Abstract
Objective
Neurodevelopmental concerns in males with sex chromosome aneuploidy (SCA) (XXY/Klinefelter syndrome, XYY, and XXYY) include many symptoms seen in autism spectrum disorder (ASD), such as speech-language impairment, verbal cognitive deficits, and social difficulties. Studies using standardized ASD assessment measures in SCA are few, and rates of ASD diagnosis in SCA using DSM-5 criteria for ASD have not been explored. We aimed to: (1) evaluate ASD characteristics in research cohorts of males with SCA using standardized assessments under DSM-IV compared to DSM-5 criteria, and (2) analyze whether type of SCA, prenatal vs. postnatal diagnosis, cognitive abilities, adaptive functioning, and socioeconomic status are associated with ASD diagnosis.
Methods
Participants from two sites with XXY/KS (n=20), XYY (n=57) and XXYY (n=21) were evaluated using protocols including medical history, cognitive testing, adaptive functioning, the Social Communication Questionnaire, Social Responsiveness Scale, Autism Diagnostic Observations Schedule (ADOS), the Autism Diagnostic Interview-Revised (ADI-R), and DSM criteria for ASD. Clinical impressions of ASD diagnostic category using the ADOS and DSM-IV criteria were compared to the ADOS-2 and DSM-5 criteria. T-tests were used to compare cognitive, adaptive, SES and prenatal vs. postnatal diagnoses between ASD versus no ASD groups determined by DSM-5 criteria.
Results
ASD rates in these research cohorts using DSM-IV criteria were 5% in XXY/KS, 33% in XYY, and 43% in XXYY. Rates of 10% in XXY/KS, 38% in XYY, and 52% in XXYY using ADOS-2 and DSM-5 criteria were not statistically higher. In both XYY and XXYY, the ASD group had lower verbal IQ and adaptive functioning compared to the group without ASD. Many children with SCA who did not meet full ASD criteria showed some deficits in social communication or social interaction skills.
Conclusion
The rate of ASD in children with SCA in this study was higher than expected compared to the general population. Males with Y chromosome aneuploidy (XYY and XXYY) were 4.8 times more likely to have a diagnosis of ASD than the XXY/KS group, and 20 times more likely than males in the general population based on the 2010 Centers for Disease Control (CDC) estimate of 1 in 42 males. ASD is an important consideration when evaluating social difficulties for children with SCA. Studies of males with SCA and Y-chromosome genes may provide insight into idiopathic ASD and male predominance in ASD.
Keywords: XXY, Klinefelter syndrome, XYY, XXYY, autism spectrum disorder, Y Chromosome
Introduction
Autism spectrum disorder (ASD) is a behaviorally defined neurodevelopmental disorder characterized by varying degrees of deficits in domains of social communication, reciprocal social interaction, and restricted, repetitive patterns of behavior. With the publication of the DSM-5 in May 2013, there was a significant change in the diagnostic criteria of ASD to reflect the commonalities and interrelationships of these core symptoms across a wide spectrum of severities, and all cases are now classified under a single diagnosis of ASD rather than the distinct disorders in the DSM-IV of Autistic Disorder, Asperger Syndrome, and Pervasive Developmental Disorder – Not Otherwise Specified (PDD-NOS). Further, the DSM-5 diagnostic criteria also emphasize the importance of neurobiological and genetic factors that may underlie the ASD symptoms, and the presence of any genetic disorders or other medical conditions known to be associated with ASD are now part of the “specifiers” to be included the diagnostic description. This has led to an increased interest in genetic disorders with behavioral phenotypes that include increased rates of ASD and other social and/or communication impairments.
Genetic studies in children with ASD identify an etiologic diagnosis in approximately 10 to up to 40% of cases, for example chromosomal duplications and deletions, fragile X syndrome, and mutations in specific genes such as MECP2 or PTEN mutations [1]. In many studies evaluating genetic etiologies in cohorts of children with ASD, cases of sex chromosome aneuploidy (SCA) conditions including XXY/Klinefelter Syndrome (XXY/KS), XYY and XXYY have been identified [2-4]. There have also been many case reports and case series of autism spectrum disorder in males with XXY/KS, XYY, and XXYY [5-8].
XXY/KS, XYY, and XXYY are the three most common forms of sex chromosome aneuploidy in males. XXY/KS occurs in approximately 1/650 male births [9, 10], and is characterized by medical features of tall stature, hypogonadism (microorchidism and testosterone deficiency), and decreased fertility. Longitudinal studies of birth cohorts have characterized the large spectrum of the psychological and behavioral phenotype in males with XXY/KS. Early speech and language delays are present in up to 70% of children with XXY/KS, and cognitive abilities are typically in the average range but often show language-based learning disabilities and a lower verbal IQ [11-13].Behavioral characteristics including shyness, social immaturity, anxiety, and other social difficulties are commonly described [14].
More recent studies specifically evaluating social functioning in XXY/KS have identified an increased rate of deficits that overlap with those seen in autism spectrum disorder. For example, Van Rijn et al. have described deficits in social cognitive functioning and increased autism traits in adult men with XXY/KS ascertained through support groups [15]. Bishop et al. reported a previous clinical diagnosis of ASD in 11% of 19 boys with prenatal diagnoses of XXY/KS [16], and, in 2012 a similar rate of 12% positive scores on an autism screening measure (Social Communication Questionnaire; SCQ) in a cohort of 34 pediatric patients with XXY/KS.[17] A similar rate of 11.6% has since been reported by Samango-Sprouse et al. in a separate cohort of boys with a prenatal diagnosis of XXY/KS using a different screening measure (Gilliam Autism Rating Scale score of “very likely.”)[18]. In 2009, Bruining et al. reported on a study of 51 males age 6-20 years with XXY/KS in the Netherlands who were evaluated using the Autism Diagnostic Interview – Revised (ADI-R), and in this cohort 27% were identified as having possible ASD [19]. To our knowledge, the Bruining et al. study has been the only study to move beyond ASD screening questionnaires and directly evaluate ASD characteristics in XXY/KS utilizing a gold-standard instrument.
XYY syndrome occurs in approximately 1/1000 male births, and shares features of tall stature, early speech and motor delays, and a high prevalence of language-based learning disabilities with XXY/KS [20]. However, males with XYY do not typically have hypogonadism or fertility problems as occurs in XXY/KS. Prospective studies on birth cohorts identified average cognitive abilities, and behavioral traits including hyperactivity and low frustration tolerance. Poor social relatedness, ADHD, Pervasive Developmental Disorder – Not Otherwise Specified and autistic behaviors have been reported in other behavioral studies and case reports in XYY syndrome, with some cohorts reporting rates of previous clinical diagnoses of ASD ranging from 20 - 30% [21, 22]. Administration of the SCQ screening questionnaire to a cohort of 22 pediatric patients with XYY led to positive results in 50% [17]. However, a study including direct diagnostic evaluation of children with XYY using standardized autism assessment tools has not previously been reported.
