an approved indication for hGH therapy. Similar applications by pharma companies to the EMA were rejected, largely due to the absence of improvement in quality of life variables. Consequently, ISS is not a reimbursable condition in countries not covered by FDA jurisdictions.

Figure 1. Investigation scheme for children with ISS showing equal status of clinical, endocrine and genetic assessment. Click to view Figure 1

The clinical entity of ISS has acquired scientific respectability, despite the absence of identification of pathogenesis in the majority of patients. Clinicians have continued to use the ISS label for their patients with unexplained short stature and a Consensus Statement on the management of ISS was published in 2008 (1), highlighting the heterogeneity of such patients and giving guidelines for their investigation and management. Patients with familial short stature (FSS), non-familial short stature (NFSS) and constitutional delay of growth and puberty (CDGP) were also included in the definition of ISS.

Since the emergence of molecular biological investigation of patients with growth disorders, which started in the late 1980s and accelerated considerably in the 1990s and through to the 21st century, genetic defects have repeatedly been discovered in children previously labelled as having ISS. Initial genetic studies were mainly in the GH–insulin-like growth factor (IGF)-1 axis, where defects in pituitary development and in genes coding for functional proteins in the pathway of GH action were identified. Examples are monogenic defects in the following genes: GH receptor (GHR); STAT5B; IGF1; IGFALS; IGF1R and PAPPA2 (2). However, recent attention has focused on genes involved in chondrogenesis and growth plate function as a key mechanism regulating normal linear growth (3). Genetic defects in the following growth plate genes have been identified in children carrying the label of ISS: SHOX; ACAN; NPR2; IHH; FGFR3.

In view of these recent discoveries, patients labelled as ISS deserve further investigation in an attempt to define their true pathogenesis. “ISS” is not a diagnosis, but a statement of diagnostic uncertainty. It is now clear that the three disciplines of diagnostic assessment, namely clinical evaluation, endocrine characterisation and genetic investigation, need to have equal status in the hierarchy of the short stature protocol (Figure 1). Clinical assessment is of crucial importance with essential documentation of family history, parental, grandparental and sibling height data, dysmorphology and body disproportion examination, in addition to developmental, feeding and behavioural milestones (4). General paediatric disorders, such as celiac disease and Turner syndrome need to be excluded. Endocrine evaluation should document GH and IGF-1 status. Genetic investigations should centre around the search for monogenic defects with major growth effects, rather than polygenic defects with multiple minor effects.

A working relationship should be established between the clinician and a molecular biologist for discussion of which short patients should undergo genetic tests, which are the appropriate techniques to use and how can the results of the tests be best interpreted. Such a relationship will bring considerable dividends in terms of academic learning, patient care and research opportunities. For example, candidate gene sequencing, indicated when a phenotype suggests a specific genetic diagnosis, is largely being replaced by the hypothesis-free approach of whole-exome sequencing using specific gene panels, which have given diagnostic yields in ~35% of unexplained short stature patients in some reports (5).

ISS is a convenient term for patients with unexplained short stature and is likely to remain in clinical usage. The scheme proposed in Figure 1 gives equality to the three diagnostic disciplines of clinical, endocrine and genetic studies to pursue the primary cause of short stature in unexplained cases. This approach has the realistic chance of clarifying the molecular pathogenesis of known or novel growth disorders. By explaining the basic mechanistic defects, genetic results can guide therapy and thereby enhance progression to optimisation of clinical management.

1. Baron J, Sävendahl L, De Luca F, Dauber A, Phillip M, et al. Short and tall stature: a new paradigm emerges. Nat Rev Endocrinol 2015; 11: 735–746. doi: 10.1038/nrendo.2015.165.

2. Cohen P, Rogol AD, Deal CL, Saenger P, Reiter EO, et al. Consensus statement on the diagnosis and treatment of children with idiopathic short stature: a summary of the Growth Hormone Research Society, the Lawson Wilkins Pediatric Endocrine Society, and the European Society for Paediatric Endocrinology Workshop. J Clin Endocrinol Metab 2008; 93: 4210–4217. doi: 10.1210/jc.2008-0509

3. Storr HL, Chatterjee S, Metherell LA, Foley C, Rosenfeld RG, et al. Nonclassical GH Insensitivity: Characterization of Mild Abnormalities of GH Action. Endocr Rev 2019; 40: 476–505. doi: 10.1210/er.2018-00146.

4. Wit JM, Kamp GA, Oostdijk W. on behalf of the Dutch Working Group on Triage and Diagnosis of Growth Disorders in Children. Towards a Rational and Efficient Diagnostic Approach in Children Referred for Growth Failure to the General Paediatrician. Horm Res Paediatr 2019; 91: 223–240 doi: 10.1159/000499915.

5. Hauer NN, Popp B, Schoeller E, Schuhmann S, Heath KE, et al. Clinical relevance of systematic phenotyping and exome sequencing in patients with short stature. Genet Med 2018; 20: 630–638. doi: 10.1038/gim.2017.159.

Kate Davies

Supporting families of paediatric patients with adrenal disease

Introduction

While adrenal disease in children is quite rare (Simpson et al., 2018), healthcare professionals today working in paediatric endocrinology departments are experts in managing the care of children with adrenal disease. Congenital adrenal hyperplasia (CAH) is the most common; primary adrenal insufficiency and Cushing’s syndrome are also rare but important disorders.

It is useful for families to learn more detail on what is involved in CAH from this helpful video from the patient support group “Living with CAH.”

The diagnosis of CAH may seem incredibly daunting for families, with all the information they are given when their child is diagnosed. Boys tend to present in the emergency room when they are around 2 weeks of age with dehydration and vomiting, whereas girls are usually diagnosed shortly after birth, as their genitalia may look more like a boy’s. This can be frightening for families, and this leaflet on differences of sex development is a comforting, useful resource, giving tips to families on how best to talk to friends, family and hospital professionals, and explanations of words that they may never have heard before.

Families should have a paediatric endocrine nurse specialist (PENS) to help with communication and to coordinate care. Such a person can act as a resource and a liaison between various members of the multidisciplinary team that they may encounter.

