Clinical and molecular analyses of isolated central congenital hypothyroidism based on a survey conducted in Japan

Abstract

Central congenital hypothyroidism (CH) can occur as an isolated deficiency or as part of combined pituitary hormone deficiency. Unlike primary CH, central CH cannot be detected by newborn screening (NBS) using dry filter paper blood TSH levels, and early diagnosis remains challenging. In this study, the clinical and genetic backgrounds of patients with isolated central CH were determined through a questionnaire-based survey among members of the Japanese Society for Pediatric Endocrinology. The known causes of isolated central CH were studied in 14 patients, including six with previously reported patient data. The results revealed IGSF1 and TBL1X pathogenic variants in nine and one patient, respectively. All six patients with low free thyroxine (FT4) levels detected in NBS carried IGSF1 pathogenic variants. Five patients with isolated central CH diagnosed after 3 months of age were variant-negative, except for one female patient with a heterozygous IGSF1 variant. Two of the four variant-negative patients and a variant-positive patient were diagnosed with pituitary hypoplasia. One and two patients with IGSF1 variant had obesity and intellectual disability, respectively. Left amblyopia was identified in the patient with a TBL1X variant. The study revalidated that IGSF1 variants comprise the most frequent pathogenic variant in patients with isolated central CH in Japan. The neonatal period is the optimal time for the diagnosis of central CH, particularly IGSF1 abnormalities, and the introduction of T4 screening should be considered in the future, taking cost-effectiveness into consideration.

CONGENITAL HYPOTHYROIDISM (CH) is one of the most common preventable causes of intellectual disability. Primary CH is associated with abnormalities in the thyroid gland, whereas, central CH causes abnormalities in the hypothalamus or pituitary gland. Newborn screening (NBS) programs for primary CH have been implemented in several countries since the 1970s, which facilitate the early detection and treatment of CH, supporting the prevention of brain damage and subsequent intellectual disability [1]. However, central CH is not targeted in NBS programs owing to its milder phenotype and rarity compared with those of primary CH. However, a T4 screening in specific areas revealed a high incidence (1 in 13,000–30,000) of central CH than its previously known abundance [2-4]. Central CH cannot be detected in newborns using the currently available filter paper blood TSH measurement system; consequently, delayed treatment causes intellectual disability in some patients with central CH [5, 6].

The etiology of central CH can be broadly classified into two groups: combined pituitary hormone deficiency (CPHD) and as isolated central CH (or isolated TSH deficiency). Pathogenic variants in TSHβ and TRHR variants are well-known genetic causes of isolated central CH. Recently, new genes, associated with isolated central CH, have been reported and the molecular basis of isolated central CH has been elucidated. While exploring central CH-related genes, abnormalities in immunoglobulin superfamily 1 (IGSF1) associated with X-linked macroorchidism [7], transducin beta-like protein 1, X-linked (TBL1X) associated with X-linked deafness [8], and insulin receptor substrate 4 (IRS4) associated with X-linked central CH have been reported [9]. IGSF1 abnormalities are also characterized by hypoprolactinemia, transient GH deficiency (childhood), delayed puberty with respect to testosterone rise but normal timing of testes growth, macroorchidism and GH excess (later in life) [10, 11].

Comprehensive genetic analyses of patients with isolated central CH have been reported in the Netherlands and Japan [12, 13]; among these studies, only one conducted in the Netherlands included a nationwide population.

Considering the need for comprehensive data based on the Japanese population, in the present study, we aimed to clarify the clinical and genetic background of patients with isolated central CH using a nationwide survey among members of the Japanese Society for Pediatric Endocrinology.

Materials and Methods

Patients

This study included a questionnaire-based survey conducted among 191 council members of the Japanese Society for Pediatric Endocrinology between 2015 and 2016. The details of this questionnaire survey have been previously reported [14]. Patients aged <18 years, diagnosed with central CH between April 2004 to March 2014, were recruited in this study. The survey response percentage was 47%. Initially, 78 patients (55 male and 23 female) with central CH were identified in Japan based on the following diagnostic criteria: 1. hypothyroidism of hypothalamic or pituitary origin; 2. disproportionately low TSH levels compared to those of thyroid hormones, even in the absence of a TRH stimulation test; 3. the patients were reevaluated at a possible later age to determine whether the condition was transient or no so as to exclude non-thyroidal illness syndrome. The exclusion criteria were as follows: 1. acquired causes, such as brain tumors or head trauma; 2. hypothyroxinemia in a preterm infant; 3. thyroxine-binding globulin (TBG) deficiency, and 4. prenatal exposure to poorly controlled maternal Graves’ disease.

