Efficacy and safety of once-weekly somatrogon in adults with growth hormone deficiency: a randomized phase 3 study

Abstract

Purpose

This study assessed the efficacy and safety of once-weekly somatrogon, a long-acting growth hormone, in adults with growth hormone deficiency (aGHD).

Methods

A phase 3 randomized, double-blind, placebo-controlled study consisted of a 26-week double-blind period (Period 1), a 26-week open-label extension (OLE; Period 2), and a multi-year OLE (Period 3). Patients were randomized 2:1 to somatrogon or placebo in Period 1, with dosing adjusted for gender, age, and estrogen therapy. All patients received somatrogon in Periods 2 and 3. Primary endpoint was change in trunk fat mass (FM; baseline-Week 26). Secondary endpoints included changes in total FM, lean body mass (LBM), percentage change in trunk FM, and trunk FM. Changes in percent trunk FM relative to total trunk mass (FM + LBM), trunk LBM, and appendicular skeletal muscle mass were also evaluated in a post-hoc supplemental analysis. Safety assessments included adverse events (AEs) and laboratory evaluations.

Results

Of 389 patients screened, 202 were randomized, and 198 received treatment (somatrogon:133; placebo:65). Mean IGF-I SDS normalized following somatrogon initiation. There was no significant difference between somatrogon and placebo in change in trunk FM from baseline to Week 26 (-0.37 vs 0.03 kg; p=0.0821; primary endpoint) or in total FM. However, somatrogon significantly improved LBM, trunk FM as a percentage of total FM and the three supplemental endpoints. AE incidence was similar between groups, with most being mild to moderate in severity.

Conclusion

Somatrogon significantly improved several body composition parameters and was well tolerated overall in adults with GHD.

Clinicaltrials.Gov

NCT01909479; registration date: 25/07/2013

Supplementary Information

The online version contains supplementary material available at 10.1007/s11102-026-01661-1.

Introduction

Daily subcutaneous injections of recombinant human growth hormone (rhGH) are the predominant treatment option for patients with adult growth hormone deficiency (aGHD) and the safety and effectiveness of rhGH (somatropin) replacement are well established [1]. The goals of daily rhGH treatment are to normalize insulin-like growth factor I (IGF-I) levels, improve symptoms, comorbidities, and quality of life (QoL), and potentially decrease the risk of cardiovascular disease. Long-term daily GH replacement in adult patients with GHD has been shown to improve body composition, resulting in a reduction in fat mass and an increase in lean body mass [2, 3]. rhGH treatment has also been shown to improve bone health and reduce fracture risk before onset of osteoporosis [4], and improve QoL in patients with aGHD [1, 5].

Although the effectiveness of daily rhGH injections is well-documented, the burden of daily administration contributes to suboptimal adherence, which can limit the therapeutic benefits of rhGH treatment [6]. A discrete choice study demonstrated that both adult and pediatric patients with GHD preferred less frequent injections for treating GHD [7], suggesting the potential for improved adherence to treatment. Thus, several long-acting rhGH (LAGH) therapies have been developed, aimed at maximizing long-term treatment outcomes by significantly reducing injection frequency to improve adherence [8–11]. Several LAGH preparations have been approved for clinical use in the past few years, in pediatric and/or adult patients [12–15].

Somatrogon is a once-weekly LAGH that is approved in over 60 countries for the treatment of children and adolescents from 3 years of age with GHD. Somatrogon consists of rhGH fused to three copies of the carboxyl-terminal peptide of human chorionic gonadotropin, which extends the half-life of rhGH [16]. Phase 2 and 3 studies in children with GHD demonstrated that weekly somatrogon effectively normalized IGF-I levels and increased height velocity and height standard deviation scores (SDS), with efficacy and safety profiles similar to once-daily rhGH [16, 17]. These findings have recently been confirmed in an open-label phase 3 extension study [18] and two real world studies [19, 20].

Somatrogon was also evaluated in patients with aGHD in three phase 1 studies, a phase 2 study, and a pivotal phase 3 study. In the phase 1 studies conducted in 103 healthy adult volunteers, somatrogon demonstrated a favorable safety and tolerability profile [21]. The phase 2 study showed that somatrogon maintained IGF-I levels within the normal range while exhibiting a favorable safety profile [22]. The phase 3 study reported here evaluated the efficacy and safety of somatrogon in patients with aGHD.

