Identification of a distinct developmental and behavioral profile in children with Dup15q syndrome

Previous research has suggested that individuals with ASD show a distinct skill profile on the VABS-II, with greatest impairment in the socialization skills with relatively spared motor and daily living skills [42, 43]. In our study, children with non-syndromic ASD demonstrated this well-described pattern of relative strengths and weaknesses on the VABS-II. By contrast, children with Dup15q syndrome did not demonstrate evidence of spared motor and daily living skills on the VABS-II. Their adaptive skills remained relatively impaired across domains, with ratings on daily living and motor skills significantly lower than the children in the ASD group. This pattern of “flat” scores across domains is consistent with previous studies of adaptive skills in children with ID [44‐46] and indeed, impairments in adaptive behavior (as opposed to cognition alone) remain central to the diagnostic criteria for ID [47]. Moreover, motor skills were significantly correlated with language, non-verbal cognition, and daily living skills in the Dup15q group, and these relationships were not present in the ASD group. It is interesting to note that these significant differences in motor skills and adaptive function emerged despite participants being matched on the overall developmental quotient. This highlights the distinction between general cognition and adaptive behavior (practical skills employed in everyday life) and suggests that children with Dup15q syndrome may have specific difficulty learning the skills necessary for daily activities.

All of the children with Dup15q syndrome met the ADOS score cutoffs for ASD or autism. However, they demonstrated lower overall ASD severity than the non-syndromic ASD group. Further investigation revealed that children with Dup15q syndrome showed relative strength in the social domain, with significantly better performance on two items: "responsive social smile" and "facial expressions directed to others". These items reflect specific instances of social communication during the administration of the ADOS. To receive a “passing score” (score of 0) on response to social smile, the child must immediately smile back to the adult who is directing a smile towards them. This item is generally administered with the adult directly in front of the child. The item "facial expressions directed to others" is somewhat broader but still involves displaying isolated instances of a skill. To receive a passing score on this item, a child must spontaneously shift gaze to direct a range of facial expressions to the adult for the purpose of communication (e.g., looking to the adult with a surprised expression to express surprise when the jack-in-the-box pops up). Unlike "social smile", this behavior must be initiated by the child but is still comprised of discrete instances. Interestingly, their performance on the "eye contact" item was poor and did not differ from the ASD group. Unlike “social smile” and “facial expressions,” the "eye contact" item is a summary score based on the child’s performance across the entire assessment and is forced to be coded as normal or abnormal, not allowing an intermediate code on this specific item. This may suggest that children with Dup15q syndrome demonstrate clear and specific instances of social interest during the ADOS but that overall, their eye contact remains infrequent and poorly modulated. Taken together, these findings indicate that while children with Dup15q syndrome are likely to meet criteria for an ASD diagnosis, they exhibit relative strengths in social interest and responsiveness through behaviors that occur discretely. As discussed below, one may ask whether an underlying social motivation is present that may be impeded or disrupted by the profound delays in other developmental domains, such as motor skills (such as difficultly in sustained head control and low truncal tone, due to their underlying neuronal pathology). Although our sample size led us to perform statistical comparisons only of items in the reciprocal social interaction subscale, visual inspection of the data suggested that the groups showed similar patterns of performance on the communication and restricted and repetitive behaviors subscales. Future research with a larger sample size will facilitate a full comparison of all ADOS items.

The strong association between motor and social communication skills in children with Dup15q syndrome, and the relative strength in measures of social interest and responsiveness, raises a question of causality vs. common pathways in the developmental phenotype of Dup15q syndrome. It is possible, particularly given the early diagnosis of hypotonia in infants with Dup15q syndrome [17], that early motor delays do limit or even obscure the development of early social communication skills, such as gesturing, eye contact, and even expressive language [48]. For instance, hypotonia and poor head control can undermine an infant’s ability to make adequate and sustained eye contact or to visually inspect his/her social environment, such as faces, which, in turn, can limit the infant’s ability to learn from these critical social cues. This hypothesis finds some support in research of infants at high risk for ASD, with risk conferred by having an older sibling with ASD, in which gross motor delays at age 6 months were associated with a later ASD diagnosis in those children who demonstrated the most severe autism symptoms [49]. An alternate possibility, however, and one articulated in a recent review by Krystal and State [50] is that duplications on 15q11.2-q13.1 have a more global impact on the developing nervous system which, in turn, impairs fundamental processes in brain development resulting in delays in the acquisition of skills across domains, from motor to social communication to language skills to risk for epilepsy. To truly disentangle these two processes, we need more refined measures of domain-specific function in early development, combined with structural and functional imaging that can link behavior to specific neural networks, and prospective studies of infants with the syndrome prior to the onset of delays. The latter study design becomes challenging given the fact that most children with Dup15q syndrome are diagnosed in the context of clinical sequelae of the syndrome, such as epilepsy or a neurodevelopmental disorder.

