Autism spectrum disorder and other neurobehavioural comorbidities in rare disorders of the Ras/MAPK pathway

Transcription

1 DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY ORIGINAL ARTICLE Autism spectrum disorder and other neurobehavioural comorbidities in rare disorders of the Ras/MAPK pathway SHRUTI GARG 1 AMI BROOKS 1 AMY BURNS 1 EMMA BURKITT-WRIGHT 2 BRONWYN KERR 2 SUSAN HUSON 2 RICHARD EMSLEY 3 JONATHAN GREEN 1 1 Division of Neuroscience & Experimental Psychology, Faculty of Biological, Medical & Health Sciences, University of Manchester and Royal Manchester Children s Hospital and Manchester Academic Health Sciences Centre, Manchester; 2 Manchester Centre for Genomic Medicine, Faculty of Biological, Medical & Health Sciences, University of Manchester and Central Manchester Foundation NHS Trust, Manchester; 3 Centre for Biostatistics, School of Health Sciences, Faculty of Biological, Medical & Health Sciences, University of Manchester, Manchester, UK. Correspondence to Jonathan Green at Jean McFarlane Building, University of Manchester, Manchester M13 9PL, UK. jonathan.green@manchester.ac.uk This article is commented on by Sznajer on page 461 of this issue. PUBLICATION DATA Accepted for publication 13th December Published online 4th February ABBREVIATIONS ADOS Autism Diagnostic Observation Scale ASD Autism spectrum disorder CFC Cardiofaciocutaneous DANVA Diagnostic Analysis of Nonverbal Accuracy NF1 Neurofibromatosis type 1 Ras/ MAPK Ras/Mitogen Activated Protein Kinase TEA-Ch Test of Everyday Attention for Children TOF Test Observation Form AIM To investigate the cognitive and behavioural phenotype in rare disorders of the Ras/ MAPK pathway, namely Noonan, cardiofaciocutaneous (CFC), and Costello syndromes, particularly prevalence of autism spectrum disorder (ASD) and attention-deficit hyperactivity disorder (ADHD). METHOD Fifty children were recruited over 10 months through the regional genetics service and advertisements. A range of parent, child, and observational measures were administered including Autism Diagnostic Interview-Revised and Autism Diagnostic Observation Scale. RESULTS Using the Collaborative Programme for Excellence in Autism criteria, 12 out of 40 children with Noonan syndrome (30%) showed ASD, and 12 out of 40 (30%) with partial ASD features and 16 out of 40 (40%) showed non-asd. The Noonan syndrome ASD group showed male dominance in a ratio of 5:1. In the CFC group, eight out of nine children met the criteria for ASD, with equal sex distribution. Additionally 19 out of 40 (48%) of the Noonan syndrome group and eight out of nine (88.9%) of the CFC group scored met clinical criteria for ADHD. Only one child was in the Costello syndrome group. INTERPRETATION This is the first systematic study to suggest a high prevalence of ASD in Noonan and CFC syndromes, and thus offers crucial evidence to support the importance of the Ras/MAPK pathway in the aetiology of ASD. Limitations include the inevitable possibility of a sampling bias in a rare disorder study of this kind. The RASopathies are a group of clinically related disorders caused by germline mutations in a number of genes that encode components or regulators of the Ras/Mitogen Activated Protein Kinase (Ras/MAPK) signalling pathway. These disorders include Noonan syndrome, cardiofaciocutaneous (CFC) syndrome, Noonan-like syndrome with multiple lentigines, Mazzanti syndrome, Costello syndrome, neurofibromatosis type 1 (NF1), and Legius syndrome. Each RASopathy exhibits characteristic features, and shares a common mechanism of Ras/MAPK pathway dysregulation and overlapping clinical manifestations. These may include a wide spectrum of cognitive deficits, cardiac malformations, similar facies, postnatal reduced growth, and an increased risk of cancer. 1 The Ras/MAPK pathway is well known for its role in oncogenesis and tumour progression and is critically involved in the regulation of the cell cycle. 2 Intellectual impairment and psychiatric morbidity are known to be common in the RASopathies but have only been systematically studied to date in NF1. 3 Within the other RASopathies, NF1 aside, Noonan syndrome is the most common and most phenotypically variable, estimated to affect between 1 in 1000 to 2000 newborn infants. 