We examined the rate of positive findings of DES in 163 patients from the ASD cohort and observed 25.8% diagnostic yield in characterized genetic etiology. Seven patients (seven of 120, 5.8%) received a report with a candidate genetic etiology (TRIP12, HDAC1, SETD5, MTOR, HTR2C, RYR3, and MN1). Three of these genes were later reclassified to characterized findings based on corroborating evidences from the literature (SETD5, TRIP12, and MTOR).²⁸x28Grozeva, D., Carss, K., Spasic-Boskovic, O. et al. De novo loss-of-function mutations in SETD5, encoding a methyltransferase in a 3p25 microdeletion syndrome critical region, cause intellectual disability. Am J Hum Genet. 2014; 94: 618–624

Abstract | Full Text | Full Text PDF | PubMed | Scopus (30)See all References In addition, we identified a suspected candidate genetic etiology in the MTOR gene likely arising from gonadal mosaicism,²⁹x29Mroske, C., Rasmussen, K., Shinde, D.N. et al. Germline activating MTOR mutation arising through gonadal mosaicism in two brothers with megalencephaly and neurodevelopmental abnormalities. BMC Med Genet. 2015; 16: 102

Crossref | PubMed | Scopus (7)See all References a gene that was recently characterized in an independent report.²⁷x27Baynam, G., Overkov, A., Davis, M. et al. A germline MTOR mutation in Aboriginal Australian siblings with intellectual disability, dysmorphism, macrocephaly, and small thoraces. Am J Med Genet A. 2015; 167: 1659–1667

Crossref | PubMed | Scopus (11)See all References One originally negative and four uncertain patients were subsequently provided with a definitive diagnosis, highlighting the value of a proactive reanalysis process that is based on new literature as well as further cosegregation analysis. Among all 42 positive or likely positive characterized genetic etiologies reported, five of the 42 unique genes were not related to ASDs based on the known clinical spectrum of the gene. One patient with an alteration in FHL1 gene received a diagnosis of Emery-Dreifuss muscular dystrophy, and another patient with an alteration in GRHL3 gene received a diagnosis of Van der Woude syndrome characterized by cleft lip and/or palate. The remaining three genes were identified in probands receiving dual diagnoses. The fourth patient with a dual diagnosis received two ASD-related diagnoses. ASDs are fairly common disorders and may be unrelated to the other diagnosis in probands with dual diagnoses. Alternatively, the additional diagnosis may represent an expansion of the clinical phenotype that has not yet been reported. The rest of the genes have previously been identified in patients with ASD, ID, and/or DD, or implicated in syndromes in which autism is a common phenotypic feature.

Interestingly, patients presenting with psychiatric conditions or ataxia and/or paraplegia in addition to autistic features were significantly more likely to receive positive results compared with patients without these features, and this may imply that patients with these combinations of phenotypes may benefit more from undergoing DES.

A few of the genes in positive findings, such as MECP2 and SHANK3, are well-established genes for ASDs, whereas the vast majority (92.5%) of the unique ASD-related genes among positive or likely positive characterized and candidate genetic etiologies were recently identified (after January 2012) as “nonclassical” ASD genes with novel pathogenic sequence variants. This fact indicates that alterations in newly characterized genes may account for a significant number of individuals with ASD or autistic features. Indeed, three of the newly characterized ASD genes were recurrent (SETD5, ADNP, and WDR45) among the positive or likely positive findings in characterized and candidate genetic etiologies in the ASD cohort. The identification of recurrent mutations in newly characterized genes from an unselected laboratory cohort corroborates their role in ASDs. However, the phenotypic spectrum and underlying disease mechanism of the recently implicated genes remain to be clarified. Overall, these observations highlight the utility of an unbiased screening method interrogating all genes in the human genome as a molecular diagnostic tool for individuals with neurodevelopmental disorders, including ASDs. In comparison, the gene content included in a gene panel may not be flexible and likely does not include the analysis of the most recently characterized genes.

The advent of cost-effective, trio-based exome sequencing has shed light on the contribution of de novo alterations in ASD incidence. It is estimated that de novo loss-of-function mutations contribute to a minimum of 10% of simplex ASD cases, and de novo missense mutations, up to 10% of affected children.³⁰x30Ronemus, M., Iossifov, I., Levy, D., and Wigler, M. The role of de novo mutations in the genetics of autism spectrum disorders. Nat Rev Genet. 2014; 15: 133–141

Crossref | PubMed | Scopus (124)See all References In our study, the majority of positive or likely positive findings in the ASD cohort were the result of de novo mutations, supporting the de novo paradigm demonstrated in ASDs. Of all the de novo alterations, 53.8% were loss-of-function mutations, and the rest 46.2%, missense alterations.

The majority of the proteins encoded by the genes reported here function in synaptic, transcriptional, and chromatin-remodeling pathways, which have been repeatedly implicated in ASDs.³¹x31De Rubeis, S., He, X., Goldberg, A.P. et al. Synaptic, transcriptional and chromatin genes disrupted in autism. Nature. 2014; 515: 209–215

Crossref | PubMed | Scopus (525)See all References^{, 32}x32Pinto, D., Delaby, E., Merico, D. et al. Convergence of genes and cellular pathways dysregulated in autism spectrum disorders. Am J Hum Genet. 2014; 94: 677–694

Abstract | Full Text | Full Text PDF | PubMed | Scopus (269)See all References Moreover, pathogenic mutations were identified in well-established ASD genes as well as in genes associated with other conditions such as ID (CTNNB1, CUL4B, DYNC1H1, ELP2, FTSJ1, IQSEC2, KCNH1, KMT2A, MECP2, MTOR, SETD5, SHANK3, TCF4, UBE3A, WDR45, ZBTB20) and epilepsy (CACNA1A, CHD2, CDKL5, SCN8A, SYN1), supporting the broad phenotypic spectrum and shared molecular pathways. Although the genetic heterogeneity poses a great challenge for developing therapies for ASDs, the few major signaling pathways implicated in ASDs, such as the PI3K-mammalian target of rapamycin signaling cascade, have been targets of active research for therapeutic implications.³³x33Sahin, M. and Sur, M. Genes, circuits, and precision therapies for autism and related neurodevelopmental disorders. Science. 2015; 350

Crossref | PubMed | Scopus (9)See all References Molecular diagnosis through DES can aid in the stratification for clinical trials and inform target identification.