XXYY syndrome occurs in approximately 1/18,000 male births. Originally described as a variant of Klinefelter syndrome (XXY) due to shared features of tall stature and hypogonadism [23], it is now recognized that XXYY is associated with more significant cognitive deficits and behavioral disorders, as well as a higher rate of associated medical problems such as seizures or congenital defects [24]. The low prevalence of XXYY has prevented prospective studies of birth cohorts like in XXY/KS and XYY. Overall cognitive abilities in XXYY syndrome are typically in the borderline range, however verbal cognition and language abilities are usually significantly lower than visual-spatial abilities. Previous studies have described behavioral characteristics including impulsivity, difficulty with social relationships, anxiety, and autistic behaviors [25-27]. In a large review of 95 individuals with XXYY by our group, 23% had been previously diagnosed with ASD upon presentation to the study [24].
Taken together, previous studies have shown that the behavioral phenotype of males with SCA includes many features that overlap with those also associated with ASD, including speech/language disorders, verbal cognitive deficits, social deficits, and other autistic behaviors. Moreover, rates of ASD characteristics in SCA are higher than expected compared to the general population when screening measures for ASD are utilized. However, with the exception of one study in XXY/KS, previous studies have not directly evaluated and compared children with SCA using standardized ASD diagnostic measures. The screening measures used in previous studies were developed for use in consideration of the DSM-IV diagnostic criteria, and diagnostic criteria for ASD changed with the publication of the DSM-5 in 2013 [28]. Further, the most commonly used standardized assessment tool for ASD called the Autism Diagnostic Observation Schedule (ADOS) was updated and released with revised scoring algorithms as the ADOS-2 in 2012 in anticipation of the changes in the DSM-5 criteria [29]. These changes in both the diagnostic criteria and assessment measures raise the question of how the new criteria have affected diagnostic rates of ASD in males with SCA.
In this project, we aimed to describe two studies that directly evaluated diagnostic rates and characteristics of ASD in research cohorts of males with SCA using standardized assessments under the DSM-IV criteria. We aimed to compare diagnostic rates in the same cohorts using updated DSM-5 diagnostic criteria and ADOS-2 assessment algorithms. We hypothesized that rates of ASD would be: (1) higher in males with SCA compared to the general population, (2) similar in the two trisomy conditions (XXY/KS and XYY), but (3) higher in XXYY compared to the trisomies given more severe neurodevelopmental involvement. Further, we hypothesized that more males with SCA would meet DSM-5 criteria for ASD compared to DSM-IV. We also aimed to analyze factors that could be associated with DSM-5 ASD diagnosis (including SCA subgroup, timing of SCA diagnosis (prenatal versus postnatal), cognitive abilities, adaptive functioning, and socioeconomic status).
Methods
Study Sites
The described studies were independently initiated and conducted at University of California (UC) Davis MIND Institute in Sacramento, California and Thomas Jefferson University (TJU) in Philadelphia, Pennsylvania, with ongoing analysis of the UC Davis cohort subsequently conducted at University of Colorado due to change in institution of the investigator (NT). Upon discovery of similar research efforts and protocols during data collection, investigators collaborated to combine results from both studies to increase sample size. Minor differences in study protocols between sites are described below and in Figure 1.
Figure 1.
Components of ASD assessment and reclassification from DSM-IV to DSM-5 criteria at each site.
Subjects
IRB approval was obtained at all institutions, parents or primary guardians signed consent forms, and participants signed assent forms as appropriate to age and developmental level. In both studies, participants were recruited from hospital-based outpatient clinics and from national SCA support organizations for participation. To minimize recruitment bias toward more affected individuals or those with concerns related to autism, recruitment materials indicated that the studies were evaluating features of health and development in SCA, and did not include specific wording that autism-related disorders were being evaluated. Inclusion required confirmation of genetic testing results showing a karyotype of nonmosaic XXY/KS, XYY, or XXYY. In the UC Davis/Colorado study, 63 male children and young adults between the ages of 3 and 25 with XXY/KS (n=20), XYY (n=22), or XXYY (n=21) participated. One male with XYY was found to also have another genetic disorder (distal 22q11.2 deletion syndrome) and was excluded from the study [30], leaving a total of 62 participants. At TJU, 41 males with XYY participated. Five participants enrolled in both studies and data from the UC Davis/Colorado was used in analysis, leaving a total of 36 participants in the TJU study.
Research Assessments
After the participants and/or their parents provided informed consent, participants were evaluated through protocols consisting of two main components: (1) psychological assessment, which included cognitive testing and autism assessments by research-reliable evaluators blinded to SCA group, and (2) medical history, developmental-behavioral history, and physical examination for all participants.
Psychological Assessments
UC Davis/Colorado study
In the UC Davis/Colorado study, cognitive levels were estimated using the Mullen Scales of Early Learning (MSEL) for children 3-5 years of age, and the Wechsler Abbreviated Scale of Intelligence (WASI) for children 6 and older. The MSEL is a standardized developmental assessment that includes subscales of development (visual reception, fine motor, expressive language, receptive language), and allows for a summary score to be calculated [31]. The WASI is an abbreviated cognitive assessment that includes four subscales, and provides estimates of Verbal IQ (VIQ), Performance IQ (PIQ), and Full Scale IQ (FSIQ) [32]. Results of the WASI VIQ, PIQ, and FSIQ have been shown to have good correlation with the more extended Wechsler Intelligence Scale for Children – Fourth Edition (WISC-IV) (r = .85, .78, and .86, respectively) [32].
Adaptive functioning was assessed using the Vineland Adaptive Behavior Scales, 2nd Edition Interview Edition. The scale is norm referenced, and standardized on a national sample, and provides standardized scores of adaptive functioning in domains of communication, daily living skills, and social skills [33].
Socioeconomic Status (SES) was assessed using the Hollingshead Two-Factor Model of parental education and occupation [34].
ASD-Specific Assessments
Assessment for ASD under the DSM-IV criteria was conducted by an interdisciplinary team including Developmental-Behavioral Pediatrics, Child Psychology and a Licensed Clinical Social Worker. All participants were evaluated using the Social Communication Questionnaire (SCQ), the Social Responsiveness Scale (SRS), and the Autism Diagnostic Observation Schedule (ADOS). The SCQ is a 40-item parent-report questionnaire designed to screen for the diagnostic features of ASD based on the ADI-R in children age 4 and above [35]. A score 15 or greater has a high sensitivity and specificity for ASD, and indicates the need for further ASD assessment. The SRS is a parent-report questionnaire for age 4 and above that evaluates social behaviors across multiple domains as they occur in natural social settings to identify the presence and degree of social deficits and determine whether symptoms are consistent with those seen in ASD [36]. The SRS yields five subscale T-scores in the domains of Social Awareness, Social Cognition, Social Communication, Social Motivation and Autistic Mannerisms, and an SRS Total score. If screening on the SCQ was positive (score of 15 or greater), if the ADOS results were in the ASD range, or if a previous concern for an autism spectrum disorder was raised by an educational, medical, or psychological professional, the Autism Diagnostic Interview-Revised (ADI-R) was also administered.