The subgroup comparisons were limited by sample size, and possible ascertainment bias, but certainly warrant further investigation. While the interstitial participants in this study did show higher average scores across domains than the children with idic(15) duplications, there were no significant differences between groups. These analyses were likely limited by the small sample size of interstitial duplications and, moreover, the fact that two thirds of the children with interstitial duplications in this sample demonstrated more severe cognitive impairment than the average participant in a larger cohort study of interstitial duplications [27]; thus, ascertainment bias may further limit interpretation.

Children with epilepsy had significantly lower cognitive and motor skill scores compared to those children without epilepsy, even when the analysis was confined to those with idic(15). The epilepsy characteristics of this group of four children did represent a range in epilepsy severity consistent with the epilepsy characteristics of children with Dup15q syndrome [29] including history of infantile spasms, onset of generalized epilepsy around puberty, as well as complex partial seizures, all requiring treatment with anti-epileptics. Notably, two of the four children in this cohort developed epilepsy in later childhood, after their developmental delays had emerged. The relationship between cognitive impairment and epilepsy in children with ASD has been well described in large cohort studies and meta-analyses [51, 52], leading to a similar question about causal relationships. While both epilepsy and developmental delay may reflect outcomes from common processes, such as excitation/inhibition imbalance from defects in GABA_A receptor function, seizures in the developing brain may also impact synaptic plasticity and cortical connectivity, which, in turn, results to developmental delays across domains [53, 54]. Further research that carefully tracks cognitive development, behavior, and epilepsy (from electrophysiological characteristics to clinical events) in this population is required to disentangle the effect of epilepsy from the underlying genetic variation in the severity of the developmental disability [55].

The relative strength observed in social interest and responsiveness in the context of impaired motor skills represents an important avenue for intervention. Intervention targeting joint attention and language within a play and engagement-based context has been established as effective for building language and social skills in children with ASD, including children with low IQ and minimal language [56, 57]. Similar intervention strategies with children with Dup15q syndrome could leverage their social interest and facilitate the development of further communication and social abilities. Additionally, intervention that focuses on fostering parent-child interactions may ultimately promote additional opportunities for children to engage in these skills throughout the day, thus increasing their opportunities for learning. Parent-mediated joint engagement-based intervention has been shown to be successful in improving social communication in toddlers with ASD [58]. Daily home routines provide an ideal context in which to promote social communication, while providing opportunities to practice motor and daily living skills. Embedding motor skills practice within an engagement-based social communication intervention has the potential to target the specific deficits observed in children with Dup15q syndrome, while building on their strength in social interest.

Given that children with epilepsy demonstrate greater cognitive impairment, behavioral intervention should be coupled with aggressive and timely treatment of seizures. Further research will be required to determine the effects of specific anti-epileptics and their timing on developmental outcomes in this cohort.

Although this study represents a uniquely detailed clinical characterization of a neurogenetic disorder, the sample size of 13 restricts statistical analyses and detection of significant effects that distinguish Dup15q syndrome, especially with regard to comparisons within the Dup15q group (by duplication type and epilepsy status), and precludes statistical correction for multiple comparisons. Furthermore, given that this was a clinic-referred sample, there exists an ascertainment bias towards children with greater symptom severity. Children in this cohort were diagnosed with Dup15q syndrome because of the sequelae of their genetic variation, including epilepsy, cognitive impairment, or ASD and not through a population screening of children with known ASD or a more general population sample, which would be ideal. Second, the wide range in ability and age of the participants necessitated variability of assessments across participants. Matching of mental age was better for verbal DQ than for non-verbal DQ, representing an inherent challenge in comparing groups with differing developmental profiles.

There are multiple ways that the findings in this study may be extended. An effective approach for other rare disorders has been multi-site studies to boost the sample size and statistical power needed to detect differences within subgroups of Dup15q syndrome. Prospective studies from early development will elucidate the temporal relationship between motor impairments and social communication skills, but these studies remain somewhat challenging given that the neurodevelopmental features of the syndrome usually precede the genetic diagnosis. However, with improved early developmental screening and increasingly widespread genetic testing, we will identify an increasing number of infants with Dup15q syndrome, prior to a formal ASD or ID diagnosis. Such a shift will facilitate prospective, developmentally informed studies. Finally, we must further explore through more refined measures the distinction between social behavior and social motivation in this cohort. The social motivation hypothesis in autism suggests that many children with ASD are less rewarded by social information (for review, see [59]). One could argue that the social communication deficit in Dup15q syndrome is not rooted in the lack of social motivation, rather in an inability (either due to cognitive delays or motor impairment) to sustain social interaction, or impairment in the quality rather than the amount of social interaction. Future studies should focus on social motivation by interrogating reward circuitry through measures of social attention and physiological responses to social stimuli.