After identifying the institutions having patients with central CH in the first survey, a secondary survey form was sent to them and clinical information was collected between July 2015 and March 2016 [14]. The response rate for the secondary survey was 85%, yielding a secondary survey list of 78 patients with central CH. Based on the description in the secondary survey, we defined isolated central CH patients, as not accompanied by impaired secretion of anterior pituitary hormones other than TSH, excluding prolactin.

Among these 78 patients, CPHD and isolated central CH were detected in 59 and 19, respectively [14]. Before the genetic analysis conducted in 2018–2019, clinical information of the 19 patients suspected of having isolated CH was further reviewed, and one patient with maternal Graves’ disease, two with transient hypothyroidism, one with TBG deficiency, and one with CPHD were excluded from this study.

Newborn screening for CH in Japan

According to a national project, NBS for CH has been conducted using filter blood TSH tests in almost all births in Japan since 1979. According to a survey conducted by the technical committee of the Japanese Society for Neonatal Screening, 11.8% of all newborns were simultaneously screened for TSH and FT4 in 2012 [14].

Collection of clinical data

In a secondary survey, clinical data, including patient background (age, sex, perinatal information, place of birth, family history of thyroid disease), the reason for the diagnosis of central CH, results of NBS, symptoms at diagnosis, information on thyroid function tests, anterior pituitary function other than TSH, brain magnetic resonance imaging (MRI), comorbidities, and psychomotor development, were collected. The clinical characteristics of the patients with and without pathogenic variants were compared. This nationwide survey was approved by the Ethics Committee of the Niigata University School of Medicine (No. 2229).

Molecular genetic analysis

Genomic DNA was extracted using peripheral venous blood and the QIAamp^® DNA Blood Mini Kit (QIAGEN, Netherlands) following protocol provided by the manufacturer. Variant analysis of known isolated central CH-inducing genes, including IGSF1, TBL1X, TSHβ, TRHR, and IRS4, was performed using the Sanger sequencing method. Primers for the amplification of these genes (GenBank accession numbers NM_001555.5, NM_005647.4, NM_000549.5, NM_003301.7, NM_001379150.1) were designed using the Primer3 Plus software. After PCR amplification and purification, amplicons were sequenced using a SeqStudio Genetic Analyzer (Applied Biosystems), following the instructions provided the manufacturer.

In female patients with heterozygous variants X-linked IGSF1, TBL1X, and IRS4, the pattern of X-chromosome inactivation was examined through a methylation assay of AR Exon 1 that includes a CAG repeat length polymorphism using leukocyte genomic DNA [15].

Sequence data were aligned with the reference genome (hg38) using the SnapGene software (GSL Biotech LLC); variants with the minor allele frequency >1% were excluded. Variants were evaluated using several programs to predict pathogenicity. Based on the American College of Medical Genetics and Genomics (ACMG) guidelines, the detected sequence variants were interpreted [16]. Moreover, this study was approved by the Genetic Ethics Committee of the Niigata University (No. G2018-0007). Written informed consent for participation in this study was obtained from either patients or their guardians.

Statistical analysis

Summary statistics for non-normally distributed quantitative variables are expressed as medians and interquartile ranges. Categorical data were summarized as numbers and percentages. Either of Mann–Whitney U test or Kruskal–Wallis test was used to compare two or more groups. Differences were considered statistically significant at a two-sided p value of less than 0.05. All statistical tests were performed using the SPSS version 28.0.0.1.

Results

Frequencies of variants in participants

Six of the 14 eligible patients, recruited in this study, were previously analyzed at other institutions, and the published results were used in this study (Case 3, 4, 6, 7, 8, and 9 in Table 1) [12, 17, 18]; hence, sequencing analyses were performed using blood collected from the remaining eight patients. Overall, we detected potential disease-causing variants (9 IGSF1 variants and a TBL1X variant) in 10 patients while analyzing five genes (Table 1, Fig. 1A). IGSF1 variants included five nonsense variants, two splice site variants, a missense variant, and a deletion; these variants included novel variants IGSF1 c.1750+1G>A (GenBank accession number 921397) and c.2548C>T (GenBank accession number 921396). IGSF1 c.2609-1G>A (dbSNP rs1246691313) is registered in the genome aggregation database (gnomAD); however, its clinical features were not described and its significance was not reported in ClinVar. The TBL1X variant was previously identified as a missense variant [8]. All variants were categorized as likely pathogenic or pathogenic, according to the American College of ACMG guidelines (Table 1).

