Novel Clustering of Items From the Autism Diagnostic Interview-Revised to Define Phenotypes Within Autism Spectrum Disorders

Department of Biochemistry and Molecular Biology, The George Washington University Medical Center, 2300 Eye St., N.W., Washington, DC 20037, USA.
Autism Research (Impact Factor: 4.33). 04/2009; 2(2):67-77. DOI: 10.1002/aur.72
Source: PubMed


Heterogeneity in phenotypic presentation of Autism spectrum disorders has been cited as one explanation for the difficulty in pinpointing specific genes involved in autism. Recent studies have attempted to reduce the "noise" in genetic and other biological data by reducing the phenotypic heterogeneity of the sample population. The current study employs multiple clustering algorithms on 123 item scores from the Autism Diagnostic Interview-Revised (ADI-R) diagnostic instrument of nearly 2,000 autistic individuals to identify subgroups of autistic probands with clinically relevant behavioral phenotypes in order to isolate more homogeneous groups of subjects for gene expression analyses. Our combined cluster analyses suggest optimal division of the autistic probands into four phenotypic clusters based on similarity of symptom severity across the 123 selected item scores. One cluster is characterized by severe language deficits, while another exhibits milder symptoms across the domains. A third group possesses a higher frequency of savant skills while the fourth group exhibited intermediate severity across all domains. Grouping autistic individuals by multivariate cluster analysis of ADI-R scores reveals meaningful phenotypes of subgroups within the autistic spectrum, which we show, in a related (accompanying) study, to be associated with distinct gene expression profiles.

  • Source
    • "elicit spontaneous behavior (Lord et al., 2000). It is commonly used in conjunction with ADI-R, which consists of semi-structured interview conducted with the parents in order to detect abnormalities that are consistent with deficits in language, social, behavioral, and cognitive functions (Akshoomoff et al., 2006; Hu and Steinberg, 2009). Since their implementation, ADOS and ADI-R have gained wide clinical use because of their ability to differentiate children with autism from those with other neurodevelopmental disorders (Reaven et al., 2008). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Autism spectrum disorder (ASD) is a set of neurodevelopmental disorders that is among the most severe in terms of prevalence, morbidity and impact to the society. It is characterized by complex behavioral phenotype and deficits in both social and cognitive functions. Although the exact cause of ASD is still not known, the main findings emphasize the role of genetic and environmental factors in the development of autistic behavior. Environmental factors are also likely to interact with the genetic profile and cause aberrant changes in brain growth, neuronal development, and functional connectivity. The past few years have seen an increase in the prevalence of ASD, as a result of enhanced clinical tests and diagnostic tools. Despite growing evidence for the involvement of endogenous biomarkers in the pathophysiology of ASD, early detection of this disorder remains a big challenge. This paper describes the main behavioral and cognitive features of ASD, as well as the symptoms that differentiate autism from other developmental disorders. An attempt will be made to integrate all the available evidence which point to reduced brain connectivity, mirror neurons deficits, and inhibition-excitation imbalance in individuals with ASD. Finally, this review discusses the main factors involved in the pathophysiology of ASD, and illustrates some of the most important markers used for the diagnosis of this debilitating disorder. Copyright © 2015. Published by Elsevier Ltd.
    Full-text · Article · Apr 2015 · International journal of developmental neuroscience: the official journal of the International Society for Developmental Neuroscience
  • Source
    • "large patient sample sizes is expected to test if current findings can be extended to different phenotypic subgroups of ASD patients. The phenotypic subgroup classification of ASD patients can be obtained via a recently developed clustering analysis method [Hu and Steinberg, 2009]. All the abnormal links shown in Figure 5b demonstrate increased connectivity. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Neuroimaging has uncovered both long-range and short-range connectivity abnormalities in the brains of individuals with autism spectrum disorders (ASD). However, the precise connectivity abnormalities and the relationship between these abnormalities and cognition and ASD symptoms have been inconsistent across studies. Indeed, studies find both increases and decreases in connectivity, suggesting that connectivity changes in the ASD brain are not merely due to abnormalities in specific connections, but rather, due to changes in the structure of the network in which the brain areas interact (i.e., network topology). In this study, we examined the differences in the network topology between high-functioning ASD patients and age and gender matched typically developing (TD) controls. After quantitatively characterizing the whole-brain connectivity network using diffusion tensor imaging (DTI) data, we searched for brain regions with different connectivity between ASD and TD. A measure of oral language ability was then correlated with the connectivity changes to determine the functional significance of such changes. Whole-brain connectivity measures demonstrated greater local connectivity and shorter path length in ASD as compared to TD. Stronger local connectivity was found in ASD, especially in regions such as the left superior parietal lobule, the precuneus and angular gyrus, and the right supramarginal gyrus. The relationship between oral language ability and local connectivity within these regions was significantly different between ASD and TD. Stronger local connectivity was associated with better performance in ASD and poorer performance in TD. This study supports the notion that increased local connectivity is compensatory for supporting cognitive function in ASD. Hum Brain Mapp, 2012. © 2012 Wiley Periodicals, Inc.
    Full-text · Article · Feb 2014 · Human Brain Mapping
  • Source
    • "Many previous subgrouping efforts also lacked ascertainment of biomarkers or comorbidities commonly observed in individuals with ASD. As quantitative traits that are associated with ASD but not required for diagnosis, this data may improve our ability to identify more genetically similar subgroups(Gottesman & Gould 2003; Hu et al. 2009; Leyfer et al. 2006; Veenstra-Vanderweele & Cook 2004; Wiggins et al. 2012). For example, multiple studies have implicated the same chromosomal region, 7q35, and candidate gene, CNTNAP2, by focusing on expression of specific language impairment (SLI) in ASD. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Autism Spectrum Disorder (ASD) is a complex neurodevelopmental disorder with strong evidence for genetic susceptibility. However, the effect sizes for implicated chromosomal loci are small, hard to replicate, and current evidence does not explain the majority of the estimated heritability. Phenotypic heterogeneity could be one phenomenon complicating identification of genetic factors. We used data from the Autism Diagnostic Interview-Revised, Autism Diagnostic Observation Schedule, Vineland Adaptive Behavior Scales, head circumferences, and ages at exams as classifying variables to identify more clinically similar subgroups of individuals with ASD. We identified two distinct subgroups of cases within the Autism Genetic Resource Exchange dataset, primarily defined by the overall severity of evaluated traits. In addition, there was significant familial clustering within subgroups (OR ≈ 1.38-1.42, p < 0.00001), and genotypes were more similar within subgroups compared to the unsubgrouped dataset (Fst = 0.17+/-0.0.0009). These results suggest that the subgroups recapitulate genetic etiology. Using the same approach in an independent dataset from the Autism Genome Project, we similarly identified two distinct subgroups of cases and confirmed this severity-based dichotomy. We also observed evidence for genetic contributions to subgroups identified in the replication dataset. Our results provide more effective methods of phenotype definition that should increase power to detect genetic factors influencing risk for ASD.
    Full-text · Article · Dec 2013 · Genes Brain and Behavior
Show more