The ADOS is a semi-structured interaction session using developmentally appropriate social and play-based interaction and standard activities in a 40-60 minute session designed to allow for the observation of behaviors relevant to the diagnosis of autism spectrum disorder. Different modules are selected based on participant age and language level, and all 4 modules were utilized in this study. The ADOS was validated across a wide range of ages and severity levels in autism with psychometric properties showing good interrater reliability [37]. The original ADOS scoring algorithms were utilized at the time of autism assessments for all participants in the UC Davis study. In the ADOS algorithms classification of Autistic Disorder or ASD required meeting a cutoff score in the two domains of Communication and Reciprocal Social Interaction.
The ADOS-2 was released in 2012 in anticipation of changes to ASD diagnostic criteria with publication of the DSM-5 [29]. The ADOS-2 utilizes the same semi-structured, play-based interaction protocol as the original ADOS, however there are revised scoring algorithms which produce an overall score comprised of the Social Affect domain (a combination of communication and reciprocal social interaction items) and the Restricted, Repetitive Behavior score (which was not included in the original ADOS scoring algorithms) [38, 39]. The ADOS-2 also introduced a new clinical severity score (CSS) which allows for the severity of ASD symptoms to be classified and compared between individuals and tracked over time [38, 40]. The CSS scoring system takes into account age and language-level (ADOS-2 Module administered), and scores are then classified into severity levels as No ASD (0-3), Non-Autism ASD (4-5), and Autism (6-10). All ADOS assessments were re-scored using the ADOS-2 algorithms in order to determine if classification of ASD changed utilizing the revised algorithms. Further, a clinical severity score was calculated for each child per the ADOS-2 scoring in order to determine and compare severity of ASD symptoms (Figure 2).
Figure 2.
ADOS-2 severity comparison scores for XXY (No ASD n=18, ASD n=2), XYY (No ASD n=35; ASD n=22), XXYY (No ASD n= 10; ASD n=11). Mean scores in the No ASD group are in the Non-ASD range as expected, however still slightly elevated especially for XYY and XXYY.
The ADI-R is a semi-structured standardized parent interview developed to assess the presence and severity of autism behaviors throughout childhood [41]. Classification of autism requires meeting a cutoff score in three domains of Qualitative Abnormalities in Reciprocal Social Interaction, Qualitative Abnormalities in Communication, and Restricted, Repetitive, and Stereotyped Patterns of Behavior. The ADI-R has been carefully validated across a wide range of ages and severity levels. Because of the time requirement of the ADI-R, only participants with a positive SCQ screen, ADOS scores in the ASD range, or individuals with a previous concern or diagnosis of ASD completed the ADI-R in the UC Davis study.
TJU study
In the TJU study, participants were 4-14 years of age, and cognitive assessment was conducted using the Differential Abilities Scale – 2nd Edition (DAS-2) [42]. The DAS-2 yields an overall score (General Conceptual Ability; GCA), as well as verbal, nonverbal, and spatial composite scores.
The composite scores on the DAS-2 are strongly correlated (r = .71 to .84) to the corresponding WISC-IV domains, i.e. verbal and VIQ scores, nonverbal and PIQ, and GCA and FSIQ scores. SES was measured using the same Hollingshead 2-Factor Model [34]. Autism assessment at TJU included neurodevelopmental history, SCQ, and SRS for all participants. Standardized autism evaluations at TJU included the ADI-R and/or ADOS (ADI-R + ADOS n=15, ADI-R only n=12, ADOS only n=9).
At both sites, the study teams first evaluated results of all measures in the assessment battery and records available for each participant and determined assignment by team consensus into one of three groups: no ASD, Pervasive Developmental Disorder - Not Otherwise Specified (PDD-NOS), or Autistic Disorder (AUT). No individuals met criteria for Asperger Syndrome since early language delays were present in all children who met criteria for ASD. Protocols were then reviewed and rescored by study teams using both the ADOS-2 algorithms and the revised DSM-5 criteria, and then reclassified into one of two diagnostic groups: ASD and no ASD. See Figure 1 for a diagram of the components of ASD assessment battery and reclassification at each site. Table 2 includes results of these analyses. The clinical severity score was also calculated for all participants using the ADOS-2 scoring system.
Table 2. Comparison of ASD rates in study participants with XXY/KS, XYY, and XXYY.
XXY/KS | XYY | XXYY | ||||
---|---|---|---|---|---|---|
DSM-IV | DSM-5 | DSM-IV | DSM-5 | DSM-IV | DSM-5 | |
UC Davis/Colorado | ||||||
No ASD | 19/20 (95%) | 18/20 (90%) | 14/21 (67%) | 12/21 (57%) | 12/21 (57%) | 10/21 (48%) |
ASD | PDD 1/20 (5%) | ASD 2/20 (10%) | PDD 6/21 (29%) | ASD 9/21 (43%) | PDD 6/21 (14%) | ASD 11/21 (52%) |
AUT 0/20 (0%) | AUT 1/21 (4.7%) | AUT 3/21 (25%) | ||||
TOTAL 1/20 (5%) | TOTAL 7/21 (33%) | TOTAL 9/21 (43%) | ||||
TJU | ||||||
No ASD | NA | 24/36 (67%) | 23/36 (64%) | NA | ||
ASD | NA | PDD 5/36 (14%) | ASD 13/36 (36%) | NA | ||
AUT 7/36 (19%) | ||||||
TOTAL 12/36 (33%) | ||||||
COMBINED SITESTOTAL ASD* | 1/20 (5%) | 2/20 (10%) | 19/57 (33%) | 22/57 (38%) | 9/21 (43%) | 11/21 (52%) |
PDD = Pervasive Developmental Disorder – Not Otherwise Specified, AUT = Autistic Disorder,
ASD = Autism Spectrum Disorder, NA=Not applicable.
For Combined Sites Total ASD Rate (DSM-5): X2 8.79, p=0.012, post-hoc results showed XYY and XXYY > XXY/KS.
Medical/Developmental History and Physical Examination
In both sites, a medical and developmental-behavioral history and physical examination was performed on each subject to characterize developmental milestones, neurodevelopmental history, and to identify other factors that may influence neurodevelopment such as prematurity or other chronic medical conditions. Genetic testing results were reviewed and age at SCA diagnosis was recorded in order to evaluate for differences between those ascertained in the prenatal versus postnatal period. Family history of ASD and other neurodevelopmental disorders were recorded.