Table 1

Summary of variants identified in 10 patients with isolated TSH deficiency

Case	Sex	Age at first visit (Month)	Nucleotide change	Protein change	ACMG classification	Protein function prediction				Frequency in normal populations			References or db SNP number
						Polyphen2 (HumDiv)1)	Mutation Taster2)	CADD PHRED score3)	provean4)	1000G (East Asian)5)	gnomAD (East Asian)6)	14KJPN7)
IGSF1 (NM_001555.5)
1	M	2	c.1750+1G>A	—	Pathogenic	—	Disease causing	23.0	—	0.0000	0.0000	0.0000	Not reported
2	F	13	c.2609–1G>A	—	Pathogenic	—	Disease causing	33.0	—	0.0000	0.0000	0.0000	rs1246691313
3	M	0	c.1918C>T	p.Gln640*	Pathogenic	—	Disease causing	35.0	Neutral	0.0000	0.0000	0.0000	Patient 4 in Nakamura et al. 2013 [17]
4	M	1	c.2553T>G	p.Tyr851*	Pathogenic	—	Disease causing	36.0	Deleterious	0.0000	0.0000	0.0000	ID2 in Table 1 within Sugisawa et al. 2019 [12]
5	M	0	c.2548C>T	p.Arg850*	Pathogenic	—	Disease causing	38.0	Neutral	0.0000	0.0000	0.0000	Not reported
6	M	0	c.2024T>C	p. Leu675Pro	Likely pathogenic	Probably Damaging	Disease causing	23.9	Deleterious	0.0000	0.0000	0.0000	ID1 in Table 1 within Sugisawa et al. 2019 [12]
7	M	0	120 bp deletion at Xp26.1	—	Pathogenic	—	—	—	—	0.0000	0.0000	0.0000	Patient 3 in Tajima et al. 2014 [18]
8	M	1	c.3230T>A	p.Val1077Glu	Likely pathogenic	Probably Damaging	Disease causing	23.7	Deleterious	0.0000	0.0000	0.0000	Patient 3 in Nakamura et al. 2013 [17]
9	M	1	c.3550C>T	p.Arg1184*	Pathogenic	—	Disease causing	38.0	Deleterious	0.0000	0.0000	0.0000	ID3 in Table 1 within Sugisawa et al. 2019 [12]
TBL1X (NM_005647.4)
10	M	2	c.1526A>G	p.Tyr509Cys	Pathogenic	Probably Damaging	Disease causing	26.6	Deleterious	0.0000	0.0000	0.0000	Heinen et al. 2016

The URLs utilized are as follows; in silico analyses and frequency in normal populations were performed by using the default parameters.

1) Polyphen2 (prediction of functional effects of human nsSNPs): http://genetics.bwh.harvard.edu/pph2/; The score ranges from 0 to 1, and the corresponding prediction is 0.15–1.0 “Possibly damaging”. 0.85–1.0 “confidently predicted to be damaging”

2) MutationTaster: https://www.genecascade.org/MutationTaster2021/#transcript (MutationTaster2021, GRCh37/Ensembl 102); Alterations are classified as disease causing, disease causing automatic, polymorphism or polymorphism automatic.

3) CADD (Combined Annotation–Dependent Depletion): http://cadd.gs.washington.edu/(Current version: 1.6, GRCh38/hg38); PHRED scores of >10–20 are regarded as deleterious, and those of >20 indicates the 1% most deleterious.

4) provean (Protein Variation Effect Analyzer) : http://provean.jcvi.org/about.php; The forecast is deleterious if score ≤–2.5.The forecast is neutral if score  ≥–2.5.damaging” >0.85;damaging” 0.85–0.15 and “benign” <0.15. “possibly

5) 1000Genomes (IGSR: The International Genome Sample Resource): https://www.internationalgenome.org/

6) gnomAD (Genome Aggregation Database): https://gnomad.broadinstitute.org/

7) 14KJPN (Whole-genome sequences of 14,000 healthy Japanese individuals and construction of the highly accurate Japanese population reference panel): https://jmorp.megabank.tohoku.ac.jp/202112/variants/

Fig. 1

Molecular genetic analysis findings

A. Identification of IGSF1 and TBL1X variants.

The chromatograms indicate the IGSF1 (NM_001555.5) and TBL1X (NM_005647.4) variants in case 1, 2, 5, and 10. Case1: IGSF1 position 1750+1 of coding DNA, Case 2: IGSF1 position 2609–1 of coding DNA, Case 5: IGSF1 position 2548 of coding DNA (p. Arg850*), Case 10: TBL1X position 1526 of coding DNA (p. Tyr509Cys). Arrows indicate the variant nucleotides.

B. Frequency of genetic variants, associated with central congenital hypothyroidism, in patients with isolated TSH deficiency.

Among patients with central CH, the percentages of isolated central CH associated with different variants were as follows: 64.3% (95% confidence interval [CI]: 38.4%–85.4%) (9/14) had IGSF1 variants, 7.1% (95% CI: 0–20.5%) (1/14) had a TBL1X variant; the IGSF1 variant was the most common one, whereas, no TSHβ, TRHR, or IRS4 variant was identified (Fig. 1B). In a female patient with IGSF1, the X-chromosome inactivation ratio was 45.1:54.9% (Case 2 in Table 1), indicating no skewed X-chromosome inactivation.

