Neurobehavioral profile of individuals with pathogenic variants in CHD3

Abstract

Snijders Blok-Campeau syndrome (SNIBCPS), a neurodevelopmental disorder first described in 2018, is caused by heterozygous pathogenic variants in CHD3. Its encoded protein plays a crucial role in the development of the nervous system of embryos. While phenotypic traits have been broadly defined, i.e., global neurodevelopmental delays such as intellectual disabilities and delayed speech acquisition, and physical features such as characteristic facial features and macrocephaly, the phenotypic spectrum has not been further assessed. We present the neurobehavioral profile of 38 individuals with variants in CHD3 and compare it to the ones of autism spectrum disorder (ASD) and Fragile X syndrome (FXS) cohorts. Profound clinical deficits were found in adaptive functioning, communication skills, and sensorimotor functioning in most SNIBCPS participants. Similarities between FXS and SNIBCPS cohorts were unveiled, characterized by diminished levels of global adaptive behavior and adaptive functioning in the social and communication domains. Nevertheless, despite profound challenges in global adaptive behavior in SNIBCPS, we reveal the social domain as showing the highest adaptive levels alongside minimal emotional/behavioral issues within the sample, suggesting relative strengths inherent to SNIBCPS. This study enriches the scarce SNIBCPS literature by delineating the neurobehavioral phenotypic spectrum of SNIBCPS and by innovating comparisons with clinically akin neurodevelopmental disorders.

Introduction

Neurodevelopmental syndromes characterized by global developmental delays and intellectual disability (ID) affect nearly 1.5% of the worldwide population [1]. In recent years, our understanding of the genetics of neurodevelopmental disorders has been accelerated by the investigation of de novo mutations in developmentally important genes. Such mutations are enriched in the genomes of individuals with severe, undiagnosed developmental disorders [2]. Snijders Blok-Campeau syndrome (SNIBCPS, OMIM #618205) is a neurodevelopmental disorder caused by de novo heterozygous pathogenic variants in Chromodomain Helicase DNA-binding 3 (CHD3, OMIM #602120). CHD3 plays a crucial role in the development of the nervous system in the embryo through its encoded protein CHD3, which regulates chromatin structure, gene transcription, and DNA repair, ultimately playing a role in neural cells’ pluripotency and lineage commitment [3–6]. CHD3 regulates the late migration of neurons and upper-layer neuron specification [4, 5]. The laminar localization of neurons and their diverse cytoarchitecture provide a framework for efficient information processing [4, 7]. Consequently, perturbations in processes of cortical migration, generation, differentiation, and wiring of cortical projections can profoundly impact cognitive functions, resulting in intellectual deficits and other cognitive delays [8].

SNIBCPS can lead to impairments in multiple organ systems. Reported physical phenotypic characteristics include macrocephaly, facial dysmorphisms, infantile hypotonia, genital anomalies, joint laxity, and vision abnormalities. Developmental and neuropsychological characteristics have also been found, such as infantile hypotonia, speech-related problems, borderline to severe ID, autistic features, and motor delays [9].

Usually, characterization of neuropsychological profiles is based on comparisons of the affected population with a neurotypical population. For novel rare disorders such as SNIBCPS, comparisons with relatively well-established neurodevelopmental conditions may enlighten such characterization. Given the estimated prevalence of autistic traits, intellectual disability, and speech delays in SNIBCPS, comparing this syndrome to well-established conditions primarily characterized by these traits provides a framework for classification. SNIBCPS shares significant speech delays and intellectual disability with Fragile X syndrome (FXS), a common cause of inherited intellectual disability. Also, contrasting SNIBCPS with autism spectrum disorder (ASD) offers an opportunity for a nuanced examination, aligning SNIBCPS’s documented ASD traits with those of a robust syndrome exhibiting similar characteristics but with minimal or no intellectual disability, specifically in this study’s ASD sample.

