Schaaf-Yang Syndrome: Clinical Phenotype and Effects of 4 years of Growth Hormone Treatment

Abstract Introduction: Schaaf-Yang syndrome (SYS) is a rare neurodevelopmental disorder caused by truncating mutations of the MAGEL2 gene, located in the Prader-Willi syndrome (PWS) region. PWS and SYS have phenotypic overlap. Patients with SYS are often treated with growth hormone (GH), but evidence for the effectiveness of the treatment in patients with SYS is limited. Methods: This study describes 7 children with SYS. We studied their phenotype, genotype, and the effect of GH treatment on height and body mass index (BMI) during 4 years and on body composition during 1 year. Results: All patients had a normal birth weight. Most patients had hypotonia and feeding difficulties after birth (86%). Full-scale IQ ranged from <50 to 92. All patients above the age of 2 years had psycho-behavioral problems. There were no apparent correlations between the phenotype and the location of the defect in the MAGEL2 gene. Mean (95% CI) height SDS increased significantly from −1.74 (−3.55; 0.07) at start to −0.05 (−1.87; 1.77) after 4 years of GH treatment. Mean (95% CI) BMI SDS decreased significantly from 2.01 (1.02; 3.00) to 1.22 (0.18; 2.26) after 6 months and remained the same during the rest of the follow-up. Fat mass percentage SDS decreased and lean body mass did not change during 1 year of treatment in 3 patients. Conclusion: Patients presented with a phenotype of hypotonia, respiratory insufficiency, and feeding difficulties after birth, endocrine disorders, intellectual disability, and behavioral problems. Treatment with GH significantly improved height SDS and BMI over the course of 4 years.


Introduction
Schaaf-Yang syndrome (SYS) is a very rare neurodevelopmental disorder, caused by a truncating mutation in the maternally imprinted, paternally expressed, singleexon MAGEL2 gene [1][2][3][4][5].The MAGEL2 gene has a role in the regulation of endosomal protein trafficking and recycling.Because the MAGEL2 gene is silenced on the maternal allele through methylation, only mutations in the paternal MAGEL2 allele lead to SYS.It has been established that it is not the absence of the MAGEL2 gene, but the presence of the truncated protein that causes SYS [6].SYS is most commonly diagnosed by whole-exome sequencing or targeted gene sequencing of the MAGEL2 gene.
First described in 2013 [1], the syndrome is characterized by developmental delay, respiratory abnormalities, neonatal hypotonia, interphalangeal joint contractures, a high prevalence of autism spectrum disorders (ASDs), and endocrine abnormalities.A subgroup of patients also developed hyperphagia and obesity [4].Due to the location of the MAGEL2 gene in the Prader-Willi syndrome (PWS), critical region on chromosome 15, SYS was originally termed a Prader-Willilike syndrome and there is phenotypic overlap between the syndromes, especially during infancy [7].However, there are also clear differences, the most striking being the difference in prevalence of ASD and joint contractures [8].MAGEL2 mutations were also identified in patients with the previously clinically diagnosed Chitayat-Hall syndrome [9].Two particular truncating mutations are associated with the severe phenotype of arthrogryposis multiplex congenita [2], leading to perinatal death.
The cognitive functioning of patients with SYS shows a large variation between patients and the most severely affected patients are often not able to be formally tested [10,11].Most patients do appear to have intellectual disability (ID) [11].The most defining feature of the SYS behavioral phenotype is ASD.Approximately 80-90% of patients with SYS are diagnosed with ASD [4,11].Other characteristics of the behavior include: hyperactivity, hyperphagia, attention problems, and social withdrawal [11].
Many patients with SYS have endocrine abnormalities, such as growth hormone deficiency (GHD), TSH deficiency, ACTH deficiency, and hypogonadism [9,12,13].Similar to patients with PWS, patients with SYS have a relatively high fat mass and low lean body mass (LBM) [13].Also, the MAGEL2-knockout mouse exhibits a higher ratio of fat to lean mass [14].In PWS, growth hormone (GH) treatment does not only significantly improve body composition and growth [15,16] but also has benefits regarding mental and motor development [17][18][19].Patients with SYS are often treated with GH, either due to the inherent similarities with PWS or because of a formal diagnose of GHD.There is only one study published on the effects of GH treatment on growth and BMI in children with SYS [20].This study included a 6 month follow-up of 14 patients.
Here, we present the phenotype and genotype of 7 patients with SYS and the effects of GH treatment over the course of 4 years.We provide anthropometric data, body composition by dual-energy X-ray absorptiometry (DXA), resting energy expenditure, and endocrine and safety parameters.First, we hypothesized that some phenotypic traits can be linked to specific locations of the individuals' mutations in the MAGEL2 gene.Second, we hypothesized that GH treatment would improve height SDS and body composition.