Clinical and biochemical features and phenotypes of patients with isolated central CH

The clinical characteristics of the patients are summarized in Table 2. Thirteen patients were male and one patient with a heterozygous IGSF1 variant was female. Nine patients with the IGSF1 variants were physically large; at birth, their length, weight and head circumference expressed as median (standard deviation score [SDS], interquartile range, SDS range) were 52.5 cm (+1.2, 50.3–54.0 cm, –0.23–2.78), 3,578 g (+1.6, 3,096–3,938 g, 0.06–3.68), and 36 cm (+2.0, 34.5–37.5 cm, –0.29–2.96), respectively. Contrastingly, the four variant-negative patients exhibited average body size, with length, weight, and head circumference, expressed as median (SDS), being 49.5 cm (+0.74), 3,187 g (+0.34), and 34.5 cm (+1.36), respectively, at birth. The median age at diagnosis was 14 months in the group carrying no variants, and 1 and 2 months in patients with IGSF1 and TBL1X variants, respectively (Fig. 2).

Table 2

Summary of clinical features of 14 patients with isolated central congenital hypothyroidism

Case	Sex	Pathogenic variant	At birth				At first visit							At this survey						References
			Length cm (SD)	Weight g (SD)	HC cm (SD)	NBS results	Age (month)	Chief Complaints	TSH (μIU/mL)	FT4 (ng/dL)	FT3 (pg/mL)	Prolactin (ng/mL)	LH/FSH (mIU/mL)	Age (years)	Height cm (SD)	Weight kg	Intellectual disability	MRI fidings	Another findings
1	M	IGSF1	53 (1.65)	3,880 (1.61)	34 (–0.29)	Low FT4*	2	Prolonged jaundice	5.40	0.82	3.71	ND	ND	6.8	117.4 (–0.16)	25.65	–	Normal	Neurofibromatosis
2	F	IGSF1	50 (0.36)	3,004 (0.06)	ND	Normal TSH	13	Poor weight gain	1.26	0.59	1.59	ND	ND	10.3	138.4 (–0.12)	26.95	–	Rathke’s cyst	—
3	M	IGSF1	54 (2.21)	4,510 (3.68)	ND	Normal TSH	0	Jaundice, Poor sucking	1.66	0.78	3.82	3.1	0.43/2.47	4.4	105.9 (0.88)	20.9	–	Normal	Amblyopia	[17]
4	M	IGSF1	54 (2.19)	3,578 (0.82)	35 (0.91)	Low FT4*	1	NBS	2.61	0.44	2.80	ND	1.6/8.3	1.3	79.5 (0.63)	13.1	–	Normal	—	[12]
5	M	IGSF1	55 (2.78)	3,996 (1.96)	37 (2.55)	Normal TSH	1	Incidentally**	0.42	0.60	ND	ND	ND	3	90.5 (–0.65)	15.85	–	Normal	—
6	M	IGSF1	51 (0.61)	3,528 (1.27)	36 (2.04)	Low FT4*	0	NBS	4.40	0.61	3.13	33.2	ND	3.6	103.9 (1.73)	24.15	–	Normal	Obesity	[12]
7	M	IGSF1	ND	3,188 (0.17)	ND	Low FT4*	0	NBS	4.00	0.56	2.35	33.88	1.01/6.47	7.7	121.2 (–0.42)	27.9	+	PH	—	[18]
8	M	IGSF1	46 (–0.23)	2,924 (1.57)	ND	Low FT4*	1	NBS	7.95	0.80	3.41	86.24	1.9/2.6	4.3	106 (0.9)	18.5	+	Normal	—	[17]
9	M	IGSF1	52 (0.74)	3,868 (1.62)	38 (2.96)	Low FT4*	1	NBS	3.02	0.72	3.13	46.25	ND	7.3	122.8 (0.43)	24.1	–	Normal	—	[12]
10	M	TBL1X	49 (–0.49)	3,146 (–0.17)	ND	Normal TSH	2	Prolonged jaundice, Poor weight gain	2.35	0.83	3.13	ND	<0.1/0.93	9.5	130.5 (–0.63)	31.9	–	Rathke’s cyst	Amblyopia
11	M	None	49.3 (1.03)	2,994 (1.14)	34.5 (1.36)	Normal TSH	6	Growth disturbance (poor weight gain)	1.44	0.62	2.78	ND	ND	9.7	138.4 (0.67)	29.1	–	Normal	—
12	M	None	ND	3,606 (1.03)	36.3 (2.05)	Normal TSH	13	Intellectual disability (Mild motor developmental delay)	3.81	0.58	4.00	6.4	0.2/1.6	6.8	109 (–1.45)	17.5	+	Rathke’s cyst	Bilateral sensorineural hearing loss
13	M	None	49.5 (0.74)	2,514 (–0.88)	34 (0.74)	Normal TSH	15	Growth disturbance (length –3.3 SD, weight –3.0 SD)	1.29	1.05	4.02	26.44	<0.07/0.43	11.5	129.4 (–2.24)	24.8	–	PH	Afebrile seizures
14	M	None	50.5 (0.24)	3,380 (–0.35)	ND	Normal TSH	37	Autism and muscle weakness	0.51	0.68	2.41	ND	<0.07/<0.30	3.8	95.9 (–0.74)	14.75	+	PH, ONH	Visual disorder

