Key words growth hormone - Turner syndrome - real world study - long acting growth hormone - IGF-1 - pituitary
Introduction
Turner syndrome (TS), also known as congenital ovarian hypoplasia syndrome, was first
reported by Turner in 1938. TS is composed of complete or partial haplo groups of
the X chromosome. The most common clinical feature is the short stature. Above
95% TS patients suffer from varying degrees of growth disorders [1 ]
[2 ],
and it is believed that its possible etiology and pathogenesis are related to a
haploinsufficiency of SHOX gene (short stature homeobox containing gene) [3 ]. In 1996, the US Food and Drug
Administration (FDA) approved growth hormone for the treatment of Turner syndrome
(TS). To date, multiple clinical studies have confirmed that GH treatment could
effectively improve the final height in TS patients [4 ]
[5 ]
[6 ]
[7 ].
At present, GH treatment is mostly administered through daily subcutaneous
injections, and poor compliance remains in long-term applications. In recent years,
the development and applications of LAGH have become a research hotspot worldwide;
most of such studies are in phases 2 and 3 of the clinical trials [8 ]
[9 ]
[10 ]
[11 ]. The publication of Miller et al. [12 ] in 2020 compared studies published between
January 2000 and June 2019, and showed that all subjects belonged to growth hormone
deficiency (GHD). Seven RCT studies were completed, and all concluded a
non-inferiority compared to daily GH, but it is still necessary to observe the
long-term compliance, safety, and effectiveness. Additionally, there had not been
any clinical studies treating TS with LAGH prior to this study.
IGF1 is recognized as an important biomarker for monitoring the efficacy and safety.
TS patients do not lack growth hormones, and usually require more than their
physiological requirements to promote growing by producing high IGF1. Both IGF1 and
IGFBP-3 have been reported to be expressed in tumor tissue, and plays an important
role in regulating cell growth, apoptosis and tumor evolution [13 ]. Although there is no clear evidence
linking rhGH to tumors [14 ]
[15 ]
[16 ],
there is a need to monitor the safety of IGF1 in GH treatment based on the genetic
background of TS children and reports of rare tumor genesis [17 ]. International guidelines on TS treatment
recommend that IGF1SDS should be maintained between 1 and 2 during rhGH treatment.
When IGF1SDS rises higher than 3, it is recommended to reduce the GH doses to ensure
safety of the treatment [18 ]. However,
titration of IGF1 was used in a study to treat TS patients, with a mean treatment
duration of 6.7 years and a final height increase of only 3.2 cm [19 ].
Our study is the first to analyze the efficacy and safety of LAGH treatment for TS
and discuss the IGF1 levels as part of the efficacy evaluation to provide basis and
references for clinical treatments.
Patients and Methods
Patients
Our data were based on a cohort study conducted at Beijing Children’s
Hospital in 2018, which has been registered and approved by the Ethics Committee
of Beijing Children’s Hospital, Capital Medical University (No.
2018–178), and are in accordance with the Declaration of Helsinki. The
parents or guardians of the pediatric patients provided Informed consent prior
to GH treatment. The primary aim was to evaluate the efficacy and safety of
recombinant human growth hormone therapy in Chinese children with short
stature.
In our study, pre-pubertal patients diagnosed with TS were included in our
analysis. Clinical diagnoses were based on the judgment of the treating
physician. Pre-puberty was defined as stage Tanner 1 for the breast
and/or having no secondary sex characteristics.
Study design
All patients enrolled were divided into four groups according to the initial
treatment: Group 1 was treated with low-dose LAGH
(0.1 mg/kg/w); Group 2 was treated with high-dose LAGH
(0.2 mg/kg/w ); Group 3 was control group (daily GH
group, 0.38 mg/kg/w); Group 4 was untreated control
group (untreated or GH had not been administered for up to 3 months), and
followed up for 2 years. But in the second year, GH dose of Group 1
(0.18 mg/kg/w) was increased to similar with Group 2
(0.2 mg/kg/w), number of Groups 3 was decreased only 7,
9 patients in Group 4 (n=15) were beginning GH therapy and 4 patients
failed to follow-up. So, study duration was divided into two phases, year 1 with
all 4 groups and year 2 with just two, Group 2 and Group 3. The study flowchart
for the analysis of the effects of GH therapy is shown in [Fig. 1 ].
Fig. 1 The study flowchart.
LAGH (Jintrolong® ) is the irreversibly PEGylated LAGH
formulation and was from Gene Science Pharmaceuticals, Changchun, China.