Despite the limited global cases (unofficial est. 150), this report aims to describe the neurobehavioral profile of SNIBCPS and to compare it to two clinical groups (FXS and ASD), examining their neurobehavioral profiles across adaptive, socioemotional, and behavioral domains. To date, this study is the first to provide a deeper analysis of the neurobehavioral profile of the SNIBCPS population and to compare it with other clinical populations. This knowledge is of utmost importance for healthcare professionals, individuals affected by the condition, and their families, as it plays a vital role in guiding clinical management, providing accurate counseling, and facilitating prognosis.

Subjects and methods

Participants

Thirty-eight participants with reported deleterious CHD3 pathogenic or likely pathogenic variants were recruited via referral, direct contact with CHU Sainte-Justine medical geneticist Dr. Philippe Campeau, and via family associations. Each participant or legal caregiver was able to complete at least one questionnaire based on an appropriate level of understanding of the English language. Two participants required a German interpreter during the Vineland-3 assessment. As such, analyses for each questionnaire were conducted with variable sample sizes with respect to their chronological age and available norms. We included 49 idiopathic ASD and 46 FXS participants from our lab’s database, using Vineland-3 and ABC-C questionnaire data. The FXS cohort underwent genetic screening and was diagnosed with the full mutation (i.e., presence of more than 200 repetitions of CGG (cytosine-guanine-guanine)). ASD participants were recruited in the context of a clinical trial on arbaclofen. They were all diagnosed with level 1 ASD by clinicians using the DSM-5 and the ADOS-2, and were required to have a level of language competence sufficient to complete modules 3 and 4 of ADOS-2, meaning that they had to be verbally fluent. ASD participants did not undergo genetic screening but had no other diagnosed genetic condition. No exclusion criteria were considered during recruitment of the SNIBCPS cohort due to the rarity of cases. The study was reviewed and approved by the Research Ethics Committee of the CHU Sainte-Justine and was carried out according to the Declaration of Helsinki. Parents were asked to electronically fill out questionnaires via a Research Electronic Data Capture (REDCap) weblink and via e-mail. Procedures were explained in detail before obtaining informed consent from participants or legal caregivers and assent from participants.

