Efficacy and safety of GH treatment in Japanese pediatric patients with SHOX deficiency: an open-label extension study

Abstract.

We have previously published efficacy and safety data up to 24 mo of GH treatment (0.35 mg/kg/wk) for SHOX-related short stature in 19 Japanese pediatric patients before pubertal entry (jRCT2080223889). Here, we report similar efficacy and safety data up to 48 mo obtained by an extension study for the same 19 patients before and after pubertal entry. The height SDS for chronological age (CA) was increased during the study period in 18 of the 19 patients and exceeded –2.0 in 17 of the 19 patients at the latest visit. Height velocity obviously increased in the first year. Bone age (BA) progressed faster than CA, and ΔBA/ΔCA remained above 1.0 throughout the study period. Serum IGF-1 SDS increased during the first 12 mo and remained relatively stable thereafter. GH treatment was generally well-tolerated with no severe adverse events, although one patient showed a markedly high serum IGF-1 value which was coped with the reduction in GH dosage to 0.23 mg/kg/wk, and two patients manifested insulin resistance. While further studies are required to clarify the long-term efficacy and safety and adult height in GH-treated patients, the results argue for the beneficial effects of GH therapy in SHOX-related short stature.

Highlights

● GH therapy was effective and safe up to 48 mo in patients with SHOX deficiency.

● The height SDS for chronological age was above –2.0 in 17 of the 19 patients at the latest visit.

● GH treatment was well-tolerated, with no severe adverse events.

Introduction

SHOX located in the short arm pseudoautosomal region of the X and Y chromosomes is a homeobox-containing gene involved in skeletal growth and development (1,2,3). SHOX deficiency is primarily caused by pathogenic sequence variants such as nonsense, missense, and indel variants in the coding exons and by pathogenic copy number variants such as microdeletions and microduplications disrupting the integrity of the SHOX coding/enhancer regions, in patients with an apparently normal karyotype (1,2,3). Heterozygous SHOX deficiency leads to diverse clinical features, including idiopathic short stature (ISS) phenotype and Léri-Weill dyschondrosteosis (LWD) phenotype characterized by Madelung deformity and relatively severe short stature (1,2,3). Previous studies have shown that the frequency of SHOX deficiency is 1.1~22.2% in patients with ISS phenotype and 50~90% in patients with LWD phenotype (1,2,3,4). In Japan, a single-center study has identified SHOX deficiency in 3.8% (12/312) of ISS patients and in 50% (8/16) of LWD patients (4).

Recombinant human growth hormone (GH) has been utilized in pediatric patients with short stature due to SHOX deficiency in multiple countries (3), promoting the statural growth to a degree similar to that in Turner syndrome (5, 6). In Japan, a multicenter, open-label, randomized, parallel-group, phase 3 short-term (24 mo) study was performed in 19 prepubertal patients with SHOX-related short stature using GH (Growject^®), revealing that the ∆height standard deviation score (SDS) for chronological age (CA) as the primary endpoint was significantly larger in ten GH-treated patients than in nine GH-untreated patients (0.92 ± 0.07 vs. 0.13 ± 0.04, P < 0.01), with no serious adverse events (AEs) (7). Consequently, Growject^® was approved as a treatment option for SHOX-related short stature in 2023. Then, the Japanese Society for Pediatric Endocrinology has published a recommendation for the diagnosis and treatment of SHOX deficiency (3).

Here, we report efficacy and safety data up to 48 mo obtained by an extension study performed for the same 19 pediatric patients.

Methods

Patients and study design

We performed a multicenter, open-label, single-arm extension study that provided continuing GH treatment to pediatric patients who had completed a 24-mo antecedent randomized study (jRCT2080223889) (7). In the antecedent study, we recruited 19 patients who satisfied the several selection criteria, including: [1] heterozygous SHOX deficiency confirmed by genetic analyses, [2] normal karyotype, [3] short stature with a height of ≤ –2.0 SD, (4) prepubertal stage, and [4] lack of other discernible diseases or medications that could affect statural growth. Subsequently, we divided the patients into the following two groups with comparable baseline characteristics including age, sex, height SDS for CA, and bone age (BA): [1] an NT-GH group (n = 9) receiving no treatment (NT) for the first 12 mo followed by GH treatment using subcutaneous injections of Growject^® at a dose of 0.35 mg/kg/wk once daily for the next 12 mo, and [2] a GH-GH group (n = 10) receiving GH treatment for 24 mo continuously (Fig. 1A). Then, we examined efficacy and safety of GH treatment using relevant data obtained before pubertal entry (thus, the data obtained after pubertal entry in three patients of the NT-GH group and in two patients of the GH-GH group were treated as missing in the antecedent study).