Methods
Assessments at baseline included chronological age, bone age, standing height,
weight, parental height, HV, and GH dose. Predicted adult
height=(father's height+mother's
height)/2–6.5. HtSDS calculation was based on the 2005 percentile table
of height and weight of children aged 0–18 in Chinese cities [20 ].
The following items were recorded during follow-ups: GH dose, height, weight,
bone age, laboratory date included IGF1, glycated hemoglobin (HbA1c), cortisol,
free thyroxine (T4 ), thyroid-stimulating hormone (TSH), lipids,
insulin, fasting blood glucose, 2-hour postprandial blood glucose, liver, and
renal function. GH related side effects including intracranial hypertension,
slippage of the femoral head, and tumor were analyses.
Serum IGF1 was determined by chemiluminescent immunoassay using an IMMULITE 2000
immunoassay system (Siemens Medical Diagnostics, Germany). IGF1 SDS was
calculated according to normal reference values for Chinese children.
BA was assessed by the clinician using an X-ray image, according to the G-P
method [21 ].
Statistical analysis
Data are presented as mean±SD and percentages. Comparison of data between
groups was performed using t -t test or one-way analysis of variance
(ANOVA). Post hot analysis was performed with the Bonferroni’s
adjustment. Chi-square test/Fisher's exact test was used to
compare the count data. Paired t -t test was used for comparing
differences between the baseline and treatment. Multivariate correlation
analysis was conducted on the influencing factors of ΔHtSDS, HV during
the LAGH treatment, and of which the results showed that p<0.05 was
statistically significant. SPSS version 22.0 (IBM, Armonk, NY, USA) was used for
statistical analysis. Graphs were plotted using GraphPad Prism version 8.2
(SanDiego, CA, USA).
Results
Baseline characteristics
Seventy-five prepubertal TS children were enrolled. The mean age was
7.9±2.9 years, and the mean bone age was 6.8±2.8 years. Baseline
characteristics are shown in [Table 1 ].
The karyotypes of the 75 patients were 45, X0 (n=16, 21%);
mosaic karyotype (n=11, 15%); chromosome structural abnormality
(n=47, 63%), and 1 case (1%) of 45, X0/46XY. The
origin of the X chromosome: there were a total of 15 cases of 45, X0; the
parental X chromosome origin of 12 of them was tested: 8 patients (67%)
had maternal X chromosomes and 4 (33%) had paternal X chromosomes. The
four groups were well balanced in terms of demographics and clinical
characteristics (p>0.05, [Table
1 ]). However, BMI showed differences between groups, (p=0.03,
[Table 1 ]), but the post hot analysis
identified no significant differences between these groups. The baseline IGF1SDS
results of the four groups were –1.10, –1.10, –0.72,
–1.15, respectively. There was no statistical difference between groups
(p=0.448).
Table 1 Baseline Characteristics of the patients with
TS.
Group 1
Group 2
Group 3
Group 4
p
n
20
27
14
13
GH dose
0.1
0.2
0.38
0
Chronological age(year)
7.8±2.4
8.0±3.2
8.9±3.1
7.9±3.7
0.75
Karyotype, n (%)
0.70
45, X0
5 (25%)
7 (25%)
3 (20%)
1 (7.7%)
X abnormal
13 (65%)
17 (61%)
7 (53%)
1 (76.9%)
Mosaic
2 (10%)
3 (11%)
4 (27%)
2 (15.4%)
Y chromosome material
0
1 (3%)
0
0
Height (cm)
109.6±11.6
109.2±14.0
115.3±17.0
109.5±16.9
0.59
HtSDS
–3.23±0.52
–3.46±0.96
–3.27±0.89
–3.23±0.79
0.60
MPHSDS
0.39±0.57
0.51±0.93
0.29±0.69
0.85±0.91c
0.32
Weight (kg)
22.58±8.17
21.00±8.49
25.91±10.77
19.57±6.97
0.23
BMI (kg/m
2
)
18.2±3.5
17.0±2.9
18.5±2.6
15.7±1.5
0.03
IGF-1 SDS
–1.10±1.24
–1.10±1.35
–0.72±0.80
–1.15±1.17
0.44
HV (cm/year)
3.2±1.1
3.9±1.5
3.4±0.9
3.9±1.4
0.18
BA (year)
#
6.8±2.5
6.8±3.0
8.2±3.2
6.7±3.2
0.67
BA-CA (year)
#
–0.6±1.0
−1.3±1.0
–1.3±1.1
–1.2±1.1
0.19
CA: Chronological age; BA: Bone age; HV: Height velocity; BMI: Body mass
index; MPH: Mid-parental height; SDS: Standard deviation score.