Methods

Study design and treatment

This phase 3 study (NCT01909479) was a randomized, double-blind, placebo-controlled, parallel-group, multicenter trial conducted in patients with aGHD. The study was sponsored by OPKO Biologics Ltd., a subsidiary of OPKO Health, Inc., and conducted between August 2013 and August 2018 at 71 sites in Australia, Austria, France, Georgia, Germany, Greece, Hungary, Israel, Poland, Romania, Russia, Serbia, Slovakia, South Korea, Spain, Taiwan, Ukraine, and the USA.

Patients were randomly assigned 2:1 (using an interactive response technology system) to receive either somatrogon or placebo. The starting dose of somatrogon/placebo was based on gender and age (Supplementary Table 1), as well as on whether female patients were taking oral estrogen therapy. The somatrogon dose was modified (Supplementary Table 2) as required to maintain IGF-I levels within the target range of − 0.5 ≤ IGF-I SDS ≤ + 1.5. To preserve blinding, dose modifications were also implemented for placebo recipients. All study treatments were self-administered as a once-weekly subcutaneous injection (using a needle and syringe) to the thighs or abdomen, with injections sites rotated. Patients received training on how to self-inject somatrogon/placebo, with the patient administering the first dose under the supervision of study site staff. Patients were provided with sufficient study drug to maintain dosing until the next scheduled study visit and were required to document their injections (injection site, dose, and vial number) in a patient diary that was regularly reviewed by study staff and at monitoring visits.

The study consisted of three treatment periods. Period 1 was a 26-week randomized, double-blind, placebo-controlled treatment period, and Period 2 was a 26-week open-label extension (OLE) followed by a 2- to 8-week washout period. During Period 2 all patients received somatrogon; patients originally randomized to placebo in Period 1 began with a starting dose of somatrogon based on their gender, age, and estrogen therapy, whereas patients originally randomized to somatrogon continued on the same dose as they had completed during Period 1. Period 3 was designed as a multiyear long-term OLE (LT-OLE) study to evaluate long-term safety. During Period 3, patients continued receiving the same dose of somatrogon received at the end of Period 2. For clarity, throughout this manuscript we refer to patients who were originally randomized to somatrogon in Period 1 as the somatrogon group and those who were randomized to placebo in Period 1 as the placebo group.

The primary objective of the study was to demonstrate the clinical superiority of somatrogon over placebo in terms of decreasing trunk fat mass (FM, in kg) in patients with aGHD; this objective was defined, in part, according to findings from a clinical study described by Hoffman et al. [23]. Secondary objectives included assessing the efficacy of somatrogon (vs. placebo) for other body composition parameters and evaluating the safety and tolerability of somatrogon compared with placebo. Additional post hocsupplemental efficacy analyses were also undertaken following discussions with regulatory agencies (see Supplementary Methods for rationale).

This study was approved by the Institutional Review Board or Independent Ethics Committee of the participating institutions and followed the Declaration of Helsinki and International Council on Harmonisation Good Clinical Practice guidelines. Signed and dated informed consent was obtained from each patient prior to any participation in study-related procedures.

Patients

Adult men and women (≥ 23 to ≤ 70 years of age) with GHD were eligible for inclusion in this study. Patients using hormonal replacement therapy for other hypothalamic-pituitary axis deficiencies could be included provided they had been on an optimized and stable treatment regimen for at least 3 months prior to screening. At screening, patients had to have IGF-I levels ≤ − 1 SDS, adjusted for age and gender [24]. Patients had to have a body mass index (BMI) between 23.0 and 35.0 kg/m² (inclusive), with no GH therapy for at least 9 months prior to the study. Patients also had to be on a stable diet and exercise regime and were not to have plans to modify their diet or exercise for ≥ 12 months at the time of enrollment. Key exclusion criteria were overt history of diabetes mellitus, evidence of growth of a benign intracranial tumor within the last 12 months, suspected or diagnosed ongoing cancer or history of any cancer, and women who were pregnant or breast-feeding. Other typical inclusion/exclusion criteria are shown in the Supplementary Methods. While participating in the study, patients were informed to avoid making changes to their diet, exercise regime, or smoking habits. However, diet and exercise were not regularly monitored during the study.