Fig. 1.

The design of the antecedent study (A) and that of the extension study (this study) (B). In the NT-GH group, the antecedent study begins with the NT observation period, whereas the extension study starts with the GH treatment. Thus, the timeline of the NT-GH group is different between the two studies. The extension study utilizes the combined data from both of the NT/GH group and GH/GH group. NT, no treatment; and GH, growth hormone.

In this extension study, we examined efficacy and safety of GH treatment performed for the same 19 pediatric patients (female patients 1–12 and male patients 13–19) before and after pubertal entry, using the data obtained from the start of GH treatment in the antecedent study (Fig. 1B). Since this extension study dealt with the combined data after GH therapy in both the NT/GH group and GH/GH group, the timeline in the NT-GH group was different between the antecedent study and the extension study (e.g., month 0, month 12, and month 24 in the antecedent study corresponded to month –12, month 0, and month 12 in the extension study, respectively). Because of the two-year enrollment duration in the antecedent study, the observation period in the extension study was variable among patients depending on the timing of the enrollment, and the number of patients examined at each study point became small with time, especially in the NT/GH group; indeed, the patient number was four at month 48 and 0 at month 60, while height data at month 54 were available in three patients 2, 3, and 15. Thus, we analyzed the data obtained up to 48 mo from the initiation of GH treatment in the antecedent study.

Assessment of efficacy

We assessed height, height SDS for CA, and ∆height SDS for CA every ~3 or ~6 mo, height velocity, ∆BA,ΔBA/ΔCA, and serum IGF-1 SDS every ~12 mo, and pubertal development every ~6 mo. Height SDS was calculated using age- and sex-matched growth references for Japanese children (8). BA was evaluated using the Tanner-Whitehouse 2 method standardized for Japanese children (9). Serum IGF-1 SDS was obtained using Japanese reference data (10). Pubertal entry was regarded as positive when breast development of Tanner stage 2 in female patients and genital development of Tanner stage 2 or testis size of ≥ 3–4 mL in male patients were observed (11). Since Shapiro-Wilk test showed normality in several raw variables and non-normality in other raw variables, we expressed the data as the mean ± SE which shows the distribution of the mean and can be utilized irrespective of the normality or non-normality of the raw variables, as employed in our previous report (7).

Assessment of safety

We evaluated the occurrence of AEs and adverse drug reactions (ADRs). AEs were coded using the preferred terms in the Medical Dictionary for Regulatory Activities Japanese translation version 26.1, and were classified as mild (not interfering with daily activities), moderate (interfering with daily activities and requiring therapeutic interventions), or severe (seriously disrupting daily activities). ADRs were defined as AEs considered to be related to the study drug by the investigators. In addition, we performed an oral glucose tolerance test (OGTT) using 1.75 g/kg of glucose (maximum of 75 g) and examined biochemical data including HbA1c and thyroid hormones every ~12 mo.

Ethics

This study was approved by the Institutional Review Board Committee of each investigator who participated in this study, and was performed after obtaining written informed consent from the parents and informed assent from the patients. This study was conducted according to the principles of the Declaration of Helsinki, Good Clinical Practice (Ordinance of the Ministry of Health and Welfare No. 28), Good Post-Marketing Study Practice (Ordinance of the Ministry of Health and Welfare No. 171), and their amendments.

Results

Patients

The baseline characteristics of the 19 patients at the initiation of GH treatment are shown in Table 1. One female patient 12 in the NT/GH group had entered puberty during the NT period (between month –12 and month –6 in Fig. 1B); her height SDS for CA increased from –2.04 at the time of enrollment (month –12) to –1.77 at the start of GH therapy (month 0), and exceeded –2.0 during the NT period. The height SDS for CA also increased from –2.04 at the time of enrollment (month –12) to –1.84 at the start of GH therapy (month 0), and exceeded –2.0 during the NT period in another female patient 10 in the NT/GH group. Thirteen of the 19 patients entered puberty during the study period. Eighteen of the 19 patients were adherent (proportion of days covered ≥ 75%), while the remaining one male patient 13 was non-adherent (proportion of days covered < 75%). GH therapy was continued in all the 19 patients, with no patient who dropped out of this study.

Table 1. Baseline characteristics of the 19 patients.

