Growth Response to Weekly Somapacitan Therapy in Children With GH Deficiency Is Related to GH Thresholds in GH Stimulation Testing

Abstract

GH deficiency (GHD) is a rare condition defined as the insufficient production or secretion of GH; when it occurs in children, it delays growth and attainment of predicted adult height. A combination of clinical history, auxological measurements, laboratory testing including IGF-I and GH stimulation tests (GHSTs), and magnetic resonance imaging studies is typically used to diagnose GHD in children [1, 2]. Despite currently being the gold standard for diagnosing GHD, GHSTs are often unreliable and lack reproducibility; thus, 2 GHSTs are typically conducted to assess peak GH concentration [2-5].

The traditionally accepted GH cutoff for diagnosing GHD in children is a peak concentration of <10 µg/L, but current consensus guidelines propose lowering this threshold for diagnosis to <7 μg/L [6], based on the harmonization of GH assay calibration resulting in a narrower target [3, 6, 7]. The <7 μg/L threshold has been adopted in most European and several other countries, including Australia, New Zealand, and Canada [3, 7-9], but global cutoffs range from 6 to 10 µg/L [10-12]. However, there is no standardized cutoff level for a GHD diagnosis, making the interpretation of GHST results arbitrary [2, 3, 13].

Although a cutoff level of <3 µg/L post-GHST is typically used to indicate severe GHD in adults in most countries [3], a clear definition of GHD severity in children is lacking [14]. Analyses of GHD concentrations in children vary across studies and countries. In a database analysis of children with GHD in Japan, severe GHD was defined as a peak GH concentration of ≤3 μg/L, moderate GHD as ≤6 μg/L, and mild GHD as >6 μg/L (but with at least 2 peak GH levels of ≤6 µg/L) [15]. A cross-sectional retrospective study of children in Turkey with idiopathic GHD treated with recombinant human GH (rhGH) therapy for at least 2 years categorized participants into 3 groups and evaluated the treatment response therein: peak GH concentration <3 μg/L, 3 to 7 μg/L, and 7 to 10 μg/L [16]. In an evaluation of patients with GHD in India over a 10-year period, magnetic resonance imaging abnormalities, such as small or absent anterior pituitary gland, thin or interrupted pituitary stalk, and ectopic posterior pituitary gland, were more commonly found in patients with severe GHD, defined as a GH peak concentration of <3 μg/L, than in children with mild or moderate GHD, classified as a GH peak concentration of 3 to 10 µg/L [17]. A study with children in Poland defined severe GHD as a GH peak concentration of ≤7 μg/L and moderate GHD as a peak concentration of >7 to <10 μg/L [18]. In a retrospective analysis of medical records of children with GHD in Brazil, severe GHD was defined as a peak GH concentration of <5 μg/L and mild GHD as a GH peak concentration of ≥5 to <10 μg/L [19].

Normal growth can usually be achieved in children with GHD with the use of GH replacement therapy. Until recently, treatment options were limited to daily subcutaneous injections of rhGH. Somapacitan (Novo Nordisk A/S) is a long-acting, once-weekly, US Food and Drug Administration–approved treatment of GHD in children and adults, administered subcutaneously [20-23]. In this subgroup analysis, we aimed to determine if there were differences in treatment response between 3 peak GH concentrations, as assessed at diagnosis, among children with GHD who were treated with daily GH or once-weekly somapacitan.

Material and Methods

The REAL4 study is a randomized, multinational, open-label, active-controlled parallel group phase 3 trial (NCT03811535) evaluating the efficacy and safety of somapacitan vs daily GH. The trial was conducted across 85 sites in 20 countries and consisted of a 52-week main phase and a 3-year safety extension. All parents or the children's legal representative provided informed consent to participate; assent was obtained from the children, as appropriate. Results from the main phase of the study and the first year of the safety extension period have been published previously [20, 24]. The trial protocol was conducted in accordance with the Declaration of Helsinki and the International Conference on Harmonization Guideline for Good Clinical Practice and was approved by local and national ethics committees, as appropriate.