# The numbers of each group with bone age: Group 1
(n=18), Group 2 (n=26), Group 3 (n=8), Group 4
(n=13).
The efficacy
Year 1
The mean HV of each Group (Group 1 to 4) was 6.7 cm/year,
8.4 cm/year, 7.7 cm/year, and
4.7 cm/year, respectively, and there was significant
difference among 4 groups, (p<0.0001). The treatment groups
increased significantly from baseline (p<0.0001) for all ([Table 2 ], [Fig 2a ]) and Group 4 was lower than
treatment group, there was no difference in the untreated group (Group 4) ,
(Table 1S) . HtSDS increased
from baseline between treatment groups, p-values for Groups 1, 2 and 3 were
0.003,<0.0001 and 0.005 respectively, and Group 4 showed no
difference, p value was 0.436 ([Table
2 ], [Fig 2b ]). Comparing
the change from baseline (ΔHtSDS) among groups as follows: there was
significant difference between groups, (p=0.0002) ([Table 2 ]). Group 2 and Group 3 had
higher ΔHtSDS than Group 4, and there was no difference between
Group 1 and 4, Group 2 and 3 had higher ΔHtSDS than Group 1, while
there was no difference between Groups 2 and 3 (p>0.99) ([Table 1S ], [Fig 2c ]), which indicated that the
effect of high-dose LAGH therapy (group 2) was similar compare to that of
daily GH, and better than low-dose LAGH.
Fig. 2
a : Comparison of height velocity (HV) among groups at
baseline and year 1; b : Comparison of height standard
deviation score (HtSDS) among groups at baseline and year 1;
c : Comparison of change in height standard deviation
score from baseline to year 1 (ΔHtSDS) among groups.
Table 2 Comparison of HtSDS, ΔHtSDS, and HV
between groups in TS at year 1.
Group 1
Group 2
Group 3
Group 4
p (ANOVA)
n
20
27
14
13
HV
6.7±1.8
8.4±1.9
7.7±2.4
4.7±1.8
<0.0001
HtSDS at baseline
–3.23±0.52
–3.46±0.96
–3.27±0.89
–3.23±0.79
–
HtSDS at year 1
–2.86±0.62
–2.91±1.22
–2.69±0.76
–3.33±0.87
–
ΔHtSDS
0.31±0.42
0.56±0.43
0.68±0.69
0.12±0.44
0.0002
HV: Height velocity; ΔHtSDS: Change in HtSDS from
baseline.
The trend of mean IGF1SDS levels between GH treatment groups at follow-up
time point is shown in [Fig. 3 ].
Compared to the baseline, the levels of IGF1SDS are improved, and rapidly
reached their peak between 6 to 9 months among treatment groups, and then
reached a plateau. IGF1 levels in great majority of the patients treated
with GH were in the normal range at follow-up time point ([Fig. 3 ]). There were 1 patient
(5%) in Group 1, 6 patients (21%) in group 2, 5 patients
(33%) in Group 3 with IGF1 levels>2 SD, and only 3 patients
(7.1%) in Group 2. One patient (7.7%) in Group 3 had IGF1
levels>3 SD.
Fig. 3 The trend of mean insulin-like growth factor 1 standard
deviation score (IGF1SDS) at visit time between GH treatment groups
(Groups 1 to 3).
The BA-CA between treatment groups (Group 1 to 3) at year 1 were
–0.6±1.0 (n=17), –1.3±1.0
(n=15), –1.3±1.1 (n=8). Compared with the
baseline –0.6±1.0 (n=17), –1.2±1.0
(n=26), –1.3±1.1 (n=8),
–1.2±1.1 (n=13), there were no statistical
differences in BA-CA between groups 1, 2, and 3. The p-values were 0.63,
0.11, and 0.13, respectively, and BA-CA did not accelerate.
Complementary additional analyses results
Multivariate correlation analysis in LAGH treatment at first year indicated
that ΔHtSDS was positively correlated with baseline variables
including baseline dose (r=0.33, p=0.02) and negatively
correlated with baseline CA (r=–0.32, p=0.03), and
BA (r=–0.54, p=0.0002). The HV was positively
correlated with baseline HtSDS (r=0.33, p=0.03), IGF1SDS
(r=0.38, p=0.01), and negatively correlated with baseline CA
(r=–0.67, p<0.0001), BA (r=–0.71,
p<0.0001), and BMI (r=–0.36, p=0.02) (Table 2S ).
