Growth hormone deficiency in children: divergence between clinical/laboratory evaluation and genetic architecture

Short stature in children is a common reason for referral to a pediatric endocrinologist. Growth hormone deficiency (GHD) is one of the less common etiologies of short stature and slow growth, but its diagnosis is often sought since there is a proper safe and effective therapy for children with this diagnosis; however, many genetic, nutritional, psychological, illness-related, and hormonal causes must be excluded before one may label a child as having GHD. GH stimulation testing remains the gold standard for accepting this diagnosis in children with slow growth.

Testing for GHD has a very long history of methods and protocols [1]. It should be noted that the basis for the diagnosis of GHD is not just “failed” stimulation tests, but one must test the appropriate children whose clinical phenotype is often growth faltering, often leading to short stature without the exclusion of all conditions that might mimic these clinical findings, such as chronic illness or a skeletal dysplasia. Yuen and colleagues [1] comprehensively reviewed GH stimulation testing and noted that the existing methods can be difficult to perform, and results can be both inaccurate and imprecise. The interpretation of these test results includes taking into account individual patient factors, such as obesity, differences in peak GH cutoffs (by age and test), testing time points, and heterogeneity of GH assays. Bright and colleagues [2] noted a significant clinical challenge with a high rate of false-positive results in stimulation tests for all short children (>97%). Thus, one is required to use additional data as noted in the present report: medical history, height velocity, insulin-like growth factor-1 (IGF-1) levels, and bone age to make a more informed decision about GH testing [3]. The present data indicate that when even that is done, there remains a large percentage of false-positive GHD diagnoses.

Here, Plachy and coinvestigators [3] have focused specifically on a group of their short, slow-growing patients, who had GH responses of less than 10 μg/L to 2 pharmacological stimulating agents, clonidine and insulin-induced hypoglycemia. The authors’ aim was “…to comprehensively evaluate the genetic etiology of short stature in children clinically diagnosed with growth hormone deficiency.” Their data show the limited reliability of their protocol, and potentially most other protocols, for pituitary stimulation.

Their study population included 233 children who met their testing criteria (stimulated GH levels of <10 μg/L) and had parental consent. These children labeled “GH deficient” were then assigned to 1 of 3 groups based on the results of next-generation genetic testing: genetic variants causing GHD; variants accounting for short stature due to an alternative genetic etiology; and no causal genetic variant. Additionally, the GH results of the stimulation tests were categorized as less than 3, 3.0 to 4.9, 5.0 to 6.9, and 7.0 to 9.9 μg/L.

Genetic causes for GHD were found in 13 of 39 (33%) of a total population of 233 individuals and alternative etiologies of growth disorders in 26 of 39 (alternative etiologies of genetic growth disorders, 67%), most notably a primary growth plate disorder (8/39, 21%); disruption of the RAS/MAPK signaling pathway (13/39, 33%), and miscellaneous genetic causes (5/39, 13%). Those “truly” GHD had the lowest maximal GH levels on stimulation: 1.8 μg/L compared to 4.8 μg/L (alternate etiology) and 5.9 μg/L (no genetic etiology). Those with combined pituitary hormone deficiencies were more likely to be in the “confirmed” GHD group 46% vs 4% (alternate genetic) vs 12% (nongenetic) and had magnetic resonance imaging (MRI) abnormalities (73 vs 14 vs 32%). Those in the alternative genetic etiology group had the least robust growth response to treatment with recombinant human GH (rhGH) adding 1.4 SD in height compared to 2.3 SD (GHD) and 1.8 SD (no genetic etiology). The groups of patients were differentiated in part at the beginning in that those with a stimulated GH peak of less than 3.0 μg/L were longer at birth, had lower pretherapy IGF-1 levels, increased incidence of neonatal hypoglycemia, as well as combined pituitary hormone deficiency, MRI abnormalities, and growth response to rhGH, as quantitated earlier.

The investigators have comprehensively evaluated the genetic etiology of short stature in a cohort of children who met the standard criteria for GHD based on medical history, physical examination, baseline laboratory and imaging data (brain pituitary MRI and bone age), and 2 pituitary stimulation tests for GH response (<10 μg/L in the United States). By these criteria they were labeled as GHD. However, there were significant differences among the various levels of response with those who had peak GH levels of less than 3.0 μg/L having distinct differences in clinical phenotype and response to rhGH. Thus what is being investigated here is essentially the fidelity of this pharmacological stimulation testing in children who meet the historical, auxological, physical, and laboratory criteria compatible with GHD [1].

In those not confirmed GHD by genetic testing, the investigators did identify a genetic etiology for their short stature in a considerable proportion, some of whom will respond less robustly to rhGH as in this study, but alternatively may respond to pharmacological agents targeted to the growth plate, such as C-type natriuretic peptide agonists, which are under study in patients having one of multiple growth plate gene variations [4].

Broadening the scope of evaluation, what can one learn from these observations for diagnoses within the much larger group of short children, including those with idiopathic (isolated) short stature? With the availability of gene panels for short stature and those more comprehensive for skeletal dysplasias, for example, one may consider performing genetic testing earlier in the diagnostic sequence to make an accurate diagnosis. That certainly will shorten the diagnostic odyssey, denote a proper diagnosis, permit proper genetic counseling, and lead to precision medicine as new and repurposed drugs become available mainly for targets at the growth plate [4, 5]. These are precisely some of the genes noted in the patients with alternate genetic etiologies as shown by Plachy et al [3]. Scalco and colleagues [6] conclude, “Genetic testing substantially improves the diagnostic process in children with isolated and syndromic short stature.”

Abbreviations

GH

growth hormone

GHD

growth hormone deficiency

IGF-1

insulin-like growth factor-1

MRI

magnetic resonance imaging

rhGH

recombinant human GH

Funding

This work did not receive funding from any agency in the public, commercial or not for profit sectors.

Disclosures

The author is a consultant to Ascendis Pharma and Tolmar Pharmaceuticals, Inc.