Early impact of X‐ and Y‐chromosome variations (XXX, XXY, XYY) on social communication and social emotional development in 1–2‐year‐old children

Abstract Sex chromosome trisomies (SCTs) are characterized by an extra X‐ or Y‐chromosome (XXX, XXY, XYY). The present study aims to investigate early signs of social communication and social emotional development in very young children with SCT. Thirty‐four children with SCT (aged 12–24 months) were included in this study, as well as 31 age‐matched controls. Social communication was measured with structured behavior observations according to the Early Social Communication Scales, and social emotional developmental level with the Bayley Social Emotional parental questionnaire. Recruitment and assessment took place in the Netherlands and in the United States. On average, 12–24‐month old children with SCT showed difficulties with early social communication, more so in responding to others as compared to initiating social communications. During social interactions, children with SCT made less frequent eye contact, compared to controls. Also, difficulties in acquiring social emotional milestones were found in 1‐year old children with SCT, with 44% of the children having social emotional vulnerabilities in the borderline or extremely low range, compared to typically developing children. In this cohort, no significant predictive effects of karyotype‐subtype (XXX, XXY, XYY) were found. Already from a very early age, SCT can be associated with increased risk for vulnerabilities in adaptive social functioning. These findings suggest that SCT impact the maturation of the social brain already from an early age, and stress the importance of early monitoring and (preventive) support early social development in young children with SCT.

compared to the typical karyotype of 46,XX in girls and 46,XY in boys.
Prevalence estimates of SCT vary from 1:650 to 1:1000 (Boyd et al., 2011). Knowledge about the impact of SCT on the neurocognitive and neurobehavioral phenotype is growing, and it is known that a disproportional high percentage of genes on the Xchromosome play a role in human cognition and brain development (Zechner et al., 2001). There is evidence suggesting that an extra X-or Y-chromosome convergently impact the maturation of a distributed and interactive network of cortical and subcortical components underlying adaptive social cognitive functioning, often described as the "social brain" (Hong & Reiss, 2014;Raznahan et al., 2016). Anatomical brain structures related to social functioning that seem to be affected by SCT include the posterior insula, the anterior cingulate, the medial prefrontal cortex, the superior temporal sulcus, and the orbitofrontal cortex (Raznahan et al., 2016). Social brain alterations in SCT fit with the increased risk for deficiencies in the processing of social information and with difficulties in social emotional development, like problems in understanding social emotional information, and in showing social adaptive behavior during interactions with others, that have been observed in school-aged children, adolescents, and adults with SCT. Although the phenotype of SCT is variable with some not having marked symptoms while others are more affected, studies that evaluated several domains of social adaptive functioning in individuals with SCT found social vulnerabilities including shyness, social immaturity, difficulties in forming adequate interpersonal relationships, increased levels of social anxiety and social impulsivity, and impairments in underlying social cognitive mechanisms (see for reviews : Freilinger et al., 2018;Tartaglia, Cordeiro, et al., 2010;Ross et al., 2012;. The severity of social behavioral and social cognitive impairments in school-aged children, adolescents, and adults with SCT is illustrated by an increased level of clinical diagnoses of Autism Spectrum Disorder (ASD). Average percentages of ASD classifications across studies vary from 15% (range 10.8%-20%) in individuals with 47,XXX; 18% (range 10%-27%) in individuals with 47,XXY; to 30% (range 19%-43%) in individuals with 47,XYY (Van Rijn, 2019).
In order to identify early markers of "at risk" social development and related social behavioral vulnerabilities later in life, it is important to study the impact of SCT on the social behavioral phenotype very early in life. The first study on social behavioral characteristics in children with SCT aged 1-5 years found social behavioral problems already at this age . Since social difficulties have a major impact on a broad range of areas of development, including language acquisition, school readiness, peer acceptance, and risk and resilience for developmental psychopathology, early identification of compromised social development in young children with SCT can help understand developmental outcome and finding targets for preventive intervention (Rao et al., 2008).
Early social development is marked by the growing ability to coordinate eye contact and to engage in reciprocal social emotional interactions (Soto-Icaza et al., 2015). Already in the first hours of life, infants prefer to look at faces that engage in eye contact (Farroni et al., 2002). Making eye contact with others is a basic biological mechanism, essential for social communication, as the visual information in eyes are important sources of information used to understand communicative goals and emotional states of others (Senju & Johnson, 2009).
Thereafter, young children develop specific communication skills underlying socially adaptive behavior, in which the following elements are of particular importance: following eye gaze and conventional gestures of others in order to achieve shared attention to an object or event (i.e., joint attention), the ability to communicate own beliefs and desires and to react to the desires of others, and being part of reciprocal social interactions. These specific communication skills involve the use of eye contact and serve to hold and coordinate attention between interactive social partners (Mundy, 2003(Mundy, /2013. In naturalistic daily life settings, social communication skills are linked to significant accomplishments of broad social emotional functioning: they shape the capacity of the child to engage with others, to comprehend emotional expressions of others, and to elaborate upon a range of feelings in social interactions (Bayley, 2006).
As the first years of life is a period in which the social brain network rapidly matures and specializes, this period serves as a key period to acquire social emotional developmental milestones (Grossmann & Johnson, 2007). Therefore, it is especially important to focus on social behavioral abilities during that stage of early childhood as a means of identifying early markers of an "at risk" social development. The current study aims to provide in that by investigating the impact of SCT on the early communicative and social emotional phenotype in a group of children, aged 12-24 months.
The present study may have specific benefit to the clinical care of individuals with SCT. As SCT is being increasingly identified during pregnancy with recent technical advances of noninvasive prenatal screening (i.e., the noninvasive prenatal screening test [NIPT]), the population of infants with prenatal diagnosis of SCT is rapidly growing . The study of very young children with SCT supports the identification of children with SCT at risk for neurobehavioral difficulties and psychopathology later on in life and will give insight in potential targets for monitoring, early prevention and intervention.
Taken together, the present study aims to investigate the impact of SCT on structured observations of early social communication and parental report of social emotional development in 1-year-old children. In addition to these main research questions, predictive effects of karyotype-subtype (XXX vs. XXY vs. XYY) on early communicative skills and social emotional skills were studied. Based on the relevance of the X-and Y-chromosome for the development of neural networks supportive of adaptive social functioning, we hypothesized that 1-year old children with SCT might show less well developed social emotional and communicative skills, compared to a control sample.