Study Population
This study describes 7 children with a molecular genetic diagnosis of SYS who visited our Dutch Reference Center for PWS/PWL between November 2015 and January 2023.Two patients started GH treatment before they received their SYS diagnosis.The data from the start of their treatment had to be collected retrospectively.

GH Treatment
Patients received a GH dose between 0.5 mg/m 2 /day and 1.0 mg/m 2 /day (~0.035 mg/kg/day) once daily at bedtime.At each visit, GH dose was adjusted according to body surface area.

Anthropometric Measurements
Standing height was measured with a Harpenden Stadiometer.Weight was assessed on a calibrated scale (Servo Balance KA-20-150S; Servo Berkel Prior, Katwijk, The Netherlands).Height, weight, and body mass index (BMI) SDS were calculated with a Growth Analyser 4.0 and were adjusted for gender and age according to Dutch reference values [21,22].Small for gestational age was defined as a birth weight below −2 standard deviations [23].Short stature was defined as a height below −2 standard deviations.Overweight was defined as a BMI above 2 standard deviations.

Endocrine Status
Growth hormone deficiency (GHD) was tested through growth hormone stimulation tests in all but the two youngest patients (patients 1 and 2).The patients underwent two growth hormone stimulation tests (GHRH-arginine test and clonidine test) and GHD was defined as a peak GH level <20 mU/L (equals approximately 6.7 μg/L).Pubertas tarda was defined as a Tanner stage M1 at the age of ≥13 years for girls and a testicular volume <4 mL and/or absent virilization at the age of ≥14 years for boys.Precocious puberty was defined as start of puberty before the age of 8 years for girls and 9 years for boys and confirmed by a gonadotropin-releasing hormone stimulation test.The hypothalamic pituitary adrenal axis was assessed by metyrapone test, and central adrenal insufficiency was defined as a peak 11-deoxycortisol level <200 nmol/L.In 1 patient, a cortisol level above 500 nmol/L during severe illness was deemed as proof of adequate adrenal function.Thyroid function was evaluated by serum TSH and free T4 measurements.Body Composition Fat mass % (FM%) and LBM were measured by dual-energy X-ray absorptiometry (DXA) (Lunar Prodigy; GE Healthcare).All scans were made on the same machine, with daily quality assurance.The intra-assay coefficients of variation were 0.41-0.88%for fat tissue and 1.57-4.49%for LBM [24].FM was expressed as percentage of total body weight (FM%).LBM was calculated as fatfree mass minus bone mineral content.Standard deviation scores of FM, FM%, and LBM age were calculated according to age-and sex-matched Dutch reference values [25].Since LBM is strongly related to height, we also calculated LBM corrected for height (LBM height ).The LBM height SD scores were computed by comparing LBM of SYS children to LBM of healthy children with the same height and sex.

Cognition and Behavior
Cognitive tests were performed by the same experienced psychologist at the Reference Center for PWS/PWL.Wechsler Preschool and Primary Scale of Intelligence, Dutch version (WPSSI-NL) was used for children below age 6 years.Wechsler Intelligence Scale for Children (WISC) versions III and V were used in children between 6 and 16 years of age.Two patients were not able to undergo an IQ test because of severe delay in development.The full-scale IQ of the entire group of patients is therefore given as a range, with the IQ of the patients with severe delay set at <50.Psycho-behavioral problems were assessed through a structured interview by the same examiner (A.F.J.).Hyperphagia was defined as insatiability, food-seeking behavior, or food hoarding and an increased focus on food.

