Language delay and developmental catch-up would be a clinical feature of pseudohypoparathyroidism type 1A during childhood

Abstract

Pseudohypoparathyroidism type 1A (PHP1A) is characterized by resistance to multiple hormones, the Albright Hereditary Osteodystrophy phenotype, obesity, and developmental delay. Developmental delay usually appears prior to hypocalcemia due to parathyroid hormone resistance and could be a clinically important feature for early diagnosis of PHP1A. To date, however, the details have not been documented. With regard to developmental delays, we conducted a multicenter retrospective study of 22 PHP1A patients from 18 families who were diagnosed clinically or genetically from 2005 to 2015. For quantitative analysis of their development, we calculated the ratios of the milestone ages of the patients to those in normal reference data. The ratio of the ages with respect to speech development, i.e., speaking a first meaningful word (median: 1.67), was significantly higher than that for gross motor development, walking unassisted (median: 1.34). The ratio of age at stringing a two-word sentence (median: 1.32) was significantly lower than that of saying a first word (median: 1.84). Ten out of 11 (91%) patients exhibited two or three of the following clinical phenotypes: developmental delay, obesity, and hyperthyrotropinemia. These results suggest two possible clinical features of developmental delays in PHP1A patients: developmental delay is more obvious in speech acquisition than in gross motor skills, and speech delays could be attenuated during later childhood. Further, the presence of multiple of three clinical symptoms could be an important indicator to differentiate the diagnosis of PHP1A during early childhood.

PSEUDOHYPOPARATHYROIDISM TYPE 1A (PHP1A) is characterized by resistance to multiple hormones, such as parathyroid hormone (PTH), thyroid-stimulating hormone (TSH), growth hormone-releasing hormone (GHRH), and gonadotropins. PHP1A is furthermore characterized by developmental delay, obesity, and typical skeletal features, referred to as Albright Hereditary Osteodystrophy (AHO), which include short stature, round face, subcutaneous calcifications, and brachydactyly [1]. Although PHP1A is a genetic disease caused by a variant of GNAS encoding a subunit protein, Gsα, which mediates PTH receptor signaling, hypocalcemia is not apparent from infancy to early childhood [2, 3]. Because of subsequent complications, such as calcification of the basal ganglia and other areas of the brain and seizures, appropriate treatment for hypocalcemia should be initiated without delay.

However, early diagnosis of PHP1A prior to the development of hypocalcemia is challenging. During the first 5 years of life, the clinical features caused by the skeletal abnormality of AHO are not specific, and is often difficult to identify by routine physical examinations. Short stature is caused by early puberty and advanced bone maturation with a blunted growth spurt [4]; thus, during childhood, short stature is not obvious. Obesity alone is not a sufficient reason to suspect the disease, because of its low specificity. Developmental delay is another possible clinical feature that would lead to a clinical diagnosis of PHP1A.

The pathological mechanisms of developmental delay in PHP1A have not been elucidated. Previously, GNAS haploinsufficiency was presumed to cause developmental delay; however, recent reports have revealed that developmental delay is more often observed in PHP1A patients than in pseudopseudohypoparathyroidism (PPHP) patients, suggesting that developmental delay in PHP1A is caused by GNAS imprinting rather than haploinsufficiency [5]. Similarly, recent mouse models have suggested that Gnas was imprinted in brain tissues, and that unbalanced expression of paternal and maternal alleles affected the phenotype of the conditional knock-out mice [6].

In spite of its pathological and clinical importance, details of developmental delay in PHP1A patients during early childhood have not been clarified. Herein, we have conducted a multicenter retrospective analysis of PHP1A patients, and our data suggest that, during early childhood, developmental delay is most obvious in language development, and patients with this delay could catch up later in childhood.

Materials and Methods

Subjects

Our study included patients who had been diagnosed with PHP1A in childhood clinically or genetically, from 2005 to 2015 and who visited our hospitals regularly. Patients were recruited from nine university hospitals and three children’s medical centers, located throughout Japan, from Hokkaido to Kyusyu district. The diagnosis of PHP1A was based on the following criteria: features of AHO, hypocalcemia (Ca: <2.1 mmol/L), and resistance to PTH (intact PTH: >3.2 pmol/L). Clinical features of AHO include short stature, round face, subcutaneous calcifications, and brachydactyly. We excluded the subjects who had vitamin D deficiency (25-hydroxy vitamin D [25OH-VitD]: <6.0 nmol/L), renal dysfunction (Cre: >+2 standard deviation [SD] for age-matched controls) or other clinical problems that could affect neurological development. Subjects or their parents provided written informed consent. The study protocol was approved by the ethics committee of the Tokyo Medical and Dental University.