| Participants
The present study is part of a larger ongoing project (the TRIXY Early Childhood Study -Leiden, the Netherlands), which includes children with SCT and nonclinical controls. A group of 34 children with SCT aged 12-24 months was included in this study (M age = 1.39, SD = 0.36), as well as 31 age-matched controls (13 boys; M age = 1.53, SD = 0.28). Mean age did not significantly differ between groups (t (63) = 1.69, p = 0.096). The SCT group consisted of 6 girls with 47,XXX (17.6%), 20 boys with 47,XXY (58.8%), and 8 boys with 47,XYY (23.5%). Gender distribution differed between the SCT and control group (χ 2 (1) = 11.37, p = 0.001). Recruitment and assessment took place on two research sites: the Trisomy of the X and Y chromosomes (TRIXY) Expert Center in the Netherlands, and the eXtraordinary Kids Clinic in Developmental Pediatrics at Children's Hospital Colorado in the United States. Children in the SCT group were recruited with the help of clinical genetics and pediatrics departments (from the Netherlands, Dutch speaking parts of Belgium and the United States), as well as through patient-advocacy groups and social media postings. One girl with XXX, five boys with XXY, and three boys with XYY were recruited and assessed in the Netherlands. Five girls with XXX, 15 boys with XXY, and 5 boys with XYY were recruited and assessed in the United States. Karyotype distribution did not differ between the two research sites (χ 2 (2) = 0.82, p = 0.664). Thirty-two children (94.1%) were diagnosed prenatally, and two children postnatally (5.9%; one girl with XXX, one boy with XXY). Eleven out of 20 boys with 47,XXY had received testosterone treatment (55%).
The diagnosis of SCT was defined by trisomy in at least 80% of the cells, which was confirmed by standard karyotyping. For the SCT group, recruitment strategy was assessed, and three subgroups were identified: (1) "active prospective follow-up," which included families who were actively followed after prenatal diagnosis (67.6% of the SCT group); (2) "information seeking parents," which included families who were actively looking for more information about SCT without having specific concerns about the behavior of their child (26.5% of the SCT group); and (3) "clinically referred cases," which included families seeking professional help based on specific concerns about their child's development (5.9% of the SCT group). The control group was recruited from the western part of the Netherlands and approached with information brochures about the study. All participants were Dutch or English speaking, had normal or corrected-to-normal vision, and did not have a history of traumatic brain injury or hearing loss.
For ethical reasons, children in the control group were not subjected to genetic screening, as these children were meant to be a representation of the general population. As the prevalence of SCT is $1 in 1000, the risk of having one or more children with SCT in the control group was considered minimal and acceptable.
Parental education and age of the primary caregiver were assessed. Parental education was assessed according to the criteria of Hollingshead (Hollingshead, 1975). Scores of this scale include: 0 (no formal education), 1 (less than seventh grade), 2 (junior high school), 3 (partial high school), 4 (high school graduate), 5 (partial college or specialized training), 6 (standard college/university graduation), and 7 (graduate/professional training). Ninety-eight percent of all parents indicated that their child has a second caregiver. If two parents were available, level of education was averaged over both parents. A Pearson χ 2 test was performed to investigate possible differences in parental education distribution between the SCT and control group.  Table 1 for a description of the nonverbal communicative behaviors within the six subscales of early social communication.