2 It is a genetically heterogenous disorder, most frequently caused by mutations in PTPN11. Mutations in a number of other genes can also cause Noonan syndrome, sometimes with a gene specific phenotype (e.g. SOS1, RAF1, SHOC2, RIT1, CBL, KRAS, NRAS, and BRAF). Mutations in several other genes (RRAS, RASA2, A2ML1, SOS2, LZRT1) have been recently reported in small numbers of patients, but their role remains under investigation. 4 CFC syndrome is a rare disorder, caused by mutations in BRAF, MAP2K1, MAP2K2, or KRAS. It affects growth and development more severely than Noonan syndrome but systematic studies describing the cognitive behavioural phenotype are still in their infancy. Costello syndrome is a rare disorder caused by mutation in HRAS, which has the highest risk of cancer amongst the RASopathies. Costello 544 DOI: /dmcn Mac Keith Press

2 syndrome is associated with mild to moderate intellectual disability and poor adaptive functioning. Emerging evidence suggests the importance of the Ras/ MAPK signalling pathway in autism spectrum disorders (ASD). 5 Within autism studies, Pinto et al. 5 have shown that genes most affected by copy number variations and single nucleotide variations converge on three neural system functional networks which include the Ras/MAPK signalling pathway. Studies investigating the behavioural phenotype in Noonan and CFC syndromes describe ASDlike symptomatology and traits. For instance, in a prospective observational study of 112 adults, Shaw et al. 6 found that poor quality of life in Noonan syndrome was related to lack of social life and inability to fit in. More recently, Adviento et al. investigated the prevalence of autistic symptomatology in children and adults (age range 1 73y) using the Social Responsiveness Scale (SRS) and Social Communication Questionnaire (SCQ) in NF1 (n=81), Costello (n=44), Noonan (n=52), and CFC (n=54) syndromes, and found significant ASD traits of 11 per cent in NF1, 21 per cent in Noonan syndrome, 26 per cent in Costello syndrome, and 54 per cent in CFC syndrome. 7 Similarly, Alfieri et al. 8 used the SCQ and Modified Checklist for Autism in Toddlers (M-CHAT) to investigate autistic symptomatology in 70 patients (age range 2 28y) with RASopathies (Noonan syndrome n=38, CFC syndrome n=11, Costello syndrome n=21) and found significant ASD traits in 12 per cent with Noonan syndrome, 44 per cent in Costello syndrome, and 64 per cent in CFC syndrome. All of these studies are therefore suggestive of the presence of ASD-like symptomatology but are severely limited by the lack of detailed ascertainment and the absence of indepth, criterion standard ASD phenotyping instruments. Comorbid cognitive and behavioural problems are known to impact the specificity of quantitative phenotypic measures such as the SRS, thus limiting interpretation of the findings on that instrument. 9 The aim of this study is to investigate the neurobehavioural phenotype of the rare disorders of the Ras pathway Noonan, CFC, and Costello syndromes and, in particular, to investigate systematically the nature and severity of social impairment and ADHD using detailed in-person phenotyping. METHODS Participants Participants were recruited between January 2014 and November 2014 from the Clinical Genetic Service in the Manchester Centre for Genomic Medicine. Participants were approached consecutively. The study was also advertised on newsletters, family information days, and on the social media of Noonan, Costello, and CFC syndrome charities. The inclusion criteria for the current study were (1) clinical diagnosis of Noonan, Costello, or CFC syndrome, made by a clinical geneticist, and (2) participants aged between 6 years and 16 years. Medical notes were reviewed to confirm the diagnosis. When mutation testing What this paper adds Both autism spectrum disorder and attention-deficit hyperactivity disorder are highly prevalent in the RASopathies. had been performed, mutation status was confirmed. Eligible participants were sent study information sheets and consent to contact forms. This study was given approval by South Manchester Research Ethics Committee, approval number 10/H1003/77. Measures and procedures The Wechsler Abbreviated Scale of Intelligence-II (WASI- II) and Mullen Scales of Early Learning were used to assess intellectual functioning, parent-rated Behaviour