Twenty-two PHP1A patients from 18 families were eligible for the present study. Their median age at the time of diagnosis was 4 years and 4 months, ranging from 8 months to 16 years. Eleven patients (50%) had developed hypocalcemia at diagnosis (Table 1), and the diagnoses in 17 patients (77%) were confirmed by genetic testing. Five out of 15 school-age patients (33%) were enrolled in special-education classes for handicapped children (Table 2).

Table 1 Clinical profiles of 22 diagnosed PHP1A cases in the present study

Case	Age at diagnosis, Gender	Gene mutation (Exon)	Height SDS	BMI SDS	AHO signs	Intact PTH level at diagnosis (pmol/L) (1.1–6.9)	Calcium level at diagnosis (mmol/L) (2.2–2.5)	Hypothyroidism
								Age at diagnosis	TSH level (mIU/L) (0.3–4.1)	fT4 level (pmol/L) (12.4–20.6)
1	0y8m, M	n. d.	–1.7	1.1	R, B, C	9.3	2.4	0y8m	6.7	11.6
2	0y10m, M	p.Val288Aspfs (Ex11)	–0.5	2.7	R, B, C	4.9	2.6	0y10m	7.7	9.5
3	1y5m, F	p. Arg165His (Ex6)	–4.4	2.2	R, B	46.7	2.5	NMS	63.5	19.6
4	1y9m, F	p.His362Tyr (Ex13)	1.1	9.2	R	8.7	2.5	1y9m	8	12
5 (a)	2y1m, F	p.Asp189Metfs (Ex7)	–1.8	4.4	R, B, C	13.8	2.7	4m	9.1	13.3
6	2y4m, M	p.Gln31X (Ex1)	–1.1	3.4	R, B, C	84.3	2	2y4m	24.9	9.4
7 (b)	2y9m, M	n. d.	–2.5	4.4	R, B, C	46.1	1.4	NMS	12.2	10.3
8 (a)	3y3m, M	p.Asp189Metfs (Ex7)	–1.7	4.5	R, B, C	27.6	2.2	5m	5.2	12.5
9 (a)	3y10m, F	p.Asp189Metfs (Ex7)	–1.1	2.4	C	34.3	2.1	1m	13.9	7.3
10 (c)	4y1m, F	p.Pro115Alafs (Ex5)	–0.8	3.4	R, B	28.9	1.7	Normal	—	—
11 (b)	4y3m, F	n. d.	–1.3	3.3	R, B, C	21.1	2.3	0m	95.4	5.4
12	4y4m, F	p.Gln35X (Ex1)	–0.8	3.4	R, B, C	71.2	2.3	4y4m	5.6	11.2
13	4y4m, F	p. Arg231His (Ex9)	1.5	4.6	R	12.3	2.1	2y9m	7.6	17
14	4y7m, F	p.Gln35X (Ex1)	–2	2.1	R, B	42.4	1.8	Normal	—	—
15	4y7m, F	p.Met1Ile (Ex1)	–2.4	0.7	B	14.2	2.6	Normal	—	—
16 (c)	5y1m, M	p.Pro115Alafs (Ex5)	–0.4	3.1	R, B	41.7	1.7	Normal	—	—
17	5y9m, M	Nonsense mutation	0.6	3.4	C	59.6	2.2	NMS	30	14.4
18	6y6m, F	2Mb deletion	1	2.3	R, B	141	1.4	2y3m	12.5	11.6
19	7y7m, M	n. d.	–1.8	1.4	R, B	107.2	1.9	7y7m	11.2	12.1
20	9y2m, F	n. d.	0.3	1.2	R, B, C	59.4	1.5	Normal	—	—
21	14y6m, M	p. Arg374Pro (Ex13)	–2.6	2.1	R, B	21.7	1.9	Normal	—	—
22	16y0m, F	p.Ala366Thr (Ex13)	–3.25	1.35	R, B	7.7	2.4	Normal	—	—

n. d., not detected; R, round head; B, brachydactyly; C, subcutaneous calcifications; NMS, newborn mass screening.

a~c: the subjects were from the same family.