| Social-emotional development: Bayley SE
The Bayley Social Emotional Questionnaire (Bayley SE;Bayley, 2006) consists of 21 or 24 items (dependent on the age of the child), and measures the acquisition of functional social-emotional milestones in naturalistic settings that broadly represent social-emotional patterns and developmental accomplishments. The Bayley SE questionnaire was administrated in the original English version (Bayley, 2006), and the translated Dutch version (Bayley SE-NL;Van Baar et al., 2014).
The items of the Bayley SE assess the attainment of age-related milestones of the child, namely the ability to engage and use various emotions, expressions and experiences, as well as the comprehension of a range of social and emotional signals and to understand and react to a feelings of others with words, gestures, or imitations. Examples of items are: "Shows you that he or she understands your actions or gestures by making an appropriate gesture in return (e.g., make funny face back at you, looks at something you point to, stops doing something when you shake your head and use a firm voice to say 'No!!' or smiles and does more of something when you nod with a big smile and say 'Yes!')" and "Uses many consecutive actions in a back-andforth way to show you what he or she wants or to have fun with you (e.g., smiles, reaches out for a hug, and, when you hug, takes your hat, puts it on his or her head, and smiles proudly OR takes your hand, leads you to the refrigerator, tugs on the handle, and, after you open it, points to something he or she likes, such as food, a bottle of juice, or milk)." The primary caregiver rates his/her child on a 5-point Likert scale ranging from 1 "none of the time" to 5 "all of the time." Raw scores were used to compare the children with SCT and the controls. Based on the guidelines of the Bayley 3rd Edition Manual, the total raw scores of the Bayley SE were converted to a composite score, ranging from 55 to 145 with a mean of 100 and standard deviation (SD) of 15 points in the norm population. The composite scores were labeled as being in the average range (composite score > 90), the "Borderline/ monitoring" range (composite score 70-89), and the "Extremely low/at risk" range (composite score < 70; Weiss et al., 2010). The Bayley SE has high internal consistency and test-retest reliability (Weiss et al., 2010), and the original validation study demonstrated that the Bayley SE distinguishes significantly between clinical groups including children with genetic syndromes or developmental disorders (Bayley, 2006).   Figure 1 for the setup of the assessment room). Verbal interactions were kept to a minimum during the ESCS. The 20-min structured assessment was videotaped, with full face view of the child and profile view of the experimenter. The parent questionnaire was completed by the primary caregiver of the child, either in Dutch or in English.

Data quality
The ESCS was successfully completed by 63 children (two children are not able to complete the task). Although all children received at least 14 out of 18 trials of the active wind-up toys and hand-operated toys, some trials of four children are not administrated (e.g., because the child was crying) or are excluded due to technical aspects (e.g., experimenter obscured camera angle). Therefore, for these children the mean value of the coded social behaviors on other trials was used in place of these missing data (=mean substitution; Kang, 2013).
Inter-rater reliability was measured based on a subsample of 10 participants and showed an intraclass correlation coefficient (ICC) of 0.84-0.96 (for the ESCS domains collapsed together) which is considered excellent reliability (Cicchetti & Sparrow, 1981).   Table 2.

Early social communication and age
Eye contact: SCT versus controls

| Social emotional development: SCT versus controls
Within the SCT group, social emotional functioning was positively cor-

| Association between early social communication and overall social emotional development
To explore the associations between early social communication behaviors and daily life social-emotional development within the SCT group, Pearson's correlations were calculated. Significant positive correlations were found between three early social communicative domains and social emotional development (initiating behavioral requests, r = 0.453; responding to joint attention, r = 0.514; and responding to social interactions, r = 0.346). See Table 3 for r and p values for all variables.

| The role of specific karyotype (XXX, XXY, XYY)
In order to investigate whether specific karyotype-subtype (XXX, XXY, XYY) was predictive of social emotional abilities and early social communication, linear regressions with dummy coding were carried out with social functioning as dependent variables. No significant predictive effects of karyotype-subtype were found. See Table 4 for exact F and p values.