Data Analysis
Before testing hypotheses, univariate and bivariate range and logic checks were performed on all variables, using both numerical and graphical summaries. Descriptive statistics were calculated for each SCA group at each site for age, IQ, SES, and psychological assessment results. In the UC Davis study, one-way analysis of variance (ANOVA) were conducted to assess for significant differences between the three SCA groups in age, IQ and adaptive domains and SES. For those with significance of p < .05, post hoc Tukey or Games-Howell analyses were conducted to determine which pairwise differences were statistically significant (Table 1). Results in the XYY groups at the two study locations were compared by independent t-tests (age, IQ subscales, SES, SCQ), and Fisher's Exact Test (ASD rates; prenatal versus postnatal diagnosis) (Table 1). The rates of ASD in the three SCA groups were compared by Chi Square with a significant p-value equal or less than 0.05 (Table 2). Data from the XYY groups from the two studies were pooled, and independent t-tests were used to compare age, SES, cognitive subscales, adaptive skills and timing of ascertainment in the XYY and XXYY groups with ASD and without ASD as determined by DSM-5 criteria (Table 3).
Table 1.
Description of Participants – Demographics and Psychological Assessments.
XXY/KS | XYYa | XXYY | Statistical comparison | |
---|---|---|---|---|
| ||||
UC Davis/Colorado | ||||
n | 20 | 21 | 21 | |
Mean Age (SD) | 9.11 (4.13) | 9.44 (4.60) | 12.47 (5.77) | F (2,59) = 2.89, p=0.064 |
Age Range | 3-17 | 3-21 | 3-22 | |
Mean SES | 53.4 | 50.8 (9.2) | 51.6 | F (2,59) = 0.20, p=0.812 |
Race (% Caucasian) | 95 | 90.5 | 95.2 | |
SCA Diagnosis: | ||||
Prenatal | 12 | 8 | 0 | XXY/KS vs. XYY: |
Postnatal | 8 | 13 | 21 | X2 (1) = 1.19, p=0.28b |
IQ (WASI), n | 15 | 16 | 17 | |
Verbal – M (SD) | 102.7 (11.3) | 90.3 (15.2) | 79.3 (11.8) | F (2,45) = 14.62, p<0.01c |
Verbal range | 76-118 | 56-113 | 64-98 | |
Performance – M (SD) | 105.9 (11.0) | 102.2 (17.4) | 91.7 (12.4) | F (2,45) = 4.57, p=0.015d |
Performance range | 89-129 | 73-136 | 65-108 | |
Full Scale – M (SD) | 104.7 (10.6) | 96.4 (15.4) | 83.2 (10.8) | F (2,45) = 12.43, p<0.01d |
Full Scale range | 81-125 | 66-127 | 61-98 | |
Early Learning | ||||
Composite (MSEL), n | 5 | 5 | 4 | |
ELC – M (SD) | 100.4 (5.7) | 85.2 (15.0) | 77.0 (24.5) | F (2,11) = 2.50, p=0.127 |
ELC range | 91-106 | 68-103 | 49-103 | |
Adaptive (Vineland), n | 20 | 21 | 21 | |
ABC – M (SD) | 89.7 (10.4) | 83.6 (16.4) | 74.4 (13.9) | F (2,59) = 6.37, p<0.01e |
ABC range | 67-107 | 59-119 | 50-101 | |
| ||||
Thomas Jefferson Univ | ||||
n | -- | 36 | -- | |
Mean Age (SD) | -- | 9.6 (3.9) | ||
Age Range | 4-14.9 | -- | ||
Mean SES (SD) | -- | 50.4 (10.2) | -- | |
Race (% Caucasian) | -- | 80% | -- | |
SCA Diagnosis: | ||||
Prenatal | -- | 14 | -- | |
Postnatal | -- | 22 | -- | |
IQ (DAS-2), n | -- | 36 | -- | |
Verbal – M (SD) | -- | 87.2 (15.3) | -- | |
Verbal range | 53-115 | |||
Nonverbal – M (SD) | -- | 93.5 (17.7) | -- | |
Nonverbal range | 56-134 | |||
GCA – M (SD) | -- | 89.2 (14.8) | -- | |
GCA range | 58-124 |
There were no significant differences between sites in age (p=0.82), SES (p=0.94), VIQ (p=0.55), PIQ (p=0.13), FSIQ/GCA (p=0.16), or proportion with prenatal diagnosis (p=0.99).
Z-test of proportions (Bonferroni control for multiple comparisons used, significant relationships are p<.0167): XXY/KS vs. XYY (z=1.44, p=1.51), XXY/KS vs. XXYY (z=5.48, p=0.000), XYY vs. XXYY (z=3.60, p=0.000)
Post-hoc TukeyXXY/KS > XYY > XXYY
Post-hoc Tukey XXY/KS, XYY > XXYY
Post-hoc Games-Howell XXY/KS vs. XYY ns (p=0.35), XXY/KS > XXYY (p<0.01), XYY vs. XXYY ns (p=0.13)
Table 3. Comparison of ASD versus No ASD in XYY and XXYY based on DSM-5 results.
XYY | XXYY | |||||
---|---|---|---|---|---|---|
Mean | ASD (n=22) | No ASD (n=35) | p | ASD (n=11) | No ASD (n=10) | p |
Age | 9.16 | 9.88 | 0.50 | 13.7 | 11.5 | 0.38 |
SES | 52.1 | 49.4 | 0.35 | 52.3 | 50.1 | 0.42 |
Verbal IQ | 84.6 | 93.7 | 0.03* | 68.9 | 83.3 | 0.03* |
Performance IQ | 91.0 | 99.5 | 0.10 | 86.5 | 94.6 | 0.21 |
Full Scale IQ | 87.2 | 94.0 | 0.11 | 76.0 | 86.3 | 0.09 |
Adaptive Functioning | 71.0 | 91.0 | <0.01* | 65.6 | 81.0 | <0.01* |
Ascertainment Pre=prenatal Post=Postnatal | Pre:7/22 Post:15/22 | Pre: 15/35 Post: 20/35 | 0.44 | Post:11/11 | Post:10/10 |
Results
Subjects
A total of 62 participants were included in data analysis (20 XXY/KS, 21 XYY, and 21 XXYY) in the UC Davis/Colorado study, and 36 participants with XYY from the TJU study. Demographics, cognitive subscales, and timing of ascertainment (prenatal versus postnatal) are reported in Table 1. There were no significant differences between the XYY groups from the two study sites by age, SES, and timing of diagnosis.
In the UC Davis/Colorado study, there were no overall significant differences in age between the three groups (F (2,59)=2.89, p=0.064). The rates of prenatal compared to postnatal diagnosis were not significantly different between XXY/KS and XYY; all subjects in the XXYY group were diagnosed after birth. There were no significant differences in SES or ethnicity between the three SCA groups. The majority of participants were Caucasian.
Results of cognitive assessment using the WASI in participants age 6 or over from UC Davis/Colorado showed a mean full-scale IQ of 104.7 (SD10.6) in the XXY/KS group, which was higher than the mean FSIQ in the XYY group of 96.4 (SD15.4), although this difference did not reach significance (p=0.16). The Verbal IQ was significantly lower in the XYY group compared to XXY/KS (p=0.03), whereas the Performance IQ was not (p=0.76). As expected due to added gene dosage effects of the tetrasomy condition of XXYY, cognitive and adaptive scores in the XXYY group were significantly lower compared to both XXY/KS and XYY (for FSIQ and VIQ, p < .001; for PIQ, p=0.01). In all 3 groups, the mean adaptive score was lower than the measured cognitive scores (XXY/KS, 16.6 points; XYY, 15.2 points; XXYY, 8.2 points).