Rating Inventory of Executive Function (BRIEF) as a measure of executive function, Autism Diagnostic Interview-Revised (ADI-R) parent interview and Autism Diagnostic Observation Schedule-2 (ADOS-2) was used to assess social communication impairments, Conners Parent Rating Scale-3rd Edition Short Form as a measure of ADHD symptomatology, Test of Everyday Attention for Children (TEA-Ch) as a measure of attention, Diagnostic Analysis of Nonverbal Accuracy (DANVA) to assess ability to identify facial expressions, and Test Observation Form (TOF) as a measure of researcher-rated behaviour. Detailed information about the measures, procedures, and diagnostic classifications of ASD and ADHD is provided in Appendix S1 (online supporting information). Statistical analysis Data were analysed in SPSS version 17 (SPSS Inc., Chicago, USA). For the Noonan syndrome and CFC syndrome groups, the group means and standard deviations were calculated for Full-scale IQ, BRIEF, and TEA-Ch measures including the rate of children scoring in the clinically impaired range. For the Noonan syndrome group, the means and standard deviation of ADI-R subdomains, ADOS, Conners, TOF, Full-scale IQ, and DANVA across the three subgroups (ASD, broad ASD, and non-asd) are described. Intelligence quotient and ADHD symptoms across the three subgroups were compared using non-parametric Kruskal Wallis test. A two-tailed p-value of less than 0.05 was considered significant. For the CFC group, the means and standard deviations of the ADI-R domains, ADOS, Conners, and TOF are described. RESULTS Patient demographics Fifty children (30 males, 20 females) received in-depth assessment. Mean age was 10 years and 4 months (standard deviation [SD] 2y 9mo, range 6y 16y 9mo) and mean IQ was 81 (SD 19.2, range ). Forty children had a diagnosis of Noonan syndrome; mean age was 10 years and 10 months (SD 2y 9mo, range 6y 16y 9mo), 25 males and 15 females. Of these, 16 out of 40 had a Special Educational Needs (SEN) Statement (a statutory document that sets out the additional educational help that a child ASD and ADHD in RASopathies Shruti Garg et al. 545

3 should receive in order to progress in school). Three had pre-existing diagnoses of ASD and one had a diagnosis of ADHD. Nine children had a diagnosis of CFC syndrome; mean age of the CFC group was 10 years and 1 month (SD 3y, range 6y 3mo 15y 7mo), five males and four females. Of these, eight out of nine had a SEN Statement. None of the children had a pre-existing diagnosis of ASD or ADHD. There was only a single child with Costello syndrome, and so the results for this child are not included below, although the results were shared with the family. Genetic status In the Noonan syndrome group, 21 participants had mutations in PTPN11, one in SOS1, one in CBL, two in SHOC2, one in KRAS, and two in RIT1; the mutation status of 12 participants was unknown. In the CFC group, eight had a BRAF mutation and one had a mutation in MAP2K1. Intelligence and executive functioning (Table I) Intelligence was measured using the WASI-II in 39 of the Noonan syndrome group and six of the CFC group. The Mullen Scale of Early Learning was used in one participant in the Noonan syndrome group and three in the CFC group. In the Noonan syndrome group, mean overall IQ was 87 (SD 13.9, range ); three out of 39 (7.7%) had Full-scale IQ below 70. In the CFC group, mean IQ using the WASI-II was 73.9 (SD 20.1, range ), mean verbal IQ 79.5 (SD 23.5), performance IQ 65.8 (SD 16.8); three out of six (50%) had Full-scale IQ below 70. The Mullen was used in three of the CFC sample; mean receptive language T scores on Mullen were (SD 5.13) and mean expressive language T scores (SD 3.46). Overall executive function was assessed using the BRIEF questionnaire (Table I). In the Noonan syndrome group, 42.8 per cent had clinically significant impairments in executive functioning with the biggest deficits seen in the working memory and initiation domains. In the CFC group, 66.7 per cent had clinically significant deficits in executive function. ASD assessment Noonan syndrome group Using the CPEA criteria, 12 children (30%; 10 males, two females) met the criteria for ASD, 12 (30%; eight males, four females) for broad ASD, and 16 (40%; seven males, Table I: Cognitive profile of Noonan and cardiofaciocutaneous (CFC) syndrome groups Noonan syndrome CFC syndrome Mean (SD) Clinical impairment % Mean (SD) Clinical impairment % WASI-II n 39 6 Full-scale IQ 87.2 (13.8) 73.8 (20.1) Verbal IQ 93.2 (13.8) 79.8 (23.5) Performance IQ 83.9 (12.9) 65.8 (16.8) Subtests Vocabulary 44.9 (8.8) 37.3 (13.3) Block design 40.4 (8.9) 27.0 (6.7) Similarities 46.7 (9.6) 39.2 (13.5) Matrix reasoning 40.2 (10.0) 32.7 (10.6) BRIEF scale T scores %T>65 %T>65 n 35 6 Inhibit 55.4 (10.6) (23.4) 55.5 Shift 63.4 (20.4) (9.7) 77.8 Emotional control 59.1 (16.3) (21.1) 66.7 Initiate 66.1 (12.5) (14.5) 88.9 Working memory 66.0 (14.5) (20.5) 77.8 Plan/organize 62.0 (15.2) (14.9) 55.5 Organization of material 58.8 (12.8) (31.1) 44.4 Self-monitor 63.0 (13.1) (17.5) 55.5 Behavioural regulation Index 59.2 (13.6) (17.5) 77.8 Metacognition index 65.4 (12.8) (18.6) 66.7 Global Executive Composite 64.5 (13.5) (17.0) 66.7 <1SD below mean TEA-Ch age-scaled scores n (%) n (%) n 32 4 Sustained attention Score! 6.7 (2.7) 15 (46.9) 2.8 (1.3) 4 (100) Sky Search dual task 6.1 (4.5) 19 (59.4) 1.0 (0.0) 4 (100) Selective attention Sky Search 7.8 (3.6) 13 (40.6) 7.0 (5.6) 2 (50) Attentional switching Creature Counting 8.1 (3.4) 11 (34.3) 7.8 (3.9) 2 (50) <1SD below mean WASI-II, Wechsler Abbreviated Scale of Intelligence-II; BRIEF, Behaviour Rating Inventory of Executive Function; TEA-Ch, Test of Everyday Attention for Children. 546 Developmental Medicine & Child Neurology 2017, 59:

4 Table II: Summary of the Noonan syndrome sample by autism spectrum disorder (ASD) diagnoses with ADI-R, ADOS, and comorbidity measures Mean (SD) ASD (n=12) Broad ASD (n=12) Non-ASD (n=16) Kruskal Wallis test Χ 2 p ASD assessment measures ADI-R Social interaction 17.3 (6.1) 14.4 (6.2) 3.9 (3.3) Social communication 14.0 (4.7) 8.9 (4.4) 4.1 (4.3) Restricted, repetitive behaviours 4.5 (3.6) 3.5 (3.0) 1.4 (1.9) ADOS Social affect 8.8 (3.6) 4.0 (3.0) 2.5 (1.7) Restrictive, repetitive behaviours 2.5 (1.2) 0.8 (0.8) 0.7 (0.7) Social affect and restrictive, 11.0 (3.2) 4.7 (3.1) 3.0 (1.9) repetitive behaviours ADHD assessment measures Conners Inattention T score 74.1 (12.1) 75.9 (16.1) 71.5 (11.9) Hyperactivity T score 72.7 (15.7) 74.4 (13.8) 64.4 (12.9) Test Observation Form Inattention T score 71.5 (12.7) 63.7 (8.8) 66.8 (10.8) Hyperactivity T score 71.5 (11.9) 64.8 (6.9) 66.8 (7.7) ADHD T score 72.0 (10.6) 64.0 (7.4) 67.1 (8.3) Full-scale IQ 84.6 (10.7) 82.0 (11.0) 92.9 (16.0) Verbal IQ 89.0 (10.3) 88.8 (11.9) 99.4 (15.4) Performance IQ 86.4 (8.5) 79.3 (12.2) 85.6 (15.5) DANVA Total errors 9.18 (4.31) 7.75 (3.79) 5.69 (3.53) High intensity errors 3.82 (2.40) 2.92 (2.06) 2.19 (1.87) Low intensity errors 5.36 (2.16) 4.83 (2.04) 3.50 (1.96) ADHD, attention-deficit hyperactivity disorder; ADI-R, Autism Diagnostic Interview-Revised; ADOS, Autism Diagnostic Observation Scale; DANVA, Diagnostic Analysis of Nonverbal Accuracy. nine females) children as non-asd. Table II shows the detailed scores on all ascertainment instruments grouped by diagnostic category. There is no overall difference in IQ across the three groups. There were no significant differences between the three groups on the rates of ADHD symptomatology on parent- and researcher-rated measures. On the DANVA, total errors were highest in the ASD group, followed by broad ASD and non-asd. Similar ordering across the groups was seen for errors on high and low intensity facial expressions. CFC group Eight out of nine children (88.9%; four males, four females) met the criteria for ASD, one out of nine for broader ASD using the CPEA criteria. On the ADI-R, the mean scores of the CFC sample across the three domains were as follows: social interaction (SD 5.89), communication (SD 4.10), and RRBs 5.56 (SD 2.45). On the ADOS, the mean scores on social affect domain were 8.22 (SD 5.09) and RRB was 4.00 (SD 1.41). The mean ADOS total scores were (SD 5.19). DANVA data was available for six participants with CFC syndrome; mean total error score on DANVA were (SD 5.35), mean high intensity error score 4.67 (SD 2.73), and mean low intensity error 6.0 (SD 2.68). ADHD Noonan syndrome group On the parent-rated Conners questionnaire, 30 out of 40 (75%) had clinically significant scores for inattention, 27 out of 40 (67.5%) for hyperactivity, and 67.5 per cent had clinically significant scores