Safety Parameters
Fasting blood samples were collected after an overnight fast and measured in the biochemistry and endocrine laboratories of the Erasmus Medical Center, Rotterdam.Fasting glucose and insulin were immediately assayed.Insulin levels were assessed using the Immulite 2000 assay (Siemens Healthcare Diagnostics) until 2018.From 2019, insulin levels were assessed using Lumipulse Q2100.A correction factor, provided by the laboratory was used for insulin levels measured by the new assay.Glucose levels were determined using the Hitachi 917 (Hitachi Device Development Center), detecting glucose levels between 0 and 42 mmol/L.Until 2019, thyroid hormone levels were measured using the Ortho VITROS ECiQ Immunodiagnostic System, and from 2019 onward, Fujirebio Lumipulse G1200 was used.IGF-1 and IGF-BP3 were measured using the IDS-iSYS (Immunodiagnostic Systems), with an interassay CV of <6.0% and <5.1%, respectively.SDS values were calculated for IGF-1 and IGF-BP3 according to age-and sexmatched reference values of the Dutch population [26].

Statistical Analysis
Statistical analyses were performed using IBM SPSS statistics 25.Categorical variables are reported as frequencies and percentages.As most data did not have a Gaussian distribution, they are expressed as median (interquartile range [IQR]).Changes in height and BMI over time were analyzed using repeated measurements analysis, in order to correct for multiple testing, with years of GH treatment as the categorical-independent variable and a first-order autoregressive heterogeneous covariance matrix for the measurements within each child.Herein, effects are presented as estimated marginal mean (95% CI).Because of the low sample size for the body composition measurements (3 patients), we could not perform statistical analyses on these numbers.These results are presented as median (IQR).

Clinical Phenotype
The clinical phenotype of the 7 patients with SYS is summarized in Table 1.The median (IQR) age at examination was 8.48 (1.64; 16.14) and the age of diagnosis was 5.90 (0.13; 12.29).Patients were all born at term with a median (IQR) gestational age of 40.0 (38.6; 40.6) weeks.There were no patients born small for gestational age.Four of the 7 patients (57%) had postnatal respiratory insufficiency, and 6 out of 7 patients (86%) had feeding difficulties and hypotonia after birth.

Anthropometry
Of the 7 patients, 3 had short stature at the start of GH treatment (43%).Median height SDS (IQR) before the start of GH treatment was −1.21 (−4.23; 0.57) and median BMI SDS (IQR) was 2.00 (1.50; 2.76).At the time of examination, 29% of the patients were overweight.

Cognition and Behavior
The range of the full-scale IQ was <50 to 92.Four out of 5 patients (80%) had ID, and 5 out of 5 (100%) were enrolled in special education.Cognitive functioning and behavior of 2 patients could not be assessed because of their young age (<2 years of age).All patients above the age of 2 years had psycho-behavioral problems.The most common were as follows: symptoms of an ASD (100%); emotional outbursts (100%); and aggressive behavior (80%).Other, less common, behavioral problems included: obsessive-compulsive behavior (60%); depressive symptoms (40%); (history of) psychosis (20%); and skin picking (60%).

Endocrine Status
The most common endocrine disorders in the patients were GHD (3/5: 60%) and central precocious puberty (2/5: 40%).One patient had central adrenal insufficiency, GHD, and delayed pubertal development.Another patient had central hypothyroidism and GHD.There was 1 patient with transient central diabetes insipidus and no other endocrine disorder.
Resting Energy Expenditure Resting energy expenditure (REE) was measured in 6 patients.The median (IQR) ratio of the measured REE versus predicted REE (REE%) was 101.0 (90.2; 105.2).There was 1 patient with a decreased measured REE.

Other Findings
In our group of SYS patients, 3 out of 7 patients (43%) had arthrogryposis and one had cryptorchidism (100% of boys).None of the patients had sleep apnea during polysomnography.
Phenotype/Genotype Correlation Figure 1 is a graphical representation of the location of the mutation in the MAGEL2 gene with a few important phenotypic traits highlighted per patient.Most patients had a mutation in the second part of the MAGEL2 gene.One The data are shown as the median (IQR) or frequency (numerator/denominator). Bone mineral density of the lower spine, corrected for bone size.† Full-scale IQ is provided as a range.patient had a mutation close to the start of the gene.This patient had no ID, no short stature, and no GH deficiency.She did, however, have severe behavioral problems.