Study assessments

Efficacy

The primary efficacy endpoint was change in trunk FM (in kg) from Baseline to Week 26, measured using dual-energy X-ray absorptiometry (DEXA). Secondary efficacy endpoints included changes (from Baseline to Week 26 and Week 52) in total FM, LBM, and percentage change in trunk FM. Change in trunk FM from Baseline to Week 52 was also a secondary endpoint. The following efficacy endpoints were evaluated using DEXA at screening, Week 15, and Week 26 in Period 1; Week 39 in Period 2; and at the end of Period 2 (Week 52/end-of-study): trunk FM, total FM, LBM, and bone density (only at screening and end-of-study OLE). IGF-I levels were measured centrally at screening, baseline and at day 3 or 4 post dose at Weeks 3, 7, 11, 15, 19, 23, 26, 32, 39, 45, and 52 by chemiluminescence IGF-I immunoassay [24]. The schedule of assessments conducted at each visit is shown in Supplementary Fig. 1.

Two post hoc supplemental efficacy analyses were conducted. The first analysis was performed to assess the impact of outliers (rationale and details provided below), and the second one evaluated additional body composition endpoints, including change in percent trunk FM (defined as 100 times trunk FM divided by the sum of trunk FM and trunk LBM), change in trunk LBM, and change in appendicular skeletal muscle mass (ASMM).

Safety

Safety evaluations included all adverse events (AEs), electrocardiogram, fundoscopy, and laboratory assessments (consisting of glucose metabolism, IGF-I levels, and immunogenicity). Treatment-emergent AEs (TEAEs) were defined as any AE that occurred on or after initiation of treatment. All AEs were coded using the Medical Dictionary for Regulatory Activities (MedDRA 19.1). The intensity or severity of an AE was characterized as mild, moderate, or severe. Local injection site reactions were recorded as AEs, and injection site pain scores of 8 or higher were recorded as an AE. Anti-drug antibodies (ADAs) were assessed from serum samples using qualitative, validated methods [16].

Statistical analyses

The safety analysis set consisted of all randomized patients who received at least one dose of somatrogon or placebo. The efficacy analysis set consisted of all patients in the safety population who had a primary efficacy evaluation at Week 26. The efficacy endpoints and changes from Baseline were summarized by descriptive statistics for each treatment group (see Supplementary Methods for details of the statistical comparisons). Safety endpoints were summarized using descriptive statistics.

Following database lock and unblinding of the data, unusually high total body mass and change in trunk FM were noted for six patients. As a result, a post hoc sensitivity analysis was conducted to assess the impact of outliers on the results of the study. These post hoc sensitivity analyses consisted of (i) identification of outliers, (ii) tipping-point analysis, and (iii) non-parametric approaches (see Supplementary Materials for further details).

Results

Patients and treatment

A total of 389 patients were screened for Period 1, with 202 patients randomized (Fig. 1) and 198 patients receiving treatment. A higher percentage of patients in the somatrogon group completed Period 1 compared with the placebo group (97.0% vs. 89.2%). The most frequently reported reasons for discontinuation in Period 1 were withdrawal of consent and AEs. In Period 2, completion rates were similar in both groups (97.7% vs. 96.6%), with discontinuations primarily due to withdrawal of consent, AEs, and being lost to follow-up. A total of 161 patients received treatment in Period 3 and 118 (73.3%) completed the LT-OLE study (up to 4 years in duration). Among the 43 patients who discontinued from the LT-OLE study, the most frequently reported reason for discontinuation was withdrawal of consent.

Fig. 1.

Patient disposition. LT-OLE=long-term open-label extension

The demographic and baseline characteristics of the safety population are presented in Table 1. Overall, baseline characteristics were well balanced across both groups; patients were predominantly White, male, and the majority were from Europe. The mean age was similar between the somatrogon and placebo groups. A higher proportion of women in the somatrogon group versus the placebo group had adult-onset GHD (41/57 [71.9%] vs. 12/25 [48.0%], respectively), and a higher proportion received oral estrogen (25/57 [43.9%] vs. 8/25 [32.0%]).

Table 1.