HC: Head circumference, ND: no data, NBS: newborn screening, FT4: free thyroxine, TSH: thyroid-stimulating hormone, PH: Pituitary hypoplasia, ONH: Optic nerve hypoplasia

* Low FT4 and normal TSH levels, **Case 5 was diagnosed incidentally for neonatal seizures.

Fig. 2

Comparative clinical characteristics in variant-positive and variant-negative patients.

The left panel shows the timing of the diagnosis of central congenital hypothyroidism (CH), and the remaining three panels demonstrate the variable thyroid function during diagnosis. Central CH was diagnosed at a significantly earlier age in the variant-positive group than in the variant-negative group, however, results of thyroid function tests indicated no differences between the two groups. Each dot represents a data point for individual patients. For each group, the horizontal lines indicate the median values, shaded squares represent the interquartile ranges (IQR), and upper and lower whiskers represent the maximum and minimum values, respectively.

Simultaneous TSH and FT4 measurements in six patients revealed low FT4 levels; the remaining eight patients underwent only TSH measurement during NBS and exhibited normal NBS results. NBS revealed low FT4 levels, and poor weight gain or prolonged jaundice in six and four patients with pathological variants, respectively. In contrast, four variant-negative patients were diagnosed with growth retardation or intellectual disability (Table 2). NBS is the most commonly used diagnostic strategy for patients with pathogenic variants. Moreover, IGSF1 variants were detected in all six NBS-positive patients.

The results of thyroid function tests at diagnosis did not significantly differ depending on the presence of variants (Fig. 2). The median IGF1 SDS at diagnosis was +0.1 (n = 4) and –1.3 (n = 3) in patients with IGSF1 variants-positive and variant-negative, respectively. In six patients with IGSF1 variants, serum prolactin levels were measured during the neonatal period, which was normal or high in all except one. Among the four IGSF1 variant-positive patients, for whom serum gonadotropin levels were measured in early infancy, FSH levels were elevated in only two at the neonatal stage (Table 2).

In the secondary questionnaire, the MRI findings were based on the description provided by primary physician. MRI of the pituitary revealed pituitary hypoplasia in one patient and Rathke’s cyst in the variant-positive group. Two of the four variant-negative patients exhibited pituitary hypoplasia and a thin pituitary stalk. During the survey, intellectual disability was noted in two IGSF1 variant-positive and two variant-negative patients. Amblyopia was observed in a variant-negative patient and a TBL1X variant-positive patient. Optic atrophy was noted in a variant-negative patient. Afebrile seizures and bilateral sensorineural hearing loss were observed in one patient each, without variants. The height SDS at the time of this survey was 0.43 (range: –0.65 to 1.73) in patients with IGSF1 variants, –0.63 in the patient with the TBL1X variant, and –1.5 (range: –2.2 to 0.67) in patients with no pathological variants. In the variant-negative group, a patient had short stature having SDS below –2.

Discussion

We studied the clinical and genetic characteristics of 14 patients (13 males and 1 female) with isolated central CH using data collected in a questionnaire-based survey conducted among members of the Japanese Society for Pediatric Endocrinology. We identified IGSF1 and TBL1X variants in nine and one patient, respectively. All six patients with isolated central CH diagnosed through low FT4 on NBS exhibited hemizygous IGSF1variants, whereas a female patient with isolated central CH diagnosed after 12 months of age exhibited a heterozygous IGSF1 variant.

To date, comprehensive genetic analyses of patients with isolated central CH have been conducted in the Netherlands and Japan. Sugisawa et al. [12] reported sequencing results for IGSF1, TBL1X, IRS4, TSHβ, and TRHR in 13 Japanese patients (11 boys and 2 girls) with isolated central CH. IGSF1 and TBL1X variants were detected in five (38.4%) and one (7.7%) patients, respectively. In the Netherlands national survey conducted among 32 patients with isolated central CH, Naafs et al. [13] reported the frequencies of IGSF1, TBL1X, IRS4, TSHβ, and TRHR variants to be 53%, 16%, 16%, 3.1%, and 3.1%, respectively. Our results indicated the presence of the IGSF1 variant in most patients, and hence, it should be considered the most frequent variant regardless of race. In Japan, the TBL1X variant is the second most common variant associated with isolated central CH, whereas, no IRS4 variant was identified. Further case studies can validate whether this finding is race-dependent. The apparent sex-associated differences in variants can be attributed to the frequency of patients with IGSF1 variants.