Efficacy

The height and height SDS for CA after GH treatment in each patient are shown in Figs. 2 and 3, respectively (such data in patients 2, 3, and 15 in whom the relevant data were also obtained at month 54 are denoted in Supplementary Fig. 1), and detailed yearly data in each patient are described in Supplementary Table 1. The height SDS for CA showed upward shift with GH therapy in all the 19 patients, including the non-adherent patient 13, before puberty and during an early phase of puberty. Furthermore, the height SDS for CA was increased during the study period in 18 of the 19 patients with the exception of female patient 12, and exceeded –2.0 in 17 of the 19 patients at the latest visit with the exception of female patients 9 and 12. In patient 9, the height SDS for CA, though it remained below –2.0 throughout the study period, increased by ~0.8 during the study period. In patient 12 who had entered puberty between month –12 and month –6, the height SDS for CA was initially increased but was subsequently decreased, and became –2.12 at month 42 (the latest visit) after 4.0–4.5 yr of pubertal period and ~3.5 yr of GH treatment.

Fig. 2.

Change in height after GH treatment in each patient. Female and male data are depicted in red and blue, respectively. The data before and after pubertal entry are shown with filled and unfilled circles, respectively. The solid and dashed lines in each figure indicate the mean and –2 SD height references of Japanese normal children, respectively.

Fig. 3.

Change in height SDS for chronological age (CA) after GH treatment in each patient. Female and male data are depicted in red and blue, respectively. The data before and after pubertal entry are shown with filled and unfilled circles, respectively. The solid and dashed lines in each figure indicate the ± 0 SD and –2 SD height references of Japanese normal children, respectively.

The effects of GH treatment in the patients examined at each yearly study point are summarized in Table 2. The ∆height SDS for CA continued to increase throughout the study period (Fig. 4). Height velocity obviously increased in the first year followed by gradual decrease in the second and the third years, and became larger in the fourth year than in the third year probably due to pubertal entry and resultant growth spurt in most patients. BA progressed faster than CA, and ΔBA/ΔCA remained above 1.0 throughout the study period, although it gradually decreased during the study period. Serum IGF-1 SDS increased during the first 12 mo, remained relatively stable in the second to third years, and further increased in the fourth year probably because of pubertal entry in most patients (Fig. 5).

Table 2. Effects of GH treatment.

Fig. 4.

Change in Δheight SDS for chronological age (CA).

Fig. 5.

Change in serum IGF-1 SDS.

Safety

The safety-related data are shown in Table 3 and Supplementary Table 2. Although AEs occurred in 100% of the patients, most of them were mild with no severe AEs. Seven patients had a total of 21 events such as mild respiratory or gastrointestinal symptoms which caused short term (1 ~ 5 d) GH interruption. ADRs were observed in three patients, i.e. a markedly high serum IGF-1 value of 712 ng/mL (+3.6 SD) at month 48 in patient 15 in whom the GH dosage was reduced from 0.35 to 0.23 mg/kg/wk thereafter (no IGF-1 value available after the reduction of GH dosage at month 48), and insulin resistance indicated by increased homeostasis model assessment-insulin resistance (HOMA-IR) values of 2.8 at month 24 in patient 1 and 2.7 at month 24 in patient 7. In addition, serum IGF-1 value exceeded +2.0 SD at least one study point in 42.1% (8/19) of the patients including patient 15. However, none showed clinically discernible features which could be related to the high GH/IGF-1 values such as acromegalic features and metabolic disorders (12), and blood HbA1c values and blood/plasma glucose values during OGTT remained within the normal range in all the patients. Furthermore, there was no report of newly developed or apparently worsened Madelung deformity from attending clinicians during the study period.

Table 3. Summary of safety-related data in the 19 patients.

Discussion

We performed the extension study up to 48 mo in the 19 pediatric patients investigated in the antecedent study, and evaluated efficacy and safety of GH treatment for SHOX-related short stature before and after pubertal entry. Notably, the height SDS for CA was increased during the study period in 18 of the 19 patients, and exceeded –2.0 in 17 of the 19 patients at the latest visit. Furthermore, the mean ∆height SDS for CA continued to increase during the study period. Thus, while the present data are complicated by pubertal entry in most patients, it is likely that GH treatment has promoted statural growth in patients with SHOX deficiency at least in the prepubertal to early pubertal period.