Patient demographics and Baseline characteristics (safety population)

	Somatrogon			Placebo
	Male N = 76	Female N = 57	Total N = 133	Male N = 40	Female N = 25	Total N = 65
Age, mean (SD), years	43.2 (13.0)	44.7 (12.0)	43.9 (12.6)	43.0 (13.9)	43.1 (11.5)	43.1 (12.9)
Race, n (%)
White	64 (84.2%)	46 (80.7%)	110 (82.7%)	30 (75.0%)	23 (92.0%)	53 (81.5%)
Black or African American	1 (1.3%)	0	1 (0.8%)	1 (2.5%)	1 (4.0%)	2 (3.1%)
Asian	7 (9.2%)	8 (14.0%)	15 (11.3%)	7 (17.5%)	1 (4.0%)	8 (12.3%)
Other	1 (1.3%)	1 (1.8%)	2 (1.5%)	0	0	0
Not Reported	3 (3.9%)	2 (3.5%)	5 (3.8%)	2 (5.0%)	0	2 (3.1%)
Weight, mean (SD), kg	80.7 (15.4)	67.8 (10.7)	75.2 (15.0)	85.6 (15.1)	65.3 (14.7)	77.8 (17.9)
BMI, mean (SD), kg/m2	27.5 (3.4)	26.3 (2.9)	27.0 (3.2)	28.3 (3.8)	26.6 (2.8)	27.7 (3.5)
GHD onset type, n (%)
Childhood-onset	38 (50.0%)	16 (28.1%)	54 (40.6%)	14 (35.0%)	13 (52.0%)	27 (41.5%)
Adult-onset	38 (50.0%)	41 (71.9%)	79 (59.4%)	26 (65.0%)	12 (48.0%)	38 (58.5%)
Oral estrogen, n (%)
Yes	0	25 (43.9%)	25 (18.8%)	0	8 (32.0%)	8 (12.3%)
No	0	32 (56.1%)	32 (24.1%)	0	17 (68.0%)	17 (26.2%)
NA (male)	76 (100.0%)	0	76 (57.1%)	40 (100.0%)	0	40 (61.5%)
Region, n (%)
USAa	15 (19.7%)	16 (28.1%)	31 (23.3%)	8 (20.0%)	7 (28.0%)	15 (23.1%)
Europeb	54 (71.1%)	34 (59.6%)	88 (66.2%)	26 (65.0%)	17 (68.0%)	43 (66.2%)
Asia	7 (9.2%)	7 (12.3%)	14 (10.5%)	6 (15.0%)	1 (4.0%)	7 (10.8%)

^a Region USA includes USA and Australia

^b Region Europe includes Europe and Israel

BMI=body mass index; GHD=growth hormone deficiency; SD=standard deviation

Efficacy

There was no statistically significant difference between the somatrogon and placebo groups in terms of change in trunk FM from Baseline to Week 26 (p = 0.0821; primary efficacy endpoint) (Fig. 2A and Supplementary Table 3). Similarly, there was no statistically significant difference between groups in terms of change in total FM from Baseline to Week 26 (p = 0.5770) (Fig. 2B). Change in total FM from Baseline to Week 52 was not statistically significant for either the somatrogon group or the placebo group (who switched to somatrogon at Week 26).

Fig. 2.

Primary and secondary endpoints of efficacy analysis. LSMs and P-value for comparison between somatrogon and placebo are from a mixed model for repeated measures (MMRM), where n represents the number of randomized patients included in the MMRM. FM = fat mass; LBM=lean body mass; LSM=least squares mean

Change in LBM from Baseline to Week 26 showed a statistically significant increase in the somatrogon group compared with the placebo group (p < 0.0001) (Fig. 2C). Change in LBM from Baseline to Week 52 was not statistically different for the somatrogon group. However, the placebo group had a statistically significant increase in LBM from Week 26 to Week 52.

Change in trunk FM as a percentage of total FM from Baseline to Week 26 showed a statistically significant decrease in the somatrogon group compared with the placebo group (p < 0.0001) (Fig. 2D). Change in trunk FM as a percentage of total FM from Baseline to Week 52 was not statistically significant for the somatrogon group, but there was a statistically significant decrease in patients originally randomized to placebo.