on both domains. On the researcher-rated TOF, 14 out of 40 (35%) had T scores greater than 70 on inattention domain, 14 out of 40 (35%) on hyperactivity domain, and 15 out of 40 (37.5%) had T scores greater than 70 on ADHD index. On the TEA-Ch subtests (Table I), significant impairments (defined as 1 SD below mean or age scaled scores <7) were seen in 46.9 per cent on Score subtest, 59.4 per cent on Sky Search dual task, 40.6 per cent on Sky Search, and 34.3 per cent on Creature Counting. Triangulating the results from the three instruments (TEA-Ch, Conners, and TOF instruments; see Appendix S1), 21 out of 40 children (52.5%) with Noonan syndrome were rated to have significant ADHD. CFC group On the parent-rated Conners, all participants had clinically significant T scores for inattention and 8 out of 9 (88.9%) for hyperactivity on the Conners. On the researcher-rated TOF, 6 out of 9 (66.7%) had T scores greater than 70 on inattention, 6 out of 9 (66.7%) on hyperactivity, and 7 out of 9 (77.8%) on combined ADHD index. On the TEA-Ch in the CFC group, significant impairments were seen in 100 per cent on Score subtest, 100 per cent on Sky Search dual task, 50 per cent on Sky Search, and 50 per cent on Creature Counting. Triangulating the results on the three instruments, all participants with CFC syndrome were considered to have significant ADHD. Other comorbidities On the TOF in the Noonan syndrome group, 19 out of 40 (47.5%) had significant scores for internalizing disorders ASD and ADHD in RASopathies Shruti Garg et al. 547

5 and 24 out of 40 (60%) for externalizing disorders. In the CFC group, 7 out of 9 (77.8%) had significant scores for internalizing disorders and 8 out of 9 (88.9%) for externalizing disorders. DISCUSSION To our knowledge, this is the first study to rigorously assess the neurobehavioural phenotype in Noonan and CFC syndromes using criterion standard, in-depth ASD phenotyping measures. The estimated prevalence of ASD in the Noonan syndrome group is 30 per cent with a further 30 per cent showing partial features. These prevalence rates are substantially higher than previously reported estimates using parent-rated questionnaire measures. 7,8 The Full-scale IQ in the Noonan syndrome ASD group was within the normal range and not significantly different to the broad ASD and non-asd group, which implies that the social impairments in Noonan syndrome group cannot be explained on the basis of cognitive deficits (Table II). As in non-syndromic ASD, within our Noonan syndrome group we found a preponderance of ASD in males with a sex ratio of 5:1; these rates are comparable to the findings in non-syndromic ASD. 10 The CFC group showed significant levels of cognitive impairment but near complete prevalence of ASD. Within this small CFC sample, the sex bias was not evident with a 1:1 sex ratio for the prevalence of ASD. Lack of ASD sex bias has been described in other genetic disorders, such as tuberous sclerosis 11 and Cornelia De Lange syndrome, 12 which are associated with significant levels of intellectual impairment. Overall intellectual functioning in the Noonan syndrome group was comparable to previously reported studies. 13,14 Executive functioning assessed using parent-rated BRIEF measure suggested clinically significant executive functioning deficits in 42 per cent of the Noonan syndrome group and 66 per cent of the CFC sample. The largest deficits were seen in the two subtests (Score! and Sky Search Dual Task) measuring sustained attention, similar to the findings reported by Pierpont et al. 14 Deficits in sustained attention on the TEA- Ch have also been found in children with idiopathic ADHD; Heaton et al. 15 investigated the use of TEA-Ch in 63 children with ADHD compared to non ADHD controls and found that the ADHD group performed significantly worse on sustained attention and attentional control but not on selective attention. These findings therefore suggest that the ADHD symptomatology in Noonan and CFC syndromes are phenomenologically similar to idiopathic ADHD. In other comorbidities, we found higher rates of internalizing and externalizing disorders than previously reported. 