A total of 1,753 children diagnosed with idiopathic neurodevelopmental disorder were recruited between 2016 and 2021 from the department of Child Health Care, Children's Hospital of Fudan University. Informed consent was obtained from patients or their guardians for genetic testing and further clinical evaluations. The recruited patients met at least one of the first two requirements and the third requirement:

– Patients with unknown global developmental disorder (GDD) or ID. GDD diagnosis was based on DQ lower than 70 in two or more developmental domains of Griffiths development scale—Chinese. ID diagnosis was based on IQ lower than 70 of Wechsler Preschool and Primary Scale of Intelligence (WPPSI) or Wechsler Intelligence Scale for Children (WISC).

The procedure of the FXS screening is illustrated in Figure 1, which has two steps. First step: Genomic DNA was extracted from peripheral blood samples in EDTA-coated Vacutainers. The 1,753 samples were first detected by using FMR1 CGG repeat PCR and processed for the capillary electrophoresis (CE) analysis. All samples were prepared for an laboratory developed test (LDT) FXS screening PCR with Roche FastStart™ Taq and a master mix containing a 10X PCR reaction buffer and 2.5 mM dNTP, a 10 mM 7-deaza-dGTP, a 25 mM MgCl2, and a GC-rich solution. The primer mix contained an fluorescein amidites (FAM)-labeled forward, reverse primer of FMR1 5′UTR and an FMR1 CGG-specific primer. A quantity of 1 μl of gDNA was transferred to the PCR master mix to amplify with an initial denaturation step of 95 °C for 5 min, 10 cycles of 97 °C for 35 s, 62 °C for 35 s, and 68 °C for 4 min, then 20 cycles of 97 °C for 35 s, 62 °C for 35 s, and 68 °C for 4 min with a 20-s extension at each cycle, and a final extension at 72 °C for 10 min. The ABI 3,130 Genetic Analyzer (Applied Biosystems, Foster City, CA, United States) was applied for FMR1 CGG repeat fragment analysis. The 0.1 μl of PCR product was transferred into a 96-well plate with 1% 500 LIZ™ Size Standard (Applied Biosystems, Foster City, CA, United States) in 10 μl Hi-Di™ Formamide (Applied Biosystems, Foster City, CA, United States). After capillary electrophoresis, the fragment analysis data interpretation was performed by using GeneMarker® (SoftGenetics, PA, United States).

Second step: A majority of the negative samples were further tested by using (WES)/panel and (array-CGH) to confirm whether there were mutations or CNVs in the genome and in FMR1.

To elucidate the clinical characteristics of Chinese FXS children, we analyzed all their medical history. Clinical data mainly included the following:

– Behavioral problems and family history were assessed and recorded by pediatrician observation and parental questionnaires.

Demographic and clinical data were analyzed using the statistical package SPSS 20.0.

Samples from 1,753 Chinese children with idiopathic NDD (male:female = 1,382:371, age: 0.5–16 years old) were first tested by tetraplet-primed PCR-capillary electrophoresis, and 41 samples with CGG over 200 were detected (male:female = 39:2). Most negative samples were further analyzed for WES/panel or array-CGH. A 5-year-old girl had a 64 kb deletion in Xq27.3, including FMR1 gene, which was screened by using array-CGH. The incidence of FXS in Chinese children with idiopathic NDD was 2.4%, and those with FXS 2.38% had a deletion (Table 1).

In this study, we determined that the incidence of FXS in Chinese children with idiopathic NDD was 2.4%. This percentage was slightly lower than that reported in Western countries. However, compared with previous studies in China, it was higher than that reported by Chen et al. and Pang et al. (12, 13), 0.93% (5/540) and 0.6% (2/324), respectively, close to that reported by Zhong et al. (14), 2.8% (32/1127), and lower than that reported by Zhang et al. and Shen et al. (10, 15), 21.2% (7/33) and 6.8% (6/88), respectively. Two vital factors can be attributed to the different percentages of incidence of FXS among the Chinese population, and these are the diversity of the recruited populations and different sample sizes of each of these studies. Previous studies conducted FXS screening among people with different degrees of ID and the varying severities of ID among the recruited populations in various studies may affect the results. In this study, FXS screening was carried out in patients with idiopathic NDD and there was an expansion of the recruited populations compared with that in previous studies. In terms of sample size, the percentage of error of research results with small sample sizes may be higher.

We also screened a girl with FXS who had a 64 kb deletion in Xq27.3, including the FMR1 gene, by using array-CGH. In those with FXS, 2.38% had a deletion. A hyper expansion of CGG trinucleotide repeats over 200 in the 5′untranslated region of the FMR1 gene is responsible for the majority of FXS cases, and less than 1% of individuals with FXS have a sequence variant, a partial deletion (18). The proportion of CNV, which was more than 1% in our study, may be attributed to the small sample size. On the other hand, this result also indicates the importance of using different research methods to elucidate atypical FMR1 gene changes.

In recent years, research on FXS-targeted drugs has developed rapidly. Several targeted drugs have been identified, which have the potential to reverse the neurobiological aspects of FXS (19–22). A screening of FMR1 full mutation provides the possibility for FXS-targeted drug therapy and further medical support.

