Effect of long-acting polyethylene glycol recombinant human growth hormone dosages in pediatric practice: A pooled analysis of two phase IV randomized controlled trials

To the Editor: Growth hormone deficiency (GHD) impairs growth and development, affecting approximately 1 in 4000 children worldwide.^[1] Recombinant human growth hormone (rhGH) has been standard practice since 1985.^[2] However, compliance has been challenged by the need for frequent injections. Polyethylene glycol rhGH (PEG-rhGH) was developed, improving protein stability and extending half-life. Some studies suggested that a weekly dosage of 0.20 mg/kg of PEG-rhGH offers improved growth rates and increased insulin-like growth factor-1 standard deviation score (IGF-1 SDS) without novel toxicities.^[3] Here, we present a pooled analysis of two phase IV trials evaluating the efficacy and safety of various PEG-rhGH dosages in GHD.

This two multicenter, randomized, open-label, dose-controlled phase IV clinical trials (https://clinicaltrials.gov; Nos. NCT02380235 and NCT02908958) recruited prepubertal children (Tanner I stage) of three years old or older who were diagnosed with GHD. Written informed consent forms were obtained from all the patients. These children had not received growth hormone (GH) treatment within at least six months. GHD confirmation involved verifying that the children’s height fell below the third percentile for their age and sex, as the 2005 Chinese national survey, a height velocity [HV] of 5.00 cm/year or less, peak plasma GH under 10.00 ng/mL, and a bone age one year or more behind the chronological age. Key exclusion criteria were abnormal liver or kidney functions, positive anti-hepatitis B core antigen (anti-HBc), hepatitis B surface antigen (HBsAg), and hepatitis B e-antigen (HBeAg) tests, high allergy propensity, serious diseases, diabetes, congenital skeletal abnormalities, or growth abnormalities such as Turner syndrome.

Participants were administered PEG-rhGH in these two trials. Using concealed randomization, the NCT02908958 trial (https://clinicaltrials.gov/study/NCT02908958) assigned participants to doses of 0.14 or 0.20 mg·kg^–1·week^–1, while the NCT02380235 trial assigned to 0.12 or 0.20 mg·kg^–1·week^–1. Participants were pooled into the high-dose group (0.20 mg·kg^–1·week^–1) and low-dose group (0.12 or 0.14 mg·kg^–1·week^–1).

Over 26 weeks, with visits at baseline, 4 weeks, 13 weeks, and 26 weeks, treatment effectiveness and safety were evaluated. Effectiveness was assessed by changes in height SDS (Ht SDS) based on age, yearly growth rate (HV), blood IGF-1 SDS, and bone maturity. The primary outcome, Ht SDS by chronological age, was calculated as (the height at evaluation–mean height of normal children with the same age and sex)/height standard deviation of normal children with the same age and sex. Safety was assessed by incidence of adverse events (AEs). Insulin resistance index (HOMA-IR) = fasting glucose (mmol/L) × fasting insulin (μU/mL) ÷ 22.50.

According to the intention-to-treat principle, all randomized participants who used at least one dose of PEG-rhGH with efficacy data constituted the full-analysis set (FAS). All participants who received at least one dose of PEG-rhGH with safety evaluation data constituted the safety set (SS). Missing data were filled using last observation carried forward. Analysis was performed with SAS 9.2 software (SAS Institute, Inc., Cary, USA). Continuous data were presented as mean ± standard deviation or median (Q1, Q3), while categorical data were presented as n (%). Continuous data were compared using paired t-test or signed rank sum test, with inter-group comparisons made via the Wilcoxon rank sum test. Categorical data comparisons utilized the chi-squared test or Fisher’s exact test. A two-tailed P value <0.05 was considered statistically significant.

The NCT02380235 trial recruited from December 2, 2014 to October 8, 2017, and completed follow-up on April 10, 2018. The NCT02908958 trail recruited from November 22, 2014 to December 25, 2017, and completed follow-up on June 27, 2018. A total of 1484 participants were recruited, with 1444 in the FAS, including 729 in the high-dose group and 715 in the low-dose group. The participants were primarily male (953/1444, 66.00%) and Han Chinese ethnicity (1384/1444, 95.84%). The children had an average age of 7.6 ± 2.5 years. Their average baseline height and weight were 112.53 ± 13.05 cm and 20.34 ± 6.28 kg, respectively. The starting growth rate was 3.17 ± 1.51 cm/year, and the baseline bone age was 5.9 ± 2.2 years. Approximately 3.39% (49/1444) of these participants had received GH treatment more than six months before enrollment. The baseline characteristics were comparable between the two dosage groups [Supplementary Table 1, http://links.lww.com/CM9/B825].