Patients with isolated central CH diagnosed through NBS are more likely to harbor IGSF1 variants, whereas, those diagnosed after infancy were less likely to harbor pathogenic variants; this difference is possibly attributed to the CPHD detected in some older patients with isolated central CH. In the present study, the gene responsible for CPHD were not analyzed. Previously, CPHD-associated structural abnormalities of the pituitary gland were reported in some patients with central CH [13]. In our study, two of four variant-negative patients with isolated central CH exhibited structural abnormalities of the pituitary gland, however, no other abnormalities in anterior pituitary function were found at the time of recruitment. These patients may require further diagnostic evaluation for CPHD. Another reason could be related to the fact that most previously reported IGSF1 variants identified in family histories were correlated with mild hypothyroidism or were asymptomatic [19]. We speculate that the neonatal period of patients with IGSF1 variants has a high demand for thyroid hormone, resulting in low FT4 levels; afterward, hypothyroidism becomes mild and less detectable. Additionally, non-thyroidal illness syndromes cannot be ruled out in patients presenting with isolated central CH after infancy.

Mostly, an isolated central CH is detected during the neonatal stage or early infancy. The levels of thyroid hormones peak in the early neonatal period and decline after the first month of life [20, 21]. Thyroid hormones have multiple survival-enhancing functions, including the stimulation of appetite center [22], promotion of bilirubin removal in physiologic jaundice [23], and postnatal organ maturation [24]. Therefore, hypothyroidism in the neonatal stage may induce poor feeding and prolonged jaundice, providing a potential opportunity for the detection of this abnormality. In the Netherlands, T4 screening is performed for almost all babies, and most patients with central CH are detected through NBS, while others are also found with the diagnosis of GHD [25]. A recent study in the Netherlands revealed that the economic benefit of T4 testing in NBS significantly outweigh the costs [26]. Therefore, the neonatal period is the optimal time to diagnose patients with isolated central CH. Moreover, NBS using FT4 can be considered economically effective.

Various clinical presentations other than central CH are associated with IGSF1 abnormalities [5, 7, 11, 17-19, 27]. Previously, patients with IGSF1 abnormalities were often reported to be overweight and large at birth [5]. In our study, IGSF1 variant positive patients were also larger in length, weight, and head circumference at birth. Additionally, although low serum prolactin levels have been reported in patients with IGSF1 abnormalities, only one patient in this study had low serum prolactin levels, probably because most patients were within the first month of life. IGF1 levels are reported to be normal in childhood and tend to be high in adulthood [10]. The mean IGF-1 levels of the four pediatric patients in this study were normal as +0.1 SD. No significant differences in thyroid function were detected between the variant-positive and variant-negative patients (Fig. 2). It is difficult to infer the results of genetic testing from thyroid function tests alone. However, the age of the variant-positive patients was significantly lower because many of them were found to be NBS positive.

IGSF1 abnormalities are often associated with GH deficiency in childhood. Furthermore, IGSF1 hemizygous variant-associated CPHD cases with pituitary structural abnormalities were reported [28]. Similarly, in the present study, case 7 with IGSF1 hemizygous variant was associated with pituitary hypoplasia, hence, GH secretion should be further evaluated. Although the common mechanism underlying effects of IGSF1 on the hypothalamus-pituitary-thyroid axis and the hypothalamus-pituitary-gonadal axis remains unknown, IGSF1-mediated regulation of the TGFβ and Activin-SMAD signaling pathways was proposed, which regulate the expression of TRHR and FSHβ [29]. During the neonatal period, referred to as mini-puberty, during which the hypothalamus-pituitary-gonadal axis is activated [30]. In this study, the high FSH levels in 50% of the tested samples collected from four patients at the neonatal stage are comparable to the reference data [30].

All participants, except one, recruited in this study were male. IGSF1 and TBL1X, the major factors involved in central CH, are located on the X chromosome, which may contribute to male predominance. However, 28% of women aged 32–80 years with IGSF1 heterozygous variants, regardless of the presence or absence of X chromosome inactivation skew, were diagnosed with central CH [10]. This suggests that patients with IGSF1 abnormalities exhibit X-chromosome dominant inheritance with low penetrance in terms of thyroid function tests. Although, the mildness of this phenotype hinders its effective clinical diagnosis. Interestingly, female patient carrying an IGSF1 heterozygous variant displayed a pronounced decrease in FT4 levels during weight loss. Although it is difficult to distinguish between central hypothyroidism and non-thyroidal illness syndromes, undernourishment may exacerbate hypothyroidism.