Table 2 Developmental details of 22 PHP1A patients

Case	Age to reach developmental milestones (Ratiod)			The results of ESID		School class
	Walking alone	Saying a word	Stringing two words sentence	Age	DQ
1	1y9m (1.57)	1y6m (1.50)	2y6m (1.32)	n. d.	n. d.	special
2	1y10m (1.64)	2y0m (2.00)	2y6m (1.32)	n. d.	n. d.	mainstream
3	1y3m (1.12)	1y3m (1.25)	2y2m (1.15)	n. d.	n. d.	pre
4	1y6m (1.34)	1y3m (1.25)	2y4m (1.23)	n. d.	n. d.	pre
5 (a)	1y6m (1.34)	3y0m (3.00)	n. d.	3y5m	81	pre
6	1y11m (1.72)	2y2m (2.17)	2y10m (1.50)	n. d.	n. d.	special
7 (b)	2y6m (2.24)	2y0m (2.00)	2y6m (1.32)	n. d.	n. d.	special
8 (a)	n. d.	2y11m (2.92)	n. d.	1y3m	73	pre
9 (a)	1y5m (1.27)	1y8m (1.67)	2y7m (1.37)	2y0m	96	mainstream
10 (c)	n. d.	n. d.	n. d.	4y1m	82	pre
11 (b)	n. d.	2y3m (2.25)	2y9m (1.45)	n. d.	n. d.	mainstream
12	1y6m (1.34)	2y0m (2.00)	2y6m (1.32)	n. d.	n. d.	mainstream
13	n. d.	n. d.	3y8m (1.94)	n. d.	n. d.	pre
14	1y6m (1.34)	n. d.	n. d.	n. d.	n. d.	pre
15	1y4m (1.19)	1y6m (1.50)	2y0m (1.06)	n. d.	n. d.	mainstream
16 (c)	1y6m (1.34)	2y0m (2.00)	4y0m (2.11)	5y1m	70	special
17	1y0m (0.90)	n. d.	n. d.	n. d.	n. d.	mainstream
18	1y4m (1.19)	1y5m (1.42)	2y2m (1.15)	n. d.	n. d.	mainstream
19	1y0m (0.90)	1y0m (1.00)	n. d.	n. d.	n. d.	mainstream
20	1y8m (1.49)	n. d.	n. d.	n. d.	n. d.	mainstream
21	1y6m (1.34)	n. d.	n. d.	n. d.	n. d.	special
22	1y2m (1.04)	n. d.	1y8m (0.88)	n. d.	n. d.	mainstream

n. d., not detected; ESID, Enjoji Scale of Infant Analytical Development.

a~c: the subjects were from the same family.

d: Ratios of the age when the subjects reached the milestones to the ages of normal reference data. Normally, walking unassisted is achieved at 13.4 months, the first meaningful word is spoken at 12.0 months, and the first two-word sentence is spoken at 22.7 months.

Developmental tests

The Enjoji Scale of Infant Analytical Development (ESID): In order to clarify the clinical details of the development in PHP1A patients, we analyzed the ESID test results, which allowed for the evaluation of the development of infants from the age of 0 month to 4 years and 7 months. Clinically, ESID is one of the most frequently employed brief tests for the assessment of child development, and is presumed to be the most reliable developmental test in Japan. The test consists of six categorized subscales (gross motor skills, fine motor skills, basic habits, social relationships, expressive language, and receptive language). Each subscale has 26 components, providing developmental quotient (DQ) scores [7].

Estimating developmental delay by exploiting normal reference data: We retrospectively analyzed the development of the subjects based on their clinical records. In order to clarify the details, we calculated the ratios of the subjects’ ages at which developmental milestones were achieved compared to those considered to be normal. For normal reference data, we employed the 50th percentile milestone ages according to the Revised Japanese Version of Denver Developmental Screening Test (JDDST-R). Normally, walking unassisted is achieved at 13.4 months, the first meaningful word is spoken at 12.0 months, and the first two-word sentence is spoken at 22.7 months.

Statistical analysis

For statistical analyses, we employed the free statistical software: EZR Ver 1.36 [8]. The significance of differences among each subscale in the Enjoji Scale of Infant Analytical Development (ESID) was evaluated by using the Kruskal-Wallis Test. We used the Wilcoxon Signed-Rank Test for other analyses. The significance level of all p-values was <0.05.