To summarize, eligible participants were children with treatment-naïve prepubertal GHD (GH peak concentration of ≤10.0 μg/L) aged 2.5 years to 10 or 11 years at screening (females and males, respectively), impaired height (at least 2.0 SD below the mean) for chronological age and sex, impaired height velocity (HV) (annualized HV below the 25th percentile) for chronological age and sex, and IGF-I < −1.0 SD score (SDS). Eligible patients had a confirmed diagnosis of GHD within 12 months before screening, as determined by 2 different unprimed GH stimulation tests (defined as a peak GH level of ≤10.0 ng/mL, without prior exposure to GH therapy and/or IGF-1 treatment). Children were excluded if they were born small for gestational age or had other conditions or concomitant treatments that may affect growth. Patients were randomized 2:1 to receive somapacitan 0.16 mg/kg/week or GH (Norditropin, Novo Nordisk) 0.034 mg/kg/day, administered subcutaneously, during the first year. All patients received subcutaneous injections of somapacitan 0.16 mg/kg/week for the second year of the trial (ie, those who received daily GH during year 1 were switched to once-weekly somapacitan for year 2). The primary study endpoint was annualized HV (cm/year) at week 52. Other growth-related assessments included HV SDS (HVSDS), height SDS (HSDS), IGF-I SDS, and safety measures (adverse events). IGF-I samples were collected at different time points of the dosing interval to characterize the IGF-I profile response to weekly somapacitan treatment, allowing for the assessment of peak, mean, and trough IGF-I values. However, the observed average IGF-I SDS for somapacitan was based on samples specifically collected at week 52 and corresponded with the model-derived average IGF-I SDS. Somapacitan demonstrated similar efficacy and safety to daily GH over 52 weeks of treatment, which was sustained between 52 and 104 weeks [20, 24].

Subgroup Analyses

In this post hoc subgroup analysis of the REAL4 trial, we assessed the primary and secondary endpoints stratified by peak GH concentration in GHSTs at diagnosis. Three GH peak concentration subgroups were evaluated: ≤3 μg/L, >3 to <7 μg/L, and ≥7 to ≤10 µg/L. An upper threshold starting at 7 μg/L was selected, as this is the threshold currently used to establish a diagnosis of GHD in most European countries [8, 10].

Statistical Methods

This subgroup analysis was conducted post hoc. Descriptive statistics were used to characterize the year-1 and year-2 data for the 3 subgroups and interpreted in a clinical context.

Results

Study Population

A total of 200 children participated in the study, 132 of whom were randomly assigned to receive once-weekly somapacitan and 68 to receive daily GH somatropin (Norditropin) during the first 52 weeks of treatment. All but 1 participant completed 52 weeks of treatment; 194 patients completed the 2-year study period. At baseline, 61 patients had a GH peak concentration of ≤3 μg/L at diagnosis, 93 patients had a GH peak concentration of >3 to <7 μg/L, and 46 patients had a GH peak concentration of ≥7 to ≤10 µg/L (Table 1). Patients in the ≥7 to ≤10 µg/L GH peak concentration group were slightly older than the other subgroups. Mean HV and mean body mass index SDS were similar for patients randomized to daily GH and somapacitan within each GH peak concentration group. At baseline, mean HVSDS, HSDS, and IGF-I SDS were lowest for the ≤3 μg/L GH peak concentration group. Treatment adherence, as assessed by electronic diaries, ranged from a mean of 78.1% (daily GH group) to 95.2% (somapacitan treatment group) at week 52 and from 91.1% (switch group) to 91.9% (somapacitan/somapacitan group) between weeks 52 and 104 for the ≤3 μg/L GH peak concentration group. For the >3 to <7 μg/L peak concentration group, adherence ranged from 95.6% (daily GH group) to 95.9% (somapacitan treatment group) at week 52 and from 88.8% (switch group) to 90.9% (somapacitan/somapacitan group) between weeks 52 and 104. For the ≥7 to ≤10 µg/L GH peak concentration group, adherence ranged from 90.2% (daily GH group) to 93.8% (somapacitan treatment group) and from 84.9% (switch group) to 81.2% (somapacitan/somapacitan group) between weeks 52 and 104.