Effects of GH Treatment on Anthropometry
Table 2 shows the effect of GH treatment on height and BMI over the course of 4 years.Mean (95% CI) height SDS increased significantly from −1.74 (−3.55; 0.07) at start to −0.70 (−2.51; 1.11) after 1 year of treatment and −0.05 (−1.87; 1.77) after 4 years.The patients with GHD showed the largest increase of height SDS and had the lowest starting height SDS.Mean (95% CI) BMI SDS decreased significantly during the first 6 months of treatment from 2.01 (1.02; 3.00) to 1.22 (0.18; 2.26).During the following few years, BMI SDS remained at the same level.

Effects of GH Treatment on Body Composition and Safety Parameters
Table 3 summarizes the effect of GH on body composition and safety parameters during 1 year of treatment.Median (IQR) FM% SDS decreased from 2.82 (2.56; -) to 2.39 (1.06; -) during 1 year of GH treatment in 3 patients.LBM age increased slightly and LBM height decreased, but the differences were small.Median (IQR) total body BMD and lower spine BMAD remained unchanged over the course of 1 year.
One patient started GH treatment at the age of 2 months.Before starting the treatment she was still receiving high-flow oxygen therapy, which she could be weaned off after starting GH treatment.

Growth Charts of Two Patients
Figure 2 shows the growth charts of 2 patients with SYS.Patient 6 started with GH treatment after she was diagnosed with GHD.After starting treatment, her height SDS normalized.She was diagnosed with SYS several years later.The second growth chart presents the growth of patient 3.This patient did not have GHD and she started GH treatment after her SYS diagnosis.

Discussion
In this study, we investigated the phenotype of 7 patients with SYS and studied the effects of GH treatment on growth and body composition.We found that patients had a similar phenotype of neonatal hypotonia, respiratory insufficiency, and feeding difficulties after birth, endocrine disorders, ID, and behavioral problems.Some patients also had short stature and arthrogryposis.Not all patients presented with these symptoms, and the phenotypic expression was variable.We found a positive effect of GH treatment on height during 4 years of treatment.GH treatment had a positive effect on body composition, but the sample size numbers were too small to perform any statistical tests.Most patients (80%) had ID, with a full-scale IQ ranging from <50 to 92.A previous study stated that 44% of patients presented with severe/profound ID, 33% with moderate ID, and 22% with mild ID [11].One patient in our study had an average cognitive development, and 2 patients could not be tested because of their severe delay in development.All school-age patients were enrolled in special education (including the patient with an average IQ), and all patients above the age of 2 years presented with behavioral problems.The most prevalent behavioral problems were symptoms of ASD, hyperphagia, aggression, and emotional outbursts.Hyperphagia is often not reported as an integral part of the syndrome.However, in adults with SYS, it appears to be more prevalent [10].In our study, a relatively large percentage of patients did have hyperphagia, perhaps due to preferential testing for SYS in patients who present with obesity and a Prader-Willi-like phenotype.
Several patients had endocrine abnormalities, with GHD being the most prevalent (60%).Some patients had multiple deficiencies.These results are in line with previous studies, which show a high prevalence of GHD [13,20] and, occasionally, (partial) hypopituitarism [27,28] in patients with SYS.One patient developed transient central diabetes insipidus, which has also been previously described [12].Data expressed as estimated marginal means (95% CI). a p < 0.01; b p < 0.001; c p < 0.05, compared to baseline.SDS, standard deviation score, according to age-and sex-matched Dutch reference values [21,22,25,26].BMI, body mass index.[21,22,25,26].N/A, not applicable; TSH, thyroid-stimulating hormone; free T4, free thyroxine; IGF-1, insulin growth factor 1; IGF-BP3, insulin growth factor binding protein 3; FM, fat mass; FM%, fat mass percentage; LBM, lean body mass; BMD TB , total body BMD; BMAD LS , lower spine adjusted BMD.Reference ranges: free T4 = 9-24 pmol/L; TSH = 0.4-4.0mU/L; IGF-1 SDS = -2 to -2; insulin <200 pmol/L.We aimed to link the phenotypic expression to the location of the mutation in the MAGEL2 gene.It has been shown that not the absence of a functional MAGEL2 gene, but the presence of a truncated protein, causes SYS [29].This could mean that a larger protein might cause a more severe phenotype with a neomorphic effect of the truncated protein [6].Patient 5 had a mutation in the first part of the MAGEL2 gene, and would, therefore, have a small truncated protein.Her phenotype could be perceived as mild as she had no signs of neonatal hypotonia or respiratory insufficiency, ID, endocrine abnormalities, or short stature.However, she did have severe behavioral problems, hyperphagia, and obesity and was enrolled in special education, despite her average IQ.As she would only have a small truncated protein as a result of the particular location of the mutation in the MAGEL2 gene, perhaps the phenotype she expresses is more due to the absence of a functional MAGEL2 copy, than to the presence of a small truncated protein.This patient is the only one with this particular mutation registered worldwide.All the other patients presented in this paper had mutations in the second half of the single-exon gene.The patients with the mutation located furthest to the end of the gene did not appear to have a more severe phenotype than the others.