Year 2
Year 2 with just two groups, Group 2 (n=25) and Group 3
(n=7), GH dose of Group 3 (0.45 mg/kg/w) was
slightly increased than year 1. The baseline HV of Group 2 was
3.3±1.5, significant increase to 6.3±2.1 at year 2 (
p=0.0001), and Group 3 was –3.46±0.96, increasing to
6.6±1.2 ( p<0.0001), The baseline HtSDS of Group 2 was
–3.27±0.89, increasing to –2.84±1.18 (
p=0.03), and Group 3 was –3.4±0.9, increasing to
–2.68±1.17 (p=0.048), ΔHtSDS at year 2
between group 2 and 3 were similar (0.88±0.74 vs. 0.89±0.76,
p=0.97). The effect of high-dose LAGH therapy was similar compared
to that of daily rhGH .
In the second year, the mean IGF1 level of 2 groups were 0.9±0.2,
1.9±1.0, p=0.1. Four patients (16%) in Group 2 and 1
patient (14.2%) in Group 3 had IGF1 levels>3 SD; the
percentage of the 2 groups had no difference.
The BA-CA at year 2 between Group 2 (–1.3±1.2, n=11)
and Group 3 (–1±0.3, n=4) was comparable
(p=0.78). Bone age delay in relation to the chronological age in
both two groups.
Safety
Over the 2 yearsʼ treatment, no serious adverse events occurred, only 4 mild
to moderate drug-related adverse events were reported. One patient from
low-dose LAGH group had elevated TSH level, but free T4 was
within the normal range, and TSH returned to normal at follow-up time after
thyroxine supplementation. One patient from high-dose LAGH group had
injection site nodules, and nodules disappeared at follow-up time when the
patient changed the injection site. The other two events occurred in daily
rhGH group, two patients had hyperinsulinemia, and one of them was
considered to have impaired glucose tolerance, which required take
metformin, no type 2 diabetes mellitus occurred, there was no injection site
acting, including erythema, lipoatrophy and notable post-injection pain
occurred. No serious adverse events including intracranial hypertension,
slipped capital femoral epiphysis, scoliosis, or tumor, etc. side effects
developed in both groups. The patient received GH treatment had normal
HbA1c, liver and kidney functions, and there were no adverse events led to
the discontinuation of GH therapy.
Discussion
Our research is the first study of LAGH therapy in children with TS. Patients
received injections of PEG rhGH weekly. Our first-year study demonstrated that
high-dose LAGH can significantly improve HtSDS and HV of TS compared to low-dose
LAGH, multiple variable analyses showed high dose positively correlated with better
outcome. For the second year, the mean dose in the low-dose LAGH group was increased
to 0.18 mg/kg/w to maintain HV more than
6 cm/year. The effect of high-dose LAGH was similar to that of the
daily rhGH at a dose of 0.38 to 0.45 mg/kg/w through a
2-year period. Both two groups showed higher HV and ΔHtSDS comparing with
low-dose group. Literature indicated that dose in first year is a major factor
contributing to total response for TS [22 ], so
our study emphasized that the highest safe dose in the first year is important for a
better outcome, especially in girls with a poor adult height prognosis [23 ].
During GH treatment, IGF1 was measured for both safety and efficacy. In our study,
IGF1 levels were monitored during the full treatment period. We discovered a rapid
IGF1 increase, which reached its peak between 6 to 9 months among treatment groups,
then maintained stable. Most of the IGF1 values were within the normal range, only 3
patients (7.1%) in Group 2 and 1 patient (7.7%) in Group 3 had got
IGF1 levels>3 SD at year 1. In the second year, 4 patients (16%) in
Group 2 and 1 patient (14.2%) in Group 3 had IGF1 levels>3 SD. The
IGF-1 level between the 2 groups were comparable, the prescribing physician did not
decrease the GH dose, and IGF-1 levels after transient were elevated, decreased, or
maintain stable at follow-up visits. Our study describes the common dosing patterns
in clinical practice. Physician chosen to fix the dosage based on the weight and
adjusted dosages according to HV. Compared with the IGF1 titration method, this
approach is effective and relatively safe in short term study. Long-term follow-up
is needed to confirm this conclusion, and if an IGF-I value is continued above
+3 SDS, GH dose should be decreased, but there is a problem concerned about
to choose an appropriate detection method and IGF1 reference.