Cognitive assessment in the TJU study was conducted using the DAS-II, a measure that has strong concurrent validity with Weschler scales (r = .71 to .84). Mean General Conceptual Abilities score in the XYY group was 89.2 (SD 14.8). There were not significant differences Verbal, Nonverbal, or Full Scale cognitive scores in the XYY groups between sites.
Autism Assessment
Results of the autism evaluation in the three SCA subgroups from both study locations are shown in Table 2, which compares the diagnostic classification results of all study participants when utilizing the original assessment battery (DSM-IV/ADOS) compared to the reclassification using DSM-5/ADOS-2 criteria applied to each case (see Figure 1). In the UC Davis/Colorado study, there was a significant difference in the overall rate of ASD between the three SCA conditions using both the DSM-IV (χ2 7.93, p = 0.016) and DSM-5 comparisons (χ2 8.79, p=0.012), with post-hoc analysis showing the difference due to XYY and XXYY >XXY/KS. Using the original DSM-IV results, only one participant (5%) of the XXY/KS group met criteria for ASD, compared to 33% (7/21) in the XYY group and 43% (9/21) in the XXYY group. After reclassification using the ADOS-2 and DSM-5 criteria, these rates increased slightly to 10% for XXY/KS, 43% for XYY, and 52% for XXYY. The TJU study showed the same results in their separate XYY cohort, with 33% (12/36) meeting DSM-IV criteria and 36% (13/36) meeting DSM-5 criteria for ASD. When combining data from both studies, males with Y chromosome aneuploidy (XYY and XXYY) were 4.8 times more likely to have a diagnosis of ASD than the group with an extra X (XXY/KS), and 20 times more likely to have a diagnosis of ASD than males in the general population based on the 2014 CDC studies estimating a rate of ASD in 1 in 42 males. [CDC, 2014]
When we specifically compared the ADOS to the ADOS-2 scoring results in the 86 participants who received this measure, 75 (87.2%) maintained their original result of No ASD, ASD, or autism, and 78 (90.7%) maintained their broader classification of either ASD or No ASD. There were 11 participants whose ADOS classification changed between the ADOS and ADOS-2 algorithms. Seven participants changed from a diagnosis of No ASD to ASD due to inclusion of the stereotyped behaviors / restricted interests domain (1 XXY/KS, 4 XYY, 2 XXYY), 3 remained on the autism spectrum however increased from a severity of ASD to AUT due to inclusion of the stereotyped behaviors / restricted interests domain (2 XYY, 1 XXYY), and 1 patient with XYY changed from an ADOS classification of ASD to No ASD on the ADOS-2 due to elimination of items A6 (using of another person's body as a tool) and B4 (integration of eye contact with words/vocalizations during social interactions) scores from the Module 1 algorithm. Overall, these results suggest that the ADOS-2 may classify slightly more children with SCA in the ASD range compared to the original ADOS algorithm (mainly because of the inclusion of stereotyped behaviors / restricted interests in the total score), however the differences in positive ASD rates comparing the two versions did not reach statistical significance.
It is important to note that the overall ASD results presented in Table 2 are based on consensus of the study team after consideration of the ADOS scores in conjunction with other available measures including the SCQ, ADI-R, SRS, and application of DSM-IV and DSM-5 criteria to behaviors. Thus, for some participants a change in ADOS diagnostic category between the ADOS and ADOS-2 did not change the overall consensus diagnostic impression of no ASD versus to ASD. In the participants who changed diagnostic classification from no ASD to ASD, applying the revised DSM-5 criteria to the participant's behavioral history was also important in supporting the change in diagnosis, especially in older patients where the DSM-5 criteria better explains manifestations of symptoms in adolescent to adult age groups, as well as due to the inclusion of sensory sensitivities as a component of the phenotype which are common in SCA. Overall, the percentage of the study group with a final classification of ASD increased slightly from 29.6% to 35.7% with the updated diagnostic tools, however this difference was not statistically significant (p=0.44).
Next, we examined ADOS-2 comparison scores between the three SCA groups as shown in Figure 2. Overall, mean autism symptom severity in boys with XYY and XXYY was higher compared to XXY/KS in both the no ASD and ASD groups. (No ASD: F (2,55)=4.20, p=0.020; ASD: F (2,48)=2.90, p=0.046). Scores in the No ASD group are expectedly lower compared to the ASD group, however, it is important to note slightly elevated mean scores in the no ASD groups as well. Within this No ASD group, 40% fell within the ASD or AUT range on one domain of the ADOS-2, however overall scores or severity of symptoms were not supportive of a diagnosis.
To better understand the factors related to a diagnosis of ASD in the XYY and XXYY groups, we compared age, SES, prenatal vs. postnatal diagnosis, cognitive factors (VIQ, PIQ, FSIQ), and adaptive scores between individuals with and without ASD as shown in Table 3. Since there were not significant differences between the XYY groups at the two study locations, data from the two studies was pooled for the XYY analysis. XXY/KS was not included due to small sample size with ASD. Results showed that in both the XYY and XXYY groups, the group with ASD had significantly lower verbal IQ (XYY 84.6 vs. 93.7, p=0.03 and XXYY 68.9 vs. 83.3, p=0.03) and adaptive functioning scores (XYY 71.0 vs 91.0, p=0.009 and XXYY 65.6 vs. 81, p=0.008) compared to the group without ASD. There were no significant differences in age, SES, performance IQ, or full scale IQ between the XYY groups with and without ASD. There was a lower proportion of ASD diagnoses in the prenatally diagnosed group (31%) compared to the postnatal diagnosis group (42%), however this difference was not statistically significant (p=0.44). Since the entire XXYY group (ASD and no ASD) was diagnosed in the postnatal period, comparisons between features in pre- versus postnatal diagnoses could not be made in this group.
Discussion
This report describes study results from 2 sites evaluating autism spectrum disorder in males with sex chromosome aneuploidy (SCA) using standardized autism assessment measures. In summary, results show that in two separate research cohorts of children with XYY, there were similar rates of ASD (approximately one third of the group), and the SCA groups with Y chromosome aneuploidy (XYY and XXYY) were more likely to meet DSM-5 criteria for ASD compared to males with XXY/KS. The presence of ASD was associated with lower verbal cognitive abilities, and lower adaptive functioning scores. Compared to use of DSM-IV criteria and ADOS algorithms, use of DSM-5 criteria and revised ADOS-2 algorithms resulted in identification of slightly more individuals with SCA who met criteria for ASD; this difference was not statistically significant.
Previous studies have reported the rate of ASD in a sample of males with XYY to be 19% based on clinical history [21], or 20% based on educational classification [16]. Our study builds on these results by directly assessing patients utilizing standardized autism measures, with very similar results between the two XYY cohorts and an overall rate of 38% for the combined group. Our rates may be slightly higher than previous studies due to identification of more ASD symptoms through the direct assessment, and also because of the change in ASD criteria and increased ASD awareness and diagnostic rates in the general population over the past 20 years.