Results

Hypothyroidism and obesity, not short stature, were common clinical features of PHP1A during childhood

Clinical profiles of the subjects are shown in Table 1. Among 22 subjects, the number of patients whose body mass index standard deviation scores (BMI SDSs) at diagnosis were more than +2 SD was 17 (77%). Hypothyroidism (TSH >5 mIU/L) was observed in 15 patients (68%), and all patients with a level of TSH >10 mIU/L, were administered levothyroxine immediately after the test. Thirteen patients (59%) exhibited both hypothyroidism and obesity, and only three patients did not exhibit either phenotype. On the other hand, short stature, less than –2 SD in height at diagnosis, was observed in only five patients (23%). Those data are consistent with those of previous reports, i.e., hypothyroidism and obesity, not short stature, are common clinical features of PHP1A during childhood [9].

PHP1A patients exhibited lower-than-average developmental quotient scores, especially in expressive language

Among the 22 patients, five patients (23%: Cases 5, 8–10, 16) underwent ESID; the median age was 3 years and 5 months, ranging from 1 year and 3 months to 5 years and 1 month. Developmental delays with DQ scores of less than 80 were observed in two subjects (40%). The average of all scores was, appropriately, 80, indicating slight delays. Although statistical significance was not detected, the speech development score was lower than those of the other subscales (Fig. 1).

Fig. 1

PHP1A patients tended to exhibit lower scores of developmental quotient (DQ) in all subscales of the Enjoji Scale of Infant Analytical Development (ESID), especially in expressive language. No statistically significant differences were found among DQ scores of the subscales (p = 0.072, Kruskal-Wallis Test).

Developmental delay in speech rather than in gross motor skills could be considered a clinical feature of PHP1A

Given the data, we hypothesized that, among the six subscales of development, according to the ESID, speech development would be the most severely affected, in PHP1A patients. In order to clarify the details, we compared speech development to gross motor development. In 13 patients (Cases 1–7, 9, 12, 15, 16, 18, 19), both clinical data were available, and we calculated the ratios of the subjects’ ages at which the first meaningful word was spoken and the first time of walking unassisted to those considered normal. The ratio was significantly higher for the ages at which the first meaningful word was spoken (median: 1.67, range: 1.00–3.00) than for those of first walking unassisted (median: 1.34, range: 0.90–2.24) (Fig. 2A).

Fig. 2

A: Speech development was delayed compared to gross motor development in PHP1A patients. The ratio of age when the first meaningful word was spoken to that in normal reference data was higher than the ratio of the age when walking unassisted began, compared to that of the normal reference data, *p = 0.011 (Wilcoxon Signed-Rank Test). B: Developmental delay in language was attenuated at later ages. The ratio of the age at which the first two-word sentence was spoken to the age in the normal reference data was significantly lower than that of the age when the first meaningful word was spoken compared to the age in the normal reference data, **p = 0.005 (Wilcoxon Signed-Rank Test).

Further, we asked whether the clinical phenotypes of AHO (other than developmental delay), such as obesity (BMI SDSs > +2 SD) and the skeletal phenotype, brachydactyly, affected the walking milestone. We did not identify any significant difference in the age at reaching the walking milestone between the two groups with or without obesity (p = 0.43), and with or without brachydactyly (p = 0.25). The results suggested that the walking milestone basically reflects the gross motor development in our subjects.

Taken together, we suggest that in PHP1A patients, speech development is delayed with respect to gross motor development.

PHP1A patients revealed a catch-up in language development

In order to clarify the details of verbal development in PHP1A patients, we longitudinally tracked the developmental course of the subjects. We compared the two points of verbal development ages, at which the first meaningful word was spoken and those when a two-word sentence was strung together. The data for the two points were available in 12 subjects (Cases 1–4, 6, 7, 9, 12, 15, 16, 18, 22), and we calculated the ratios of the ages at which the two milestones were achieved to those in normal reference data. The ratio of the age of stringing a two-word sentence (median: 1.32, range: 1.06–2.11) was significantly lower than that at which a meaningful word was first spoken (median: 1.84, range: 1.25–2.25) (Fig. 2B), suggesting that, in PHP1A patients, developmental delay in language could be mitigated with age.

