Lipoprotein(a) and growth hormone: is the puzzle solved?

Abstract

Recombinant human growth hormone (rhGH) is now used routinely in the treatment of congenital and acquired GH deficiency (GHD). Additional indications for long-term administration of rhGH such as osteoporosis and aging have been proposed, but concerns regarding the safety of such prolonged courses of rhGH have been raised. Administration of rhGH is associated with glucose intolerance, higher plasma insulin concentrations and fluid retention. To add to these potential problems is the finding that rhGH may increase plasma concentrations of the atherogenic lipoprotein(a) (Lp(a)) particle (1, 2). In Caucasians, increased plasma concentrations of Lp(a) are associated with an increased risk of cardiovascular disease (3), but the mechanism by which Lp(a) is atherogenic has not been fully elucidated. Lp(a) is present only in hedgehogs, great apes and humans (4). In humans, plasma concentrations of Lp(a) vary over a 1000-fold range between individuals. Plasma Lp(a) concentrations are primarily determined by sequences at or close to the gene encoding apolipoprotein(a) (apo(a)), a highly polymorphic glycoprotein that is covalently attached to apolipoprotein B-100 (apoB100) of low-density lipoprotein (LDL) in Lp(a) (5). apo(a) contains variable numbers of a tandemly repeated ,114 amino acid motif called kringle-4. Overall, there is an inverse relationship between the number of kringle-4 repeats encoded by the apo(a) gene and the associated plasma Lp(a) concentration (6, 7). Only a restricted number of physiological, pharmacological or environmental factors influence plasma Lp(a) concentrations (8); GH is one of these factors. Over the past 5 years, more than 20 studies, including six placebo-controlled double-blind studies, have examined the effect of the administration of rhGH on plasma Lp(a) concentrations. Most studies show that rhGH administration is associated with a 110–200% increase in plasma Lp(a) concentrations, but in two studies, plasma Lp(a) concentrations remained unchanged after 12 months of rhGH therapy (9, 10). Both of these studies were performed in Japanese children, which may reflect inter-ethnic differences in the effect of rhGH on plasma Lp(a) concentrations. In this issue of European Journal of Endocrinology, Nolte et al. (11) report the effect of rhGH administration on plasma Lp(a) concentrations in a randomised, placebo-controlled, double-blind 12-month study involving 32 adults with GHD. The authors found that rhGH administration induced an ,140% increase in plasma Lp(a) concentrations and an ,14% decrease in LDL-cholesterol concentrations. This well-designed and well-executed study demonstrates once again that administration of rhGH to GHD adults is associated with a significant increase in plasma Lp(a) concentrations. Should we, as clinicians, be concerned about the Lp(a)-increasing effect of rhGH? The answer to this question is not as straightforward as one might suppose. The increase in plasma Lp(a) concentrations associated with rhGH substitution in GHD patients is modest, especially in relationship to the wide inter-individual variability in plasma Lp(a) concentrations in the general population. A direct relationship exists between plasma LDL-cholesterol concentrations and the incidence of coronary atherosclerosis. In contrast, the atherogenicity of Lp(a) does not appear to be concentrationdependent over the entire range of plasma Lp(a) concentrations. Lp(a) is associated with a significant risk when plasma Lp(a) concentrations are greater than about 20–30 mg/dl (3). If subjects in the study by Nolte’s group were classified as having low Lp(a) concentrations (i.e. 20 mg/dl), only two of 32 subjects treated with rhGH moved from the lowto the high-risk category with treatment. The Lp(a)-increasing effect of rhGH is coupled with a significant reduction in the plasma LDL-cholesterol concentrations (1, 2, 11). GHD patients tend to have higher plasma LDL-cholesterol concentrations than healthy controls and are at a significantly greater risk of developing cardiovascular events (12), despite their lower plasma Lp(a) concentrations. The beneficial LDL-decreasing effect of rhGH in GHD individuals may counterbalance the potentially detrimental effect of the increase in plasma Lp(a) concentrations. Moreover, in non-GHD subjects, reduction of plasma LDL-cholesterol concentrations using pharmacological agents that do not affect plasma Lp(a) significantly reduces coronary risk (13). Even in individuals with high plasma concentrations of Lp(a), it appears that decreasing plasma LDL-cholesterol concentrations may obviate the atherogenic effects of a high plasma Lp(a) concentration (14). European Journal of Endocrinology (1997) 137 450–452 ISSN 0804-4643