Growth Hormone in Non-GH-Deficient Short Stature: Open Questions Support Long-Term Follow-up

Laurie Barclay, MD

July 01, 2003

July 1, 2003 — New research suggests questions remain about the long-term impact of using growth hormone (GH) to treat short stature in non–GH deficient children, an indication that a U.S. Food and Drug Administration panel recently endorsed.

"GH is increasingly used in non–GH deficient short children, with conditions such as Turner's syndrome and more recently in short children who lack a specific diagnosis on their cause of short stature," Jean-Claude Carel, MD, a professor of pediatric endocrinology at Groupe Hospitalier Cochin in Paris, France, told Medscape. "Studies performed so far have focused on longitudinal growth and adult height and several studies have now demonstrated a modest increase in adult height with long-term use of GH in such children."

However, GH may also increase cross-sectional growth and thereby increase bone mineral density (BMD) and strength, according to an oral presentation June 19 at the 85th annual meeting of The Endocrine Society (Endo 2003) in Philadelphia, Pennsylvania.

"In children with non-growth hormone-deficient short stature (NGHDSS), GH treatment stimulates longitudinal bone growth and thus height velocity," write Rachel I. Gafni, MD, from the National Institute of Child Health and Human Development of the National Institutes of Health (NIH) in Bethesda, Maryland, and colleagues. "Longitudinal growth occurs by endochondral bone formation whereas cross-sectional growth occurs by intramembranous bone formation."

In this double-blind trial, 39 boys with NGHDSS, aged 10 to 20 years (mean 12.8 ± 1.6 years), were randomized to receive subcutaneous GH (Humatrope; Eli Lilly), 0.074 mg/kg, three times weekly or placebo for 4.2 ± 1.3 years.

Metacarpal width, cortical thickness, and cortical BMD measured at six-month intervals by computer-assisted radiogrammetry of bone age radiographs significantly increased with age. Independent of age, these indices also increased with pubertal stage reflected in testicular volume or in Tanner stage for pubic hair ( P < .001).

Compared with placebo-treated controls, the rate of increase was greater in GH-treated subjects for metacarpal width (169% of controls; P < .001), cortical thickness (150%; P < .005), and BMD (144%, P < .01).

The authors concluded that GH treatment increases cortical thickness and bone width in boys with NGHDSS, and that this acceleration in cross-sectional bone growth may contribute to an increase in BMD and bone strength.

"These results were not unexpected given findings from recent clinical studies and animal models," coauthor Jeffrey Baron, MD, head of the Unit of Growth and Development from NIH, told Medscape. "Previous evidence has suggested that GH can increase cortical bone growth, and this study confirmed that in the metacarpal bone."

Dr. Carel, who was not involved in this study, commented that increased bone width was not surprising given changes seen in acromegaly, in which excess GH after growth plate fusion increases bone width as well as hand and shoe size.

"The implications of the findings are therefore not completely straightforward, since it is not easy to delineate the extra width that is expected to be associated with the increased length from a truly excessive width," he said. "In any case, these findings indicate that there are still many unsolved questions in the field of GH treatment."

One of the questions unanswered by the NIH study, according to George Werther, MD, MSc, FRACP, is whether acceleration by GH of the normal pubertal increases in metacarpal width, cortical thickness, and bone density leads to any ultimate difference in these parameters at the end of puberty.

"GH has in some studies been shown to increase the tempo of puberty, and this may be the mechanism operating here," said Dr. Werther, who reviewed the abstract for Medscape. He is a professor of endocrinology and diabetes and director of the Centre for Hormone Research at Royal Children's Hospital in Australia. "It is not clear that there is a sustained effect of GH therapy on bone strength in GH-intact short adolescent boys." Dr. Werther therefore recommended long-term follow-up of bone density and fracture frequency in this cohort.

In Australia, more stringent entry and exit criteria for use of GH were introduced in 1993-1994. According to Dr. Werther's article in the May issue of the Journal of Pediatric Endocrinology and Metabolism, patient numbers and drug costs were halved after the guidelines were implemented. Of 1,250 Australian children currently being treated with GH, 26% have idiopathic short stature, with final adult height -1.8 standard deviation score (SDS), or improvement of 1.1 SDS.

"The only documented effects of treating non-GH deficient children with GH is an initial increase in growth velocity, allowing some catch-up to peers, while final height data in most studies only shows a gain of 3 to 5 cm," Dr. Werther said. "There is no reason to expect any impairment in bone health in such children, so that I do not believe that GH-induced acceleration of normal pubertal increases in bone size and strength at this stage provides increased support for GH usage in this population."