Questionnaires

Data for the SNIBCPS cohort were gathered from ten standardized questionnaires. (1) An in-house questionnaire was used to collect demographic information, pathogenic variant, and overall developmental skills. Three global adaptive functioning questionnaires were used: (2) The Ages & Stages Questionnaires, Third Edition (ASQ-3) measures developmental progress determined by cut-off scores on five areas: Communication, Gross Motor, Fine Motor, Problem Solving, and Personal Social [10]. (3) Adaptive Behavior Assessment System-Second edition (ABAS-II) was used to assess adaptive functioning [11]. Three composites were calculated to obtain the adaptive domains: Conceptual, Social, and Practical. A General adaptive composite (GAC) was also calculated. Standardized scores (mean = 100 and SD = 15) were obtained from a chronological age normative sample. Higher scores indicate better adaptive functioning. (4) The Vineland Adaptive Behavior Scale – Third Edition Comprehensive Interview Form (Vineland-3) was conducted by videoconference to estimate the level of support required in daily life. The Vineland-3 is known to be the leading instrument for supporting the diagnosis of intellectual and developmental disabilities [12]. Three composites were calculated to obtain the adaptive domains: Communication, Daily Living Skills, and Socialization. An adaptive behavior composite (ABC) was also calculated. The Vineland-3 and the ABAS-II follow the same standardized method (mean = 100 and SD = 15). Two questionnaires and other pertinent adaptive domains were used to screen and assess sensorimotor functioning: (5) The Developmental Coordination Disorder Questionnaire (DCDQ) was used to screen for Developmental Coordination Disorder (DCD) according to DSM-IV criteria and based on cut-off scores. Lower scores indicate greater indication of DCD [13]. (6) Sensory Profile 2 and Adolescent/Adult Sensory Profile were used to assess sensory processing patterns. Scores from four sensory profiles and processing domains were categorized based on an age-normative sample. Greater or lower scores indicate a processing pattern more or less similar to the majority [14, 15]. Four questionnaires were used to screen social, emotional, and behavioral difficulties: (7) The Aberrant Behavior Checklist for Community (ABC-C; 16). The ABC-C is a questionnaire especially designed for populations with ID, measuring different problematic behaviors, namely irritability, lethargy, stereotypy, hyperactivity, and inappropriate speech. (8) The Social Responsiveness Scale – Second Edition (SRS-II) has been used to assess the presence and severity of autistic traits [17–19]. Norms were based on a normative sample, and T-scores (mean = 50 and SD = 10) were calculated. Higher scores indicate greater autistic symptomatology. (9) The Conners 3rd Edition (Conners 3) or the Conners’ Adult ADHD Rating Scales–Self Report (CAARS – S:L for older participants) was used to assess attention-deficit/hyperactivity disorder (ADHD) and its most common comorbid disorders [20]. T-scores (mean = 50 and SD = 10) were calculated for six domains: Conners-3 Global Index Total (GI), Learning Problems, DSM-V ADHD Inattentive, DSM-V ADHD Hyperactive-Impulsive, DSM-V Conduct Disorder, and DSM-V Oppositional Defiant Disorders. Higher scores indicate more ADHD symptoms. (10) Children Behavior Checklist (CBCL) or Adult Self-Report (ASR for older participants) measure general behavioral and emotional issues, and provide an overview on psychopathology [21–24]. Standardized scores were obtained from chronological age CBCL/ASR normative samples. Higher scores indicate more psychopathology symptoms. Supplementary Table S1 contains information about the questionnaires.

Statistical analyses

Statistical analyses were performed using SPSS Statistics, version 23 (IBM Corp., Armonk, NY, USA) for comparisons between the three clinical cohorts using ABC-C and Vineland-3 data. Data distribution was verified using histograms as well as skewness and kurtosis criteria (values within −1 and 1 were considered acceptable). For each between-group analysis, Z-scores falling beyond 3.29 SD (0.001%) were considered as outliers and were excluded. The significance level was set to 5% (p = 0.05). Between-group comparisons were performed using Kruskal-Wallis H tests. Significant differences were investigated with post-hoc pairwise comparisons using Dunn’s tests with a Bonferroni correction for multiple comparisons. Additional non-parametric covariance analyses controlling for adaptive behavior (Vineland-3 ABC standard scores) were conducted using Quade’s tests [25]. Descriptive analyses were also conducted for the SNIBCPS cohort across all ten questionnaires.

Results

SNIBCPS cohort and other clinical cohorts

The average age of the SNIBCPS cohort was 9.17, with the youngest patient being 2 months old and the oldest being 47.5 years old. For the between-group statistical analyses, only participants falling between the range of 3–20 years old were selected. The FXS cohort’s average age is 12.72 (5–20 years old), and the ASD cohort’s average age is 12.80 (6–17 years old). Age differences were found between the groups (p = 0.011). Table 1 provides more general demographics on the clinical cohorts, and Supplementary Tables S2–S4 provide demographics for the participants included in the between-group analyses and p-values for age differences.

Table 1.

Demographics with comorbidity and medication information of the study populations.