We also observed the safety of GH therapy. The common side effects are injection site
acting, headache, and muscle or joint pain, serious side effects including benign
intracranial hypertension, type 2 diabetes mellitus (T2DM), and slipped capital
femoral epiphysis (SCFE) and tumors. LAGH related adverse events include edema
injection-site lipoatrophy, etc. No serious side effects occurred in our study. Only
4 mild to moderate drug-related adverse events were reported. One patient from
low-dose LAGH group had elevating TSH level, and normal free T4 level,
and TSH returned to normal at follow up time after thyroxine supplementation. TS is
susceptible to immune diseases, including thyroid disease. So, whether or not
beginning growth hormone therapy, it is important to measure (free) T4 and TSH
levels. No injection-site lipoatrophy was observed, but 1 patient from high-dose
LAGH group had injection site nodules, when the patient changed the injection site
the nodules disappeared. In the long-acting PEGylated rhGH phase III and phase IV
randomized controlled trails, there is no injection-site lipoatrophy reported [24 ]
[25 ],
but some clinical study showed that at 13 weeks after GH treatment, injection-site
lipoatrophy occurred, when the injection-site was changed and to avoid repeated
injections on the same site, injection-site lipoatrophy recovered after 3–6
months. The other two events occurred in daily rhGH group, two patients had
hyperinsulinemia, no type 2 diabetes mellitus occurred. Previous study indicated
that incidence of diabetes mellitus and impaired glucose tolerance in children with
GH treatment are higher than not treated [26 ].
But it is still controversial, some study suggested that GH treatment reduced
abdominal adiposity and significantly improved glucose tolerance. In our study,
HbA1c, liver, and kidney function, total cholesterol was within normal range among
all treatment groups. There is no acceleration in BA.
The previous studies have shown that the efficacy of Turner syndrome is related to
the dose of GH, age of treatment initiation, and duration of treatment. Younger age,
longer treatment duration and higher initial GH dose, may result in relative better
outcomes [27 ]. In our study, multivariate
correlation analysis of LAGH treatment came to the similar conclusion, which was
consistent with the relevant studies of daily rhGH. We also found that the
therapeutic response in year 1 was positive correlated with baseline IGF1, HtSDS,
and negative correlated with baseline BMI. Previous studies [28 ] also found that baseline weight is GH
treatment response predictor (the lower baseline weight associated with the increase
of height SD score), as well as the risk of metabolism in TS patients are closely
associated with weight gain, we then emphasized the importance of weight
control.
Compared with previous studies, our study found several correlation factors affecting
the efficacy of LAGH were as follows: in year 1, patients with younger age, younger
bone age and higher IGF1SDS at baseline received a better effect, which was
confirmed in the study of GHD and daily GH treatment of TS.
In actual clinical settings, clinicians generally adopt treatment plans based on the
weight-based treatment regiments. Compared to IGF1 titration, it showed benefit of
height gain. During the two-year treatment, IGF1 of most patients were within the
normal range. Even if it occasionally exceeded 3SD, it could be reduced to normal
level without reducing dose.
Strengths and Limitations
The strength of the study is the fairly large size of the study cohort and
first-time evaluation efficacy of LAGH in TS. The limitations are the relatively
short observation period and not obtaining the ultimate height of patients. We
would further investigate the long-term efficacy and safety for LAGH treatment
in the future.
Conclusion
Our research first assessed the efficacy and safety of LAGH treatment in Turner
syndrome. Our study demonstrated that the effect of high-dose LAGH
(0.2 mg/kg/w) was similar compared with daily rhGH (0.38 to
0.45 mg/kg/w) in TS treatment, concerning about practical
questions, such as dose adjustment of GH, and IGF1 monitoring. Compared with IGF1
titration method, a fixed GH dose according to weight and adjust GH dose according
to therapeutic response could effectively improve height gain. Sufficient GH dose,
early diagnosis, and early treatment, high HtSDS and IGF1 level, lower BMI improved
the first year outcomes. GH therapy does not accelerate epiphyseal healing. However,
with continuous high dosage, monitoring IGF1 is important, especially in TS
treatment, and the benefit of height gain should be weighed against the GH-related
side effects, cost, and tolerance. There were no serious adverse effects. For TS,
short-term LAGH treatment was effective and safe.
Notice
This article was changed according to the following Erratum
on January 13th 2023.
Erratum
In the article mentioned above the shared first authorship was
incorrectly declared. Correct is: Xinying Gao and Jiajia Chen
contributed equally to this work.