In this study, we present the first cross-sectional nationwide survey among members of the Japanese Society for Pediatric Endocrinology elucidating the clinical and genetic background of patients with isolated central CH. However, this study had some limitations: 1) FT4 screening is performed in a few municipalities of Japan, hence, many patients with isolated central CH possible remain undiagnosed. 2) The rate of questionnaire collection was 47%, which did not cover the entire population of isolated central CH in Japan. Therefore, this study cannot reflect the nationwide data. 3) As patients were recruited based on cross-sectional data collected using questionnaires, patients with transient central CH and CPHD may not completely excluded. 4) Structural abnormalities in the pituitary, detected through MRI, were analyzed using secondary survey questionnaires based on vague definitions. 5) Finally, we did not perform gene analysis for CPHD nor exome sequencing.

In conclusion, in this study, IGSF1 pathogenic variants were identified in more than half of the patients with isolated central CH, revalidating previously known data. The neonatal period is the optimal time for the diagnosis of central CH, particularly IGSF1 abnormalities, and the introduction of T4 screening should be considered in the future, taking cost-effectiveness into consideration.

Acknowledgments

We thank the patients and their family members for supporting the study. We would also like to thank Toshihiro Tajima for his cooperation with the national survey questionnaire for central congenital hypothyroidism as a leader of the Mass Screening Committee of the Japanese Society for Pediatric Endocrinology.

