Clinical experience in identical twins discordant for adult-onset growth hormone deficiency.

The syndrome of adult-onset growth hormone (GH) deficiency is well established. However uncertainties exist regarding monitoring of the therapeutic response. Currently the Growth Hormone Research Society consensus statement recommends monitoring GH replacement by taking a history, with attention to questions relating to quality of life, physical examination, body composition (by a suitable method) and measurement of insulin-like growth factor (IGF)-I. With the advent of commercially available assays for IGF binding protein (IGFBP)-3 and acid-labile subunit (ALS) it was hoped that it might prove possible to develop a more sensitive protocol for monitoring GH replacement therapy. We are in the unique position of having three patients with GH deficiency who are one of a pair of monozygotic twins, in which the second twin is healthy. We studied the effect of recombinant GH (rhGH) therapy (0.0375 U/kg) in the GH-deficient twin, and compared pre- and post-rhGH variables with similar observations in the healthy twin. Body composition, IGF-I, IGFBP-3 and ALS levels were measured before and after 3 months of rhGH therapy. Body composition changed as expected most impressively in twin 1; twins 1 and 2 had low IGF-I, IGFBP-3 and ALS prior to treatment with GH, but normal concentrations that were similar to those of their healthy twin were attained after 3 months of GH treatment. Twin 3 had normal IGF-I, IGFBP-3 and ALS concentrations before GH treatment, which were similar to those of her twin. Interestingly she also had a high fasting insulin level. After GH replacement these concentrations increased but remained within the normal range, similar to those of her healthy twin. Insulin and GH have both been shown to stimulate hepatic production of IGF-I and these results suggest that insulin may also be involved in the regulation of IGFBP-3 and ALS in vivo. Insulin has already been implicated in the regulation of IGFBP-3 and ALS in vitro and there is some evidence to suggest that this is also the case in vivo. We conclude that clinical state, body composition and IGF-I assessed together remain the best methods of monitoring GH replacement therapy and that measurements of IGFBP-3 and ALS do not confer any advantage over IGF-I.