Hypothyroidism did not affect the development of PHP1A patients

We asked whether hypothyroidism affected development in PHP1A patients. Normal thyroid metabolism is essential for human development. Thus, we compared the patients with and without a history of hyperthyrotropinemia (TSH >5 mIU/L). All patients were administered levothyroxine immediately after the diagnosis of hyperthyrotropinemia. We found no statistically significant difference between the two groups in the age of first walking unassisted (p = 0.51), first saying a meaningful word (p = 0.95), or stringing two words together (p = 0.48). Those data suggest that properly managed hypothyroidism will not affect development in PHP1A patients.

Discussion

This is the first report to analyze the precise phenotypes of developmental delay during early childhood in PHP1A patients. Regarding the development of PHP1A patients, we would like to highlight two points: firstly, in PHP1A patients, developmental delay could be more obvious in language than in gross motor skills; secondly, the verbal developmental delay could be attenuated during later childhood. Further, our study suggests, in a patient who exhibits two out of three clinical features, i.e., developmental delay, hypothyroidism, and obesity, there is need for careful follow-up for early identification of PHP1A.

Several studies in the literature have reported that variable degree of developmental delay/mental retardation is a feature of PHP1A [10]; however, there is an apparent discrepancy in frequency between the adult (27%) and the pediatric (64%) population [11]. Although, the cause of this discrepancy could be attributable to differences in criteria or methods of analysis, we presume that the tendency of the patients to catch up in development could be another possible explanation.

The tendency for some patients to catch up in their development is intriguing, and this could be explained by a temporal dependency of GNAS imprinting in the brain. The imprinting of GNAS is presumably spatially dependent and occurs in limited tissues; however, the mechanisms behind this process are unknown. The proximal tubule in the kidney is one targeted organ, as is the brain [6, 12]. In addition to a spatial dependency, some experimental models suggest that genomic imprinting could also be temporally dependent [13]. Despite its genetic cause, hypocalcemia of PHP1A is apparent, not during infancy but, at a later stage, early to middle childhood [2, 14]. This speculation, while not excluding other possibilities, such as the alternation of compensative mechanisms for GNAS and the delayed start of speech, which is a variation of normal language development [15], is based on clinical observations.

Because of the limited number of patients who underwent IQ/DQ tests including ESID, our cohort may not have had enough statistical power to reveal all clinical features of development in PHP1A patients. Alternatively, we employed the ratio of the ages, when developmental milestones were reached in the subjects, compared with the normal reference data subjects. However, our analysis was based on just three key milestones: walking unassisted, saying a first meaningful word, and stringing a sentence of two or more words, and did not elucidate the details of the development in PHP1A patients. Further, the clinical information that could affect neurological development, such as family history and environmental factors were not available, resulting in another limitation to our study.

Based on the results of our study, we propose that, in patients who exhibit two out of three clinical symptoms, i.e., hypothyroidism, obesity, and developmental delay, serum levels of PTH should be frequently monitored. Our present study is a retrospective analysis, and evaluating the specificity of each factor was beyond its scope; however, we presume that a combination of two or three of the above-mentioned clinical features will increase the specificity for the diagnosis of PHP1A. In our study, 67% of the patients presented with developmental delays. Obesity and hypothyroidism were observed in 77% and 68% of the subjects, respectively. In 11 of our subjects, the three clinical records regarding development, BMI SDS, and thyroid function, were available, and 10 out of 11 (91%: Cases 1–4, 6, 7, 9, 12, 15, 16, 18) patients exhibited at least two out of the three clinical symptoms listed above. It is known that, in PHP1A patients, the serum level of PTH does not increase from infancy through early childhood [16]. We, therefore, recommend that, in patients who exhibit two out of the three clinical features, differentiating the diagnosis of PHP1A with careful follow-up should be considered.

In summary, we retrospectively conducted a multicenter study of PHP1A focusing on developmental profiles. Our study revealed that developmental delay was obvious in language acquisition, and this delay tended to be temporal. Furthermore, based on clinical profiles of the subjects, we propose that PTH levels should be evaluated when two out of three of the clinical symptoms are present: hypothyroidism, obesity, and developmental delay. Our data should be interpreted with caution, because the present study was based on limited data, employed only three key milestones for developmental analysis, and had small sized subject population. Regardless, our study would provide valuable insight into the clinical management and identification of the pathophysiology of PHP1A. For elucidating the details of the development in PHP1A patients, accumulation of the cases and a large-scale cohort study is necessary.

Acknowledgements

We are indebted to TTT (Trace The Turner) study group, in which ten authors are involved, for fruitful discussion with the members and recruitment of the patients.

Disclosure

None of the authors have any potential conflicts of interest associated with this research.

