|
|
Long-term Growth Hormone Use: Safety Profile and Advese Events« Back to Volume 26, Issue 1, September 2010 - Table of Contents Roberto Lanes, MDBell and colleagues recently reported on the safety profile and adverse events detected with the use of recombinant human growth hormone (rhGH) during 20 years of post-marketing surveillance by the National Cooperative Growth Study (NCGS) of Genentech.1 Additionally an editorial on the subject by Allen was published simultaneously.2 These two articles are important and should be carefully studied. They describe and interpret the data on the cumulative enrollment of patients treated with rhGH followed by the NCGS. There were 54,996 patients (65% males, 35% females) treated from December 1985 to January 2006. This included 195,419 patient-years of treatment with Genentech's rhGH products. While the overall safety profile of rhGH continues to be favorable, this analysis highlights new areas of concern, while it tends to discard other safety issues. There were 1559 serious adverse events, including 174 deaths, most of which were unrelated to rhGH. The most common cause of death in the 19 cases believed to be rhGH related were central nervous system (CNS) tumors, particularly occurring in patients with organic GH deficiency (GHD). There were also 5 deaths due to aortic dissection/rupture in patients with Turner Syndrome (TS) and 2 deaths probably due to respiratory/cardiac problems in patients with Prader Willi Syndrome (PWS). There were 11 events consistent with acute adrenal insufficiency (AI), leading to 4 deaths. Of the 4 fatalities, 3 appeared to be associated with infection as were 5 of the nonfatal cases of serious AI. GH is known to affect the metabolism of glucocorticoids and it has a modulating effect on hepatic 11B-hydroxysteroid dehydrogenase decreasing the conversion of cortisone to cortisol. Therefore, endogenous cortisol secretion may decrease after rhGH is initiated in hypopituitary patients and previously unsuspected central hypoadrenalism may become apparent during rhGH treatment. Patients who are begun on rhGH may need to consider glucocorticoid replacement, particularly during stress with increased doses above physiologic maintenance. Patients with hypopituitarism are at a lifelong risk of developing AI, regardless of rhGH use, and need to be counseled and to receive appropriate medical attention during illness as these patients have an increased risk of sudden death. In the past, leukemia was believed to be a major safety issue associated with rhGH administration. However, there were very few patients with new-onset leukemia in the series. Thus, the data confirmed other reports that therapy with rhGH does not appear to increase the incidence of this cancer in children who do not have any other risk factors that are known to be associated with leukemia.3 However, there was an increased risk of second malignancies detected by the NCGS. The patients who received irradiation were at a higher risk of developing second malignancies. Second tumors were seen in 49 of 2500 patients with a prior history of malignancy (excluding craniopharyngioma), or 4.6 cases per 1000 patient years of rhGH treatment. The most commonly detected secondary neoplasms were CNS tumors followed by osteogenic sarcoma. There were 4 malignancies and one meningioma that developed in 16 patients with retinoblastoma. Although the risk of developing a new tumor is increased in any patient with a prior malignancy, regardless of rhGH treatment, this risk seems to be further increased by rhGH. Thus, patients and families need to be made aware of this risk. There are theoretical risks that may account for the increased risk of post-treatment tumor development in patients who have received rhGH. The mitogenic and anti-apoptotic actions of GH and insulin-like growth factor (IGF)-I suggest that high-normal levels of free IGF-I may increase the rate of cancers of the breast and prostrate. IGF-I concentrations in the high-normal range are often detected in children and adolescents treated with rhGH, particularly for non-classical indications in which supra-physiological rhGH doses are often administered. The potential relationship between neoplasms, GH use and increased serum IGF-I levels clearly needs to be considered. Targeted events reported with an infrequent incidence of <1% included scoliosis and slipped capital femoral epiphysis (SCFE), probably associated with rapid growth. New onset cases of scoliosis were not serious and were detected mainly in patients with TS, known to have an increased incidence, independent of rhGH treatment. SCFE was found in PWS associated with obesity, untreated endocrine conditions (hypothyroidism and GHD), trauma, radiation and growth during puberty. Also occurring in the population of rhGH treated patients was intracranial hypertension (IH), diabetes mellitus (DM), AI and pancreatitis. IH has been previously documented with rhGH treatment, but its mechanism is not clear. It seems to be more frequent in distinct groups of patients who were at a higher risk for this complication, ie, those with chronic renal insufficiency and TS. These patients are known to have a higher risk of IH independent of rhGH therapy. The incidence of type 1 DM was not increased with rhGH administration, while type 2 DM and insulin resistance seemed to be associated with rhGH use. These alterations appeared to be transient and reversible when GH was discontinued. Pancreatitis was detected in 3 patients with TS and in 7 other patients treated with rhGH; the mechanism linking pancreatitis to rhGH administration is unknown. There are potential pitfalls on relying on data obtained from post-marketing surveillance studies. Enrollment of treated patients is incomplete, drug exposure is variable, inconsistent compliance may lead to underreporting of adverse events by physicians, and finally reporting of adverse events in these surveillance studies is limited to the period of rhGH treatment, while detection and reporting of subsequent adverse effects depends on reports to monitoring agencies by the physician. Editor's CommentThe post-marketing surveillance study reported by Bell et al1 and reviewed in this issue of GGH by Roberto Lanes, was established, managed and supported by Genentech Inc, manufactures of the first rhGH that was approved for clinical use by the FDA in 1985. This very large and comprehensive project, carried out under the NCGS, is not the only project of this nature. The Pfizer International Growth Study database (KIGS) has also been ongoing and collected data in over 58,000 patients treated with their rhGH product. The post-marketing efforts of the manufactures of rhGH are very important, but they are not scientifically designed studies. Therefore, these data need to be carefully interpreted. There may also be differences in the results between the two large post-marketing studies that should be considered and evaluated to properly understand the differences. For example the KIGS database did not find that rhGH treatment was associated with an increase in the incidence of malignancies4; patients with no medical history of risks known to increase the risk of cancer were not at a higher cancer risk with rhGH treatment. However, how long are they planning to look for secondary malignancies later in life? In this issue of GGH Yoshikazu Nishi also points out the potential weakness of such surveillance studies; he calls our attention to the possibility of hepatoblastomas in low birth weight children treated with rhGh. More recently, there have been a number of publications that denote interest of the manufactures of rhGH to address the challenges of adherence to the medication regimen in patients receiving rhGH. Non-compliance with rhGH therapy is high5 and this must be considered in the interpretation of data regarding growth response and/or adverse events and complications of rhGH. For example, it is known that there is lower concordance of height velocity with the duration of rhGH therapy, choice of delivery device and short prescription durations.6 Adherence to medication administration has been difficult to assess and often determined indirectly by clinical subjective assessments, although newer electronic devices are being used to improve adherence.7 It may be expected that with improved adherence to rhGH treatment there may be a better growth response − but there may also be more adverse events. Fima Lifshitz, MD References - (linked to
|
Volume 26 Number 1
September 2010
www.GGHjournal.com
ISSN 1932-9032
)