Both the high-dose and low-dose groups demonstrated a significant increase in Ht SDS over the course of 26 weeks. There was a statistically significant difference between the two groups at 26 weeks (t = –3.40, P <0.01) [Figure 1A]. Besides, the high-dose group exhibited a more noticeable rise in Ht SDS from baseline compared to the low-dose group at 13 weeks (Wilcoxon rank sum test value = –3.29, P <0.01) and 26 weeks (Wilcoxon rank sum test value = –8.46, P <0.01) [Figure 1B]. Similarly, the annual growth rates also increased significantly from baseline to 26 weeks in both groups. The high-dose group displayed a statistically significantly higher annual growth rate than the low-dose group at 4 weeks (t = –2.77, P <0.01), 13 weeks (t = –7.16, P <0.01), and 26 weeks (t = –8.59, P <0.01), respectively [Figure 1C]. The increases of annual growth rate from baseline level were also higher in the high-dose group than in the low-dose group at 4 weeks (Wilcoxon rank sum test value = –2.10, P <0.05), 13 weeks (Wilcoxon rank sum test value = –5.51, P <0.01), and 26 weeks (Wilcoxon rank sum test value = –6.10, P <0.01), respectively [Figure 1D]. Regarding the IGF-1 SDS, there were significant increases in both groups from baseline to all follow-up points. Notably, IGF-1 SDS was consistently higher in the high-dose group than in the low-dose group at all visit points (t = –4.33 at week 4, t = –4.45 at week 13, t = –4.32 at week 26, respectively, all P <0.01) [Figure 1E]. The increases of IGF-1 SDS from the baseline was also more significant in the high-dose group (the values of Wilcoxon rank sum test were –5.61, –5.30, –5.10, respectively, all P <0.01) [Figure 1F]. Finally, after 26 weeks of treatment, bone age assessment revealed no statistically significant differences between the two groups. The change rates in bone maturity were similar in both groups, indicating that there was no accelerated bone age progression.

Figure 1.

Effectiveness results of the high-dose group and low-dose group at 4 weeks, 13 weeks, 26 weeks, and/or baseline. (A) Ht SDS. (B) The changes of Ht SDS. (C) The annual growth rates. (D) The changes of annual growth rate. (E) The IGF-1 SDS. (F) The changes of IGF-1 SDS. Values are shown was mean ± standard error. ^*P <0.01; ^†P <0.05. ΔHt SDS: Height standard deviation score change; ΔHV: Height velocity change; ΔIGF-1 SDS: Insulin-like growth factor-1 standard deviation score change.

In the SS, composed of 1451 participants, the overall occurrence of AEs was 50.03% (726/1451), serious AEs (SAEs) were 1.03% (15/1451), drug-related AEs occurred in 8.55% (124/1451) of participants, and drug-related SAEs were only 0.14% (2/1451), with no statistically significant difference between the high-dose and low-dose groups. Hypothyroidism was the most common AE, reported in 30 patients mostly around 2 months after treatment. Some experienced elevated thyroid-stimulating hormone, while others had decreased. In the latter cases, patients were given levothyroxine sodium but continued their treatment. Thyroid function remained stable between the two groups after 26 weeks. Transient edema occurred in 19 participants, primarily (13/23) within the first week. It resolved within a week without long-term effects. Mild local reactions at the injection site occurred in 15 participants. Scoliosis was reported in 18 participants (9 in each group), typically appearing 6 months after treatment. These children’s annual growth rate was not significantly different from those without scoliosis. Mild lipoatrophy at the injection site occurred in two participants due to fixed injection site. After changing the injection site, lipoatrophy was resolved within three months. The treatments of these two patients were not interrupted. Fasting insulin, fasting blood glucose, and glycated hemoglobin levels did not show significant changes and remained within normal limits. No diabetes cases were reported during the treatment. The HOMA-IR showed a slight increase in the high-dose group but stayed stable. Lipid metabolism indicators (total cholesterol, triglycerides, and high-density lipoprotein cholesterol) remained stable and within normal limits in both groups. No anti-drug antibodies were observed in either group after 26 weeks of treatment.

This study suggested that higher dose of PEG-rhGH (0.20 mg·kg^–1·week^–1) is more effective in increasing growth velocity, without accelerating bone maturity, indicating an appropriate dose for future studies and clinical applications. The study also evaluated the serum IGF-1 levels, a standard method to assess the efficacy and safety of rhGH treatment.^[4] The levels significantly increased but stayed within the age-appropriate range, proving that the high-dose treatment did not cause long-term excessive IGF-1 levels. The differences in dosage were not associated with any significant changes in safety profile. Common problems with long-acting rhGH preparations, scoliosis, and lipoatrophy were observed,^[5] but these were manageable and did not disrupt treatment. The study was limited by relatively short follow-up. In conclusion, PEG-rhGH proved efficacy and safety in children with GHD.

Conflicts of interest

None.