Post hoc sensitivity analyses

A post hoc sensitivity analysis was conducted to assess the impact of outliers on the study results. Six outliers (two patients from the somatrogon group and four from the placebo group) were identified in this analysis. Following exclusion of the most extreme outlier as part of the tipping point analysis, the somatrogon group showed a statistically significant difference (p = 0.0082) in change in trunk FM from Baseline to Week 26 compared with the placebo group. The difference between the somatrogon and placebo groups continued to be statistically significant following exclusion of the most extreme outlier pair (p = 0.0131) as well as with each additional outlier pair.

Several non-parametric approaches (rank-based ANOVA, stratified Wilcoxon rank-sum, and aligned rank test) were used to analyze the primary endpoint, and four approaches showed a statistically significant decrease in trunk FM (kg) from Baseline to Week 26 in the somatrogon group compared with the placebo group. For the three secondary endpoints (change in total FM, change in LBM, and change in trunk FM as a percentage of total FM), sensitivity analyses for outliers did not reverse the original interpretation of the study results (reported above).

Supplemental efficacy analyses

The somatrogon group showed a statistically significant decrease in percent trunk FM from Baseline to Week 26, compared with the change observed in the placebo group (− 1.33 vs. 0.19; p = 0.0002) (Fig. 3A and Supplementary Table 4). The somatrogon group had a statistically significant increase in trunk LBM (from Baseline to Week 26) compared with the placebo group (0.66 vs. − 0.17; p < 0.0001) (Fig. 3B and Supplementary Table 4). Similarly, the change in ASMM from Baseline to Week 26 showed a significantly greater increase in the somatrogon group compared with the placebo group (0.62 vs. 0.21; p = 0.0191) (Fig. 3C and Supplementary Table 4).

Fig. 3.

Supplemental efficacy analysis. ASMM=appendicular skeletal muscle mass; CI=confidence interval; FM = fat mass; LBM=lean body mass; LSM=least squares mean

IGF-I

 Mean (SD) IGF-I SDS values at Baseline for the somatrogon and placebo groups were similar, being − 2.63 (1.13) and − 2.56 (1.24), respectively. Up to Week 26, 97.7% of somatrogon-treated patients achieved IGF-I SDS within the target range ([− 0.5, 1.5], inclusive) at least once. Over the same period, only 6.2% of patients in the placebo group achieved IGF-I SDS within the target range at least once; the difference between the somatrogon and placebo groups was statistically significant (p < 0.0001). Mean IGF-I SDS values in the somatrogon group increased to within the target range approximately 7 weeks after the initial dose and remained at or above zero through Week 26 and Week 52 (Supplementary Fig. 2). In contrast, mean IGF-I SDS values in the placebo group remained below − 0.5 to Week 26 (Supplementary Fig. 2). After switching to somatrogon at the beginning of Period 2, mean IGF-I SDS values in the placebo group increased to within the target range between Weeks 39 and 52 (Supplementary Fig. 2). Mean IGF-I remained between − 0.5 and 2 for both groups during Period 3 (Supplementary Fig. 2).

Safety

In Periods 1–3 the incidence of all-causality AEs was similar between the somatrogon and placebo groups (Period 1: 63.9% vs. 69.2%; Period 2: 56.4% vs. 60%; Period 3: 64.9% vs. 60.0%, respectively) (Table 2). Across all three periods the majority of AEs (> 80%) in each treatment group were mild to moderate in severity (Table 2), and most AEs were considered unrelated or unlikely related to treatment. In Period 1 the three most frequently reported TEAEs (somatrogon vs. placebo) were injection site pain (9.0% vs. 13.8%), headache (8.3% vs. 7.7%), and upper respiratory tract infection (4.5% vs. 6.2%) (Supplementary Table 5). The corresponding TEAEs in Period 2 were nasopharyngitis (7.5% vs. 6.2%), headache (8.3% vs. 3.1%), and arthralgia (2.3% vs. 9.2%) (Supplementary Table 6). All patients received somatrogon in Period 2, and in terms of AEs there were no differences in the frequency, severity, and/or the relationship to study drug compared to that observed during Period 1 and no differences between cohorts (i.e., patients originally randomized to somatrogon vs. placebo). The three most frequently reported TEAEs (somatrogon vs. placebo) in Period 3 were nasopharyngitis (11.7% vs. 10.0%), upper respiratory tract infection (9.0% vs. 6.0%), and headache (6.3% vs. 8.0%) (Supplementary Table 7). Compared with Period 1, the incidence of injection site pain was lower in Periods 2 and 3, during which all patients received somatrogon.

