Growth Hormone and Related Growth Factor Measurements:
Increasing Demand with Diverse Clinical Applications
by Michael J. Bennett, Ph.D., FRCPath, FACB
Professor of Pathology and Pediatrics
University of Texas Southwestern Medical Center at Dallas
Director of Clinical Chemistry
Children's Medical Center of Dallas, TX, US

The measurement of human growth hormone (GH) in serum has been useful for many years in the clinical evaluation of short stature and growth failure in children due to gross pituitary failure and isolated GH deficiency. In adults and older children, the measurement of GH has been valuable for the diagnosis of acromegaly, an abnormal growth acceleration usually due to pituitary adenomas (tumors), which secrete excessive amounts of GH.

Although the clinical measurement of GH has been a standard practice for some time, the mechanisms of action and regulation have not been fully defined. Increasingly, interactions of GH with its target hormones, previously known as somatomedins but now referred to as insulin-like growth factors (IGFs), and its inhibitory element, somatostatin, have been implicated in a wide array of disease states including somatic growth defects and cancer.

This overview will focus specifically upon the interactions and pathophysiology of GH and IGF-I and will address the clinical importance of measuring these components in serum.

GH is a polypeptide (191 amino acids) secreted from the somatotrope cells of the anterior pituitary in response to various stimuli. These stimuli include the hypothalamic release of a polypeptide (40 amino acids) called growth hormone-releasing hormone (GHrH). Release of GH is also modulated by higher centers in the brain. Stimuli such as stress and hypoglycemia, which result in increased catecholamine production, also promote GH release into the circulation (Figure 1). GH release by the pituitary is inhibited by the action of somatostatin, otherwise known as growth hormone-inhibiting hormone, a hypothalamic peptide (14 amino acids) that also has an inhibitory effect on TSH secretion by the anterior pituitary, and on insulin and glucagon secretion by the pancreas.

Figure 1. Physiological control of GH, IGF and IGFBP production and release.

The overall effect of GH is to promote growth of bone and cartilage and to promote anabolic (growth) pathways in all tissues. The effects are mediated by IGFs, with IGF-I being quantitatively the most significant of these growth factors. The polypeptides IGF-I (17 kDa) and IGF-II (22 kDa) are alternately spliced products of a single gene located on chromosome 12q22-q24.1. They are synthesized by the liver in response to GH stimulation via a GH receptor. IGFs are transported in the circulation by members of a family of GH-dependent IGF-binding glycoproteins, of which insulin-like growth factor binding protein-3 (IGFBP-3) is the most significant. IGFBP-3 is responsible for the transport of 75 percent of circulating IGF-I. IGFs have structures that are remarkably similar to insulin, but they have far greater growth properties. Their mechanism of action is through cell membrane receptors that are similar to insulin receptors.

Clinical utility of GH measurement
GH levels fluctuate considerably under baseline conditions, as they depend upon many factors. Consequently, random samples may lead to erroneous conclusions in cases of both hypersecretion and deficiency. A number of inhibition/stimulation tests have been devised to better determine GH status.

Hypersecretion of GH
Individuals with acromegaly or pituitary gigantism due to pituitary adenoma typically demonstrate a flat GH response to an oral glucose load, whereas normal individuals demonstrate GH levels that fall to less than 2 ng/mL, 30 to 60 minutes after an oral glucose load.

Hyposecretion of GH
Diagnosing growth hormone deficiency in children who are failing to grow at the normal rate is crucial. Pediatric endocrinologists must often distinguish short children who are truly GH deficient and will achieve their expected growth potential through expensive GH therapy from children who are constitutively short and will not realize therapeutic benefits. It is also very important to identify children with a GH deficiency before epiphyseal fusion of the long bones takes place, as treatment is unlikely to be effective after this time. Random GH levels are generally uninformative and a number of stimulation tests have been devised to test GH reserves. These include insulin-induced hypoglycemia or glucagon stimulation. A peak GH level in excess of 10 ng/mL is regarded a a normal response whereas a peak level below 10 ng/mL is generally regarded as a case of GH deficiency (Figure 2).