8 On researcher-rated TOF, we found 47.5 per cent of the Noonan syndrome group and 77.8 per cent of the CFC group had significant rates of internalizing disorder. Similarly 60 per cent of the Noonan syndrome group and 88.9 per cent of the CFC group had significant rates of externalizing disorders. How clinically significant are these findings? The NIHCD CPEA criteria produce categories equivalent to clinical diagnoses, with the broad ASD group containing partial traits that are considered also as having clinical significance. 16 There does not appear to be any unusual pattern of ASD symptoms within the Noonan syndrome or CFC groups. Moreover, as in idiopathic autism, the Noonan syndrome and CFC groups show difficulties in recognition of facial expressions particularly on low intensity expressions. 17 Idiopathic autism shows a striking male preponderance in a ratio of 4:1 but in samples with intellectual disability, males and females are more equally represented. 18 ASD sex ratios of 5:1 in Noonan syndrome and 1:1 in CFC syndrome thus mirror the findings in idiopathic autism literature. These results are therefore suggestive of phenomenological similarities of ASD in Noonan syndrome and CFC syndrome with idiopathic ASD. Given the extreme variability and underdiagnosis of Noonan syndrome and the extreme rarity of CFC and Costello syndromes, and hence their patchy geographical spread throughout the UK, it was not possible to do systematic population-based ascertainment of the kind we undertook in our NF1 study. 19 The recruitment of the sample through clinic referral and self-referral from advertisement through charity networks inevitably produces the possibility of referral bias in the sample and ascertainment bias in the results and is a limitation regarding representativeness. The same difficulties are also inherent in previous questionnaire-based studies of these disorders. Further systematic study using complementary sampling methods will be useful to test replication of the results. Other limitations of the study include small sample sizes, in particular of the CFC group. Nonetheless despite the genetic heterogeneity of the sample, there appears to be a consistent set of behavioural symptomatology in these individuals with disorders affecting the Ras/MAPK pathway. Mutation status was known in 28 of the 40 patients with a clinical diagnosis of Noonan syndrome. The NSEuronet database ( contains mutation information on 1714 patients with Noonan syndrome. In that cohort, PTPN11 mutations account for 61 per cent of patients, SOS1 16 per cent, RAF1 7 per cent, RIT1 3 per cent, and all other genes less than 1 per cent. Although the small numbers make direct comparison impossible, the relatively high proportion of PTPN11 mutations in the mutation positive patients in this cohort (75%) may reflect that the fact that cognitive and behavioural difficulties are more common in Noonan syndrome because of PTPN11 mutations. Further study will be required to determine the frequency of neurobehavioural difficulties with specific genotypes. The strengths of this study are the use of rigorous, criterion standard measures of psychopathology, including triangulation between parent-rated, in-person child assessment, and researcher-rated measurements. These methods minimize rater biasing of results and make our results comparable with similar studies of other related disorders, which will be crucial in order to learn about which aspects of phenotype may be specific to particular disorders or general across them. 548 Developmental Medicine & Child Neurology 2017, 59:

6 The Ras/MAPK pathway is well studied in cancer and is considered an attractive therapeutic target for cancer treatment. 20 The results from this study complement our similar approach to measurement of ASD and ADHD in NF1, 19 offering empirical support that multiple members of a well-defined biochemical pathway contribute to ASD. The Ras/MAPK pathway has thus the potential to provide a unifying theory of autism, 21 with the possibility of targeted interventions capable to ameliorating the neurobehavioural phenotype. 