Adult growth hormone replacement: lessons learned and future direction.

Abstract

After publication of the first two randomized placebocontrolled studies of GH replacement in GH-deficient adults in 1989 (1, 2), it became clear that GH plays an important role in the maintenance of physiological processes within the adult. The beneficial effect of GH on body composition, serum lipids, and quality of life in these two studies has been confirmed in a number of similarly designed, placebocontrolled studies. Further evidence indicating that GH may be even more important in the adult than suggested by these initial studies is derived from several large retrospective cohort studies, which demonstrated an increased standardized mortality rate in hypopituitary adults with untreated GH deficiency, in the region of 1.8, compared with the general population (3, 4). Further retrospective cohort studies have also indicated an increased prevalence of fractures in GH-deficient hypopituitary adults (5). In 1996, on the basis of the beneficial effects observed in the above-described, randomized, placebo-controlled studies and the possibility that adult GH replacement therapy would reverse the reported increases in mortality and fracture rates of hypopituitary adults, GH received a license for use as replacement therapy in the hypopituitary adult. In adult hypopituitarism all of the evidence for the beneficial effect of GH replacement had been obtained from placebo-controlled studies using GH dosing regimens based on weight derived from experience in the pediatric setting. As a consequence, side effects were common, necessitating withdrawal of patients from the study or significant dose reductions in a high proportion of the study population. Unsurprisingly, serum IGF-I values were above the agerelated reference range in many patients, bringing into question whether the observed beneficial effects of GH were the result of replacement therapy or of pharmacological intervention. In retrospect, the use of weight-based dosing regimens was counterintuitive, given that we know that spontaneous GH secretion is reduced with increasing body mass index, is approximately 3-fold greater in females compared with males, and is reduced by approximately 14%/decade of adult life (6). Thus, when using a weight-based dosing regimen, complex algorithms would be required to compensate for these given variables. In 1998, the Growth Hormone Research Society issued guidelines for the diagnosis and treatment of GH-deficient adults (7). It was recommended that adult GH replacement be commenced at a low dose and individually optimized according to the clinical and biochemical responses. The ageadjusted IGF-I is generally accepted as the best biochemical marker to allow monitoring of the GH dose and thus prevent overreplacement of GH in an individual. Despite these guidelines the medical literature contains few studies of adult GH replacement in which the GH dose has been slowly increased from the outset of the study according to the serum IGF-I level and the maintenance GH dose individually optimized. Establishing the benefit of GH replacement in the adult using weight-based dosing regimens made it ethically difficult to perform further randomized, placebo-controlled studies using an individually optimized GH dosing regimen. Such studies would deny a proportion of the patients a therapy with which benefits had purportedly been substantiated. Thus, there are no randomized, double-blind, placebocontrolled studies using an individually optimized GH dosing regimen. At first glance, several placebo-controlled studies appear to reflect individualized GH dosing; in these studies, however, GH replacement has been initiated using a weight-based regimen with dose reduction to individualize the GH dose according to the serum IGF-I level only occurring later (8–10). Thus, it can be argued that the observed beneficial effects result from the earlier higher GH doses consequent on the weight-based dosing regimen. Reports of the possible benefits of this more physiological GH replacement regimen are thus dependent on audit of practice studies and therefore are subject to criticism associated with this study design (11–16); these latter studies have confirmed that GH has a beneficial effect on body composition, serum lipids, bone mineral density, and quality of life. In general, however, the beneficial changes observed with individualized GH dosing, quality of life being the one exception, have been of lesser magnitude than in previous weight-based replacement regimens. Furthermore, interpretation of the results is potentially confounded by the placebo effect and regression toward the mean, which are difficult to quantify. Nonetheless, individually optimized GH replacement has been shown to produce significant improvements in quality of life using both generic and disease-specific questionnaires, a reduction in total and low density lipoprotein cholesterol in the region of 0.4 mmol, a 2% fall in body fat mass with an increase in lean body mass of similar magnitude, and a 3.5% increase in lumbar spine bone mineral density (BMD) with 18-month replacement. Abbreviation: BMD, Bone mineral density. 0013-7227/02/$15.00/0 The Journal of Clinical Endocrinology & Metabolism 87(10):4427–4428 Printed in U.S.A. Copyright © 2002 by The Endocrine Society doi: 10.1210/jc.2002-020430