References

• 1   Long  DN,  McGuire  S,  Levine  MA,  Weinstein  LS,  Germain-Lee  EL (2007) Body mass index differences in pseudohypoparathyroidism type 1a versus pseudopseudohypoparathyroidism may implicate paternal imprinting of Gαs in the development of human obesity. J Clin Endocrinol Metab 92: 1073–1079.
• 2   De Sanctis  L,  Giachero  F,  Mantovani  G,  Weber  G,  Salerno  M, et al. (2016) Genetic and epigenetic alterations in the GNAS locus and clinical consequences in Pseudohypoparathyroidism: Italian common healthcare pathways adoption. Ital J Pediatr 42: 101.
• 3   Usardi  A,  Mamoune  A,  Nattes  E,  Carel  JC,  Rothenbuhler  A, et al. (2017) Progressive development of PTH resistance in patients with inactivating mutations on the maternal allele of GNAS. J Clin Endocrinol Metab 102: 1844–1850.
• 4   de Wijin  EM,  Steendijk  R (1982) Growth and maturation in pseudohypoparathyroidism: a longitudinal study in 5 patients. Acta Endocrinol (Copenh) 101: 223–226.
• 5   Mouallem  M,  Shaharabany  M,  Weintrob  N,  Shalitin  S,  Nagelberg  N, et al. (2008) Cognitive impairment is prevalent in pseudohypoparathyroidism type Ia, but not in pseudopseudohypoparathyroidism: possible cerebral imprinting of Gsα. Clin Endocrinol (Oxf) 68: 233–239.
• 6   Chen  M,  Shrestha  YB,  Podyma  B,  Cui  Z,  Naglieri  B, et al. (2017) Gsα deficiency in the dorsomedial hypothalamus underlies obesity associated with Gsα mutations. J Clin Invest 127: 500–510.
• 7   Mori  S,  Nakamoto  H,  Mizuochi  H,  Ikudome  S,  Gabbard  C (2013) Influence of affordances in the home environment on motor development of young children in Japan. Child Dev Res 2013: 1–5.
• 8  Kanda Y (2013) Investigation of the freely available easy-to-use software EZR for medical statistics. Bone Marrow Transplant 48: 452–458.
• 9   Kayemba  KS,  Tripon  C,  Heron  A,  Hindmarsh  P (2016) Pseudohypoparathyroidism type 1A-subclinical hypothyroidism and rapid weight gain as early clinical signs: a clinical review of 10 cases. J Clin Res Pediatr Endocrinol 8: 432–438.
• 10   Perez  KM,  Lee  EB,  Kahanda  S,  Duis  J,  Reyes  M, et al. (2018) Cognitive and behavioural phenotype of children with pseudohypoparathyroidism type 1A. Am J Med Genet A 176: 283–289.
• 11   Farfel  Z,  Friedman  E (1986) Mental deficiency in pseudohypoparathyroidism type 1 is associated with Ns-protein deficiency. Ann Intern Med 105: 197–199.
• 12   Zheng  H,  Radeva  G,  McCann  JA,  Hendy  GN,  Goodyer  CG (2001) Gαs transcripts are biallelically expressed in the human kidney cortex: implications for pseudohypoparathyroidism type 1b. J Clin Endocrinol Metab 86: 4627–4629.
• 13   Xie  T,  Plagge  A,  Gavrilova  O,  Pack  S,  Jou  W, et al. (2006) The alternative stimulatory G protein α-subunit XLαs is a critical regulator of energy and glucose metabolism and sympathetic nerve activity in adult mice. J Biol Chem 281: 18989–18999.
• 14   Sano  S,  Nakamura  A,  Matsubara  K,  Nagasaki  K,  Fukami  M, et al. (2017) (Epi)genotype-phenotype analysis in 69 Japanese patients with pseudohypoparathyroidism type 1. J Endocr Soc 2: 9–23.
• 15   Eun  JJ,  Lee  HJ,  Kim  JK (2014) Developmental profiles of preschool children with delayed language development. Korean J Pediatr 57: 363–369.
• 16   Turan  S,  Fernandez  RE,  Aydin  C,  Zoto  T,  Reyes  M, et al. (2014) Postnatal establishment of allelic Gαs silencing as a plausible explanation for delayed onset of parathyroid hormone resistance owing to heterozygous Gαs disruption. J Bone Miner Res 29: 749–760.