Phenotype of SYS and Treatment with GH
We evaluated the effects of 4 years of GH treatment on height and BMI.Height SDS improved significantly over the course of 4 years.The improvement of height SDS was largest in the group of patients with GHD.The growth chart of 1 of these patients, shown in Figure 2, portrays an impressive increase in height, resulting in a normal adult stature.The patients with GHD also had the lowest starting height SDS.Perhaps GHD is the main cause for short stature in children with SYS.
Our study was the first to use body composition measurements by DXA scan to test the effects of GH treatment in patients with SYS.We found that FM% decreased during treatment but not significantly.The effect on LBM appeared smaller and LBM height even decreased, implying that the increase of LBM is mainly due to the increase in height in this group of patients.Both total body BMD and lower spine BMD remained the same during treatment.Only one study on the effect of GH in patients with SYS has been performed previously [20].This retrospective study found a significant increase of height SDS during 6 months of treatment.BMI was used as a surrogate marker of body composition in this study and showed a significant decrease during 6 months of treatment, which is in line with our results.The effect of treatment with GH in patients with SYS appears to be less pronounced when compared to patients with PWS, especially with regard to the effect on body composition [15,16,30,31].
There were no adverse effects of GH treatment, and all safety parameters remained within normal range.Notably, no patients presented with sleep apnea, which is a common feature of SYS [4,10].All patients participating in the current study underwent polysomnography before the start of treatment and no sleep apnea was detected.One patient was still suffering from respiratory problems at the age of 2 months.After the start of GH treatment, she could be weaned off oxygen therapy.This suggests a potential benefit of starting GH treatment early in patients with SYS and respiratory problems after birth.
Our study must be considered in the light of some limitations.First, we had a small sample size.This is unavoidable in a syndrome as rare as SYS but might have been the cause of some outcome measures not reaching significance.Second, as our reference center is focused on patients with PWS and Prader-Willi-like disorders, there might have been selection bias in which patients were referred to us.Finally, for 2 patients who started GH treatment after a diagnosis of GHD (and before they received a diagnosis of SYS), we had to rely on longitudinally acquired but retrospective data to complement the prospective data.
In conclusion, our 7 patients with SYS presented with a similar phenotype of hypotonia, respiratory insufficiency, and feeding difficulties after birth, endocrine disorders, ID, and behavioral problems.No convincing correlation between phenotypic expression and location of the MAGEL2 defect could be detected.However, the patient with the mutation closest to the start of the single-exon gene and therefore, with the shortest truncated protein, appeared to have a more mild phenotype with regard to stature, cognitive development, and endocrine status.This patient did have severe behavioral problems, including hyperphagia.Treatment with GH improved height SDS over the course of 4 years.The effect of GH treatment on body composition seemed positive and FM% did decrease in the 3 patients who underwent body composition measurements.However, due to the small sample size, we could not perform a statistical analysis on these results.Further studies in patients with SYS are needed to elucidate the genotypephenotype correlation and effects of GH, especially on body composition, preferably in a larger group of patients, in a controlled setting and with a longer follow-up.

Fig. 1 .
Fig. 1.Location of the patients' mutation in the MAGEL2 gene and the most important phenotypic traits per patient.GH, growth hormone; ID, intellectual disability; IQ, intelligence quotient; ASD, autism spectrum disorder; N/A, not applicable (due to young age).

Fig. 2 .
Fig. 2. Growth charts for patients 6 (left) and 3 (right) with the start of GH treatment marked.Patient 6 was diagnosed with GHD and treated with GH accordingly.Patient 3 did not have GHD.Pink represents the height SDS of the Dutch reference population.Blue represents the target height range.

Table 2 .
Effect of GH treatment on height and BMI over the course of 4 years

Table 3 .
Effect of GH treatment on body composition and safety parameters over the course of 1 year