Figure 2. Normal and deficient responses during a glucagon stimulation test for GH deficiency.

The pathophysiological nature of the GH deficiency must be identified, as this will assist in determining the appropriate treatment regimen. Organic deficiency of GH, due to isolated failure of the pituitary to secrete GH or other pituitary hormones (panhypopituitarism), generally responds well to GH therapy. In the case of panhypopituitarism, other aspects of the endocrine system need to be investigated including the thyroid, gonadotropic and adrenocortical axes. Nonorganic causes of GH deficiency include malnutrition and psychosocial deficiency due to poor social contact. These causes may be reversible without the need for GH therapy if the adverse conditions are addressed.

Experience of performing in-house GH measurement in a tertiary care pediatric center
Prior to the introduction of an in-house service for GH measurement in Children's Medical Center (CMC) of Dallas, Texas, samples were sent out to a reference laboratory with a concomitant delay in turnaround. In 1999, the CMC introduced a chemiluminescence-based GH assay and in-house service (initially IMMULITE^®, currently IMMULITE^® 2000). Most of the requests for GH measurement have been for the evaluation of short stature. This requires a glucagon stimulation test that is performed in the clinic, with samples being collected at 0, 30, 60, 120 and 180 minutes. The five samples are analyzed sequentially upon receipt of the 180-minute sample in the laboratory and results are reported within minutes. The CMC endocrinologists are now able to act immediately, as results often become available while the patient is still in the clinic. This improved turnaround has resulted in an impressive increase in orders for this test (Figure 3).

Figure 3. Increase in the number of orders for GH testing at Children's Medical Center in Dallas after the 1999 introduction of in-house service using the IMMULITE®, and later, the IMMULITE® 2000 system.

Measurement of IGF-I and IGFBP-3 in the clinical laboratory
Both IGF-I and IGFBP-3 secretion are dependent upon the pituitary release of GH. In the cases of hyper- and hyposecretion of GH, the result is an increase or decrease in the hepatic synthesis or release of IGF-I and IGFBP-3. While there is still only a relatively small amount of data on the utility of IGFBP-3 measurement, it has been suggested that measurement of IGF-I is superior to that of GH in evaluating both acromegaly and GH deficiency states. IGF-I levels do not appear to fluctuate as much as GH, although other endocrine disorders may also result in low IGF-I levels. The increased availability of sensitive assays for these important growth factors will provide the opportunity to evaluate the best determinants of abnormal growth.

Mutations in the IGF gene
A number of mutations in the IGF gene have been described, which result in severe growth retardation in the homozygous-deficient state. Carriers for these mutations also tend to be short in stature, suggesting a possible heterozygous gene dose effect, which may potentially be a very important determinant of growth in a larger population. Paradoxically, the homozygous-affected individuals tend to have high GH levels, as do patients who have mutations in the GH receptor, so that even high GH levels may eventually provide clues to the etiology of short stature.

IGF-I and cancer
High circulating IGF-I levels have been implicated in prostate growth, hyperplasia and malignancy. Prostate-specific antigen (PSA) is still the most powerful independent predictor of prostate cancer, but adding IGF-I to the receiver operator curve for PSA has been shown to improve the predictive value of PSA measurement. Other studies have shown a strong relationship between circulating IGF-I concentrations and breast cancer in premenopausal but not postmenopausal women. Clearly, further studies on the role of IGF-I and cancer could provide critical insights into the mechanisms of oncogenesis.

Conclusion
Serum GH measurements have demonstrated long-standing utility in the clinical evaluation of specific growth abnormalities. Inhibition/stimulation tests are generally more valuable than basal determinations, as GH levels are in constant flux. Related growth factors such as IGF-I and IGFBP-3, synthesized in response to GH secretion, exhibit more stable plasma levels. Many investigators have indicated that IGF measurements may actually provide more useful information than those of GH in the evaluation of acromegaly and GH deficiency states. In addition, high IGF-I serum levels may be strongly predictive of risk for a variety of malignancies. Currently, the clinical issues surrounding GH are generating more questions than answers. It is hoped that further research will eventually lead to a more conclusive body of evidence in this area.