Efficacy Results

Height Velocity

The greatest HV was observed for the ≤3 μg/L GH peak concentration group at both weeks 52 and 104; HV increases were similar for the >3 to <7 μg/L and the ≥7 to ≤10 µg/L GH peak concentration groups (Fig. 1). At week 52 for those treated with daily GH and somapacitan, annualized mean HV ranged from 14.1 to 12.8 for the ≤3 μg/L GH peak concentration group, 10.9 to 10.5 for the >3 to <7 μg/L GH peak concentration group, and 10.0 to 10.6 cm/year for ≥7 to ≤10 µg/L GH peak concentration groups [Table 2 and Supplementary Table S1 (stratified by sex)] [25]. At week 104 for those in the daily GH/somapacitan and somapacitan/somapacitan groups, estimated mean HV ranged from 9.9 to 9.2, 8.1 to 8.0, and 8.0 to 7.9 cm/year, respectively, for the 3 peak GH concentration groups [Table 3 and Supplementary Table S2 (stratified by sex)] [25]. Figure 2A and 2B present HV and change in HV plotted against peak GH concentration. Pearson correlation coefficients are shown in Supplementary Table S3 [25].

Figure 1.

Observed HV from baseline to weeks 52 and 104 by GH peak concentration groups. Mean (SD) observed HV (cm/year) at baseline (week 0), week 52, and week 104 for the switch (daily GH/soma) and soma (soma/soma) groups, by GH peak concentration: (A) GH peak concentration ≤3 μg/L; (B) GH peak concentration >3 to <7 μg/L; (C) GH peak concentration ≥7 to ≤10 µg/L. Data are presented as mean with error bars representing SD.

Abbreviations: HV, height velocity; soma, somapacitan.

Open in new tabDownload slide

Figure 2.

HV (A) and change in HV (B) to GH peak concentration at week 52 and week 104 for the switch (daily GH/soma) and soma (soma/soma) groups.

Abbreviations: HV, height velocity; soma, somapacitan.

Open in new tabDownload slide

Other Growth-related and IGF-I Assessments

As with HV, the greatest changes from baseline and increases in mean HVSDS and HSDS at weeks 52 and 104 were seen in the ≤3 μg/L GH peak concentration group (Tables 2 and 3 and Supplementary Tables S1 and S2 [25]). Mean IGF-I SDS at week 52 ranged from −0.21 to 0.01, 0.32 to 0.29, and 0.15 to 0.66 for the ≤3 μg/L, >3 to <7 μg/L, and ≥7 to ≤10 µg/L GH peak concentration groups, respectively; change in IGF-I SDS from baseline ranged from 3.13 to 3.01, 2.11 to 1.96, and 1.87 to 2.26 at week 52 (Table 2 and Supplementary Table S1 [25]). At week 104, mean IGF-I SDS values were also within the normal range (−2 to +2), with the change in baseline varying from 2.81 to 2.11, 1.85 to 1.62, and 1.28 to 1.71 for the 3 GH peak concentration groups, respectively (Table 3 and Supplementary Table S2 [25]).

Safety Results

As previously reported, most adverse events experienced by study participants were mild or moderate in severity, and the majority were deemed unrelated to the trial product [20, 24]. Some of the adverse events reported for the patients in the ≤3 μg/L GH peak concentration group taking somapacitan throughout the entire 2-year period were endocrine disorders (n = 8; 21.6%), including secondary hypothyroidism, hypothyroidism, adrenocorticotropic hormone deficiency, and secondary adrenocortical insufficiency. There were no endocrine disorder–related adverse events reported in the >3 to <7 μg/L GH peak concentration group and only 1 in the >7 to ≤10 µg/L GH peak concentration group, suggesting that treatment unmasked some secondary endocrine disorders in the ≤3 μg/L GH peak subgroup. In reviewing the adverse events by IGF-I SDS, we did not observe that patients with IGF-I SDS >2 at 2 consecutive visits were more likely to have adverse events than patients with IGF-I SDS ≤2.

We also evaluated IGF-I SDS for low responders, defined as patients with HV <8 cm/year at week 52 based on previous literature [26, 27]. Of the n = 17 low responders, 6 were in the daily GH/somapacitan group (n = 68, 8.8%) and n = 11 were in the somapacitan/somapacitan group (n = 132, 8.3%). The low responders consisted of n = 5, n = 9, and n = 3 patients in the ≤3 μg/L, >3 to <7 μg/L, and ≥7 to ≤10 µg/L peak concentration groups, respectively. At baseline, the mean (SD) IGF-I SDS was −1.51 (0.67) and −1.50 (0.56) for each of the 2 treatment groups. Mean IGF-I SDS was −0.23 (1.15) for the daily GH/somapacitan group and −0.34 (1.02) for the somapacitan/somapacitan group at week 52; at week 104, mean IGF-I SDS was −0.47 (0.55) and −0.27 (0.80), respectively (n = 5 for the daily GH/somapacitan group and n = 8 for the somapacitan/somapacitan group at week 104).