It is very important that the prescribed medications are taken correctly and on time, as children may have a life-threatening adrenal crisis if they do not have enough of the hormone cortisol in their system (Miller et al., 2020). Babies will have to take extra liquid salt supplements every day until they are fully weaned (Padidela & Hindmarsh, 2010), and it is important that parents know the correct pharmacy to obtain this from. Fludrocortisone tablets are given to treat reduced levels of aldosterone, and hydrocortisone replacement is also needed for basic cortisol replacement, given at different times throughout the day. The tablets can be crushed and mixed with water to obtain the correct dose, which can be quite small in children (Watson et al., 2018), or there is a paediatric-specific formulation, which comes as capsules that can be broken open and the small grains administered (Porter, Withe, & Ross, 2018).

When children are unwell, hydrocortisone doses will need to be doubled or trebled until clinical improvement occurs. In an emergency, an injection of hydrocortisone needs to be administered, and the PENS can help train the family for this eventuality. A useful app (Tollerfield, Atterbury, Langham, Morris, & Farrell, 2017) formulated by the team at Great Ormond Street Hospital also demonstrates the injection (see image), and the team have other useful information at the following links:

Education in the primary disorder will underpin the understanding of families to have a full stock of the medication at home. The PENS can help liaise with nurseries and schools and agree on a management plan regarding medication and potential emergency scenarios. A steroid therapy card should always be with the child and the PENS can formulate hand-held emergency management plans to be kept at school, with both parents, at grandparents’ houses etc. Medic alert bracelets are also advised to be worn at all times.

Travel advice can also be given by the PENS, focusing on when to take the medication across different time zones, advice for travelling with an emergency kit on an aeroplane, emergency plans in relevant languages, and possibly the contact details of the local paediatric endocrine team (Moloney, Murphy, & Collin, 2015).

It is vital that families and healthcare professionals from a multidisciplinary team work together: the diagnosis can be alarming, but support from other families and support groups can be invaluable (Moloney et al., 2015), with education being the key to optimum and safe management (Bowden & Henry, 2018).

Bowden, S. A., & Henry, R. (2018). Pediatric Adrenal Insufficiency: Diagnosis, Management, and New Therapies. Int J Pediatr 2018; 1739831. doi:10.1155/2018/1739831

Miller, B. S., Spencer, S. P., Geffner, M. E., Gourgari, E., Lahoti, A., Kamboj, M. K., Sarafoglou, K. Emergency management of adrenal insufficiency in children: advocating for treatment options in outpatient and field settings. J Investig Med 2020; 68: 16–25. doi:10.1136/jim-2019-000999

Moloney, S., Murphy, N., & Collin, J. An overview of the nursing issues involved in caring for a child with adrenal insufficiency. Nurs Child Young People 2015; 27; 28–36. doi:10.7748/ncyp.27.7.28.e609

Padidela, R., & Hindmarsh, P. C. Mineralocorticoid deficiency and treatment in congenital adrenal hyperplasia. Int J Pediatr Endocrinol 2010; 656925. doi:10.1155/2010/656925

Porter, J., Withe, M., & Ross, R. J. Immediate-release granule formulation of hydrocortisone, Alkindi(R), for treatment of paediatric adrenal insufficiency (Infacort development programme). Expert Rev Endocrinol Metab 2018; 13: 119–124. doi:10.1080/17446651.2018.1455496

Simpson, A., Ross, R., Porter, J., Dixon, S., Whitaker, M. J., & Hunter, A. Adrenal Insufficiency in Young Children: a Mixed Methods Study of Parents’ Experiences. J Genet Couns 2018; 27: 1447–1458. doi:10.1007/s10897-018-0278-9

Tollerfield, S., Atterbury, A., Langham, S., Morris, S., & Farrell, N. C1.4 My cortisol, development of a life saving app. Arch Dis Child 2017; 102; A5. doi:10.1136/archdischild-2017-084620.12

Watson, C., Webb, E. A., Kerr, S., Davies, J. H., Stirling, H., & Batchelor, H. How close is the dose? Manipulation of 10mg hydrocortisone tablets to provide appropriate doses to children. Int J Pharm 2018; 545: 57–63. doi:10.1016/j.ijpharm.2018.04.054

Professor Ieuan Hughes and Dr Rieko Tadokoro-Cuccaro

Medical management of congenital adrenal hyperplasia: can conventional therapy be replaced by novel treatments?

Introduction

Congenital adrenal hyperplasia (CAH) is a disorder of decreased cortisol production coupled with an increase in androgens from a compensatory adrenocorticotropic hormone (ACTH) drive. Glucocorticoid and mineralocorticoid replacement is the standard treatment for CAH. The optimal goal is to ensure well-being, prevent adrenal crises and suppress excess androgens but this is unrealistic using conventional treatment. The price paid is childhood growth suppression and adult morbidity associated with obesity and cardiovascular disease. How can this conundrum be solved?

Hydrocortisone (HC) is the preferred glucocorticoid for children with CAH. A short half-life requires dosing three times daily, which is a challenge in infants. Longer-acting steroids have their place but generally only for adults. Despite cortisone being introduced 70 years ago to treat CAH, replicating the cortisol diurnal rhythm remains a challenge.

Immediate-release HC granules, Alkindi (Diurnal Ltd, UK), were developed for infants and children (1). They provide small dosing, flexibility and easy administration. Pharmacokinetic studies showed bioequivalence of Alkindi to the HC tablet.

Two modified-release HC preparations are in clinical trials to try to match endogenous cortisol rhythm: Plenadren (immediate and delayed release; Shire Pharmaceuticals Ltd, UK) and Chronocort (delayed and prolonged release; Diurnal Ltd, UK). A morning dose of Plenadren in 16 healthy adults produced a similar cortisol exposure to three times daily HC, except later in the day (2). A supplementary dose of HC would still be needed.

Chronocort was designed to mimic increasing cortisol levels overnight, peaking on awakening and falling to a nadir later in the day. Chronocort is taken twice daily in “toothbrush” mode at bedtime and on awakening. A phase 2 study in 16 adults with CAH showed profiles similar to endogenous cortisol; this resulted in reduced conventional HC doses and decreased androgens (3). A larger phase 3 study in 122 patients failed to confirm superiority over conventional treatment but the manufacturers of this agent have nevertheless been given permission to apply for EMA marketing authorization.