	SNIBCPS	FXS	ASD
N	38	46	49
Males (n, %)	25 (65.79%)	34 (73.91%)	36 (73.47%)
Females (n, %)	13 (34.21%)	12 (26.09%)	13 (26.53%)
Age
Mean ± SD	9.17 ± 9.56	12.72 ± 4.50	12.80 ± 3.30
Range	0.17–47.5	5–20	6–17
Age category
Preschool (0–4 yr 11 mo)	16	0	0
School (5–18 yr)	18	36	49
Adults (18+)	4	10	0
Comorbid diagnoses
ASD	5	18	NA
ID	13	27	5
Epilepsy	4	3	1
ADHD	5	19	25
Cerebral Palsy	3	-	-
Medication
Antipsychotics	3	6	4
Antidepressants	3	13	13
Anxiolytics	1	1	1
Psychostimulants	4	21	21
Anticonvulsants	4	0	1
CBD	0	3	0

Global adaptive functioning

Among the SNIBCPS participants that completed the Vineland-3, two-thirds (67%; 16/24) exhibited profound global impairment in adaptive functioning, and one-third (33%; 8/24) showed borderline global impairment (Fig. 1A). Results show Low Adaptive Level of Daily Living Skills in 71% (17/24) of participants. Similarly, the ABAS-II revealed GAC scores in the Extremely Low range 95% of participants aged between 5 and 21 years old (18/19). The preschool-aged individuals (0–5 y.o.) showed significantly delayed developmental milestones, as all 14 of them fell below the cut-off on at least one developmental category of the ASQ-3.

Fig. 1. Vineland-3 Adaptive Behavior Composite standard scores by age and gender in the SNIBCPS cohort and comparisons of Vineland-3 Adaptive Behavior Composite and Vineland-3 Daily Living Skills standard scores between ASD, FXS and SNIBCPS cohorts.

A ABC adaptive behavior composite, SS standard scores. The population plotted is SNIBCPS. The normative area is 100 ± 15. The dashed line presents the threshold for the clinical range (below 2 SD) interpreted as the Low Adaptive Level. Higher scores indicate better functioning. B ABC adaptive behavior composite, DLS daily living skills. Whiskers represent minimum and maximum values and the horizontal line in the box represents the median. Higher scores indicate better functioning.

A difference was found in the Vineland-3 ABC scores between the groups (H = 19.877, p < 0.001), with a mean rank score of 42 for SNIBCPS, 34.76 for FXS, and 62.74 for ASD. Pairwise comparisons revealed higher adaptive functioning in the ASD cohort compared to FXS (Z = 27.987, p < 0.001) and SNIBCPS (Z = 20.745, p = 0.007), and no difference between FXS and SNIBCPS (p = 1.000). The same tendency was found for the Daily Living Skills domain (H = 14.854, p < 0.001), with a mean rank score of 37.84 for SNIBCPS, 39.18 for FXS, and 60.78 for ASD. Higher scores were found in ASD when compared to FXS (Z = 21.604, p = 0.003) and to SNIBCPS (Z = 22.939, p = 0.009). The FXS and SNIBCPS groups were not significantly different (p = 1.000). One ASD participant had an extreme score (Z = 4.2) and was excluded from the between-group comparisons. Results are shown in Fig. 1B.

To better understand developmental stages in SNIBCPS compared to ASD and FXS, the analyses were also conducted among participants aged 10 years old and below (SNIBCPS: N = 15; ASD: N = 13; FXS: N = 11). A significant difference was found on ABC scores (H = 6.650, p = 0.036), with higher adaptive functioning in ASD than SNIBCPS (Z = −10.610, p = 0.042), but no difference was found on Daily Living Skills (H = 2.858, p = 0.240).

Communication level

64% (9/14) of the preschool SNIBCPS participants were clinically delayed on the Communication domain of the ASQ-3. 71% (17/24) showed a Low Adaptive level and 29% (7/24) showed a Moderate Adaptive Level on the Vineland-3 Communication domain. No SNIBCPS participant showed normotypical communication skills (0/24 within the norms).

A significant between-group difference was found in the Vineland-3 Communication scores (H = 28.049, p < 0.001), with a mean rank score of 36.68 for SNIBCPS, 32.79 for FXS, and 65.28 for level 1 ASD. Post-hoc pairwise comparisons showed higher communication skills in the ASD group than in the FXS (Z = 32.485, p < 0.001) and SNIBCPS (Z = 26.591, p = 0.002) groups. The FXS and SNIBCPS groups were not significantly different (p = 1.000; Fig. 2A).

