IMT staff
Official Sponsor
Medscape: Could you discuss the side effects of GH treatment?
Dr Rosenfeld: We have to begin by saying that pituitary-derived GH was used for many, many years worldwide with people thinking that it was a safe medication, only to find that a number of recipients of pituitary-derived GH developed a devastating disease called Creutzfeldt-Jakob disease that inevitably led to progressive neurological compromises, dementia, and death. On that basis, in 1983, pituitary GH was withdrawn from the market in the United States.
Creutzfeldt-Jakob disease is not a complication of synthetic recombinant DNA-derived GH, which was approved for use in the United States in 1985. Its use was followed by more careful surveillance for potential side effects, and the track record of its safety over the ensuing 25 years has been remarkably good. Reported complications have been rare. They include an orthopedic condition called slipped capital femoral epiphysis and a self-limited neurological condition called pseudotumor cerebri. In patients who have active diabetes, there can be an exacerbation of the diabetes, sometimes requiring higher doses of insulin. I can say that in my own experience, I have never had to take a patient off synthetic GH for any complication. I have been impressed by its safety.
The safety issue has been complicated by the publication of a preliminary report of a portion of the EU SAGhE study.[2] In this French study, the investigators followed almost 7000 adults who had been treated with recombinant GH during childhood in the late 1980s and 1990s. (Children had been diagnosed with either idiopathic isolated GH deficiency, neurosecretory dysfunction, idiopathic short stature, or born short for gestational age.) The investigators looked at long-term mortality and compared it with a French database of normal children. They reported a 30% higher mortality rate in the GH recipients, and in particular, they singled out cardiovascular and cerebrovascular deaths as well as deaths due to certain kinds of cancers. The authors also asserted that children who had received a higher dosage of GH (a dosage commonly used in the United States) were at greater risk of mortality.
I think the French authors are to be commended for performing a very difficult and labor-intensive study. Having said that, the French study was actually part of a 7-nation collaborative study, and it's unclear to me, as it is unclear to some of their European colleagues, why they submitted their findings for publication separately from the rest of their collaborators.
The study has significant methodological limitations. First, the control group was a normal pediatric population, so patients who were treated with GH, often because they have an underlying clinical disorder, were compared with otherwise normal children. It may be that children who have underlying disorders -- whether GH deficiency or idiopathic short stature or intrauterine growth retardation -- are at higher risk for death for a variety of reasons compared with the normal population. There may very well have been undiagnosed genetic, chromosomal, or metabolic disorders that made these children short to begin with and put them at higher risk for mortality. To a certain extent, this is an unavoidable criticism of the study, because one is not going to find a control group of 7000 children with GH deficiency who were not treated with GH. But it should be recognized that this is a very real methodological limitation.
Second, the number of subjects in the study doesn't allow for the kind of statistical evaluation that the authors performed. The total number of cancers in the control group and the GH recipient group were actually identical. If the total numbers were identical, but 1 subgroup of tumors was overrepresented, then, by necessity, other groups of tumors must have been underrepresented, a point that the authors did not mention. The same limitation applies to the other disorders described, such as cerebrovascular and cardiovascular disorders, which actually represent a heterogeneous group of clinical problems without a single, clear underlying etiology or predisposition.
In terms of the dosage warning, the number of patients who received the "high dosage" of growth hormone was only 281 out of the 7000. Most of those children had intrauterine growth retardation, a group that may very well be at increased risk of early death for a variety of reasons.
Having said that, I do think that SAGhE is an important study. I think it was analyzed and published prematurely, but I think we need to recognize that it is an important study. I also think that in retrospect, the endocrine community should have insisted on better long-term surveillance of children treated with GH.
My colleagues and I have published a commentary in which we recommend that even though we missed the opportunity for long-term surveillance, it is not too late.[3] It is our recommendation that all children and adults who receive GH be entered into a lifetime registry that is organized, supervised, and conducted independently of industry oversight. In the commentary, we conclude that "establishment of carefully considered cohort(s) must be a priority, if one accepts that long-term safety of GH and IGF-1 treatment constitutes an obligation of the healthcare community, pharmaceutical industry, and regulatory agencies." Our recommendations were endorsed in an accompanying editorial.[4]
Medscape: It sounds as if the community of endocrinologists is being faced with interesting challenges.
Dr Rosenfeld: I believe this to be true. GH has an enviable track record of both safety and efficacy. We should not allow this, however, to make us cavalier in our assessment of this therapy. We need to examine ways to maximize its efficacy in the appropriate populations and to guarantee its long-term safety.
Medscape: Medscape Access