As an alternative to suppressing ACTH-induced androgens with modified HC, why not use a top-down approach to block ACTH? (Figure). The hypothalamic–pituitary–adrenal (HPA) axis is orchestrated by corticotrophin-releasing hormone or factor (CRH or CRF). Consequently, blocking this factor or the receptor to which it binds on the cell membrane of pituitary corticotroph cells would inhibit ACTH synthesis and reduce adrenal steroid production. A CRF1 receptor antagonist was studied in eight women with CAH in a phase 1b clinical trial (4), where a bedtime dose significantly reduced morning ACTH levels in a dose-dependent manner. In turn, this decreased concentrations of precursor steroids such as 17OH-progesterone and induced a reduction of androgens. Further studies of a novel approach to medical adrenalectomy are required but the results are promising.

Perhaps a complete shutdown of adrenocortical function can be achieved safely and replacement doses of HC and a mineralocorticoid added. This form of block and replace regimen would be akin to the non-surgical treatment of hyperthyroidism.

Other examples of non-conventional treatments for CAH include inhibiting the enzyme for adrenal androgen biosynthesis with abiraterone (5), blocking the action of androgens and oestrogens with steroid receptor antagonists (6), continuous infusion of HC by pump therapy similar to that used to treat insulin-dependent diabetes (7) and bilateral adrenalectomy (8). Such therapies have merit at an individual level, but are not suited for routine management of CAH.

Use of HC granules sprinkled on soft food is a practical step forward for infants and young children with CAH. Attempts to deliver HC to replicate endogenous cortisol secretion are progressing with some success. The more radical option of a neuroendocrine approach to the HPA axis by blocking CRH action also has exciting potential for the future medical management of CAH.

Support from the National Institute for Health Research Cambridge Biomedical Centre. R T-C is supported by the Cambridge Children’s Kidney Research Fund.

  1. Neumann U, Whitaker MJ, Wiegand S, et al. Absorption and tolerability of taste-masked hydrocortisone granules in neonates, infants and children under 6 years of age with adrenal insufficiency. Clin Endocrinol 2018; doi:10.1111/cen.13447
  2. Johannsson G, Bergthorsdottir R, Nilsson AG, et al.Improving glucocorticoid replacement therapy using a novel modified-release hydrocortisone tablet: a pharmacokinetic study. Eur J Endocrinol 2009; doi: 10.1530/EJE-09-0170
  3. Mallappa A, Sinaii N, Kumar P, et al. A phase 2 study of Chronocort, a modified-release formulation of hydrocortisone, in the treatment of adults with classic congenital adrenal hyperplasia. J Clin Endocrinol Metab 2015; doi:10.1210/jc.2014-3809
  4. Turcu AF, Spencer-Segal JL, Farber RH, et al. Single-dose study of a corticotropin-releasing factor receptor-1 antagonist in women with 21-hydroxylase deficiency. J Clin Endocrinol Metab; doi:10.1210/jc.2015-3574
  5. Auchus RJ, Buschur EO, Chang AY, et al.Abiraterone acetate to lower androgens in women with classic 21-hydroxylase deficiency. J Clin Endocrinol Metab 2014; doi:10.1210/jc.2014-1258
  6. Auchus RJ.Management considerations for the adult with congenital adrenal hyperplasia. Mol Cell Endocrinol 2015; doi:10.1016/j.mce.2015.01.039
  7. Nella AA, Mallappa A, Perritt AF, et al. A phase 2 Study of continuous subcutaneous hydrocortisone infusion in adults with congenital adrenal hyperplasia. J Clin Endocrinol Metab 2016; doi: 10.1210/jc.2016-1916
  8. MacKay D, Nordenström A, Falhammar H. Bilateral adrenalectomy in congenital adrenal hyperplasia: A systematic review and meta-analysis. J Clin Endocrinol Metab 2018; doi: 10.1210/jc.2018-00217

Dr Alexander A L Jorge

Introduction

Noonan syndrome (NS) is a relatively common autosomal dominant disorder (1:1000 to 1:2500 live births) characterised by facial dysmorphism, short stature, chest deformities and congenital heart defects. Variable developmental delay and intellectual disability are also observed in some patients.

Disease-causing mutations in genes of the RAS/MAPK pathway are identified in 70–80% of affected patients (1). PTPN11 (protein tyrosine phosphatase, nonreceptor type 11) gene was the first causative gene identified in this condition and encodes a tyrosine phosphatase protein. Nearly 40–50% of patients with NS harbour heterozygous pathogenic variants in this gene.

The diagnosis of NS is based on classical clinic features and can be confirmed by the identification of a heterozygous pathogenic variant in one of the causative genes.

Currently, the diagnosis has been established mainly by multigene sequencing analysis (whole exome or target panel sequencing).

Patients with NS frequently have proportional postnatal short stature. Birth weight is usually normal, but there is a trend of lower birth length. Short stature (height standard deviation score [SDS] <–2] is observed in approximately 70–80% of children with NS. The growth deficit is more pronounced at pubertal ages due to the delayed puberty commonly observed in these patients (2). Interestingly, patients harbouring mutations in some specific NS-causative genes, mainly SOS1 and SOS2, have more preserved postnatal growth in comparison with other genotypes. As a group, the adult height of non-treated patients with NS is around –2.5 SDS (2).

The exact physiopathology of growth impairment remains unclear and probably involves multiple mechanisms, including reduction in IGF-1 generation and direct effect on the growth plate (1). Nevertheless, it is accepted that human recombinant growth hormone (rhGH) treatment can increase the short-term height velocity and improve the adult height of these patients (3, 4). Growth hormone therapy is licensed for NS by the US Food & Drug Administration (5) but not by the European Medicines Agency. The total increment in height SDS after long-term rhGH therapy ranges from 0.6 to 1.7 in different studies.

Early studies suggested that patients with PTPN11 mutations could have a lower short-term growth response to rhGH treatment (6), although these findings have not been reproduced by other studies (3). Only three studies evaluated the effect of rhGH therapy on adult height regarding the patients’ genotype. In these studies, the majority of the patients harbour the PTPN11 mutation and no clear difference in height gain is observed regarding genotypes.

rhGH has been shown to be safe for patients with NS, mainly based on data from retrospective studies with a limited number of patients (3). For this reason, some concerns still remain about an increase in cancer risk and worsening of hypertrophic myocardiopathy in patients with NS treated with rhGH (4). Patients with specific mutations in PTPN11, KRAS and RIT1 can have a high rate of myeloproliferative disorder during the first 5 years of life. In those patients with genetic variants highly associated with myeloproliferative disorders, the decision to start rhGH therapy should be carefully discussed and only begun after the age of 5 years. Future studies are necessary to better define the impact of specific genotypes on growth response and safety of patients with NS treated with rhGH. However, some insights are already possible by identifying the molecular basis of a patient with NS.