Fig. 2. Comparisons of Vineland-3 Communication standard scores and ABC-C Inappropriate Speech Z-scores between ASD, FXS and SNIBCPS cohorts.

A SS standard scores. Whiskers represent minimum and maximum values, and the horizontal line in the box represents the median. Higher scores indicate better functioning. B ABC-C aberrant behavior checklist for community. Whiskers represent minimum and maximum values, and the horizontal line in the box represents the median. Higher scores indicate higher severity.

There was also a difference in ABC-C inappropriate speech scores between the groups (H = 6.612, p = 0.037), with a mean rank inappropriate speech score of 53.67 for SNIBCPS, 71.52 for FXS, and 55.38 for level 1 ASD. Post-hoc pairwise comparisons with the Bonferroni correction did not reveal between-group differences. However, without the Bonferroni correction, inappropriate speech was higher in the FXS group than in the SNIBCPS (Z = 17.856, p = 0.035) and level 1 ASD (Z = 16.145, p = 0.024; Fig. 2B) groups. Differences were confirmed when controlling for adaptive behavior using Quade’s test (F = 3.742, p = 0.027), with higher inappropriate speech in FXS than SNIBCPS (t = 2.527, p = 0.013) and ASD (t = 2.152, p = 0.034), and no difference between SNIBCPS and ASD (p = 0.373).

Sensorimotor functioning

The Gross Motor domain was the ASQ-3 domain in which the preschool participants showed the highest rate of developmental delay, with 93% (13/14) falling below the cut-off score. The Fine Motor domain was also impaired in 64% (9/14). Similarly, DCDQ results revealing clinically significant impairments in the SNIBCPS sample suggested suspected DCD in all participants (17/17). Likewise, 72% (13/18) showed a Low Adaptive Level in the Vineland-3 Motor domain.

The Sensory Profile results revealed extremely low registration of the sensory cues in 46% (12/26) of the SNIBCPS group and normotypical sensation avoidance in 62% (16/26). Body position processing was the subdomain showing the most impairment in participants aged below 14 years old, with 64% (14/22) falling in the clinical range.

Screening for the social, emotional, and behavioral domains

Socialization level

71% (10/14) of the preschool SNIBCPS participants showed clinically significant delays in the Personal Social domain of the ASQ-3, and 57% (8/14) in the Social domain of the ABAS-II. Despite such results, the Socialization domain of the Vineland-3 had smallest number of participants falling in the clinical range. In fact, 58% (14/24) of the participants showed Low Adaptive Levels of Socialization skills, 33% (8/24) showed Moderate Adaptive Levels, and 8% (2/24) showed Adequate Adaptive Levels. When compared to global adaptive scores (ABC), 79% (19/24) of participants had a higher Socialization domain score (mean score difference = 5.13; SD = 6.46). Within the Socialization domain, the Interpersonal Relationship Subdomain was the one with the highest number of participants falling within the norm (21%; 5/24).

A significant difference was found when comparing the Vineland-3 Socialization scores between the groups (H = 13.765, p = 0.001), with a mean rank score of 45.32 for SNIBCPS, 36.76 for FXS, and 57.73 for ASD. The ASD group had higher social skills than the FXS group (Z = 20.97, p = 0.004), but the SNIBCPS group did not significantly differ from FXS (p = 0.874) nor from ASD (p = 0.306; Fig. 3).

Fig. 3. Comparisons of Vineland-3 Socialization standard scores between ASD, FXS and SNIBCPS cohorts.

SS standard scores. Whiskers represent minimum and maximum values, and the horizontal line in the box represents the median. Higher scores indicate better functioning.

