Effect of recombinant human growth hormone replacement therapy on the incidence of diabetes in children and adolescents
Growth hormone (GH) replacement therapy is the only treatment that has proven to be effective in patients with growth hormone deficiency [1⁃2], and has significant growth-promoting effects after GH replacement therapy. As early as 1958, the case of successful treatment of growth hormone deficiency (GHD) in children with growth hormone was first reported abroad . In the 1980s, “inclusion body technology” made it possible to obtain recombinant human growth hormone (rhGH). DNA recombination technology enabled rhGH to be mass-produced, overcoming the yield limitation of pituitary extracts, and also solved some application safety problems. . RhGH has been used clinically for more than 20 years, and its effectiveness has been affirmed, but safety has received wide attention. On the one hand, GH antagonizes the effect of insulin on glucose metabolism by reducing peripheral blood glucose intake, promoting glycogenolysis and gluconeogenesis; on the other hand, GH can also promote lipid decomposition, thereby promoting lipid oxidation and inhibiting glucose decomposition  . Due to the antagonistic effect of growth hormone on insulin and the effect of increasing blood sugar, its influence on the incidence of diabetes has been paid more and more attention. This paper will discuss the effects of growth hormone on glucose metabolism and diabetes incidence in children and adolescents in the existing research, in order to cause Physicians are concerned about changes in glucose metabolism in patients receiving GH replacement therapy.
1. Sources of follow-up data for children and adolescents with growth hormone replacement therapy
It is not easy to evaluate the effect of growth hormone replacement therapy on the incidence of diabetes. First, the incidence of diabetes after growth hormone replacement therapy is relatively low. It has been reported that the unit of measurement is every 100 000 treatment years . To obtain statistical significance, there must be a large sample size and long-term follow-up. Secondly, the onset time of diabetes can be separated from the use of growth hormone for several decades, and it is affected by many factors such as heredity, eating habits and drugs, which brings great difficulty to clinical research. Retrospective studies can count the incidence of early growth hormone replacement therapy in children and adolescents, but inevitably bring in various biases. Although prospective studies can avoid bias as much as possible, the time to follow up patients is relatively short. Finally, the incidence of diabetes itself is susceptible to factors such as genetic background, lifestyle, and age of life. Therefore, to scientifically evaluate the relationship between diabetes mellitus and growth hormone use, it is necessary to find a relatively matched control population.
To objectively evaluate the risk of growth hormone replacement therapy for long-term diabetes, first understand the source of the original follow-up data. At present, there are several database and clinical research projects of rhGH in the world. These databases are generally derived from the safety supervision of pharmaceutical companies after listing. The cumulative number of patients is tens of thousands, and the follow-up time is generally more than 10 years. They are respectively Pfizer. , Eli Lilly, Genentech and Novartis. These databases conduct drug safety studies at regular intervals, including investigations into the incidence of diabetes. (1) Pfizer International Growth Database (KIGS): A patient database established by Famasia Pu Cong Pharmaceutical Co., a company that produces rhGH, which was incorporated into Pfizer in 2003. It is a large international drug epidemiology. The main purpose of the survey was to monitor the safety of growth hormone replacement therapy in children and adolescents. Among them, there were only 23,000 children and adolescents enrolled in the 10 years from 1987 to 1997, and nearly 60% (about 14,000) were children and adolescents with growth hormone deficiency [6⁃7]. (2) International Study of Genetics and Neuroendocrinology in Dwarfs (GeNeSIS): A multi-country follow-up survey funded by Eli Lilly and Company, which was administered to children and adolescents treated with Eli Lilly’s humatrope growth hormone. The survey enrolled more than 11,000 children and adolescents between 1997 and 2007, including about 5 500 children and adolescents in the United States . (3) National Growth Collaborative Study (NCGS): A long-term prospective follow-up from 1985 to 2010, followed by patients who underwent growth hormone in Genentech under the age of 17 in North America (USA and Canada). Since 2006, the study has been conducted in France, Germany, Italy, Romania, and the United Kingdom, and has been renamed International Growth Research (iNCGS), which targets 3,000 subjects under 18 years of age .
(4) PATR Children Study (PATRO): A multi-country, multi-center follow-up survey of adverse drug events in children and adolescents using omnitrope (a rhGH) in 10 European countries .