  1. Tajan M, Paccoud R, Branka S, et al. The RASopathy Family: Consequences of Germline Activation of the RAS/MAPK Pathway. Endocr Rev doi: 10.1210/er.2017-00232
  2. Malaquias AC, Brasil AS, Pereira AC, et al. Growth standards of patients with Noonan and Noonan-like syndromes with mutations in the RAS/MAPK pathway. Am J Med Genet A 2012; doi: 10.1002/ajmg/a/35519
  3. Noonan JA, Kappelgaard AM. The efficacy and safety of growth hormone therapy in children with noonan syndrome: a review of the evidence. Horm Res Paediatr 2015; doi: 10.1159/000369012
  4. Malaquias AC, Noronha RM, Souza TTO, et al. Impact of Growth Hormone Therapy on Adult Height in Patients with PTPN11 Mutations Related to Noonan Syndrome. Horm Res Paediatr 2019: doi: 10.1159/000500264
  5. US Food & Drug Administration. Somatropin Information, as of 23 July 2015. https://www.fda.gov/drugs/postmarket-drug-safety-information-patients-and-providers/somatropin-information
  6. Binder G, Wittekindt N, Ranke MB. Noonan Syndrome: Genetics and Responsiveness to Growth Hormone Therapy. Horm Res Paediatr 2007; doi: 10.1159/000097552

Dr Sasha Howard

Introduction

Delayed puberty (DP) is common in the developed world, affecting over 2% of adolescents, and is associated with adverse health outcomes including short stature, reduced bone mineral density and compromised psychosocial health. The majority of patients have constitutional or self-limited DP, which is often familial, most commonly segregating in an autosomal dominant pattern. However, the key genetic regulators in self-limited DP are largely unknown [1], Figure 1.

Figure 1 – Schematic representing the genes (encircled) known to be involved in the pathogenesis of self-limited DP, and their overlap with cHH and loci identified by genome-wide association studies of age of puberty in the general population. Important genes identified in precocious puberty (DLK1 and MKRN3) and the Kisspeptin gene (KISS1) and its receptor (KISS1R) are also included.

Insights into the genetic mutations that lead to familial DP have come from sequencing genes within the gonadotropin-releasing hormone (GnRH) pathway known to cause pubertal failure. Recently, via next generation sequencing, mutations in HS6ST1, GNRHR, IL17RD, SEMA3A, TACR3 and TAC3 [2] have been found in patients with DP and spontaneous onset of puberty. One persuasive hypothesis is that a single deleterious mutation may lead to a phenotype of DP, whilst two or more mutations may be required to cause absent puberty, for example in congenital hypogonadotropic hypogonadism (cHH).

By extension, other pathways related to GnRH neuronal development and function have been explored in the search for the genetic basis of DP. Mutations in IGSF10 provoke a dysregulation of GnRH neuronal migration during embryonic development, which first presents in adolescence as DP without previous constitutional delay in growth [3]. Pathogenic IGSF10 mutations leading to disrupted IGSF10 signalling potentially result in reduced numbers or mistimed arrival of GnRH neurons at the hypothalamus; this produces a functional defect in the GnRH neuroendocrine network and thus an increased “threshold” for the onset of puberty.

Upstream transcriptional regulators of GnRH signalling, such as KISS1, OCT2, TTF1, YY1 and EAP1, which act as a pubertal “brake” through a balance of activating and repressive inputs, have been proposed as attractive candidates for the pathogenesis of DP. EAP1 is known to contribute to the initiation of female puberty through transactivation of the GnRH promoter, and mutations in EAP1 have very recently been found in families with self-limited DP [4]. New evidence has also identified small non-coding RNAs important for the murine critical period (or mini-puberty). The microRNA (miR)-200/429 family and miR-155 both act as epigenetic up-regulators of GnRH transcription [5], whilst miR-7a2 has been demonstrated to be essential for normal hypothalamic–pituitary–gonadal function, with deletion in mice leading to cHH.

Administration of GH to the patient’s fibroblasts failed to produce the expected increase in IGF-I, leading the researchers to focus on the signalling pathways between these two molecules. This revealed a homozygous missense mutation in STAT5b, which rendered the protein incapable of activation by GH, despite being stably expressed.

Activated STAT5b in turn starts a signalling cascade that leads to increased transcription of IGF-I, IGF binding protein 3 and ALS. Together these molecules form a ternary complex, in which ALS stabilises the binding of IGF-I to IGF binding protein 3 and extends the half-life of circulating IGF-I. As the majority (80–85%) of IGF-I in the body is in this complex form, it was thought to be the main pathway through which GH exerts its effects.3

However, the case of a boy who entirely lacked ALS challenged this notion. The patient had a frameshift point mutation, which prevented production of ALS; he therefore had no ternary complexes and his plasma level of IGF-I was more than 5 SD below the average.

Yet, although the patient was referred to Horacio Domené (Ricardo Gutiérrez Children’s Hospital, Buenos Aires, Argentina) and colleagues partly for growth restriction, this was less severe than might be expected, at 2.05 SD below average. The team speculated that growth in their patient had been sustained by free or locally produced IGF-I, implying that total circulating IGF-I levels are less critical than previously believed.

Large-scale genome-wide association studies in the general population have highlighted hundreds of potential loci associated with the timing of puberty. Few genes at these loci have been shown to be causal in DP, although several loci are in or near to genes implicated in rare disorders of puberty (LEPR, GNRH1, KISS1 and TACR3) and pituitary function (POU1F1, TENM2 and LGR4), and two imprinted central precocious puberty genes (MKRN3 and DLK1). Genes involved in BMI control including FTO were also seen, and rare heterozygous variants in FTO have been identified in families with self-limited DP with extreme low BMI and maturational delay in growth.

Evidence for the role of the IGF-I receptor in pre- and postnatal growth came from Steven Chernausek (Cincinnati Children’s Hospital Medical Center, Ohio, USA) and co-workers, who sequenced the gene in 51 children with heights more than 2 SD below average.5 They identified two children, who both had intrauterine and postnatal growth restriction and both had mutations in the IGF-I receptor gene, one resulting in reduced IGF-I receptor function and the other in a reduced number of IGF-I receptors on the fibroblasts.