Screening for ASD traits

Regarding the SRS-2, among the four ranges of deficiencies in social interaction, 8/31 (26%) participants fell within the Normal range, 7/31 (23%) within the Mild range, 10/31 (32%) within the Moderate range, and 6/31 (19%) within the Severe range. The preschool population seemed to exhibit less social impairment. In fact, 4 of the 6 preschool participants (67%) fell within the Normal range, and none showed severe deficiencies in social interaction.

Screening for ADHD traits

Fifty six percent (10/18) of the SNIBCPS cohort showed clinical levels of behavioral, academic, and emotional issues associated with ADHD as reflected by the Global Index Total (GI) scores of the Conners-3 questionnaire (Fig. 4). DSM-V Inattentive and Hyperactive-Impulsive symptomatology of ADHD were observed as significant areas of concern in 50% (9/18) and 61% (11/18) of the sample, respectively. The academic area was also indicated as a significant area of concern in 69% (11/16) of the participants aged 18 and below. Most participants scored within the norm in the DSM-V Conduct Disorder (88%; 14/16) and DSM-V Oppositional Defiant Disorders (81%; 13/16).

Fig. 4. Conners-3 Global Index Total T-Scores by age and gender in the SNIBCPS cohort.

GI global index. The population plotted is SNIBCPS. The normative area is 50 ± 10. The lower dashed line presents the threshold for the borderline range (“Elevated” and “High Average” area of concern; between 1 and 2 SD). The upper dashed line presents the threshold for the clinical range (“Very Elevated” area of concern; above 2 SD). Higher scores indicate higher symptomatology.

Emotional functioning and Psychopathology

65% (22/34) of the SNIBCPS group had CBCL/ASR Total scores within the norm, and 26% (9/34) were in the clinical range.

No significant difference was found between the SNIBCPS, ASD, and FXS groups for the ABC-C composite score (H = 1.074, p = 0.585), hyperactivity subscale (H = 3.466, p = 0.177), irritability subscale (H = 1.300, p = 0.522), lethargy subscale (H = 0.373, p = 0.830), and stereotypy subscale (H = 2.426, p = 0.297). When controlling for adaptive behavior, the same tendency was found: no significant difference on the composite score (F = 0.430, p = 0.652), nor on the hyperactivity (F = 0.676, p = 0.511), irritability (F = 0.488, p = 0.616), lethargy (F = 0.367, p = 0.694), and stereotypy (F = 0.472, p = 0.625) subscales. We further compared behavioral traits using two-way ANOVAs to look for interactions between group and gender on ABC-C subscales. Since it is already established in FXS that behavioral problems are worse in males, we only included the SNIBCPS and FXS cohorts [26]. We found an interaction on lethargy scores (F = 4.812, p = 0.32), with females showing higher scores than males in the ASD group (F = 6.191, p = 0.017), but not in the SNIBCPS group (F = 0.815, p = 0.375).

Discussion

Our study aimed to define the neurobehavioral phenotype of the SNIBCPS population and provide insights into the strengths and challenges of this developmental disorder through comparisons with similar clinical populations. In a new cohort of 38 SNIBCPS participants, we found severe impairments in adaptive functioning, communication skills, and gross and fine motor abilities. We highlight relatively high socialization levels and low prevalence of emotional and behavioral problems in our sample, which is proposed as relative strengths of this syndrome. Phenotypic similarities and differences are reported between the FXS, level 1 ASD, and SNIBCPS populations.

Global adaptive functioning

Overall, the SNIBCPS population shows significant impairments in adaptive functioning as reflected by the Vineland-3, a widely used instrument for supporting the diagnosis of intellectual and developmental disabilities [12]. While formal diagnoses of ID and developmental delay were not feasible within the scope of our study, the results shed light on the challenges faced by this population in terms of independence and managing daily demands. Additionally, their level of support required for daily functioning is comparable with that one of FXS.