Discussion

In this post hoc analysis of the REAL4 clinical trial, substantial clinical variability in response to GH therapy was observed in patients across the continuum of GHD. We found the greatest HV in the ≤3 μg/L GH peak concentration group, particularly at year 1. We also observed that increases in other growth-related assessments, including HVSDS and HSDS, were greatest in the ≤3 μg/L GH peak concentration group at both year 1 and year 2. Additionally, mean IGF-I SDS values were in the normal range for all 3 GH peak concentration subgroups at the end of the second year of the trial.

It is important for clinicians treating GHD to discuss with the parents and the child both the rationale for starting GH therapy as well as the anticipated growth response. As a result of this need to predict potential response to GH therapy, several prediction models have been developed over the past 20 years [28-30]. Additionally, models predicting the first-year HV and total height gain for patients with various GH-related disorders have been described in the literature, including those with GHD [30-33]. This REAL4 subgroup analysis is noteworthy, as it indicates that patients with more severe GHD benefit most from treatment. However, it is important to note that this analysis cannot speak to the nature of a relationship between GH peak concentration at baseline and the effect of treatment on adult height. Additionally, the REAL4 trial was designed to test treatment noninferiority differences between somatropin (Norditropin) and somapacitan and, thus, was not designed to test statistical differences between the peak GH concentration subgroups. Further, other factors, such as the child's age at diagnosis and etiology of GHD, may also affect treatment response.

Although the cutoffs may be slightly arbitrary and vary by country, including the GH peak concentration, subgroups of >3 to <7 μg/L and ≥7 to ≤10 µg/L reveal important insights into the spectrum of GHD and potential response to therapy. The children in the REAL4 trial with a GH peak concentration of ≤3 μg/L were typically diagnosed younger (at 6 years of age) than those in the highest GH peak concentration group and had more severe short stature, as indicated by their baseline characteristics; this underscores the importance of early referral and diagnosis. In their analysis of children with GHD treated with daily rhGH therapy for at least 2 years, Donbaloğlu et al demonstrated that response to treatment was greater in patients with lower GH peak values [16]. A longitudinal retrospective study of children treated for idiopathic GHD showed that the main factors in achieving a good long-term response were the response in the first year of treatment and having severe GHD, defined as a GH peak concentration of <3 μg/L [34]. In that study, Soliman et al found that after 1 year of GH therapy, changes in IGF-I SDS and HSDS from baseline were significantly higher in the patients with peak GH concentration of <3 µg/L than in those with peak GH concentration of >3 to <7 µg/L [35]. However, in a small study of children in Brazil with GHD treated with rhGH, Cardoso et al found a similar HV after 1 year in the patients with severe and mild GHD [19]. In our study, no difference in the growth response was seen between children with a GH peak between 3 and 7 μg/L or those with a GH peak of >7 μg/L. Analysis of the first-year HV against GH peak concentration suggests that a threshold peak GH exists above which the growth response plateaus. Confirming this would require analysis of a larger group of children with GHD. Additionally, the group with a peak GH concentration of <3 µg/L had the highest prevalence of organic pituitary disease than the other 2 peak GH concentration groups. Additional research would be needed to further understand how this may have affected the findings.

It is important to understand the different thresholds for assessing GHD severity to tailor treatment approaches accordingly. The traditionally accepted GH cutoff for diagnosing GHD in children is a peak concentration of <10 µg/L; however, improvements in GH assay methodology require a more standardized definition of GHD [6]. Further, a consensus statement by a collaboration of the Growth Hormone Research Society, the International Federation of Clinical Chemistry and Laboratory Medicine, the International Society for IGF Research, and The Pituitary Society made recommendations for the standardization of GH assays [36]. A specific understanding of thresholds for identifying GHD severity allows for tailoring therapy for individual patients to disease severity and clinical treatment goals, including the following: (1) growth promotion, which strives for optimal HV; (2) metabolic and cardiovascular health, which balances metabolic homeostasis; and (3) psychosocial and functional outcomes. Additionally, a more nuanced approach to GH therapy raises awareness of how treating GHD can potentially uncover other hormone deficiencies. Several clinical conditions may result in GHD, and the approach to further segment degrees of GHD may be valuable (eg, by providing insights into the potential development of secondary endocrinopathies). Moreover, because hypothyroidism may cause GHD, it is important first to exclude this diagnosis before testing for GHD.