22 From a clinical perspective, these findings add important specificity to the knowledge about the clinical psychopathology of the disorders. The high prevalence of ASD and cooccurrence of ADHD have important implications for clinical management and educational planning. Our data on preexisting diagnosis make it clear that ASD and ADHD are underrecognized in these groups possibly because of diagnostic overshadowing, where the behavioural difficulties are ascribed to the primary disorder itself. Our findings suggest that screening for ASD and ADHD should be undertaken in all children with Noonan and CFC syndromes, and that these developmental disorders should be considered as part of the differential diagnoses when children present in the clinic with problems of behaviour, social adjustment, or relationships. ACKNOWLEDGEMENTS This study was funded by the Newlife Foundation for Disabled Children grant reference 10-11/10. The authors want to thank Karen Tricker for her assistance with the ethics application, and the Noonan Syndrome Association (UK) and Costello Kids for their help with recruitment. We also thank patients and families who participated in this study and were incredibly generous with their time. The authors have stated that they had no interests that might be perceived as posing a conflict or bias. SUPPORTING INFORMATION The following additional material may be found online: Appendix S1: Measures and procedures. REFERENCES 1. Digilio MC, Lepri F, Baban A, et al. RASopathies: clinical diagnosis in the first year of life. Mol Syndromol 2011; 1: Rauen KA. The RASopathies. Annu Rev Genomics Hum Genet 2013; 14: Garg S, Green J, Leadbitter K, et al. Neurofibromatosis type 1 and autism spectrum disorder. Pediatrics 2013; 132: e Aoki Y, Niihori T, Inoue S, Matsubara Y. Recent advances in RASopathies. J Hum Genet 2016; 61: Pinto D, Delaby E, Merico D, et al. Convergence of genes and cellular pathways dysregulated in autism spectrum disorders. Am J Hum Genet 2014; 94: Shaw AC, Kalidas K, Crosby AH, Jeffery S, Patton MA. The natural history of Noonan syndrome: a long-term follow-up study. Arch Dis Child 2007; 92: Adviento B, Corbin IL, Widjaja F, et al. Autism traits in the RASopathies. J Med Genet 2014; 51: Alfieri P, Piccini G, Caciolo C, et al. Behavioral profile in RASopathies. Am J Med Genet A 2014; 164A: Hus V, Bishop S, Gotham K, Huerta M, Lord C. Factors influencing scores on the social responsiveness scale. J Child Psychol Psychiatry 2013; 54: Baird G, Simonoff E, Pickles A, et al. Prevalence of disorders of the autism spectrum in a population cohort of children in South Thames: the Special Needs and Autism Project (SNAP). Lancet 2006; 15: Jeste SS, Sahin M, Bolton P, Ploubidis GB, Humphrey A. Characterization of autism in young children with tuberous sclerosis complex. J Child Neurol 2008; 23: Moss J, Howlin P, Magiati I, Oliver C. Characteristics of autism spectrum disorder in Cornelia de Lange syndrome. J Child Psychol Psychiatry 2012; 53: Lee DA, Portnoy S, Hill P, Gillberg C, Patton MA. Psychological profile of children with Noonan syndrome. Dev Med Child Neurol 2005; 47: Pierpont EI, Tworog-Dube E, Roberts AE. Attention skills and executive functioning in children with Noonan syndrome and their unaffected siblings. Dev Med Child Neurol 2015; 57: Heaton SC, Reader SK, Preston AS, et al. The Test of Everyday Attention for Children (TEA-Ch): patterns of performance in children with ADHD and clinical controls. Child Neuropsychol 2001; 7: Lainhart JE, Bigler ED, Bocian M, et al. Head circumference and height in autism: a study by the Collaborative Program of Excellence in Autism. Am J Med Genet A 2006; 140: Grossman RB, Tager-Flusberg H. Who said that? Matching of low- and high-intensity emotional prosody to facial expressions by adolescents with ASD. J Autism Dev Disord 2012; 42: Fombonne E. The prevalence of autism. JAMA 2003; 289: Garg S, Lehtonen A, Huson SM, et al. Autism and other psychiatric comorbidity in neurofibromatosis type 1: evidence from a population-based study. Dev Med Child Neurol 2013; 55: Santarpia L, Lippman SM, El-Naggar AK. Targeting the MAPK-RAS-RAF signaling pathway in cancer therapy. Expert Opin Ther Targets 2012; 16: Packer A. Ras pathway, a potentially unifying theory of autism. Simons Foundation Autism Research Initiative. 13 March Available at: lly-unifying-theory-of-autism (accessed 9 January 2014). 22. Li W, Cui Y, Kushner S, et al. The HMG-CoA reductase inhibitor lovastatin reverses the learning and attention deficits in a mouse model of neurofibromatosis type 1. Curr Biol 2005; 15: ASD and ADHD in RASopathies Shruti Garg et al. 549