Communication

The present study adds to the existing body of research by providing robust evidence that supports the presence of clinically impaired speech abilities in the SNIBCPS population. Notably, none of the participants exhibited normotypical communication skills, thus highlighting the substantial nature of speech delays within this cohort. The presence of speech delays in this cohort aligns with the consistent diagnosis of Childhood Apraxia of Speech (CAS) in the initial cohort, which led to the discovery of de novo mutations in CHD3, providing insights into the genetic basis of the SNIBCPS. CAS is a rare neurodevelopmental disorder characterized by impaired acquisition of synchronized sequences of mouth and facial movements essential for fluent speech.

Overall, the FXS and SNIBCPS groups showed comparable communication skills, but FXS had a tendency towards higher inappropriateness of speech than SNIBCPS and level 1 ASD, which was confirmed when controlling for adaptive behavior.

Sensorimotor functioning

This study refines previous findings by elucidating motor-based speech delays and hypotonia as fundamental characteristics of SNIBCPS. The preschool cohort exhibits notable impairments in gross and fine motor skills, underscoring early developmental motor deficits. Furthermore, consistent with additional investigations, a high prevalence of suspected DCD is evident across the entire age range studied (5–15 years), with most individuals displaying clinical adaptive motor levels. Complementary evidence from the Sensory Profile questionnaire substantiates hypotonia as a distinctive trait in the SNIBCPS population. Notably, the subdomain focused on body position processing, comprising proprioception-related items, demonstrates the most pronounced impairments. Low scores on such items seem to drive low scores on the Registration domain, which states that this population misses sensory input at a higher rate than neurotypicals. Such difficulties in processing and detection of sensory input seem to be associated with disruptions of proprioceptive senses known to signal body shape, body position, and movement, and muscle force [27]. Disruptions in body signaling have also been reported in previous SNIBCPS cohorts, with several individuals experiencing pain insensitivity. Existing literature postulates compelling hypotheses highlighting the interplay between proprioception and pain, including the sensory-motor incongruity of pain and the gate control theory [28–30]. Further investigation is warranted to unravel the intricate connections among proprioception, sensory processing, and pain within the SNIBCPS population. Intriguingly, our study reveals a paradoxical finding wherein most participants miss a substantial amount of sensory input while avoiding it at normal levels. The items within the sensation avoidance domain primarily focus on the social-emotional aspects of sensory processing, aligning with our study’s observation of mainly normotypical socioemotional functioning in the SNIBCPS population.

Social behavior

The considerable heterogeneity in research findings regarding social skills in SNIBCPS has prompted a comprehensive exploration of this aspect. Sparse information on social skills stems from two previous cohorts, which only reported autistic-like traits in approximately one-third of the population based on clinical observations [9, 31]. In the current study, we systematically investigated social behavior associated with ASD and general socialization abilities. The general socialization skills of the SNIBCPS cohort were found to be comparable to the ASD and FXS populations. Despite most of the cohort showing low social functioning, this domain seems to be a relative strength as it surpasses their global adaptive skills, and it stands out from the frequency of clinical-level deficits found in other life domains. Their socialization abilities, therefore, seem to transcend the challenges posed by reduced motor abilities and significant speech delays.

Emotional functioning and psychopathology

No other study has investigated socioemotional and behavioral issues with a psychopathological approach (i.e., DSM-related scales) within the SNIBCPS population. Approximately half of the population exhibited clinical levels on ADHD scales. Additionally, learning difficulties were revealed as an area of concern within most of the population, which aligns with previous evidence of intellectual disabilities (ID) and current findings of severely impaired global adaptive behavior.