2. Analysis of the impact of GH application on the incidence of diabetes
Looking at the follow-up study of rhGH for nearly 20 years, it can be found that not only the conclusions of the research between different databases will be different, but also the data analyzed in a database with different time periods will be inconsistent [6⁃10]. This indicates that the objective and accurate analysis of the impact of rhGH on the risk of long-term diabetes can not be concluded only by a short-term, single-database study, which requires a long period of follow-up to accumulate enough data samples, and needs to integrate various databases. The research report looks at it.
1. GH application has no effect on the incidence of diabetes: 1991
Czernichow et al  analyzed follow-up records of 8 100 children and adolescents receiving growth hormone replacement therapy in the KIGS database and found that only one child and adolescent had type 1 diabetes (T1DM) after using growth hormone, and two patients developed type 2 diabetes (T1DM). Diabetes (T2DM), and one of them has high risk factors such as obesity and family genetic predisposition. The short follow-up of this study inevitably ignored some of the children and adolescents who had not developed diabetes at the time.
In 2000, Maneatis et al  analyzed more than 33,000 patients in the NCGS database from 1985 to 1999 and found that there was no increase in the risk of diabetes compared with the normal control population. As of September 2012, PATRO initial results analysis found
In 184 research centers in 10 European countries, 1 837 patients with growth hormone omnitrope were treated with no confirmed diabetes in 10 children . In 2013, Shigeru et al  analyzed 1 129 GHD and 90 patients with Turner syndrome in the (GeNeSIS) database using riptropin GH for 4 years, and the GH dose was 0.175 and 0.35 mg·kg in GHD patients and TS patients, respectively. 1·Week-1, the effect on glucose metabolism was found to be very low. In the GHD group, 7 patients had impaired glucose tolerance, only 2 patients developed T2DM related to GH therapy, and only 1 patient in the TS group developed T2DM. Not related to GH applications. Ricardo et al  found that T⁃ cells mediate autoimmune disease in a T1DM mouse model, GH overexpression can regulate immune response, reverse the progression of diabetes, the mechanism may be GH acting on β cell proliferation and apoptosis, by triggering environmental cells The factor promotes polarization of anti-inflammatory macrophages, maintains T cell activity, and inhibits Th17 cell plasticity.
2. GH application increases the incidence of diabetes: In 2000, Cutfield et al  retrospectively analyzed the KIGS database, followed up the participants between 1987 and 1997, and collected data related to glucose metabolism such as family history, hematology Diabetes risk factors (obesity, Turner syndrome, Pradaer⁃ Willi syndrome, etc.) and related drug use, each patient received growth hormone therapy for more than 3 months. The “expectation rate” of type 1 and type 2 diabetes in the KIGS group was calculated by demographic data by matching the national distribution of KIGS. The study found that the incidence of T1DM in children treated with growth hormone replacement therapy did not increase compared to the “expectation rate”; the incidence of T2DM was 33.4 per 100 000 treatment years, which was six times higher than the “expectation rate”. About 80% of adolescents with T2DM are obese at the time of diagnosis, while those with obesity are very rare in patients with T2DM after receiving growth hormone therapy, so the increase in the incidence of T2DM in patients receiving growth hormone therapy is not significantly affected by obesity. It is worth noting that in those patients with T2DM, diabetes is not relieved by stopping the use of growth hormone, indicating that the symptoms of T2DM are not caused by the transient effects of growth hormone. The authors believe that this may be related to the growth of hormones in patients who have not received growth hormone therapy.
In 2010, Bell et al  re-examined the NCGS database, 55 000 patients received growth hormone replacement therapy in 1985 and 2006, and found 33 patients with T1DM, the incidence was not significantly higher than the general population. Twenty patients had T2DM (including 2 patients with Papill’s syndrome) during or after the use of growth hormone, with an incidence of 14 per 100,000 treatment years. Although it is lower than Cutfield’s research, and due to the difference in objective conditions, the patient’s weight, lifestyle and other data could not be systematically counted. This incidence is still significantly higher than the T2DM rate in children in the United States. Personally, these two data are relatively poorly comparable. Maneatis’s research target is mainly NCGS database. Most of the patients are enrolled in the US and Canada. The Cutfield study is mainly based on the KIGS database. The patients enrolled are more geographically rich in glucose metabolism. There is a difference in the expected value of the abnormality.