In a randomized trial of 168 short children born small for gestational age, reported in the April issue of the Journal of Clinical Endocrinology and Metabolism, Dr. Carel's group found that GH treatment, 0.067 mg/kg/day until attainment of adult height, increased adult height by at least 0.6 SDS compared with no treatment.

At study entry, mean age was 10.5 years for girls and 12.5 years for boys, and mean height was -3.2 SD. Treatment duration was 2.7 ± 0.6 years. Height gain was 0.5 ± 0.8 SDS in the control group and 1.1 ± 0.9 SDS in the GH group ( P = .002).

Tolerance of GH treatment was excellent, but Dr. Carel's group concluded that there was limited potential for spontaneous catch-up in height in adolescents born small for gestational age.

Eli Lilly and Company helped support the NIH study and employs one of its authors. Dr. Carel has received reimbursements for attending conferences and speakers' fees and/or has been an investigator in clinical trials sponsored by GH manufacturers, including Eli Lilly, Pharmacia, and Sanofi.

On June 10, the FDA's Endocrinologic and Metabolic Drugs Advisory Committee (EMDAC) supported approval of Lilly's Humatrope, a somatropin of recombinant DNA origin, for a novel indication in patients with NGHDSS, with labeling restricted to patients with a height cut-off of -2.25 SDS.

The committee voted 7 to 3 that the safety of Humatrope in NGHDSS had been sufficiently characterized, and they recommended a mandatory patient registry. But the committee was deadlocked 5 to 5 over whether available data from the studies presented was sufficient to guide the safe and effective use of Humatrope in patients with NGHDSS, because adequate dose and age to treat were still uncertain.

By an 8 to 2 vote, they recommended FDA approval for the use of GH in NGHDSS as proposed by Lilly, with restricted labeling. Before initiation of therapy, the EMDAC recommended determining pretreatment height velocity, bone age, chronological age, and serum IGF-1 levels.

Based on Lilly's estimates of approximately 4,000 U.S. children between the ages of 7 to 15 years with NGHDSS, the committee calculated that as many as 30,000 to 40,000 U.S. patients could be on GH treatment at five years after approval, and they therefore suggested an age restriction on labeling.

Lilly is currently conducting a long-term observational study of Humatrope patients, called the Genetics and Neuroendocrinology of Short Stature International Study (GeNeSIS), which is ongoingin 30 countries worldwide, including 1,400 U.S. sites. If Humatrope is approved for NGHDSS, EMDAC has suggested a mandatory patient registry to detect long-term effects.

Current indications for Humatrope are pediatric growth failure caused by inadequate secretion of normal endogenous GH, short stature associated with Turner syndrome in children whose epiphyses are not closed, and replacement of endogenous GH in adults with GH deficiency.

The proposed indication for Humatrope is for the long-term treatment of NGHDSS in children whose epiphyses are not closed and in whom diagnostic evaluation excludes causes of short stature that should be treated by other means. The proposed dose is up to 0.37 mg/kg weekly, given by injection 3 to 7 times per week.

Features of a comprehensive risk management program proposed by Lilly to prevent misuse of GH include labeling to limit access to patients with a height cut-off of -2.25 SDS, educating physicians about label restrictions and benefit-risk analyses, no direct-to-consumer advertising, detailing only of pediatric endocrinologists for this indication, and controlled distribution.

To support Humatrope treatment of NGHDSS, Lilly submitted to the FDA data from two phase III clinical trials with different designs and dosing regimens, and overall patient exposure of approximately 1,200 patient-years. No new adverse effects were identified in these trials; the safety profile of Humatrope was similar to that in earlier trials; and there were no patient deaths related to Humatrope treatment.Two patients developed a neoplasm, which external oncologists determined were pre-existing to enrollment.

The magnitude of the Humatrope effect in the pivotal trial was 0.51 ± 0.20 SDS, equivalent to 3.7 cm or 1.44 inches. In the supportive trial, mean final height exceeded the mean baseline predicted height by an average of 7.2 cm for high-dose Humatrope and 5.4 cm for low-dose Humatrope. At two years of treatment, height was at the lower limits of normal for age and sex.

Endo 2003: Abstract OR8-6. Presented June 19, 2003.
J Clin Endocrinol Metab. 2003;88(40):1587-1593
J Pediatr Endocrinol Metab. 2003;16(suppl 3):613-618

Reviewed by Gary D. Vogin, MD