This study suggests that the SNIBCPS population does not exhibit apparent emotional and behavioral problems, which represents a noteworthy strength within this population. Most of the participants do not exhibit actively disruptive behaviors associated with conduct or oppositional defiant disorders. Although such findings align with the relatively high levels of socialization found in our study, literature supports the opposite. In fact, rates of psychopathology are much higher for individuals with ID compared to the general population [32]. The prevalence of psychiatric disturbances among young individuals with ID is found to be three to four times higher compared to children of average intelligence, as 40 to 50% suffer from a psychiatric disorder [33, 34]. Our study reveals evidence for lower disruptive behaviors in the SNIBCPS group, contradicting current literature and revealing a unique characteristic of the syndrome. Further investigation regarding this population’s psychiatric disorders is necessary to support the low level of psychopathology as a core feature of the syndrome. Such evidence would alleviate negative impacts associated with psychopathology, such as high levels of family stress and parental mental health problems [34–36].

Limitations

The retrospective design of the study introduced several limitations. Firstly, considering the rarity of SNIBCPS, no exclusion criteria were included, so the study lacked control over medication use and behavioral and educational interventions, which may have underestimated or overestimated neurobehavioral ratings. Antipsychotics, antidepressants, anxiolytics, psychostimulants, and anticonvulsants can positively or negatively alter cognition, mood, behavior, and sensorimotor functioning [37–41]. Notably, psychostimulant use may have underestimated ADHD symptomatology rates. Secondly, comparisons with the ASD cohort must be interpreted with care, as this study’s ASD participants represent a specific subgroup of the heterogeneous ASD spectrum, since all ASD participants had level 1 ASD and were verbally fluent. Moreover, the validity of standardized questionnaires can be highly influenced by different factors. Studies have shown that results tend to underestimate internalizing (i.e., broadly referring to symptoms of anxiety, depression, and somatic symptoms), lethargy, and irritability symptoms in children with language impairment relative to verbally fluent children, and either over- or underestimate behavior problems in children under 5 on the ABC-C [42, 43]. Diagnosis of neurodevelopmental disorders can also bias validity, which explains the use of clinical norms and comparisons between different clinical populations in this study. Parent perception bias could have also influenced results. Future research should integrate expert assessments, including formal evaluations conducted by medical professionals, neuropsychologists, and therapists. It is also important to note that between-group differences on the Vineland-3 are based on comparisons of the groups’ raw scores to neurotypical norms, whereas differences on the ABC-C are based on comparisons of the groups’ raw scores to all participants. The differences found regarding inappropriate speech while controlling for adaptive behavior should also be interpreted with care, as the adaptive behavior covariate (Vineland-3 ABC) includes social and communication skills.

Additionally, future studies with larger sample sizes and additional measures (e.g., genetic analysis, IQ/neuropsychological assessments, MRI, and EEG) would permit a more complete phenotypic profile of SNIBCPS, which in turn could be used to develop the sparse and unclear literature of the genotype-phenotype association within this population. Two families with inherited variants in CHD3 participated in the study, which highlights the importance to investigate inheritance of CHD3 variants and their phenotypic expressivity. Further efforts of analysing variability within the population could provide valuable information for better personalized care, as generalization of traits can aid diagnosis and follow-up, but can also impede personalized recommendations and interventions

Author contributions

AI designed the study, contributed to data collection of the SNIBCPS population, data analysis and wrote the manuscript; EML contributed to data analysis and to the writing of the manuscript; MPL contributed to data collection of the ASD and FXS populations, data analysis and provided feedback on the report; PMC and SL contributed to the study design, supervision of the study and provided feedback on the manuscript; all the other authors contributed to data collection of the ASD and FXS populations.

Funding

Funding by the Fonds de Recherche du Québec en Santé (SL) enabled this study.

Data availability

Data from the FXS and SNIBCPS groups is attached as supplementary material. Data from the ASD group can be found at https://www.braincode.ca under the project POND.

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

Consent for participation in this study was obtained via the individual themselves or the individuals’ legal guardian(s), and the study was approved by the Research Ethics Committee of the CHU Sainte-Justine (#2016-962) and was carried out according to the Declaration of Helsinki.

Supplementary information

The online version contains supplementary material available at 10.1038/s41431-025-01926-6.