| The role of recruitment strategy
Within the SCT group we tested whether recruitment strategy was predictive of social emotion abilities and early social communication.
We used a linear regression with dummy coding for the three recruit-

| DISCUSSION
The aim of the current study is to investigate the early impact of SCT No karyotype-specific differences (XXX, XXY, XYY) were found.
We used systematic behavior observations to explore social communication behaviors in 1-year-old children with SCT when they were actually exposed to social interactions in a structured play situation.  (Tartaglia, Cordeiro, et al., 2010). However, it should be noted that these studies utilized different parents report measures and had differences in sample ascertainment. Follow-up studies are needed to explore the developing pathways of motivation for social communication and underlying social cognitive and motivational mechanisms in young children with SCT from the first years of life into childhood. This is especially important in order to find targets of early/preventive support and intervention, based on the idea that early intact motivation for social contact should be preserved over the course of development.
The impact of SCT on social communication and social emotional development early in life as found in this study has implications for our understanding of brain-behavior pathways leading to these difficulties. The biological predisposition of SCT allows us to study early social development of a homogeneous group of children, which may serve as "high risk" group when it comes to neurobehavioral social development. Raznahan et al. (2016) found that in a group of participants with sex chromosome aneuploidies aged 5-25 years old, the Xand Y-chromosome congruently impact the functionality of cortical areas that support adaptive socio-emotional functioning and social communication (e.g., medial prefrontal cortex, anterior cingulate, and superior temporal sulcus). The difficulties we detect on the social domain in 12-24-month-old children with SCT may reflect an impact of SCT on the maturation of an integrated social brain network already from infancy on, as these social difficulties are behavioral expressions of an impaired maturation of cortical brain networks that underlie social development (Johnson et al., 2005). The finding that there are no differences between karyotype-subtypes (XXX, XXY,

XYY) on social emotional and communicative difficulties is in line with
a convergent influence of the extra X-and Y-chromosome on social brain maturation and associated social behavioral functioning. However, it was shown that the social behavioral profile in boys with 47,XYY is more vulnerable as compared to girls and boys with an extra X chromosome, which is illustrated by a higher risk for ASD (Cordeiro et al., 2012;Ross et al., 2012;Tartaglia et al., 2017). This more pronounced vulnerability in the XYY group was not found in the current study. However, our sample size in the XYY group was small. These findings call for more research into the nature of early social cognitive  (Mundy, 2017) and therefore also serves as an important target to monitor during early development in young children with SCT.
Deficits in early social communication may not only be associated with impairments in subsequent social development, but also with an altered language acquisition, speech delay and behavioral problems during early development (Pickard & Ingersoll, 2015;Schietecatte et al., 2012). It is found in nonclinical samples that early social skills are associated with school readiness and early school success (Ziv, 2013 The current study has both strengths and limitations. Strengths include the structured observations of social interactions, which allow us to measure observed social communicative behaviors of 1-year-old children with SCT in interaction with a social partner. Second, the study sample consisted mainly (94%) of children that were prospectively studied after a prenatal diagnosis of SCT which suggest that our findings are highly representative for the group of diagnosed children.
In this study, social outcomes were largely not dependent on recruitment strategy (i.e., prospective follow-up group, information seeking parents group, or clinically referred cases group), which suggests that on average our findings are representative for this group of diagnosed children. However, it should be taken into account that the sample sizes were small and vary between the recruitment groups. Similar, although no predictive effect of karyotype-subtype was found on social outcomes, it is important to consider the small sample sizes and the small number of participants in the XXX and XYY group which calls for a careful interpretation of the nonsignificant differences between karyotype-subtype. Additional studies with larger sample sizes are needed to investigate the individual X-and Y-chromosome influences on social functioning very early in life of children with SCT.
Next, parents of children with SCT are aware of their child's diagnoses of SCT. As a consequence, they may rate the items of the Bayley SE questionnaire differently compared to parents of children in the control group, as parents of children with SCT are more or less aware what to expect in terms of their child's developmental outcome. This may have biased the outcome of parent reports.
The findings of this study deserve further investigation of the longitudinal effects of early difficulties in social competence on social cognitive and behavioral outcomes and related neurodevelopmental psychopathology in children with SCT. These longitudinal effects will be further investigated in this population with prospective follow-up.
As it was beyond the scope of this study to investigate the role of testosterone treatment in boys with 47,XXY, future studies with an applicable design (i.e., Randomized Controlled Trials) are needed to explore the influence of these variables in association with social behavioral functioning in young children with SCT.

| CONCLUSION
In summary, the results of this study show that already very early in development, that is, at the age of 1 year, children with SCT have vul-

ACKNOWLEDGMENTS
The authors warmly thank the families that participated in our study, and the research assistants and students in Colorado and Leiden (the Netherlands) for their help with data collection and processing. Special thanks to the students who were involved in coding the ESCS

CONFLICT OF INTEREST
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported. design; interpretation of the data; final approval of the manuscript.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.