Table 2.

All-causality treatment-emergent AEs

Number (%) of patients	Period 1 (Baseline to Week 26)		Period 2 (Week 27 to Week 52)		Period 3 (LT-OLE)
	Somatrogon (n = 133)	Placebo (n = 65)	Somatrogon (n = 133)	Placebo (n = 65)	Somatrogon (n = 111)	Placebo (n = 50)
Number of AEs	313	160	224	145	431	246
Patients with AEs	85 (63.9)	45 (69.2)	75 (56.4)	39 (60.0)	72 (64.9)	30 (60.0)
Patients with serious AEs	4 (3.0)	5 (7.7)	5 (3.8)	3 (4.6)	10 (9.0)	8 (16.0)
Patients with severe AEs	2 (1.5)	5 (7.7)	1 (0.8)	2 (3.1)	7 (6.3)	5 (10.0)
Patients with permanent discontinuation due to AEs	3 (2.3)	2 (3.1)	0 (0.0)	1 (0.5)	4 (3.6)	3 (6.0)
Patients with temporary discontinuation or dose reduced due to AEs	8 (6.0)	1 (1.5)	0 (0.0)	4 (6.2)	7 (6.3)	6 (12.0)

AE adverse event, LT-OLE long-term open-label extension

In Periods 1 and 2 the incidence of serious AEs (SAEs) ranged from 3.0% to 7.7% in each group (Table 2); none were considered treatment-related. In Period 3, SAEs were reported in 9.0% of the somatrogon group and 16.0% of the placebo group (Table 2), with three SAEs (hemianopia, pituitary-dependent Cushing’s syndrome, and hemangioma) considered treatment-related. No deaths were reported for either treatment group in Periods 1 and 2. In Period 3, one death due to myocardial infarction was reported and was not considered related to the treatment or any study procedure.

Overall, 13 patients were permanently discontinued from the study owing to TEAEs (all-causality). Treatment-related AEs that resulted in permanent discontinuation were reported in three patients in Period 1 (pain, carpal tunnel syndrome, and injection site pain) and two patients in Period 3 (face oedema and two instances of periorbital oedema). Treatment-related AEs that resulted in temporary discontinuation or dose reduction were reported in three patients in Period 1 (nerve root compression, rash pruritic, vertebral foraminal stenosis, vertebral osteophyte, headache, fatigue, and ocular discomfort), one patient in Period 2 (joint stiffness), and one patient in Period 3 (paresthesia).

Across all three periods there were no discernible differences among treatment groups (from Baseline values) in glucose, HbA1C, or insulin (Supplementary Results). Additional, infrequent AEs related to electrocardiograms and fundoscopy are detailed in Supplementary Results.

Immunogenicity

During Periods 1 and 2 the overall incidence of ADAs was 14.3% (19/133) in the somatrogon group and 7.7% (5/65) in the placebo group. In Period 3 the incidence of anti-somatrogon and anti-hGH ADAs in all patients at Year 4 Month 6 was 17.4% (28/161) and 15.5% (25/161), respectively. Across all three periods there was largely overlapping specificity between anti-somatrogon ADA and anti-hGH antibodies. Neutralizing antibodies (NAbs) against somatrogon and hGH were detected in a single patient in Periods 1 and 2. In this patient the development of NAbs was associated with loss of IGF-I response over time despite maintenance of drug exposure (dose increased); this patient did not demonstrate any improvement in body composition parameters. Two patients tested positive for NAbs to hGH in Period 3, with NAb positivity persisting from Visit 18 (Year 1 Month 6) until the last study visit, Visit 30 (Year 4 Month 6); for both of these patients, the development of NAbs did not affect IGF-I levels or body composition parameters. For the majority of patients, IGF-I levels and somatrogon efficacy did not appear to be affected by the presence of ADAs.

Discussion

This phase 3, randomized, double-blind, placebo-controlled study did not meet its primary endpoint as the change in trunk FM from Baseline to Week 26 was not significantly different between the somatrogon and placebo groups. Secondary endpoints such as change in total FM (Baseline to Week 26) and percent change in trunk FM (Baseline to Week 52) were also not significantly different between treatment groups. Although the primary endpoint was not met, somatrogon treatment significantly improved several other body composition parameters (versus placebo), including LBM, trunk FM as a percentage of total FM, percent trunk FM, trunk LBM, and ASMM, and maintained IGF-I SDS within normal limits for the vast majority of patients.