This study and the Geerts et al. report (2003) compare the rates of ASD in prenatal versus postnally ascertained cases of XYY [21], as this comparison is important in consideration of ascertainment bias and relevant to prenatal genetic counseling. While our results show a slightly higher percentage of ASD in those with postnatal ascertainment (32% prenatal compared to 43% postnatal), this difference was not statistically significant. The Bishop et al. study describes a rate of ASD of 20% in a prenatal XYY sample [16]. Together, these results support that an elevated risk for ASD exists even within a prenatally diagnosed group with XYY. It is important to note that these reported rates of 20-32% are likely somewhat higher than would be expected from a true prenatal birth cohort, since there is still ascertainment bias (study participation bias) represented in the small samples evaluated for the research.
In comparison to previous studies of autism features in males with XXY/Klinefelter syndrome, Bruining et al. (2009) evaluated 51 children with XXY/KS age 6-19 using the ADI-R parent interview, with results showing that 27% met criteria for ASD [19]. In our study only 3 participants with XXY/KS with high SCQ scores received the ADI-R, and 1 of the 3 (33%) had a positive score. This rate of 33% is much higher than our final diagnoses of 5-10% obtained following a more comprehensive assessment including the ADOS/ADOS-2, DSM criteria, and team consensus. These discrepancies suggest that a more comprehensive assessment that includes direct evaluation of social interactions with the child using a measure like the ADOS may be important to differentiate between core social deficits of ASD versus many of the other neurodevelopmental differences associated with SCA such as language disorders or anxiety that may contribute to the ASD-like behaviors reported by parents on the ADI-R.
Another important finding in our study is that autism severity scores were elevated even in the SCA groups without ASD, and over one-third of SCA individuals without ASD met criteria in one domain of the ADOS. These results indicate that children with SCA can have a behavioral phenotype that overlaps with behaviors seen in ASD. Many of these overlapping behaviors, including language delays, poor eye contact, oversensitivity to sensory stimuli, poor social relatedness, and restricted interests, have been previously reported in children with SCA [43, 44]. A 2008 study by van Rijn et al included 31 adults with XXY/KS reported significantly more distress during social interactions, less frequent social interactions, and more autistic traits including difficulties with communication, imagination, and social skills [15]. This increased risk for ASD may be the result of similar underlying neurodevelopmental differences in core areas also known to be affected in SCA (including language, verbal cognition, and executive functioning). However, there is also evidence showing both neuroanatomical and functional MRI differences between males with XXY/KS and idiopathic autism, despite similar scores on clinical rating scales such as the SRS [45, 46]. This suggests that at least some of the underlying neurobiologic abnormalities leading to clinical ASD symptoms may be different in XXY/KS. Future similar studies will also be important in other SCAs with higher reported rates of ASD such as XYY and XXYY. Further evaluation of other contributing genetic or environmental factors leading to a higher risk for ASD in all SCA subtypes is important.
From a clinical perspective, these results support that children with SCA are at increased risk for ASD, and that ASD should be considered when evaluating behavioral, social, educational and emotional concerns. The neurodevelopmental features in some individuals with SCA can be complex and involve deficits in cognition/learning, language, and executive functioning that overlap with ASD features. Evidence-based treatments for ASD should be offered to children with SCA and ASD symptoms, however studies are needed to determine if standard ASD treatments need to be modified for SCA. Many services and programs are well established in communities to support children with ASD and their families, and these services can potentially be advantageous for SCA as well.
Our study results support that males with Y chromosome aneuploidy (XYY and XXYY) have an increased prevalence of ASD compared to males with XXY/KS and compared to XY males in the general population. This is reflected not only in higher rates of ASD in the XYY and XXYY groups, but also in the higher clinical severity scores in the groups without ASD, where the XYY group was higher compared to the XXY/KS group. The role of Y chromosome genes that contribute to the susceptibility of ASD and their mechanism in leading to neurodevelopmental changes are important for understanding the pathogenesis of ASD and may be very important to understanding the male predominance in ASD. There has been limited ASD research on Y-chromosome specific genes. Polymorphisms in the neuroligin 4Y (NLGN4Y) gene involved in neurodevelopment and synaptic function have been associated with ASD [47, 48], and Ross et al. recently published genetic data demonstrating a positive correlation between NLGN4Y expression and ASD symptoms in children with XYY [49].
Another important Y-specific gene is SRY which is well known for its important role in triggering sexual differentiation. However SRY is also expressed in human brain regions into adulthood and has been shown to directly modulate enzymes involved in both production and breakdown of important catecholamine neurotransmitters. More specifically, SRY has been shown to upregulate tyrosine hydroxylase (the rate-limiting enzyme in the synthesis of catecholamines dopamine, epinephrine, and norepinephrine), beta-endorphin production, and monoamine oxidase-A (MAO-A), the enzyme that breaks down monoamine neurotransmitters including the catecholamines and serotonin.[50-53] Polymorphisms in MAO-A and abnormal levels of catecholamines, serotonin, and endorphin have been reported in various ASD studies and may thus be involved in the higher rates of ASD in XYY and XXYY as well.[54-56] Further investigation of the neurodevelopmental effects of Y-chromosome genes is important as it will contribute to understanding both the male predominance in ASD and the increased rate of ASD in males with Y chromosome aneuploidy.
Ascertainment bias is a limitation in this study and an important consideration in all SCA research. Bias exists both in the ascertainment (diagnosis) of patients with SCA due to more severe clinical symptoms (thus in the general population those with more significant involvement are diagnosed), as well as the bias of those who choose to participate in our research studies or who we reach through advertisements through support organizations. It is important to describe that in both studies, recruitment was not from a clinical sample, and recruitment flyers did not mention evaluation for ASD but a more general study of health and development. While sample size and limited racial/ethnic diversity are additional limitations in this study, it is an important study in that it is the largest study to utilize and compare standardized ASD clinical assessment in a research cohort of males with SCA. Larger studies on a sample ascertained prenatally and followed prospectively are needed, and may soon be possible with recent advances in prenatal genetic testing which can identify SCA through genetic testing of maternal blood samples. This noninvasive prenatal testing has already started to notably increase diagnosis rates of infants with SCA.[57, 58]
In summary, results support that ASD should be considered for males with SCA with social difficulties in order to help direct treatment recommendations, although other aspects of the phenotype including language and learning disorders, executive function, attention deficits, and emotional factors need to be considered during the diagnostic process and in developing a treatment plan. By comparing SCA subtypes, we have shown that Y polysomies are more likely to have autism compared to X chromosome polysomies, indicating the importance of future study of Y chromosome genes and Y chromosome gene dosage in potentially understanding the etiology and male predominance of ASD.