We elaborated the clinical characteristics of Chinese children with FXS in terms of five factors: general growth and development, repetitive behaviors, social and emotional problems, complications, and facial features.

The height of the children was within the normal range. However, we found that two boys were overweight. The frequency of overweight/obesity was lower than that (53%–61%) previously reported (23). The average IQ/DQ of all FXS patients was 48, which was similar to the mean value of previous reports (23). We showed that the age of meaningful words and walking alone were the two most accurate milestones recorded from the memory of parents. The average age of meaningful words was 2 years and 10 months and that of walking alone was 1 year and 7 months, which were slightly different from those of previous reports (23). The frequency of the age of meaningful words and walking alone older than one and a half years was 88% and 48%, respectively, suggesting that the delay in reaching these two milestones is common in children with FXS.

FXS is the most prevalent monogenic cause of ASD. Individuals with FXS presented several behaviors of autism, such as social abnormalities and repetitive behaviors. Social withdrawal, social anxiety, shyness, hyperarousal to sensory stimulation, and hypersensitivity to changes were frequent in our study (≥50%). Repetitive motor behaviors such as hand-flapping and hand-biting were at 50% and 25%, respectively, which were lower than those (70.8% and 54.2%) reported previously (24). In addition, emotional problems are common in FXS. A total of 56% and 58% of FXS children in this cohort were found to be emotionally labile and prone to temper tantrums. Self-injury accounted for 19%, which was greater than that (15%) in another study (25). Aggressiveness occurred in 28%, which was much lower than that (90%) reported in a previous review (25). This was likely related to cultural differences.

Among complications of FXS, ADHD is considered one of the most common comorbidities of FXS. In our cohort, ADHD accounted for 64%, which was slightly higher than that (54%–59%) reported previously (23). The incidence of vomiting was high at 33% in our cohort. The proportion of other gastrointestinal symptoms was relatively low, and diarrhea and constipation accounted for 8%. Gastroesophageal reflux was found in two young children and the frequency was lower than that (11%) of individuals with FXS in another study (26). Another important complication that occurs in FXS is epilepsy. Other studies have reported the prevalence of epilepsy in FXS individuals at 10%–20% in boys and 5%–10% of girls (23). In our study, 8% of FXS children had epilepsy, and this percentage was similar to that of previous data. Strabismus was seen in 19% of children with FXS, which equaled that of a previous review (26). A large-scale parental survey presented that 32% of FXS children experienced sleeping difficulties, including difficulty in falling asleep and frequent night awakenings (27). Similarly, 22% of FXS children in our study had sleep-related problems. In a previous review, recurrent otitis media was found to be a common medical problem in FXS. However, none of the 36 children had recurrent otitis media in our study, and only three had adenoid hypertrophy with frequent respiratory infections. Problems of connective tissue were also notable in FXS, which may affect the cardiovascular system, limbs, and spine. We found that 11 FXS children had finger joints that were hyperextensible and five had flat feet. However, we did not find any FXS children with mitral valve prolapse, aortic root dilation, or scoliosis after performing an echocardiogram and spinal x-ray. In addition, we found that a boy (9 years and 11 months) had increased testicular volume (25 ml).

In terms of facial features, we found in our study that a narrow and elongated face and large or prominent ears were found in 92% of patients, followed by a prominent jaw (69%). This result was consistent with the typical facial features of FXS reported previously (23). In addition, we found that some facial features were common, such as a long palpebral fissure (58%), a broad or prominent forehead (53%), and periorbital puffiness (39%). Several studies in Japan, Thailand, and Korea reported the facial features of FXS, and some in these countries also found the elongated facial characteristic of FXS to be less significant than that of the American population (28, 29). In sum, the statistics of facial features are subjective to some extent and the facial features become more obvious and noticeable with age.

This is the first FXS clinical study conducted in China to elucidate the clinical characteristics of this syndrome in detail. The clinical features of FXS children obtained in this study will serve to increase the understanding and diagnosis of FXS. These FXS patients' induced pluripotent stem cells (iPSC) derived neurons could be studied in the future as several phenotypic variations were observed compared with published reports. This could lead to a better understanding of the molecular pathogenesis of the disorder.

The first limitation of this study is possibly related to the recruitment of children with NDD from the Department of Child Health Care of Children's Hospital of Fudan University in Shanghai, China. This indicates that this is a single-center study, the data of which may be biased because of the lack of a full coverage of the incidence of FXS in NDD in China. The second limitation is that we did not include all 42 children with FXS in the statistical data of clinical phenotypes because of incomplete data, and the total count stood at 36. The sample size of our study is not very large, which may not reflect the facial features of a majority of Chinese children. In addition, clinical data were obtained mainly from pediatricians’ assessments and parental questionnaires, and there is the possibility of these questionnaires being a little subjective in nature. In summary, further studies are needed to continue the exercise of large-scale screening of the FMR1 gene in the NDD population by involving multiple medical centers in order to provide more accurate epidemiological data. In addition, pediatricians need to master the clinical characteristics of FXS in practice to promote early diagnosis and interventions.