Previous reports assessing the diagnostic yield of DES in ASDs have utilized relatively small samples and reported the highest diagnostic yields among the more complex ASD phenotypes.⁷x7Tammimies, K., Marshall, C.R., Walker, S. et al. Molecular Diagnostic Yield of Chromosomal Microarray Analysis and Whole-Exome Sequencing in Children With Autism Spectrum Disorder. JAMA. 2015; 314: 895–903

Crossref | PubMed | Scopus (53)See all References^{, 10}x10Lee, H., Deignan, J.L., Dorrani, N. et al. Clinical exome sequencing for genetic identification of rare Mendelian disorders. JAMA. 2014; 312: 1880–1887

Crossref | PubMed | Scopus (219)See all References Tammimies et al.⁷x7Tammimies, K., Marshall, C.R., Walker, S. et al. Molecular Diagnostic Yield of Chromosomal Microarray Analysis and Whole-Exome Sequencing in Children With Autism Spectrum Disorder. JAMA. 2015; 314: 895–903

Crossref | PubMed | Scopus (53)See all References reported a diagnostic yield of 16.7% among the most complex cases and 28.6% for those with less evidence of dysmorphology, whereas cases with the least complex phenotype received a molecular diagnosis 3.1% of the time. In our ASD cohort, multiple congenital anomalies were observed in 13.5% of patients and positive findings were slightly more frequent among patients with multiple congenital anomalies when compared with patients without these clinical features (45.5% vs 22.7%; P = 0.03). These results support the findings by Tammimies et al.⁷x7Tammimies, K., Marshall, C.R., Walker, S. et al. Molecular Diagnostic Yield of Chromosomal Microarray Analysis and Whole-Exome Sequencing in Children With Autism Spectrum Disorder. JAMA. 2015; 314: 895–903

Crossref | PubMed | Scopus (53)See all References on the utility of whole exome sequencing as a first-tier ASD test especially for children with physical and congenital anomalies. The majority of the patients in the ASD cohort have an additional neurological phenotype, such as an epilepsy and/or seizure phenotype or a psychiatric condition in addition to autistic features. Moreover, we observed a higher rate of ID in our ASD cohort compared with the varying estimates (16.7% to 84%) generally reported.³⁴x34Postorino, V., Fatta, L.M., Sanges, V. et al. Intellectual disability in Autism Spectrum Disorder: Investigation of prevalence in an Italian sample of children and adolescents. Res Dev Disabil. 2016; 48: 193–201

Crossref | PubMed | Scopus (4)See all References These findings imply that our cohort represents the complex ASD cases that are referred for exome sequencing after long and extensive testing process, and therefore the diagnostic yield reported here may be closer to what is observed in genetics clinics for patients with no recognizable syndromes. The rate of ASD and autistic features is higher in our cohort than in the population (1% to 2%), which was expected because the indication for testing for most individuals was neurological dysfunction.

In addition to the “first-tier” testing (karyotype, CMA, and fragile X), the current ACMG practice guidelines recommend testing for single genes such as UBE3A, PTEN, and MECP2 among many others when suspecting ASDs.³⁵x35Schaefer, G.B. and Mendelsohn, N.J. Professional Practice Guidelines Committee. Clinical genetics evaluation in identifying the etiology of autism spectrum disorders: 2013 guideline revisions. Genet Med. 2013; 15: 399–407

Crossref | PubMed | Scopus (106)See all References In our cohort, however, positive or likely positive findings in characterized genes and candidate genetic etiologies in all these genes were underrepresented (7.1%) likely because patients with mutations in these genes might already have a positive result through gene or gene panel sequencing and thus there was no need for DES.

Nearly all the patients in our cohort had a “first-tier” test performed, and of these, over 90% had some form of an array before testing in our laboratory. These findings imply that our cohort represents patients undiagnosed with the tests commonly recommended for the evaluation of neurodevelopmental disabilities and generally covered by insurance, whereas DES coverage varies widely between insurance policies. However, our study shows a significant increase in the diagnostic rate for neurodevelopmental disorders compared with traditional “first-tier” testing. This observation is of importance to the patients and their families because a diagnosis may alter the clinical management, help predict recurrence risks, inform prognosis, and end their long and invasive “diagnostic odyssey.” The information is significant to the clinicians who have been unable to provide answers to these patients. Moreover, the high diagnostic rate of DES reflects the potential medical-economic savings to both patients and insurance companies.³⁶x36Valencia, C.A., Husami, A., Holle, J. et al. Clinical Impact and Cost-Effectiveness of Whole Exome Sequencing as a Diagnostic Tool: A Pediatric Center's Experience. Front Pediatr. 2015; 3: 67

Crossref | PubMedSee all References^{, 37}x37Monroe, G.R., Frederix, G.W., Savelberg, S.M. et al. Effectiveness of whole-exome sequencing and costs of the traditional diagnostic trajectory in children with intellectual disability. Genet Med. 2016; 18: 949–956

Crossref | PubMed | Scopus (26)See all References