Although closely related to IGF-I, and well-recognised as a major influence on intrauterine growth, IGF-II has been considered less important for postnatal growth. However, a very recent paper detailed how a paternally inherited IGF-II nonsense mutation resulted in Silver–Russell syndrome.6

Thomas Eggermann (University Hospital, Aachen, Germany) and co-researchers found their patients lacked mutations known to cause Silver–Russell syndrome, but detected the IGF-II mutation using exome and Sanger sequencing. Two of the three affected children were very small at birth but responded to later growth hormone treatment and achieved an adult height close to the third centile; the third did not respond and had continued poor growth.

Another aspect of these case studies was the finding of other, non-growth symptoms that the researchers attributed to the mutations in the GH/IGF-I axis, implying pleiotropic actions of this pathway. For example, the boy with truncated IGF-I had, among other symptoms, profound sensorineural deafness and mental retardation, the girl with mutated STAT5b had a compromised immune response, and the Silver–Russell patients had symptoms including severe postnatal feeding problems, delayed development, mental retardation and hypotonia.

Thus, the genetic basis of DP is highly heterogeneous, with gene defects producing pathogenic mechanisms that act from early foetal life into adolescence, all converging on a common pathway of pubertal delay.

  1. Howard SR. Genes underlying delayed puberty. Mol Cell Endocrinol 2018; doi: 10.1016/j.mce.2018.05.001
  2. Zhu J, Choa RE, Guo MH, et al. A Shared Genetic Basis for Self-Limited Delayed Puberty and Idiopathic Hypogonadotropic Hypogonadism. J Clin Endocrinol Metab 2015; doi: 10.1210/jc.2015-1080
  3. Howard SR, Guasti L, Ruiz-Babot G, et al. IGSF10 mutations dysregulate gonadotropin-releasing hormone neuronal migration resulting in delayed puberty. EMBO Mol Med 2016; doi: 10.15252/emmm.201606250
  4. Mancini A, Howard SR, Cabrera CP, et al. EAP1 regulation of GnRH promoter activity is important for human pubertal timing. Hum Mol Genet  2019; doi: 10.1093/hmg/ddy451
  5. Messina A, Langlet F, Chachlaki K, et al. A microRNA switch regulates the rise in hypothalamic GnRH production before puberty. Nat Neurosci 2016; doi: 10.1038/nn.4298

Ieuan Hughes, MD FRCP FRCPCH FMedSci

Congenital adrenal hyperplasia: optimising cardiometabolic outcomes

The introduction of glucocorticoid replacement for congenital adrenal hyperplasia (CAH) has revolutionised the outlook for patients with this lifelong condition, preventing life-threatening salt-wasting crises. However, accumulating evidence shows persistently increased morbidity, and even mortality, in CAH patients, and links it not only to the disease, but also to the treatment.

Here we review key publications, selected by Ieuan Hughes, Emeritus Professor of Paediatrics from Addenbrooke’s Hospital in Cambridge, UK, which summarise the current knowledge and show how clinicians and the pharmaceutical industry are beginning to respond to this challenge.

For patients with the salt-wasting (classical) form of CAH, the introduction of glucocorticoid treatment in the 1950s turned the condition into a chronic illness, rather than a guarantee of early death.

And Hughes says that CAH was initially viewed as simple to treat: “Take your hydrocortisone, go away, no problem.”

Yet a 2014 Swedish registry study showed that the problem of salt-wasting crises was reduced, rather than entirely vanquished.1 The mortality rate was 3.9% among 588 CAH patients, compared with 1.6% among nearly 59,000 controls, and 42% of the deaths in CAH patients were due to adrenal crises. Other causes of death were cardiovascular, cancer, accident and suicide, and occurred at a similar rate as in controls. However, study authors Henrik Falhammar (Karolinska University Hospital, Stockholm, Sweden) and colleagues noted that half the cardiovascular deaths in CAH patients occurred concurrently with a severe infection, suggesting that an adrenal crisis could have contributed to these deaths.

The findings imply suboptimal glucocorticoid treatment, a point highlighted in a study of 203 CAH patients treated at specialist endocrine centres in the UK.2 Levels of the androgen precursor andostenedione were suppressed in between 10% and 29% of patients and elevated in around a third. Researcher Wiebke Arlt (University of Birmingham, UK) and colleagues found that, overall, only a third of patients had levels within the target range, and the same was true of renin levels in patients receiving mineralocorticoid replacement.

“Paediatricians have prided themselves – allegedly anyway – including myself, that we’ve done a good job,” says Hughes.

Paediatricians were not only keeping CAH patients alive, he says, but also achieving fairly good growth, helping them attain adult heights that, although lower than expected for their family, were still well within population norms.

“And then of course we hand them over to the adult physicians and they come back to us – quite rightly…”

Because the adult physicians, making detailed study of their patients’ health, were becoming aware of increased cardiometabolic morbidity in adult CAH patients. For example, Arlt et al found CAH patients were more often obese relative to the UK population, and frequently had metabolic abnormalities, with 46% having hypercholesterolaemia and 29% having insulin resistance, while around a third had osteopenia.

Similarly, Falhammar et al followed up their mortality study with a look at cardiometabolic morbidities, finding that these were almost fourfold more likely to occur in CAH patients relative to controls from the Swedish population, while cardiovascular disease was nearly threefold more common.3 Specific conditions that were elevated in CAH patients included thyrotoxicosis, venous thromboembolism, atrial fibrillation, obesity and diabetes.

Another study found evidence of cardiovascular morbidity in CAH patients at a worryingly young age.4 Ivani Silva (Universidade Federal de Minas Gerais, Belo Horizonte, Brazil) and team assessed 38 pubertal CAH patients, aged 20 years or younger, and controls matched for age, gender and pubertal status. They showed that, relative to the controls, the CAH patients had significantly increased carotid intima-media thickness – an early sign of atherosclerosis. This was not restricted to overweight patients, but was seen only in females.

So cardiometabolic morbidity is “something that’s now being taken very seriously,” says Hughes.