Acknowledgments
This study was supported by The Children's Miracle Network of the UC Davis Medical Center, NIH/NCATS Colorado CTSA Grant Number UL1 TR001082, NIH Pediatric LRP and NINDS K23NS070337 to NRT; and NIH MH77554 to DH.JR acknowledges support from the Thomas Jefferson University Dean's Transformational Science Award, US Department of Defense CDMRP AR140197, and 1 R21 MH109158. Contents are the authors' sole responsibility and do not necessarily represent official NIH views.
Abbreviations
- SCA
Sex Chromosome Aneuploidy
- ASD
Autism Spectrum Disorder
- PDD-NOS
Pervasive Developmental Disorder-Not Otherwise Specified
- VIQ
Verbal Intelligence Quotient
- PIQ
Performance Intelligence Quotient
- FSIQ
Full-Scale Intelligence Quotient
- SES
Socioeconomic status
Footnotes
The authors declare no conflicts of interest related to this project
References
- 1.Schaefer GB, Mendelsohn NJ P.P.a.G. Committee. Clinical genetics evaluation in identifying the etiology of autism spectrum disorders: 2013 guideline revisions. Genet Med. 2013;15(5):399–407. doi: 10.1038/gim.2013.32. [DOI] [PubMed] [Google Scholar]
- 2.Hagerman RJ. Neurodevelopmental Disorders: Diagnosis and Treatment. New York: Oxford University Press; 1999. [Google Scholar]
- 3.Konstantareas M, Homatidis S. Chromosomal Abnormalities in a Series of Children with Autistic Disorders. Journal of Autism and Developmental Disorders. 1999;29(4):275–285. doi: 10.1023/a:1022155201662. [DOI] [PubMed] [Google Scholar]
- 4.Kielinen M, et al. Associated medical disorders and disabilities in children with autistic disorder - a population based study. Autism. 2004;8(1):49–60. doi: 10.1177/1362361304040638. [DOI] [PubMed] [Google Scholar]
- 5.Stuart SW, King CH, Pai GS. Autism spectrum disorder, Klinefelter syndrome, and chromosome 3p21.31 duplication: a case report. MedGenMed. 2007;9(4):60. [PMC free article] [PubMed] [Google Scholar]
- 6.Nicolson R, Bhalerao S, Sloman L. 47, XYY karyotypes and pervasive developmental disorders. Can J Psychiatry. 1998;43:619–22. doi: 10.1177/070674379804300611. [DOI] [PubMed] [Google Scholar]
- 7.Merhar SL, Manning-Courtney P. Two boys with 47, XXY and autism. J Autism Dev Disord. 2007;37(5):840–6. doi: 10.1007/s10803-006-0211-1. [DOI] [PubMed] [Google Scholar]
- 8.Jha P, Sheth D, Ghaziuddin M. Autism spectrum disorder and Klinefelter syndrome. Eur Child Adolesc Psychiatry. 2007;16(5):305–8. doi: 10.1007/s00787-007-0601-8. [DOI] [PubMed] [Google Scholar]
- 9.Coffee B, et al. Incidence of fragile X syndrome by newborn screening for methylated FMR1 DNA. Am J Hum Genet. 2009;85(4):503–14. doi: 10.1016/j.ajhg.2009.09.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bojesen A, Juul S, Gravholt CH. Prenatal and postnatal prevalence of Klinefelter syndrome: a national registry study. J Clin Endocrinol Metab. 2003;88(2):622–6. doi: 10.1210/jc.2002-021491. [DOI] [PubMed] [Google Scholar]
- 11.Bender B, et al. Speech and language development in 41 children with sex chromosome anomalies. Pediatrics. 1983;71(2):262–7. [PubMed] [Google Scholar]
- 12.Boada R, et al. The cognitive phenotype in Klinefelter syndrome: a review of the literature including genetic and hormonal factors. Dev Disabil Res Rev. 2009;15(4):284–94. doi: 10.1002/ddrr.83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Pennington BF, et al. Learning disabilities in children with sex chromosome anomalies. Child Dev. 1982;53(5):1182–92. [PubMed] [Google Scholar]
- 14.Geschwind D, Dykens E. Neurobehavioral and psychosocial issues in klinefelter syndrome. Learning Disabilities Research & Practice. 2004;19(3):166–173. [Google Scholar]
- 15.Van Rijn S, et al. Social Behavior and Autism Traits in a Sex Chromosomal Disorder: Klinefelter (47XXY) Syndrome. J Autism Dev Disord. 2008;38(9):1634–41. doi: 10.1007/s10803-008-0542-1. [DOI] [PubMed] [Google Scholar]
- 16.Bishop DV, et al. Autism, language and communication in children with sex chromosome trisomies. Arch Dis Child. 2011;96(10):954–9. doi: 10.1136/adc.2009.179747. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ross JL, et al. Behavioral and social phenotypes in boys with 47, XYY syndrome or 47, XXY Klinefelter syndrome. Pediatrics. 2012;129(4):769–78. doi: 10.1542/peds.2011-0719. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Samango-Sprouse CA, et al. Is it all the X: familial learning dysfunction and the impact of behavioral aspects of the phenotypic presentation of XXY? Am J Med Genet C Semin Med Genet. 2013;163C(1):27–34. doi: 10.1002/ajmg.c.31353. [DOI] [PubMed] [Google Scholar]
- 19.Bruining H, et al. Psychiatric characteristics in a self-selected sample of boys with Klinefelter syndrome. Pediatrics. 2009;123(5):e865–70. doi: 10.1542/peds.2008-1954. [DOI] [PubMed] [Google Scholar]
- 20.Ratcliffe S. Long-term outcome in children of sex chromosome abnormalities. Arch Dis Child. 1999;80(2):192–5. doi: 10.1136/adc.80.2.192. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Geerts M, Steyaert J, Fryns JP. The XYY syndrome: a follow-up study on 38 boys. Genet Couns. 2003;14(3):267–79. [PubMed] [Google Scholar]
- 22.Bishop DV, et al. Autism, language and communication in children with sex chromosome trisomies. Arch Dis Child. 2010 doi: 10.1136/adc.2009.179747. p. epub ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Muldal S, Ockey CH. Double male:New chromosome constitution in Klinefelter's syndrome. Lancet. 1960;2:492–3. doi: 10.1530/acta.0.0390183. [DOI] [PubMed] [Google Scholar]
- 24.Tartaglia N, et al. A New Look at XXYY Syndrome: Medical and Psychological Features. American Journal of Medical Genetics Part A. 2008;146A(12):1509–22. doi: 10.1002/ajmg.a.32366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Schlegel RJ, et al. Studies on a Boy with Xxyy Chromosome Constitution. Pediatrics. 1965;36:113–9. [PubMed] [Google Scholar]
- 26.Sorensen K, et al. The 48, XXYY syndrome. J Ment Defic Res. 1978;22(3):197–205. [PubMed] [Google Scholar]
- 27.Fryns JP, et al. XYY syndrome and other Y chromosome polysomies. Mental status and psychosocial functioning. Genet Couns. 1995;6(3):197–206. [PubMed] [Google Scholar]