Disclosure

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

References

• 1   Ford  G,  LaFranchi  SH (2014) Screening for congenital hypothyroidism: a worldwide view of strategies. Best Pract Res Clin Endocrinol Metab 28: 175–187.
• 2   Lanting  CI,  van Tijn  DA,  Loeber  JG,  Vulsma  T,  de Vijlder  JJM, et al. (2005) Clinical effectiveness and cost-effectiveness of the use of the thyroxine/thyroxine-binding globulin ratio to detect congenital hypothyroidism of thyroidal and central origin in a neonatal screening program. Pediatrics 116: 168–173.
• 3   Fujiwara  F,  Fujikura  K,  Okuhara  K,  Tsubaki  J,  Fukushi  M, et al. (2008) Central congenital hypothyroidism detected by neonatal screening in Sapporo, Japan (2000–2004): it’s prevalence and clinical characteristics. Clin Pediatr Endocrinol 17: 65–69.
• 4   Adachi  M,  Soneda  A,  Asakura  Y,  Muroya  K,  Yamagami  Y, et al. (2012) Mass screening of newborns for congenital hypothyroidism of central origin by free thyroxine measurement of blood samples on filter paper. Eur J Endocrinol 166: 829–838.
• 5   Zwaveling-Soonawala  N,  van Trotsenburg  AS,  Verkerk  PH (2015) The severity of congenital hypothyroidism of central origin should not be underestimated. J Clin Endocrinol Metab 100: E297–E300.
• 6   Naafs  JC,  Marchal  JP,  Fliers  E,  Verkerk  PH,  Luijten  MAJ, et al. (2021) Cognitive and motor outcome in patients with early-detected central congenital hypothyroidism compared with siblings. J Clin Endocrinol Metab 106: e1231–e1239.
• 7   Sun  Y,  Bak  B,  Schoenmakers  N,  van Trotsenburg  AS,  Oostdijk  W, et al. (2012) Loss-of-function mutations in IGSF1 cause an X-linked syndrome of central hypothyroidism and testicular enlargement. Nat Genet 44: 1375–1381.
• 8   Heinen  CA,  Losekoot  M,  Sun  Y,  Watson  PJ,  Fairall  L, et al. (2016) Mutations in TBL1X are associated with central hypothyroidism. J Clin Endocrinol Metab 101: 4564–4573.
• 9   Heinen  CA,  de Vries  EM,  Alders  M,  Bikker  H,  Zwaveling-Soonawala  N, et al. (2018) Mutations in IRS4 are associated with central hypothyroidism. J Med Genet 55: 693–700.
• 10   Joustra  SD,  Schoenmakers  N,  Persani  L,  Campi  I,  Bonomi  M, et al. (2013) The IGSF1 deficiency syndrome: characteristics of male and female patients. J Clin Endocrinol Metab 98: 4942–4952.
• 11   Joustra  SD,  Roelfsema  F,  van Trotsenburg  ASP,  Schneider  HJ,  Kosilek  RP, et al. (2020) IGSF1 Deficiency results in human and murine somatotrope neurosecretory hyperfunction. J Clin Endocrinol Metab 105: e70–e84.
• 12   Sugisawa  C,  Takamizawa  T,  Abe  K,  Hasegawa  T,  Shiga  K, et al. (2019) Genetics of congenital isolated TSH deficiency: mutation screening of the known causative genes and a literature review. J Clin Endocrinol Metab 104: 6229–6237.
• 13   Naafs  JC,  Verkerk  P,  Fliers  E,  van Trotsenburg  ASP,  Zwaveling-Soonawala  N (2020) Clinical and genetic characteristics of Dutch children with central congenital hypothyroidism, early detected by neonatal screening. Eur J Endocrinol 183: 627–636.
• 14   Nagasaki  K,  Kubota  T,  Kobayashi  H,  Sawada  H,  Numakura  C, et al. (2017) National survey for congenital central hypothyroidism in Japan. Jpn J Mass Screening 27: 9–15 (In Japanese).
• 15   Allen  RC,  Zoghbi  HY,  Moseley  AB,  Rosenblatt  HM,  Belmont  JW (1992) Methylation of HpaII and HhaI sites near the polymorphic CAG repeat in the human androgen-receptor gene correlates with X chromosome inactivation. Am J Hum Genet 51: 1229–1239.
• 16   Richards  S,  Aziz  N,  Bale  S,  Bick  D,  Das  S, et al. (2015) Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med 17: 405–424.
• 17   Nakamura  A,  Bak  B,  Silander  TL,  Lam  J,  Hotsubo  T, et al. (2013) Three novel IGSF1 mutations in four Japanese patients with X-linked congenital central hypothyroidism. J Clin Endocrinol Metab 98: E1682–E1691.
• 18   Tajima  T,  Nakamura  A,  Morikawa  S,  Ishizu  K (2014) Neonatal screening and a new cause of congenital central hypothyroidism. Ann Pediatr Endocrinol Metab 19: 117–121.
• 19   Lauffer  P,  Zwaveling-Soonawala  N,  Naafs  JC,  Boelen  A,  van Trotsenburg  ASP (2021) Diagnosis and management of central congenital hypothyroidism. Front Endocrinol (Lausanne) 12: 686317.
• 20   Jayasuriya  MS,  Choy  KW,  Chin  LK,  Doery  J,  Stewart  A, et al. (2018) Reference intervals for neonatal thyroid function tests in the first 7 days of life. J Pediatr Endocrinol Metab 31: 1113–1116.
• 21   Naafs  JC,  Heinen  CA,  Zwaveling-Soonawala  N,  van der Schoor  SRD,  van Tellingen  V, et al. (2020) Age-specific reference intervals for plasma free thyroxine and thyrotropin in term neonates during the first two weeks of life. Thyroid 30: 1106–1111.
• 22   Kong  WM,  Martin  NM,  Smith  KL,  Gardiner  JV,  Connoley  IP, et al. (2004) Triiodothyronine stimulates food intake via the hypothalamic ventromedial nucleus independent of changes in energy expenditure. Endocrinology 145: 5252–5258.
• 23   Hamoud  AR,  Weaver  L,  Stec  DE,  Hinds  TD Jr (2018) Bilirubin in the liver-gut signaling axis. Trends Endocrinol Metab 29: 140–150.
• 24   Fonseca  TL,  Garcia  T,  Fernandes  GW,  Nair  TM,  Bianco  AC (2021) Neonatal thyroxine activation modifies epigenetic programming of the liver. Nat Commun 12: 4446.
• 25   van Iersel  L,  van Santen  HM,  Zandwijken  GRJ,  Zwaveling-Soonawala  N,  Hokken-Koelega  ACS, et al. (2018) Low FT4 concentrations around the start of recombinant human growth hormone treatment: predictor of congenital structural hypothalamic-pituitary abnormalities? Horm Res Paediatr 89: 98–107.
• 26   Stroek  K,  Heijboer  AC,  Bouva  MJ,  van der Ploeg  CPB,  Heijnen  MA, et al. (2020) Critical evaluation of the newborn screening for congenital hypothyroidism in the Netherlands. Eur J Endocrinol 183: 265–273.
• 27   Tajima  T (2017) A new cause of congenital central hypothyroidism-immunoglobulin superfamily member 1 gene deficiency. Journal of the Japan Thyroid Association 8: 33–36 (In Japanese).
• 28   Asakura  Y,  Abe  K,  Muroya  K,  Hanakawa  J,  Oto  Y, et al. (2015) Combined growth hormone and thyroid-stimulating hormone deficiency in a Japanese patient with a novel frameshift mutation in IGSF1. Horm Res Paediatr 84: 349–354.
• 29   Garcia  M,  Barrio  R,  Garcia-Lavandeira  M,  Garcia-Rendueles  AR,  Escudero  A, et al. (2017) The syndrome of central hypothyroidism and macroorchidism: IGSF1 controls TRHR and FSHB expression by differential modulation of pituitary TGFbeta and Activin pathways. Sci Rep 7: 42937.
• 30   Busch  AS,  Ljubicic  ML,  Upners  EN,  Fischer  MB,  Raket  LL, et al. (2022) Dynamic changes of reproductive hormones in male minipuberty: temporal dissociation of Leydig and Sertoli cell activity. J Clin Endocrinol Metab 107: 1560–1568.