These studies have not only flagged high cardiometabolic risk in CAH patients, they have also begun to link it directly to CAH treatment. A cross-sectional study of 196 adults with CAH found that patients with more severe disease received higher glucocorticoid doses without achieving better disease control – higher dose was actually associated with higher androgen levels.5 Moreover, the increased glucocorticoid doses were also associated with elevated blood pressure.

Although researcher Richard Ross (Royal Hallamshire Hospital, Sheffield, UK) and team found that dexamethasone – the most potent of the glucocorticoids used in the study patients – was associated with the lowest androgen levels, this came at the expense of greater insulin resistance.

A study from Hughes’ own group took a closer look at insulin resistance in 37 CAH patients and 41 healthy controls.6 The 25 patients with classical CAH, which is diagnosed and treated early in life, had significantly greater fat mass than the controls, which the researchers attributed to the long-term effects of glucocorticoid treatment.

“Add to that the obesity epidemic and you’ve got a pretty powerful cocktail of problems,” observes Hughes.

By contrast, the 12 children with nonclassical CAH, which is usually diagnosed later in childhood, had greater lean body mass and blood pressure than the controls, and significant increases in several measures of insulin resistance. Hughes attributes these changes to the patients’ prolonged exposure to excess androgen levels in early childhood, saying that androgen “in itself is a pretty potent stimulator of insulin resistance.”

The most recent guidelines for CAH management, from 2010, acknowledge the difficulty of optimising treatment, calling it “a difficult balance between hyperandrogenism and hypercortisolism”, and noting that efforts to completely normalise androgen levels “typically result in overtreatment”.7

The guidelines also advise against long-acting glucocorticoids in children, because of their growth-suppressing effects, instead recommending hydrocortisone three times daily at the lowest possible dose to avoid compromising growth.

Hughes believes that paediatric endocrinologists “are doing much better now than we used to”, saying that the daily hydrocortisone dose, which “in the past has been too much”, has been brought down. “And that’s being translated into better growth and outcome in the short term.”

However, paediatric CAH patients experience the usual deluge of childhood colds and illnesses, each one placing the body under stress and prompting a temporary increase in hydrocortisone dose. Necessary though this is, Hughes wonders if “collectively, we have been overdosing them.”

Added to that is the highly variable pharmacokinetics of hydrocortisone, not only between different stages of childhood, but also between different children, aptly demonstrated in a paper from Peter Hindmarsh (UCL Institute of Child Health, London, UK) and Evangelia Charmandari (University of Athens Medical School, Greece).8 The researchers gave 48 children the same hydrocortisone bolus, but found that its half-life in the different patients ranged from 40 to 225 minutes. Although all children attained a similar peak plasma cortisol concentration, the speed of absorption varied widely, with time to peak concentration ranging from 20 to 118 minutes and the additional time taken for the concentration to fall below 100 nmol/L ranging from 140 to 540 minutes.

This suggests that hydrocortisone dosing should be highly individualised, and perhaps more frequent, with the study authors moving towards four doses per day on the basis of their findings.

But in some cases, clearance of hydrocortisone may be so rapid that even a six-times-daily dosing schedule is insufficient to achieve androgen control. An ingenious solution for this is continuous delivery via an adapted insulin pump, as illustrated in a case study, also by Hindmarsh.9

Pump delivery avoids gaps in hydrocortisone exposure and increased need for hydrocortisone during physiological stress can be managed simply by increasing the infusion rate or using the bolus function. Hindmarsh’s team has implemented this in three children, achieving 24-hour cortisol levels within the normal range, normalised 17-hydroxyprogesterone (17-OHP) levels and large improvements in school attendance and quality of life.

Although an effective means of mimicking physiological hormone production, an insulin pump cannot match the simplicity of oral treatment. And pharmaceutical companies are beginning to develop oral drug formulations that more closely replicate physiological production. Two slow-release formulations are currently undergoing clinical testing: Plenadren (ViroPharma SPRL, Brussels, Belgium) and Chronocort (Diurnal Ltd, Cardiff, UK).

In a phase II study of Chronocort, Ashwini Mallappa (National Institutes of Health Clinical Center, Bethesda, Maryland, USA) and colleagues found that the rates of elevated androstenedione and 17-OHP levels among 16 adult CAH patients decreased significantly with Chronocort treatment relative to conventional therapy (33.7 vs 12.0% and 33.2 vs 12.0%, respectively).10 After 6 months of Chronocort treatment, 73% and 59% had normal androstenedione and 17-OHP levels, respectively.

Hughes highlights that the more physiological replacement appeared to allow further dose reductions, with eight of the 16 patients requiring a dose reduction (although two other patients needed a dose increase).

Chronocort is taken twice daily, and therefore suppresses the overnight rise in adrenocorticotrophic hormone (ACTH) seen with Plenadren, which is taken just once daily.11

Hughes says that combating the overnight ACTH rise “probably is important, but on the other hand, if you want to be a pragmatist you’re more likely to take your medicine if it’s a once-daily thing as opposed to a twice-daily thing.”

Plenadran is a little further down the clinical testing route than Chronocort, being approved and now the subject of a post-authorisation safety registry, which includes CAH patients.12 But both formulations still need to be tested in paediatric patients.

“I hope that perhaps getting these so-called more physiological replacement regimens will in due course improve morbidity in adult life,” says Hughes.

But he adds: “It will take a generation to see that.”