In 2011, Child et al  analyzed more than 11,000 children and adolescents using growth hormone replacement therapy in the GeNeSIS database, compared with the average children in the US SEARCH for Diabetes in Youth Study database and found median treatment of average growth hormone. With a life of 1.8 years, the incidence of T1DM did not increase, and the incidence of T2DM was 6.5 per thousand, much higher than the average, further supporting the above conclusions.
Although the above studies do not support the use of growth hormone therapy to aggravate the risk of T1DM, a controlled study by Bonfig et al  found that exogenous growth hormone may aggravate the symptoms of T1DM hyperglycemia, affecting age, gender, duration of diabetes, ethnicity, etc. After matching the factors, children and adolescents with GHD and T1DM need to use a larger dose of insulin (1 U·kg⁃1·d⁃1) to control blood glucose, while children and adolescents with T1DM only need 0.75 units·kg⁃1 · insulin dose of d⁃1.
Third, the effect of growth hormone replacement therapy on the incidence of diabetes
Growth hormone antagonizes the regulation of insulin on blood glucose under physiological conditions, and has the effect of raising blood sugar. Therefore, neonates with growth hormone deficiency often have clinical symptoms such as hypoglycemia . The incidence of diabetes in patients with acromegaly with excessive growth hormone secretion is significantly increased, reaching 20% 50%, confirming that growth hormone can increase insulin resistance and affect glucose metabolism . A lot of research suggests that
The glycemic effect of GH may mainly produce insulin resistance by inhibiting the oxidation of glucose by peripheral tissues and the non-oxidative pathway of glucose in the body, increasing insulin release and FFA excretion. Because diabetes is affected by factors such as age, race, eating habits, environment, family history, body mass index, etc., different doses, monitoring methods, treatment time, etc. will affect the results of glucose metabolism, and the conclusions of GH on blood glucose are inconsistent. Studies have also found that GH has an insulin-like effect, and short-term application can lower blood sugar and inhibit fat degradation . Long-term GH treatment may improve lipid metabolism, decreased fat mass, and may even improve glucose metabolism and insulin sensitivity [20⁃21].
Based on the above research report, we can see some common conclusions: (1) There is no evidence to support the risk of T1DM and because of the increase in growth hormone replacement therapy, but growth hormone deficiency in children and adolescents with T1DM requires greater doses of insulin control. Blood glucose; (2) The current research supports the incidence of T2DM after treatment with growth hormone, and the occurrence of diabetes does not disappear with the withdrawal of growth hormone, suggesting that this effect may occur in advance with growth hormone-promoting diabetes-susceptible population Diabetes-related, because the occurrence of T2DM often takes a relatively long time to show significant differences, so objective assessment requires longer follow-up. It is worth noting that the above studies also have certain limitations. First, because the incidence of diabetes has increased year by year in the past two or three decades, and the incidence of different physiological and social backgrounds is different, it is difficult to estimate the incidence of diabetes in the general population. Second, since most studies are retrospective, there is inevitably bias, and the effectiveness and safety of GH need to be studied through large-scale, long-term, prospective, randomized, double-blind aspects, but this requires a lot of financial support. Finally, all cohort subjects were not included in the Asian population on a large scale. Whether the above conclusions are applicable to children and adolescents in China is subject to scientific research and clinical practice testing.
The impact of growth hormone replacement therapy on glucose metabolism in children is currently lacking in uniform conclusions, and further large-scale long-term experiments are needed to explore the safety of treatment. By analyzing the database of large-scale epidemiological surveys, the incidence of various types of diabetes in children with growth hormone replacement therapy is calculated, and the risk of diabetes in children is more concerned, which provides a reference for clinically focused patients. Summarizing existing research data, GH application may increase the risk of diabetes incidence, and weigh the pros and cons before applying GH replacement therapy. Before using growth hormone therapy, we need to routinely evaluate patients’ blood glucose levels, including glycated hemoglobin, fasting blood glucose and insulin concentration. For children with high risk factors for diabetes such as obesity, Turner syndrome, intrauterine growth retardation, and Papillon’s syndrome, children and adolescents need to pay attention to follow-up changes in blood glucose.