We consider that the primary endpoint may not have been met due to the presence of outliers in both treatment groups. The post hoc sensitivity analyses identified six outliers, and exclusion of the most extreme outlier pair resulted in a statistically significant difference between the somatrogon and placebo group in terms of change in trunk FM from Baseline to Week 26. Although patient demographics and clinical characteristics were similar between treatment groups at Baseline, significant alterations in diet or physical activity during the study might have affected the study outcomes. As an example, a patient who was randomized to the placebo group significantly increased their physical activity after the Baseline examination, which resulted in substantial weight loss over the course of the study; it is important to emphasize that the study was not sufficiently powered to account for outliers of this magnitude. Another factor that might have contributed to the primary endpoint not being met was the somatrogon group having a higher proportion of female patients with adult-onset GHD (somatrogon: 71.9%; placebo: 48.0%) and a higher proportion of female patients receiving oral estrogen (somatrogon: 43.9%; placebo: 32.0%). There is evidence to indicate that women with GHD have a lower response to GH therapy compared with men with GHD and this is especially pronounced in women on oral estrogen [25–27]. Lastly, the dose of somatrogon used in this study may have also been a factor in the primary endpoint not being met. Previous studies of daily rhGH treatments demonstrated that higher doses of GH resulted in greater reductions in FM [28–30]. Although it would not be appropriate to make direct comparisons between daily rhGH treatments and somatrogon (due to differences in formulations, dosing paradigms, and exposure profiles), these studies suggest that the dosage of somatrogon employed in the current study may have also contributed to the absence of a statistically significant effect on trunk FM in somatrogon-treated patients.

When the protocol for this study was being developed, choosing to use change in trunk FM as the primary endpoint was driven largely by findings from a study of patients with aGHD treated with the LAGH depot GH [23], which found the largest treatment effect was reductions in trunk and visceral FM. More recently there has been greater recognition of the need to evaluate all body composition parameters to enable a more comprehensive assessment of GH efficacy. After this study was completed, a post hoc supplemental analysis of additional body composition endpoints was conducted. In combination with the efficacy data generated from the prespecified analysis, data from this post hoc supplemental analysis provided a more complete assessment of the efficacy of somatrogon for the treatment of aGHD. The recent development of a minimum dataset for aGHD studies should help researchers identify the most appropriate endpoints in future studies and enable comparability of data across studies [31].

Somatrogon treatment resulted in a statistically significant increase in both LBM and trunk LBM. This is consistent with results from other clinical studies with somapacitan [13] and lonapegsomatropin [32], two LAGH therapies approved for the treatment of aGHD. The phase 3 study of somapacitan showed that treatment with somapacitan and daily rhGH significantly increased mean LBM compared with placebo [13]. The mean increase in LBM for somapacitan and daily rhGH was of similar magnitude to that observed for somatrogon (+ 1.38 kg vs. + 1.44 kg vs. + 1.30 kg). Similarly, treatment with lonapegsomatropin also resulted in a significant increase in mean LBM (1.6 kg) compared with placebo (− 0.1 kg). A metanalysis of placebo-controlled trials of GH treatment in patients with aGHD also reported a significant increase in LBM [33]. Although LBM is not a direct cardiovascular risk factor, increased LBM is associated with a lower mortality risk in elderly patients and is negatively associated with cardiometabolic disorders, particularly in women [34–36]. The significant increase in mean ASMM (+ 0.6 kg) following somatrogon treatment was also observed in the somapacitan study; mean ASMM increased by + 0.56 and + 0.51 in patients treated with somapacitan and daily rhGH, respectively [13].

This study also revealed that somatrogon was generally well tolerated. Across all three treatment periods the incidence of AEs was similar between both treatment groups, and the majority of AEs were mild or moderate in severity. Most AEs that occurred with the highest frequency during somatrogon treatment were also observed in previous studies of daily rhGH and other LAGH treatments. The safety profile of somatrogon in this study was reflected in previous studies of somatrogon in pediatric populations [16, 17, 37, 38], with no new safety signals observed. Taken together, although aGHD is a different treatment population from pediatric GHD, the overall safety profile of somatrogon in patients with aGHD was consistent with that established for daily rhGH products in previous studies. Additionally, although ADAs and NAbs were observed in a proportion of the treatment population, their presence did not affect the treatment efficacy or IGF-I levels in the vast majority of patients.