  1. Falhammar H, Frisén L, Norrby C, et al. Increased mortality in patients with congenital adrenal hyperplasia due to 21-hydroxylase deficiency. J Clin Endocrinol Metab 2014; 99: E2715–E2721
    http://dx.doi.org/10.1210/jc.2014-2957
  2. Arlt W, Willis DS, Wild SH, et al. Health status of adults with congenital adrenal hyperplasia: a cohort study of 203 patients. J Clin Endocrinol Metab 2010; 95: 5110–5121
    http://dx.doi.org/10.1210/jc.2010-0917
  3. Falhammar H, Frisén L, Hirschberg AL, et al. Increased cardiovascular and metabolic morbidity in patients with 21-hydroxylase deficiency: a Swedish population-based national cohort study. J Clin Endocrinol Metab 2015; 100: 3520–3528.
    http://dx.doi.org/10.1210/JC.2015-2093
  4. Rodrigues TM, Barra CB, Santos JL, et al. Cardiovascular risk factors and increased carotid intima-media thickness in young patients with congenital adrenal hyperplasia due to 21-hydroxylase deficiency. Arch Endocrinol Metab 2015; 59: 541–547
    http://www.scielo.br/scielo.php
  5. Han TS, Stimson RH, Rees DA, et al. Glucocorticoid treatment regimen and health outcomes in adults with congenital adrenal hyperplasia. Clin Endocrinol 2013; 78: 197–203.
    http://onlinelibrary.wiley.com/doi/10.1111/cen.12045/abstract
  6. Williams RM, Deeb A, Ong KK, et al. Insulin sensitivity and body composition in children with classical and nonclassical congenital adrenal hyperplasia. Clin Endocrinol 2010; 72: 155–160.
    http://dx.doi.org/10.1111/j.1365-2265.2009.03587.x
  7. Speiser PW, Azziz R, Baskin LS, Ghizzoni L, et al. Congenital adrenal hyperplasia due to steroid 21-hydroxylase deficiency: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2010; 95: 4133–4160.
    http://dx.doi.org/10.1210/jc.2009-2631
  8. Hindmarsh PC, Charmandari E. Variation in absorption and half-life of hydrocortisone influence plasma cortisol concentrations. Clin Endocrinol 2015; 82: 557–561.
    http://dx.doi.org/10.1111/cen.12653
  9. Hindmarsh PC. The child with difficult to control congenital adrenal hyperplasia: is there a place for continuous subcutaneous hydrocortisone therapy. Clin Endocrinol 2014; 81: 15–18.
    http://onlinelibrary.wiley.com/doi/10.1111/cen.12453/abstract
  10. Mallappa A, Sinaii N, Kumar P, et al. A phase 2 study of Chronocort, a modified-release formulation of hydrocortisone, in the treatment of adults with classic congenital adrenal hyperplasia. J Clin Endocrinol Metab 2015; 100: 1137–1145.
    http://dx.doi.org/10.1210/jc.2014-3809
  11. Johannsson G, Nilsson AG, Bergthorsdottir R, et al. Improved cortisol exposure-time profile and outcome in patients with adrenal insufficiency: a prospective randomized trial of a novel hydrocortisone dual-release formulation. J Clin Endocrinol Metab 2012
    http://dx.doi.org/10.1210/jc.2011-1926
  12. Ekman B, Fitts D, Marelli C, et al. European Adrenal Insufficiency Registry (EU-AIR): a comparative observational study of glucocorticoid replacement therapy. BMC Endocr Disord 2014; 14: 40
    http://bmcendocrdisord.biomedcentral.com/articles/10.1186/1472-6823-14-40

Andrew Dauber, MD, MMSc

The genetic toolbox: dissecting pathways to growth

The role of growth hormone (GH) has been established for many decades, but it is only with the advent of advanced molecular genetic techniques that researchers have started to piece together the precise pathways leading to growth.

Here we review seven key papers, chosen by Professor Andrew Dauber, from Cincinnati Children’s Hospital Medical Center in Ohio, USA, which document the pivotal findings that shed light on the process of signalling through the GH/insulin-like growth factor (IGF)-I axis.

The research largely represents the extreme end of the growth disorder spectrum, with a single mutation having a profound effect on a patient’s height, but Dauber stresses that “they teach us a lot about the underlying physiology and the role that those genes play in normal biology.”

Laron dwarfism was first described in 1966, as an autosomal recessive syndrome with many clinical similarities to isolated GH deficiency. The characteristic low levels of IGF-I despite increased levels of GH implied a role for the GH receptor, but it was not until 1989 that researchers provided direct evidence of this hypothesis. Michel Goossens (CHU Henri Mondor, Créteil, France) and colleagues identified a likely culprit mutation in the GH receptor gene of a family with consanguineous parents with normal phenotypes who had four children, all affected by Laron dwarfism.1

All of the children were homozygous for a single base-pair substitution resulting in serine being present at a position in the extracellular domain of the GH receptor where phenylalanine is invariably found across multiple species. This, combined with the lack of GH-binding activity in the study patients, led the team to conclude that the mutation was probably causative, providing the first direct evidence for the role of the GH receptor in growth.

In contrast to this study, which actively sought the genetic basis for a condition, another two landmark reports detail more serendipitous findings; direct evidence for the involvement of STAT5 and the IGF-I receptor in growth signalling both emerged from researchers’ efforts to identify the cause of short stature in an individual child, while the surprising findings of another case study challenged the importance of the acid-labile subunit (ALS) in GH/IGF-I signalling.

In one case, Ron Rosenfeld (Oregon Health and Science University, Portland, USA) and co-researchers studied a girl whose height was 7.5 standard deviations (SD) below the average for her age, of 16.5 years.2 She had elevated GH levels and low IGF-I levels, consistent with GH resistance, yet her GH receptor gene had no mutations.

Administration of GH to the patient’s fibroblasts failed to produce the expected increase in IGF-I, leading the researchers to focus on the signalling pathways between these two molecules. This revealed a homozygous missense mutation in STAT5b, which rendered the protein incapable of activation by GH, despite being stably expressed.

Activated STAT5b in turn starts a signalling cascade that leads to increased transcription of IGF-I, IGF binding protein 3 and ALS. Together these molecules form a ternary complex, in which ALS stabilises the binding of IGF-I to IGF binding protein 3 and extends the half-life of circulating IGF-I. As the majority (80–85%) of IGF-I in the body is in this complex form, it was thought to be the main pathway through which GH exerts its effects.3

However, the case of a boy who entirely lacked ALS challenged this notion. The patient had a frameshift point mutation, which prevented production of ALS; he therefore had no ternary complexes and his plasma level of IGF-I was more than 5 SD below the average.

Yet, although the patient was referred to Horacio Domené (Ricardo Gutiérrez Children’s Hospital, Buenos Aires, Argentina) and colleagues partly for growth restriction, this was less severe than might be expected, at 2.05 SD below average. The team speculated that growth in their patient had been sustained by free or locally produced IGF-I, implying that total circulating IGF-I levels are less critical than previously believed.

A third case study supplied direct evidence that IGF-I is crucial not just for postnatal growth but also for intrauterine growth. The report featured a 15.8-year-old boy who had a birth length 5.4 SD below average and continued to have severe growth retardation throughout childhood, so that, by the time of the report, his height was 6.9 SD below average.4

This patient had elevated GH levels and undetectable serum IGF-I, and Adrian Clark and colleagues, from St Bartholomew’s Hospital in London, UK, found that this was caused by a mutation of the IGF-I gene itself. Their patient was missing exons 4 and 5, resulting in a truncated IGF-I protein.