One noteworthy area of the study was the small number of patients who were low responders to therapy despite their GH peak concentration subgroup status. It is difficult to predict how patients with mild GHD will respond to treatment. Additionally, we found that patients with IGF-I SDS >2 were not more likely to have adverse events than patients with lower IGF-I SDS. Gene expression could potentially be developed into a clinically applicable pretreatment test to offer insight into the mechanisms underlying the variable prediction of growth response to therapy and thereby increase the accuracy with which we can predict expected responses for individual patients [37].

Greater awareness among clinicians is needed regarding the identification of patients with the greatest severity of GHD and who may benefit the most from GH therapy. Further subgroup analysis of patient outcomes may be warranted as these individuals with GHD continue to mature. Future research with larger patient cohorts could potentially lead to a better understanding of the response to therapy across regions and countries; how the response may vary by GH peak concentration, age, and GHD etiology; and how to identify GH peak concentration groups that can be consistently assessed in clinical practice. Lastly, consideration should be given to using GH peak concentration as the primary definition of GHD severity.

This subgroup analysis has some limitations. Because this was a post hoc analysis, we cannot determine if the differences seen are due to treatment or differences in individuals that are driving responses to treatment. Further, the study population was skewed male, as seen in other studies of children with GHD [38, 39], limiting the understanding of differences across the 3 GH peak concentration subgroups by sex. Additionally, the lack of racial/ethnic diversity in the study may limit the generalizability to the overall population of children with GHD. Additionally, because the REAL4 trial was conducted at 85 sites across 20 countries, there may be variability in the different GH assays used. However, the trial employed rigid inclusion criteria for diagnosing GHD, and thus the variation in GH assays used is not expected to be a considerable factor. The plain language summary of this work is available at Backeljauw et al [25].

Conclusions

Patients with GHD in the ≤3 μg/L GH peak concentration group had increased HV and greater changes in height SDS and IGF-I SDS from baseline compared with the >3 to <7 μg/L and the ≥7 to ≤10 μg/L GH peak concentration groups at weeks 52 and 104 of the REAL4 phase 3 trial in children with GHD. There were no clear differences in growth response or IGF-I SDS between children with a GH peak concentration of >3 to <7 µg/L and ≥7 to ≤10 µg/L. Understanding the different thresholds for assessing GHD severity is important for the individualization of GH therapy. In future clinical trials, GHD severity stratification may help identify additional gaps or unmet needs to further improve the outcome of GH therapy in children with GHD.

Acknowledgments

The authors thank Rebecca Hahn, MPH, and Stephanie Burkhead, MPH, of KJT Group, Inc. (Rochester, NY, USA), for providing medical writing support, which was funded by Novo Nordisk in accordance with Good Publication Practice (2022) guidelines.

Funding

This research was funded by Novo Nordisk (Søborg, Denmark), and medical writing support was funded by Novo Nordisk (Plainsboro, NJ, USA).

Disclosures

P.F.B. is currently a consultant for/has received honoraria from Ascendis Pharma, BioMarin Pharmaceutical, Cavalry Biosciences, Ipsen Biopharmaceuticals, Novo Nordisk, and Upsher-Smith Laboratories. He also receives research support from the National Institutes of Health/National Heart, Lung, and Blood Institute (U01HL133883). B.S.M. is a consultant for AbbVie, Amgen, Ascendis Pharma, BioMarin, Endo Pharmaceuticals, Novo Nordisk, Pfizer, Sanofi, and Tolmar. He has received research support from Alexion, AbbVie, Aeterna Zentaris, Foresee, Lumos Pharma, Novo Nordisk, OPKO Health, Pfizer, and Sangamo Therapeutics. M.H., N.K., M.L.J., and A.P. are employees and stockholders of Novo Nordisk. J.M. has received research funding from Novo Nordisk, Pfizer, JCR Pharmaceuticals, and Ascendis Pharma. He has received honoraria from Novo Nordisk, Pfizer, and JCR Pharmaceuticals. C.B. is a consultant for/has received honoraria from Novo Nordisk and Sanofi. She has received research support from Merck.

Data Availability

Some or all datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.

Clinical Trial Registration

NCT03811535

References