Strengths of this study include the large patient population and extended study duration, with most patients evaluated for at least 3 years. The study also had a diverse patient population, recruited from multiple countries across several global regions. However, as previously discussed, a key limitation of the study was the use of a single body composition parameter (trunk FM) as the prespecified primary endpoint. Another limitation of the study design was that diet and exercise were not carefully monitored during the study so that it is possible that changes in body fat could have been influenced by patient behavior to a greater extent in one group relative to the other. Other limitations were the presence of an unexpected number of outliers (which increased variability and reduced the precision of the estimated treatment effect), the exclusion of patients with diabetes mellitus, and the lack of comparison with daily rhGH treatment, although it should be noted that the use of a placebo arm was a regulatory requirement at the time of the study.

Conclusion

This study did not meet its primary endpoint as there was no statistical difference between the somatrogon and placebo groups in terms of change in trunk FM from Baseline to Week 26. A post hoc analysis suggested that somatrogon’s potential efficacy may have been masked by the presence of outliers. Notably, somatrogon treatment resulted in significant improvements in several other body composition parameters (LBM, percent trunk FM, trunk LBM, and ASMM) compared with placebo. Somatrogon was well tolerated overall, and its safety profile in aGHD was generally consistent with that observed for daily rhGH and with other LAGH preparations. Requiring less frequent dosing, but with similar safety and tolerability to once-daily rhGH, once-weekly somatrogon may provide a more convenient treatment option and may enhance adherence to GH replacement in patient with aGHD.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

G.H. conceived of the study and G.H. and A.B-I designed the study. A.M., J.C., and M.C. analyzed the data. M.F., B.M.K.B., S.M.W., C.J.S., C.E.G., M.B., C.W., I.B., F.F.C., A.G-J., and K.C.J.Y. were investigators in the study. All authors interpreted the data. All authors participated in writing the original draft and reviewed, edited, and approved the manuscript for submission.

Funding

OPKO Health, Inc. (study); Pfizer Inc (medical writing support).

Data availability

Upon request, and subject to review, Pfizer will provide the data that support the findings of this study. Subject to certain criteria, conditions and exceptions, Pfizer may also provide access to the related individual de-identified participant data. See https://www.pfizer.com/science/clinical-trials/trial-data-and-results for more information.

Declarations

Disclosure

Pfizer’s generative artificial intelligence (AI)-assisted technology was used in the production of this manuscript. Specifically, the AI tool was used to help develop the first draft of the methods and results in the manuscript. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Clinical Trial Registration number

NCT01909479.

Competing interests

MF: Research grant: Ascendis; Consulting fee: Novo Nordisk; Member of Pituitary editorial board. BMKB: Research grant: Ascendis; Consultant: Ascendis, Pfizer, Novo Nordisk; at the time of this study which closed over 6 years ago, BMKB served as the PI of a research grant to Massachusetts General Hospital from Prolor and as an occasional consultant to the company. SMW, CG, IB, FFC: No conflicts of interest. CJS: Research grant: OPKO Health, Inc.; Consultant/speaker: Prolor/OPKO Health Inc., Novo Nordisk, Sandoz Hexal, Pfizer, Ascendis, Crinetics. MB: Research grant: OPKO Health, Inc.; Speaker/consultant: Novo Nordisk, Sandoz Hexal, Pfizer, Ascendis, IDS, Roche, Diasorin. CW: Research grant: Ascendis. AG-J: Speaker: Recordati, Novo Nordisk, Ipsen, Ascendis, Pfizer. GH, AB-I: Employee of OPKO Biologics. AM: Employee of OPKO Pharmaceuticals LLC. JC: Employee of OPKO Health Inc. MC: Employee of Pfizer. DLT, CTT: Employee of and owns stock in Pfizer. KCJY: Research grant: Ascendis, Novo Nordisk; Consultant: Novo Nordisk, Ascendis, Pfizer; Member of Pituitary editorial board.