Evidence for the role of the IGF-I receptor in pre- and postnatal growth came from Steven Chernausek (Cincinnati Children’s Hospital Medical Center, Ohio, USA) and co-workers, who sequenced the gene in 51 children with heights more than 2 SD below average.5 They identified two children, who both had intrauterine and postnatal growth restriction and both had mutations in the IGF-I receptor gene, one resulting in reduced IGF-I receptor function and the other in a reduced number of IGF-I receptors on the fibroblasts.

Although closely related to IGF-I, and well-recognised as a major influence on intrauterine growth, IGF-II has been considered less important for postnatal growth. However, a very recent paper detailed how a paternally inherited IGF-II nonsense mutation resulted in Silver–Russell syndrome.6

Thomas Eggermann (University Hospital, Aachen, Germany) and co-researchers found their patients lacked mutations known to cause Silver–Russell syndrome, but detected the IGF-II mutation using exome and Sanger sequencing. Two of the three affected children were very small at birth but responded to later growth hormone treatment and achieved an adult height close to the third centile; the third did not respond and had continued poor growth.

Another aspect of these case studies was the finding of other, non-growth symptoms that the researchers attributed to the mutations in the GH/IGF-I axis, implying pleiotropic actions of this pathway. For example, the boy with truncated IGF-I had, among other symptoms, profound sensorineural deafness and mental retardation, the girl with mutated STAT5b had a compromised immune response, and the Silver–Russell patients had symptoms including severe postnatal feeding problems, delayed development, mental retardation and hypotonia.

Postnatal growth is not regulated only by the GH/IGF-I axis. A notable example outside of this pathway is the SHOX (short stature homeobox-containing) gene, a transcription factor that is expressed in the growth plate.

SHOX was first described in 1997, by Gudrun Rappold (Heidelberg University, Germany) and colleagues.7 The cause of short stature in girls with Turner syndrome had at this time been attributed to monosomy of genes in the sex chromosomes, and narrowed down to the pseudoautosomal region. By studying patients with partial monosomy of this region and adults’ heights ranging from normal to more than 3 SD below average, the team refined this to a 170 kb section, in which they discovered SHOX.

Besides explaining the short stature of Turner syndrome patients, SHOX may also account for a small percentage of idiopathic short stature; the researchers discovered a mutation in one of 91 such patients tested.

And SHOX is just the tip of the iceberg, according to Dauber. He says that although the reviewed papers have provided “a tremendous amount of insight” into the GH/IGF-I axis, “there’s much more to growth than that one biological pathway.”

Illustrating this, Dauber points out that all reported patients with mutations in the GH/IGF axis number only a few hundred worldwide.

“But if you look at the number of kids who come to be evaluated by an endocrinologist, even with heights below minus 3 standard deviations or so, there are tens of thousands of those and we really only explain a tiny percentage of those with bona fide mutations in the GH/IGF axis.”

This suggests that most patients with idiopathic short stature will prove to have mutations in genes that are not directly involved in this pathway.

Not that Dauber believes the influence of the GH/IGF pathway is fully mapped out; he sees research in that area starting to focus more on mutations in regulatory regions of genes, and in epigenetics.

But he says: “I think that the challenge for endocrinologists in the coming years is going to be how to learn and think about the defects in growth that are outside that classic hormonal pathway.”

A complicating factor is the large grey area between polygenic short stature – caused by the cumulative effect of the thousands of variants with, at most, a millimetre effect on height – and monogenic short stature – caused by a mutation in a specific critical gene.

“The long-term challenge is going to be finding therapies that don’t affect the GH/IGF axis but do affect growth,” says Dauber.

Such therapies are coming into play, however; Dauber gives the example of the NPR2 gene, which encodes a receptor (natriuretic protein receptor B) that acts directly in the growth plate and is essential for endochondral ossification. An NPR2 agonist is currently in development for patients with achondroplasia, he says.

“That’s a totally different type of growth-promoting therapy, in a disorder that’s more commonly handled by geneticists than endocrinologists.”

  1. Amselem S, Duquesnoy P, Attree O, et al. Laron dwarfism and mutations of the growth hormone-receptor gene. N Engl J Med 1989; 321: 989–995
    http://www.nejm.org/doi/full/10.1056/NEJM198910123211501
  2. Kofoed EM, Hwa V, Little B, et al. Growth hormone insensitivity associated with a STAT5b mutation. N Engl J Med 2003; 349: 1139–1147
    http://www.nejm.org/doi/full/10.1056/NEJMoa022926
  3. Domené HM, Bengolea SV, Martínez AS, et al. Deficiency of the circulating insulin-like growth factor system associated with inactivation of the acid-labile subunit gene. N Engl J Med 2004; 350: 570–577
    http://www.nejm.org/doi/full/10.1056/NEJMoa013100
  4. Woods KA, Camacho-Hübner C, Savage MO, Clark AJ. Intrauterine growth retardation and postnatal growth failure associated with deletion of the insulin-like growth factor I gene. N Engl J Med 1996; 335: 1363–1367
    http://www.nejm.org/doi/full/10.1056/NEJM199610313351805
  5. Abuzzahab MJ, Schneider A, Goddard A, et al. IGF-I receptor mutations resulting in intrauterine and postnatal growth retardation. N Engl J Med 2003; 349: 2211–2222
    http://www.nejm.org/doi/full/10.1056/NEJMoa010107
  6. Begemann M, Zirn B, Santen G, et al. Paternally inherited IGF2 mutation and growth restriction. N Engl J Med 2015; 373:349–356
    http://www.nejm.org/doi/full/10.1056/NEJMoa1415227
  7. Rao E, Weiss B, Fukami M, et al. Pseudoautosomal deletions encompassing a novel homeobox gene cause growth failure in idiopathic short stature and Turner syndrome. Nat Genet 1997; 16: 54–63
    http://www.nature.com/ng/journal/v16/n1/pdf/ng0597-54.pdf
  8. Picture courtesy of Cincinnati Children’s Hospital Medical Center
    http://www.cincinnatichildrens.org/default.htm