- 28.Association AP. Diagnostic and statistical manual of mental disorders (5th ed) (DSM-5) Arlington, VA: American Psychiatric Publishing; 2013. [Google Scholar]
- 29.Lord C, Rutter M. Autism Diagnostic Observation Scales - 2nd Edition (ADOS-2) Torrance, CA: Western Psychological Services; 2012. [Google Scholar]
- 30.Ben-Shachar S, et al. 22q11.2 distal deletion: a recurrent genomic disorder distinct from DiGeorge syndrome and velocardiofacial syndrome. American Journal of Human Genetics. 2008;82(1):214–221. doi: 10.1016/j.ajhg.2007.09.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Mullen E. Mullen Scales of Early Learning. Cranston, RI: TOTAL Child, Inc; 1989. [Google Scholar]
- 32.Wechsler D. Weschler Abbreviated Scale of Intelligence. San Antonio: The Psychological Corporation; 1999. [Google Scholar]
- 33.Sparrow S, Cicchetti D, Balla D. Vineland Adaptive Behavior Scales - 2nd Edition. Circle Pines, MN: American Guidance Service; 2005. [Google Scholar]
- 34.Hollingshead A, Redlich F. Social class - 2 Factor Index of Social Position. New York: John Wiley; 1958. [Google Scholar]
- 35.Rutter M, et al. Social Communication Questionnaire (SCQ) Los Angeles, CA: Western Psychological Services; 2003. [Google Scholar]
- 36.Constantino JN. The Social Responsiveness Scale. Los Angeles, CA: Western Psychological Services; 2005. [Google Scholar]
- 37.Lord C. The autism diagnostic observation schedule-generic:a standard measure of social and communication deficits associated with the spectrum of autism. J Autism Dev Disord. 2000;30(3):205–23. [PubMed] [Google Scholar]
- 38.Hus V, Lord C. The autism diagnostic observation schedule, module 4: revised algorithm and standardized severity scores. J Autism Dev Disord. 2014;44(8):1996–2012. doi: 10.1007/s10803-014-2080-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Gotham K, et al. The Autism Diagnostic Observation Schedule: revised algorithms for improved diagnostic validity. J Autism Dev Disord. 2007;37(4):613–27. doi: 10.1007/s10803-006-0280-1. [DOI] [PubMed] [Google Scholar]
- 40.Gotham K, Pickles A, Lord C. Standardizing ADOS scores for a measure of severity in autism spectrum disorders. J Autism Dev Disord. 2009;39(5):693–705. doi: 10.1007/s10803-008-0674-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Lord C. Autism Diagnostic Interview- Revised: a revised version of a diagnostic interview for caregivers of individulas with possible pervasive developmental disorders. J Autism Dev Disord. 1994;24(5):659–85. doi: 10.1007/BF02172145. [DOI] [PubMed] [Google Scholar]
- 42.Elliott C. Differential Abilities Scales - 2nd Edition (DAS-II) San Antonio, TX: Pearson; 2007. [Google Scholar]
- 43.Geschwind DH, et al. Neurobehavioral phenotype of Klinefelter syndrome. Ment Retard Dev Disabil Res Rev. 2000;6(2):107–16. doi: 10.1002/1098-2779(2000)6:2<107::AID-MRDD4>3.0.CO;2-2. [DOI] [PubMed] [Google Scholar]
- 44.van Rijn S, et al. Social behavior and autism traits in a sex chromosomal disorder: Klinefelter (47XXY) syndrome. J Autism Dev Disord. 2008;38(9):1634–41. doi: 10.1007/s10803-008-0542-1. [DOI] [PubMed] [Google Scholar]
- 45.van Rijn S, et al. Social cognition and underlying cognitive mechanisms in children with an extra X chromosome: a comparison with autism spectrum disorder. Genes Brain Behav. 2014;13(5):459–67. doi: 10.1111/gbb.12134. [DOI] [PubMed] [Google Scholar]
- 46.Brandenburg-Goddard MN, et al. A comparison of neural correlates underlying social cognition in Klinefelter syndrome and autism. Soc Cogn Affect Neurosci. 2014;9(12):1926–33. doi: 10.1093/scan/nst190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Serajee FJ, Mahbubul Huq AH. Association of Y chromosome haplotypes with autism. J Child Neurol. 2009;24(10):1258–61. doi: 10.1177/0883073809333530. [DOI] [PubMed] [Google Scholar]
- 48.Yan J, et al. Analysis of the neuroligin 4Y gene in patients with autism. Psychiatr Genet. 2008;18(4):204–7. doi: 10.1097/YPG.0b013e3282fb7fe6. [DOI] [PubMed] [Google Scholar]
- 49.Ross JL, et al. Behavioral phenotypes in males with XYY and possible role of increased NLGN4Y expression in autism features. Genes Brain Behav. 2015;14(2):137–44. doi: 10.1111/gbb.12200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Cohen I, et al. Association of autism severity with a monoamine oxidase A functional polymorphism. Clinical Genetics. 2003;64:190–197. doi: 10.1034/j.1399-0004.2003.00115.x. [DOI] [PubMed] [Google Scholar]
- 51.Mayer A, et al. The Y-chromosomal genes SRY and ZFY are transcribed in adult human brain. Neurogenetics. 1998;1(4):281–8. doi: 10.1007/s100480050042. [DOI] [PubMed] [Google Scholar]
- 52.Wu JB, et al. Regulation of monoamine oxidase A by the SRY gene on the Y chromosome. FASEB J. 2009;23(11):4029–38. doi: 10.1096/fj.09-139097. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Milsted A, et al. Regulation of tyrosine hydroxylase gene transcription by Sry. Neurosci Lett. 2004;369(3):203–7. doi: 10.1016/j.neulet.2004.07.052. [DOI] [PubMed] [Google Scholar]
- 54.Tassone A, et al. MAOA, DBH, and SCL6A4 variants in CHARGE: a case-control study of autism spectrum disorders. Autism Res. 2011;4(4):250–61. doi: 10.1002/aur.196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Cohen I, et al. Autism severity is associated with child and maternal MAOA genotypes. Clin Genetics. 2011;79(4):355–62. doi: 10.1111/j.1399-0004.2010.01471.x. [DOI] [PubMed] [Google Scholar]
- 56.Muller C, et al. The serotonin system in autism spectrum disorder: from biomarker to animal model. Neuroscience. 2016 May;3(321):24–41. doi: 10.1016/j.neuroscience.2015.11.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Song Y, et al. Noninvasive prenatal testing for fetal aneuploidies in the first trimester of pregnancy. Ultrasound Obstet Gynecol. 2014 doi: 10.1002/uog.13460. [DOI] [PubMed] [Google Scholar]
- 58.Zimmermann B, et al. Noninvasive prenatal aneuploidy testing of chromosomes 13, 18, 21, X, and Y, using targeted sequencing of polymorphic loci. Prenat Diagn. 2012;32(13):1233–41. doi: 10.1002/pd.3993. [DOI] [PMC free article] [PubMed] [Google Scholar]