The height SDS for CA in female patient 12 was somewhat lower at the latest visit (–2.12) than at the time of enrollment (–2.04) and at the start of GH therapy (–1.77), after 4.0–4.5 yr of pubertal period and ~3.5 yr of GH treatment. This may suggest that the height SDS for CA is decreased with pubertal progression even under the GH treatment because of accelerated bone maturation, especially in female patients. Indeed, it has been suggested that gonadal estrogens exert a maturational effect on distal limb skeletal tissues that are susceptible to unbalanced premature epiphyseal fusion because of SHOX deficiency, facilitating growth plate fusion and the development of LWD in a female-dominant and pubertal tempo-influenced fashion (1). Consistent with this notion, the previous longitudinal study in seven GH-untreated female patients with SHOX deficiency has shown that height SDS is decreased from –2.3 ± 0.5 in childhood to –2.9 ± 0.8 in adulthood (P = 0.0060) (decrease in height SDS of ~ 0.6 during puberty) (13). Considering that such a relatively large growth reduction has been observed during puberty in GH-untreated patients, it is expected that GH treatment still exerts a beneficial effect on statural growth during puberty. However, since the height SDS for CA would be somewhat decreased with pubertal progression especially in female patients, it is strongly recommended starting GH therapy as early as possible, to ensure a sufficient time before pubertal entry.

It remains to be studied whether GH therapy ameliorates the adult height in patients with SHOX deficiency. Indeed, BA progressed faster than CA throughout the study period, with the mean ΔBA/ΔCA being > 1.0. However, previous studies have revealed that 57% of patients with SHOX deficiency achieved adult height above –2 SDS after GH treatment (6), and that the adult height of patients with SHOX deficiency is 6.3 cm taller in GH-treated patients than in GH-untreated patients in a non-randomized controlled trial (14). Thus, GH treatment would increase the adult height in Japanese patients with SHOX deficiency as well.

Although GH treatment was generally well-tolerated with no severe AEs, two findings would be worth pointing out. First, this study identified markedly increased IGF-1 in one patient and IGF-1 SDS of > + 2.0 in 42.1% (8/19) of the patients. The previous study using the same dosage of recombinant GH (Humatrope^®) has also revealed at least one occurrence of serum IGF-1 SDS of > + 2.0 in 46.9% (23/49) of the patients with SHOX deficiency (6). Second, insulin resistance was identified in the two patients in this study. However, clinical features potentially related to increased IGF-1 and insulin resistance, such as acromegalic features and diabetes mellites (DM), have not been documented in this study and in the previous studies reporting GH treatment in patients with SHOX deficiency (5, 6, 14). Furthermore, despite an increased risk of metabolic disorders such as insulin resistance in patients born small for gestational age (SGA), development of DM has not virtually been reported in GH-treated SGA patients, probably because of the compensatory effect of increased insulin secretion (15,16,17). Thus, the pharmacological dosage of GH would usually be well tolerated, although careful monitoring of IGF-1 and glucose metabolism would be required as well as adequate modification of GH dosage when considered necessary.

Newly developed or apparently worsened Madelung deformity was not reported in this study, although the assessment was mostly subjective without objective data such as the measurement of carpal angle. This is primarily consistent with the previous report that 2-yr GH treatment from prepubertal period did not increase the frequency of Madelung deformity in patients with SHOX deficiency (5). It is inferred, therefore, that GH therapy in the prepubertal to early pubertal period exerts no major effect on the skeletal phenotype in patients with SHOX deficiency. In this regard, some patients, especially female patients, may manifest newly recognized or apparently worsened Madelung deformity with pubertal progression. However, such a phenomenon would primarily be due to increased gonadal estrogen production (1), and the relevance of GH treatment would remain minor, if any.

Conclusion

This study indicates the efficacy and safety of GH therapy up to 48 mo in Japanese pediatric patients with SHOX-related short stature. However, the study period remains short, and the patient number is small, especially at month 48. Thus, further studies are required to clarify the long-term efficacy and safety, as well as the adult height in GH-treated patients with SHOX-related short stature.

Conflict of interests

N.T., H.H., R.I., O.Y., and T.Y. are employees and/or stockholders of the Japan Development Unit or Development Division of JCR Pharmaceuticals Co., Ltd., Ashiya, Japan. T.O. received lecture fees from JCR Pharmaceuticals Co., Ltd.; S. D. received lecture fees from JCR Pharmaceuticals Co., Ltd., Novo Nordisk A/S, and Pfizer Inc.

Funding Statement

This study was funded by JCR Pharmaceuticals Co., Ltd.